Polymeric nanomedicines for image-guided drug delivery and tumor-targeted combination therapy

Nano Today (2010) 5, 197—212

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nanotoday

REVIEW

Polymeric nanomedicines for image-guided drug delivery and tumor-targeted combination therapy Twan Lammers a,b,c,∗, Vladimir Subr d, Karel Ulbrich d, Wim E. Hennink b, Gert Storm b, Fabian Kiessling a a

Department of Experimental Molecular Imaging, RWTH — Aachen University, Pauwelsstrasse 30, 52074 Aachen, Germany Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Sorbonnelaan 16, 3584 CA Utrecht, The Netherlands c Department of Innovative Cancer Diagnosis and Therapy, Clinical Cooperation Unit Radiotherapeutic Oncology, German Cancer Research Center, Im Neuenheimer Feld 280, 69120 Heidelberg, Germany d Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic, Heyrovsky Square 2, 162 06 Prague 6, Czech Republic b

Received 5 February 2010; received in revised form 9 April 2010; accepted 3 May 2010 Available online 31 May 2010

KEYWORDS Nanomedicines; Drug targeting; Polymer therapeutics; Image-guided drug delivery; Theranostics; Combination therapy

Summary Nanomedicine systems aim to improve the biodistribution and the target site accumulation of systemically applied (chemo-) therapeutics. By increasing drug localization at the pathological site, and by reducing its accumulation in healthy organs and tissues, the efficacy of the intervention can often be increased, while its toxicity can be attenuated. Using HPMA copolymers as a model macromolecular nanomedicine system, and several selected examples of recently published proof-of-principle reports, we here show that besides for standard drug delivery purposes, polymeric nanomedicines can also be employed for image-guided drug delivery, as well as for tumor-targeted combination therapy. © 2010 Elsevier Ltd. All rights reserved.

Introduction Many routinely used (chemo-) therapeutic agents suffer from poor pharmacokinetics and from an inappropriate biodistribution. Because of their low molecular weight, for instance,

∗ Corresponding author at: Department of Experimental Molecular Imaging, University Clinic, RWTH Aachen, Pauwelstrasse 30, 52074 Aachen, Germany. Tel.: +49 241 8080254. E-mail addresses: [email protected], [email protected] (T. Lammers).

intravenously administered anticancer drugs are generally rapidly cleared from the circulation (by means of renal filtration), and they do not accumulate well in tumors and in tumor cells. In addition to this, because of their small size and/or their high hydrophobicity, chemotherapeutic drugs generally present with a large volume of distribution, and they tend to accumulate in and cause toxicity towards many different healthy tissues. To assist i.v. applied anticancer agents in achieving proper circulation times and tumor concentrations, and to improve the balance between their efficacy and their toxicity, a large number of drug delivery systems have

1748-0132/$ — see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.nantod.2010.05.001

198

T. Lammers et al.

Figure 1 Drug targeting systems and strategies. (A—E) Examples of clinically used tumor-targeted nanomedicines. Liposomes and liposomal bilayers are depicted in grey, polymers and polymer-coatings in green, linkers allowing for drug release and for sheddable stealth coatings in blue (rectangles), targeting ligands in yellow (arrows), antibodies and antibody fragment in purple, radionuclides in orange (suns) and conjugated or entrapped active agents in red (stars). (F—H) Principles of passive and active drug targeting to tumors. (F) Upon i.v. injection, a low molecular weight anticancer agent is generally rapidly cleared from the circulation, and only low levels of the drug accumulate in tumors and in tumor cells. At the same, due to its small size, its high hydrophobicity and/or its large volume of distribution, significant levels of the agent accumulate in healthy tissues. (G) Upon encapsulation in (or conjugation to) a long-circulating and passively tumor-targeted drug delivery system, the concentration of the active agent in tumors can be increased substantially (by means of the Enhanced Permeability and Retention (EPR) effect), while its accumulation in healthy organs and tissues can be attenuated. (H) Upon the incorporation of targeting ligands, like antibodies and peptides, the interaction between the drug and/or drug delivery system and cancer cells can be improved, resulting in a more selective target site localization and/or target cell uptake. Figure adapted, with permission, from [4].

been designed and evaluated over the years [1—4]. Clinically relevant examples of such nanometer-sized carrier materials are liposomes, polymers, micelles and antibodies (Fig. 1A—E). The former three nanomedicine systems pri-

marily aim to improve the circulation time of the conjugated or entrapped active agent, and to improve its tumor accumulation by means of Enhanced Permeability and Retention(EPR-) mediated passive drug targeting (Fig. 1G). The latter,

Figure 2 Image-guided drug delivery. Schematic representation of the advantages and applications resulting from the cofunctionalization of drug delivery systems with diagnostic and therapeutic agents. By besides pharmacologically active agents also incorporating contrast agents in nanomedicine formulations, it is possible to predict and improve the outcome of the intervention, and to visualize and better understand various important aspects of the drug delivery process. See text for details.

Image-guided polymeric nanomedicines for combination therapy Table 1

199

Clinically relevant examples of intravenously administered polymeric nanomedicines.

Compound

Name

Indication

Status

Albumin-paclitaxel Albumin-methotrexate Dextran-doxorubicin Dextran-camptothecin PEG-L-asparaginase PEG-camptothecin PEG-IFN␣2a/-IFN␣2b PHPMA-doxorubicin Galactosamine-targeted PK1 PHPMA-oxaliplatin PGA-paclitaxel PGA-camptothecin

Abraxane MTX-HSA DOX-OXD Delimotecan Oncaspar Prothecan PegAsys/PegIntron PK1 PK2 ProLindac Xyotax CT-2106

Breast Kidney Various Various Leukemia Lung Melanoma, leukemia Breast, lung, colon Liver Ovarian Lung, ovarian Colon, ovarian

Approved Phase II Phase I Phase I Approved Phase II Phase I/II Phase II Phase I/II Phase II Phase III Phase II

on the other hand, are primarily designed to improve target cell recognition and target cell uptake, and to improve the efficacy of the attached active agent by means of active drug targeting (Fig. 1H). In addition to this, various different types of hybrid systems have been developed over the past two decades, including e.g. antibody-modified liposomes or peptide-functionalized polymers, which are intended to combine passive and active targeting, and to initially localize to tumors by means of EPR, and to subsequently stimulate tumor cell-specific uptake by selectively binding to surface receptors (over-) expressed by cancer cells. The overall aim of these approaches is to more selectively kill cancer cells. Besides for therapeutic interventions, however, actively and passively targeted nanomedicine systems have also been more and more used for diagnostic interventions. Antibodies, liposomes, polymers and micelles carrying contrast agents, such as radionuclides, MR imaging probes and near-infrared fluorophores, have on several occasions already been shown to hold significant potential for detecting diseases, as well as for visualizing various important aspects of the drug delivery process (i.e. for image-guided drug delivery; see Fig. 2). In addition to this, because of their beneficial biodistribution, tumor-targeted nanomedicines have also been more and more used for combined modality interventions, e.g. for combinations with surgery, with radiotherapy and with standard chemotherapy. In the present manuscript, using HPMA copolymers as a model macromolecular nanomedicine system, we summarize the most important evidence obtained in our labs in the past five years with regard to these two areas of research, and we show that these prototypic polymeric drug carriers hold significant potential both for image-guided drug delivery and for tumor-targeted combination therapy.

Drug targeting to tumors using HPMA copolymers As outlined in a large number of previous reports and reviews, passive drug targeting is a validated method for improving the delivery of low molecular weight (chemo) therapeutic agents to tumors [5—8]. Various different types of passively tumor-targeted drug delivery systems

have been designed and evaluated over the years, including besides liposomes [9,10], polymers [11,12] and micelles [13,14], also more exotic carrier materials, like nanotubes [15], nanospheres [16], nanoshells [17], MiniCells [18] and NanoCells [19]. Given their versatility, their biocompatibility and their beneficial biodistribution, we have decided to work with polymeric drug delivery systems. Several different polymer-based anticancer agents have been approved for clinical use, including e.g. Gliadel (i.e. carmustine-containing polymeric wafers; for surgical implantation in case of brain cancer), Lupron Depot (i.e. LHRH-containing polymeric microspheres; for local prostate cancer therapy) and Oncaspar (i.e. PEG-L-asparaginase; for the treatment of leukemia) [4,11,12,20]. For passive tumor targeting, however, essentially only Abraxane, i.e. albumin-based paclitaxel, has managed to gain FDA and EMEA approval (for metastatic breast cancer) [21]. This in spite of the large number of passively tumor-targeted polymeric drug carriers that have been evaluated in animal models and in patients in the past two decades. Prominent examples of macromolecular drug carrier systems evaluated in patients are albumin, poly(ethylene glycol), dextran, poly(L-glutamic acid) and poly[N-(2hydroxypropyl)methacrylamide] [4,11,12]. Several clinically relevant examples of intravenously administered polymeric nanomedicines are summarized in Table 1. The most advanced of these formulations currently is Xyotax, i.e. poly(L-glutamic acid)-based paclitaxel, which is in phase III trials for breast, ovarian and lung cancer [22]. The first clinical trial in which an i.v. applied passively tumor-targeted polymeric prodrug was evaluated in patients was initiated in 1994. The results of this trial, in which copolymers based on N-(2-hydroxypropyl)methacrylamide (i.e. HPMA) were used to improve the tumor-directed delivery of doxorubicin, were published in 1999, and they demonstrated that long-circulating and passively tumor-targeted polymeric drug carriers are able to beneficially affect the therapeutic index of doxorubicin-based chemotherapy: the maximum tolerated dose of poly(HPMAGFLG-doxorubicin) was found to be 4—5 times higher than that of free doxorubicin, no cardiotoxicity was detected (in spite of the relatively high overall doses administered), and several clear responses were observed, in patients with

200

T. Lammers et al.

Figure 3 HPMA copolymers circulate for prolonged periods of time, localize to tumors both effectively and selectively, and improve the tumor-directed delivery of doxorubicin. (A) Blood concentration vs. time curves of three differently sized iodine-131-labeled HPMA copolymers upon i.v. injection into Copenhagen rats bearing subcutaneously transplanted Dunning AT1 tumors, exemplifying that they are retained in systemic circulation significantly more effectively than is the free label (i.e. free iodine-131). (B) Schematic representation of the regions of interest and the corresponding organs used in the scintigraphic analyses. (C) Biodistribution of the three radiolabeled copolymers mentioned above, visualized using 2D ␥-scintigraphy, showing that they effectively localize to tumors over time. (D) Analysis of the tumor and organ concentrations of six differently sized HPMA copolymers at 168 h p.i., exemplifying selective tumor accumulation. Except for spleen, the levels of the copolymers in tumors were always significantly higher than their levels in the majority of healthy tissues. (E) Fluorescence microscopy analysis of the amount of doxorubicin localized to AT1 tumors at 24 h p.i. (F) HPLC analysis of the amount of doxorubicin in tumors at 24 p.i., exemplifying the beneficial effect of drug targeting. Figure adapted, with permission, from [43,65].

non-small cell lung cancer, with colon cancer and with doxorubicin-resistant breast cancer [23]. Following these not unpromising proof-of-principle findings obtained for PK1 (i.e. ‘Prague-Keele 1’; poly(HPMAGFLG-doxorubicin)), about a handful of other HPMA-based polymeric prodrugs entered clinical trials, including e.g. PK2 (i.e. galactosamine-targeted PK1), PNU166945 (i.e. polymer-bound paclitaxel), AP5280 (i.e. a polymer-bound cisplatin-derivative) and AP5346 (i.e. polymer-bound oxaliplatin) [4,11]. In addition to this, under the acronym PK3, also an HPMA-based diagnostic agent has been evaluated in patients. This ∼25 kDa-sized tyrosinamide-modified copolymer can be radiolabeled with iodine, and it can consequently be used for ␥-scintigraphy, for PET and for SPECT. At the preclinical level, a large number of additional HPMA-based nanodiagnostics and nanotherapeutics have been evaluated over the years, functionalized besides with standard chemotherapeutics, like doxorubicin, cisplatin and paclitaxel, also with more recently developed (and more sophisticatedly acting) drugs, such as the heat shock proteininhibitor geldanamycin [24], and the antiangiogenic agent TNP-470 [25]. Moreover, besides passively tumor-targeted polymeric nanomedicines, also a large number of actively targeted HPMA copolymers have been designed and evalu-

ated [26,27]. Furthermore, both in animal models and in patients, certain antibody-targeted polymer—drug conjugates with immunomodulating properties have been tested [28]. And finally, significant progress has also been made with respect to the development of novel drug linkers, like pH-sensitive hydrazone spacers, which are much more effective than GFLG spacers in releasing doxorubicin, and which consequently much more strongly inhibit tumor growth [29—31]. In addition to this, HPMA copolymers have also been used for overcoming multi-drug resistance [32], for coating viral and non-viral gene delivery systems [33], for imaging purposes [34,35], and for making block copolymer micelles, which can be core-crosslinked to improve their circulation times [36]. Furthermore, HPMA copolymers have recently also been employed to improve the treatment of non-cancerous disorders, such as rheumatoid arthritis and bacterial infections, which are also characterized by a strong inflammation component and leaky blood vessels, and which are therefore also highly amenable to EPR-mediated passive drug targeting [37]. And finally, again primarily for cancer, a significant number of studies have focused on the combination of HPMA copolymer-based nanomedicine formulations with other treatment modalities, such as with

Image-guided polymeric nanomedicines for combination therapy

201

Figure 4 Real-time MRI-based biodistributional analysis of pHPMA-gadolinium. (A—C) MR angiography scans of the chest and head region of a rat (A), of a tumor-bearing paw (B), and of an AT1 tumor (C), obtained at 0.5 h after the i.v. injection of a 25 kDa-sized gadolinium-containing HPMA copolymer. (D) Color-coded maximal intensity projection (MIP) of the image in panel (C). (E) Dynamic color-coded MRI T1 determinations obtained for AT1 tumor, for kidney and for liver before contrast agent administration and at various time points after the i.v. injection of a low (0.5 kDa Gd-DTPA-BMA; Omniscan) and high (25 kDa Gd-pHPMA) molecular weight MR contrast agent. Note that contrast agent accumulation corresponds to a decrease in the T1 signal, and that the polymer-based contrast agent slowly but steadily accumulates in tumors over time by means of EPR. (F) Quantification of the concentrations of gadolinium in AT1 tumor, kidney and liver upon the i.v. injection of Gd-DTPA-BMA and Gd-pHPMA. Values represent average ± S.D. (*) p < 0.05 (paired t-test). Figure reproduced, with permission, from [65].

surgery, with radiotherapy, with hyperthermia, with photodynamic therapy and with chemotherapy [38,39]. Together, these insights and efforts, and the wealth of information available in the literature [4—6,11,12,23—42], illustrate that HPMA copolymers are highly suitable systems for drug targeting to tumors. In addition, they indicate that by improving the pharmacokinetic and biodistributional properties of attached diagnostic and therapeutic active agents, long-circulating and passively tumor-targeted polymeric drug carriers might be suitable systems for image-guided drug delivery, as well as for improving the efficacy of combined modality anticancer therapy.

Image-guided polymeric nanomedicines for visualizing drug targeting to tumors Upon (co-) functionalizing polymeric drug carriers with contrast agents, such as with radionuclides or with mag-

netic resonance imaging (MRI) probes, information can be obtained on the biodistribution and the target site accumulation of these nanomaterials. An important advantage of using radionuclides or MR contrast agents over fluorophores is that information can be obtained non-invasively, enabling longitudinal analyses in the same animal. Such real-time biodistribution studies not only substantially increase the amount (and the relevance) of the information that can be obtained, but they also substantially reduce the number of animals needed in such analyses. Another important advantage of using radionuclides and MR imaging agents over fluorophores relates to the quantitative nature of the information that is obtained. Even though several advanced near-infrared fluorophores have been shown to be highly suitable for semi-quantitative biodistribution studies, really meaningful real-time results (i.e. truly quantitative comparisons; e.g. in % injected dose per gram tissue) can — because of physical constraints — still only be obtained upon using radionuclides or MR imaging probes.

202 In the majority of our recent experiments, the aim of which was to better understand and to improve drug targeting to tumors using prototypic polymeric drug carriers, we have therefore focused on biodistributional analyses involving ␥-scintigraphy and MRI. Iodine-131-labeled HPMA copolymers were used for radiographic monitoring (Fig. 3A—D), and gadolinium-labeled HPMA copolymers for MR analyses (Fig. 4). A small number of studies were also performed using fluorophore-labeled copolymers, and in some cases, also drug (i.e. doxorubicin) levels in tumors and organs were quantified, by means of CLSM and HPLC, in order to confirm the results obtained using ␥-scintigraphy and MRI (Fig. 3E—F). To better understand drug targeting to tumors using polymeric nanomedicines, in one of our initial experimental setups, we have set out to compare the biodistribution and the target site accumulation of functionalized and unfunctionalized HPMA copolymers. Hereto, thirteen physicochemically different copolymers were synthesized, varying in size, in charge and in the nature and amount of functional groups introduced, and upon radiolabeling, their circulation times, their tumor accumulation and their organ distribution were evaluated in rats bearing subcutaneously transplanted Dunning AT1 tumors [43]. In line with the literature [5,44], it was found that larger HPMA copolymers circulated longer and accumulated in tumors stronger than did smaller copolymers (Fig. 3). In addition, it was found that the tumor accumulation of copolymers with average molecular weights well below the renal clearance threshold of ∼45 kDa peaked at about 24 h after i.v. injection, and remained relatively constant afterwards, whereas for larger copolymers, levels localizing to tumors continued to increase up until 168 h. The introduction of carboxyl and hydrazide groups, as well as that of spacer-, drug- and peptide-moieties, reduced the long-circulation properties of the copolymers, and it lowered their tumor concentrations. Interestingly, however, levels in the majority of healthy tissues were also found to be reduced, and when calculating tumor-to-organ ratios, hardly any difference was observed between functionalized and unfunctionalized copolymers (with in the majority of cases, higher levels in tumors than in 6—8 out of 9 healthy organs analyzed) [43]. This is an important finding, as it demonstrates that also physicochemically modified HPMA copolymers localize to tumors relatively selectively, and as it illustrates that introducing functional groups into HPMA copolymers does not negatively affect their beneficial biodistribution. The only organ for which a clear deviation from the above trend was observed was the kidney: all ten physicochemically modified HPMA copolymers accumulated in the kidney more strongly than did size-matched non-modified controls [43]. Especially the introduction of peptide moieties was found to result in a strong increase in kidney accumulation, as exemplified by the high concentrations observed for both GFLG-containing copolymers, and for all three PHSCNcontaining copolymers. Using HPMA copolymers modified with 3 and 8 mol% of carboxyl and hydrazide groups, it could furthermore be demonstrated that the degree of functionalization correlates with the degree of kidney accumulation. This notion was confirmed in later experiments with P-GemDox, in which relatively high amounts of doxorubicin and gemcitabine were co-conjugated to the same copolymer,

T. Lammers et al. and in which also very high kidney concentrations were observed [45]. It furthermore is in line with experiments in which HPMA copolymers were functionalized with different amounts of RGDfK (i.e. a cyclic pentapeptide used for blood vessel targeting), and in which also a clear correlation between the degree of functionalization and the degree of kidney accumulation was observed [46]. Also in patients injected with radiolabeled PK1 [23], and in some of our own experiments with radiolabeled PK1 [47,48], indications for a more or less selective localization to the kidney were obtained. Strikingly however, when quantifying the levels of doxorubicin in the kidney, only relatively low levels of the drug were detected, and no tendency towards a more preferential localization to kidney was observed [49]. This either indicates that kidney accumulation takes place after the drug is released or, more likely, that the intact conjugate localizes to the kidney, and that doxorubicin is subsequently rapidly released and excreted (note that relatively high levels of cathepsin B are present in the kidney [50]). The finding that drug levels in the kidney are relatively low is substantiated by the fact that no major kidney-related side effects have been reported thus far, either in animal models, or in patients [11,12,23,40—42]. In spite of this important notion, however, it is advisable to always keep kidney accumulation in mind when working with polymer—drug conjugates, and to always monitor kidney function when treating patients with polymer therapeutics (especially in case of platins; which are highly nephrotoxic). Conversely, because of their strong tendency to accumulate in the kidney, one could also argue that certain densely modified and/or peptide-‘targeted’ HPMA copolymers could be used to improve drug delivery to the kidney, and to enhance the treatment of certain severe kidney disorders, such as renal fibrosis and renal cell carcinoma. Other efforts with regard to better understanding drug targeting to tumors using polymeric nanomedicines have elaborated on the development of a long-circulating gadolinium-containing contrast agent, which can be used for MR angiography (i.e. for visualizing blood vessels; Fig. 4) [34]. Like the abovementioned iodine-labeled copolymers, this gadolinium-containing copolymer holds significant potential for image-guided drug delivery. Image-guided drug delivery refers to the combination of (cancer) diagnosis and therapy, and it not only aims to improve treatment outcome, but also to better understand several important aspects of the drug delivery process (see Fig. 2). Regarding the latter, image-guided drug delivery can be used to visualize the biodistribution of the conjugated or entrapped active agent, to monitor its tumor accumulation, to visualize drug release, to monitor its intratumoral distribution, and to non-invasively assess its therapeutic efficacy (e.g. in case of orthotopic tumors). Regarding the former, i.e. improving treatment outcome, image-guided drug delivery can be used to monitor and to predict therapeutic responses. PK3, for instance, could have been used to pre-screen patients assigned to PK1, in order to identify which tumors are amenable to EPR-mediated drug targeting and which are not, and to predict which patients are likely to respond to PK1 therapy and which are not [23]. By mixing in trace amounts of PK3 with (e.g. every second cycle of) PK1 during follow-up, and by continuously subjecting patients to 2D ␥scintigraphy or 3D PET, it would furthermore be possible to

Image-guided polymeric nanomedicines for combination therapy visualize the efficacy of the intervention in real-time, and to provide important information for assisting in deciding whether or not to (dis-) continue therapy, and whether or not to adjust drug doses. Image-guided drug delivery might thereby assist in realizing the potential of ‘personalized medicine’, i.e. tailor-made therapy for individual patients, which besides on the study of genetic polymorphisms and biomarkers, also relies on the development of visual methods for measuring and predicting treatment outcome. An additional important aspect of image-guided drug delivery is that it substantially facilitates preclinical testing, enabling not only more informative and less invasive biodistribution studies, but also more elegant and more relevant efficacy analyses, in which e.g. genetically modified mice and orthotopic tumor models are used to study disease progression and treatment outcome in real-time. A final important consideration to take into account with regard to image-guided drug delivery relates to the fact that it can be used to better understand tumor physiology. Not only tumor perfusion and tumor blood vessel density can be visualized using imageguided nanomedicines, but e.g. also the extravasation of agents across tumor blood vessels, and their penetration into the (less well-vascularized areas in the) cores of the tumors. Our recent studies have contributed to efforts in this regard by synthesizing several different image-guided polymeric nanomedicines. Iodine-labeled HPMA copolymers can be used for ␥-scintigraphy, for PET and for SPECT, and they enable a highly detailed analysis of the pharmacokinetics and the biodistribution of the attached active agents. Gadolinium-labeled HPMA copolymers can be used both for biodistributional and for functional analyses, enabling e.g. MR angiography to study blood vessel status. Both radionuclide- and gadolinium-modified polymeric nanomedicines are considered to be highly suitable systems for image-guided drug delivery: the former can be used for non-invasively assessing target site accumulation, for monitoring intratumoral distribution and for predicting treatment outcome, and the latter for visualizing tumor penetration and for tracking treatment efficacy (e.g. in case of antiangiogenic therapy). Interesting future directions in this regard relate to the use of an iodine-labeled hydrazonebased polymeric prodrug for predicting tumor accumulation and treatment efficacy in patients, the use of a gadoliniumlabeled HPMA copolymer for visualizing the antiangiogenic effects of (co-conjugated) TNP-470, and the use of polymerbased contrast agents for multimodality imaging (e.g. upon co-functionalizing HPMA copolymers with two or three different MR-, PET-, SPECT- and/or NIR-probes). Together, the above insights and efforts demonstrate that HPMA copolymers are suitable systems for image-guided drug delivery, and they illustrate that iodine- and gadolinium-labeled HPMA copolymers can be used to better understand drug targeting to tumors.

Improving drug targeting to tumors using image-guided polymeric nanomedicines Attempts to improve drug targeting to tumors using HPMA copolymer-based nanomedicine formulations have encompassed both passive and active targeting approaches.

203

Regarding the latter, various different targeting ligands have been evaluated over the years (including e.g. aminosugars, hormones, antibodies and peptides), and several different targeting strategies have been tested (e.g. cancer cell targeting, endothelial cell targeting and enhancing cellular uptake) [26,27,51,52]. Regarding the former, on the other hand, experiments have thus far essentially only aimed at establishing biodegradable high molecular weight grafts, which because of their large size and their prolonged circulation times accumulate in tumors significantly more effectively than do standard small HPMA copolymers (i.e. non-biodegradable polymer therapeutics, like PK1, with average molecular weights well below the renal clearance threshold of ∼45 kDa) [31,49,53]. In our recent studies, as an alternative strategy for (actively) improving passive drug targeting, we have set out to combine HPMA copolymers with other treatment modalities. Both radiotherapy and hyperthermia were used for this purpose [48], and also agents known to affect tumor blood flow and tumor blood vessels were tested (i.e. angiotensin II and prostaglandin E2). Regarding the latter, i.e. co-treatment with the vasoconstrictor ATII and the extravasation-enhancer PGE2, no indications for an improvement in tumor accumulation were observed, at least not for a 31 kDa-sized drug-free control copolymer, and not in the Dunning AT1-sc rat prostate carcinoma model. This can be explained by taking into account that blood vessels in AT1-sc tumors are poorly supported by pericytes and smooth muscle cells, and therefore already are relatively leaky, and likely do not respond well to treatment with vasoconstricting agents (Fig. 5A). As it seems obvious that other tumor models, in which blood vessels are less tortuous and less leaky (i.e. better differentiated), respond very differently to treatment with such compounds, no definite conclusions can be drawn from these studies yet, and additional analyses are necessary to resolve this issue. Alternative interesting pharmacological means for improving passive drug targeting to tumors relate to the use of antiangiogenic agents, like Avastin and Sutent, which normalize the tumor vasculature, and to the use of TNF␣ and inhibitors of TGF␤, which enhance extravasation. The above notion of tumor heterogeneity was taken firmly into account when assessing the effects of radiotherapy and hyperthermia on the tumor accumulation of HPMA copolymers. Before actually combining them with radiotherapy and with hyperthermia, we therefore first evaluated the biodistribution and the tumor accumulation of two different HPMA copolymers in three different tumor models, i.e. in AT1-sc, AT1-im and H-sc (Fig. 5). Dunning AT1 tumors transplanted subcutaneously (AT1-sc) were used as a reference model and as a positive control, as they are known to possess relatively leaky blood vessels, and to accumulate HPMA copolymers relatively well [43]. Dunning AT1 tumors inoculated intramuscularly (AT1-im) were used to reduce the impact of subcutaneous tumor encapsulation. Because of the more infiltrative growth and the substantial increase in tumor-muscle surface area, the clearance of HPMA copolymers from AT1-im tumors was expected to be less unidirectional and, therefore, to be more substantial than for AT1-sc tumors. Dunning H tumors transplanted subcutaneously (H-sc) were used to investigate the impact of the morphology of tumor blood vessels on the accumu-

204

T. Lammers et al.

Figure 5 HPMA copolymer-based nanomedicine systems effectively accumulate in three different tumor models. (A) Characterization of the three tumor models used. Immunohistochemical analysis of subcutaneously transplanted Dunning AT1 tumors (AT1-sc), intramuscularly inoculated Dunning AT1 tumors (AT1-im) and subcutaneously transplanted Dunning H tumors (H-sc). Methanol/acetone-fixed cryosections were stained using H+E, CD31 (red) and ␣-smooth muscle actin (green), and imaged at magnifications of 400× (H+E) and 600× (CD31 and ␣SMA). Arrows indicate muscle fibers. (B) Scintigraphic analysis of the biodistribution and the tumor accumulation of two differently sized radiolabeled HPMA copolymers in Copenhagen rats bearing AT1-sc, AT1-im and H-sc tumors. In the scintigrams obtained at 0.5 h p.i., localization to heart (i.e. in circulation (1)) and to bladder (2) was observed. In the images obtained at 24 h and 168 h p.i., the accumulation of the copolymers was most prominent in spleen (3), in liver (4) and in tumor (5). Figure adapted, with permission, from [48].

lation of HPMA copolymers. H-sc tumors grow much slower than AT1-sc tumors, and they consequently present with a slowly matured and well-differentiated vasculature, which is much less leaky than that in AT1-sc tumors (Fig. 5A). Two differently sized HPMA copolymers, i.e. a 31 kDa and a 65 kDa drug-free control copolymer containing ∼1 mol% of tyrosinamide (for radiolabeling), were used in these initial investigations, and in line with our expectations, it was found that HPMA copolymers accumulated most effectively in AT1-sc tumors. Levels in AT1-im and H-sc tumors were always approximately 50% lower, but in spite of this, evidence in favor of effective EPR-mediated passive drug targeting could be obtained in all cases, as exemplified e.g. by the fact that tumor concentrations were always substantially higher for the 65 kDa copolymer than for the 31 kDa copolymer (Fig. 5B) [48]. This is an important finding, as it demonstrates that HPMA copolymers effectively accumulate in various different types of tumors, and as it confirms the beneficial biodistributional properties of HPMA copolymers. In these three models, we then set out to evaluate the effects of three different doses of hyperthermia (i.e. 41, 42 and 42.5 ◦ C; applied for 1 h; concomitant to i.v. administration) and four different regimens of radiotherapy (i.e. 2 and 20 Gy; applied 1 and 24 h prior to i.v. injection) on the tumor accumulation of the copolymers. As opposed to hyperthermia, which only increased the tumor accumulation of the 65 kDa copolymer in H-sc tumors, 20 Gy of radiotherapy, applied 24 h prior to i.v. injection, improved the tumor

accumulation of both copolymers in all three tumor models (Fig. 6) [48]. This notion suggests that radiotherapy affects the permeability of tumor blood vessels through one or more general (patho-) physiological mechanisms. It has been shown in this regard, for instance, that radiotherapy upregulates the expression of vascular endothelial growth factor (VEGF) [54] and fibroblast growth factor (FGF) [55], which are both known to be able to increase the permeability of (tumor) blood vessels towards nanosized (carrier) materials [56]. Also by inducing apoptosis in endothelial cells, radiotherapy might enhance the extravasation of nanomedicine systems [57]. In addition to this, radiotherapy has been shown to be able to decrease the interstitial fluid pressure in tumors, i.e. the outward-bound pressure gradient that is typical of solid malignancies and that is, at least in part, responsible for the poor penetration of blood-borne agents into tumors [58]. And furthermore, radiotherapy has been shown to be able to reduce the cell density in tumors [59], which is also known to be an important (negative) determinant of drug penetration and drug distribution in solid malignancies. Our findings are in line with the small number of other reports that have been published on the effects of radiotherapy on the tumor accumulation of passively targeted nanomedicines, and it is interesting to note that for similar systems, i.e. for PEGylated liposomal doxorubicin and for poly(L-glutamic acid)-based paclitaxel, similar levels of improvement have been observed (ranging from

Image-guided polymeric nanomedicines for combination therapy

Figure 6 Effect of radiotherapy and hyperthermia on the tumor accumulation of HPMA copolymers. (A) Scintigraphic analysis of the effects of radiotherapy (20 Gy; 24 h prior to i.v. injection) and hyperthermia (42 ◦ C; applied for 1 h; concomitant to i.v. injection) on the tumor accumulation of the 31 kDa-sized HPMA copolymer. (B—C) Analysis of the effect of different doses of radiotherapy on the tumor accumulation of the 31 kDa (B) and the 65 kDa (C) copolymer at 24 h after i.v. injection, exemplifying that radiotherapy improves the tumor accumulation of HPMA copolymers independent of the tumor model used. (D—E) Analysis of the effect of different doses of hyperthermia on the tumor accumulation of the 31 kDa (D) and the 65 kDa (E) copolymer at 24 h after i.v. injection, exemplifying that hyperthermia only improves the tumor accumulation of the 65 kDa copolymer in the H-sc model. Values represent average ± S.D. (*) p < 0.05 (unpaired t-test). Figure reproduced, with permission, from [48].

∼25% to ∼100%; depending on polymer size and on the tumor model used) [60,61]. Together, these findings convincingly demonstrate that radiotherapy is an effective means for improving drug targeting to tumors. Interesting future directions in this regard relate to the combina-

205

tion of passively tumor-targeted polymeric prodrugs with clinically relevant regimens of radiotherapy (see below), and to the development of fibrinogen- and RGD-modified actively targeted nanomedicines, which selectively bind to radiotherapy-induced membrane receptors, such as ␣v ␤3 and ␣2b ␤3 integrins [62]. As opposed to radiotherapy, hyperthermia was only found to be able to improve the concentration of one HPMA copolymer in one tumor model. In AT1-sc and AT1-im tumors, which already possess a relatively leaky vasculature, hyperthermia had no obvious effect on the amount of copolymer localizing to tumors [48]. In slowly growing H-sc tumors, on the other hand, which possess a properly differentiated endothelial lining, hyperthermia did increase the concentrations of the 31 kDa (∼25%; not significant) and the 65 kDa (∼80%; p < 0.05) copolymer (Fig. 6). In principle, these findings are in line with the results obtained for liposomes, for which it has been demonstrated that hyperthermia can increase the pore cut-off size of tumor blood vessels from less than 100 nm to values beyond 400 nm [63,64]. As compared to polymers, however, liposomes are relatively large (i.e. 5—10 nm vs. 100—200 nm, respectively). This implies that polymer therapeutics might already extravasate effectively at pore cut-off sizes of less than 10 nm, whereas for liposomes, pore cut-off sizes have to be at least 100 nm. It can consequently be reasoned that the more leaky tumors already are under non-hyperthermic conditions, the less likely it is that hyperthermia will have a beneficial effect on the tumor accumulation of polymeric prodrugs. As liposomes are much larger than polymers, and as liposomes therefore much more depend on pore cut-off size for effective extravasation, it seems as if hyperthermia might be more useful for improving the tumor accumulation of liposomes than for improving that of polymers. It should be noted in this regard, however, that in general, blood vessels in animal models tend to be (much) more leaky than blood vessels in human tumors, and it would therefore not be surprising if in patients, hyperthermia would turn out to be able to improve the tumor accumulation of polymers. These findings demonstrate, at least for HPMA copolymers, that the effects of hyperthermia depend significantly on the tumor model used, and they seem to suggest that tumor grade inversely correlates with responsiveness to hyperthermia. More promising progress with regard to the combination of tumor-targeted nanomedicines with hyperthermia has recently been made by developing temperature-sensitive drug delivery systems. By carefully optimizing lipid composition, for instance, doxorubicin-containing PEGylated liposomes can be prepared which effectively release the entrapped active agent upon heating to temperatures exceeding 38 ◦ C. This formulation, i.e. ThermoDox, is currently under phase III evaluation for the treatment of patients with primary liver carcinomas, and the results of this trial are eagerly awaited.

Polymeric nanomedicines for tumor-targeted combination therapies To improve the therapeutic efficacy of polymeric nanomedicines, and to broaden their clinical applicability, several different HPMA copolymer-based anticancer agents have in recent years been combined with other

206

T. Lammers et al.

Figure 7 Carrier-based radiochemotherapy. (A) Carrier-based radiochemotherapy relies on the notion that drug targeting systems and radiotherapy interact synergistically, with on the one hand, radiotherapy improving the tumor accumulation of drug targeting systems, and with on the other hand, drug targeting systems improving the temporal and spatial interaction between i.v. applied (weekly) chemotherapy and clinically relevant (daily) radiotherapy. (B) Schematic representation of the in vivo interaction between daily radiotherapy and weekly chemotherapy upon standard radiochemotherapy and upon carrier-based radiochemotherapy, exemplifying that in case of the latter, both the temporal and the spatial interaction between the two treatment modalities is improved. The biodistribution of an i.v. applied standard (left panel) and passively tumor-targeted (right panel) chemotherapeutic agent is depicted in red, the application of fractionated radiotherapy in yellow (lightning), and tumors as squares. (C) Growth inhibition of subcutaneous Dunning AT1 tumors induced by three i.v. injections (days 1, 8 and 15; vertical arrows) of saline, of free doxorubicin, of HPMA copolymer-bound doxorubicin and of an IgG-modified version of HPMA copolymer-bound doxorubicin. (D) Upon combining three weekly i.v. injections (days 1, 8 and 15; vertical arrows) of free doxorubicin, of HPMA copolymer-bound doxorubicin and of IgG-targeted HPMA copolymer-bound doxorubicin with 20 daily doses of fractionated radiotherapy (vertical lines), both passively tumor-targeted polymeric prodrugs were found to be significantly more effective in inhibiting tumor growth than was the free drug. In addition, they were also found to be significantly less toxic (see [65]). Values represent average ± S.E.M. (*) p < 0.05 vs. control, (#) p < 0.05 vs. free Dox, and (†) p < 0.005 vs. free Dox (Mann—Whitney U test; Bonferroni—Holm post hoc analysis). Figure adapted, with permission, from [65].

treatment modalities, e.g. with surgery, with radiotherapy and with chemotherapy. Regarding surgery, the effect of intratumoral injection on the biodistribution and the therapeutic potential of HPMA copolymer-based drug delivery systems was evaluated [47]. This was done to demonstrate that long-circulating and passively tumor-targeted polymeric drug carriers can be used to improve the retention of an i.t. applied low molecular chemotherapeutic agent at the target site. To assess the impact of i.t. injection on the biodistribution and the therapeutic potential of HPMA copolymers, three different copolymers were injected both i.v. and i.t., and their circulation kinetics, tumor concentrations, tumor-to-organ ratios, antitumor efficacy and toxicity were compared. It was found that as compared to i.v. injection, i.t. injection improved both the

tumor concentrations and the tumor-to-organ ratios of the copolymers substantially. In addition, as compared to i.v. and i.t. applied free doxorubicin and to i.v. applied PK1 (i.e. poly(HPMA)-GFLG-doxorubicin), the i.t. administered polymeric prodrug was found to be significantly more effective, and to be less toxic. These findings suggest that when tumors are easily accessible, e.g. during surgery, intratumorally injected carrier-based chemotherapeutics should be considered as interesting alternatives for the routinely used chemotherapy regimens and routes of administration. Following up on the abovementioned beneficial effects of radiotherapy on the tumor accumulation of polymeric nanomedicines, we have furthermore set out to evaluate the potential of ‘carrier-based radiochemotherapy’ [65].

Image-guided polymeric nanomedicines for combination therapy Carrier-based radiochemotherapy is based on the notion that the temporal and spatial interaction between i.v. applied weekly chemotherapy and clinically relevant daily radiotherapy is suboptimal, and on the assumption that drug targeting systems are able to improve the temporal and spatial parameters of this interaction (Fig. 7). HPMA copolymers were used as a model drug delivery system, doxorubicin and gemcitabine as model drugs, and the syngeneic and radio- and chemoresistant Dunning AT1 rat prostate carcinoma as a model tumor model. Using ␥-scintigraphy, MRI, fluorescence microscopy and HPLC, the polymeric drug carriers were first shown to circulate for prolonged periods of time, to localize to tumors both effectively and selectively, and to improve the tumor-directed delivery of doxorubicin. Subsequently, they were then shown to interact synergistically with radiotherapy, with radiotherapy increasing the tumor accumulation of the copolymers, and with the copolymers increasing the therapeutic index of radiochemotherapy (both for doxorubicin and for gemcitabine). Regarding the former, i.e. enhancing the tumor accumulation of HPMA copolymers, we have extended previous efforts by working both with iodine- and with gadolinium-labeled copolymers, and by evaluating their tumor concentrations both at 24 and at 168 h post i.v. injection. And regarding the latter, previous efforts were extended by attempting to generalize the concept of carrier-based radiochemotherapy, by implementing clinically relevant regimens of radio- and chemotherapy, by using a radio- and chemoresistant tumor model, and by directly comparing the radiosensitizing potential of polymeric prodrugs to that of free chemotherapeutic agents (Fig. 7). Together, these insights and efforts strongly suggest that carrier-based radiochemotherapy holds significant potential for improving the treatment of advanced solid malignancies. Some initial clinical proof-of-principle in favor of carrierbased radiochemotherapy has recently been provided by Dipetrillo and colleagues, who treated 12 patients with localized esophageal and gastric cancer with the combination of poly(L-glutamic acid)-bound paclitaxel (Xyotax; 6 doses; weekly) and fractionated radiotherapy (28 cycles; 1.8 Gy; daily), and who observed 4 complete responses and an additional 7 partial responses (with reductions in tumor size of more than 50%) [66]. Prior to this trial, preclinical studies had already identified Xyotax as a highly potent radiosensitizer: when a single i.v. injection of Xyotax was combined with a single dose of radiotherapy, for instance, the dose required to produce 50% tumor cures (i.e. the TCD50 ) could be reduced substantially, from 53.9 to 7.5 Gy [67]. When radiotherapy was delivered as 5 daily fractions, the effect of Xyotax was even more pronounced, reducing the TCD50 from 66.6 to 7.9 Gy [67]. In line with this, in a follow-up study in the same tumor model, similar results were reported for Abraxane, i.e. for albumin-based paclitaxel, which also beneficially combined both with single dose and with fractionated radiotherapy, and which did not increase normal tissue radiotoxicity [68]. In comparable analyses, Harrington et al. demonstrated that also liposomes hold significant potential for combination with radiotherapy [69]. They combined both PEGylated liposomal doxorubicin and PEGylated liposomal cisplatin with both single dose (4.5 and 9 Gy) and fractionated (3 × 3 Gy) radiotherapy, and they showed that animals

207

treated with carrier-based radiochemotherapy survived for significantly longer periods of time than did animals treated with standard radiochemotherapy. Davies and colleagues recently confirmed and extended these findings, showing that Doxil, i.e. PEGylated liposomal doxorubicin, significantly improves the efficacy of both single dose (8 Gy) and fractionated (3 × 3.6 Gy) radiotherapy, and that it does so, at least in part, by improving the penetration and the intratumoral distribution of the drug [61]. In experiments with Xyotax, it had also already been observed that radiotherapy increases the tumor accumulation of passively targeted nanomedicines, attributing at least part of the supra-additively improved efficacy of PGA-paclitaxel (i.e. Xyotax) and radiotherapy to a radiotherapy-induced increase in tumor localization [60]. It is interesting to note in this regard that the overall improvement in the tumor concentration of PGA-paclitaxel was virtually identical to that observed in our studies for (size-matched) HPMA copolymers, with as compared to sham-irradiated controls, increases ranging from ∼25 to 50%. Also in line with our findings, it was demonstrated that this radiotherapy-induced increase in the tumor accumulation of Xyotax could already be observed almost immediately upon i.v. injection (i.e. at 1 h p.i., vs. at 30 min p.i.), and that it remained relatively constant over time (i.e. up to 24, and up to 168 h p.i.) [60,65]. Together with the abovementioned findings on the enhanced penetration and the improved intratumoral distribution of PEGylated liposomal doxorubicin in response to radiotherapy [61], these observations suggest that radiotherapy beneficially affects both arms of the EPR effect: on the one hand, by enhancing the expression of VEGF and FGF [54—56] and by inducing endothelial cell apoptosis [57], it likely increases the permeability of tumor blood vessels towards long-circulating nanomedicines, and on the other hand, by reducing the tumor cell density [59] and the interstitial fluid pressure [58], it likely improves their retention and their intratumoral distribution. Though clinically clearly less important than the improved therapeutic indices resulting from carrier-based radiochemotherapy, these improvements in EPR-mediated drug targeting to tumors provide an additional indication for combining long-circulating and passively tumor-targeted (polymeric) nanomedicines with radiotherapy. Based on these findings, and on the fact that its principles are likely broadly applicable (and e.g. also hold for liposomes and for micelles), we propose carrier-based radiochemotherapy as a novel, a rational, a translational and a relatively straightforward approach for improving the treatment of advanced solid malignancies. Clinical trials in which nanomedicine formulations are combined with fractionated radiotherapy are therefore strongly encouraged. To finally also provide experimental evidence for using polymeric nanomedicines to improve the efficacy of chemotherapy combinations, and to for the first time provide in vivo proof-of-principle for ‘polymerbased multi-drug targeting’ (Fig. 8A), a polymer—drug conjugate carrying 6.4 wt.% of gemcitabine, 5.7 wt.% of doxorubicin and 1 mol% of tyrosinamide (to allow for radiolabeling) was synthesized [45]. The resulting 24 kDa construct, i.e. poly(HPMA-co-MA-GFLG-gemcitabine-co-MAGFLG-doxorubicin-co-MA-TyrNH2 ) (Fig. 8B), was termed P-Gem-Dox, and its properties were evaluated both in vitro

208

T. Lammers et al.

Figure 8 Polymer-based multi-drug targeting. (A) Rationale for polymer-based multi-drug targeting. Left image: upon the concomitant i.v. injection of two low molecular weight chemotherapeutic drugs, relatively low levels of the two agents accumulate in tumors, in tumor cells, and in their nuclei (N). In addition to this, because of differences in the pharmacokinetics, the tumor accumulation and the intratumoral distribution of the two agents, many cells within a solid tumor will only be exposed to either one of them. As a consequence, in such cases, DNA repair generally prevails over apoptosis and/or necrosis, and the majority of cells eventually escape from chemotherapy-induced cell death. Right image: upon the i.v. injection of a polymeric prodrug carrying two different chemotherapeutic agents simultaneously, the overall levels of the agents in tumors, in tumor cells and in their nuclei are much higher. In addition to this, the two attached active agents are in all cases delivered (in-) to the same cells, thereby assuring that the cells are exposed to both agents at the same time, and thereby lowering the threshold for the induction of apoptosis and/or necrosis. As a consequence, tumors and tumor cells treated with two-drug-containing (HPMA) copolymers will be less able to escape from apoptosis and/or necrosis, and a substantially higher fraction of them will eventually be eliminated. (B—D) Simultaneous delivery of gemcitabine and doxorubicin to tumors in vivo using HPMA copolymers. (B) Structure of P-Gem-Dox. i.e. poly(HPMA-co-MA-GFLG-gemcitabine-co-MA-GFLG-doxorubicin-co-MA-TyrNH2 ). This 24 kDa-sized copolymer contained 5.7 wt.% of gemcitabine, 6.4 wt.% of doxorubicin and 1.0 mol% of tyrosinamide (to allow for radiolabeling). (C) Growth inhibition of subcutaneously transplanted Dunning AT1 tumors induced by three i.v. injections (days 1, 8 and 15) of saline, of 1.4 mg/kg gemcitabine plus 1.25 mg/kg doxorubicin (Gem + Dox: 50% MTD), of 12.8 mg of P-Gem plus 14.4 mg/kg of P-Dox (10.9 wt.% and 8.7 wt.%, respectively; P-Gem + P-Dox: 50% MTD), and of 21.9 mg/kg of P-Gem-Dox (5.7 and 6.4 wt.%, respectively; P-Gem-Dox: 50% MTD). For the 25% and 75% MTD regimens, P-Gem-Dox was administered at 3 × 11.0 mg/kg and at 3 × 32.9 mg/kg, respectively. Values represent average ± S.E.M. (*) p < 0.05 vs. control, (#) p < 0.005 vs. control, (†) p < 0.05 vs. free gemcitabine plus free doxorubicin, (‡) p < 0.005 vs. free gemcitabine plus free doxorubicin, and (§) p < 0.05 vs. P-Gem plus P-Dox (Mann—Whitney U test; Bonferroni—Holm post hoc analysis). (D) Effect of drug treatment on angiogenesis and apoptosis. Cryosections were obtained one week

Image-guided polymeric nanomedicines for combination therapy and in vivo. In vitro, P-Gem-Dox was found to effectively release both agents over time, and to kill cancer cells at low nanomolar concentrations. In vivo, using 2D ␥-scintigraphy, P-Gem-Dox was shown to circulate for prolonged periods of time, and to localize to tumors relatively selectively, with higher levels in tumors than in 7 out of 9 healthy tissues. In addition, P-Gem-Dox effectively inhibited tumor growth: as compared to control regimens, it increased the efficacy of the combination of gemcitabine and doxorubicin without increasing its toxicity, and it more strongly inhibited angiogenesis and induced apoptosis (Fig. 8C and D). Together with the pioneering work reported by Vicent, Duncan and coworkers on the co-conjugation of doxorubicin and aminogluthetimide to HPMA copolymers [38,70], these findings demonstrate that polymers, as e.g. liposomes, can be used to deliver two different drugs to tumors simultaneously, and they thereby set the stage for more elaborate analyses on the potential of polymer-based multi-drug targeting.

Future perspectives Together, the above insights and efforts, and the wealth of information available in the literature, demonstrate that HPMA copolymers are versatile and multifunctional drug carriers that circulate for prolonged periods of time, that localize to tumors both effectively and selectively, and that are able to improve the tumor-directed delivery of low molecular weight agents. Because of their multifunctional nature, their biocompatibility and their beneficial biodistribution, HPMA copolymer-based nanomedicines are considered to be highly suitable systems for image-guided drug delivery and for tumor-targeted combination therapy. Besides HPMA copolymers, however, there are a large number of other nanomedicine systems that are — at least in theory — highly suitable for image-guided drug delivery and combined modality anticancer therapy. Not only other types of polymeric carrier materials can be used for such purposes, but also liposomes, micelles and antibodies, and likely also some of the more exotic nanomedicine systems that have been designed and evaluated in the past few years (such as nanotubes, nanospheres, nanoshells, MiniCells and NanoCells) [9—19]. In addition to this, also materials with an intrinsic ability to be used for imaging purposes, such as iron oxides [71—73] and quantum dots [74—76], seem to hold significant potential for use as image-guided nanomedicines. These systems can in principle be relatively easily co-loaded with several different types of drugs and imaging agents, and they can consequently be used both for multimodality imaging and for tumor-targeted combination therapy [77—79]. Thus far, however, not many efforts have been undertaken in this regard, but it is expected that in the next decade, there will be a steep increase in the number of multifunctional theranostic nanomedicines. Consequently, significant advances are foreseen in the fields of multimodal (molecular) imaging, of image-guided drug delivery, and of image-guided and tumor-targeted combination therapies.

209

Another interesting application of image-guided drug delivery that will likely receive more and more attention in the years to come relates to the real-time monitoring of drug release from nanomedicine systems. Liposomal formulations can for instance be developed in which pharmacologically active agents are co-entrapped with low molecular weight gadolinium-containing compounds (which do not generate contrast when shielded from external water), and which upon release from the liposomes allow for a non-invasive visualization of drug activation. Similarly, liposomes, polymers, micelles and several other types of nanomedicine systems can be co-loaded with drugs and near-infrared dyes, which are quenched at high concentrations and/or in close proximity, and which upon liberation from the carrier materials can be used to optically monitor drug release. Various other concepts and constructs can be envisioned in which such theranostic nanomedicine formulations are used to visualize drug activation. Such systems are expected to be of considerable importance for optimizing the efficacy of therapeutic interventions in which external stimuli are used to trigger drug release. An interesting example in this regard relates to the use of temperature-sensitive liposomes co-loaded with doxorubicin, with a CEST-agent and with fluorine-19. When implemented in MR-guided HIFU (i.e. magnetic resonance-guided high intensity focused ultrasound), this formulation can be used to non-invasively assess its biodistribution and its target site accumulation (by means of the CEST-signal), and at the same time also to visualize HIFU-mediated hyperthermia-induced drug release at the target site (by loss of the CEST-signal and generation of the fluorine-19-signal) [80]. Similar types of systems can be generated using other drug delivery platforms, and the next decade will likely see a considerable increase in the number of nanomedicine systems which can be visualized in vivo in real-time, which can be triggered to release their contents, and which can be used to non-invasively assess the efficacy of triggered drug release. A final important aspect that needs to be taken into account with regard to image-guided drug delivery relates to predicting and monitoring therapeutic responses. As already mentioned above, in the clinical trials with PK1, trace amounts of PK3 (i.e. radiolabeled PK1) could have been used to pre-screen patients, in order to identify those tumors which are amenable to EPR-mediated drug targeting and which are not, and to predict which patients are likely to respond to PK1 therapy and which are not. By at the same time also mixing trace amounts of PK3 with e.g. every second or third cycle of PK1 during follow-up, and by continuously subjecting patients to 2D ␥-scintigraphy or 3D PET, it would have furthermore been possible to visualize the efficacy of the intervention in real-time, and to provide important information for assisting in deciding whether or not to (dis-) continue therapy, and whether or not to adjust drug doses. Similar theranostic strategies can be envisioned for virtually all other types of nanomedicine systems, and

after the end of therapy (i.e. at day 22), and blood vessels and apoptotic cells were stained using antibodies against CD31 (red) and TUNEL (green), respectively. Images were counterstained with DAPI. Note that according the concept proposed in panel (A), polymer-based multi-drug targeting decreased the number of blood vessels and increased the number of apoptotic cells.

Figure adapted, with permission, from [39,45].

210 it is expected that such image-guided therapeutic interventions will gain more and more popularity in the years to come. With regard to nanomedicine-based combination therapies, several advanced formulations have already entered the initial phases of clinical evaluation. Mayer and coworkers, for instance, have co-encapsulated (optimal ratios of) doxorubicin and vincristine, irinotecan and floxuridine, and daunorubicin and cytarabine into liposomes [81—83], and they are currently evaluating the potential of the latter two formulations in patients. Comparable combination constructs can be generated using polymers and micelles, and also antibodies with intrinsic antitumor activity, such as Erbitux and Herceptin, can in principle be relatively easily further functionalized with low molecular weight drugs. It is important in this regard to carefully select the pharmacologically active agents which are used in such systems, in order to ensure synergistic therapeutic activity when administered in vivo. The HPMA copolymer-based nanomedicine formulation mentioned above, for instance, which contains comparable amounts of doxorubicin and gemcitabine, might not be the most ideal system to transfer to the clinic. This because the intrinsic sensitivity of cancer cells to doxorubicin and to gemcitabine varies quite substantially, with for gemcitabine, IC50 -values ranging from pico- to nanomoles, and for doxorubicine, IC50 -values ranging from nano- to micromoles. The above study should therefore be considered more as a proof-of-principle approach, in order to demonstrate that the simultaneous delivery of doxorubicin and gemcitabine to tumors in vivo is feasible, rather than as a directly clinically relevant approach. The combination of doxorubicin with aminoglutethimide, on the other hand, would be much more rational in this regard, as it has already been show to produce highly synergistic cytotoxic effects in vitro [38,70]. The results of vivo experiments in which the efficacy and the tolerability of this combination conjugate are compared to those of the combination of free doxorubicin plus free aminoglutethimide, as well as to those of the combination of polymer-bound doxorubicin plus polymerbound aminoglutethimide, are therefore eagerly awaited. Alternative combinatorial efforts involving targeted nanomedicines are also strongly encouraged. Besides the abovementioned formulations, for instance, which are designed to deliver two different drugs to tumors simultaneously, also rational combinations of one-drugcontaining nanomedicine systems might lead to substantial improvements in therapeutic efficacy. The combination of liposomal corticosteroids, which primarily affect tumorassociated macrophages and inhibit angiogenesis [84,85], with polymer-based chemotherapeutic agents, which primarily kill cancer cells, would for instance be an interesting approach in this regard. By at the same time also adding in radiotherapy, or certain specific monoclonal antibodies [86,87] or polymer- or lipid-based siRNA formulations [88—90], one could take this approach even one step further, all with the aim of eradicating even more cancer and endothelial cells, and of attacking the disease from various different angles. Taking the above insights and advances into account, it can be concluded that both polymeric and several other types of nanomedicine systems hold significant potential for improving the treatment of advanced solid malignancies.

T. Lammers et al. These formulations can not only be used for improving the tumor-directed delivery of low molecular weight (chemo) therapeutic agents, and to thereby beneficially affect the balance between the efficacy and the toxicity of the intervention, but also for combining disease diagnosis and therapy, and for improving the efficacy of combined modality anticancer therapy. There is still quite some work that needs to be done, however, before such advanced formulations can be administered to patients, but given the fact that ever more academic, industrial and governmental research initiatives are being launched to develop nanotechnological means to detect, to diagnose and to treat cancer, and that ever more scientific disciplines are cooperating to realize the potential of personalized nanomedicine, we are confident that significant progress will be made in this area of research in the years to come.

Acknowledgements This work was supported by the German-Israeli Cooperation Program in Cancer Research, by the Wieland-Stiftung, by the Program of Research Centers of the Ministry of Education, Youth and Sports of the Czech Republic (IM4635608802), and by the European Commission (FP6-MediTrans).

References [1] F.X. Gu, R. Karnik, A.Z. Wang, F. Alexis, E. Levy-Nissenbaum, S. Hong, et al., Nano Today 2 (2007) 14. [2] D. Peer, J.M. Karp, S. Hong, O. Farokhzad, R. Margalit, R. Langer, Nat. Nanotechnol. 2 (2007) 751. [3] M.E. Davis, Z.G. Chen, D.M. Shin, Nat. Rev. Drug Discov. 7 (2008) 771. [4] T. Lammers, W.E. Hennink, G. Storm, Br. J. Cancer 99 (2008) 392. [5] L.W. Seymour, Crit. Rev. Ther. Drug Carrier Syst. 9 (1992) 135. [6] R. Langer, Nature 392 (1998) S5. [7] V.P. Torchilin, Eur. J. Pharm. Sci. 11 (2000) S81. [8] T.M. Allen, P.R. Cullis, Science 303 (2004) 1818. [9] V.P. Torchilin, Nat. Rev. Drug Discov. 4 (2005) 145. [10] D.C. Drummond, O. Meyer, K. Hong, D.B. Kirpotin, D. Papahadjopoulos, Pharm. Rev. 51 (1999) 691. [11] J. Kopecek, P. Kopeckova, T. Minko, Z. Lu, Eur. J. Pharm. Biopharm. 50 (2000) 61. [12] R. Duncan, Nat. Rev. Cancer 6 (2006) 688. [13] K. Kataoka, H. Harada, Y. Nagasaki, Adv. Drug Deliv. Rev. 41 (2001) 113. [14] V.P. Torchilin, Pharm. Res. 24 (2007) 1. [15] K. Kostarelos, A. Bianco, M. Prato, Nat. Nanotechnol. 4 (2009) 627. [16] R. Gref, Y. Minamitake, M.T. Peracchia, V. Trubetskoy, V. Torchilin, R. Langer, Science 263 (1994) 1600. [17] C. Loo, A. Lowery, N. Halas, J. West, R. Drezek, Nano Lett. 5 (2005) 709. [18] J.A. MacDiarmid, N.B. Mugridge, J.C. Weiss, L. Phillips, A.L. Burn, R.P. Paulin, et al., Cancer Cell 11 (2007) 431. [19] S. Sengupta, D. Eavarone, I. Capila, G. Zhao, N. Watson, T. Kiziltepe, et al., Nature 436 (2005) 568. [20] M.A. Moses, H. Brem, R. Langer, Cancer Cell 4 (2003) 337. [21] W.J. Gradishar, S. Tjulandin, N. Davidson, H. Shaw, N. Desai, P. Bhar, et al., J. Clin. Oncol. 23 (2005) 7794. [22] L. Paz-Ares, H. Ross, M. O’Brien, A. Riviere, U. Gatzemeier, J. Von Pawel, et al., Br. J. Cancer 98 (2008) 1608.

Image-guided polymeric nanomedicines for combination therapy [23] P.A. Vasey, S.B. Kaye, R. Morrison, C. Twelves, P. Wilson, R. Duncan, et al., Clin. Cancer Res. 5 (1999) 83. [24] N. Nishiyama, A. Nori, A. Malugin, Y. Kasuya, P. Kopeckova, J. Kopecek, Cancer Res. 63 (2003) 7876. [25] R. Satchi-Fainaro, M. Puder, J.W. Davies, H.T. Tran, D.A. Sampson, A.K. Greene, et al., Nat. Med. 10 (2004) 255. [26] V. Omelyanenko, P. Kopeckova, C. Gentry, J. Kopecek, J. Control. Rel. 53 (1998) 25. [27] D.B. Pike, H. Ghandehari, Adv. Drug Deliv. Rev. 62 (2010) 167. [28] B. Rihova, M. Kovar, Adv. Drug Deliv. Rev. 62 (2010) 184. [29] T. Etrych, M. Jelinkova, B. Rihova, K. Ulbrich, J. Control. Release 73 (2001) 89. [30] K. Ulbrich, V. Subr, Adv. Drug Deliv. Rev. 56 (2004) 1023. [31] K. Ulbrich, V. Subr, Adv. Drug Deliv. Rev. 62 (2010) 150. [32] T. Minko, Adv. Drug Deliv. Rev. 62 (2010) 192. [33] K.D. Fisher, L.W. Seymour, Adv. Drug Deliv. Rev. 62 (2010) 240. [34] F. Kiessling, M. Heilmann, T. Lammers, K. Ulbrich, V. Subr, P. Peschke, et al., Bioconjugate Chem. 17 (2006) 42. [35] Z.R. Lu, Adv. Drug Deliv. Rev. 62 (2010) 246. [36] M. Talelli, C.J.F. Rijcken, C.F. van Nostrum, G. Storm, W.E. Hennink, Adv. Drug Deliv. Rev. 62 (2010) 231. [37] X.M. Liu, S.C. Miller, D. Wang, Adv. Drug Deliv. Rev. 62 (2010) 258. [38] F. Greco, M.J. Vicent, Adv. Drug Deliv. Rev. 61 (2009) 1203. [39] T. Lammers, Adv. Drug Deliv. Rev. 62 (2010) 203. [40] M.J. Vicent, H. Ringsdorf, R. Duncan, Adv. Drug Deliv. Rev. 61 (2009) 1117. [41] R. Duncan, Adv. Drug Deliv. Rev. 61 (2009) 1131. [42] R. Duncan, M.J. Vicent, Adv. Drug Deliv. Rev. 62 (2010) 272. [43] T. Lammers, R. Kühnlein, M. Kissel, V. Subr, T. Etrych, R. Pola, et al., J. Control. Release 110 (2005) 103. [44] L.W. Seymour, Y. Miyamoto, H. Maeda, M. Brereton, J. Strohalm, K. Ulbrich, et al., Eur. J. Cancer 31 (1995) 766. [45] T. Lammers, V. Subr, K. Ulbrich, P. Peschke, P.E. Huber, W.E. Hennink, et al., Biomaterials 30 (2009) 3466. [46] M.P. Borgman, T. Coleman, R.B. Kolhatkar, S. Geyser-Stoops, B.R. Line, H. Ghandehari, J. Control. Release 132 (2008) 193. [47] T. Lammers, P. Peschke, R. Kühnlein, V. Subr, K. Ulbrich, P. Huber, et al., Neoplasia 8 (2006) 788. [48] T. Lammers, P. Peschke, R. Kühnlein, V. Subr, K. Ulbrich, J. Debus, et al., J. Control. Release 117 (2007) 333. [49] J.G. Shiah, M. Dvorak, P. Kopeckova, Y. Sun, C.M. Peterson, J. Kopecek, Eur. J. Cancer 37 (2001) 131. [50] E. Kominami, T. Tsukahara, Y. Bando, N. Katunuma, J. Biochem. 98 (1985) 87. [51] P.A. Flanagan, P. Kopeckova, J. Kopecek, R. Duncan, Biochim. Biophys. Acta 993 (1989) 83. [52] A. Nori, J. Kopecek, Adv. Drug Deliv. Rev. 57 (2005) 609. [53] T. Etrych, P. Chytil, T. Mrkvan, M. Sirova, B. Rihova, K. Ulbrich, J. Control. Release 132 (2008) 184. [54] J.S. Park, L. Qiao, Z.Z. Su, D. Hinman, K. Willoughby, R. McKinstry, et al., Oncogene 25 (2001) 3266. [55] Y.J. Lee, S.S. Galoforo, C.M. Berns, G. Erdos, A.K. Gupta, D.K. Ways, et al., J. Biol. Chem. 270 (1995) 28790. [56] F. Samaniego, P.D. Markham, R. Gendelman, Y. Watanabe, V. Kao, K. Kowalski, et al., Am. J. Pathol. 152 (1998) 1433. [57] M. Garcia-Barros, F. Paris, C. Cordon-Cardo, D. Lyden, S. Rafii, A. Haimovitz-Friedman, et al., Science 300 (2003) 1155. [58] C.A. Znati, M. Rosenstein, Y. Boucher, M.W. Epperly, W.D. Bloomer, R.K. Jain, Cancer Res. 56 (1996) 964. [59] P. Peschke, V. Klein, G. Wolber, E. Friedrich, E.W. Hahn, Histol. Histopathol. 14 (1999) 461. [60] C. Li, S. Ke, Q.P. Wu, W. Tansey, N. Hunter, L.M. Buchmiller, et al., Clin. Cancer Res. 6 (2000) 2829. [61] C.L. Davies, L.M. Lundstrom, J. Frengen, L. Eikenes, O.S. Bruland, O. Kaalhus, et al., Cancer Res. 64 (2004) 47. [62] D. Hallahan, L. Geng, S. Qu, C. Scarfone, T. Giorgio, E. Donnelly, et al., Cancer Cell 3 (2003) 63.

211

[63] G. Kong, R.D. Braun, M.W. Dewhirst, Cancer Res. 60 (2000) 4440. [64] G. Kong, R.D. Braun, M.W. Dewhirst, Cancer Res. 61 (2001) 3027. [65] T. Lammers, V. Subr, P. Peschke, R. Kühnlein, W.E. Hennink, K. Ulbrich, et al., Br. J. Cancer 99 (2008) 900. [66] T. Dipetrillo, L. Milas, D. Evans, P. Akerman, T. Ng, T. Miner, et al., Am. J. Clin. Oncol. 29 (2006) 376. [67] L. Milas, K. Mason, N. Hunter, C. Li, S. Wallace, Int. J. Radiat. Oncol. Biol. Phys. 55 (2003) 707. [68] N. Wiedenmann, D. Valdecanas, N. Hunter, S. Hyde, T.A. Buchholz, L. Milas, et al., Clin. Cancer Res. 131 (2007) 868. [69] K.J. Harrington, G. Rowlinson-Busza, K.N. Syrigos, R.G. Vile, P.G. Uster, A.M. Peters, et al., Clin. Cancer Res. 6 (2000) 4939. [70] M.J. Vicent, F. Greco, R.I. Nicholson, A. Paul, P.C. Griffiths, R. Duncan, Angew. Chem.-Int. Edit. 44 (2005) 4061. [71] A.K. Gupta, M. Gupta, Biomaterials 26 (2005) 3995. [72] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, et al., Nano Lett. 6 (2006) 2427. [73] M. Talelli, C.J. Rijcken, T. Lammers, P.R. Seevinck, G. Storm, C.F. van Nostrum, et al., Langmuir 25 (2009) 2060. [74] V. Bagalkot, L. Zhang, E. Levy-Nissenbaum, S. Jon, P.W. Kantoff, R. Langer, et al., Nano Lett. 7 (2007) 3065. [75] H. Zhang, D. Yee, C. Wang, Nanomedicine 3 (2008) 83. [76] P. Zrazhevskiy, X. Gao, Nano Today 4 (2009) 414. [77] W.T. Al-Jamal, K. Kostarelos, Nanomedicine 2 (2007) 85. [78] M. Arruebo, R. Fernandez-Pacheco, M.R. Ibarra, J. Santamaria, Nano Today 2 (2007) 22. [79] W.J. Mulder, G.J. Strijkers, G.A. van Tilborg, D.P. Cormode, Z.A. Fayad, K. Nicolay, Acc. Chem. Res. 42 (2009) 904. [80] S. Langereis, J. Keupp, J.L. van Velthoven, I.H. de Roos, D. Burdinski, J.A. Pikkemaat, et al., J. Am. Chem. Soc. 131 (2009) 1380. [81] L.D. Mayer, T.O. Harasym, P.G. Tardi, N.L. Harasym, C.R. Shew, S.A. Johnstone, et al., Mol. Cancer Ther. 5 (2006) 1854. [82] S.A. Abraham, C. McKenzie, D. Masin, R. Ng, T.O. Harasym, L.D. Mayer, et al., Clin. Cancer Res. 10 (2004) 728. [83] P.G. Tardi, R.C. Gallagher, S. Johnstone, N. Harasym, M. Webb, M.B. Bally, et al., Leuk. Res. 33 (2009) 129. [84] M. Banciu, J.M. Metselaar, R.M. Schiffelers, G. Storm, J. Steroid Biochem. Mol. Biol. 111 (2008) 101. [85] M. Banciu, J.M. Metselaar, R.M. Schiffelers, G. Storm, Neoplasia 10 (2008) 108. [86] G.P. Adams, L.M. Weiner, Nat. Biotechnol. 23 (2005) 1147. [87] R.M. Sharkey, D.M. Goldenberg, CA Cancer J. Clin. 56 (2006) 226. [88] R.M. Schiffelers, A. Ansari, J. Xu, Q. Zhou, Q. Tang, G. Storm, et al., Nucleic Acids Res. 32 (2004) e149. [89] Y.H. Huang, Y. Bao, W. Peng, M. Goldberg, K. Love, D.A. Bumcrot, et al., Proc. Natl. Acad. Sci. U.S.A. 106 (2009) 3426. [90] A.D. Judge, M. Robbins, I. Tavakoli, J. Levi, L. Hu, A. Fronda, et al., J. Clin. Invest. 119 (2009) 661. Twan Lammers studied pharmacy at the University of Utrecht. He obtained a DSc degree in Radiation Oncology from Heidelberg University in 2008 (‘Role of PP2Ca in growth regulation, in cellular stress signaling and in tumorigenesis’), and a PhD degree in pharmaceutics from Utrecht University in 2009 (‘Drug targeting to tumors using HPMA copolymers’). Since 2007, he has been appointed as a postdoctoral fellow at the Department of Pharmaceutics at Utrecht University, where he predominantly works on the MediTrans project (FP6: ‘Targeted delivery of nanomedicines’). Since 2009, he has also been appointed as a junior group leader at the Department of Experimental Molecular Imaging at the University of Aachen (RWTH). His primary research

212 interests include drug targeting to tumors, image-guided drug delivery and tumor-targeted combination therapy. Vladimir ˇ Subr obtained his master degree in polymer science from Institute of Chemical Technology Prague in 1981. He earned his PhD degree in polymer science from the Institute of Macromolecular Chemistry of the Czechoslovak Academy of Sciences in 1986. He was a postdoctoral fellow with Professor Ruth Duncan at the University of Keele in 1987. From 1988 to 1992, he was a research scientist in Institute of Macromolecular Chemistry of the Czechoslovak Academy of Sciences and from 1992 till now he is a senior research scientist in the Laboratory of Biomedical Polymers, Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic. Karel Ulbrich is a head of the Department of Biomedical Polymers at the Institute of Macromolecular Chemistry, Academy of Sciences of the Czech Republic and professor of macromolecular chemistry at Chemical Technology University in Prague. He received his MS degree in polymer chemistry from the Prague Institute of Polymer Technology in 1970, his PhD and DSc degree in macromolecular sciences from Academy of Sciences of the Czech Republic in 1975 and 2000. He acted as a visiting lecturer at the Johannes-Guttenberg University, Mainz in 1983 and as a visiting professor at the University of Utah (in 1989 and 1992). Karel Ulbrich is a member of the Learned Society of the Czech Republic and vice president of the Scientific Board of the Academy of Sciences of the Czech Republic. His research interests have focused on biocompatible hydrophilic polymers and copolymers for biomedical applications, development of polymer drug conjugates, targeting of drugs and development of polymer gene delivery vectors. He is the author or co-author of more than 220 papers published in peer-reviewed journals and 33 patents, mostly from the area of controlled drug delivery. Professor Wim E. Hennink obtained his PhD degree in 1985 at the Twente University of Technology on a thesis on a biomaterials topic. From 1985 until 1992 he had different positions in the industry. In 1992 he was appointed as professor at the Faculty of Pharmacy of the University of Utrecht. From 1996 on he is head of the Department of Pharmaceutics. Since 2010 he is vice dean of the Faculty of Science. His main research interests are in the field of polymeric drug delivery systems. He published over 350 papers and book chapters and is the inventor of 20 patents.

T. Lammers et al. Professor Gert Storm studied biology at the Utrecht University, The Netherlands. He graduated in 1983. He obtained his PhD degree in 1987 at the Dept. of Pharmaceutics of the same university. In 1988—1989 he was a visiting scientist at Liposome Technology Inc. in Menlo Park, USA, and visiting assistant professor at the School of Pharmacy, UCSF, San Francisco. In 1990—1991 he was senior research scientist at Pharma Bio-Research Consultancy B.V. in Zuidlaren, The Netherlands. During this period he contributed to the design, co-ordination and evaluation of clinical pharmacological studies. In September 1991 he took up his present position. In 2000, he was appointed as a professor (Drug Targeting Chair) at Utrecht University. In addition to this, he was appointed an adjunct professor at the Royal School of Pharmacy, Copenhagen, in 1999, and from July 2009 onwards, he is an honorary professor in biomacromolecular drug delivery at the University of Copenhagen. He is the author/co-author of more than 300 original articles, reviews and book chapters, in the field of advanced drug delivery/drug targeting (in particular with liposomal systems), and theme (co-)editor of Advanced Drug Delivery Reviews and the book ‘‘Long Circulating Liposomes. Old Drugs, New Therapeutics’’. Fabian Kiessling studied Medicine and did his thesis at the University of Heidelberg. Until the end of 2002, he worked as a resident at the Department of Radiology at the German Cancer Research Center (DKFZ) in Heidelberg. In 2003, he changed to the Department of Medical Physics in Radiology at the DKFZ, as a leader of the Molecular Imaging Group. In parallel, he did his clinical training at the University of Heidelberg and received the board certification as a radiologist. Fabian Kiessling finished his habilitation in experimental radiology in 2006. He is the author of more than 80 publications and book chapters, and received several research awards. Since 2008, he is heading the Department of Experimental Molecular Imaging at the Helmholtz Center of Applied Engineering of the RWTH-University in Aachen. With a particular focus on angiogenesis, the aim of his research is the development of novel diagnostic probes and imaging tools for disease-specific diagnosis and therapy monitoring.