Immune reactions after modern lamellar (DALK, DSAEK, DMEK) versus conventional penetrating corneal transplantation

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Progress in Retinal and Eye Research journal homepage: www.elsevier.com/locate/preteyeres

Immune reactions after modern lamellar (DALK, DSAEK, DMEK) versus conventional penetrating corneal transplantation Deniz Hosa,b, Mario Matthaeia, Felix Bocka,b, Kazuichi Maruyamac, Maria Notaraa,b, Thomas Clahsena, Yanhong Houa, Viet Nhat Hung Lea,d, Ann-Charlott Salabarriaa, Jens Horstmanna, Bjoern O. Bachmanna, Claus Cursiefena,b,* a

Department of Ophthalmology, University of Cologne, Cologne, Germany Center for Molecular Medicine Cologne (CMMC), University of Cologne, Cologne, Germany c Department of Innovative Visual Science, Graduate School of Medicine, Osaka University, Japan d Department of Ophthalmology, Hue College of Medicine and Pharmacy, Hue University, Viet Nam b

A R T I C LE I N FO

A B S T R A C T

Keywords: Corneal transplantation Keratoplasty Graft rejection DALK DSAEK DMEK Lymphangiogenesis Stem cell transplantation

In the past decade, novel lamellar keratoplasty techniques such as Deep Anterior Lamellar Keratoplasty (DALK) for anterior keratoplasty and Descemet stripping automated endothelial keratoplasty (DSAEK)/Descemet membrane endothelial keratoplasty (DMEK) for posterior keratoplasty have been developed. DALK eliminates the possibility of endothelial allograft rejection, which is the main reason for graft failure after penetrating keratoplasty (PK). Compared to PK, the risk of endothelial graft rejection is significantly reduced after DSAEK/ DMEK. Thus, with modern lamellar techniques, the clinical problem of endothelial graft rejection seems to be nearly solved in the low-risk situation. However, even with lamellar grafts there are epithelial, subepithelial and stromal immune reactions in DALK and endothelial immune reactions in DSAEK/DMEK, and not all keratoplasties can be performed in a lamellar fashion. Therefore, endothelial graft rejection in PK is still highly relevant, especially in the “high-risk” setting, where the cornea's (lymph)angiogenic and immune privilege is lost due to severe inflammation and pathological neovascularization. For these eyes, currently available treatment options are still unsatisfactory. In this review, we will describe currently used keratoplasty techniques, namely PK, DALK, DSAEK, and DMEK. We will summarize their indications, provide surgical descriptions, and comment on their complications and outcomes. Furthermore, we will give an overview on corneal transplant immunology. A specific focus will be placed on endothelial graft rejection and we will report on its incidence, clinical presentation, and current/future treatment and prevention options. Finally, we will speculate how the field of keratoplasty and prevention of corneal allograft rejection will develop in the future.

1. Introduction 1.1. Anatomy of the cornea The cornea is the transparent and avascular front part of the eye playing a central role in light refraction and providing a physical barrier to the external environment. It consists of 5 layers (as depicted in Fig. 1): the epithelium, Bowman's layer, stroma, Descemet membrane and endothelium (DelMonte and Kim, 2011). The outermost layer of the cornea, the epithelium, comprises of 5–6 layers of stratified non-keratinized cells measuring approximately 50 μm in humans. The cells in the uppermost layers form intercellular tight junctions which prevent the invasion of microorganisms and other

*

potentially harmful exogenous factors (Eghrari et al., 2015). The corneal epithelium is maintained by a population of stem cells residing at the basal layers of the vascularized junction between the cornea and the conjunctiva, namely the limbus (Notara and Daniels, 2008). The highly (lymph)vascularized Palisades of Vogt are stromal invaginations located in the limbus, and are considered a putative limbal epithelial stem cell niche (Notara et al., 2010). The limbus is the border between physiologically avascular cornea and heavily hem- and lymphvascularized conjunctiva. The corneal basal epithelial cell layer rests on the acellular Bowman's layer which is a strong membrane consisting of randomly oriented collagen fibrils (Komai and Ushiki, 1991). Contrary, the corneal stroma features a highly organized structure consisting of highly

Corresponding author. Department of Ophthalmology, University of Cologne, Kerpener Str. 62, 50924, Cologne, Germany. E-mail address: [email protected] (C. Cursiefen).

https://doi.org/10.1016/j.preteyeres.2019.07.001 Received 4 December 2018; Received in revised form 1 July 2019; Accepted 2 July 2019 1350-9462/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).

Please cite this article as: Deniz Hos, et al., Progress in Retinal and Eye Research, https://doi.org/10.1016/j.preteyeres.2019.07.001

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

improvements in surgical techniques resulted in the modern era of PK. Thus, keratoplasty techniques have evolved from lamellar to penetrating approaches, which was the mainstay of corneal transplantation surgery until recently. In the past decade, however, novel minimallyinvasive lamellar keratoplasty techniques for a variety of corneal diseases have been developed, which result in generally better outcomes when compared to full-thickness PK. This led to fundamental changes in corneal transplantation, and the majority of keratoplasties can be performed in a lamellar fashion today. The current technique for anterior corneal diseases is Deep Anterior Lamellar Keratoplasty (DALK) using e.g. the “big bubble technique” introduced by Anwar (Anwar and Teichmann, 2002). DALK eliminates the possibility of endothelial graft rejection, which is the main reason for graft failure after PK, but does not prevent epithelial, subepithelial or stromal immune reactions (Gonzalez et al., 2017). The most popular techniques for posterior lamellar keratoplasty today are Descemet stripping automated endothelial keratoplasty (DSAEK) and Descemet membrane endothelial keratoplasty (DMEK), the latter introduced by Melles (Melles et al., 2006). The risk of developing an endothelial graft rejection after posterior lamellar keratoplasty is significantly reduced when compared to PK, especially after DMEK (Anshu et al., 2012; Hos et al., 2017b). Thus, even with modern lamellar grafting techniques there are epithelial, subepithelial and stromal immune reactions in DALK and endothelial immune reactions in DSAEK/DMEK. Furthermore, not all corneal transplantations can be performed in a lamellar fashion, because these novel techniques are not available worldwide and there are still corneal pathologies that require the replacement of all corneal layers. In these cases, immune-mediated endothelial graft rejection is highly relevant, especially in pathologically vascularized high-risk eyes constituting around 10% of all grafts, but currently available treatment options are still unsatisfactory. We therefore want to first review currently used keratoplasty techniques, namely PK, DALK, DSAEK, and DMEK. We will summarize their indications, describe the surgical techniques, and comment on their complications and outcomes. Second, we will give an overview on corneal transplant immunology. Furthermore, we will mainly focus on endothelial graft rejection and report on its incidence, clinical presentation, and current/future treatment and prevention options (except for DALK, where we will briefly summarize recent findings regarding epithelial, subepithelial, and stromal rejections). Finally, we will speculate on how the field of corneal transplantation and prevention of corneal allograft rejection will develop in the future.

Fig. 1. Anatomy of the cornea. Schematic representation of the cornea depicting its distinct anatomical layers, namely the epithelium, Bowman's layer, the stroma, the Descemet membrane and the endothelium.

organized collagen I and V fibers to which the cornea owes its transparency and biomechanical strength. The corneal keratocytes, mainly localized in the anterior stroma, produce these collagen fibrils while remaining relatively sparse within the stromal tissue (Hassell and Birk, 2010). The Descemet membrane underlies the corneal stroma and consists mainly of collagen IV which is produced by a monolayer of quiescent endothelial cells, which preform the key function of keeping the cornea dehydrated by continuously pumping fluid out of the stroma and into the aqueous humour (AqH) (Yu et al., 2011). In addition to the previously described structures, the cornea contains numerous nerves, immune cells, putative mesenchymal stem cells, and hem- and lymphvascular sprouts close to the limbus. If corneal transparency, integrity or function are compromised, corneal transplantation is the therapy of choice. 1.2. History of keratoplasty - from lamellar to penetrating to modern lamellar surgery

2. Types of keratoplasty, complications, preventive measures, and outcomes

Corneal transplantation (keratoplasty) is the oldest and most successful form of tissue transplantation. First attempts to surgically treat opaque corneas (by superficial keratectomy) can be traced back to 100200 A.D. by Galen (Crawford et al., 2013; Mannis and Krachmer, 1981; Moffatt et al., 2005). Techniques for anterior lamellar keratoplasty were established during the late 1800s: von Walther and Koenigshofer recommended the excision of the anterior cornea without grafting the deeper layers and the Descemet's membrane. In 1886, Von Hippel performed the first partially successful lamellar (xeno)transplant, where a full-thickness cornea of a rabbit was transplanted onto the lamellar corneal bed of a human eye. The vision of the patient transiently improved, but the graft became opaque afterwards (Crawford et al., 2013; Mannis and Krachmer, 1981; Moffatt et al., 2005). It was Eduard Zirm in 1905, who performed the first successful keratoplasty where the graft remained clear. The graft was derived from a human eye that was enucleated directly prior to full-thickness penetrating keratoplasty (PK), where the graft was secured with overlay sutures. At 6 months postoperatively, the patient had a visual acuity of 6/36 (Zirm, 1906). In the following decades, there was no significant further progress in corneal transplantation due to the lack of knowledge of general principles of antisepsis and immunology, as well as surgical techniques. By the mid1950s, the introduction of antibiotics, corticosteroids and

2.1. Penetrating keratoplasty (PK) Although the number of performed PK procedures is decreasing because newer lamellar transplantation methods have been developed (Flockerzi et al., 2018), PK is still an important and necessary surgical procedure, particularly when newer surgical methods are not reasonable or fail. 2.1.1. Indications and surgery PK is normally applied to recover the visual function of opaque or morphologically abnormal corneas. Moreover, corneal perforation due to infections or immunological diseases such as rheumatoid arthritis is an indication for PK. Prolonged endothelial dysfunction can also lead to stromal scarring and neovascularization and PK will become the first choice. One further indication is corneal opacification due to bacterial, viral, or fungal infection, which can damage not only the anterior cornea but also the corneal endothelium. Patients with keratoconus have also been treated by PK, but DALK has become the first-line treatment here. PK surgery is different from other types of ophthalmic procedures 2

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

techniques aiming at replacing diseased corneal stroma by healthy donor stroma retaining only the recipient's Descemet's membrane with endothelial cells. In general, all corneal stromal diseases with healthy corneal endothelium can be treated by DALK. The most common indication for DALK is advanced keratoconus that cannot be treated by contact lenses or corneal collagen crosslinking (CXL) e.g. due to scarring or extensive thinning (Schaub et al., 2017d). Especially young patients with less severe but progressive disease might first be treated by CXL, as a global expert panel recently stated (Gomes et al., 2015). By far less numbers of patients become treated by DALK for other reasons like ectasia after refractive corneal surgery, corneal stromal scars and stromal thinning after trauma or inflammation or stromal obscurations in corneal dystrophies (Unal et al., 2013). If corneal transplantation is necessary in case of acute keratitis a DALK rather than a PK can be considered under certain circumstances to lower the risk of intraocular inflammation and the risk for graft rejection. Donor preparation in DALK comprises trephination similar to PK after removal of the Descemet membrane. Donor parameters like donor age, death to preservation time, preservation time or preservation method (cold or warm storage) do not correlate with functional outcomes (Schaub et al., 2017a). Different techniques for the removal of the recipient's corneal stroma exist. Corneal stroma can be either removed by manual lamellar dissection or by high-pressure dissection of the Descemet's membrane from the stroma followed by stromal removal. The beginning of both approaches is similar to PK in which after marking the cornea with a radial marker a trephination is performed. In contrast to PK, it is important that the trephination ends inside the corneal stroma close to the Descemet membrane without perforation (incomplete trephination). Depending on the trephination system the actual resulting trephination depth can be estimated to a certain degree only. Trephination with a femtosecond laser allows a very precise trephination depth, making femtosecond laser-assisted surgery a promising tool for DALK (Buzzonetti et al., 2010; Salouti et al., 2019). In manual lamellar dissection crescent blades are usually used for lamellar dissection. In a variation of lamellar dissection, a blunt instrument (e.g. a spatula) can be introduced into a deep stromal pocket, close to the Descemet membrane. From that point, a blunt dissection between deep stromal layers can be performed. If available, visualization of the corneal stroma and the Descemet membrane can be improved by the use of intraoperative optical coherence tomography (iOCT) (Siebelmann et al., 2016). iOCT machines are available as microscope integrated imaging devices allowing the use of OCT in the sterile surgical field. It has been shown that iOCT allows visualizing all different steps of preparation during lamellar corneal surgery. The injection of air into deep stromal layers is a successful and very common method for the separation of corneal stroma from the Descemet membrane (Anwar and Teichmann, 2002; Riss et al., 2012, 2013). It has been shown that the dissection line after successful formation of a big bubble can be between the Descemet membrane and corneal stroma as well as in between very deep stromal lamellae (Dua et al., 2013). The latter is called a type I bubble and the first a type II bubble. The small layer of deep corneal stroma attached to the Descemet membrane in type I bubble formation was named after its discoverer (Dua's layer) (Dua et al., 2013). However, there is no clear evidence that this layer is an anatomical structure rather than an artificial formation of a stromal layer caused by pneumodissection (Schlotzer-Schrehardt et al., 2015).

such as cataract or vitreous surgery, which usually proceed in a closed space. In contrast, parts of the PK procedure are implemented with the eye wide open. Therefore, surgery time is an important issue since the risk of severe complications such as expulsive bleeding increases over time. The shape of the eye is usually maintained by intraocular pressure, but the eye will not be able to maintain intraocular pressure during PK and will collapse especially in vitrectomized and aphacic eyes. Moreover, a vitreous body with high pressure can suddenly flow out during corneal resection and choroidal bleeding may develop. The size of the donor cornea should in our experience be 0.25 mm larger than the excised recipient cornea. The donor cornea is prepared using e.g. a Baron vacuum trephine and cut from the endothelial side. Alternatively, nonmechanical trephination techniques can be used (Preclik et al., 2010; Reimer et al., 2012). In the recipient, a scleral ring (Flieringa ring) might be used to prevent collapse after corneal resection (especially after vitrectomy or in aphakia). Afterwards, the center of the cornea is determined and marked. For resection of the recipient cornea, a Baron vacuum trephine and punches are usually used. Excision of (both donor and) host tissue can also be performed non-mechanically e.g. by using excimer or femtosecond lasers (Daniel et al., 2016; Seitz et al., 1999). After excision of the patient's cornea, surgical iridectomy is key to avoid Urrets-Zavalia syndrome. Then, the donor cornea is sutured to the recipient. Initially, eight temporary sutures are placed to stabilize the anterior chamber, followed by two continuous or interrupted sutures. Interrupted sutures might be used for easier manipulation if sutures should loosen afterwards in a weak site, e.g. after infections keratitis. Astigmatic adjustment during surgery has a serious effect on visual acuity after PK, and a Mallory/Placido disc ring can be used to adjust suture strength at the end of PK surgery to reduce astigmatism. The reader should note that the description of the PK procedure here is based on the authors' own preferred technique and that many variations exist. For example, many surgeons often do not perform an iridectomy, use a same size trephine for the donor and recipient cornea, and do not always use a Flieringa ring after vitrectomy or in aphakia. 2.1.2. Complications, preventive measures and outcomes The risk of vitreous body outflow and expulsive hemorrhage can be reduced by lowering vitreous body pressure before excising the host cornea. Appropriate anesthesia also reduces the risk of these complications. High intraocular pressure in the immediate postoperative period is usually caused by viscoelastic material remaining in the anterior chamber. Appropriate treatment is also required for infections or graft dehiscence with aqueous humor leakage through the graft-host interface. Emergency suturing is required for graft dehiscence to prevent further complications such as intraocular infection or bleeding. Drug toxicity or infections can cause persistent epithelial defects after transplantation. Therefore, close follow-up observations are needed to determine the cause and to ensure appropriate treatment. We have previously analyzed the long-term visual outcome after PK for keratoconus. In this retrospective study that included 456 patients, mean best spectacle corrected visual acuity (BSCVA) at eight years postoperatively was 20/25, with a mean (topographic) astigmatism of 4.2 diopters (Preclik et al., 2010). In a further retrospective study analyzing the long-term visual outcome after PK for bullous keratopathy in 413 patients, we have shown that mean BSCVA increased to 20/32 with a mean (topographic) astigmatism of 3.9 diopters at eight years postoperatively (Reimer et al., 2012). Interestingly, eight-year BSCVA of patients with preoperative BSCVA ≤20/200 was significantly reduced when compared to patients with better preoperative BSCVA in both studies (Preclik et al., 2010; Reimer et al., 2012).

2.2.2. Complications, preventive measures and outcomes The spectrum of complications in DALK shares some similarities to complications occurring during PK. However, in contrast to PK primary graft failure due to endothelial decompensation is less likely to occur after DALK. Moreover, the risk of intraoperative hypotension and its associated risk of suprachoroidal bleeding is considered to be lower in DALK compared to PK. However, there are some DALK specific complications which are related to the Descemet membrane-stroma-

2.2. Deep anterior lamellar keratoplasty (DALK) 2.2.1. Indications and surgery Deep anterior lamellar keratoplasty (DALK) comprises surgical 3

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

dysfunction. DSAEK includes selective removal of the diseased Descemet and corneal endothelium and transplantation of a stroma-Descemet-endothelium lamella, whereas in DMEK, only the donor-Descemet with endothelium is grafted.

interface that does not exist in PK. A very common intraoperative complication is the inability of Descemet membrane separation from the corneal stroma. The drawback of pneumatic dissection is the reduced success rate varying roughly between 50 and 90% even in the hands of experienced surgeons (Goweida, 2015; Sarnicola et al., 2016). After unsuccessful attempts of air injection without big bubble formation, a manual dissection of corneal stroma can be considered which results in a low overall conversion rate to PK (Riss et al., 2013; Schaub et al., 2017d). During manual lamellar preparation and pneumatic dissection of the Descemet membrane, perforations of the Descemet membrane can occur. In this case, preparation in that area should be stopped and if some deeper stromal tissue is still in place it should be left for sealing the perforated area. At the end of the surgery a so called double anterior chamber formation, which simply describes the distance of the Descemet membrane from the corneal stroma after DALK, can be avoided by inflation of the anterior chamber with air or sulfur hexafluoride (SF6) gas at 20% concentration. Before inflation it needs to be ensured that a sufficient iridotomy or iridectomy with an opening at 6 o'clock position is present. It is generally recommended to perform a YAG-laser iridotomy at least the day before the planned DALK. Once the anterior chamber is filled with gas, residual fluid between the Descemet membrane and the interface can be removed via the stroma-stroma-interface. Postoperative complications comprise of suture-related problems like suture loosening and suture related infections, high astigmatism, persistent corneal erosion as well as recurrences of the underlying disease in corneal dystrophies in similar frequency as after PK. The most common DALK specific postoperative complication in the early phase after surgery is the formation of a double anterior chamber due to the detachment of the Descemet membrane. Reasons for persistent Descemet membrane detachment after DALK include macro- and microperforations of the Descemet membrane, which can be treated by a intracameral application of an air bubble to increase the outflow of fluid from the Descemet membrane-stroma-interface. It is generally accepted that the endothelial cell density after DALK is higher than after PK (Borderie et al., 2012; Kubaloglu et al., 2011; Reinhart et al., 2011). Studies comparing DALK and PK for keratoconus demonstrated less endothelial cell loss when the patient's own Descemet membrane could be preserved (Borderie et al., 2011, 2012; Cheng et al., 2011; Kubaloglu et al., 2011). Moreover, the continuing annual rates of endothelial cell loss are less after DALK (−3.9%) than after PK (−7.8%). Most studies demonstrate no postoperative difference in astigmatism, spherical equivalent or maximum keratometry (Cheng et al., 2011; Schaub et al., 2017d; Sogutlu Sari et al., 2012). If separation of the Descemet membrane and the corneal stroma succeeds by pneumatic dissection, visual acuity and central corneal thickness are equal after DALK and PK (Borderie et al., 2012; Reinhart et al., 2011). However, when manual dissection for stroma removal in DALK is needed and not all stromal tissue can be removed, the resulting visual acuity is reduced and central corneal thickness is thicker compared to post-PK (Borderie et al., 2012).

2.3.1. Descemet stripping automated endothelial keratoplasty (DSAEK) 2.3.1.1. Indications and surgery. DSAEK is performed for the treatment of corneal endothelial disorders without advanced and irreversible scarification of the corneal stroma. Indications are mostly congruent to the spectrum of DMEK indications and will be specified in more detail in section 2.3.2. DMEK is generally considered to provide better functional and more cost-effective results than DSAEK (Gibbons et al., 2019). However, due to its increased thickness a DSAEK graft may be easier to unfold and may provide advantages in EK with complicated anatomical and surgical preconditions. This is mostly applicable in eyes of infants/newborn and eyes with hypotony and silicone oil. Tissue preparation for DSAEK may occur in an intraoperative setting by the ophthalmic surgeon or in a preoperative setting at the eye bank (Price et al., 2008). The corneal epithelium and the anterior stroma are removed at a nominal thickness of 300, 350 or 400 μm either manually for DSEK surgery or using a microkeratome for DSAEK surgery (Cursiefen and Kruse, 2008; Cursiefen et al., 2009b). Microkeratomecut, and particularly ultra-thin grafts have been linked to faster visual recovery and better visual outcome (Busin et al., 2013; Schaub et al., 2017b). The graft is routinely excised from the prepared lamella with an 8, 8.5 or 9 mm trephine (Cursiefen et al., 2009b). In the recipient, a clear-cornea tunnel and two paracenteses are created. A preexisting iridotomy may be spread open or an iridectomy may be performed. A large descemetorhexis is torn under air insufflation (Melles et al., 2004). The variety of implantation procedures may be divided into three methods: forceps, glides and mechanical injectors (Price et al., 2017). When implanted using forceps, the tissue is two- or trifolded and pulled with forceps into the anterior chamber. When implanted with a glide, unfolding of the graft is facilitated as it is drawn into the anterior chamber (Fig. 2). Insertion devices may be preloaded and may show significant differences in design, construction material, insertion technique, dimensions, incision requirements and endothelial cell loss (Khan et al., 2015). The anterior chamber is restored. The inserted stroma-Descemet-endothelium lamella is positioned centrally in the recipient bed and fixed by injection of air or a 20% SF6 gas bubble. The

2.3. Endothelial keratoplasty (EK) Endothelial keratoplasty (EK) procedures comprise selective replacement of Descemet membrane and the adherent corneal endothelial cell (CEC) layer. EK is generally performed in a minimally invasive fashion avoiding an “open sky” situation. Moreover, only few or no sutures are used to close the surgical incisions. Important benefits of EK compared to PK include better functional outcome, faster visual recovery, reduced risk of hemorrhage and infection, less astigmatism, less corneal denervation, a tectonically more stable result, less vessel growth, and fewer graft rejection episodes (Anshu et al., 2012; Bachmann et al., 2008b, 2017; Baydoun et al., 2016a; Hos et al., 2017b). Two EK procedures, namely DSAEK and DMEK have emerged as standard techniques for the treatment of corneal endothelial

Fig. 2. Descemet stripping automated endothelial keratoplasty (DSAEK) –surgical technique. a) Preparation of a corneoscleral tunnel (arrows). b) – e) Marking of the central 9 mm, descemetorhexis and removal. f) – h) Implantation of the graft with a glide (arrows in g) and unfolding in the anterior chamber (arrows in h). i) Fixation of the graft with an air bubble. From (Cursiefen et al., 2008). 4

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

incisions are closed by hydrogenation or by sutures. Corneal surface massage and small stromal incisions may facilitate the removal of residual fluid from the interface (Price and Price, 2006). The reader should note that the DSEAK procedure described in this section is based on the authors' own preferences and that many other variations exist. 2.3.1.2. Complications, preventive measures and outcomes. The most common complication after DSAEK surgery is inadequate graft attachment. However, there is a significant decrease in complications within the learning curve of the surgeon (Price and Price, 2006). Due to spontaneous reattachments, a re-bubbling procedure is only required to treat advanced or complete graft detachments (Price et al., 2017). In addition, primary or secondary graft failure, rejection, carryover of debris into the interface, pupillary block, increased intraocular pressure and Urrets-Zavalia syndrome, epithelial invasion, retinal detachment, cystoid macular edema, and hemorrhage are less common complications (Suh et al., 2008). Graft survival after DSAEK seems to be at least as good as after PK (Ang et al., 2016; Hjortdal and Ehlers, 2009; Price et al., 2011). A hyperopic shift of about 1.13 diopters needs to be considered (Covert and Koenig, 2007; Price et al., 2017). After DSAEK, a significantly faster visual rehabilitation may be expected compared to PK. Li et al. found that the percentage of patients achieving 20/25 BCVA was 36.1% at 6 months and 70.4% at 3 years after DSAEK (Li et al., 2012a). The percentage of patients reaching BCVA of 20/20 was 11.1% at 6 months and 47.2% at 3 years (Li et al., 2012a). Wacker et al. observed BCVA of 20/25 or better in 26% of DSAEK patients at 1 year and in 56% of patients at 5 years (Wacker et al., 2016). These studies demonstrated that visual improvement does not plateau and suggest a continual remodeling process of the corneal tissue even five years after DSAEK (Li et al., 2012a; Wacker et al., 2016). CEC loss after DSAEK is about 32% at one year and about 7–9% annually over the following years (Wacker et al., 2016) resulting in a total CEC loss of about 47–55% at 5 years (Price et al., 2011, 2018; Wacker et al., 2016). Of note, although the total number of posterior lamellar keratoplasties performed in Germany is continuously increasing due to the dramatical increase in DMEK surgeries, the number of performed DSAEK surgeries decreased over the past years. Of 4169 posterior lamellar keratoplasties carried out in Germany during 2016, 319 were still performed as DSAEK (Flockerzi et al., 2018). Ultrathin (UT)-DSAEK may become an ideal combination of pros and cons of both DSAEK and DMEK in the future (Bucher et al., 2013).

Fig. 3. Descemet membrane endothelial keratoplasty (DMEK) – graft preparation and surgical technique. a) After circular scoring of the peripheral Descemet membrane and staining with trypan blue, the edge of the Descemet membrane is circularly lifted; b) The Descemet membrane is peeled with 2 forceps leaving the center unstripped; c) Trephination of the Descemet endothelium complex with 8 mm trephine; d and e) Application of graft orientation marks for intraoperative orientation with 1 mm trephine (green arrows); f) complete detachment of the graft; g) and h) loading of the graft into the shooter cartridge; i) Clear cornea tunnel in the recipient eye; j) Spreading of the iridotomy; k) Descemetorhexis under air insufflation; l) Injection of the graft; m) Unrolling of the graft; n) Control of correct positioning of the graft orientation marks (green arrows); o) fixation of the graft with an air bubble; p) flooding with 20% sulfur hexafluoride (SF6) gas. From (Matthaei et al., 2018).

in a second step. A variety of graft dissection methods and DMEK surgical techniques have been previously described, and the description of the DMEK procedure here is based on the authors' own preferred procedure and experience. We perform graft dissection at our institution as described previously (Kruse et al., 2014; Price et al., 2009a). Here, the edge of the Descemet membrane is circularly lifted towards the center with two pairs of suturing forceps leaving a central area of about 2 × 2mm unstripped. The Descemet endothelium complex is then trephined and stained with trypan blue. Finally, the graft is completely peeled off and is loaded into the cartridge of an intraocular lens shooter (Fig. 3). After graft dissection for DMEK in a first recipient, the anterior part of the same donor tissue may be used for a DALK-procedure in a second recipient (Heindl et al., 2011b; Schaub et al., 2015). In this process called split-cornea transplantation, a single graft may be used for one DMEK and one DALK surgery, respectively, thus contributing to the reduction of global donor tissue shortage (Heindl et al., 2011a, 2011b) (Fig. 4). It has been demonstrated by several studies that split corneal transplantation may be integrated as a safe procedure into clinical practice (Heindl et al., 2011b, 2013; Schaub et al., 2015). In the recipient, an iridotomy or iridectomy is required to prevent postoperative pupillary block after anterior chamber endotamponade. We usually perform a YAG-iridotomy the day before surgery in order to minimize the risk of intraoperative bleeding. Initially, a clear-cornea tunnel and two paracenteses are created. A preexisting iridotomy may be spread open or an iridectomy may be cut. A large Descemetorhexis is torn under air insufflation using a reversed Sinskey hook or Price hook. Under irrigation, the graft is then injected into the anterior chamber

2.3.2. Descemet membrane endothelial keratoplasty (DMEK) DMEK comprises selective transplantation of the corneal endothelium and the Descemet membrane (Cursiefen and Kruse, 2010). Main advantages of DMEK over previous posterior lamellar keratoplasty procedures are the very low graft rejection rate, the rapid visual recovery, and the fact that the surgical procedure itself requires little technical equipment. However, DMEK is considered an intervention with higher difficulty level and technically more demanding than DSAEK. 2.3.2.1. Indications and surgery. Indications for DMEK include all corneal disorders with advanced corneal endothelial dysfunction that lack irreversible scarring of the deep corneal stroma, such as Fuchs endothelial corneal dystrophy (FECD), pseudophakic/aphakic bullous keratopathy as well as endothelial decompensation following intraocular surgery and graft failure following previous keratoplasty (Cursiefen and Kruse, 2010). More rare indications include pseudoexfoliation keratopathy, irido-corneo-endothelial (ICE) syndrome, central Haab striae-related edema in buphthalmus, and endothelial dystrophies other than FECD such as posterior polymorphic corneal dystrophy or congenital hereditary endothelial dystrophy (Matthaei et al., 2018). Scarring of the anterior cornea is not a general contraindication for DMEK and often regresses postoperatively or may be removed by phototherapeutic keratectomy 5

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Cystoid macular edema (CME) occurs after DMEK with an incidence of about 10% and is therefore a common complication (Heinzelmann et al., 2015; Hoerster et al., 2016). A significant drop in incidence to almost complete elimination of CME as a complication after DMEK was seen after the introduction of intensified topical postoperative corticosteroid therapy, with hourly topical corticosteroids in the first postoperative week (Hoerster et al., 2016). So far, there is no report on the use of non-steroidal anti-inflammatory drugs (NSAIDs) to reduce CME after DMEK/Triple-DMEK surgery. It is imaginable that NSAIDs could be useful, as in cataract surgery (Ray and D'Amico, 2002; Wielders et al., 2017). Nevertheless, clinical trials would be needed to provide evidence whether NSAIDS are effective in reducing CME after DMEK/ Triple-DMEK surgery. Particularly when DMEK is combined with cataract surgery, a hyperopic shift needs to be considered in the selection of the IOL. Refractive changes occur most likely due regression of corneal edema. This hyperopic shift is usually about 0.6 diopters and counteracted by many surgeons through myopic target refraction. The risk of hyperopic shift seems to be increased especially in a central flat, oblate posterior cornea (Fritz et al., 2019). The average hyperopic shift appears to be slightly lower after DMEK than after DSAEK (Deng et al., 2018; Droutsas et al., 2016; Hamzaoglu et al., 2015; Tourtas et al., 2012). Our own data demonstrate a mean increase in BCVA from 20/62 before DMEK to 20/27 at 6 months and to 20/25 at 12 months after DMEK (Schrittenlocher et al., 2018). Ham et al. reported BCVA of 20/25 or better in 73% of eyes and of 20/20 or better in 44% of eyes at 6 months and of 20/25 or better in 83% of eyes and of 20/20 or better in 54% of eyes at 4 years (Ham et al., 2016). Peraza-Nieves reported BCVA of 20/ 25 or better in 81% of eyes and of 20/20 or better in 49% of eyes at 12 months and of 20/25 or better in 82% of eyes and of 20/20 or better in 52% of eyes at 2 years (Peraza-Nieves et al., 2017). Endothelial cell loss in the postoperative course seems to be similar between DMEK and DSAEK (Price et al., 2018). Endothelial cell loss after DMEK in our own studies was about 38% at 12 months (Schrittenlocher et al., 2018). Ham et al. reported endothelial cell loss after DMEK of 33.9% at 6 months, 37,6% at 1 year and 52,6% at 4 years (Ham et al., 2016). Peraza-Nieves reported endothelial cell loss after DMEK of 37% at 6 months, 40% at 1 year and 45% at 2 years (Peraza-Nieves et al., 2017), while Price et al. reported an endothelial cell loss of 48% at 5 years (Price et al., 2018). Visual recovery after DMEK has generally reached a stable level after six months and is considered to be faster than after DSAEK (Singh et al., 2017). In direct comparison, DMEK also provides better long-term visual acuity outcome than DSAEK (Stuart et al., 2018). In a current study, we could demonstrate that even eyes with poor preoperative visual acuity reach very good postoperative results, although visual acuity of < 20/80 preoperatively was associated with significantly longer recovery and reduced final BCVA (Schrittenlocher et al., 2019).

Fig. 4. Principle of split cornea transplantation to reduce human corneal tissue shortage. One donor cornea is used for two recipients by combining deep anterior lamellar keratoplasty (DALK) e.g. for a keratoconus patient and Descemet's membrane endothelial keratoplasty (DMEK) e.g. for a Fuchs' endothelial dystrophy patient. From (Heindl et al., 2011).

and is unrolled by gentle tapping on the cornea, and it may be helpful to vary the depth of the anterior chamber. Afterwards, the graft is fixed in the recipient bed with an air bubble and the anterior chamber is filled to about 80% with 20% SF6 gas (Fig. 3). 2.3.2.2. Complications, preventive measures and outcomes. Complications of DMEK include graft detachment, hemorrhage, secondary glaucoma, Urrets-Zavalia syndrome, primary and secondary graft failure,intraocular lens (IOL) calcifications, and macular edema (Gorovoy, 2014; Schrittenlocher et al., 2018). Graft detachment represents the most common complication after DMEK and seems to occur more often than after DSAEK (Stuart et al., 2018). The number of re-bubblings was significantly reduced after changing the endotamponade from room air to SF6 20% gas (Guell et al., 2015; Schaub et al., 2017c) and is now less than 3% depending on the study (Gorovoy, 2014; Rodriguez-Calvo-de-Mora et al., 2015). Small laminar detachments often attach spontaneously. Larger detachments, especially if they show rolled-up edges, often need to be treated by re-bubbling (Bucher et al., 2015). However, there are still no standardized criteria for the necessity and timing of re-bubbling. Increases in intraocular pressure after DMEK in the shorter interval are usually caused by pupillary block. The occurrence of a pupillary block can be prevented by a basal iridotomy or iridectomy, by an early release of gas from the anterior chamber or by incomplete endotamponade during the procedure. Calcification of the IOL is a rare complication with an incidence of about 2% (Schrittenlocher et al., 2017, 2018). The pathomechanism of the IOL calcification is still not fully understood.

3. Immunology of corneal transplantation and corneal (lymph) angiogenic privilege 3.1. Immune privilege and (lymph)angiogenic privilege of the cornea For the immune response after organ or tissue transplantation, Wayne Streilein introduced an analogy to the neural reflex arc, namely the “immune reflex arc” (Streilein, 1999). Herein the afferent arm is defined as an antigenic signal captured in the periphery and transmitted to secondary lymphoid tissues via lymphatic vessels. At the central processing mechanism the antigenic signal is transduced into effector modalities like B-lymphocyte derived antibodies and effector T lymphocytes. The efferent arm then represents the delivery of these effectors to the site of antigen (usually via the blood circulation), leading to its elimination (Cursiefen et al., 2003; Streilein, 1993, 1999, 2003). Inflammatory insults usually lead to an altered immune response at the cornea, which is due to loss of the so called immune privilege of the cornea. 6

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

This phenomenon of an altered immune response in the eye was first experimentally analyzed by Sir Peter Medawar in 1948 where he implanted skin grafts into the anterior chamber of rabbits. He observed that skin grafts adjacent to the corneal endothelium survived whereas skin grafts attached to the iris were rejected. He explained these findings with the additional observation that only the iris associated grafts were penetrated by blood vessels. Thus, Medawar concluded already at that time that vessels are crucial for a sufficient immune response against foreign tissue (Medawar, 1948). Later on, Medawar defined the principal of acquired immune tolerance against self- and foreign antigens (Billingham et al., 1953) for which he together with Frank Burnet received the Nobel Prize in Physiology or Medicine in 1960. Thereby, the importance of transplantation models to explore basic and translational questions in immunology was demonstrated. In contrast to non-immune privileged sites, the presentation of antigens to the anterior chamber (AC) provokes a series of immunological responses, which are characterized by the generation of non-complement antibodies (IgG1 isotope in mice), the formation of cytotoxic T lymphocyte (CTL) precursors and an absence of a normal delayed type hypersensitivity (DTH) response (Ferguson and Griffith, 1997; Niederkorn, 1990; Streilein, 1993; Streilein et al., 1992). For example, P815 mastocytoma cells (derived from DBA/2 mice) with a characteristic fast immunogenic effect after subcutaneous transplantation into allogenic BALB/c mice can escape an immunologic rejection in the AC (Niederkorn et al., 1981; Streilein et al., 1980). The fact that P815 tumor allografts were consequently rejected outside the eye shows that the growth of these highly immunogenic tumors in the eye is facilitated by the down regulation of a cell-mediated immunity. So, CD8+ CTL precursors invading in the intraocular tumor are not able to differentiate completely to gain their cytolytic function (Ksander et al., 1991; Ksander and Streilein, 1990). The immune privilege of the anterior part of the eye is achieved by several unique features of this site:

(DC) in an immature state are present (Hamrah et al., 2002). These mechanisms are responsible for the exceptionally high success of corneal transplantation when compared to other tissue and organ transplantations. Nevertheless, immune-mediated corneal allograft rejection can still occur, especially after PK. Although the early outcomes of corneal transplantation in avascular so-called low-risk recipients are typically very good (90% at 1 year after surgery) (Williams et al., 2006), the situation is very different in case of an inflamed and vascularized recipient cornea, e.g. due to chemical or thermal burns, herpes infections or history of previous graft rejection. In these cases, the one-year survival rate can decrease to below 50% (Bachmann et al., 2010; Williams et al., 2006), with immune-mediated allograft rejection being the most common cause for graft failure. In fact, there is a direct correlation between the number of corneal quadrants being vascularized and the risk of immune reactions after keratoplasty (Bachmann et al., 2010). For molecular details on corneal (lymph)angiogenic privilege see chapter 4.3.3. 3.2. Role of direct and indirect allo-sensitization in corneal transplant immunology Corneal allograft rejection is histologically characterized by infiltration of CD4+ T-cells, macrophages and the invasion of blood and lymphatic vessels. Significant numbers of neutrophils and natural killer cells (NK cells) were also found in the cellular infiltrates (Bradley, 2002; Forrester et al., 2010; Hegde et al., 2005; Kuffova et al., 2001; Torres et al., 1996). Furthermore, CD11c+ CD11b+ DCs are present throughout the anterior stroma. Inflammatory stimuli determine the maturation stage of DCs and their immunogenic properties. It is known that an increasing number of MHC class II-bearing, antigen specific DCs in cervical draining lymph nodes arise only a few hours after transplantation (Forrester et al., 2010). This suggests that DCs at different maturation stages might be the main initiators of allograft rejection (Forrester et al., 2010; Hamrah et al., 2002; Maruyama et al., 2005; Novak et al., 2003; Yamagami and Amano, 2003). The origin of these antigen presenting cells (APCs) can be from the donor tissue or the host. Thereby allo-antigens from the grafted tissues travel via lymphatic vessels towards the draining lymph nodes of the recipient. In this context, we could show previously that lymphatic rather than blood vessels determine the high-risk status of a corneal graft recipient (Dietrich et al., 2010). Allo-antigens in the lymph nodes are recognized by host T cells either by the direct pathway, which involves the recognition of intact donor MHC molecules on donor-derived APCs, or by the indirect pathway, which involves the recognition of donor major or minor MHC-derived peptides, processed and presented by host APCs to T cells (Fig. 5). In the low-risk setting the indirect pathway is predominant, whereas in the high-risk setting both the direct as well as the indirect pathway are involved (Huq et al., 2004; Sano et al., 2000). Nowadays it is widely accepted that resident MHC class II− APC populations are present in the cornea at constitutive state (Hamrah et al., 2003a, 2003b). These cells are capable of expressing MHC class II antigen and accessory molecules (CD40, CD80) after transplantation (Boisgerault et al., 2009; Huq et al., 2004; Liu et al., 2002). It has been demonstrated that allograft rejection in the high-risk situation can be triggered by the above described “direct allo-sensitization pathway” where APCs travel from the donor tissue via corneal pathological lymphatic vessels into the local draining lymph nodes of the recipient and present the foreign antigen. Using TSP-1 −/− mice we could show that allo-sensitization of the host against the graft can be also mediated by donor derived APCs. Using donor tissue from TSP-1−/ − mice resulted in a significantly increased rejection rate compared to the wild type control grafts. The graft derived TSP-1 negatively regulated the direct type of allo-sensitization, but not the indirect type (Saban et al., 2010). Taken together, both (direct and indirect) allo-sensitization

a) Absence of blood and lymphatic vessels in the central cornea (corneal [lymph]angiogenic privilege; reviewed in chapter 4.3.3.) (Cursiefen, 2007). b) Presence of secretory immunomodulatory factors. Soluble immunomodulatory factors in the AqH such as α-Melanocyte-stimulating hormone (α-MSH) and Vasoactive intestinal peptide (VIP) are secreted from the ciliary body. AqH has been shown to have an inhibitory effect on a variety of inflammatory cells (Streilein and Stein-Streilein, 2000; Taylor, 1999): CD4+ T cells in their proliferation and secretion of effector cytokines; polymorphonuclear neutrophilic granulocytes in their destructive potential; macrophages in the production of reactive oxygen intermediates and nitric oxide; natural killer cells in their capacity to lyse target cells. On the other hand, AqH does not impair the capacity of functional cytotoxic T cells to lyse their specific targets and neutralizing, noncomplement-fixing antibodies retain their properties when AqH is present. The immunosuppressive factors transforming growth factor-β (TGFβ), thrombospondin-1 and -2 (TSP-1/TSP-2) are secreted from the cornea, the iris pigment epithelium (IPE) and the retinal pigment epithelium (RPE) (Cursiefen et al., 2004c; Horie et al., 2009; Jeong et al., 2019; Tandon et al., 2010). It was shown in this context, that T cells are altered in their immunological function when passing the ocular pigment epithelium (Wenkel and Streilein, 2000; Yoshida et al., 2000a, 2000b) c) Corneal endothelial cells promote apoptosis of CD95 + T cells by the CD95 ligand (Osawa et al., 2004; Stuart et al., 1997). Additionally, corneal endothelial cells express only low levels of major histocompatibility complex (MHC) class I molecules thereby preventing the recognition of corneal endothelial cells by T cells. This prevents destruction of the corneal endothelium and protects the visual axis (Streilein and Stein-Streilein, 2000). Also, in the normal uninflamed cornea MHC class II-negative Langerhans cell-type dendritic cells 7

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

be. In the setting of corneal transplantation, however, the most important prognostic factor is not the HLA type matching (Bohringer et al., 2018) but the status of the recipient bed: Non-vascularized and uninflamed recipient beds have a high chance of accepting the corneal graft (Bachmann et al., 2010; Zheng et al., 2011). This high success rate is achievable without HLA matching or profound systemic immune suppression, being a major difference between corneal transplantations and solid tissue transplantations (Amouzegar et al., 2016; SanchezFueyo and Strom, 2011; Sheldon and Poulton, 2006). In contrast, grafts transplanted into prevascularized corneal recipients which show a breakdown of the cornea's angiogenic and immune privilege, are rejected in up to 50% despite of immunosuppressive therapy. In these high-risk eyes, HLA matching might be beneficial, although outcome is still unsatisfactory (van Essen et al., 2015). 3.4. Interaction of corneal cells and the immune system in corneal allograft rejection 3.4.1. Interaction of corneal vascular endothelial cells and the immune system Neovascularization is mainly driven by members of the vascular endothelial growth factor (VEGF) family, whose expression is induced by hypoxia, inflammatory stimuli, injury or underlying diseases (Ellis and Hicklin, 2008). The most important members of the VEGF family involved in corneal neovascularization are VEGF-A, -C and -D. VEGF-A induces both hem- and lymphangiogenesis by binding to VEGF receptor (VEGFR)1, present on blood endothelial cells (BECs) and VEGFR2, present on both BECs and lymphatic endothelial cells (LECs) (Amadio et al., 2016; Ellis and Hicklin, 2008). In addition, it was shown that VEGF-A can also indirectly mediate lymphangiogenesis by the recruitment of VEGFR1+ macrophages, which subsequently secrete VEGF-C and -D to induce LEC proliferation (Cursiefen et al., 2004b). VEGF-C and -D preferentially bind to VEGFR2 and VEGFR3, which are mainly present on LECs and specifically stimulate lymphangiogenesis (Ellis and Hicklin, 2008). Besides VEGFs, there are several other important players in the angiogenic cascade, and corneal neovascularization is mainly a result of an imbalance between many pro- and anti-angiogenic factors (Abdelfattah et al., 2016). For a detailed description of further known regulators of corneal hem- and lymphangiogenesis, we refer to chapter 4.3.3. Pathological ingrowth of blood and lymphatic vessels into the cornea causes collateral tissue alterations such as lipid deposition, stromal edema and invasion of immune cells thereby reducing vision and interrupting corneal immune privilege. This significantly decreases subsequent corneal transplantation success (Bachmann et al., 2009; Bock et al., 2013; Cursiefen, 2007; Cursiefen et al., 2004a; Hos et al., 2015; Regenfuss et al., 2012; Zheng et al., 2011). Inhibition of both hem- and lymphangiogenesis by trapping VEGF-A in experimental studies can significantly improve corneal graft survival (Cursiefen et al., 2004a). Furthermore, by selectively inhibiting the ingrowth of lymphatic but not blood vessels into the cornea by the blockade of VEGFR3 (Bock et al., 2008; Cursiefen et al., 2006) or integrin α5β1 (Dietrich et al., 2007), our group and others could demonstrate that especially lymphatic vessels play a crucial role in the mediation of corneal transplant allo-immunity in the high-risk setting (Chen et al., 2004; Dietrich et al., 2010) (Fig. 6). Recently, it could be demonstrated that the blockade of the prolymphangiogenic growth factor VEGF-C after corneal transplantation not only reduces corneal lymphangiogenesis, but also results in less APC maturation and migration to the draining lymph nodes in mice (Hajrasouliha et al., 2012). In addition, it is well known that lymphatic vessel increase the expression of the chemoattractive cytokine CCchemokine ligand (CCL) 19 and CCL21 under inflammatory conditions, which is sensed by mature APCs via CC-chemokine receptor 7 (CCR7) (Ohl et al., 2004). In this context, we could recently show that local blockade of the chemokine receptor CCR7 by administration of eye

Fig. 5. Direct and indirect allograft recognition after corneal transplantation. Direct recognition: The direct recognition of the corneal graft takes place via T cells carrying T cell receptors specific for allogeneic MHC class I or class II molecules (donor MHC) in combination with a peptide (graft antigen). Donor APCs (green) expressing the allogeneic MHC molecule (donor MHC, purple) and costimulatory molecules (CSM) activate the alloreactive T cells of the recipient. Indirect recognition: The indirect recognition takes place via T cells carrying receptors specific for allogeneic peptides of the transplant. Proteins from the graft (red dots) are processed by APCs of the recipient (Recipient APC; blue) and presented by recipient MHC class I or II molecules (recipient MHC,blue).

pathways can be employed by the host's immune system to induce allograft rejection, with the direct pathway being mostly responsible for the faster and more robust rejection reaction in the high-risk setting. The proinflammatory microenvironment in the high-risk setting increasing the maturation status of APCs (Huq et al., 2004; Salabarria et al., 2019) as well as the presence of a vascular network providing faster egress of APCs seem to support this pathway. 3.3. Similarities and differences between keratoplasty and other tissue/solid organ transplantations All allogenic tissue or solid organ transplantations introduce genetically different tissue to the recipient, which typically causes a T cell mediated immune response that leads to rejection and destruction of the transplant. The cornea represents a unique situation for tissue transplantation because of its (lymph)angiogenic and immune privilege (Cursiefen, 2007; Niederkorn, 2007; Sonoda and Streilein, 1992). However, injury and disease can override these privileges. This is then termed “high-risk corneal transplantation” and the underlying immunologic mechanisms are comparable with other tissue or solid organ transplantations. This implies that knowledge gained in vascularized high-risk eyes may also be applicable to other sites of tissue and organ transplantation. As described above, antigens of the graft can be presented by donor or recipient APCs to activate recipient T cells against the graft. After transplantation, the major molecules that initiate graft rejections are the MHC molecules. If a cell expresses another MHC type than the body's own (“self”) it will be identified as “non-self” and get rejected. Therefore, patients and donors are usually screened for their human leucocyte antigen (HLA) types prior to transplantation (Sanchez-Fueyo and Strom, 2011; Sheldon and Poulton, 2006). The better the similarity between the HLA types the better the transplant survival is predicted to 8

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Fig. 6. Absence of lymphatic vessels in the high-risk recipient cornea prior to transplantation significantly improves murine graft survival. a) Kaplan-Mayer survival curves in differentially vascularized recipient beds (BV: blood vessels; LV: lymphatic vessels); b-d) corneal whole mounts stained with CD31 (blood vessels; green) and LYVE-1 (lymphatic vessels, red) in different recipient beds. This study shows LV to be the prime mediator of immune responses after corneal transplantation. Modified from (Dietrich et al., 2010).

extensive morphological changes during passage into the vessel lumen (Stoitzner et al., 2002, 2003). Thus, independent of the route of entry, it is likely that intimate contacts are formed between leukocytes and inflamed lymphatic endothelium which likely regulates cell transit and immune status. Podgrabinska et al. demonstrated that tumor necrosis factor α (TNF-α)-stimulated lymphatic endothelium suppresses the maturation and function of DCs in vitro showing for the first time that LECs can actively influence the function of immune cells (Podgrabinska et al., 2009). This indicates that the interaction between lymphatic endothelium and APCs can be altered by different inflammatory, pro- or antiangiogenic stimuli, and that this APC-lymphatic interaction is key for allo-sensitization and graft rejection.

drops containing a CCR7 blocking fusion protein (CCL19 fused to the Fc part of human IgG1), leads to significantly reduced APC migration to the local lymph nodes of mice and significantly enhances corneal graft survival even in avascular corneas in the low-risk setting (Hos et al., 2016b). Furthermore, it was shown that CCL19 and CCL21 are able to further promote APC maturation, indicating that blockade of migratory signals to APCs might also alter the immunological status of these cells (Marsland et al., 2005). APCs transmigrate through the lymphatic endothelium similar to the process of rolling and sticking of leukocytes in blood vessels by interaction with Intercellular Adhesion Molecule-1 (ICAM- 1) and Vascular Cell Adhesion Molecule (VCAM) (Johnson et al., 2006). In an earlier study we visualized the transmigration of highly motile single cells into LYVE-1 labeled lymphatic vessels in the inflamed murine cornea using two-photon microscopy (Steven et al., 2011) (Fig. 7). Most recently it was shown that Dcs enter lymph vessels by hyaluronanmediated docking to the endothelial receptor LYVE-1 (Johnson et al., 2017). Furthermore, Activated leukocyte cell adhesion molecule (ALCAM) was recently shown to be another key player in DC transmigration into afferent corneal lymphatic vessels thus experimentally allowing the promotion of graft survival by targeting ALCAM (Willrodt et al., 2019). In other studies, leukocytes were seen to undergo

3.4.2. Interaction of corneal nerves and the immune system The cornea is richly supplied by sensory nerve fibers derived from the ophthalmic division of the trigeminal nerve, mainly via the long ciliary nerves. Branches radiate into the anterior corneal stroma from an annular plexus near the limbus, whereupon they lose their myelin sheaths and form a subepithelial plexus, from which fine axons, devoid of Schwann cells, pierce the Bowman's layer to form a intraepithelial subbasal nerve plexus. Nowadays it is known that the nervous and the immune system interact intimately. Recently, this was also Fig. 7. Transmigration of immune cells into the lymph vessel lumen at certain entry points a–e) Lymphatic vessel (LV), crossing blood vessel (BV, dotted line), corneal epithelium (blue) are imaged simultaneously by 2-photon in vivo microscopy. An individual cell (mC, red circle) migrates (yellow line) into the lymphatic vessel via a presumed gate (white star) that demonstrates an enhanced LYVE-1-antibody signal. From (Steven et al., 2011).

9

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

demonstrated for the cornea. In this regard, Paunicka et al. have shown in a mouse model of allogenic corneal transplantation, that whereas 50% of first-time corneal allografts enjoyed long-term survival, more than 90% of subsequent allografts placed in the opposite eye were rejected (Paunicka et al., 2015). Further analysis revealed that this was due to a loss of regulatory T cell (T reg) function caused by the ablation of corneal nerves and subsequent secretion of the neuropeptide substance P in both eyes. Thus, this study indicates that also the corneal immune privilege is possibly co-regulated by the nervous system. This also opens new potential future treatment avenues. 3.5. Imaging of clinically invisible corneal lymphatic vessels In contrast to pathologic corneal blood vessels, lymphatic vessels are clinically invisible. This has hampered their detection until specific histologic markers for lymphatic endothelium became available (Cursiefen et al., 2002). In vivo imaging of lymphatics is challenging due to their specific anatomic features, namely their thin and discontinuous vessel walls and the optical transparency of the lymphatic fluid. Two-photon microscopy was the first tool to image lymphatics and lymphatic–immune cell interactions experimentally in vivo (Steven et al., 2011) (Fig. 7). Imaging lymphatics in the clinical setting was not possible until very recently two novel approaches for easy in vivo imaging of lymphatics were described experimentally in mice: (i) Microscopic optical coherence tomography (mOCT) imaging with an axial and lateral resolution of 1 μm allows for reliable differentiation of corneal lymphatic versus blood vessels by contrasting morphology and flow features at a cellular level (Horstmann et al., 2017) (Fig. 8). This complex technology has recently been integrated into a slit lamp-based imaging system. Next, clinical trials are necessary to confirm the promising experimental results, e.g. by analyzing lymphatic vessels in pathologically prevascularized corneas prior to keratoplasty and subsequent validation by immunohistochemistry. (ii) Intrastromal fluorescein dye injection and subsequent imaging with angiography devices allow discrimination between pathologic blood and dye-uptaking lymphatics (Le et al., 2018a) (Fig. 8). Fig. 8. In vivo imaging of corneal lymphatic vessels. A+B) Microscopic optical coherence tomography (mOCT) vertical cross-sectional scan of a corneal lymphatic (A) and a corneal blood vessel (B). Lymphatic vessels appear as dark, mainly cell-free structures, whereas blood vessels are densely populated with fast-flowing cells. In contrast with lymphatic vessels, blood vessels exhibit hyperreflectivity in the endothelium and vessel wall. C+E) Visualization of lymphatic vessels with indirect fluorescein lymphangiography after intrastromal fluorescein injection. The fluorescein-labeled lymphatic vessels (arrows in E) are clearly detectable under the fluorescein angiography mode (C+E), but undetectable under the slit-lamp microscope (D+F). Blood vessels (arrowheads in E+F) are detectable in both fluorescein and color modes (C-F). From (Horstmann et al., 2017) and (Le et al., 2018).

These two novel techniques will be key for future clinical trials using e.g. lymphatic vessel regression prior to high-risk PK as endpoints. For further validation of these methods and a broader diagnostic spectrum, a multimodal system including 2-photon-microscopy is conceivable. Furthermore, these techniques might be helpful to answer fundamental, so far unknown questions in the context of lamellar keratoplasty, for instance whether lymphatic vessels actually occur in allograft rejection after lamellar keratoplasty. 4. Immune reactions after penetrating keratoplasty (PK) The cornea is among the few immune privileged sites of the body (previously described in chapter 3). Thus, corneal transplantation differs immunologically e.g. from renal, cardiac, or liver transplantation because it does not necessarily require matching of MHC-class antigens. In this context, it has previously been demonstrated in a rat model of allogenic corneal transplantation in which MHC-class antigens and minor histocompatibility (H) loci completely differed (donor: WistarFurth; recipient: Lewis), that rejection rates were about 50%. Moreover, in a model of MHC-class II mismatched transplantation (donor: DA.MR; recipient: DA.BDV), long-term allograft survival was > 90% (Ross et al., 1991a). From these standpoints, success rates are higher for corneal than for other types of tissue transplantation (Ross et al., 1991b). Furthermore, most corneal transplants require no systemic immunosuppressive agents or only for a very short period; topical immunosuppressive eye drops (mainly corticosteroids) are usually sufficient.

4.1. Low-risk penetrating keratoplasty (PK) When PK is performed to treat non-inflammatory and non-vascular diseases like keratoconus or Fuchs' endothelial dystrophy, the mechanisms of corneal immune and (lymph)angiogenic privilege are intact and lead to very satisfactory graft survival rates with low risk of graft rejection (“low-risk keratoplasty”). The incidence of a graft rejection episode after low-risk PK reported in the literature varies from 4% to 20% (Alldredge and Krachmer, 1981; Boisjoly et al., 1993; Inoue et al., 2001; Williams et al., 2015). We have also previously analyzed the incidence of endothelial graft rejection following low-risk PK. In 397 patients with a median follow-up of 18 months, endothelial graft rejection was observed in 22 cases, with most graft rejection episodes occurring between 11 and 18 months postoperatively. The percentage of grafts without any episode of endothelial graft rejection was 95% 10

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

corneal graft rejection. Blockade of T cell activation significantly reduced graft rejection rates even in vascularized high-risk corneas to similar levels seen in low-risk corneas (Amescua et al., 2008). Thus, vascularization of the recipient cornea impairs corneal immune privilege also via the recruitment of immune effector cells and secreted immune effector elements, leading to rapid rejection of grafted corneas in the absence of aggressive immunosuppressive therapy. Therefore, high-risk PK is usually performed after an uninflamed and quiescent interval, when blood and lymphatic vessels have partially regressed. In this context, we have previously shown in the experimental setting that regressed corneal blood and lymphatic vessels are reformed after grafting, and that transient postoperative VEGF-neutralization improves graft survival in these corneas (Bachmann et al., 2009). In addition, anti-VEGF treatment also suppresses inflammatory cell recruitment, e.g. of innate immune cells such as macrophages (Cursiefen et al., 2004a, 2004b). Therefore, anti-VEGF therapy represents an important option for reducing graft rejection rates in vascularized high-risk corneas. Corneal graft rejection was thought to be triggered mainly via Th1 immune responses such as interferon-γ (IFN-γ) secreting, activated T cells. However, a study by Hargrave et al. demonstrated that the rate of corneal transplant rejection in IFN-γ deficient mice and wild-type mice did not differ when MHC-mismatched corneal grafts were transplanted, indicating that MHC-mismatched corneal allografts can undergo rejection also in the absence of IFN-γ. By contrast, corneal allografts that were MHC-matched but mismatched at all minor-H loci were rejected in about 50% of the wild-type hosts, but were not rejected in any of the IFN-γ hosts, indicating that only immune responses against minor-H depend on Th1 immune responses (Hargrave et al., 2004). As already mentioned, atopic dermatitis is an established risk factor for graft rejection after PK even without clinically detectable ocular involvement (Nguyen et al., 2009; Reinhard et al., 1999). Allergic disease is also regarded as an important risk factor for immune reactions after PK (Ghoraishi et al., 1995). In the experimental setting, mouse models of allergic asthma or airway hyperactivity usually reject transplanted corneas at rates of 90–100% (Niederkorn et al., 2009). Both Th1 and Th2 cytokines are highly activated in allergic states (Randolph et al., 1999; Stern et al., 2005). In fact, only Th2 cell transfer could prolong graft survival in host mice compared with control mice, compared with transferred Th1 cells or Th1 combined with Th2 cells (Niederkorn et al., 2010). Therefore, even if the cornea of an allergic host is not vascularized, there is a high-risk of graft rejection, and persistent inflammation should be suppressed before undergoing transplantation. Infection with herpes simplex virus-1 (HSV-1) is also a high-risk phenotype. Corneal neovascularization, the most critical factor in allograft rejection, frequently occurs during infection with HSV-1. Furthermore, corneal HSV-1 infection results in persistent corneal inflammation related to both innate and adaptive immune responses (Kuffova et al., 2016). Therefore, corneal neovascularization and innate and adaptive immune responses must be controlled at the time of PK into an HSV-1 infected host. In this regard, it has previously been shown that recipients with history of but not active HSV-1 infection show significantly better graft survival than recipients with active HSV-1 infection (history of HSV-1: 84% survival at 3 years postoperatively vs. active HSV-1 infection: 64% survival at 3 years postoperatively) (Williams et al., 2015). Besides these major risk factors, further factors associated with increased risk of graft rejection are any other type of keratitis/inflammation, clinically manifest tear insufficiency/dry eye disease, young recipient age, and history of previous graft rejection (Maguire et al., 1994; Panda et al., 2007). Increased graft size or excentric grafts are also associated with increased risk of graft rejection, arguably due to the proximity to the limbal blood and lymphatic vasculature. Preoperative glaucoma, previous anterior segment surgery and, anterior iris synechiae also seem to increase the risk of graft rejection (Maguire

after 12 months, 89% after 18 months, and 86.5% after 24 months (Kuchle et al., 2002). Atopic dermatitis and clinically manifest tear insufficiency were two factors significantly associated with graft rejection in this study. After intensified topical and systemic corticosteroid treatment, only one graft decompensated. The remaining grafts regained clarity (Kuchle et al., 2002). In this context, we have also analyzed the impact of duration of topical corticosteroid therapy on the incidence of endothelial graft rejection after low-risk PK (Nguyen et al., 2007). In this study including 406 eyes, postoperative treatment started with prednisolone acetate 1% eye drops five times daily and was tapered over the first six months. Afterwards, patients were randomized into either short-term (stop topical steroid treatment) or long-term treatment (continue steroids once daily for 12 months). During followup, 29 eyes (7.1%) developed an endothelial graft rejection episode. Graft rejections were significantly more common in the short-term (19 of 202; 9.1%) compared with the long-term treatment group (10 of 204: 4.9%), indicating that long-term, low-dose topical corticosteroid treatment significantly reduces the risk of graft rejection (Nguyen et al., 2007). Similar to the study by Kuchle et al., most rejection episodes were observed between 11 and 25 months postoperatively (Kuchle et al., 2002; Nguyen et al., 2007). 4.2. High-risk penetrating keratoplasty (PK) As described, the host cornea in a low-risk PK setting has an immune privileged and avascular status, which contributes to low rejection rates after transplantation. In contrast, if the corneal immune and (lymph) angiogenic privilege is lost due to severe inflammation and vascularization and corneas are grafted into non-immune privileged, prevascularized recipients (“high-risk keratoplasty”), the risk of graft rejection significantly increases, even with HLA-matching and systemic immunosuppression (Cursiefen et al., 2003; Niederkorn, 2010) (Fig. 9). Corneal neovascularization seems to be the most important risk factor for graft rejection in high-risk corneas. Williams and colleagues have previously reported on the largest cohort followed after PK and have determined graft survival rates in non-vascularized vs. vascularized recipients (Williams et al., 2015). In this cohort, graft survival rates were 87% after 3 years and 62% after 9 years in recipients without corneal neovascularization, whereas graft survival was significantly reduced to 70% after 3 years and 36% after 9 years in recipients with corneal neovascularization. If the recipient cornea contains preexisting pathological lymphatic vessels before transplantation, APCs can easily migrate through these vessels to regional lymph nodes, leading to accelerated allograft sensitization. Furthermore, vascularized corneas produce chemokines that can activate T cells, which have been shown to be critical mediators of

Fig. 9. Graft rejection after (high-risk) penetrating keratoplasty. Endothelial immune reaction after penetrating keratoplasty presenting with corneal edema, endothelial precipitates and a classical “Khodadoust-line” (black arrows). Note the corneal blood vessel (black arrowhead). 11

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

The study situation regarding the outcome of systemic CsA therapy in the context of post keratoplasty management is controversial (Hill, 1994, 1995; Shimazaki et al., 2011; Ziaei et al., 2016). Individual studies raise doubts about the benefits of systemic CsA therapy (Poon et al., 2001; Shimazaki et al., 2011). Due to the numerous side effects, regular monitoring is crucial.

et al., 1994). Loose sutures have also been identified as risk factor of graft rejection, both in the low- and high-risk setting, arguably due to induction of inflammation and infection. Accumulated inflammatory cells then become triggers for antigen presentation due to upregulated local HLA and minor H antigens (Jonas et al., 2002). Taken together, the corneal immune and (lymph)angiogenic privilege are abrogated in the high-risk setting, generating an immunological scenario which is similar to transplantations in any other, non-immune privileged tissue or organs.

4.3.1.3. Mycophenolate Mofetil (MMF). The antimetabolite MMF has been used since the 1990s as an alternative to CsA for the treatment of high-risk PK. Additional applications in ophthalmology include inflammatory diseases such as uveitis, scleritis, and mucous membrane pemphigoid (Thorne et al., 2005). MMF is a prodrug of the active substance mycophenolic acid (MPA). MMF inhibits inosine monophosphate dehydrogenase and interferes with the synthesis of guanosine nucleotides. It has a selective cytostatic effect on T and B lymphocytes. In a prospective randomized clinical trial Reis et al. demonstrated that 2g MMF daily is as effective as CSA in preventing acute rejection following high-risk PK within the follow-up period of approximately 1 year (Reis et al., 1999). A retrospective study by Birnbaum et al. demonstrated a superior effect of MMF compared to CsA in preventing immune reactions after high-risk PK and comparable clear graft survival for both agents (Birnbaum et al., 2005). The positive effect of MMF on reduction of immune reactions and on graft survival after high-risk PK has been confirmed in further studies (Birnbaum et al., 2009; Reinhard et al., 2005; Szaflik et al., 2016). In particular, the fact that only few serious side effects occur under MMF treatment proves to be advantageous. The side effects of MMF include gastrointestinal disorders, liver dysfunction, and bone marrow suppression (Thorne et al., 2005). Compared to CsA, MMF therapy has only little effect on renal function, and features a broad therapeutic range with better dosing and less requirement for clinical monitoring.

4.3. Prevention of immune reactions after PK: current state of the art and future options 4.3.1. Immunomodulatory therapy in high-risk penetrating keratoplasty (PK) High-risk PK often requires immunomodulatory therapy over several months. Immunosuppression is generally assured by topical and systemic therapeutics. In addition to corticosteroids, which may be considered as gold-standard in ocular immunomodulatory therapy, routinely applied substances particularly include Cyclosporine A (CsA) and Mycophenolate Mofetil (MMF). Other agents, including tacrolimus and rapamycin, were not yet able to fully assert themselves in clinical practice which may be mostly explained by an increased range of side effects. Importantly, a fully defined treatment strategy for pharmacological immunomodulatory therapy in high-risk PK has not yet been established. Novel agents are still under development or clinical investigation (Abud et al., 2017). 4.3.1.1. Corticosteroids. Corticosteroids are the mainstay in ocular immunomodulatory therapy. The effect of corticosteroids is mediated by the intracellular glucocorticoid receptor, which is a member of the nuclear receptor family of ligand-activated transcription factors acting through genomic and non-genomic pathways (Ramamoorthy and Cidlowski, 2016). The various forms of corticosteroids may be applied via the topical (intrabulbar, subconjunctival injection, eyedrops/ointment) and systemic (oral or intravenous) route. Corticosteroids are part of the standard high-risk PK therapy and are generally applied short-term systemically and longer-term topically. The systemic dosage initially amounts to about 1 mg per kg body weight and is then successively tapered down. Systemic side effects include diabetes, impaired wound healing, skin atrophy, muscle atrophy, peptic ulcers, hypertension, metabolic syndrome, osteoporosis, and water/electrolyte imbalance (Ramamoorthy and Cidlowski, 2016). Ocular side effects include cataract induction, wound healing disorders, and secondary glaucoma. The multiple side effects limit long-term application of systemic steroids and their topical application is generally considered insufficient as sole long-term immunotherapy in high-risk PK.

4.3.1.4. Rapamycin (Sirolimus). Rapamycin is also a macrolide compound such as CsA but in contrast to CsA no calcineurin inhibitor. It binds to the cytosolic protein FK-binding protein 12. This complex inhibits the mammalian target of rapamycin (mTOR) pathway and inhibits T cell and B cell activation and secretion of IL-2 (Mukherjee and Mukherjee, 2009). The number of studies on the use of Rapamycin in conjunction with high-risk PK is still very limited: Birnbaum and colleagues compared Rapamycin (10 patients) and MMF (24 patients) after high-risk PK in a pilot study. The authors found similar efficacy of both agents in preventing immune reactions, especially during the first 6 postoperative months (Birnbaum et al., 2006). However, they also showed side effects in all of the rapamycin-treated patients, in particular hypercholesterolemia, furunculosis, exanthema, hypertriglyceridemia, elevation of Lactatdehydrogenase (LDH), as well as gastrointestinal side effects (Birnbaum et al., 2006). Chatel and Larkin investigated the dual use of Rapamycin and MMF (Sirolimus + MMF for 1 year, then Rapamycin alone for 2 years) in six patients with high-risk PK in a prospective interventional case series (Chatel and Larkin, 2010). Over a follow-up interval of at least 13 months, 5 out of 6 grafts remained clear. There were no further significant adverse effects except for MMF-related hepatotoxicity in one case. Nevertheless, Rapamycin induced hypercholesterolemia in all patients. In addition, furunculosis and possibly immunosuppression related measles infection were found as side effects (Chatel and Larkin, 2010). Rapamycin itself exerts no nephrotoxicity.

4.3.1.2. Cyclosporine A (CsA). CsA is a macrolide drug and has been applied in the context of corneal transplantation since the 1980s (Hill, 1989; Hoffmann and Wiederholt, 1986). It belongs to the substance group of calcineurin inhibitors. CsA binds to cytoplasmic cyclophilin A or D thereby inhibiting calcineurin, which results in a downregulation of the interleukin (IL)-2 pathway and helper and cytotoxic T cell differentiation and activation (Ziaei et al., 2016). Topical and systemic forms of application are popular in the treatment of a variety of ocular disorders. As with corticosteroids, there is no consensus on the therapeutic effectiveness and thus its recommended application (Abud et al., 2017). Blood levels of about 200 μg/L are required for the effective treatment of graft rejection. Due to the numerous side effects, however, blood levels of 130–170 μg/L are often preferred (Ziaei et al., 2016). The most common side effects include nephrotoxicity, hypertension, gingival hyperplasia, hepatotoxicity, gastrointestinal toxicity, increased susceptibility to infections, lymphoproliferative disorders, anemia, hyperlipidemia, and hirsutism (Ziaei et al., 2016).

4.3.1.5. Tacrolimus (FK506). Tacrolimus (FK506) is also a macrolide drug and belongs to the substance group of calcineurin inhibitors. In addition to high-risk PK other applications in ophthalmology include inflammatory conjunctival and corneal disorders, uveitis and graftversus-host disease (Zhai et al., 2011). Its mechanism of action is similar to that of CsA and prevents T cell activation. Studies in the areas of liver, kidney, and pulmonary transplantation attest that tacrolimus has a higher immunosuppressive effect and a lower spectrum of side 12

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

(Dobrowolski et al., 2015; Eslani et al., 2012; Soma et al., 2014). However, in most cases, stromal clarity can only be restored by keratoplasty. Clinical observations demonstrate that corneal transplantation, without prior restoration of limbal stem cell function, leads to increased graft rejection in patients with severe or total LSCD (Maguire et al., 1994). The only alternative in this case is keratoprosthesis surgery, e.g. Boston KPro (Schaub et al., 2018). It is perceived that the stem cells play a key role in stopping the progression of blood and lymphatic vessels at the limbus thus keeping the cornea avascular. This limbal barrier prevents the development of inflammatory events including the migration of immune cells such as macrophages and T cells through the vessels, which play a key role in further progression of neovascularization as well as graft rejection (Notara et al., 2018). While the exact mechanisms by which the limbal barrier is maintained is not clear, we have recently demonstrated that loss of the limbal stem cell character following UV damage is associated with the upregulation of proinflammatory factors in limbal epithelial cells thus contributing to immune cell infiltration and neovascularization (Notara et al., 2015, 2016). Although there are not many relevant studies, recent reports document the improved survival of corneal grafts following limbal tissue or stem cell transplantation. Recently, Figueiredo et al. investigated patients with unilateral total LSCD which were treated with cultured LSC transplantation followed by PK after approximately 19 months (Figueiredo et al., 2018). Visual acuity was significantly increased following keratoplasty, while mean graft survival at 12 months was 91% and was majorly improved compared to previous reports for high-risk keratoplasty (Abud et al., 2017; Chow et al., 2015; Szaflik et al., 2016), while it was 72% at 32 months, which was lower than reported previously (Figueiredo et al., 2015a). Basu et al. on the other hand, reported an 80% graft survival at 12 months in younger patients suffering alkali burns who previously received LSC transplantation (Basu et al., 2011). Specifically, corneal survival rates were majorly improved in patients who received 2-step limbal and corneal transplantation (80 ± 6%; median survival, at 4 years follow-up) compared to single-step limbal and corneal transplantation (25 ± 13% median survival at 6 months) (Basu et al., 2011), while Gupta et al., reported similar early success rates (85% at 6 months) of keratoplasty followed by simple limbal epithelial transplantation (SLET) (Gupta et al., 2018). Currently there is no established consensus regarding the timing of these two-stage treatments and whether they could be also combined into a single-step procedure; based on data from different studies a twostage approach with a gap between the two steps ranging from 6 weeks to 19 months has been reported (Baradaran-Rafii et al., 2010; Basu et al., 2011; Cernak et al., 2002; Figueiredo et al., 2018; Gupta et al., 2018; Sangwan et al., 2005). Further studies are necessary to elucidate the mechanisms by which good LSC function enables corneal graft success and to establish the guidelines for the timing between the LSC transplantation and keratoplasty. Furthermore, future studies will have to reveal the specific mechanisms whereby LSCs inhibit hem- and lymphangiogenesis, since these factors could also be used therapeutically.

effects compared to CsA (Zhai et al., 2011). Joseph et al. used systemic tacrolimus in high-risk PK at a daily dose of 2.5 mg over a follow-up period of about 34 months and observed clear grafts in 65% (Joseph et al., 2007). In a non-comparative case series, Sloper et al. treated patients after high-risk PK with systemic tacrolimus at a mean dosage of 4.4 mg daily. Under this treatment regime, none of the seventeen patients (23 grafts) had irreversible graft rejection over a follow-up period of 24 months (Sloper et al., 2001). A study by Yamazoe et al. used systemic tacrolimus in 10 eyes of 10 patients with graft failure despite CsA treatment and found significantly fewer rejection episodes and longer graft survival under tacrolimus compared to CsA (Yamazoe et al., 2014). A retrospective case series by Dhaliwal et al. examined topical tacrolimus 0.03% in high-risk PK as a second-line treatment in steroid. During a follow-up of 33 months, none of the four patients showed a graft rejection episode during (Dhaliwal et al., 2008). 4.3.1.6. Novel immunomodulatory agents. A very wide range of further immunomodulatory agents is currently under development and investigation. In addition to further members of the above mentioned substance groups, these include antibody-related therapeutics, small molecule inhibitors, receptor antagonists, interferons, RNA interfering compounds including morpholino oligonucleotides, and others (Abud et al., 2017). 4.3.2. Role of limbal stem cells and their repair for success of high-risk grafting In severe conditions affecting the ocular surface including injuries (such as chemical or thermal burns), genetic conditions (including aniridia) or following extreme immune reactions (Stevens Johnson Syndrome), the limbal stem cells (LSCs) can be destroyed causing limbal stem cell deficiency (LSCD). The LSCs are situated in the vascularized junction between the cornea and the conjunctiva, namely the limbus (Cotsarelis et al., 1989; Notara et al., 2010, 2018; Notara and Daniels, 2008). Their main role is to continuously replenish the corneal epithelium as they migrate and proliferate to accommodate for injuries or normal ware and tare of the tissue, thus sustaining cornea clarity. LSCD is classified as mild (or partial), where the damage extends in only part of the limbus and in severe (or total), where more than two quadrants of the limbus are involved and the central cornea is also affected (Nubile et al., 2013). In severe LSCD other symptoms are also present including persistent inflammation and epithelial breakdown, corneal (lymph)angiogenesis and opacification; if the visual axis is affected there is considerable visual loss and ultimately blindness (OseiBempong et al., 2013; Pellegrini and De Luca, 2014; Rama et al., 2010). These patients benefit from LSC transplantation during which (e.g. ex vivo expanded) LSCs are grafted on the diseased corneas. In January 2015, Holoclar (ex vivo expanded autologous human corneal epithelial cells containing stem cells), was the first EU-approved limbal stem cell treatment to be granted marketing authorization (Pellegrini et al., 2018). In severe LSCD cases, for example in severe alkali burns, LSC alone cannot restore corneal clarity as the stroma may be scarred and the endothelium is also affected as endothelial to mesenchymal transdifferentiation may occur (EnMT). Recent reports showed potential benefits of Smad7 overexpression in animal models by preventing EnMT following an alkali burn (Miyamoto et al., 2010; Sumioka et al., 2008). The prevention of stromal scarring, however, still remains an unresolved issue. In these severe cases, it is often necessary to perform a two-step treatment using LSC transplantation followed by (penetrating) keratoplasty in order to restore visual acuity (Baradaran-Rafii et al., 2010; Basu et al., 2011; Figueiredo et al., 2015a, 2018; Gupta et al., 2018). Notably, ex vivo expanded LSC or limbal tissue transplantation alone may only partially improve vision in these patients. This is routinely carried out by either transplanting ex vivo cultivated autologous or allogeneic limbal epithelial cells (Rama et al., 2010; Shortt et al., 2014) or autologous cells from other sources such as the oral mucosa

4.3.3. Endogenous regulators of corneal hem- and lymphangiogenesis and their role in corneal transplantation In order to maintain corneal avascularity, the precise balance of proand anti-(lymph)angiogenic factors plays a very important role. Hemand lymphangiogenesis are mainly induced by members of the VEGF family (reviewed in chapter 3.4.1). In addition to VEGFs, also fibroblast growth factor-2 (FGF-2), placental growth factor (PIGF), hepatocyte growth factor (HGF), platelet-derived growth factor (PDGF), Neuropilins, and angiopoetin have been shown to have hem- and lymphangiogenic properties (Achen et al., 1998; Bussolino et al., 1996; Heldin and Westermark, 1999; Jitariu et al., 2015; Kajiya et al., 2005; Oh et al., 1997; Otrock et al., 2007; Pepper et al., 1994; Tammela et al., 2005; Veikkola and Alitalo, 1999; Yuen et al., 2014). Cytokines 13

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

et al., 2014). Direct blocking of VEGF-C with anti-VEGF-C (VGX-100) in a murine model of high-risk transplantation was also effective in reducing graft lymphangiogenesis and this treatment also significantly improved survival compared to controls (Dohlman et al., 2015). In addition, it has been shown that the administration of anti-VEGF-C not only suppressed corneal lymphangiogenesis but also inhibited trafficking and maturation of APCs (Hajrasouliha et al., 2012). Integrins, a family of heterodimeric transmembrane glycoproteins mediating cell–cell and cell–ECM connections, have also been shown to regulate hem- and lymphangiogenesis (Garmy-Susini and Varner, 2008). Integrin α5β1 is expressed in lymphatic vessels of the inflamed cornea and low molecular weight antagonists of Integrin α5β1 reduce inflammatory corneal lymphangiogenesis in mice whereas hemangiogenesis is not significantly affected (Dietrich et al., 2007). Further studies have identified integrin α9 as regulator of lymphatic valve formation in the cornea (Kang et al., 2016). in this context, Kang et al. have demonstrated that treatment with anti-integrin α9 blocking antibody suppressed lymphatic valvulogenesis after corneal transplantation and resulted in a higher rate of corneal graft survival (Kang et al., 2016). Semaphorins are known guidance factors for axons and also play an important role in the proper orientation of blood vessels (Fukushima et al., 2011). Within the Semaphorin family, Semaphorin 3f (Sema3F), which belongs to the class-3 semaphorins, was shown to contribute to the anti-angiogenic barriers of the eye (Buehler et al., 2013; Sun et al., 2017). We could recently show that Sema3F is also highly expressed in the murine and human corneal epithelium and that its expression is downregulated under inflamed and vascularized conditions. Corneal epithelium-derived Sema3F suppressed lymphatic endothelial sprouting, and exogenous application of recombinant Sema3F selectively inhibited the growth of corneal lymphatic vessels after injury in mice, whereas the growth of blood vessels was not affected. Furthermore, Sema3F treatment also increased corneal graft survival after high-risk corneal transplantation in mice. Thus, Sema3F might contribute to the lymphangiogenic privilege of the cornea and could be used therapeutically to improve graft survival after high-risk corneal transplantation (Reuer et al., 2018). Podoplanin is a membrane-bound glycoprotein, which is strongly expressed on lymphatic endothelial cells, but not on blood endothelial cells (Baluk and McDonald, 2008). Its role in corneal transplantation has recently been investigated. Administration of an anti-podoplanin antibody inhibited lymphatic vessel growth, led to a dramatically reduced infiltration of macrophages and significantly suppressed graft rejection in the murine corneal transplantation model (Maruyama et al., 2014). Taken together, there is substantial (preclinical) evidence that inhibition of corneal neovascularization improves subsequent corneal graft survival. Nevertheless, it is important to realize that the role of corneal neovascularization after keratoplasty has been mainly studied in the context of PK and not of lamellar keratoplasty. To our knowledge, there are no published reports on the role of corneal neovascularization in corneal graft rejection after lamellar keratoplasty so far.

including IL-1 and TNF-α, IL-4, IL-6 and IL-10 also regulate hem- and lymphangiogenesis (Alitalo et al., 2005; Hos et al., 2016a; Jiang et al., 2018; Peppicelli et al., 2014; Savetsky et al., 2015). Besides these activators of hem- and lymphangiogenesis, also several endogenous inhibitors of hem- and lymphangiogenesis are expressed in the cornea to prevent the pathological ingrowth of blood and/or lymphatic vessels by minor vascular stimuli. These inhibitors include soluble VEGFR1 (sVEGFR1), which serves as decoy receptor for VEGF-A, pigment epithelium-derived factor, angiostatin, endostatin, TSP-1, and TSP-2 (Ambati et al., 2006; Cursiefen, 2007; Ribatti, 2009). The corneal epithelium also expresses soluble VEGFR2 (sVEGFR2) and soluble VEGFR3 (sVEGFR3), which trap VEGF-C and -D and specifically inhibit lymphangiogenesis (Albuquerque et al., 2009; Singh et al., 2013). It has been shown that the binding of VEGF-C to sVEGFR3 not only inhibits lymphangiogenesis, but also causes a regression of already formed lymphatic vessels (Makinen et al., 2001; Singh et al., 2013). In addition to these sVEGFRs, Cursiefen et al. demonstrated that corneal epithelial cells express membrane-bound VEGFR3, which is able to bind VEGF-C and thereby maintain corneal avascularity (Cursiefen et al., 2006). Many experimental studies have demonstrated that modulating hem- and lymphangiogenesis after corneal transplantation can increase transplant survival: in the mouse model of low- and high-risk corneal transplantation, our group and others have previously shown that the use of VEGF Trap(R1R2) to neutralize VEGF-A inhibits corneal hem- and lymphangiogenesis and promotes corneal allograft survival (Bachmann et al., 2008a; Cursiefen et al., 2004a). VEGF-trap(R1R2) also significantly inhibited immune cell infiltration, including macrophages and CD3+ T cells (Dohlman et al., 2015). By using a neutralizing antibody against VEGF-A (Bevacizumab) in a high-risk corneal transplantation model, Dastjerdi et al. identified that subconjunctival application of Bevacizumab significantly reduced corneal neovascular area and prolonged graft survival (Dastjerdi et al., 2010). Recently, this approach has been translated in the clinic, and studies have already demonstrated promising results for the treatment of corneal neovascularization in PK patients with Bevacizumab. In particular, a significant reduction of corneal neovascularization in patients topically treated with Bevacizumab after corneal transplantation has been reported (Bhatti et al., 2013b; Dekaris et al., 2015; Koenig et al., 2009). In addition, Bhatti et al. showed that subconjunctival injection of Bevacizumab was more effective than the topical application in reducing corneal neovascularization (Bhatti et al., 2013a). Fasciani et al. showed the safety and efficacy of preoperative treatment of corneal neovascularization with Bevacizumab (Fasciani et al., 2015) and Dekaris et al. identified that 70% of high-risk grafts remained clear during three years of follow-up after a combined subconjunctival and topical Bevacizumab treatment (Dekaris et al., 2015). In order to investigate the specific role of lymphatic vessels in corneal graft rejection, we have previously generated different experimental conditions in recipient mice: 1. low-risk (healthy avascular recipient bed), 2. high-risk (recipient bed vascularized with blood and lymph vessels), and 3. alymphatic high-risk (recipient bed vascularized only with blood vessels; the outgrowth of lymphatic vessels was pharmacologically inhibited). Graft survival after transplantation was similar between low-risk and alymphatic high-risk recipients, indicating that existing lymphatic vessels in the recipient have a much greater impact on graft rejection than blood vessels (Dietrich et al., 2010) (Fig. 6). Albuquerque et al. also observed in a corneal transplantation model, that administration of sVEGFR2 reduced corneal lymphangiogenesis and enhanced corneal allograft survival (Albuquerque et al., 2009). Emami-Naeini et al. investigated the effect of VEGF-C and VEGFD blockade via sVEGFR3 on corneal neovascularization and transplant survival. They demonstrated that treatment with sVEGFR3 resulted in a significant blockade of lymphangiogenesis and significantly better corneal allograft survival (Emami-Naeini et al., 2014). In addition, treatment with sVEGFR3 significantly reduced the frequencies of allosensitized T cells and IFN-γ–producing CD4+ T cells (Emami-Naeini

4.3.3.1. Recently identified novel endogenous regulators of lymphangiogenesis. Our laboratory and others have recently identified several novel regulators of endogenous lymphangiogenesis. These discoveries have substantially contributed to the better understanding of fundamental principles of lymphangiogenesis, and the identified targets might be used clinically to prevent corneal graft rejection in the future. In addition to sVEGFR2 (Albuquerque et al., 2009), sVEGFR3 (Makinen et al., 2001; Singh et al., 2013), and membrane-bound VEGFR3 (Cursiefen et al., 2006), TSP-1 (Cursiefen et al., 2011) and vasohibin-1 (Heishi et al., 2010) were also identified as endogenous inhibitors of corneal lymphangiogenesis. TSP-1 belongs to a family of high molecular weight glycoproteins that are secreted by various cell 14

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

lymphangiogenesis in a corneal alkali-burn rat model (Oh et al., 2018). Recently, Grimaldo et al. identified that the expression of miR-184 is reduced during corneal lymphangiogenesis. In addition, overexpression of miR-184 in LECs suppressed their adhesion and migratory capacity and reduced tube formation, indicating that also miR-184 is a negative regulator of lymphangiogenesis (Grimaldo et al., 2015). Taken together, these findings indicate that miRNAs may be promising therapeutic targets to control corneal lymphangiogenesis in the setting of corneal transplantation in the future.

types. We have previously shown that TSP-1 inhibits corneal hemangiogenesis in vivo (Cursiefen et al., 2004c). More recently, our group provided evidence that TSP-1 also inhibits lymphangiogenesis by binding to CD36 expressed on the surface of macrophages. This binding leads to an inhibition of VEGF-C production in macrophages, which in turn leads to reduced corneal lymphangiogenesis (Cursiefen et al., 2011). Vasohibin-1 was originally identified as an endothelium-derived negative feedback regulator of hemangiogenesis (Watanabe et al., 2004). Moreover, it has been shown that Vasohibin-1 also inhibits lymphangiogenesis in tumors and in the cornea (Heishi et al., 2010). Other newly identified regulators of lymphangiogenesis include matrix metalloproteinases (MMPs). MMPs belong to a family of zincdependent enzymes, which play key roles in degrading the extracellular matrix; their action associated with metastasis, angiogenesis and tumor invasion (Egeblad and Werb, 2002; McCawley and Matrisian, 2000). Membrane type-1-matrix metalloproteinase (MT1-MMP) is a membrane-bound enzyme which is essential for diverse physiological processes like extracellular matrix remodeling and pericellular proteolysis (Barbolina and Stack, 2008). MT1-MMP-deficient mice exhibit defective FGF2-induced corneal hemangiogenesis (Wong et al., 2012; Zhou et al., 2000). Wong et al. recently demonstrated that MT1-MMP is also an endogenous inhibitor of corneal lymphangiogenesis (Wong et al., 2016). MT1-MMP directly cleaves LYVE-1 on lymphatic endothelial cells and thereby inhibits LYVE-1-mediated lymphangiogenic responses (Wong et al., 2016). Recently, Du et al. identified that MMP-2 and MMP-9 are also critically involved in inflammatory corneal lymphangiogenesis (Du et al., 2017). The AqH not only supplies the lens and the cornea with nutrients but also facilitates the removal of potentially harmful agents and contains also several immunomodulatory factors contributing to the immune privilege of the cornea (see chapter 3). Recently it has been shown that the AqH contributes to the (lymph)angiogenic privilege of the cornea and that the α-MSH and VIP partially mediate this effect (Bock et al., 2016). AqH also contains high amounts of TGF-β. It has previously been shown that inhibition of endogenous TGF-β signaling enhances lymphangiogenesis in a mouse model of chronic peritonitis (Oka et al., 2008). Thus, it is likely that TGF-β is also an important regulator of the (lymph)angiogenic privilege of the cornea. MicroRNAs (miRNAs), which are non-coding RNAs of approximately 21–24 nucleotides, posttranscriptionally regulate gene expression by inhibition of translation or mRNA degradation (Bartel, 2004; Filipowicz et al., 2008). Recently, it has been shown that miRNAs are involved in the regulation of the transcription factor Prospero Homeobox-1 (Prox-1), which is the first identified gene whose activity is specifically important for lymphatic vessel development and maintenance (Hong and Detmar, 2003; Hong et al., 2002; Wigle et al., 2002). Prox-1 is mainly expressed in LECs and has no influence on the development and function of BECs (Wigle et al., 2002). Despite the importance of Prox-1 for lymphangiogenesis, very little is known about the mechanisms by which Prox-1 expression is controlled. Kazenwadel et al. demonstrated that miR-181a binds the 3′ untranslated region (UTR) of Prox-1, resulting in rapid and efficient transcript degradation and translation inhibition (Kazenwadel et al., 2010). miR-181a expression was significantly higher in embryonic BECs compared to LECs, identifying an important mechanism to restrict Prox-1 expression to the lymphatic vasculature (Kazenwadel et al., 2010). Recently, Seo et al. identified another miRNA (miR-466) which binds to the 3’ UTR of Prox1 and suppresses Prox-1 mRNA and protein expression (Seo et al., 2015). MiRNAs have also been demonstrated to (indirectly) regulate VEGFR-3. Jones et al. identified that the inflammatory cytokine IL-1β leads to induction of miR-1236 in LECs which leads to a translational inhibition of VEGFR-3 (Jones et al., 2012). Recently, Oh et al. identified that miR-199a/b-5p in lymphatic endothelial cells reduces DDR1 expression, a known positive regulator of lymphangiogenesis (Oh et al., 2018; Xiao et al., 2015). As expected, treatment with a miR-199a/b-5p mimic not only suppressed DDR1 expression, but also inhibited

4.3.3.2. The cornea as a model system to identify novel endogenous regulators of lymphangiogenesis. By comparing different inbred and wild-derived mouse strains, our group has shown that the number of physiological lymphatic vascular extensions into the peripheral cornea in naive eyes showed significant differences between the different inbred and wild-type strains (Regenfuss et al., 2010). We have also demonstrated significant differences in the lymphangiogenic response after an inflammatory stimulus between several mouse strains, suggesting that underlying genetic factors might influence the lymphangiogenic response (Regenfuss et al., 2010). By performing pathway-specific expression analyses of the cornea of lowlymphangiogenic BALB/c with high-lymphangiogenic C57BL/6 mice, two novel endogenous regulators of lymphangiogenesis could already be identified, namely tumor necrosis factor (ligand) superfamily member 10 (Tnfsf10/Trail) and the tissue-specific plasminogen activator (Plat/tPA). Both factors are expressed in the naive cornea and we have shown that they contribute to the (lymph)angiogenic privilege of the cornea (Regenfuss et al., 2015). More recently, we have also identified the tyrosinase gene, which is partly responsible for the differences in the limbal lymph vessel architecture of BALB/c and C57BL/6 mice (Buttner et al., 2018). Furthermore, by comparing C57BL/6 mice with albino C57BL/6 (B6N-TyrcBrd) mice which harbor a spontaneous mutation the gene encoding for Tyrosinase, we demonstrated that Tyrosinase not only is a novel endogenous regulator of developmental lymphangiogenesis, but also of inflammatory lymphangiogenesis (Buttner et al., 2018). These results demonstrate that the cornea is a suitable model to identify novel regulators of lymphangiogenesis. 4.3.4. Current and future options for inhibition of pathological corneal blood and lymphatic vessels to promote graft survival As outlined above, corneal neovascularization disrupts the immune privilege of the cornea and is recognized as a risk factor for corneal graft rejection after PK (Bachmann et al., 2010). Various approaches have already been reported to reduce progressive corneal neovascularization: 4.3.4.1. Corticosteroids. Corticosteroid therapy is the standard antiinflammatory treatment at the cornea. Corticosteroids are also used for antiangiogenic treatment at the cornea, as it has been shown that these drugs are potent inhibitors of corneal hemangiogenesis and, as we have experimentally demonstrated, also corneal lymphangiogenesis (Hos et al., 2011b). The anti(lymph)angiogenic effect of corticosteroids is mainly indirect due to the strong reduction of macrophage infiltration into the inflamed cornea. Furthermore, corticosteroids also inhibit the expression of proinflammatory cytokines and also show direct inhibition of vascular endothelial cell proliferation (Hos et al., 2011b). The anti(lymph)angiogenic effect of corticosteroids varies with prednisolone and dexamethasone having the strongest effects. Weaker corticosteroids including fluorometholone show less inhibition of inflammatory corneal hem- and lymphangiogenesis. However, the (prolonged) use of corticosteroids may cause several severe side effects such as delayed corneal wound healing, cataract or elevated intraocular pressure. Furthermore, although corticosteroids inhibit the progression of newly developing corneal blood and lymphatic vessels very potently, they are less 15

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Fig. 10. Corneal collagen crosslinking (CXL) regresses mature corneal blood and lymphatic vessels and improves graft survival after subsequent murine high-risk keratoplasty. A-C: Corneal vessels were double stained in corneal whole mounts 4 days after CXL treatment (merged images). D-F: Blood vessels (BV: with white arrow) were stained with CD31 (green). G-I: Lymphatic vessels (LV: with white arrow head) were stained with LYVE-1 (red). J-K: Quantification of corneal vessels on day 4 and day 8 post 6 or 9 min CXL treatment (n = 5; *p < 0.05, **p < 0.01). L: Corneal graft survival was significantly improved in high-risk eyes preoperatively treated with CXL (p < 0.05). Modified from (Hou et al., 2018).

neovascularization in the clinical setting (Ferrari et al., 2013). Bevacizumab and Ranibizumab showed comparable antihemangiogenic effects and both were recently reported to also inhibit corneal lymphangiogenesis, at least in the experimental setting (Bock et al., 2007; Bucher et al., 2012). However, as (blood) vessel maturation and pericyte coverage is usually accompanied by reduced dependence on VEGFs, treatment strategies targeting the VEGF pathway usually fail to regress already maturated corneal neovessels.

effective in regressing already present, mature vessels. 4.3.4.2. Anti-VEGF agents. Several anti(lymph)angiogenic treatment strategies specifically targeting the VEGF pathway at the cornea have been tested preclinically in the past, including Aflibercept, Bevacizumab, Ranibizumab, Pegaptanib, VEGF receptor tyrosine kinase inhibitors, and others (Bock et al., 2007, 2008; Bucher et al., 2012; Cursiefen et al., 2004a; Detry et al., 2013; Hos et al., 2008; Lipp et al., 2014). Despite this enormous progress in the experimental setting, none of these inhibitors has an FDA approval for the use at the cornea so far. Nevertheless, Bevacizumab is frequently used off label to treat patients with corneal neovascular diseases also in the context of keratoplasty, and several groups have already proven its effectiveness and safety (Koenig et al., 2012). In addition, recent studies have also shown the effectiveness of Ranibizumab in inhibiting corneal

4.3.4.3. Inhibition of insulin receptor substrate-1 (IRS-1). Insulin receptor substrate-1 (IRS-1) is a cytosolic scaffolding protein also known to interact with the VEGF-receptor complex. It was recently demonstrated that IRS-1 is also expressed in the cornea and plays a role in blood vessel development (Andrieu-Soler et al., 2005). It has been demonstrated that the blockade of IRS-1 signalling by GS-101 16

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

neovascularization and the reduction of inflammatory cells in CXL treated corneas. In addition, the population of T regs in draining cervical lymph nodes was significantly increased in CXL treated corneas after keratoplasty, which might have also contributed to the improved graft survival. In summary, CXL using riboflavin and UV-A is an effective presurgical (lymph)angioregressive treatment strategy to improve graft survival in high-risk corneal transplantation (Hou et al., 2018). Furthermore, these studies show the feasibility of the concept of (lymph) angioregressive pretreatment prior to high-risk keratoplasty to promote corneal graft survival. In addition, this approach of a temporary and selective regression of resident lymphatic vessels at the site of transplantation could also be used in other transplantation fields. Nevertheless, it should be noted that the beneficial impact of pre-surgical CXL on high-risk corneal transplant survival has so far only shown in experimental studies. Clinical studies are missing so far, and have to prove the efficacy of this promising approach.

(Aganirsen), an antisense oligonucleotide against IRS-1, inhibits inflammatory corneal hem-, and as we have recently shown, also lymphangiogenesis in animal models (Andrieu-Soler et al., 2005; Hos et al., 2011a). Mechanisms of action are the inhibition of corneal macrophage infiltration and proinflammatory cytokine expression as well as reduction of macrophage-derived expression of VEGF-A and VEGF-C (Andrieu-Soler et al., 2005; Hos et al., 2011a). Additionally, Aganirsen also directly impairs vascular endothelial cell proliferation (Hos et al., 2011a). Aganirsen has also already successfully been tested as eye drops in phase II and III clinical trials (Cursiefen et al., 2009a, 2014). These data demonstrate that Aganirsen is safe, well-tolerated and effective in inhibiting progressive corneal neovascularization (Cursiefen et al., 2009a, 2014). Furthermore, Aganirsen seems to be able to reduce the need for keratoplasty in patients with viral keratitisassociated central corneal neovascularization (Cursiefen et al., 2014). Thus, Aganirsen might be the first drug that will be approved for the topical treatment of corneal neovascularization by the FDA. One limitation of the treatment with Aganirsen is that mature vessels are only marginally affected. As mentioned, the above summarized anti(lymph)angiogenic treatment strategies are able to potently inhibit progressive corneal neovascularization. In contrast, the regression of already established, mature blood and lymphatic vessels is still a major challenge. Recently, we have therefore analyzed the impact of various already clinically available techniques to regress corneal neovascularization in the experimental setting, namely corneal collagen crosslinking (CXL), photodynamic therapy (PDT), and fine needle diathermy (FND).

4.3.4.5. Photodynamic therapy (PDT). Photodynamic therapy (PDT) is an anti-angiogenic strategy utilizing a photosensitizer and light source. After the application of the photosensitizer verteporfin (a benzoporphyrin derivative), this molecule is specifically absorbed and retained in neovascular structures. By treatment with non-thermal 689 nm red-light, the accumulated verteporfin can be selectively activated, which leads to the liberation of cytotoxic free radicals causing occlusion of neovascular endothelial cells (Gomer et al., 1988; Gupta and Illingworth, 2011; Henderson and Dougherty, 1992; Reed et al., 1988; Weiss et al., 2012). In ophthalmology, PDT is currently used in the clinic to treat several neovascular diseases in the posterior pole of the eye, such as subfoveal choroidal neovascularization, choroidal hemangioma, and polypoidal choroidal vasculopathy (Verteporfin in Photodynamic Therapy Study Group, 2001; Harding, 2001; Reinke et al., 1999; Ziemssen and Heimann, 2012). In the cornea, PDT using systemically applied verteporfin as a photosensitizer has been shown to regress pathologic blood vessels in rabbit models, but the effect of PDT on corneal lymphatic vessels was uncharacterized (Holzer et al., 2003; Yoon et al., 2006). Holzer et al. found that when using low laser energy (such as 17 or 50 J/cm2), PDT was able to achieve a 30-50% regression of corneal blood vessels, although vessels reoccurred 3–5 days after treatment (Holzer et al., 2003). When the laser energy was increased to 150 J/cm2, the average regression of corneal blood vessels was 56% and no regrowth was observed at 10 days post PDT (Holzer et al., 2003). We have previously shown in mice that when verteporfin is used intrastromally, corneal lymphatic vessels can be selectively regressed without having an effect on mature corneal blood vessels (Bucher et al., 2014). More recently, our group also showed in a murine corneal neovascularization model, that with PDT using intravenously applied verteporfin, both mature corneal blood and lymphatic vessels could time-dependently be regressed (Hou et al., 2017) (Fig. 11). PDT decreased 60% of the corneal lymphatic and 40% of the mature blood vessels without notable side effect on adjacent structures. Importantly, via pre-transplant PDT treatment (using intravenous verteporfin and red-light irradiation 24 h later), allograft survival in prevascularized high-risk corneas was significantly improved (PDT: 75% survival vs. controls: 27%; Fig. 11) (Hou et al., 2017). As PDT is already used in the clinic to treat patients with neovascular diseases at the posterior segment, this promising strategy might also be used for angio-regressive therapy at the cornea in the near future.

4.3.4.4. Corneal collagen crosslinking (CXL). Corneal collagen crosslinking (CXL) is usually used in several corneal ectatic diseases for stabilization of the cornea via bonding of one corneal collagen fiber to another adjacent fiber (Raiskup and Spoerl, 2013). CXL has been applied to treat progressive keratoconus since 2003 (Wollensak et al., 2003). In 2016, riboflavin ophthalmic solution combined with ultraviolet - A (UV-A) irradiation, which is used in CXL, received approval from the US Food and Drug Administration (FDA). CXL using riboflavin topically followed by UV-A illumination was reported to reduce corneal keratocyte density by inducing keratocyte apoptosis (Mencucci et al., 2010; Spoerl et al., 2007; Wollensak et al., 2004a, 2004b). This indicated that CXL might also regress pathological vessels in the cornea by having a similar effect on vascular endothelial cells, as these vessels are also located in the corneal stroma. Indeed, our group showed for the first time that CXL using riboflavin and UV-A successfully regressed pre-existing blood and lymphatic vessels in the murine cornea (Hou et al., 2018) (Fig. 10). After topical application of riboflavin solution and UV-A irradiation in the mouse model of sutureinduced inflammatory corneal neovascularization, the CXL treated corneas showed significantly less pathological blood and lymphatic vessels. In addition, TUNEL+ cells were observed in both corneal blood and lymphatic vascular endothelium, indicating that corneal vessels regressed via the induction of cell death in vascular endothelial cells (Hou et al., 2018). This may be caused by the reaction of applied riboflavin with UV-A, which results in the release of highly reactive singlet oxygen and other reactive oxygen radicals and subsequent cell death of vascular endothelial cells (Spoerl et al., 2007; Wollensak et al., 2004a, 2004b). CXL treatment also decreased the number of macrophages and of CD45 + cells in inflamed corneas. The decrease of macrophages may contribute to the regression of pathologic corneal vessels, as macrophages essentially contribute to inflammatory lymphangiogenesis and lymphatic vessel maintenance (Kataru et al., 2009; Kerjaschki et al., 2006; Li et al., 2012d; Maruyama et al., 2005, 2007, 2012; Song et al., 2018). Importantly, via preoperative treatment of vascularized high-risk recipient beds with CXL, subsequent corneal allograft survival was significantly enhanced (Hou et al., 2018) (Fig. 10). This effect seems to be mainly attributed to the regression of pathological corneal

4.3.4.6. Fine needle diathermy (FND). Fine needle diathermy (FND) of corneal blood vessels is a simple and inexpensive procedure that was first described and clinically applied in 2000 (Pillai et al., 2000). Here, corneal neovascularization was coagulated by using a stainless steel 3/8 needle mounted on a 10/0 black nylon suture and diathermy current. The needle was inserted intrastromally near the limbus, parallel to and 17

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Fig. 11. Time-dependent regression of corneal blood and lymphatic vessels after PDT using intravenous verteporfin promotes murine high-risk graft survival. A+B: Quantification of corneal blood (A) and lymphatic (B) vessels after PDT using intravenous verteporfin. Corneal neovascularization was previously induced by suture placement (*:p < 0.05; **:p < 0.01; ***:p < 0.001). C: Promotion of corneal allograft survival after subsequent high-risk keratoplasty (Control: without preoperative treatment, PDT: with preoperative PDT treatment; *: p < 0.05). Modified from (Hou et al., 2017).

subdivided into five species (Netrin 1, 3, 4, G1, G2). Netrin-4 is expressed in almost all tissues of the eye, and the basement membranes of the lens and the cornea, the inner limiting membrane, the vascular basement membrane and the Bruch's membrane of the retina express very high levels of Netrin-4 (Li et al., 2012c). In the cornea, it was shown, that topical application of recombinant Netrin-4 supports wound healing after chemical burn. Treatment with Netrin-4 post injury resulted in a decreased level of neovascularization and inflammation. Furthermore, Netrin-4 was able to promote the regression of preexisting blood vessels (Han et al., 2015). So far, no effect on corneal suture-induced lymphangiogenesis in Netrin-4 knockout mice could be shown (Maier et al., 2017). However, Reuten et al. showed a dose dependent effect of Netrin-4 on BECs as well as LECs, where high concentrations led to a destabilization of vessel-like tube formation and maintenance (Reuten et al., 2016). 4.3.4.7.2. Soluble CD83 (sCD83). The expression of costimulatory membrane-bound molecules determines the capacity of DCs to activate T cells (Dilioglou et al., 2003; Pardoll, 2002). The membrane-bound glycoprotein CD83 (mCD83) is a well-known surface marker for mature DCs in humans and mice (Berchtold et al., 1999; Twist et al., 1998). On mature DCs, mCD83 acts as a costimulatory molecule and increases T cell activation (Aerts-Toegaert et al., 2007; Prechtel et al., 2007). Hock et al. reported on the existence of a soluble form of CD83 (sCD83) released by activated DCs and B cells, which was also detectable in normal human sera (Hock et al., 2001) and, in increased concentrations, in sera of patients with hematological malignancies (Hock et al., 2004). A direct correlation between increased sCD83 levels and treatment free survival was observed in patients suffering from chronic lymphocytic leukemia, indicating that sCD83 may modulate anti-tumoral immune responses (Hock et al., 2009). In subsequent studies, using recombinant sCD83 molecules, it was shown that sCD83 has indeed very prominent immunomodulatory properties. In this respect, it was reported that recombinant sCD83 inhibits DC maturation and blocks T cell stimulation (Lan et al., 2010; Lechmann et al., 2001; Scholler et al., 2002). Further reports also demonstrated that CD83 is a regulator of B cell function (Breloer et al., 2007; Kretschmer et al., 2009; Krzyzak et al., 2016). Thus, considering the limitations and failures of current therapeutics in preventing (long-term) corneal transplant rejection on the one hand, and the long-lasting tolerogenic effects of sCD83 on the other hand, we investigated whether sCD83 might be a novel therapeutic option in corneal transplantation. Importantly, we could demonstrate that sCD83 suppresses corneal allograft rejection when applied systemically or even topically in form of eye drops (Fig. 13). In both cases, the tolerogenic effect of sCD83 was associated with an increased frequency of T regs in the eye-draining lymph nodes, which was induced by DCs (Fig. 13). Furthermore, we

at the same level of the blood vessels. In case of multiple, closely located vessels, the needle was inserted tangentially to occlude as many vessels as possible in a single shot. Subsequently, the diathermy probe was brought into contact with the shaft of the surgical needle. The contact was maintained until mild blanching of the corneal stroma and corneal shrinkage occurred. FND can reach and occlude both afferent and efferent vessels at any corneal depth and is suitable for progressive as well as already mature vessels (Faraj et al., 2014). After FND, retreatment might be required due to reperfusion of occluded vessels and success rates vary from 60 to 100% after one single treatment, depending on the study (Faraj et al., 2014; Pillai et al., 2000; Thatte, 2011). The long-term efficacy and safety of FND is well documented, and common side effects are transient corneal opacification, and subconjunctival and intracorneal hemorrhages. These complications completely resolve after one day to a couple of weeks. Corneal microperforation is considered as a more serious complication that can occur during passaging of the needle, especially in thin corneas (Faraj et al., 2014; Koenig et al., 2012; Pillai et al., 2000; Thatte, 2011). When used alone, FND can induce the release of proangiogenic factors, especially when applied to the majority of vessels (Pillai et al., 2000). Hence, Spiteri et al. proposed the use of corneal angiography to help and guide FND to selectively close the afferent vessels, minimizing the numbers of intrastromal needle insertion and diathermy without altering the effectiveness of FND (Spiteri et al., 2015). Additionally, we and others suggested the combination of FND and topical or subconjunctival Bevacizumab to antagonize newly released VEGF (Elbaz et al., 2015; Hussain and Savant, 2017; Koenig et al., 2012). We have recently identified that FND treatment not only targets visible blood, but also clinically invisible lymphatic vessels in mice (Le et al., 2018b) (Fig. 12). The most pronounced effect of FND on regression of both blood and lymphatic vessels was observed at day 7 after FND treatment. Importantly, pretreatment of pathologically prevascularized high-risk eyes with FND significantly improved graft survival after subsequent keratoplasty in mice (Fig. 12) (Le et al., 2018b). As FND to treat corneal neovascularization is already in clinical use, patients studies that also demonstrate an improvement of corneal allograft survival will likely follow. Indeed, our group has recently reported initial results of a clinical pilot study, which appear promising (Hos et al., 2019a). 4.3.4.7. Novel experimental options. In addition to the already partially clinically available vessel regression methods presented above, novel molecular strategies that interfere with the afferent and efferent arm of the immune reflex arc are under development. These therapeutic strategies may effectively reduce the rates of corneal allograft rejection. 4.3.4.7.1. Netrin-4. Netrins are laminin-like secreted proteins 18

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Fig. 12. Fine needle-diathermy (FND) targets corneal blood and lymphatic vessels and promotes murine high-risk graft survival. A+B): Effect of FND on regression of blood vessels (BV; A) and lymphatic vessels (LV; B) at four time points after treatment (14 days after corneal suture placement). Compared to the nontreated group (NON: red column), the FND treated group (green column) resulted in significant regression of both BV and LV at all analyzed time points (p < 0.05). The most obvious effect of FND was observable at day 7 with the reduction of BV by 60% (p < 0.0001) and LV by approximately 80% (p < 0.0001) as compared to the NON group (ANTI: anti-inflammatory therapy). C) Two weeks after corneal suture placement, mice were treated with FND and anti-inflammatory eye drops (FND; green line) or with anti-inflammatory eye drops only (ANTI; purple line). Graft survival was significantly improved when compared to controls that did not receive FND or anti-inflammatory therapy (NON; red line; survival rates: FND versus NON: 60.9% vs. 8.3%, p < 0.0001; ANTI versus NON: 33.3% vs. 8.3%, p = 0.0164). FND treatment also significantly improved graft survival in comparison to isolated anti-inflammatory therapy (FND versus ANTI: 60.9% vs. 33.3%, p < 0.05). Modified from (Le at al. 2018).

5. Immune reactions after modern lamellar keratoplasty: immunology, clinical presentation, prevention, treatment, and outcomes

demonstrated that indoleamine 2,3-dioxygenase (IDO) and TGF-β are the major mediators of the sCD83 effect. Thus, these findings open interesting new therapeutic options not only in the field of corneal transplantation but also for other solid organ and tissue transplantation studies that require long-term tolerance.

5.1. Stromal immune reactions after DALK After DALK, endothelial immune reactions cannot occur, as the Descemet membrane and the endothelium of the recipient cornea are retained. Thus, only epithelial, subepithelial, or stromal immune

Fig. 13. sCD83 improves murine corneal allograft survival. Top left: Topical application of sCD83 promotes corneal allograft survival. Top right: sCD83 induces Foxp3-positive regulatory T cells in the lymph node, but not in the spleen. Bottom left: Blockade of indoleamine 2,3-dioxygenase (IDO) activity by 1-MT reverses the effect of sCD83 on corneal allograft survival, indicating that IDO is crucial for the immunoregulatory effect of sCD83. Bottom right: In vitro treatment of DC-T cell cocultures with sCD83 leads to a decrease of pro-inflammatory cytokines and an increase of TGF-β expression. Adapted from (Bock et al., 2013). 19

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

rejection risk. As the graft is introduced into the anterior chamber, the mechanisms of ACAID can be effective, and the exposure to APCs of the recipient cornea, which are mainly located in the anterior stroma, is reduced. Similarly, significantly fewer APCs from the donor, which also have been shown to contribute to graft rejection, are transplanted. Furthermore, the transplanted donor tissue might be lees immunogenic, because the donor epithelium and the majority of the donor stroma are not transplanted. Another important factor is the absence of corneal sutures, which may become loose in case of PK and induce inflammation and a secondary immune reaction.

reactions can occur. Interestingly, the risk of developing an epithelial/ subepithelial/stromal immune reaction seems to be considerably higher when compared to PK (Giannaccare et al., 2018; Gonzalez et al., 2017; Olson et al., 2012). This might be due to the fact that in case of DALK, donor antigens are not introduced into the anterior chamber and the mechanisms of ACAID cannot be effective. In a study by Giannaccare et al., 20 of 377 eyes (5.3%) experienced an episode of stromal rejection (Giannaccare et al., 2018). Stromal rejection was symptomatic in 7 of 20 cases (35%), with blurred vision, photophobia, and/or ocular redness as main complaints, whereas the remaining 13 cases (65%) were asymptomatic and stromal rejection was diagnosed during a scheduled follow-up examination (Giannaccare et al., 2018). The majority of the rejection episodes (75%) occurred within the first postoperative year and almost all patients were off steroidal treatment. Patients with stromal rejection were treated with dexamethasone 0.1% eye drops every 2 h for 14 days and then slowly tapered off over a 6-month period. All episodes resolved within 6 months after the onset, with no significant differences between pre-rejection and 6-month post-rejection values of visual acuity, central corneal thickness, and endothelial cell density (Giannaccare et al., 2018). Gonzalez et al. followed 251 DALK procedures, where the risk of stromal rejection was 14% with 18month median follow up (Gonzalez et al., 2017). Episodes of graft rejection were treated with prednisolone acetate 1% or tobramycin/ dexamethasone ointment 0.1% at various dosing intervals depending on severity, and the dose was tapered over weeks to months. By this treatment, stromal rejection resolved in all eyes (Gonzalez et al., 2017). Thus, non-endothelial, mainly stromal immune reactions after DALK occur more frequent than after PK. These immune reactions can successfully be treated with intensified topical corticosteroids. Furthermore, it seems that with time the risk of epithelial, subepithelial and stromal rejection episodes might decrease, as the host epithelial cells and keratocytes repopulate the donor tissue, although late rejections after 3-4 years have been reported (Roberts et al., 2016). For the prevention of immune reactions after DALK, it is advisable to start with topical corticosteroids (e.g. prednisolone acetate 1%) at least 5 times a day, followed by a monthly reduction by 1 drop to once daily at least for the first postoperative year. However, this recommendation is mainly based on the authors’ own experiences, and other groups also use shorter postoperative corticosteroid regimens (Borderie et al., 2011; Noble et al., 2007; Romano et al., 2015).

5.2.1.1. Incidence and clinical presentation. Endothelial immune responses after DSAEK are clinically more subtle than after PK. In a study by Price and colleagues, 35% of the patients were asymptomatic and immune reaction was only diagnosed during a routine check (Price et al., 2009b). The signs of the endothelial immune responses also differ from immune responses after PK, as there are often only isolated precipitates, which might be focal or diffuse (60-70%). Further signs of graft rejection are corneal edema (10-25%) and anterior chamber cells (25%). Classical endothelial rejection lines are rare, but can occasionally occur (Fiorentzis et al., 2015; Jordan et al., 2009; Price et al., 2009b; Saelens et al., 2011; Wu et al., 2012). The graft rejection risk after DSEAK is higher for African Americans, in eyes with preexisting glaucoma and in steroid responders (Price et al., 2009b). Several studies have reported on the time between surgery and graft rejection: In a study by Wu et al. the mean time to rejection following DSAEK was 13 ± 10 months (Wu et al., 2012). Li et al. also have shown that the greatest number of rejections occurred between 12 and 18 months postoperatively (Li et al., 2012b). Similarly, a study by Sepsakos et al. have shown that the greatest number of rejection episodes occurred between postoperative months 19 and 30 and that there was a statistically significant association of developing a rejection episode when a patient was off of topical steroids (Sepsakos et al., 2016). 5.2.1.2. Clinical course. Work from Price and colleagues have demonstrated that endothelial immune responses after DSAEK occur in about 7% of patients within the first postoperative year (Price et al., 2009b). In a more recent study by Price and colleagues, the incidence of endothelial graft rejection during the first 5 postoperative years in a large FED cohort was retrospectively evaluated (Price et al., 2018). From 1312 DSAEK cases, 80 cases experienced a rejection episode, resulting in a cumulative 5-year rejection episode rates of 7.9%. Interestingly, only a minority of these graft rejection episodes resulted in secondary graft failure requiring a re-graft (Price et al., 2018), and experiencing a rejection episode was not a predictor of graft failure within 5 years of surgery in this study, which is in contrast to the study of Figueiredo et al. where 60-80% of grafts experiencing a rejection episode failed (Figueiredo et al., 2015b). Thus, reports regarding the risk of graft failure after immune rejection after DSEAK show significant variation.

5.2. Endothelial immune reactions after endothelial keratoplasty (EK) As previously mentioned, the risk of developing an immune reaction after EK is considerably lower when compared to PK, especially after DMEK (Allan et al., 2007; Anshu et al., 2012; Hos et al., 2017b; Li et al., 2012b; Nguyen et al., 2007; Price et al., 2009b; Sepsakos et al., 2016). In addition, there are so far only few relevant studies reporting on graft rejection rates in EK patients who would have been considered as highrisk hosts in PK. These (smaller) studies demonstrate relatively low graft rejection rates when compared to high-risk PK (between 0% and 17%) (Anshu et al., 2011; Baydoun et al., 2015b; Heinzelmann et al., 2017; Kim et al., 2012; Mitry et al., 2014), and there is no general consensus whether a high-risk scenario in EK even exists so far. It will be interesting to see whether studies in larger cohorts with longer follow-up can recapitulate this fact and whether it will be necessary to further sub-define “low-low-risk” and “high-low-risk” scenarios in the already low-risk setting of EK in the future.

5.2.1.3. Prevention and treatment. Although the risk of immune reactions after DSAEK seems to be slightly lower, most reports follow the postoperative treatment scheme for PK. In a study by Ezon et al., patients were initially treated with prednisolone acetate 1% 6 to 8 times per day. This was maintained during the first week of follow-up, then tapered to 3 to 4 times daily for the first postoperative month and then tapered further by 1 drop daily every 4–6 months (Ezon et al., 2013). In other studies, the topical corticosteroid regimen used to prevent graft rejection was prednisolone acetate 1% dosed 4 times daily, tapered by 1 drop every 3 months to once daily (Hamzaoglu et al., 2015). From our point of view, it is advisable to start at least 5 times a day with corticosteroid eye drops, followed by a monthly reduction by 1 drop to once daily, which should then be continued for at least another 18 months or even indefinitely.

5.2.1. Endothelial immune reactions after DSAEK Within the first 2 years after keratoplasty, studies have shown that the risk of endothelial immune rejection is 5–17% for low-risk PK and 8-14% for DSAEK (Allan et al., 2007; Li et al., 2012b; Nguyen et al., 2007; Price et al., 2009b; Sepsakos et al., 2016). Thus the risk of developing an immune rejection episode after DSAEK seems to be reduced when compared to PK. There might be several reasons for the reduced 20

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

anticipate a rejection episode and/or to prevent an allograft rejection from clinically manifesting itself (Monnereau et al., 2014). Patients presenting with corneal graft rejection after DMEK might often be asymptomatic, especially in the absence of corneal edema despite clearly recognizable endothelial precipitates (Hos et al., 2017b). Furthermore, patients can present with ocular discomfort, redness and reduced visual acuity. Since most episodes are asymptomatic, a regular follow-up of patients after DMEK seems to be feasible. Baydoun et al. reported on 17 eyes with rejection, where 12 were asymptomatic, and 4 had no subjective complaints (Baydoun et al., 2016b). We have previously reported on 12 out of 1000 patients with rejection episodes, where five patients were symptomatic and 7 did not note the rejection episode (Hos et al., 2017b). These data suggest that DMEK transplant recipients should be examined regularly to actively look for any signs of rejection. We suggest that lamellar keratoplasty recipients (both after DSAEK and DMEK) should be controlled at 1, 3, 6, 12, 18 and 24 months postoperatively and annually thereafter.

If a graft rejection episode occurs after DSAEK, it is mainly treated by intensive topical corticosteroid therapy. Wu and colleagues used prednisolone acetate 1% or difluprednate 0.05% hourly (Wu et al., 2012). In this study, some eyes were additionally treated with intracameral or sub-Tenon's corticosteroid injections or even systemic prednisolone, although reports on the beneficial impact of these strategies and a consensus whether systemic corticosteroid treatment should be initiated are missing so far. Li et al. also used topical prednisolone acetate 1% up to hourly doses, depending on the severity of the rejection. Oral steroids were not typically used to treat graft rejection episodes. Topical steroids were then gradually tapered down to a maintenance dose of 1 drop per day according to the speed of resolution of the rejection episode (Li et al., 2012b). 5.2.2. Endothelial immune reactions after DMEK 5.2.2.1. Incidence and clinical presentation. Several studies have reported on the incidence of graft rejection after DMEK: Anshu et al. reported on 1 graft rejection episode in 140 DMEK cases (Anshu et al., 2012), Guerra et al. reported on 7 episodes in 131 DMEK cases (Guerra et al., 2011), and Baydoun et al. reported on 352 eyes where 2 rejection episodes occurred (Baydoun et al., 2015a). In another study by Baydoun and colleagues, 17 from 750 eyes were reported to develop a rejection episode (Baydoun et al., 2016b). In the largest report of consecutive DMEK cases so far, we also observed a similar incidence of graft rejection after DMEK (Hos et al., 2017b). In our cohort, 12 of 905 included eyes developed a rejection episode, with estimated probabilities of rejection of 0.9% at 1 year and 2.3% at 4 years. In summary, these studies demonstrate that the risk of developing a graft rejection episode after DMEK is significantly lower than after PK or DSAEK. Whether the concepts of ACAID and immune privilege for corneal transplantation established in animal models also hold true for patients and lamellar keratoplasty is still not clear. In this context, it has e.g. been shown experimentally in rodents that removal of the spleen leads to increased rejection of corneal allografts after corneal transplantation. However, we have recently observed a splenectomized patient accepting a posterior lamellar graft without immune rejection with a follow-up of more than 5 years, indicating that an intact spleen might not be necessary for allograft acceptance after posterior lamellar keratoplasty in humans (Hos et al., 2019b). The clinical picture of graft rejection after DMEK can be very subtle. Immune reactions after DMEK may present with a classical Khodadoust line, but mostly show diffuse endothelial precipitates (Hos et al., 2017b). Endothelial precipitates are usually confined to the DMEK grafts but can also occur on peripheral corneal areas that were initially denuded by descemetorhexis but not covered by the donor Descemet membrane and donor endothelium (Hos et al., 2014) (Fig. 14). Melanin precipitates in areas of irregularities at the periphery of the graft should not be mistaken as precipitates due to an immune response. Additional clinical signs of acute graft rejection can include anterior chamber reaction and corneal edema. Interestingly, Monnereau and colleagues have retrospectively shown that changes in endothelial cell morphology might be detected even before an allograft rejection after DMEK occurs (Monnereau et al., 2014). From 500 eyes that underwent DMEK, 7 eyes developed typical clinical signs of an allograft rejection. Interestingly, specular microscopy images of these eyes that developed an allograft rejection surprisingly showed that characteristic endothelial cell changes could already be observed up to several months before the rejection became clinically apparent. Compared to asymptomatic DMEK control eyes, that showed homogenous reflectivity, fairly regular cell shape and distribution, and an invisible cell nucleus, eyes with a developing allograft rejection showed a clearly different image, with disorganized cell shape, size, and distribution, high reflectivity, and pronounced cell nuclei. This study suggests that allograft rejection may not be an acute event, but rather a slow-onset immune response and that, monitoring donor endothelium after DMEK may be used to

5.2.2.2. Clinical course. We have recently reported on the clinical course of patients with immune reactions after DMEK (Hos et al., 2017b). After the rejection episodes, visual acuity and central corneal thickness remained stable in most cases when rejection was immediately treated with intensified corticosteroids. However, endothelial cell densities (ECDs) were significantly lower when compared to the last visit before diagnosis of graft rejection. ECD values 3 months and 1 year after the graft rejection episodes were comparable without further decrease, indicating that ECD seems to be stable after the graft rejection episodes has subsided (Hos et al., 2017b). Also mid-/long-term stability after treatment of endothelial immune reactions after DMEK seems to be good (Schaub et al., 2019). 5.2.2.3. Prevention and treatment. Although the rate of graft rejection after DMEK is considerably low, local corticosteroid therapy is the mainstay in the prevention of graft rejection. Price and colleagues have analyzed the incidence of graft rejection when topical corticosteroids were discontinued 1 year after DMEK surgery. Here, 6% of the patients who stopped topical corticosteroids experienced a possible or probable rejection episode, whereas no single rejection episode occurred when topical corticosteroids were continued (Price et al., 2016). A comparison of different corticosteroids has also shown that rejection rates after DMEK are similar for lower effective topical steroids such as Fluorometholone 0.1% or Loteprednol etabonat 0.5%, when compared to prednisolone acetate 1%, which might therefore preferentially used when a steroid response is known (Price et al., 2014, 2015). In line with this, we also have previously demonstrated that the risk of graft rejection seems to be increased in patients without local corticosteroid treatment (Hos et al., 2017b). However, there are also reports where the majority of observed graft rejection episodes occured in patients that still receiving fluorometholone (Baydoun et al., 2016b). Since the report by Price and colleagues demonstrated that the risk of rejection following DMEK increased after discontinuation of local steroid therapy at the end of the first postoperative year (Price et al., 2016), and our studies also showed that the majority of graft rejection episodes occur within the first two postoperative years (Hos et al., 2017b), it is recommend that (low-dose) corticosteroids should be continued until at least the end of the second postoperative year, or even indefinitely (Hos et al., 2017b; Price et al., 2016; Quilendrino et al., 2017). However, it is currently not known whether the use of topical steroids in low doses represents an effective long-term prophylaxis of rejection reactions beyond the second year. Irrespective of whether a patient with newly diagnosed graft rejection is still on (low-dose) corticosteroid therapy or had discontinued local corticosteroids, the graft rejection episode is usually treated with intensified local corticosteroids, although reports on the exact treatment scheme are missing so far. In a study by Price et al., treatment of graft rejection ranged from topical fluorometholone 0.1% 4 times daily 21

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Fig. 14. Clinical presentation of an immune reaction after DMEK. A, Slitlamp photograph showing diffuse endothelial precipitates; red box: magnified area in (B). B, Magnification of the peripheral corneal area that was initially denuded and not covered by the donor Descemet membrane; precipitates are also detectable on the denuded area (arrowheads; arrow indicates the margin of the donor Descemet membrane). C, After 2 weeks of intensive treatment with topical corticosteroids, the immune reaction resolved and precipitates disappeared. Magnification of the peripheral denuded area; the arrows indicate the margin of the donor Descemet membrane (left) and the margin of descemetorhexis (right), respectively. D, Schematic drawing illustrating the distribution of endothelial precipitates. From (Hos et al., 2013).

keratoplasty as it minimizes surgical trauma and may solve the issue of donor shortage. However, the methods of in vitro CEC expansion as well as the cell delivery and treatment protocols remain to be improved. To successfully promote CEC proliferation in vitro while preserving their phenotype, it is necessary to use cytokines and growth factors to lift the mitotic block as well as prevent EnMT. Traditional ex vivo cell expansion methods including the enzymatic dissociation of corneal endothelial donor tissue using trypsin (Choi et al., 2014; Su et al., 2015), or the use of FGF were reported to induce loss of tight junctions and eventually EnMT (Lee and Kay, 2012). The use of the correct composition of culture media and substrate (such as laminin (Okumura et al., 2015) and collagen IV (Choi et al., 2014; Sugino et al., 2011)) as well as the genetic manipulation of the cells (for example by the use of p120 and Kaiso siRNAs leading to CEC proliferation while maintaining pumping function and hexagonal shape (Zhu et al., 2014)) may enable the successful culture and passaging of CECs. The use of Rho-kinase (ROCK) inhibitors has been a pivotal step in CEC culture and cell therapy. Okumura reported in 2009 that ROCK pathway inhibition promoted CEC proliferation in vitro while maintaining their functional phenotype (Okumura et al., 2009). Very importantly, ROCK inhibitor administered together with cultured CECs, which were injected in the anterior chamber, also enforced the homing and functionality of the cells in vivo (Okumura et al., 2012, 2016) as well as in bullous keratopathy patients leading to decreased corneal thickness and improved visual acuity (Kinoshita et al., 2018; Okumura et al., 2009). In principal, CEC transplantation could involve a lower risk for graft rejection in comparison to PK as the corneal endothelium was shown to be less involved in alloimmunity in contrast to corneal stromal and epithelial tissue (Hori and Streilein, 2001; Khodadoust and Silverstein, 1969; Saban et al., 2009), but the same holds true for DMEK. However, as described in the previous chapters, endothelial graft rejections might rarely occur (Anshu et al., 2012; Hos et al., 2017b). The ocular immune privilege and ACAID, the selective deficiency of antigen-specific DTH following the introduction of allogeneic cells or antigens into the anterior chamber (see chapter 3), is thought to be responsible for the

to prednisolone acetate 1% hourly, depending on the severity of graft rejection (Price et al., 2016). Using this treatment, 13 of 14 rejection episodes resolved, whereas 1 eye decompensated and required a regraft. Afterwards, topical corticosteroid dosing was tapered to once per day over several weeks and continued indefinitely. Anshu et al. reported on a similar treatment regimen, where eyes with a rejection episode were treated with topical prednisolone acetate 1% in a dose ranging from 8 times daily to hourly, depending on the severity of rejection (Anshu et al., 2012). The steroid was then gradually tapered, based on response to therapy, down to a maintenance dose of once daily. Our own treatment scheme for graft rejection is prednisolone acetate 1%; every half hour for 3 days, afterwards tapered by the following scheme: hourly for 1 week, 6 times daily for 1 week, followed by a weekly reduction by 1 drop to once daily, which is then continued indefinitely. By this treatment scheme, the majority of graft rejection episodes can be successfully treated and resolve (Hos et al., 2017b). In summary, the risk of developing an immune reaction after DMEK is very low and often does not lead to secondary graft failure with the requirement of a re-graft when timely treated with local intensified corticosteroids. For the prophylaxis of graft rejection, low-dose corticosteroid therapy should be continued until at least the end of the second postoperative year, or even indefinitely. 5.3. Perspective: immune reactions after endothelial cell transplantation Human corneal endothelial cells (CEC) do not proliferate in vivo as they are arrested at the G1 phase of their cell cycle (Joyce et al., 2002) and, unsurprisingly, their in vitro expansion is also challenging. Due to disease, injury or ageing, CEC count may reduce or the cells can become dysfunctional leading to vision loss (Abell et al., 2014; Alqudah et al., 2013; McLaren et al., 2014). Bullous keratopathy, resulting e.g. from FECD or (e.g. cataract-surgery related) endothelial trauma (Bourne and McLaren, 2004), are the leading causes of corneal blindness which may be treated with corneal endothelial cell transplantation (Flockerzi et al., 2018). CEC transplantation is an attractive alternative to posterior lamellar 22

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

(Niederkorn et al., 1981; Streilein et al., 1980). Thus, it is conceivable that corneal transplantation may be used to prevent subsequent solid organ transplant rejection (from the same donor). However, although the experimental groundwork of this fascinating concept was laid in the 1980's, a further transition into the clinic is still missing. So there not only is an unmet need in tackling the high rejection rates in vascularized high-risk grafts in patients, but also in translating that to other fields of solid organ transplantation. For better diagnosis and treatment of patients with corneal allograft rejection, it would also be of great interest to identify specific biomarkers e.g. in the serum or AqH of patients that indicate or even precede graft rejection episodes. In this context, a study by The Collaborative Corneal Transplantation Studies Research Group has shown that anti-class I lymphocytotoxic antibodies directed against donor class I HLA antigens are associated with graft rejection episodes in high risk patients, and might therefore be an indicator of allograft rejection (Hahn et al., 1995). Furthermore, Foster and colleagues have demonstrated that serum levels of soluble IL-2 receptor (sIL-2R), which has already been used as a marker for early rejection in solid organ transplant recipients, were also significantly elevated during acute rejection episodes in corneal transplant recipients when compared to prerejection and post-rejection levels (Foster et al., 1993). Recently, it has also been shown that preoperative elevation of AqH cytokine levels, e.g. of monocyte chemotactic protein 1 (MCP-1) and IFN-γ are associated with reduced ECDs after PK or DSAEK (Yagi-Yaguchi et al., 2017; Yazu et al., 2018). In addition, preoperatively elevated cytokine levels were associated with primary graft failure in PK and DSAEK eyes (Yamaguchi et al., 2018). Whether this applies also for graft rejection and not only graft failure, is so far unknown. Furthermore, an elevated innate immune response was recently observed in the AqH of patients with failed DMEK grafts, although it is unclear whether this is the result and not the cause of DMEK failure (Luznik et al., 2019). Evidently, larger prospective trials are necessary to determine the sensitivity, specificity and overall value of anti-class I lymphocytotoxic antibodies, sIL-2R or AqH cytokine levels as predictive tools for corneal allograft rejection. With molecular diagnostic tools becoming more widely and easier available, it is also imaginable that in the future, a “personalized approach” in high-risk patients to determine the individual rejection risk profile will be possible. One could e.g. analyze cytokine levels in the AqH and might specifically counteract dysregulated factors pre-, intra- or postoperatively. Currently, corticosteroids are still the mainstay for the treatment of graft rejection after keratoplasty, although there is no general consensus regarding the intensity of topical corticosteroid therapy and whether topical corticosteroids should be used in combination with oral or even intravenous corticosteroids. In addition, it is also unclear whether there is a benefit of additional intracameral corticosteroids for the treatment of acute graft rejection. Also for the prevention of graft rejection, there is no general agreement on how long corticosteroids should be used postoperatively to avoid immune reactions, although studies have shown that the risk of immune reactions is increased when corticosteroids are discontinued e.g. after 6 months or 1 year (irrespective whether PK or posterior lamellar keratoplasty was performed) (Nguyen et al., 2007; Price et al., 2016). Whether this also holds true when corticosteroids are stopped, e.g. after 2 years or later, is currently unknown. As outlined in chapter 4.3.1, alternative, steroid-reducing/replacing medications including topical or systemic CsA, MMF, Rapamycin, Tacrolimus, and others have also been demonstrated to reduce the incidence of graft rejection after (high-risk) PK. However, also these drugs seem to be only partially effective and graft rejection episodes might still occur. In this context, it should also be said that there is still an unmet need for a licensed topical inhibitor of (lymph)angiogenesis available for use in the clinic as outlined in a recent consensus paper (Cursiefen et al., 2012). In DMEK surgery, further modifications of DMEK such as HemiDMEK/Quarter-DMEK have been developed recently (Lam et al., 2014;

low risk of allograft rejection after endothelial transplantation (Niederkorn, 1990). Allo-antigens situated on the surface of corneal endothelial cells may induce ACAID and thus promote the survival and successful homing of grafted CECs. This is supported by the observation that survival of corneal allografts is increased by triggering ACAID via injecting allogeneic splenocytes or CECs lines into the anterior chamber and the acquisition of donor-specific tolerance in a mouse PK model (Niederkorn and Mellon, 1996). Recently, Yamada et al. investigated the ability of murine eyes which received murine CEC grafts to enable ACAID and to promote transplant tolerance; it was demonstrated that the injection of CECs into the anterior chamber did not induce any allorejection (Yamada et al., 2016). These data, along with clinical observations, indicate that CEC grafts are less likely to be rejected in comparison to PK. However, whether a difference to DMEK exists is currently unknown. Despite progress, CEC transplantation still remains challenging. Ongoing problems include culture optimization and the choice of cell delivery method. Even though the CEC injection in the anterior chamber appears to be a promising approach, it requires that the patient remains in a prone position for several hours. On the other hand, the use of CEC sheets cultured on substrates including amniotic membrane (Ishino et al., 2004), or collagen sheets (Koizumi et al., 2007; Mimura et al., 2004; Yamaguchi et al., 2016), can also be problematic as technical issues during the surgery, carrier opacity and cell sheet detachment are likely to occur. The development of safe CEC delivery in the clinic has progressed significantly as recent studies have proposed protocols avoiding xenobiotic components including bovine serum (Soh et al., 2017), however, the development of Good Manufacturing Practice (GMP)-compliant methods for cell expansion and the optimization of cell delivery surely requires further investigation. Importantly, a recent study where a ROCK inhibitor was co-injected with cultured CECs into the anterior chamber of bullous keratopathy patients showed promise for improved outcome of future endothelial cell therapy treatments (Kinoshita et al., 2018). The cells were expanded ex vivo and they were injected into the anterior chamber of the patients who subsequently remained in prone position for 3 h. Notably, there was restoration of corneal transparency, with a central corneal endothelial cell density of > 500 cells/mm2 at approximately 6 months following treatment. Corneal thickness was also reduced under 630 μm and the majority of patients (9 out of 11) had significantly improved visual acuity. The results of this study demonstrate the clear benefits of injection of CECs supplemented with a ROCK inhibitor for patients with bullous keratopathy and potentially other corneal endothelial disorders (Kinoshita et al., 2018). 6. Future directions Corneal transplantation immunology in the low-risk setting differs significantly from other tissue and solid organ grafting, as the cornea is immunologically privileged and avascular (Cursiefen, 2007). However, the high-risk setting, where pathological blood and lymphatic vessels are present prior to surgery and thus the cornea's immune privilege is lost, permits comparison to transplantations performed in primarily vascularized tissues and organs. Indeed, findings from grafting experiments at the vascularized cornea have contributed to the unraveling of novel fundamental principles in transplantation immunology that are also relevant outside the eye (Hos et al., 2015). The majority of our knowledge regarding e.g. the role of lymphatic vessels in allograft rejection is deduced from experimental studies at the cornea. Beyond that, it is imaginable that other tissue or solid organ transplantations could benefit not only from knowledge derived from the eye, but also practically: as mentioned in chapter 3, the introduction of (allo-)antigens into the anterior chamber induces a systemic inhibition of adaptive immune responses against these antigens (ACAID), leading to acceptance e.g. of skin allografts that would have been rejected without the previous introduction of antigens into the anterior chamber 23

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Conflicts of interest

Zygoura et al., 2018). This is of great interest to counteract the global donor shortage as one donor cornea can be used for two (in HemiDMEK) or even four (Quarter-DMEK) recipients. However, studies indicate that ECDs are lower than after (full) DMEK surgery although BCVA and central corneal thickness are comparable to (full) DMEK (Birbal et al., 2018; Zygoura et al., 2018). It will be important to determine whether the risk of immune reactions after Hemi/-QuarterDMEK is further reduced when compared to (full) DMEK (arguably because of less antigenic tissue transplanted) and whether more recipients of Hemi/-Quarter-DMEK decompensate earlier when compared to DMEK in mid-/long-term. In addition to Hemi-/Quarter-DMEK, a further novel concept of Descemet Stripping Without Endothelial Keratoplasty (“DWEK”) has been developed, where the diseased central Descemet membrane is removed without grafting a donor Descemet membrane and endothelium (Arbelaez et al., 2014; Borkar et al., 2016; Iovieno et al., 2017; Koenig, 2015). This completely eliminates the risk of immune reactions. However, in DWEK, visual recovery seems to be slow, midand longterm results are still missing, and studies show very variable and partially unpredictable outcomes (Arbelaez et al., 2014; Borkar et al., 2016; Iovieno et al., 2017; Koenig, 2015). Nevertheless, there will be definitely further modifications and refinements of current endothelial keratoplasty techniques in the future, which might lead to superior outcomes or even lower graft rejection rates when compared to DMEK surgery. In addition, it will be interesting to study how endothelial keratoplasty techniques will perform for alternative indications, such as DMEK for Descemet membrane rupture in acute keratoconus (Tu, 2017). Furthermore, comparative trails will be needed to determine whether the exciting concept of CEC transplantation, which has been very recently transferred into the clinic (Kinoshita et al., 2018), is equal e.g. to DMEK surgery in terms of efficacy, long-term outcomes, safety and the risk of immune reactions. Finally, therapeutic temporary induction of lymphangiogenesis may become a tool to treat corneal stromal edema (Hos et al., 2017a). In eyes with severely altered ocular surface conditions e.g. after chemical burn or in traumatized eyes, only an artificial cornea/keratoprosthesis can help. In that context, a yet unmet clinical problem is the inflammation-driven destruction of the carrier cornea in Boston keratoprosthesis surgery, the worldwide most commonly used from of keratoprosthesis (Ahmad et al., 2016; Lee et al., 2015; Schaub et al., 2018). Pre-transplant (Kanellopoulos and Asimellis, 2014) or in situ crosslinking (Toth et al., 2016) may be a future strategy to avoid corneal melting after Boston Kpro surgery. In addition, immune-modulating approaches are also potentially useful strategies (Robert et al., 2017). Furthermore, the problem of inability to precisely measure intraocular pressure after Boston keratoprosthesis may be avoided using telemetric IOP measurements with an intraocular pressure sensor, as has recently been tested successfully in a clinical trial (Enders et al., 2019). Taken together, modern lamellar corneal transplant surgery has revolutionized not only outcomes in keratoplasty, but also the field of corneal transplant immunology itself, with endothelial immune reactions having lost most of their threat. Nonetheless, in vascularized highrisk eyes there is still a huge unmet need to prevent rejections and to promote long-term graft survival.

The authors declare no conflict of interest. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.preteyeres.2019.07.001. References Abdelfattah, N.S., Amgad, M., Zayed, A.A., Hussein, H., Abd El-Baky, N., 2016. Molecular underpinnings of corneal angiogenesis: advances over the past decade. Int. J. Ophthalmol. 9, 768–779. Abell, R.G., Kerr, N.M., Howie, A.R., Mustaffa Kamal, M.A., Allen, P.L., Vote, B.J., 2014. Effect of femtosecond laser-assisted cataract surgery on the corneal endothelium. J. Cataract Refract. Surg. 40, 1777–1783. Abud, T.B., Di Zazzo, A., Kheirkhah, A., Dana, R., 2017. Systemic immunomodulatory strategies in high-risk corneal transplantation. J. Ophthalmic Vis. Res. 12, 81–92. Achen, M.G., Jeltsch, M., Kukk, E., Makinen, T., Vitali, A., Wilks, A.F., Alitalo, K., Stacker, S.A., 1998. Vascular endothelial growth factor D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4). Proc. Natl. Acad. Sci. U. S. A 95, 548–553. Aerts-Toegaert, C., Heirman, C., Tuyaerts, S., Corthals, J., Aerts, J.L., Bonehill, A., Thielemans, K., Breckpot, K., 2007. CD83 expression on dendritic cells and T cells: correlation with effective immune responses. Eur. J. Immunol. 37, 686–695. Ahmad, S., Mathews, P.M., Lindsley, K., Alkharashi, M., Hwang, F.S., Ng, S.M., Aldave, A.J., Akpek, E.K., 2016. Boston type 1 keratoprosthesis versus repeat donor keratoplasty for corneal graft failure: a systematic review and meta-analysis. Ophthalmology 123, 165–177. Albuquerque, R.J., Hayashi, T., Cho, W.G., Kleinman, M.E., Dridi, S., Takeda, A., Baffi, J.Z., Yamada, K., Kaneko, H., Green, M.G., Chappell, J., Wilting, J., Weich, H.A., Yamagami, S., Amano, S., Mizuki, N., Alexander, J.S., Peterson, M.L., Brekken, R.A., Hirashima, M., Capoor, S., Usui, T., Ambati, B.K., Ambati, J., 2009. Alternatively spliced vascular endothelial growth factor receptor-2 is an essential endogenous inhibitor of lymphatic vessel growth. Nat. Med. 15, 1023–1030. Alitalo, K., Tammela, T., Petrova, T.V., 2005. Lymphangiogenesis in development and human disease. Nature 438, 946–953. Allan, B.D., Terry, M.A., Price Jr., F.W., Price, M.O., Griffin, N.B., Claesson, M., 2007. Corneal transplant rejection rate and severity after endothelial keratoplasty. Cornea 26, 1039–1042. Alldredge, O.C., Krachmer, J.H., 1981. Clinical types of corneal transplant rejection. Their manifestations, frequency, preoperative correlates, and treatment. Arch. Ophthalmol. 99, 599–604. Alqudah, A.A., Terry, M.A., Straiko, M.D., Greiner, M.A., Davis-Boozer, D., 2013. Immediate endothelial cell loss after penetrating keratoplasty. Cornea 32, 1587–1590. Amadio, M., Govoni, S., Pascale, A., 2016. Targeting VEGF in eye neovascularization: what's new?: a comprehensive review on current therapies and oligonucleotide-based interventions under development. Pharmacol. Res. 103, 253–269. Ambati, B.K., Nozaki, M., Singh, N., Takeda, A., Jani, P.D., Suthar, T., Albuquerque, R.J., Richter, E., Sakurai, E., Newcomb, M.T., Kleinman, M.E., Caldwell, R.B., Lin, Q., Ogura, Y., Orecchia, A., Samuelson, D.A., Agnew, D.W., St Leger, J., Green, W.R., Mahasreshti, P.J., Curiel, D.T., Kwan, D., Marsh, H., Ikeda, S., Leiper, L.J., Collinson, J.M., Bogdanovich, S., Khurana, T.S., Shibuya, M., Baldwin, M.E., Ferrara, N., Gerber, H.P., De Falco, S., Witta, J., Baffi, J.Z., Raisler, B.J., Ambati, J., 2006. Corneal avascularity is due to soluble VEGF receptor-1. Nature 443, 993–997. Amescua, G., Collings, F., Sidani, A., Bonfield, T.L., Rodriguez, J.P., Galor, A., Medina, C., Yang, X., Perez, V.L., 2008. Effect of CXCL-1/KC production in high risk vascularized corneal allografts on T cell recruitment and graft rejection. Transplantation 85, 615–625. Amouzegar, A., Chauhan, S.K., Dana, R., 2016. Alloimmunity and tolerance in corneal transplantation. J. Immunol. 196, 3983–3991. Andrieu-Soler, C., Berdugo, M., Doat, M., Courtois, Y., BenEzra, D., Behar-Cohen, F., 2005. Downregulation of IRS-1 expression causes inhibition of corneal angiogenesis. Investig. Ophthalmol. Vis. Sci. 46, 4072–4078. Ang, M., Soh, Y., Htoon, H.M., Mehta, J.S., Tan, D., 2016. Five-year graft survival comparing descemet stripping automated endothelial keratoplasty and penetrating keratoplasty. Ophthalmology 123, 1646–1652. Anshu, A., Price, M.O., Price Jr., F.W., 2011. Descemet's stripping endothelial keratoplasty under failed penetrating keratoplasty: visual rehabilitation and graft survival rate. Ophthalmology 118, 2155–2160. Anshu, A., Price, M.O., Price Jr., F.W., 2012. Risk of corneal transplant rejection significantly reduced with Descemet's membrane endothelial keratoplasty. Ophthalmology 119, 536–540. Anwar, M., Teichmann, K.D., 2002. Big-bubble technique to bare Descemet's membrane in anterior lamellar keratoplasty. J. Cataract Refract. Surg. 28, 398–403. Arbelaez, J.G., Price, M.O., Price Jr., F.W., 2014. Long-term follow-up and complications of stripping descemet membrane without placement of graft in eyes with Fuchs endothelial dystrophy. Cornea 33, 1295–1299. Bachmann, B., Taylor, R.S., Cursiefen, C., 2010. Corneal neovascularization as a risk factor for graft failure and rejection after keratoplasty: an evidence-based meta-

Funding German Research Foundation (DFG) FOR2240 “(Lymph)angiogenesis and Cellular Immunity in Inflammatory Diseases of the Eye”: HO 5556/1-2 (to DH), CL751/1-1 (to TC), Cu 47/9-1 (to CC; www.for2240. de); DFG MA 5110/5-1 (to MM); Center for Molecular Medicine Cologne (CMMC B3 to DH, FB and CC; CMMC CAP 11 to DH); EU Arrest Blindness (www.arrestblindness.eu; to MN and CC).

24

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Bernier, J., Roy, R., 1993. Risk factors of corneal graft failure. Ophthalmology 100, 1728–1735. Borderie, V.M., Guilbert, E., Touzeau, O., Laroche, L., 2011. Graft rejection and graft failure after anterior lamellar versus penetrating keratoplasty. Am. J. Ophthalmol. 151, 1024–1029 e1021. Borderie, V.M., Sandali, O., Bullet, J., Gaujoux, T., Touzeau, O., Laroche, L., 2012. Longterm results of deep anterior lamellar versus penetrating keratoplasty. Ophthalmology 119, 249–255. Borkar, D.S., Veldman, P., Colby, K.A., 2016. Treatment of Fuchs endothelial dystrophy by descemet stripping without endothelial keratoplasty. Cornea 35, 1267–1273. Bourne, W.M., McLaren, J.W., 2004. Clinical responses of the corneal endothelium. Exp. Eye Res. 78, 561–572. Bradley, B.A., 2002. Rejection and recipient age. Transpl. Immunol. 10, 125–132. Breloer, M., Kretschmer, B., Luthje, K., Ehrlich, S., Ritter, U., Bickert, T., Steeg, C., Fillatreau, S., Hoehlig, K., Lampropoulou, V., Fleischer, B., 2007. CD83 is a regulator of murine B cell function in vivo. Eur. J. Immunol. 37, 634–648. Bucher, F., Bi, Y., Gehlsen, U., Hos, D., Cursiefen, C., Bock, F., 2014. Regression of mature lymphatic vessels in the cornea by photodynamic therapy. Br. J. Ophthalmol. 98, 391–395. Bucher, F., Hos, D., Muller-Schwefe, S., Steven, P., Cursiefen, C., Heindl, L.M., 2015. Spontaneous long-term course of persistent peripheral graft detachments after Descemet's membrane endothelial keratoplasty. Br. J. Ophthalmol. 99, 768–772. Bucher, F., Parthasarathy, A., Bergua, A., Onderka, J., Regenfuss, B., Cursiefen, C., Bock, F., 2012. Topical Ranibizumab inhibits inflammatory corneal hem- and lymphangiogenesis. Acta Ophthalmol. 92, 143–148. Bucher, F., Roters, S., Mellein, A., Hos, D., Heindl, L.M., Cursiefen, C., Hermann, M., 2013. OSMO-UT-DSAEK" using THIN-C medium. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 251, 2181–2185. Buehler, A., Sitaras, N., Favret, S., Bucher, F., Berger, S., Pielen, A., Joyal, J.S., Juan, A.M., Martin, G., Schlunck, G., Agostini, H.T., Klagsbrun, M., Smith, L.E., Sapieha, P., Stahl, A., 2013. Semaphorin 3F forms an anti-angiogenic barrier in outer retina. FEBS Lett. 587, 1650–1655. Busin, M., Madi, S., Santorum, P., Scorcia, V., Beltz, J., 2013. Ultrathin descemet's stripping automated endothelial keratoplasty with the microkeratome double-pass technique: two-year outcomes. Ophthalmology 120, 1186–1194. Bussolino, F., Albini, A., Camussi, G., Presta, M., Viglietto, G., Ziche, M., Persico, G., 1996. Role of soluble mediators in angiogenesis. Eur. J. Cancer 32A, 2401–2412. Buttner, C., Clahsen, T., Regenfuss, B., Dreisow, M.L., Steiber, Z., Bock, F., Reis, A., Cursiefen, C., 2018. Tyrosinase is a novel endogenous regulator of developmental and inflammatory lymphangiogenesis. Am. J. Pathol. 189, 440–448. Buzzonetti, L., Laborante, A., Petrocelli, G., 2010. Standardized big-bubble technique in deep anterior lamellar keratoplasty assisted by the femtosecond laser. J. Cataract Refract. Surg. 36, 1631–1636. Cernak, A., Cernak, M., Liskova, P., 2002. [Treatment of severe caustic injuries of the anterior segment of the eye with limbal stem cell transplantation, penetrating keratoplasty and amnionic membrane transplantation]. Ceska a Slov. Oftalmol. : Cas. Ceske Talmologicke Spolecnosti Slovenske Talmologicke spolecnosti 58, 16–20. Chatel, M.A., Larkin, D.F., 2010. Sirolimus and mycophenolate as combination prophylaxis in corneal transplant recipients at high rejection risk. Am. J. Ophthalmol. 150, 179–184. Chen, L., Hamrah, P., Cursiefen, C., Zhang, Q., Pytowski, B., Streilein, J.W., Dana, M.R., 2004. Vascular endothelial growth factor receptor-3 mediates induction of corneal alloimmunity. Nat. Med. 10, 813–815. Cheng, Y.Y., Visser, N., Schouten, J.S., Wijdh, R.J., Pels, E., van Cleynenbreugel, H., Eggink, C.A., Zaal, M.J., Rijneveld, W.J., Nuijts, R.M., 2011. Endothelial cell loss and visual outcome of deep anterior lamellar keratoplasty versus penetrating keratoplasty: a randomized multicenter clinical trial. Ophthalmology 118, 302–309. Choi, J.S., Kim, E.Y., Kim, M.J., Khan, F.A., Giegengack, M., D'Agostino Jr., R., Criswell, T., Khang, G., Soker, S., 2014. Factors affecting successful isolation of human corneal endothelial cells for clinical use. Cell Transplant. 23, 845–854. Chow, S.P., Cook, S.D., Tole, D.M., 2015. Long-term outcomes of high-risk keratoplasty in patients receiving systemic immunosuppression. Cornea 34, 1395–1399. Cotsarelis, G., Cheng, S.Z., Dong, G., Sun, T.T., Lavker, R.M., 1989. Existence of slowcycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57, 201–209. Covert, D.J., Koenig, S.B., 2007. New triple procedure: descemet's stripping and automated endothelial keratoplasty combined with phacoemulsification and intraocular lens implantation. Ophthalmology 114, 1272–1277. Crawford, A.Z., Patel, D.V., McGhee, C., 2013. A brief history of corneal transplantation: from ancient to modern. Oman J. Ophthalmol. 6, S12–S17. Cursiefen, C., 2007. Immune privilege and angiogenic privilege of the cornea. Chem. Immunol. Allergy 92, 50–57. Cursiefen, C., Bock, F., Horn, F.K., Kruse, F.E., Seitz, B., Borderie, V., Fruh, B., Thiel, M.A., Wilhelm, F., Geudelin, B., Descohand, I., Steuhl, K.P., Hahn, A., Meller, D., 2009a. GS-101 antisense oligonucleotide eye drops inhibit corneal neovascularization: interim results of a randomized phase II trial. Ophthalmology 116, 1630–1637. Cursiefen, C., Cao, J., Chen, L., Liu, Y., Maruyama, K., Jackson, D., Kruse, F.E., Wiegand, S.J., Dana, M.R., Streilein, J.W., 2004a. Inhibition of hemangiogenesis and lymphangiogenesis after normal-risk corneal transplantation by neutralizing VEGF promotes graft survival. Investig. Ophthalmol. Vis. Sci. 45, 2666–2673. Cursiefen, C., Chen, L., Borges, L.P., Jackson, D., Cao, J., Radziejewski, C., D'Amore, P.A., Dana, M.R., Wiegand, S.J., Streilein, J.W., 2004b. VEGF-A stimulates lymphangiogenesis and hemangiogenesis in inflammatory neovascularization via macrophage recruitment. J. Clin. Investig. 113, 1040–1050. Cursiefen, C., Chen, L., Dana, M.R., Streilein, J.W., 2003. Corneal lymphangiogenesis: evidence, mechanisms, and implications for corneal transplant immunology. Cornea

analysis. Ophthalmology 117, 1300–1305 e1307. Bachmann, B.O., Bock, F., Wiegand, S.J., Maruyama, K., Dana, M.R., Kruse, F.E., LuetjenDrecoll, E., Cursiefen, C., 2008a. Promotion of graft survival by vascular endothelial growth factor a neutralization after high-risk corneal transplantation. Arch. Ophthalmol. 126, 71–77. Bachmann, B.O., Luetjen-Drecoll, E., Bock, F., Wiegand, S.J., Hos, D., Dana, R., Kruse, F.E., Cursiefen, C., 2009. Transient postoperative vascular endothelial growth factor (VEGF)-neutralisation improves graft survival in corneas with partly regressed inflammatory neovascularisation. Br. J. Ophthalmol. 93, 1075–1080. Bachmann, B.O., Pogorelov, P., Kruse, F.E., Cursiefen, C., 2008b. Patientenzufriedenheit nach posteriorer lamellärer Keratoplastik (DSAEK). Klin. Monatsblatter Augenheilkd. 225, 577–581. Bachmann, B.O., Schrittenlocher, S.A., Schaub, F., Siebelmann, S., Matthaei, M., Cursiefen, C., 2017. DMEK: probleme vermeiden, erkennen, lösen. Klin. Monatsblatter Augenheilkd. 234, 1354–1361. Baluk, P., McDonald, D.M., 2008. Markers for microscopic imaging of lymphangiogenesis and angiogenesis. Ann. N. Y. Acad. Sci. 1131, 1–12. Baradaran-Rafii, A., Ebrahimi, M., Kanavi, M.R., Taghi-Abadi, E., Aghdami, N., Eslani, M., Bakhtiari, P., Einollahi, B., Baharvand, H., Javadi, M.A., 2010. Midterm outcomes of autologous cultivated limbal stem cell transplantation with or without penetrating keratoplasty. Cornea 29, 502–509. Barbolina, M.V., Stack, M.S., 2008. Membrane type 1-matrix metalloproteinase: substrate diversity in pericellular proteolysis. Semin. Cell Dev. Biol. 19, 24–33. Bartel, D.P., 2004. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116, 281–297. Basu, S., Mohamed, A., Chaurasia, S., Sejpal, K., Vemuganti, G.K., Sangwan, V.S., 2011. Clinical outcomes of penetrating keratoplasty after autologous cultivated limbal epithelial transplantation for ocular surface burns. Am. J. Ophthalmol. 152, 917–924 e911. Baydoun, L., Dapena, I., Melles, G., 2016a. Evolution of posterior lamellar keratoplasty: PK - DLEK - DSEK/DSAEK - DMEK - DMET. In: Cursiefen, C., Jun, A.S. (Eds.), Current Treatment Options for Fuchs Endothelial Dystrophy. Springer, pp. 73–85. Baydoun, L., Ham, L., Borderie, V., Dapena, I., Hou, J., Frank, L.E., Oellerich, S., Melles, G.R., 2015a. Endothelial survival after descemet membrane endothelial keratoplasty: effect of surgical indication and graft adherence status. JAMA Ophthalmol. 133, 1277–1285. Baydoun, L., Livny, E., Ham, L., Bruinsma, M., Melles, G.R., 2016b. 360-Degree scheimpflug imaging to predict allograft rejection after descemet membrane endothelial keratoplasty. Cornea 35, 1385–1390. Baydoun, L., van Dijk, K., Dapena, I., Musa, F.U., Liarakos, V.S., Ham, L., Melles, G.R., 2015b. Repeat Descemet membrane endothelial keratoplasty after complicated primary Descemet membrane endothelial keratoplasty. Ophthalmology 122, 8–16. Berchtold, S., Muhl-Zurbes, P., Heufler, C., Winklehner, P., Schuler, G., Steinkasserer, A., 1999. Cloning, recombinant expression and biochemical characterization of the murine CD83 molecule which is specifically upregulated during dendritic cell maturation. FEBS Lett. 461, 211–216. Bhatti, N., Qidwai, U., Hussain, M., Kazi, A., 2013a. Efficacy of sub-conjunctival and topical bevacizumab in high-risk corneal transplant survival. J. Pak. Med. Assoc. 63, 1256–1259. Bhatti, N., Qidwai, U., Hussain, M., Kazi, A., 2013b. Efficacy of topical bevacizumab in high-risk corneal transplant survival. Pak J Med Sci 29, 519–522. Billingham, R.E., Brent, L., Medawar, P.B., 1953. Actively acquired tolerance of foreign cells. Nature 172, 603–606. Birbal, R.S., Hsien, S., Zygoura, V., Parker, J.S., Ham, L., van Dijk, K., Dapena, I., Baydoun, L., Melles, G.R.J., 2018. Outcomes of hemi-descemet membrane endothelial keratoplasty for Fuchs endothelial corneal dystrophy. Cornea 37, 854–858. Birnbaum, F., Bohringer, D., Sokolovska, Y., Sundmacher, R., Reinhard, T., 2005. Immunosuppression with cyclosporine A and mycophenolate mofetil after penetrating high-risk keratoplasty: a retrospective study. Transplantation 79, 964–968. Birnbaum, F., Mayweg, S., Reis, A., Bohringer, D., Seitz, B., Engelmann, K., Messmer, E.M., Reinhard, T., 2009. Mycophenolate mofetil (MMF) following penetrating highrisk keratoplasty: long-term results of a prospective, randomised, multicentre study. Eye 23, 2063–2070. Birnbaum, F., Reis, A., Bohringer, D., Sokolowska, Y., Mayer, K., Voiculescu, A., Oellerich, M., Sundmacher, R., Reinhard, T., 2006. An open prospective pilot study on the use of rapamycin after penetrating high-risk keratoplasty. Transplantation 81, 767–772. Bock, F., Maruyama, K., Regenfuss, B., Hos, D., Steven, P., Heindl, L.M., Cursiefen, C., 2013. Novel anti(lymph)angiogenic treatment strategies for corneal and ocular surface diseases. Prog. Retin. Eye Res. 34, 89–124. Bock, F., Onderka, J., Braun, G., Schneider, A.C., Hos, D., Bi, Y., Bachmann, B.O., Cursiefen, C., 2016. Identification of novel endogenous anti(lymph)angiogenic factors in the aqueous humor. Investig. Ophthalmol. Vis. Sci. 57, 6554–6560. Bock, F., Onderka, J., Dietrich, T., Bachmann, B., Kruse, F.E., Paschke, M., Zahn, G., Cursiefen, C., 2007. Bevacizumab as a potent inhibitor of inflammatory corneal angiogenesis and lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 48, 2545–2552. Bock, F., Onderka, J., Dietrich, T., Bachmann, B., Pytowski, B., Cursiefen, C., 2008. Blockade of VEGFR3-signalling specifically inhibits lymphangiogenesis in inflammatory corneal neovascularisation. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 246, 115–119. Bohringer, D., Grotejohann, B., Ihorst, G., Reinshagen, H., Spierings, E., Reinhard, T., 2018. Rejection prophylaxis in corneal transplant. Deutsches Arzteblatt Int. 115, 259–265. Boisgerault, F., Liu, Y., Anosova, N., Dana, R., Benichou, G., 2009. Differential roles of direct and indirect allorecognition pathways in the rejection of skin and corneal transplants. Transplantation 87, 16–23. Boisjoly, H.M., Tourigny, R., Bazin, R., Laughrea, P.A., Dube, I., Chamberland, G.,

25

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

773–777. Ellis, L.M., Hicklin, D.J., 2008. VEGF-targeted therapy: mechanisms of anti-tumour activity. Nat. Rev. Canc. 8, 579–591. Emami-Naeini, P., Dohlman, T.H., Omoto, M., Hattori, T., Chen, Y., Lee, H.S., Chauhan, S.K., Dana, R., 2014. Soluble vascular endothelial growth factor receptor-3 suppresses allosensitization and promotes corneal allograft survival. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 252, 1755–1762. Enders, P., Hall, J., Bornhauser, M., Mansouri, K., Altay, L., Schrader, S., Dietlein, T.S., Bachmann, B.O., Neuhann, T., Cursiefen, C., 2019. Telemetric intraocular pressure monitoring after Boston keratoprosthesis surgery. Ophthalmology 126, 322–324. Eslani, M., Baradaran-Rafii, A., Ahmad, S., 2012. Cultivated limbal and oral mucosal epithelial transplantation. Semin. Ophthalmol. 27, 80–93. Ezon, I., Shih, C.Y., Rosen, L.M., Suthar, T., Udell, I.J., 2013. Immunologic graft rejection in descemet's stripping endothelial keratoplasty and penetrating keratoplasty for endothelial disease. Ophthalmology 120, 1360–1365. Faraj, L.A., Elalfy, M.S., Said, D.G., Dua, H.S., 2014. Fine needle diathermy occlusion of corneal vessels. Br. J. Ophthalmol. 98, 1287–1290. Fasciani, R., Mosca, L., Giannico, M.I., Ambrogio, S.A., Balestrazzi, E., 2015. Subconjunctival and/or intrastromal bevacizumab injections as preconditioning therapy to promote corneal graft survival. Int. Ophthalmol. 35, 221–227. Ferguson, T.A., Griffith, T.S., 1997. A vision of cell death: insights into immune privilege. Immunol. Rev. 156, 167–184. Ferrari, G., Dastjerdi, M.H., Okanobo, A., Cheng, S.F., Amparo, F., Nallasamy, N., Dana, R., 2013. Topical Ranibizumab as a treatment of corneal neovascularization. Cornea 32, 992–997. Figueiredo, G.S., Jones, M.N., Krishna, Y., Figueiredo, F.C., Larkin, D.F., Kaye, S.B., 2015a. Transplant rejection following endothelial keratoplasty and penetrating keratoplasty in the United Kingdom: incidence and survival. Am. J. Ophthalmol. 160, 416–421. Figueiredo, G.S., Jones, M.N., Krishna, Y., Figueiredo, F.C., Larkin, D.F., Kaye, S.B., National Health Service, B., Transplant Ocular Tissue Advisory, G., 2015. Transplant rejection following endothelial keratoplasty and penetrating keratoplasty in the United Kingdom: incidence and survival. Am. J. Ophthalmol. 160, 416–421. Figueiredo, G.S., Salvador-Culla, B., Baylis, O.J., Mudhar, H.S., Lako, M., Figueiredo, F.C., 2018. Outcomes of penetrating keratoplasty following autologous cultivated limbal epithelial stem cell transplantation. Stem cells (Dayton, Ohio) 36, 925–931. Filipowicz, W., Bhattacharyya, S.N., Sonenberg, N., 2008. Mechanisms of post-transcriptional regulation by microRNAs: are the answers in sight? Nat. Rev. Genet. 9, 102–114. Fiorentzis, M., Viestenz, A., Seitz, B., 2015. [Immunological graft rejection with Khodadoust line after "Descemet stripping automated endothelial keratoplasty" (DSAEK)]. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 112, 783–787. Flockerzi, E., Maier, P., Bohringer, D., Reinshagen, H., Kruse, F., Cursiefen, C., Reinhard, T., Geerling, G., Torun, N., Seitz, B., all German Keratoplasty Registry, C., 2018. Trends in corneal transplantation from 2001 to 2016 in Germany: a report of the DOG-section cornea and its keratoplasty registry. Am. J. Ophthalmol. 188, 91–98. Forrester, J.V., Xu, H., Kuffova, L., Dick, A.D., McMenamin, P.G., 2010. Dendritic cell physiology and function in the eye. Immunol. Rev. 234, 282–304. Foster, C.S., Wu, H.K., Merchant, A., 1993. Systemic (serum) soluble interleukin-2 receptor levels in corneal transplant recipients. Doc. Ophthalmol. Adv. Ophthalmol. 83, 83–89. Fritz, M., Grewing, V., Bohringer, D., Lapp, T., Maier, P., Reinhard, T., Wacker, K., 2019. Avoiding hyperopic surprises after Descemet membrane endothelial keratoplasty in Fuchs dystrophy eyes by assessing corneal shape. Am. J. Ophthalmol. 197, 1–6. Fukushima, Y., Okada, M., Kataoka, H., Hirashima, M., Yoshida, Y., Mann, F., Gomi, F., Nishida, K., Nishikawa, S., Uemura, A., 2011. Sema3E-PlexinD1 signaling selectively suppresses disoriented angiogenesis in ischemic retinopathy in mice. J. Clin. Investig. 121, 1974–1985. Garmy-Susini, B., Varner, J.A., 2008. Roles of integrins in tumor angiogenesis and lymphangiogenesis. Lymphatic Res. Biol. 6, 155–163. Ghoraishi, M., Akova, Y.A., Tugal-Tutkun, I., Foster, C.S., 1995. Penetrating keratoplasty in atopic keratoconjunctivitis. Cornea 14, 610–613. Giannaccare, G., Weiss, J.S., Sapigni, L., Bovone, C., Mattioli, L., Campos, E.C., Busin, M., 2018. Immunologic stromal rejection after deep anterior lamellar keratoplasty with grafts of a larger size (9 mm) for various stromal diseases. Cornea 37, 967–972. Gibbons, A., Leung, E.H., Yoo, S.H., 2019. Cost-effectiveness analysis of descemet membrane endothelial keratoplasty compared to descemet stripping endothelial keratoplasty in the United States of America. Ophthalmology 126, 207–213. Gomer, C.J., Ferrario, A., Hayashi, N., Rucker, N., Szirth, B.C., Murphree, A.L., 1988. Molecular, cellular, and tissue responses following photodynamic therapy. Lasers Surg. Med. 8, 450–463. Gomes, J.A., Tan, D., Rapuano, C.J., Belin, M.W., Ambrosio Jr., R., Guell, J.L., Malecaze, F., Nishida, K., Sangwan, V.S., Group of Panelists for the Global Delphi Panel of, K.,Ectatic, D., 2015. Global consensus on keratoconus and ectatic diseases. Cornea 34, 359–369. Gonzalez, A., Price, M.O., Feng, M.T., Lee, C., Arbelaez, J.G., Price Jr., F.W., 2017. Immunologic rejection episodes after deep anterior lamellar keratoplasty: incidence and risk factors. Cornea 36, 1076–1082. Gorovoy, M.S., 2014. DMEK complications. Cornea 33, 101–104. Goweida, M.B., 2015. Intraoperative review of different bubble types formed during pneumodissection (big-bubble) deep anterior lamellar keratoplasty. Cornea 34, 621–624. Grimaldo, S., Yuen, D., Theis, J., Ng, M., Ecoiffier, T., Chen, L., 2015. MicroRNA-184 regulates corneal lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 56, 7209–7213. Guell, J.L., Morral, M., Gris, O., Elies, D., Manero, F., 2015. Comparison of sulfur hexafluoride 20% versus air tamponade in descemet membrane endothelial keratoplasty.

22, 273–281. Cursiefen, C., Chen, L., Saint-Geniez, M., Hamrah, P., Jin, Y., Rashid, S., Pytowski, B., Persaud, K., Wu, Y., Streilein, J.W., Dana, R., 2006. Nonvascular VEGF receptor 3 expression by corneal epithelium maintains avascularity and vision. Proc. Natl. Acad. Sci. U. S. A 103, 11405–11410. Cursiefen, C., Colin, J., Dana, R., Diaz-Llopis, M., Faraj, L.A., Garcia-Delpech, S., Geerling, G., Price, F.W., Remeijer, L., Rouse, B.T., Seitz, B., Udaondo, P., Meller, D., Dua, H., 2012. Consensus statement on indications for anti-angiogenic therapy in the management of corneal diseases associated with neovascularisation: outcome of an expert roundtable. Br. J. Ophthalmol. 96, 3–9. Cursiefen, C., Kruse, F.E., 2008. [Descemet's stripping with automated endothelial keratoplasty (DSAEK)]. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 105, 183–190 192. Cursiefen, C., Kruse, F.E., 2010. DMEK: posteriore lamelläre Keratoplastiktechnik. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 107, 370–376. Cursiefen, C., Kruse, F.E., Erlanger, D.G., 2009b. [Descemet's stripping automated endothelial keratoplasty (DSAEK)]. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 106, 939–952 quiz 953. Cursiefen, C., Maruyama, K., Bock, F., Saban, D., Sadrai, Z., Lawler, J., Dana, R., Masli, S., 2011. Thrombospondin 1 inhibits inflammatory lymphangiogenesis by CD36 ligation on monocytes. J. Exp. Med. 208, 1083–1092. Cursiefen, C., Masli, S., Ng, T.F., Dana, M.R., Bornstein, P., Lawler, J., Streilein, J.W., 2004c. Roles of thrombospondin-1 and -2 in regulating corneal and iris angiogenesis. Investig. Ophthalmol. Vis. Sci. 45, 1117–1124. Cursiefen, C., Schlotzer-Schrehardt, U., Kuchle, M., Sorokin, L., Breiteneder-Geleff, S., Alitalo, K., Jackson, D., 2002. Lymphatic vessels in vascularized human corneas: immunohistochemical investigation using LYVE-1 and podoplanin. Investig. Ophthalmol. Vis. Sci. 43, 2127–2135. Cursiefen, C., Viaud, E., Bock, F., Geudelin, B., Ferry, A., Kadlecova, P., Levy, M., Al Mahmood, S., Colin, S., Thorin, E., Majo, F., Frueh, B., Wilhelm, F., Meyer-Ter-Vehn, T., Geerling, G., Bohringer, D., Reinhard, T., Meller, D., Pleyer, U., Bachmann, B., Seitz, B., 2014. Aganirsen antisense oligonucleotide eye drops inhibit keratitis-induced corneal neovascularization and reduce need for transplantation: the I-can study. Ophthalmology 121, 1683–1692. Daniel, M.C., Bohringer, D., Maier, P., Eberwein, P., Birnbaum, F., Reinhard, T., 2016. Comparison of long-term outcomes of femtosecond laser-assisted keratoplasty with conventional keratoplasty. Cornea 35, 293–298. Dastjerdi, M.H., Saban, D.R., Okanobo, A., Nallasamy, N., Sadrai, Z., Chauhan, S.K., Hajrasouliha, A.R., Dana, R., 2010. Effects of topical and subconjunctival bevacizumab in high-risk corneal transplant survival. Investig. Ophthalmol. Vis. Sci. 51, 2411–2417. Dekaris, I., Gabric, N., Draca, N., Pauk-Gulic, M., Milicic, N., 2015. Three-year corneal graft survival rate in high-risk cases treated with subconjunctival and topical bevacizumab. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 253, 287–294. DelMonte, D.W., Kim, T., 2011. Anatomy and physiology of the cornea. J. Cataract Refract. Surg. 37, 588–598. Deng, S.X., Lee, W.B., Hammersmith, K.M., Kuo, A.N., Li, J.Y., Shen, J.F., Weikert, M.P., Shtein, R.M., 2018. Descemet membrane endothelial keratoplasty: safety and outcomes: a report by the American academy of ophthalmology. Ophthalmology 125, 295–310. Detry, B., Blacher, S., Erpicum, C., Paupert, J., Maertens, L., Maillard, C., Munaut, C., Sounni, N.E., Lambert, V., Foidart, J.M., Rakic, J.M., Cataldo, D., Noel, A., 2013. Sunitinib inhibits inflammatory corneal lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 54, 3082–3093. Dhaliwal, J.S., Mason, B.F., Kaufman, S.C., 2008. Long-term use of topical tacrolimus (FK506) in high-risk penetrating keratoplasty. Cornea 27, 488–493. Dietrich, T., Bock, F., Yuen, D., Hos, D., Bachmann, B.O., Zahn, G., Wiegand, S., Chen, L., Cursiefen, C., 2010. Cutting edge: lymphatic vessels, not blood vessels, primarily mediate immune rejections after transplantation. J. Immunol. 184, 535–539. Dietrich, T., Onderka, J., Bock, F., Kruse, F.E., Vossmeyer, D., Stragies, R., Zahn, G., Cursiefen, C., 2007. Inhibition of inflammatory lymphangiogenesis by integrin alpha5 blockade. Am. J. Pathol. 171, 361–372. Dilioglou, S., Cruse, J.M., Lewis, R.E., 2003. Function of CD80 and CD86 on monocyteand stem cell-derived dendritic cells. Exp. Mol. Pathol. 75, 217–227. Dobrowolski, D., Orzechowska-Wylegala, B., Wowra, B., Wroblewska-Czajka, E., Grolik, M., Szczubialka, K., Nowakowska, M., Puzzolo, D., Wylegala, E.A., Micali, A., Aragona, P., 2015. Cultivated oral mucosa epithelium in ocular surface reconstruction in aniridia patients. BioMed Res. Int. 2015, 281870. Dohlman, T.H., Omoto, M., Hua, J., Stevenson, W., Lee, S.M., Chauhan, S.K., Dana, R., 2015. VEGF-trap aflibercept significantly improves long-term graft survival in highrisk corneal transplantation. Transplantation 99, 678–686. Droutsas, K., Lazaridis, A., Papaconstantinou, D., Brouzas, D., Moschos, M.M., Schulze, S., Sekundo, W., 2016. Visual outcomes after descemet membrane endothelial keratoplasty versus descemet stripping automated endothelial keratoplasty-comparison of specific matched pairs. Cornea 35, 765–771. Du, H.T., Du, L.L., Tang, X.L., Ge, H.Y., Liu, P., 2017. Blockade of MMP-2 and MMP-9 inhibits corneal lymphangiogenesis. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 255, 1573–1579. Dua, H.S., Faraj, L.A., Said, D.G., Gray, T., Lowe, J., 2013. Human corneal anatomy redefined: a novel pre-Descemet's layer (Dua's layer). Ophthalmology 120, 1778–1785. Egeblad, M., Werb, Z., 2002. New functions for the matrix metalloproteinases in cancer progression. Nat. Rev. Canc. 2, 161–174. Eghrari, A.O., Riazuddin, S.A., Gottsch, J.D., 2015. Overview of the cornea: structure, function, and development. Prog. Mol. Biol. Transl. Sci. 134, 7–23. Elbaz, U., Mireskandari, K., Shen, C., Ali, A., 2015. Corneal fine needle diathermy with adjuvant bevacizumab to treat corneal neovascularization in children. Cornea 34,

26

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Hock, B.D., Kato, M., McKenzie, J.L., Hart, D.N., 2001. A soluble form of CD83 is released from activated dendritic cells and B lymphocytes, and is detectable in normal human sera. Int. Immunol. 13, 959–967. Hoerster, R., Stanzel, T.P., Bachmann, B.O., Siebelmann, S., Felsch, M., Cursiefen, C., 2016. Intensified topical steroids as prophylaxis for macular edema after posterior lamellar keratoplasty combined with cataract surgery. Am. J. Ophthalmol. 163, 174–179 e172. Hoffmann, F., Wiederholt, M., 1986. Topical cyclosporin A in the treatment of corneal graft reaction. Cornea 5, 129. Holzer, M.P., Solomon, K.D., Vroman, D.T., Sandoval, H.P., Margaron, P., Kasper, T.J., Crosson, C.E., 2003. Photodynamic therapy with verteporfin in a rabbit model of corneal neovascularization. Investig. Ophthalmol. Vis. Sci. 44, 2954–2958. Hong, Y.K., Detmar, M., 2003. Prox1, master regulator of the lymphatic vasculature phenotype. Cell Tissue Res. 314, 85–92. Hong, Y.K., Harvey, N., Noh, Y.H., Schacht, V., Hirakawa, S., Detmar, M., Oliver, G., 2002. Prox1 is a master control gene in the program specifying lymphatic endothelial cell fate. Dev. Dynam. 225, 351–357. Hori, J., Streilein, J.W., 2001. Role of recipient epithelium in promoting survival of orthotopic corneal allografts in mice. Investig. Ophthalmol. Vis. Sci. 42, 720–726. Horie, S., Sugita, S., Futagami, Y., Kawaguchi, T., Kamoi, K., Shirato, S., Mochizuki, M., 2009. Human iris pigment epithelium suppresses activation of bystander T cells via TGFbeta-TGFbeta receptor interaction. Exp. Eye Res. 88, 1033–1042. Horstmann, J., Schulz-Hildebrandt, H., Bock, F., Siebelmann, S., Lankenau, E., Huttmann, G., Steven, P., Cursiefen, C., 2017. Label-free in vivo imaging of corneal lymphatic vessels using microscopic optical coherence tomography. Investig. Ophthalmol. Vis. Sci. 58, 5880–5886. Hos, D., Bock, F., Dietrich, T., Onderka, J., Kruse, F.E., Thierauch, K.H., Cursiefen, C., 2008. Inflammatory corneal (lymph)angiogenesis is blocked by VEGFR-tyrosine kinase inhibitor ZK 261991, resulting in improved graft survival after corneal transplantation. Investig. Ophthalmol. Vis. Sci. 49, 1836–1842. Hos, D., Bucher, F., Regenfuss, B., Dreisow, M.L., Bock, F., Heindl, L.M., Eming, S.A., Cursiefen, C., 2016a. IL-10 indirectly regulates corneal lymphangiogenesis and resolution of inflammation via macrophages. Am. J. Pathol. 186, 159–171. Hos, D., Bukowiecki, A., Horstmann, J., Bock, F., Bucher, F., Heindl, L.M., Siebelmann, S., Steven, P., Dana, R., Eming, S.A., Cursiefen, C., 2017a. Transient ingrowth of lymphatic vessels into the physiologically avascular cornea regulates corneal edema and transparency. Sci. Rep (in press). Hos, D., Dorrie, J., Schaft, N., Bock, F., Notara, M., Kruse, F.E., Krautwald, S., Cursiefen, C., Bachmann, B.O., 2016b. Blockade of CCR7 leads to decreased dendritic cell migration to draining lymph nodes and promotes graft survival in low-risk corneal transplantation. Exp. Eye Res. 146, 1–6. Hos, D., Heindl, L.M., Bucher, F., Cursiefen, C., 2014. Evidence of donor corneal endothelial cell migration from immune reactions occurring after descemet membrane endothelial keratoplasty. Cornea 33, 331–334. Hos, D., Le, V.N.H., Hellmich, M., Siebelmann, S., Roters, S., Bachmann, B.O., Cursiefen, C., 2019a. Risk of corneal graft rejection after high-risk keratoplasty following fineneedle vessel coagulation of corneal neovascularization combined with bevacizumab: a pilot study. Transplant. Dir. 5, e452. Hos, D., Regenfuss, B., Bock, F., Onderka, J., Cursiefen, C., 2011a. Blockade of insulin receptor substrate-1 inhibits corneal lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 52, 5778–5785. Hos, D., Saban, D.R., Bock, F., Regenfuss, B., Onderka, J., Masli, S., Cursiefen, C., 2011b. Suppression of inflammatory corneal lymphangiogenesis by application of topical corticosteroids. Arch. Ophthalmol. 129, 445–452. Hos, D., Schaub, F., Cursiefen, C., 2019b. Does anterior chamber-associated immune deviation (ACAID) play a role in posterior lamellar keratoplasty? Case report of a splenectomized patient. BMC Ophthalmol. 19, 100. Hos, D., Schlereth, S.L., Bock, F., Heindl, L.M., Cursiefen, C., 2015. Antilymphangiogenic therapy to promote transplant survival and to reduce cancer metastasis: what can we learn from the eye? Semin. Cell Dev. Biol. 38, 117–130. Hos, D., Tuac, O., Schaub, F., Stanzel, T.P., Schrittenlocher, S., Hellmich, M., Bachmann, B.O., Cursiefen, C., 2017b. Incidence and clinical course of immune reactions after descemet membrane endothelial keratoplasty: retrospective analysis of 1000 consecutive eyes. Ophthalmology 124, 512–518. Hou, Y., Le, V.N.H., Clahsen, T., Schneider, A.C., Bock, F., Cursiefen, C., 2017. Photodynamic therapy leads to time-dependent regression of pathologic corneal (lymph) angiogenesis and promotes high-risk corneal allograft survival. Investig. Ophthalmol. Vis. Sci. 58, 5862–5869. Hou, Y., Le, V.N.H., Toth, G., Siebelmann, S., Horstmann, J., Gabriel, T., Bock, F., Cursiefen, C., 2018. UV light crosslinking regresses mature corneal blood and lymphatic vessels and promotes subsequent high-risk corneal transplant survival. Am. J. Transplant. 18, 2873–2884. Huq, S., Liu, Y., Benichou, G., Dana, M.R., 2004. Relevance of the direct pathway of sensitization in corneal transplantation is dictated by the graft bed microenvironment. J. Immunol. 173, 4464–4469. Hussain, R., Savant, V., 2017. Fine-needle diathermy with simultaneous subconjunctival bevacizumab. Semin. Ophthalmol. 32, 550–552. Inoue, K., Amano, S., Oshika, T., Tsuru, T., 2001. Risk factors for corneal graft failure and rejection in penetrating keratoplasty. Acta Ophthalmol. Scand. 79, 251–255. Iovieno, A., Neri, A., Soldani, A.M., Adani, C., Fontana, L., 2017. Descemetorhexis without graft placement for the treatment of Fuchs endothelial dystrophy: preliminary results and review of the literature. Cornea 36, 637–641. Ishino, Y., Sano, Y., Nakamura, T., Connon, C.J., Rigby, H., Fullwood, N.J., Kinoshita, S., 2004. Amniotic membrane as a carrier for cultivated human corneal endothelial cell transplantation. Investig. Ophthalmol. Vis. Sci. 45, 800–806. Jeong, H.S., Yun, J.H., Lee, D.H., Lee, E.H., Cho, C.H., 2019. Retinal pigment epithelium-

Ophthalmology 122, 1757–1764. Guerra, F.P., Anshu, A., Price, M.O., Giebel, A.W., Price, F.W., 2011. Descemet's membrane endothelial keratoplasty: prospective study of 1-year visual outcomes, graft survival, and endothelial cell loss. Ophthalmology 118, 2368–2373. Gupta, D., Illingworth, C., 2011. Treatments for corneal neovascularization: a review. Cornea 30, 927–938. Gupta, N., Farooqui, J.H., Patel, N., Mathur, U., 2018. Early results of penetrating keratoplasty in patients with unilateral chemical injury after simple limbal epithelial transplantation. Cornea 37, 1249–1254. Hahn, A.B., Foulks, G.N., Enger, C., Fink, N., Stark, W.J., Hopkins, K.A., Sanfilippo, F., 1995. The association of lymphocytotoxic antibodies with corneal allograft rejection in high risk patients. Transplantation 59, 21–27 The Collaborative Corneal Transplantation Studies Research Group. Hajrasouliha, A.R., Funaki, T., Sadrai, Z., Hattori, T., Chauhan, S.K., Dana, R., 2012. Vascular endothelial growth factor-C promotes alloimmunity by amplifying antigenpresenting cell maturation and lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 53, 1244–1250. Ham, L., Dapena, I., Liarakos, V.S., Baydoun, L., van Dijk, K., Ilyas, A., Oellerich, S., Melles, G.R., 2016. Midterm results of descemet membrane endothelial keratoplasty: 4 to 7 Years clinical outcome. Am. J. Ophthalmol. 171, 113–121. Hamrah, P., Huq, S.O., Liu, Y., Zhang, Q., Dana, M.R., 2003a. Corneal immunity is mediated by heterogeneous population of antigen-presenting cells. J. Leukoc. Biol. 74, 172–178. Hamrah, P., Liu, Y., Zhang, Q., Dana, M.R., 2003b. The corneal stroma is endowed with a significant number of resident dendritic cells. Investig. Ophthalmol. Vis. Sci. 44, 581–589. Hamrah, P., Zhang, Q., Liu, Y., Dana, M.R., 2002. Novel characterization of MHC class IInegative population of resident corneal Langerhans cell-type dendritic cells. Investig. Ophthalmol. Vis. Sci. 43, 639–646. Hamzaoglu, E.C., Straiko, M.D., Mayko, Z.M., Sales, C.S., Terry, M.A., 2015. The first 100 eyes of standardized descemet stripping automated endothelial keratoplasty versus standardized descemet membrane endothelial keratoplasty. Ophthalmology 122, 2193–2199. Han, Y., Shao, Y., Liu, T., Qu, Y.L., Li, W., Liu, Z., 2015. Therapeutic effects of topical netrin-4 inhibits corneal neovascularization in alkali-burn rats. PLoS One 10, e0122951. Harding, S., 2001. Photodynamic therapy in the treatment of subfoveal choroidal neovascularisation. Eye 15, 407–412. Hargrave, S.L., Hay, C., Mellon, J., Mayhew, E., Niederkorn, J.Y., 2004. Fate of MHCmatched corneal allografts in Th1-deficient hosts. Investig. Ophthalmol. Vis. Sci. 45, 1188–1193. Hassell, J.R., Birk, D.E., 2010. The molecular basis of corneal transparency. Exp. Eye Res. 91, 326–335. Hegde, S., Beauregard, C., Mayhew, E., Niederkorn, J.Y., 2005. CD4(+) T-cell-mediated mechanisms of corneal allograft rejection: role of Fas-induced apoptosis. Transplantation 79, 23–31. Heindl, L.M., Riss, S., Adler, W., Bucher, F., Hos, D., Cursiefen, C., 2013. Split cornea transplantation: relationship between storage time of split donor tissue and outcome. Ophthalmology 120, 899–907. Heindl, L.M., Riss, S., Bachmann, B.O., Laaser, K., Kruse, F.E., Cursiefen, C., 2011a. Split cornea transplantation for 2 recipients: a new strategy to reduce corneal tissue cost and shortage. Ophthalmology 118, 294–301. Heindl, L.M., Riss, S., Laaser, K., Bachmann, B.O., Kruse, F.E., Cursiefen, C., 2011b. Split cornea transplantation for 2 recipients - review of the first 100 consecutive patients. Am. J. Ophthalmol. 152, 523–532 e522. Heinzelmann, S., Bohringer, D., Eberwein, P., Lapp, T., Reinhard, T., Maier, P., 2017. Descemet membrane endothelial keratoplasty for graft failure following penetrating keratoplasty. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 255, 979–985. Heinzelmann, S., Maier, P., Bohringer, D., Huther, S., Eberwein, P., Reinhard, T., 2015. Cystoid macular oedema following Descemet membrane endothelial keratoplasty. Br. J. Ophthalmol. 99, 98–102. Heishi, T., Hosaka, T., Suzuki, Y., Miyashita, H., Oike, Y., Takahashi, T., Nakamura, T., Arioka, S., Mitsuda, Y., Takakura, T., Hojo, K., Matsumoto, M., Yamauchi, C., Ohta, H., Sonoda, H., Sato, Y., 2010. Endogenous angiogenesis inhibitor vasohibin1 exhibits broad-spectrum antilymphangiogenic activity and suppresses lymph node metastasis. Am. J. Pathol. 176, 1950–1958. Heldin, C.H., Westermark, B., 1999. Mechanism of action and in vivo role of plateletderived growth factor. Physiol. Rev. 79, 1283–1316. Henderson, B.W., Dougherty, T.J., 1992. How does photodynamic therapy work? Photochem. Photobiol. 55, 145–157. Hill, J.C., 1989. The use of cyclosporine in high-risk keratoplasty. Am. J. Ophthalmol. 107, 506–510. Hill, J.C., 1994. Systemic cyclosporine in high-risk keratoplasty. Short- versus long-term therapy. Ophthalmology 101, 128–133. Hill, J.C., 1995. Systemic cyclosporine in high-risk keratoplasty: long-term results. Eye 9 (Pt 4), 422–428. Hjortdal, J., Ehlers, N., 2009. Descemet's stripping automated endothelial keratoplasty and penetrating keratoplasty for Fuchs' endothelial dystrophy. Acta Ophthalmol. 87, 310–314. Hock, B.D., Fernyhough, L.J., Gough, S.M., Steinkasserer, A., Cox, A.G., McKenzie, J.L., 2009. Release and clinical significance of soluble CD83 in chronic lymphocytic leukemia. Leuk. Res. 33, 1089–1095. Hock, B.D., Haring, L.F., Steinkasserer, A., Taylor, K.G., Patton, W.N., McKenzie, J.L., 2004. The soluble form of CD83 is present at elevated levels in a number of hematological malignancies. Leuk. Res. 28, 237–241.

27

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Kubaloglu, A., Sari, E.S., Unal, M., Koytak, A., Kurnaz, E., Cinar, Y., Ozerturk, Y., 2011. Long-term results of deep anterior lamellar keratoplasty for the treatment of keratoconus. Am. J. Ophthalmol. 151, 760–767 e761. Kuchle, M., Cursiefen, C., Nguyen, N.X., Langenbucher, A., Seitz, B., Wenkel, H., Martus, P., Naumann, G.O., 2002. Risk factors for corneal allograft rejection: intermediate results of a prospective normal-risk keratoplasty study. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 240, 580–584. Kuffova, L., Knickelbein, J.E., Yu, T., Medina, C., Amescua, G., Rowe, A.M., Hendricks, R.L., Forrester, J.V., 2016. High-risk corneal graft rejection in the setting of previous corneal herpes simplex virus (HSV)-1 infection. Investig. Ophthalmol. Vis. Sci. 57, 1578–1587. Kuffova, L., Lumsden, L., Vesela, V., Taylor, J.A., Filipec, M., Holan, V., Dick, A.D., Forrester, J.V., 2001. Kinetics of leukocyte and myeloid cell traffic in the murine corneal allograft response. Transplantation 72, 1292–1298. Lam, F.C., Baydoun, L., Dirisamer, M., Lie, J., Dapena, I., Melles, G.R., 2014. HemiDescemet membrane endothelial keratoplasty transplantation: a potential method for increasing the pool of endothelial graft tissue. JAMA Ophthalmol. 132, 1469–1473. Lan, Z., Lian, D., Liu, W., Arp, J., Charlton, B., Ge, W., Brand, S., Healey, D., DeBenedette, M., Nicolette, C., Garcia, B., Wang, H., 2010. Prevention of chronic renal allograft rejection by soluble CD83. Transplantation 90, 1278–1285. Le, V.N.H., Hou, Y., Horstmann, J., Bock, F., Cursiefen, C., 2018a. Novel method to detect corneal lymphatic vessels in vivo by intrastromal injection of fluorescein. Cornea 37, 267–271. Le, V.N.H., Schneider, A.C., Scholz, R., Bock, F., Cursiefen, C., 2018b. Fine needle-diathermy regresses pathological corneal (Lymph)Angiogenesis and promotes high-risk corneal transplant survival. Sci. Rep. 8, 5707. Lechmann, M., Krooshoop, D.J., Dudziak, D., Kremmer, E., Kuhnt, C., Figdor, C.G., Schuler, G., Steinkasserer, A., 2001. The extracellular domain of CD83 inhibits dendritic cell-mediated T cell stimulation and binds to a ligand on dendritic cells. J. Exp. Med. 194, 1813–1821. Lee, J.G., Kay, E.P., 2012. NF-kappaB is the transcription factor for FGF-2 that causes endothelial mesenchymal transformation in cornea. Investig. Ophthalmol. Vis. Sci. 53, 1530–1538. Lee, W.B., Shtein, R.M., Kaufman, S.C., Deng, S.X., Rosenblatt, M.I., 2015. Boston keratoprosthesis: outcomes and complications: a report by the American academy of ophthalmology. Ophthalmology 122, 1504–1511. Li, J.Y., Terry, M.A., Goshe, J., Davis-Boozer, D., Shamie, N., 2012a. Three-year visual acuity outcomes after Descemet's stripping automated endothelial keratoplasty. Ophthalmology 119, 1126–1129. Li, J.Y., Terry, M.A., Goshe, J., Shamie, N., Davis-Boozer, D., 2012b. Graft rejection after Descemet's stripping automated endothelial keratoplasty: graft survival and endothelial cell loss. Ophthalmology 119, 90–94. Li, Y.N., Pinzon-Duarte, G., Dattilo, M., Claudepierre, T., Koch, M., Brunken, W.J., 2012c. The expression and function of netrin-4 in murine ocular tissues. Exp. Eye Res. 96, 24–35. Li, Z.R., Li, Y.P., Lin, M.L., Su, W.R., Zhang, W.X., Zhang, Y., Yao, L., Liang, D., 2012d. Activated macrophages induce neovascularization through upregulation of MMP-9 and VEGF in rat corneas. Cornea 31, 1028–1035. Lipp, M., Bucher, F., Parthasarathy, A., Hos, D., Onderka, J., Cursiefen, C., Bock, F., 2014. Blockade of the VEGF isoforms in inflammatory corneal hemangiogenesis and lymphangiogenesis. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 252, 943–949. Liu, Y., Hamrah, P., Zhang, Q., Taylor, A.W., Dana, M.R., 2002. Draining lymph nodes of corneal transplant hosts exhibit evidence for donor major histocompatibility complex (MHC) class II-positive dendritic cells derived from MHC class II-negative grafts. J. Exp. Med. 195, 259–268. Luznik, Z., Oellerich, S., Roesch, K., Yin, J., Zumbansen, M., Franken, L., Melles, G.R.J., Dana, R., 2019. Descemet membrane endothelial keratoplasty failure associated with innate immune activation. Ophthalmology. https://doi.org/10.1016/j.ophtha.2019. 04.016. In press. Maguire, M.G., Stark, W.J., Gottsch, J.D., Stulting, R.D., Sugar, A., Fink, N.E., Schwartz, A., 1994. Risk factors for corneal graft failure and rejection in the collaborative corneal transplantation studies. Collaborative Corneal Transplantation Studies Research Group. Ophthalmology 101, 1536–1547. Maier, A.B., Klein, S., Kociok, N., Riechardt, A.I., Gundlach, E., Reichhart, N., Strauss, O., Joussen, A.M., 2017. Netrin-4 mediates corneal hemangiogenesis but not lymphangiogenesis in the mouse-model of suture-induced neovascularization. Investig. Ophthalmol. Vis. Sci. 58, 1387–1396. Makinen, T., Jussila, L., Veikkola, T., Karpanen, T., Kettunen, M.I., Pulkkanen, K.J., Kauppinen, R., Jackson, D.G., Kubo, H., Nishikawa, S., Yla-Herttuala, S., Alitalo, K., 2001. Inhibition of lymphangiogenesis with resulting lymphedema in transgenic mice expressing soluble VEGF receptor-3. Nat. Med. 7, 199–205. Mannis, M.J., Krachmer, J.H., 1981. Keratoplasty: a historical perspective. Surv. Ophthalmol. 25, 333–338. Marsland, B.J., Battig, P., Bauer, M., Ruedl, C., Lassing, U., Beerli, R.R., Dietmeier, K., Ivanova, L., Pfister, T., Vogt, L., Nakano, H., Nembrini, C., Saudan, P., Kopf, M., Bachmann, M.F., 2005. CCL19 and CCL21 induce a potent proinflammatory differentiation program in licensed dendritic cells. Immunity 22, 493–505. Maruyama, K., Asai, J., Ii, M., Thorne, T., Losordo, D.W., D'Amore, P.A., 2007. Decreased macrophage number and activation lead to reduced lymphatic vessel formation and contribute to impaired diabetic wound healing. Am. J. Pathol. 170, 1178–1191. Maruyama, K., Ii, M., Cursiefen, C., Jackson, D.G., Keino, H., Tomita, M., Van Rooijen, N., Takenaka, H., D'Amore, P.A., Stein-Streilein, J., Losordo, D.W., Streilein, J.W., 2005. Inflammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages. J. Clin. Investig. 115, 2363–2372. Maruyama, K., Nakazawa, T., Cursiefen, C., Maruyama, Y., Van Rooijen, N., D'Amore,

derived transforming growth factor-beta2 inhibits the angiogenic response of endothelial cells by decreasing vascular endothelial growth factor receptor-2 expression. J. Cell. Physiol. 234, 3837–3849. Jiang, X., Nicolls, M.R., Tian, W., Rockson, S.G., 2018. Lymphatic dysfunction, leukotrienes, and lymphedema. Annu. Rev. Physiol. 80, 49–70. Jitariu, A.A., Cimpean, A.M., Kundnani, N.R., Raica, M., 2015. Platelet-derived growth factors induced lymphangiogenesis: evidence, unanswered questions and upcoming challenges. Arch. Med. Sci. 11, 57–66. Johnson, L.A., Banerji, S., Lawrance, W., Gileadi, U., Prota, G., Holder, K.A., Roshorm, Y.M., Hanke, T., Cerundolo, V., Gale, N.W., Jackson, D.G., 2017. Dendritic cells enter lymph vessels by hyaluronan-mediated docking to the endothelial receptor LYVE-1. Nat. Immunol. 18, 762–770. Johnson, L.A., Clasper, S., Holt, A.P., Lalor, P.F., Baban, D., Jackson, D.G., 2006. An inflammation-induced mechanism for leukocyte transmigration across lymphatic vessel endothelium. J. Exp. Med. 203, 2763–2777. Jonas, J.B., Rank, R.M., Budde, W.M., 2002. Immunologic graft reactions after allogenic penetrating keratoplasty. Am. J. Ophthalmol. 133, 437–443. Jones, D., Li, Y., He, Y., Xu, Z., Chen, H., Min, W., 2012. Mirtron microRNA-1236 inhibits VEGFR-3 signaling during inflammatory lymphangiogenesis. Arterioscler. Thromb. Vasc. Biol. 32, 633–642. Jordan, C.S., Price, M.O., Trespalacios, R., Price Jr., F.W., 2009. Graft rejection episodes after Descemet stripping with endothelial keratoplasty: part one: clinical signs and symptoms. Br. J. Ophthalmol. 93, 387–390. Joseph, A., Raj, D., Shanmuganathan, V., Powell, R.J., Dua, H.S., 2007. Tacrolimus immunosuppression in high-risk corneal grafts. Br. J. Ophthalmol. 91, 51–55. Joyce, N.C., Harris, D.L., Mello, D.M., 2002. Mechanisms of mitotic inhibition in corneal endothelium: contact inhibition and TGF-beta2. Investig. Ophthalmol. Vis. Sci. 43, 2152–2159. Kajiya, K., Hirakawa, S., Ma, B., Drinnenberg, I., Detmar, M., 2005. Hepatocyte growth factor promotes lymphatic vessel formation and function. EMBO J. 24, 2885–2895. Kanellopoulos, A.J., Asimellis, G., 2014. Long-term safety and efficacy of high-fluence collagen crosslinking of the vehicle cornea in Boston keratoprosthesis type 1. Cornea 33, 914–918. Kang, G.J., Truong, T., Huang, E., Su, V., Ge, S., Chen, L., 2016. Integrin alpha 9 blockade suppresses lymphatic valve formation and promotes transplant survival. Investig. Ophthalmol. Vis. Sci. 57, 5935–5939. Kataru, R.P., Jung, K., Jang, C., Yang, H., Schwendener, R.A., Baik, J.E., Han, S.H., Alitalo, K., Koh, G.Y., 2009. Critical role of CD11b+ macrophages and VEGF in inflammatory lymphangiogenesis, antigen clearance, and inflammation resolution. Blood 113, 5650–5659. Kazenwadel, J., Michael, M.Z., Harvey, N.L., 2010. Prox1 expression is negatively regulated by miR-181 in endothelial cells. Blood 116, 2395–2401. Kerjaschki, D., Huttary, N., Raab, I., Regele, H., Bojarski-Nagy, K., Bartel, G., Krober, S.M., Greinix, H., Rosenmaier, A., Karlhofer, F., Wick, N., Mazal, P.R., 2006. Lymphatic endothelial progenitor cells contribute to de novo lymphangiogenesis in human renal transplants. Nat. Med. 12, 230–234. Khan, S.N., Shiakolas, P.S., Mootha, V.V., 2015. Descemet's stripping automated endothelial keratoplasty tissue insertion devices. J. Ophthalmic Vis. Res. 10, 461–468. Khodadoust, A.A., Silverstein, A.M., 1969. Transplantation and rejection of individual cell layers of the cornea. Investig. Ophthalmol. 8, 180–195. Kim, P., Yeung, S.N., Lichtinger, A., Amiran, M.D., Shanmugam, S.V., Iovieno, A., Slomovic, A.R., Rootman, D.S., 2012. Outcomes of repeat endothelial keratoplasty in patients with failed Descemet stripping endothelial keratoplasty. Cornea 31, 1154–1157. Kinoshita, S., Koizumi, N., Ueno, M., Okumura, N., Imai, K., Tanaka, H., Yamamoto, Y., Nakamura, T., Inatomi, T., Bush, J., Toda, M., Hagiya, M., Yokota, I., Teramukai, S., Sotozono, C., Hamuro, J., 2018. Injection of cultured cells with a ROCK inhibitor for bullous keratopathy. N. Engl. J. Med. 378, 995–1003. Koenig, S.B., 2015. Planned descemetorhexis without endothelial keratoplasty in eyes with Fuchs corneal endothelial dystrophy. Cornea 34, 1149–1151. Koenig, Y., Bock, F., Horn, F., Kruse, F., Straub, K., Cursiefen, C., 2009. Short- and longterm safety profile and efficacy of topical bevacizumab (Avastin) eye drops against corneal neovascularization. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 247, 1375–1382. Koenig, Y., Bock, F., Kruse, F.E., Stock, K., Cursiefen, C., 2012. Angioregressive pretreatment of mature corneal blood vessels before keratoplasty: fine-needle vessel coagulation combined with anti-VEGFs. Cornea 31, 887–892. Koizumi, N., Sakamoto, Y., Okumura, N., Okahara, N., Tsuchiya, H., Torii, R., Cooper, L.J., Ban, Y., Tanioka, H., Kinoshita, S., 2007. Cultivated corneal endothelial cell sheet transplantation in a primate model. Investig. Ophthalmol. Vis. Sci. 48, 4519–4526. Komai, Y., Ushiki, T., 1991. The three-dimensional organization of collagen fibrils in the human cornea and sclera. Investig. Ophthalmol. Vis. Sci. 32, 2244–2258. Kretschmer, B., Luthje, K., Schneider, S., Fleischer, B., Breloer, M., 2009. Engagement of CD83 on B cells modulates B cell function in vivo. J. Immunol. 182, 2827–2834. Kruse, F.E., Schrehardt, U.S., Tourtas, T., 2014. Optimizing outcomes with Descemet's membrane endothelial keratoplasty. Curr. Opin. Ophthalmol. 25, 325–334. Krzyzak, L., Seitz, C., Urbat, A., Hutzler, S., Ostalecki, C., Glasner, J., Hiergeist, A., Gessner, A., Winkler, T.H., Steinkasserer, A., Nitschke, L., 2016. CD83 modulates B cell activation and germinal center responses. J. Immunol. 196, 3581–3594. Ksander, B.R., Bando, Y., Acevedo, J., Streilein, J.W., 1991. Infiltration and accumulation of precursor cytotoxic T-cells increase with time in progressively growing ocular tumors. Cancer Res. 51, 3153–3158. Ksander, B.R., Streilein, J.W., 1990. Failure of infiltrating precursor cytotoxic T cells to acquire direct cytotoxic function in immunologically privileged sites. J. Immunol. 145, 2057–2063.

28

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Oh, S., Seo, M., Choi, J.S., Joo, C.K., Lee, S.K., 2018. MiR-199a/b-5p inhibits lymphangiogenesis by targeting discoidin domain receptor 1 in corneal injury. Mol. Cells 41, 93–102. Oh, S.J., Jeltsch, M.M., Birkenhager, R., McCarthy, J.E., Weich, H.A., Christ, B., Alitalo, K., Wilting, J., 1997. VEGF and VEGF-C: specific induction of angiogenesis and lymphangiogenesis in the differentiated avian chorioallantoic membrane. Dev. Biol. 188, 96–109. Ohl, L., Mohaupt, M., Czeloth, N., Hintzen, G., Kiafard, Z., Zwirner, J., Blankenstein, T., Henning, G., Forster, R., 2004. CCR7 governs skin dendritic cell migration under inflammatory and steady-state conditions. Immunity 21, 279–288. Oka, M., Iwata, C., Suzuki, H.I., Kiyono, K., Morishita, Y., Watabe, T., Komuro, A., Kano, M.R., Miyazono, K., 2008. Inhibition of endogenous TGF-beta signaling enhances lymphangiogenesis. Blood 111, 4571–4579. Okumura, N., Kakutani, K., Numata, R., Nakahara, M., Schlotzer-Schrehardt, U., Kruse, F., Kinoshita, S., Koizumi, N., 2015. Laminin-511 and -521 enable efficient in vitro expansion of human corneal endothelial cells. Investig. Ophthalmol. Vis. Sci. 56, 2933–2942. Okumura, N., Koizumi, N., Ueno, M., Sakamoto, Y., Takahashi, H., Tsuchiya, H., Hamuro, J., Kinoshita, S., 2012. ROCK inhibitor converts corneal endothelial cells into a phenotype capable of regenerating in vivo endothelial tissue. Am. J. Pathol. 181, 268–277. Okumura, N., Sakamoto, Y., Fujii, K., Kitano, J., Nakano, S., Tsujimoto, Y., Nakamura, S., Ueno, M., Hagiya, M., Hamuro, J., Matsuyama, A., Suzuki, S., Shiina, T., Kinoshita, S., Koizumi, N., 2016. Rho kinase inhibitor enables cell-based therapy for corneal endothelial dysfunction. Sci. Rep. 6, 26113. Okumura, N., Ueno, M., Koizumi, N., Sakamoto, Y., Hirata, K., Hamuro, J., Kinoshita, S., 2009. Enhancement on primate corneal endothelial cell survival in vitro by a ROCK inhibitor. Investig. Ophthalmol. Vis. Sci. 50, 3680–3687. Olson, E.A., Tu, E.Y., Basti, S., 2012. Stromal rejection following deep anterior lamellar keratoplasty: implications for postoperative care. Cornea 31, 969–973. Osawa, H., Maruyama, K., Streilein, J.W., 2004. CD95 ligand expression on corneal epithelium and endothelium influences the fates of orthotopic and heterotopic corneal allografts in mice. Investig. Ophthalmol. Vis. Sci. 45, 1908–1915. Osei-Bempong, C., Figueiredo, F.C., Lako, M., 2013. The limbal epithelium of the eye–a review of limbal stem cell biology, disease and treatment. Bioessays : News Rev. Mol. Cell Dev. Biol. 35, 211–219. Otrock, Z.K., Makarem, J.A., Shamseddine, A.I., 2007. Vascular endothelial growth factor family of ligands and receptors: review. Blood Cells Mol. Dis. 38, 258–268. Panda, A., Vanathi, M., Kumar, A., Dash, Y., Priya, S., 2007. Corneal graft rejection. Surv. Ophthalmol. 52, 375–396. Pardoll, D.M., 2002. Spinning molecular immunology into successful immunotherapy. Nat. Rev. Immunol. 2, 227–238. Paunicka, K.J., Mellon, J., Robertson, D., Petroll, M., Brown, J.R., Niederkorn, J.Y., 2015. Severing corneal nerves in one eye induces sympathetic loss of immune privilege and promotes rejection of future corneal allografts placed in either eye. Am. J. Transplant. : Off. J. Am. Soc. Transplant. Am. Soc. Transplant Surg. 15, 1490–1501. Pellegrini, G., Ardigo, D., Milazzo, G., Iotti, G., Guatelli, P., Pelosi, D., De Luca, M., 2018. Navigating market authorization: the path holoclar took to become the first stem cell product approved in the European Union. Stem Cell. Transl. Med. 7, 146–154. Pellegrini, G., De Luca, M., 2014. Eyes on the prize: limbal stem cells and corneal restoration. Cell stem cell 15, 121–122. Pepper, M.S., Wasi, S., Ferrara, N., Orci, L., Montesano, R., 1994. In vitro angiogenic and proteolytic properties of bovine lymphatic endothelial cells. Exp. Cell Res. 210, 298–305. Peppicelli, S., Bianchini, F., Calorini, L., 2014. Inflammatory cytokines induce vascular endothelial growth factor-C expression in melanoma-associated macrophages and stimulate melanoma lymph node metastasis. Oncol Lett 8, 1133–1138. Peraza-Nieves, J., Baydoun, L., Dapena, I., Ilyas, A., Frank, L.E., Luceri, S., Ham, L., Oellerich, S., Melles, G.R.J., 2017. Two-year clinical outcome of 500 consecutive cases undergoing descemet membrane endothelial keratoplasty. Cornea 36, 655–660. Pillai, C.T., Dua, H.S., Hossain, P., 2000. Fine needle diathermy occlusion of corneal vessels. Investig. Ophthalmol. Vis. Sci. 41, 2148–2153. Podgrabinska, S., Kamalu, O., Mayer, L., Shimaoka, M., Snoeck, H., Randolph, G.J., Skobe, M., 2009. Inflamed lymphatic endothelium suppresses dendritic cell maturation and function via Mac-1/ICAM-1-dependent mechanism. J. Immunol. 183, 1767–1779. Poon, A.C., Forbes, J.E., Dart, J.K., Subramaniam, S., Bunce, C., Madison, P., Ficker, L.A., Tuft, S.J., Gartry, D.S., Buckley, R.J., 2001. Systemic cyclosporin A in high risk penetrating keratoplasties: a case-control study. Br. J. Ophthalmol. 85, 1464–1469. Prechtel, A.T., Turza, N.M., Theodoridis, A.A., Steinkasserer, A., 2007. CD83 knockdown in monocyte-derived dendritic cells by small interfering RNA leads to a diminished T cell stimulation. J. Immunol. 178, 5454–5464. Preclik, A., Langenbucher, A., Seitz, B., Cursiefen, C., 2010. [Long-term outcome after penetrating keratoplasty for keratoconus–impact of preoperative corneal curvature and best corrected visual acuity on the functional results]. Klin. Monatsblatter Augenheilkd. 227, 199–207. Price, D.A., Kelley, M., Price Jr., F.W., Price, M.O., 2018. Five-year graft survival of descemet membrane endothelial keratoplasty (EK) versus descemet stripping EK and the effect of donor sex matching. Ophthalmology 125, 1508–1514. Price Jr., F.W., Price, M.O., 2006. Descemet's stripping with endothelial keratoplasty in 200 eyes: early challenges and techniques to enhance donor adherence. J. Cataract Refract. Surg. 32, 411–418. Price, M.O., Baig, K.M., Brubaker, J.W., Price Jr., F.W., 2008. Randomized, prospective comparison of precut vs surgeon-dissected grafts for descemet stripping automated endothelial keratoplasty. Am. J. Ophthalmol. 146, 36–41. Price, M.O., Fairchild, K.M., Price, D.A., Price Jr., F.W., 2011. Descemet's stripping

P.A., Kinoshita, S., 2012. The maintenance of lymphatic vessels in the cornea is dependent on the presence of macrophages. Investig. Ophthalmol. Vis. Sci. 53, 3145–3153. Maruyama, Y., Maruyama, K., Kato, Y., Kajiya, K., Moritoh, S., Yamamoto, K., Matsumoto, Y., Sawane, M., Kerjaschki, D., Nakazawa, T., Kinoshita, S., 2014. The effect of podoplanin inhibition on lymphangiogenesis under pathological conditions. Investig. Ophthalmol. Vis. Sci. 55, 4813–4822. Matthaei, M., Bachmann, B., Siebelmann, S., Cursiefen, C., 2018. Technik der „Descemet membrane endothelial keratoplasty“ (DMEK): videobeitrag. Der Ophthalmologe. Z. Dtsch. Ophthalmol. Ges. 115, 778–784. McCawley, L.J., Matrisian, L.M., 2000. Matrix metalloproteinases: multifunctional contributors to tumor progression. Mol. Med. Today 6, 149–156. McLaren, J.W., Bachman, L.A., Kane, K.M., Patel, S.V., 2014. Objective assessment of the corneal endothelium in Fuchs' endothelial dystrophy. Investig. Ophthalmol. Vis. Sci. 55, 1184–1190. Medawar, P.B., 1948. Immunity to homologous grafted skin; the fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br. J. Exp. Pathol. 29, 58–69. Melles, G.R., Ong, T.S., Ververs, B., van der Wees, J., 2006. Descemet membrane endothelial keratoplasty (DMEK). Cornea 25, 987–990. Melles, G.R., Wijdh, R.H., Nieuwendaal, C.P., 2004. A technique to excise the descemet membrane from a recipient cornea (descemetorhexis). Cornea 23, 286–288. Mencucci, R., Marini, M., Paladini, I., Sarchielli, E., Sgambati, E., Menchini, U., Vannelli, G.B., 2010. Effects of riboflavin/UVA corneal cross-linking on keratocytes and collagen fibres in human cornea. Clin. Exp. Ophthalmol. 38, 49–56. Mimura, T., Yamagami, S., Yokoo, S., Usui, T., Tanaka, K., Hattori, S., Irie, S., Miyata, K., Araie, M., Amano, S., 2004. Cultured human corneal endothelial cell transplantation with a collagen sheet in a rabbit model. Investig. Ophthalmol. Vis. Sci. 45, 2992–2997. Mitry, D., Bhogal, M., Patel, A.K., Lee, B.S., Chai, S.M., Price, M.O., Price Jr., F.W., Jun, A.S., Aldave, A.J., Mehta, J.S., Busin, M., Allan, B.D., 2014. Descemet stripping automated endothelial keratoplasty after failed penetrating keratoplasty: survival, rejection risk, and visual outcome. JAMA Ophthalmol. 132, 742–749. Miyamoto, T., Sumioka, T., Saika, S., 2010. Endothelial mesenchymal transition: a therapeutic target in retrocorneal membrane. Cornea 29 (Suppl. 1), S52–S56. Moffatt, S.L., Cartwright, V.A., Stumpf, T.H., 2005. Centennial review of corneal transplantation. Clin. Exp. Ophthalmol. 33, 642–657. Monnereau, C., Bruinsma, M., Ham, L., Baydoun, L., Oellerich, S., Melles, G.R., 2014. Endothelial cell changes as an indicator for upcoming allograft rejection following descemet membrane endothelial keratoplasty. Am. J. Ophthalmol. 158, 485–495. Mukherjee, S., Mukherjee, U., 2009. A comprehensive review of immunosuppression used for liver transplantation. J. Transplant. 2009, 701464. Nguyen, N.X., Martus, P., Seitz, B., Cursiefen, C., 2009. Atopic dermatitis as a risk factor for graft rejection following normal-risk keratoplasty. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 247, 573–574. Nguyen, N.X., Seitz, B., Martus, P., Langenbucher, A., Cursiefen, C., 2007. Long-term topical steroid treatment improves graft survival following normal-risk penetrating keratoplasty. Am. J. Ophthalmol. 144, 318–319. Niederkorn, J., Streilein, J.W., Shadduck, J.A., 1981. Deviant immune responses to allogeneic tumors injected intracamerally and subcutaneously in mice. Investig. Ophthalmol. Vis. Sci. 20, 355–363. Niederkorn, J.Y., 1990. Immune privilege and immune regulation in the eye. Adv. Immunol. 48, 191–226. Niederkorn, J.Y., 2007. Immune mechanisms of corneal allograft rejection. Curr. Eye Res. 32, 1005–1016. Niederkorn, J.Y., Chen, P.W., Mellon, J., Stevens, C., Mayhew, E., 2009. Allergic airway hyperreactivity increases the risk for corneal allograft rejection. Am. J. Transplant. : Off. J. Am. Soc. Transplant. Am. Soc. Transplant Surg. 9, 1017–1026. Niederkorn, J.Y., Chen, P.W., Mellon, J., Stevens, C., Mayhew, E., 2010. Allergic conjunctivitis exacerbates corneal allograft rejection by activating Th1 and th2 alloimmune responses. J. Immunol. 184, 6076–6083. Niederkorn, J.Y., Mellon, J., 1996. Anterior chamber-associated immune deviation promotes corneal allograft survival. Investig. Ophthalmol. Vis. Sci. 37, 2700–2707. Noble, B.A., Agrawal, A., Collins, C., Saldana, M., Brogden, P.R., Zuberbuhler, B., 2007. Deep Anterior Lamellar Keratoplasty (DALK): visual outcome and complications for a heterogeneous group of corneal pathologies. Cornea 26, 59–64. Notara, M., Alatza, A., Gilfillan, J., Harris, A.R., Levis, H.J., Schrader, S., Vernon, A., Daniels, J.T., 2010. In sickness and in health: corneal epithelial stem cell biology, pathology and therapy. Exp. Eye Res. 90, 188–195. Notara, M., Daniels, J.T., 2008. Biological principals and clinical potentials of limbal epithelial stem cells. Cell Tissue Res. 331, 135–143. Notara, M., Lentzsch, A., Coroneo, M., Cursiefen, C., 2018. The role of limbal epithelial stem cells in regulating corneal (Lymph)angiogenic privilege and the micromilieu of the limbal niche following UV exposure. Stem Cell. Int. 2018, 8620172. Notara, M., Refaian, N., Braun, G., Steven, P., Bock, F., Cursiefen, C., 2015. Short-term uvb-irradiation leads to putative limbal stem cell damage and niche cell-mediated upregulation of macrophage recruiting cytokines. Stem Cell Res. 15, 643–654. Notara, M., Refaian, N., Braun, G., Steven, P., Bock, F., Cursiefen, C., 2016. Short-term ultraviolet a irradiation leads to dysfunction of the limbal niche cells and an antilymphangiogenic and anti-inflammatory micromilieu. Investig. Ophthalmol. Vis. Sci. 57, 928–939. Novak, N., Siepmann, K., Zierhut, M., Bieber, T., 2003. The good, the bad and the ugly– APCs of the eye. Trends Immunol. 24, 570–574. Nubile, M., Lanzini, M., Miri, A., Pocobelli, A., Calienno, R., Curcio, C., Mastropasqua, R., Dua, H.S., Mastropasqua, L., 2013. In vivo confocal microscopy in diagnosis of limbal stem cell deficiency. Am. J. Ophthalmol. 155, 220–232.

29

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al. endothelial keratoplasty five-year graft survival and endothelial cell loss. Ophthalmology 118, 725–729. Price, M.O., Feng, M.T., Scanameo, A., Price Jr., F.W., 2015. Loteprednol etabonate 0.5% gel vs. Prednisolone acetate 1% solution after descemet membrane endothelial keratoplasty: prospective randomized trial. Cornea 34, 853–858. Price, M.O., Giebel, A.W., Fairchild, K.M., Price Jr., F.W., 2009a. Descemet's membrane endothelial keratoplasty: prospective multicenter study of visual and refractive outcomes and endothelial survival. Ophthalmology 116, 2361–2368. Price, M.O., Gupta, P., Lass, J., Price Jr., F.W., 2017. EK (DLEK, DSEK, DMEK): new frontier in cornea surgery. Annu. Rev. Vis. Sci. 3, 69–90. Price, M.O., Jordan, C.S., Moore, G., Price Jr., F.W., 2009b. Graft rejection episodes after Descemet stripping with endothelial keratoplasty: part two: the statistical analysis of probability and risk factors. Br. J. Ophthalmol. 93, 391–395. Price, M.O., Price Jr., F.W., Kruse, F.E., Bachmann, B.O., Tourtas, T., 2014. Randomized comparison of topical prednisolone acetate 1% versus fluorometholone 0.1% in the first year after descemet membrane endothelial keratoplasty. Cornea 33, 880–886. Price, M.O., Scanameo, A., Feng, M.T., Price Jr., F.W., 2016. Descemet's membrane endothelial keratoplasty: risk of immunologic rejection episodes after discontinuing topical corticosteroids. Ophthalmology 123, 1232–1236. Quilendrino, R., Rodriguez-Calvo de Mora, M., Baydoun, L., Ham, L., van Dijk, K., Dapena, I., Oellerich, S., Melles, G.R.J., 2017. Prevention and management of descemet membrane endothelial keratoplasty complications. Cornea 36, 1089–1095. Raiskup, F., Spoerl, E., 2013. Corneal crosslinking with riboflavin and ultraviolet A. I. Principles. Ocul. Surf. 11, 65–74. Rama, P., Matuska, S., Paganoni, G., Spinelli, A., De Luca, M., Pellegrini, G., 2010. Limbal stem-cell therapy and long-term corneal regeneration. N. Engl. J. Med. 363, 147–155. Ramamoorthy, S., Cidlowski, J.A., 2016. Corticosteroids: mechanisms of action in health and disease. Rheum. Dis. Clin. N. Am. 42, 15–31 (vii). Randolph, D.A., Stephens, R., Carruthers, C.J., Chaplin, D.D., 1999. Cooperation between Th1 and Th2 cells in a murine model of eosinophilic airway inflammation. J. Clin. Investig. 104, 1021–1029. Ray, S., D'Amico, D.J., 2002. Pseudophakic cystoid macular edema. Semin. Ophthalmol. 17, 167–180. Reed, M.W., Miller, F.N., Wieman, T.J., Tseng, M.T., Pietsch, C.G., 1988. The effect of photodynamic therapy on the microcirculation. J. Surg. Res. 45, 452–459. Regenfuss, B., Bock, F., Cursiefen, C., 2012. Corneal angiogenesis and lymphangiogenesis. Curr. Opin. Allergy Clin. Immunol. 12, 548–554. Regenfuss, B., Dreisow, M.L., Hos, D., Masli, S., Bock, F., Cursiefen, C., 2015. The naive murine cornea as a model system to identify novel endogenous regulators of lymphangiogenesis: TRAIL and rtPA. Lymphatic Res. Biol. 13, 76–84. Regenfuss, B., Onderka, J., Bock, F., Hos, D., Maruyama, K., Cursiefen, C., 2010. Genetic heterogeneity of lymphangiogenesis in different mouse strains. Am. J. Pathol. 177, 501–510. Reimer, A., Langenbucher, A., Cursiefen, C., 2012. [Long-term outcome after penetrating keratoplasty for bullous keratopathy - influence of preoperative visual acuity on final outcome]. Klin. Monatsblatter Augenheilkd. 229, 149–157. Reinhard, T., Mayweg, S., Sokolovska, Y., Seitz, B., Mittelviefhaus, H., Engelmann, K., Voiculescu, A., Godehardt, E., Sundmacher, R., 2005. Systemic mycophenolate mofetil avoids immune reactions in penetrating high-risk keratoplasty: preliminary results of an ongoing prospectively randomized multicentre study. Transpl. Int. : Off. J. Eur. Soc. Organ Transplant. 18, 703–708. Reinhard, T., Moller, M., Sundmacher, R., 1999. Penetrating keratoplasty in patients with atopic dermatitis with and without systemic cyclosporin A. Cornea 18, 645–651. Reinhart, W.J., Musch, D.C., Jacobs, D.S., Lee, W.B., Kaufman, S.C., Shtein, R.M., 2011. Deep anterior lamellar keratoplasty as an alternative to penetrating keratoplasty a report by the american academy of ophthalmology. Ophthalmology 118, 209–218. Reinke, M.H., Canakis, C., Husain, D., Michaud, N., Flotte, T.J., Gragoudas, E.S., Miller, J.W., 1999. Verteporfin photodynamic therapy retreatment of normal retina and choroid in the cynomolgus monkey. Ophthalmology 106, 1915–1923. Reis, A., Reinhard, T., Voiculescu, A., Kutkuhn, B., Godehardt, E., Spelsberg, H., Althaus, C., Sundmacher, R., 1999. Mycophenolate mofetil versus cyclosporin A in high risk keratoplasty patients: a prospectively randomised clinical trial. Br. J. Ophthalmol. 83, 1268–1271. Reuer, T., Schneider, A.C., Cakir, B., Buhler, A.D., Walz, J.M., Lapp, T., Lange, C., Agostini, H., Schlunck, G., Cursiefen, C., Reinhard, T., Bock, F., Stahl, A., 2018. Semaphorin 3F modulates corneal lymphangiogenesis and promotes corneal graft survival. Investig. Ophthalmol. Vis. Sci. 59, 5277–5284. Reuten, R., Patel, T.R., McDougall, M., Rama, N., Nikodemus, D., Gibert, B., Delcros, J.G., Prein, C., Meier, M., Metzger, S., Zhou, Z., Kaltenberg, J., McKee, K.K., Bald, T., Tuting, T., Zigrino, P., Djonov, V., Bloch, W., Clausen-Schaumann, H., Poschl, E., Yurchenco, P.D., Ehrbar, M., Mehlen, P., Stetefeld, J., Koch, M., 2016. Structural decoding of netrin-4 reveals a regulatory function towards mature basement membranes. Nat. Commun. 7, 13515. Ribatti, D., 2009. Endogenous inhibitors of angiogenesis: a historical review. Leuk. Res. 33, 638–644. Riss, S., Heindl, L.M., Bachmann, B.O., Kruse, F.E., Cursiefen, C., 2012. Pentacam-based big bubble deep anterior lamellar keratoplasty in patients with keratoconus. Cornea 31, 627–632. Riss, S., Heindl, L.M., Bachmann, B.O., Kruse, F.E., Cursiefen, C., 2013. Microbubble incision as a new rescue technique for big-bubble deep anterior lamellar keratoplasty with failed bubble formation. Cornea 32, 125–129. Robert, M.C., Crnej, A., Shen, L.Q., Papaliodis, G.N., Dana, R., Foster, C.S., Chodosh, J., Dohlman, C.H., 2017. Infliximab after Boston keratoprosthesis in stevens-johnson syndrome: an update. Ocul. Immunol. Inflamm. 25, 413–417. Roberts, H.W., Maycock, N.J., O'Brart, D.P., 2016. Late stromal rejection in deep anterior lamellar keratoplasty: a case series. Cornea 35, 1179–1181.

Rodriguez-Calvo-de-Mora, M., Quilendrino, R., Ham, L., Liarakos, V.S., van Dijk, K., Baydoun, L., Dapena, I., Oellerich, S., Melles, G.R., 2015. Clinical outcome of 500 consecutive cases undergoing Descemet's membrane endothelial keratoplasty. Ophthalmology 122, 464–470. Romano, V., Iovieno, A., Parente, G., Soldani, A.M., Fontana, L., 2015. Long-term clinical outcomes of deep anterior lamellar keratoplasty in patients with keratoconus. Am. J. Ophthalmol. 159, 505–511. Ross, J., Callanan, D., Kunz, H., Niederkorn, J., 1991a. Evidence that the fate of class IIdisparate corneal grafts is determined by the timing of class II expression. Transplantation 51, 532–536. Ross, J., He, Y.G., Niederkorn, J.Y., 1991b. Class I disparate corneal grafts enjoy afferent but not efferent blockade of the immune response. Curr. Eye Res. 10, 889–892. Saban, D.R., Bock, F., Chauhan, S.K., Masli, S., Dana, R., 2010. Thrombospondin-1 derived from APCs regulates their capacity for allosensitization. J. Immunol. 185, 4691–4697. Saban, D.R., Chauhan, S.K., Zhang, X., El Annan, J., Jin, Y., Dana, R., 2009. 'Chimeric' grafts assembled from multiple allodisparate donors enjoy enhanced transplant survival. Am. J. Transplant. : Off. J. Am. Soc. Transplant. Am. Soc. Transplant Surg. 9, 473–482. Saelens, I.E., Bleyen, I., Bartels, M.C., Van Rij, G., 2011. A posterior khodadoust line in a graft rejection episode after descemet stripping automated endothelial keratoplasty. Cornea 30, 245–246. Salabarria, A.C., Braun, G., Heykants, M., Koch, M., Reuten, R., Mahabir, E., Cursiefen, C., Bock, F., 2019. Local VEGF-A blockade modulates the microenvironment of the corneal graft bed. Am. J. Transplant. https://doi.org/10.1111/ajt.15331. In press. Salouti, R., Zamani, M., Ghoreyshi, M., Dapena, I., Melles, G.R.J., Nowroozzadeh, M.H., 2019. Comparison between manual trephination versus femtosecond laser-assisted deep anterior lamellar keratoplasty for keratoconus. Br. J. Ophthalmol. https://doi. org/10.1136/bjophthalmol-2018-313365. In press. Sanchez-Fueyo, A., Strom, T.B., 2011. Immunologic basis of graft rejection and tolerance following transplantation of liver or other solid organs. Gastroenterology 140, 51–64. Sangwan, V.S., Fernandes, M., Bansal, A.K., Vemuganti, G.K., Rao, G.N., 2005. Early results of penetrating keratoplasty following limbal stem cell transplantation. Indian J. Ophthalmol. 53, 31–35. Sano, Y., Ksander, B.R., Streilein, J.W., 2000. Langerhans cells, orthotopic corneal allografts, and direct and indirect pathways of T-cell allorecognition. Investig. Ophthalmol. Vis. Sci. 41, 1422–1431. Sarnicola, E., Sarnicola, C., Sabatino, F., Tosi, G.M., Perri, P., Sarnicola, V., 2016. Cannula DALK versus needle DALK for keratoconus. Cornea 35, 1508–1511. Savetsky, I.L., Ghanta, S., Gardenier, J.C., Torrisi, J.S., Garcia Nores, G.D., Hespe, G.E., Nitti, M.D., Kataru, R.P., Mehrara, B.J., 2015. Th2 cytokines inhibit lymphangiogenesis. PLoS One 10, e0126908. Schaub, F., Bachmann, B.O., Cursiefen, C., Hos, D., 2019. Midterm follow-up of immune reactions after Descemet membrane endothelial keratoplasty (DMEK). Graefe’s Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. https:// doi.org/10.1007/s00417-019-04357-4. In press. Schaub, F., Cursiefen, C., Heindl, L.M., 2015. Retrospective appraisal of split-cornea transplantation: an audit of 1141 donor corneas. JAMA Ophthalmol. 133, 1086–1087. Schaub, F., Enders, P., Adler, W., Bachmann, B.O., Cursiefen, C., Heindl, L.M., 2017a. Impact of donor graft quality on deep anterior lamellar Keratoplasty (DALK). BMC Ophthalmol. 17, 204. Schaub, F., Enders, P., Roters, S., Cursiefen, C., Bachmann, B.O., 2017b. Single-pass ultrathin DSAEK (UT-DSAEK) with the SLc expert microkeratome(R). Acta Ophthalmol. 95, e160–e161. Schaub, F., Enders, P., Snijders, K., Schrittenlocher, S., Siebelmann, S., Heindl, L.M., Bachmann, B.O., Cursiefen, C., 2017c. One-year outcome after Descemet membrane endothelial keratoplasty (DMEK) comparing sulfur hexafluoride (SF6) 20% versus 100% air for anterior chamber tamponade. Br. J. Ophthalmol. 101, 902–908. Schaub, F., Heindl, L.M., Enders, P., Roters, S., Bachmann, B.O., Cursiefen, C., 2017d. [Deep anterior lamellar keratoplasty : experiences and results of the first 100 consecutive DALK from the university eye hospital of Cologne]. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 114, 1019–1026. Schaub, F., Neuhann, I., Enders, P., Bachmann, B.O., Koller, B., Neuhann, T., Cursiefen, C., 2018. [Boston keratoprosthesis: 73 eyes from Germany : an overview of experiences from two centers]. Ophthalmologe : Z. Dtsch. Ophthalmol. Ges. 115, 744–753. Schlotzer-Schrehardt, U., Bachmann, B.O., Tourtas, T., Torricelli, A.A., Singh, A., Gonzalez, S., Mei, H., Deng, S.X., Wilson, S.E., Kruse, F.E., 2015. Ultrastructure of the posterior corneal stroma. Ophthalmology 122, 693–699. Scholler, N., Hayden-Ledbetter, M., Dahlin, A., Hellstrom, I., Hellstrom, K.E., Ledbetter, J.A., 2002. Cutting edge: CD83 regulates the development of cellular immunity. J. Immunol. 168, 2599–2602. Schrittenlocher, S., Bachmann, B., Tiurbe, A.M., Tuac, O., Velten, K., Schmidt, D., Cursiefen, C., 2019. Impact of preoperative visual acuity on Descemet Membrane Endothelial Keratoplasty (DMEK) outcome. Graefe's archive for clinical and experimental ophthalmology. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 257, 321–329. Schrittenlocher, S., Penier, M., Schaub, F., Bock, F., Cursiefen, C., Bachmann, B., 2017. Intraocular lens calcifications after (triple-) Descemet membrane endothelial keratoplasty. Am. J. Ophthalmol. 179, 129–136. Schrittenlocher, S., Schaub, F., Hos, D., Siebelmann, S., Cursiefen, C., Bachmann, B., 2018. Evolution of consecutive Descemet membrane endothelial keratoplasty outcomes throughout a 5-Year period performed by two experienced surgeons. Am. J. Ophthalmol. 190, 171–178. Seitz, B., Langenbucher, A., Kus, M.M., Kuchle, M., Naumann, G.O., 1999. Nonmechanical corneal trephination with the excimer laser improves outcome after penetrating

30

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Suh, L.H., Yoo, S.H., Deobhakta, A., Donaldson, K.E., Alfonso, E.C., Culbertson, W.W., O'Brien, T.P., 2008. Complications of Descemet's stripping with automated endothelial keratoplasty: survey of 118 eyes at One Institute. Ophthalmology 115, 1517–1524. Sumioka, T., Ikeda, K., Okada, Y., Yamanaka, O., Kitano, A., Saika, S., 2008. Inhibitory effect of blocking TGF-beta/Smad signal on injury-induced fibrosis of corneal endothelium. Mol. Vis. 14, 2272–2281. Sun, Y., Liegl, R., Gong, Y., Buhler, A., Cakir, B., Meng, S.S., Burnim, S.B., Liu, C.H., Reuer, T., Zhang, P., Walz, J.M., Ludwig, F., Lange, C., Agostini, H., Bohringer, D., Schlunck, G., Smith, L.E.H., Stahl, A., 2017. Sema3f protects against subretinal neovascularization in vivo. EBioMedicine 18, 281–287. Szaflik, J.P., Major, J., Izdebska, J., Lao, M., Szaflik, J., 2016. Systemic immunosuppression with mycophenolate mofetil to prevent corneal graft rejection after high-risk penetrating keratoplasty: a 2-year follow-up study. Graefe's Arch. Clin. Exp. Ophthalmol. Albrecht Von Graefes Arch. Klin. Exp. Ophthalmol. 254, 307–314. Tammela, T., Saaristo, A., Lohela, M., Morisada, T., Tornberg, J., Norrmen, C., Oike, Y., Pajusola, K., Thurston, G., Suda, T., Yla-Herttuala, S., Alitalo, K., 2005. Angiopoietin1 promotes lymphatic sprouting and hyperplasia. Blood 105, 4642–4648. Tandon, A., Tovey, J.C., Sharma, A., Gupta, R., Mohan, R.R., 2010. Role of transforming growth factor Beta in corneal function, biology and pathology. Curr. Mol. Med. 10, 565–578. Taylor, A.W., 1999. Ocular immunosuppressive microenvironment. Chem. Immunol. 73, 72–89. Thatte, S., 2011. Fine needle diathermy - a choice for managing corneal vascularization. Nepal. J. Ophthalmol. : A Biannu. Peer-Reviewed Acad. J. Nepal Ophthalmic Soc. 3, 23–26. Thorne, J.E., Jabs, D.A., Qazi, F.A., Nguyen, Q.D., Kempen, J.H., Dunn, J.P., 2005. Mycophenolate mofetil therapy for inflammatory eye disease. Ophthalmology 112, 1472–1477. Torres, P.F., De Vos, A.F., van der Gaag, R., Martins, B., Kijlstra, A., 1996. Cytokine mRNA expression during experimental corneal allograft rejection. Exp. Eye Res. 63, 453–461. Toth, G., Bucher, F., Siebelmann, S., Bachmann, B., Hermann, M., Szentmary, N., Nagy, Z.Z., Cursiefen, C., 2016. In situ corneal cross-linking for recurrent corneal melting after Boston type 1 keratoprosthesis. Cornea 35, 884–887. Tourtas, T., Laaser, K., Bachmann, B.O., Cursiefen, C., Kruse, F.E., 2012. Descemet membrane endothelial keratoplasty versus descemet stripping automated endothelial keratoplasty. Am. J. Ophthalmol. 153, 1082–1090 e1082. Tu, E.Y., 2017. Descemet membrane endothelial keratoplasty patch for persistent corneal hydrops. Cornea 36, 1559–1561. Twist, C.J., Beier, D.R., Disteche, C.M., Edelhoff, S., Tedder, T.F., 1998. The mouse Cd83 gene: structure, domain organization, and chromosome localization. Immunogenetics 48, 383–393. Unal, M., Arslan, O.S., Atalay, E., Mangan, M.S., Bilgin, A.B., 2013. Deep anterior lamellar keratoplasty for the treatment of stromal corneal dystrophies. Cornea 32, 301–305. van Essen, T.H., Roelen, D.L., Williams, K.A., Jager, M.J., 2015. Matching for Human Leukocyte Antigens (HLA) in corneal transplantation - to do or not to do. Prog. Retin. Eye Res. 46, 84–110. Veikkola, T., Alitalo, K., 1999. VEGFs, receptors and angiogenesis. Semin. Canc. Biol. 9, 211–220. Verteporfin in Photodynamic Therapy Study Group, 2001. Photodynamic therapy of subfoveal choroidal neovascularization in pathologic myopia with verteporfin. 1-year results of a randomized clinical trial–VIP report no. 1. Ophthalmology 108, 841–852. Wacker, K., Baratz, K.H., Maguire, L.J., McLaren, J.W., Patel, S.V., 2016. Descemet stripping endothelial keratoplasty for Fuchs' endothelial corneal dystrophy: five-year results of a prospective study. Ophthalmology 123, 154–160. Watanabe, K., Hasegawa, Y., Yamashita, H., Shimizu, K., Ding, Y., Abe, M., Ohta, H., Imagawa, K., Hojo, K., Maki, H., Sonoda, H., Sato, Y., 2004. Vasohibin as an endothelium-derived negative feedback regulator of angiogenesis. J. Clin. Investig. 114, 898–907. Weiss, A., den Bergh, H., Griffioen, A.W., Nowak-Sliwinska, P., 2012. Angiogenesis inhibition for the improvement of photodynamic therapy: the revival of a promising idea. Biochim. Biophys. Acta 1826, 53–70. Wenkel, H., Streilein, J.W., 2000. Evidence that retinal pigment epithelium functions as an immune-privileged tissue. Investig. Ophthalmol. Vis. Sci. 41, 3467–3473. Wielders, L.H., Schouten, J.S., Aberle, M.R., Lambermont, V.A., van den Biggelaar, F.J., Winkens, B., Simons, R.W., Nuijts, R.M., 2017. Treatment of cystoid macular edema after cataract surgery. J. Cataract Refract. Surg. 43, 276–284. Wigle, J.T., Harvey, N., Detmar, M., Lagutina, I., Grosveld, G., Gunn, M.D., Jackson, D.G., Oliver, G., 2002. An essential role for Prox1 in the induction of the lymphatic endothelial cell phenotype. EMBO J. 21, 1505–1513. Williams, K.A., Esterman, A.J., Bartlett, C., Holland, H., Hornsby, N.B., Coster, D.J., 2006. How effective is penetrating corneal transplantation? Factors influencing long-term outcome in multivariate analysis. Transplantation 81, 896–901. Williams, K.A., Keane, M.C., Galettis, R.A., Jones, V.J., Mills, R.A., Coster, D.J., 2015. The Australian Corneal Graft Registry 2015 Report. . http://hdl.handle.net/2328/35402. Willrodt, A.H., Salabarria, A.C., Schineis, P., Ignatova, D., Hunter, M.C., Vranova, M., Golding-Ochsenbein, A.M., Sigmund, E., Romagna, A., Strassberger, V., Fabbi, M., Ferrini, S., Cursiefen, C., Neri, D., Guenova, E., Bock, F., Halin, C., 2019. ALCAM mediates DC migration through afferent lymphatics and promotes allospecific immune reactions. Front. Immunol. 10, 759. Wollensak, G., Spoerl, E., Reber, F., Seiler, T., 2004a. Keratocyte cytotoxicity of riboflavin/UVA-treatment in vitro. Eye 18, 718–722. Wollensak, G., Spoerl, E., Seiler, T., 2003. Riboflavin/ultraviolet-a-induced collagen crosslinking for the treatment of keratoconus. Am. J. Ophthalmol. 135, 620–627. Wollensak, G., Spoerl, E., Wilsch, M., Seiler, T., 2004b. Keratocyte apoptosis after corneal

keratoplasty. Ophthalmology 106, 1156–1164 discussion 1165. Seo, M., Choi, J.S., Rho, C.R., Joo, C.K., Lee, S.K., 2015. MicroRNA miR-466 inhibits Lymphangiogenesis by targeting prospero-related homeobox 1 in the alkali burn corneal injury model. J. Biomed. Sci. 22, 3. Sepsakos, L., Shah, K., Lindquist, T.P., Lee, W.B., Holland, E., 2016. Rate of Rejection after Descemet Stripping Automated Endothelial Keratoplasty in Fuchs Dystrophy: ThreeYear Follow-Up. Cornea. Sheldon, S., Poulton, K., 2006. HLA typing and its influence on organ transplantation. Methods Mol. Biol. 333, 157–174. Shimazaki, J., Den, S., Omoto, M., Satake, Y., Shimmura, S., Tsubota, K., 2011. Prospective, randomized study of the efficacy of systemic cyclosporine in high-risk corneal transplantation. Am. J. Ophthalmol. 152, 33–39. Shortt, A.J., Bunce, C., Levis, H.J., Blows, P., Dore, C.J., Vernon, A., Secker, G.A., Tuft, S.J., Daniels, J.T., 2014. Three-year outcomes of cultured limbal epithelial allografts in aniridia and Stevens-Johnson syndrome evaluated using the Clinical Outcome Assessment in Surgical Trials assessment tool. Stem Cell. Transl. Med. 3, 265–275. Siebelmann, S., Steven, P., Cursiefen, C., 2016. [Intraoperative optical coherence tomography in deep anterior lamellar keratoplasty]. Klin. Monatsblatter Augenheilkd. 233, 717–721. Singh, A., Zarei-Ghanavati, M., Avadhanam, V., Liu, C., 2017. Systematic review and meta-analysis of clinical outcomes of descemet membrane endothelial keratoplasty versus descemet stripping endothelial keratoplasty/descemet stripping automated endothelial keratoplasty. Cornea 36, 1437–1443. Singh, N., Tiem, M., Watkins, R., Cho, Y.K., Wang, Y., Olsen, T., Uehara, H., Mamalis, C., Luo, L., Oakey, Z., Ambati, B.K., 2013. Soluble vascular endothelial growth factor receptor 3 is essential for corneal alymphaticity. Blood 121, 4242–4249. Sloper, C.M., Powell, R.J., Dua, H.S., 2001. Tacrolimus (FK506) in the management of high-risk corneal and limbal grafts. Ophthalmology 108, 1838–1844. Sogutlu Sari, E., Kubaloglu, A., Unal, M., Pinero Llorens, D., Koytak, A., Ofluoglu, A.N., Ozerturk, Y., 2012. Penetrating keratoplasty versus deep anterior lamellar keratoplasty: comparison of optical and visual quality outcomes. Br. J. Ophthalmol. 96, 1063–1067. Soh, Y.Q., Peh, G.S.L., Mehta, J.S., 2017. Translational issues for human corneal endothelial tissue engineering. J. Tissue Eng. Regenerat. Med. 11, 2425–2442. Soma, T., Hayashi, R., Sugiyama, H., Tsujikawa, M., Kanayama, S., Oie, Y., Nishida, K., 2014. Maintenance and distribution of epithelial stem/progenitor cells after corneal reconstruction using oral mucosal epithelial cell sheets. PLoS One 9, e110987. Song, H.B., Park, S.Y., Ko, J.H., Park, J.W., Yoon, C.H., Kim, D.H., Kim, J.H., Kim, M.K., Lee, R.H., Prockop, D.J., Oh, J.Y., 2018. Mesenchymal stromal cells inhibit inflammatory lymphangiogenesis in the cornea by suppressing macrophage in a TSG-6dependent manner. Mol. Ther. : J. Am. Soc. Gene Ther. 26, 162–172. Sonoda, Y., Streilein, J.W., 1992. Orthotopic corneal transplantation in mice–evidence that the immunogenetic rules of rejection do not apply. Transplantation 54, 694–704. Spiteri, N., Romano, V., Zheng, Y., Yadav, S., Dwivedi, R., Chen, J., Ahmad, S., Willoughby, C.E., Kaye, S.B., 2015. Corneal angiography for guiding and evaluating fine-needle diathermy treatment of corneal neovascularization. Ophthalmology 122, 1079–1084. Spoerl, E., Mrochen, M., Sliney, D., Trokel, S., Seiler, T., 2007. Safety of UVA-riboflavin cross-linking of the cornea. Cornea 26, 385–389. Stern, M.E., Siemasko, K., Gao, J., Duong, A., Beauregard, C., Calder, V., Niederkorn, J.Y., 2005. Role of interferon-gamma in a mouse model of allergic conjunctivitis. Investig. Ophthalmol. Vis. Sci. 46, 3239–3246. Steven, P., Bock, F., Huttmann, G., Cursiefen, C., 2011. Intravital two-photon microscopy of immune cell dynamics in corneal lymphatic vessels. PLoS One 6, e26253. Stoitzner, P., Holzmann, S., McLellan, A.D., Ivarsson, L., Stossel, H., Kapp, M., Kammerer, U., Douillard, P., Kampgen, E., Koch, F., Saeland, S., Romani, N., 2003. Visualization and characterization of migratory Langerhans cells in murine skin and lymph nodes by antibodies against Langerin/CD207. J. Investig. Dermatol. 120, 266–274. Stoitzner, P., Pfaller, K., Stossel, H., Romani, N., 2002. A close-up view of migrating Langerhans cells in the skin. J. Investig. Dermatol. 118, 117–125. Streilein, J.W., 1993. Immune privilege as the result of local tissue barriers and immunosuppressive microenvironments. Curr. Opin. Immunol. 5, 428–432. Streilein, J.W., 1999. Regional immunity and ocular immune privilege. Chem. Immunol. 73, 11–38. Streilein, J.W., 2003. Ocular immune privilege: the eye takes a dim but practical view of immunity and inflammation. J. Leukoc. Biol. 74, 179–185. Streilein, J.W., Niederkorn, J.Y., Shadduck, J.A., 1980. Systemic immune unresponsiveness induced in adult mice by anterior chamber presentation of minor histocompatibility antigens. J. Exp. Med. 152, 1121–1125. Streilein, J.W., Stein-Streilein, J., 2000. Does innate immune privilege exist? J. Leukoc. Biol. 67, 479–487. Streilein, J.W., Wilbanks, G.A., Cousins, S.W., 1992. Immunoregulatory mechanisms of the eye. J. Neuroimmunol. 39, 185–200. Stuart, A.J., Romano, V., Virgili, G., Shortt, A.J., 2018. Descemet's membrane endothelial keratoplasty (DMEK) versus Descemet's stripping automated endothelial keratoplasty (DSAEK) for corneal endothelial failure. Cochrane Database Syst. Rev. 6, CD012097. Stuart, P.M., Griffith, T.S., Usui, N., Pepose, J., Yu, X., Ferguson, T.A., 1997. CD95 ligand (FasL)-induced apoptosis is necessary for corneal allograft survival. J. Clin. Investig. 99, 396–402. Su, C.C., Chen, C.W., Ho, W.T., Hu, F.R., Lee, S.H., Wang, I.J., 2015. Phenotypes of trypsin- and collagenase-prepared bovine corneal endothelial cells in the presence of a selective Rho kinase inhibitor, Y-27632. Mol. Vis. 21, 633–643. Sugino, I.K., Rapista, A., Sun, Q., Wang, J., Nunes, C.F., Cheewatrakoolpong, N., Zarbin, M.A., 2011. A method to enhance cell survival on Bruch's membrane in eyes affected by age and age-related macular degeneration. Investig. Ophthalmol. Vis. Sci. 52, 9598–9609.

31

Progress in Retinal and Eye Research xxx (xxxx) xxx–xxx

D. Hos, et al.

Yoshida, M., Kezuka, T., Streilein, J.W., 2000a. Participation of pigment epithelium of iris and ciliary body in ocular immune privilege. 2. Generation of TGF-beta-producing regulatory T cells. Investig. Ophthalmol. Vis. Sci. 41, 3862–3870. Yoshida, M., Takeuchi, M., Streilein, J.W., 2000b. Participation of pigment epithelium of iris and ciliary body in ocular immune privilege. 1. Inhibition of T-cell activation in vitro by direct cell-to-cell contact. Investig. Ophthalmol. Vis. Sci. 41, 811–821. Yu, W.Y., Sheridan, C., Grierson, I., Mason, S., Kearns, V., Lo, A.C., Wong, D., 2011. Progenitors for the corneal endothelium and trabecular meshwork: a potential source for personalized stem cell therapy in corneal endothelial diseases and glaucoma. J. Biomed. Biotechnol. 2011, 412743. Yuen, D., Grimaldo, S., Sessa, R., Ecoiffier, T., Truong, T., Huang, E., Bernas, M., Daley, S., Witte, M., Chen, L., 2014. Role of angiopoietin-2 in corneal lymphangiogenesis. Investig. Ophthalmol. Vis. Sci. 55, 3320–3327. Zhai, J., Gu, J., Yuan, J., Chen, J., 2011. Tacrolimus in the treatment of ocular diseases. BioDrugs : Clin. Immunother. Biopharm. Gene Ther. 25, 89–103. Zheng, Y., Lin, H., Ling, S., 2011. Clinicopathological correlation analysis of (lymph) angiogenesis and corneal graft rejection. Mol. Vis. 17, 1694–1700. Zhou, Z., Apte, S.S., Soininen, R., Cao, R., Baaklini, G.Y., Rauser, R.W., Wang, J., Cao, Y., Tryggvason, K., 2000. Impaired endochondral ossification and angiogenesis in mice deficient in membrane-type matrix metalloproteinase I. Proc. Natl. Acad. Sci. U. S. A 97, 4052–4057. Zhu, Y.T., Li, F., Han, B., Tighe, S., Zhang, S., Chen, S.Y., Liu, X., Tseng, S.C., 2014. Activation of RhoA-ROCK-BMP signaling reprograms adult human corneal endothelial cells. J. Cell Biol. 206, 799–811. Ziaei, M., Ziaei, F., Manzouri, B., 2016. Systemic cyclosporine and corneal transplantation. Int. Ophthalmol. 36, 139–146. Ziemssen, F., Heimann, H., 2012. Evaluation of verteporfin pharmakokinetics–redefining the need of photosensitizers in ophthalmology. Expert Opin. Drug Metabol. Toxicol. 8, 1023–1041. Zirm, E., 1906. Eine erfolgreiche totale Keratoplastik. Albr. Graefe's Arch. Ophthalmol. 64, 580–593. Zygoura, V., Baydoun, L., Ham, L., Bourgonje, V.J.A., van Dijk, K., Lie, J.T., Dapena, I., Oellerich, S., Melles, G.R.J., 2018. Quarter-Descemet membrane endothelial keratoplasty (Quarter-DMEK) for Fuchs endothelial corneal dystrophy: 6 months clinical outcome. Br. J. Ophthalmol. 102, 1425–1430.

collagen cross-linking using riboflavin/UVA treatment. Cornea 23, 43–49. Wong, H.L., Cao, R., Jin, G., Chan, K.M., Cao, Y., Zhou, Z., 2012. When MT1-MMP meets ADAMs. Cell Cycle 11, 2793–2798. Wong, H.L., Jin, G., Cao, R., Zhang, S., Cao, Y., Zhou, Z., 2016. MT1-MMP sheds LYVE-1 on lymphatic endothelial cells and suppresses VEGF-C production to inhibit lymphangiogenesis. Nat. Commun. 7, 10824. Wu, E.I., Ritterband, D.C., Yu, G., Shields, R.A., Seedor, J.A., 2012. Graft rejection following descemet stripping automated endothelial keratoplasty: features, risk factors, and outcomes. Am. J. Ophthalmol. 153, 949–957 e941. Xiao, Q., Jiang, Y., Liu, Q., Yue, J., Liu, C., Zhao, X., Qiao, Y., Ji, H., Chen, J., Ge, G., 2015. Minor type IV collagen alpha5 chain promotes cancer progression through discoidin domain receptor-1. PLoS Genet. 11, e1005249. Yagi-Yaguchi, Y., Yamaguchi, T., Higa, K., Suzuki, T., Yazu, H., Aketa, N., Satake, Y., Tsubota, K., Shimazaki, J., 2017. Preoperative aqueous cytokine levels are associated with a rapid reduction in endothelial cells after penetrating keratoplasty. Am. J. Ophthalmol. 181, 166–173. Yamada, J., Ueno, M., Toda, M., Shinomiya, K., Sotozono, C., Kinoshita, S., Hamuro, J., 2016. Allogeneic sensitization and tolerance induction after corneal endothelial cell transplantation in mice. Investig. Ophthalmol. Vis. Sci. 57, 4572–4580. Yamagami, S., Amano, S., 2003. Role of resident corneal leukocytes and draining cervical lymph nodes in corneal allograft rejection. Cornea 22, S61–S65. Yamaguchi, M., Shima, N., Kimoto, M., Ebihara, N., Murakami, A., Yamagami, S., 2016. Optimization of cultured human corneal endothelial cell sheet transplantation and post-operative sheet evaluation in a rabbit model. Curr. Eye Res. 41, 1178–1184. Yamaguchi, T., Higa, K., Tsubota, K., Shimazaki, J., 2018. Elevation of preoperative recipient aqueous cytokine levels in eyes with primary graft failure after corneal transplantation. Mol. Vis. 24, 613–620. Yamazoe, K., Yamazoe, K., Yamaguchi, T., Omoto, M., Shimazaki, J., 2014. Efficacy and safety of systemic tacrolimus in high-risk penetrating keratoplasty after graft failure with systemic cyclosporine. Cornea 33, 1157–1163. Yazu, H., Yamaguchi, T., Aketa, N., Higa, K., Suzuki, T., Yagi-Yaguchi, Y., Satake, Y., Abe, T., Tsubota, K., Shimazaki, J., 2018. Preoperative aqueous cytokine levels are associated with endothelial cell loss after descemet's stripping automated endothelial keratoplasty. Investig. Ophthalmol. Vis. Sci. 59, 612–620. Yoon, K.C., Ahn, K.Y., Lee, S.E., Kim, K.K., Im, S.K., Oh, H.J., Jeong, I.Y., Park, S.W., Park, Y.G., Nah, H.J., Im, W.B., 2006. Experimental inhibition of corneal neovascularization by photodynamic therapy with verteporfin. Curr. Eye Res. 31, 215–224.

32