Vascular targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions: A preliminary study

Photodiagnosis and Photodynamic Therapy (2012) 9, 109—117

Available online at www.sciencedirect.com

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

Vascular targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions: A preliminary study Haixia Qiu a,1, Yongping Mao b,1, Ying Gu MD a,∗, Ying Wang a, Jianguo Zhu a, Jing Zeng a a

Department of Laser Medicine, Chinese PLA General Hospital, No. 28 Fuxing Road, Beijing 100853, China Department of Gastroenterology and Hepatology, Chinese PLA General Hospital, No. 28 Fuxing Road, Beijing 100853, China Available online 16 December 2011

b

KEYWORDS Gastrointestinal; Vascular lesions; Photodynamic therapy; Vascular-targeted

∗ 1

Summary Background: Vascular-targeted photodynamic therapy (V-PDT) has shown good selectivity and efficacy in the treatment of certain types of vascular disease, including port wine stains, agerelated macular degeneration, and esophageal varices. This study was conducted to test the efficacy and safety of V-PDT in the treatment of gastrointestinal (GI) bleeding caused by the abnormal dilatation of capillaries. Methods: Patients with bleeding GI mucosal vascular lesions treated with V-PDT were included in this retrospective study. The efficiency of V-PDT was analyzed by comparing the documented endoscopy results, hemoglobin levels, and transfusion requirements before and at 6 months after the last V-PDT. Side effects during and after V-PDT and follow-up results were also analyzed. Results: Seven patients with bleeding GI mucosal vascular lesions were treated with V-PDT. After 1—4 V-PDT sessions, all patients no longer needed transfusions to maintain a stable hemoglobin level during the follow-up period of 6 months. The mean hemoglobin level of the seven patients improved from 6.21 ± 1.48 g/dl to 11.66 ± 1.21 g/dl (p < 0.001), and the GI bleeding and melena of all the patients disappeared. No perforations, strictures, scars, or episodes of photosensitization occurred in the seven patients, and there were no recurrences of GI bleeding during the 1—21 months of further follow-up. Conclusions: This preliminary study indicated that V-PDT is a highly selective, safe, welltolerated, and effective treatment modality for bleeding GI mucosal vascular lesions. However, prospective studies with larger sample sizes are needed to confirm this finding. © 2011 Elsevier B.V. All rights reserved.

Corresponding author. Tel.: +86 10 6693 9394; fax: +86 10 6822 2584. E-mail addresses: [email protected] (H. Qiu), [email protected] (Y. Mao), [email protected] (Y. Gu). These authors contributed equally to this work.

1572-1000/$ — see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.pdpdt.2011.11.003

110

Introduction Many gastrointestinal (GI) mucosal vascular lesions, including gastric antral vascular ectasia (watermelon stomach), various vascular malformations, and radiation telangiectasias (including radiation gastritis, radiation gastroenteritis, and radiation proctitis), can cause recurrent GI bleeding and lead to anemia. Endoscopic therapy has come to the forefront in the treatment of such disease and has included Nd:YAG laser photocoagulation therapy [1], bipolar electrocoagulation [2], heat probe [3], and in recent years, argon plasma coagulation (APC) [4]. In the methods listed above, thermal effects are used to cause nonselective damage to the target tissue [5]. Unfortunately, it is very difficult to treat diffuse lesions and vessels too small to be recognized during endoscopy with these methods. Photodynamic therapy (PDT) differs quite substantially from these techniques in that it does not involve any thermal or mechanical damage. It is based on the concept that a photosensitizer (PS) can be excited by light irradiation to produce activated oxygen that causes irreversible oxidative damage to intracellular target molecules. PDT has dual selectivity in that the diseased tissue can preferentially take up PS and PS activation can be confined to the diseased tissue by restricting the illumination to a specific region [6]. Therefore, PDT leaves the normal tissue intact, while selectively destroying the diseased tissue. PDT was originally developed as a cancer regimen and has been approved for the treatment of multiorgan tumors in many countries. Extensive studies have shown that PDT damages tumors not only through direct tumor cell killing, but also by the destruction of the tumor-nourishing microvasculature [7]. Recently, there has been growing interest in PDT-induced microvasculature damage [8,9]. Vascular-targeted photodynamic therapy (V-PDT) is based on the concept that the level of PDT-induced blood vessel damage is correlated to the serum level of PS [10]. As the biodistribution of PS changes rapidly with time, irradiating when the PS vascular tissue to normal tissue ratio is near its maximal level can lead to PDT selectively damaging the target tissue. V-PDT was first introduced by Gu et al. [11] in China in 1991 and has been successfully used in China for the treatment of port wine stains (PWS), congenital vascular malformations characterized by multiple dilated vessels in the dermis. Subsequently, V-PDT has been widely used outside China as the treatment for exudative age-related macular degeneration (AMD) [12], a condition in which the choroidal neovascularization causes severe irreversible central vision loss. V-PDT can selectively occlude both the dilated vessels in PWS in the dermis and the choroidal neovasculature in AMD without damaging the overlying epidermis or retina, respectively [11,13]. More recently, V-PDT has shown promise as a treatment of esophageal varices [14]. In this study, we conducted a retrospective study to test the safety and efficacy of V-PDT in the treatment of bleeding GI mucosal lesions caused by the abnormal dilatation of capillaries.

H. Qiu et al.

Materials and methods Study design Seven Chinese patients who underwent V-PDT in the Laser Medicine Department of the Chinese PLA General Hospital from September 2005 to January 2011 for the treatment of diffuse mucosal vascular lesions were included in this retrospective study, and the medical records for all seven were reviewed.

Patients The seven patients were admitted for iron-deficiency anemia, gastrointestinal bleeding, or melena. Among the seven patients, two were diagnosed with gastric antral vascular ectasias, one had a vascular malformation of the rectum (diffuse telangiectasia), one had radiation gastritis, one had diffuse radiation gastroenteritis, and two had radiation proctitis. All patients had overt or occult bleeding accompanied by moderate to severe anemia and had been treated with transfusions, hemostasis, or iron supplement therapy (for the anemia) without success. None of the patients had received endoscopic therapy previously, and informed written consent was obtained from each patient.

V-PDT protocol A domestically produced photosensitizer Photocarcinorin (PSD-007, Institute of Pharmacochemistry of the Second Military Medical University, Shanghai, China) was used in this study. Photocarcinorin is derived from hematoporphyrin derivatives (HpD) and contains a mixture of six different porphyrins. The chemical composition of Photocarcinorin and the structures of its main components have been described in detail by Xu [15]. The stock solution of Photocarcinorin was at a concentration of 100 mg/10 ml and kept in the dark at −20 ◦ C. Before V-PDT was performed on a patient, a PSD-007 skin test was performed. All seven patients showed no reactivity to PSD-007. Patients fasted overnight prior to the procedure, and for patients who required a colonoscopy, an enema was prepared. A standard endoscopic examination was performed carefully with a gastroscope (GIF-H260; Olympus) or a colonoscope (CF-H260AI; Olympus) to identify the lesions. A flexible quartz fiber was then inserted through the endoscopic biopsy channel and positioned along the lesions to be treated. A cylindrical fiber (2.0 cm or 3.0 cm, Medlight, S.A. Switzerland) or a flat cut fiber (Medlight, S.A. Switzerland) was used according to the location and characteristics of a particular lesion. Next, Photocarcinorin (100 mg/10 ml) was intravenously injected at a dose of 5 mg/kg body weight (injection usually took 5 min to complete). Laser irradiation was initiated immediately after the end of Photocarcinorin administration. A 5 W continuous-wave diode laser (Beijing Newraysing Laser Tech Co., Ltd., China) with a wavelength of 532 nm was used as the PDT light source. Laser power density varied between 60 and 150 mW/cm2 . For lesions in which multiple spots were illuminated, V-PDT was usually started at the lower end of the lesion due to the fear

Vascular-targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions Table 1

111

Characteristics of the patients with bleeding gastrointestinal mucosal vascular lesions.

Case no.

Sex

Age (years)

Co-existing medical conditions

Diagnosis

1

F

52

—

2

F

42

—

3

F

79

4

M

56

Chronic renal failure, hepatic cirrhosis, hypertension Diabetes, hepatic cirrhosis, hepatic cancer

Radiation proctitis (due to radiotherapy after operation for cervical cancer) Vascular malformation of the rectum (diffuse telangiectasia) Gastric antral vascular ectasias

5 6

M M

58 63

7

F

53

Hepatic cirrhosis Pancreatic neuroendocrine carcinoma accompanied with peritoneal and neck nymph node metastasis —

that edema could occur in the irradiation area. Six of the patients tolerated the V-PDT treatment well without general anesthesia or sedation, and the other patient (Case #6 in Table 2) was treated under general anesthesia. After the intravenous administration of the Photocarcinorin, patients were instructed to avoid direct exposure to sunlight for at least 30 days. For all patients, V-PDT was conducted until the dilated vessel markedly darkened and the irradiated GI mucosa became edematous during each session. For patients overt bleeding, the bleeding of the treated area usually ceased at 5—10 min since the beginning of V-PDT. The exposure time of an irradiated area varied from 10 to 30 min. If bleeding continued after a treatment session of V-PDT or if a lesion was so diffuse that it was too large to be irradiated in one PDT session, another treatment session was conducted. In such cases, the interval between V-PDT treatments was 1—2 months. Patients with upper GI lesions were given an oral proton pump inhibitor for at least 2 weeks after V-PDT, and blood transfusions were given during V-PDT sessions.

Evaluation of efficiency The efficiency of the V-PDT treatment was evaluated by reviewing the medical records in order to compare the documented endoscopy results, the hemoglobin levels, and the transfusion requirements of the seven patients before and at 6 months after the last V-PDT treatment session. Further follow-up results were also collected both from the medical records and through contacting the patients. The criteria for a successful treatment were the cessation of GI bleeding and the maintenance of a stable hemoglobin level without transfusions. The side effects during and post V-PDT were also recorded.

Radiation gastritis (due to radiotherapy after operation for liver cancer) Gastric antral vascular ectasias Radiation gastroenteritis (due to radiotherapy after operation for pancreatic neuroendocrine carcinoma) Radiation proctitis (due to radiotherapy after operation for cervical cancer)

Statistical analysis SPSS (version 13.0) software was used for the statistical analysis. The data is presented as mean ± SD. Differences in hemoglobin levels before and at 6 months after the last V-PDT were analyzed with the non-parametric Mann—Whitney’s U test (Wilcoxon rank-sum test). Differences were considered statistically significant when p < 0.05.

Results The patients’ characteristics are listed in Table 1. Of the seven patients, four were females, and three were males. The average age of the patients was 57.57 ± 11.44 (range 42—79) years. All seven patients were successfully treated by V-PDT, and the mean number of treatment sessions required for the cessation of bleeding was 1.86 ± 1.21 sessions. Among the patients, four patients required one treatment session, one needed two sessions, one needed three sessions, and one required four sessions. The V-PDT treatment parameters are shown in detail in Table 2. Endoscopy examinations at 1 week post-V-PDT showed that the dilated vessels had partially disappeared, there was thrombosis in parts of the dilated vessels, and there was a small amount of superficial ulceration in two of the patients. At 4—6 weeks after V-PDT, there was complete healing of the GI mucosa without any evidence of scarring. At 6 months after the last V-PDT session, endoscopic examinations revealed that the dilated vessels had almost disappeared in five of the seven patients (Fig. 1) and were significantly decreased in the other two patients (Fig. 2). Before the V-PDT treatment, all seven patients had moderate to severe anemia and had been treated with transfusions, hemostasis, and iron supplements without success. The mean transfusion requirements were 9.43 ± 6.29

112

Table 2

Details of the V-PDT treatment parameters and follow-up. Hemoglobin level (g/dl) before and 6 months after the last V-PDT session

V-PDT session

Power density (mW/cm2 )

Fiber type

Irradiation time (min)

Irradiation area

Bleeding

Time of follow-up (months)

1

7.6/11.2

1

100

Flat cut fiber

D1: 20 D2: 20 D3: 20

Overt

27a

2

7.5/13.4

1

150

Flat cut fiber

D1: 15 D2: 10 D3: 15

Overt

24

3

6.8/12

1

60

D1: 20 D2: 15

Occult

22

4

5.6/12.1

1

80

D1: 2.0-cm Cylindrical fiber D2: flat cut fiber 2.0-cm Cylindrical fiber

D1: rectal lesions at 20 cm from anus D2: rectal lesions at 16 cm from anus D3: rectal lesions at 5 cm from anus D1: rectal lesions at 10 cm from anus D2: rectal lesions at 8 cm from the anus D3: rectal lesions at 6 cm from the anus D1: gastric body D2:gastric antrum

Overt

14b

1

100

30

Occult

12

2 1

150 100

2.0-cm Cylindrical fiber Flat cut fiber D1: 2.0-cm cylindrical fiber D2: 2.0-cm cylindrical fiber D3: flat cut fiber 2.0-cm Cylindrical fiber D1: 2.0-cm cylindrical fiber D2: 2.0-cm cylindrical fiber D3: flat cut fiber

Overt

12

5

6

3.6/10.6

5.2/9.8

2

80

3

100

D1: 15 D2: 20

D1: pylorus and duodenum bulb D2: gastric antrum and gastric body Gastric antrum

30 D1: 15 D2: 20 D3: 20

Gastric antrum D1: duodenal bulb and descending part D2: gastric body D3: gastric antrum

20

Gastric antrum and duodenal bulb D1: duodenal bulb and descending D2: gastric antrum and pylorus D3: lesser curvature of gastric antrum

D1: 20 D2: 20 D3: 30

H. Qiu et al.

Case no.

Case no.

7

Hemoglobin level (g/dl) before and 6 months after the last V-PDT session

V-PDT session

Power density (mW/cm2 )

Fiber type

Irradiation time (min)

Irradiation area

1

100

3.0-cm Cylindrical fiber

D1: 20 D2: 20 D3: 20

D1: rectal lesions at 10 cm from anus D2: rectal lesions at 7 cm from anus D3: rectal lesions at 4 cm from anus D1: rectal lesions at 10 cm from anus D2: rectal lesions at 4 cm from anus D1: rectal lesions at 10 cm from anus D2: rectal lesions at 7 cm from anus D3: rectal lesions at 4 cm from anus D1: rectal lesions at 7 cm from anus D2: rectal lesions at 4 cm from the anus

7.2/12.5

2

100

2.0-cm Cylindrical fiber

D1: 20 D2: 20

3

100

2.0-cm Cylindrical fiber

D1: 15 D2: 10 D3: 20

4

100

2.0-cm Cylindrical fiber

D1: 15 D2: 15

Bleeding

Time of follow-up (months)

Overt

7

Vascular-targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions

Table 2 (Continued)

Note: D1, first spot of illumination; D2, second spot of illumination; D3, third spot of illumination; V-PDT, vascular-targeted photodynamic therapy. a Patient lost to follow-up in January 2008. b Patient died of liver cancer in December 2010.

113

114

H. Qiu et al.

Figure 1 Endoscopic images of a 58-year-old male patient (Case #5) with gastric antral vascular ectasias before (A), during (B), and at 6 months after (C) two treatment sessions of V-PDT.

Figure 2 Endoscopic images of a 53-year-old female patient (Case #7) with radiation proctitis before (A), during (B), and at 6 months after (C) after four treatment sessions of V-PDT.

Vascular-targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions units of packed red cells (number of transfusion packed red cells units per month). After the 1—4 treatments sessions, the patients no longer required transfusions to maintain a stable hemoglobin level (as measured at the 6-month follow-up). The mean hemoglobin level before and at 6 months after the last treatment session was 6.21 ± 1.48 g/dl and 11.66 ± 1.21 g/dl (p < 0.001, Table 1), respectively. Furthermore, melena was completely absent in all patients at 6 months after the last treatment. The mean follow-up time of the seven patients was 16.14 ± 8.15 months (7—27 months), and none of them had any episodes of recurrent bleeding or melena during this period. The V-PDT treatment was well tolerated by all seven patients. No fevers, signs of cutaneous photosensitization, perforations, strictures, or scars were found in any of the seven patients during the 7—27 months of follow-up. During the follow-up period, one patient died of liver cancer 20 months after the treatment, and another was lost to follow-up 27 months post-V-PDT.

Discussion The drug-light interval (DLI) is a crucial factor for the therapeutic selectivity of PDT [9]. As PS is cleared at different rates from different tissues, irradiating at the time of the maximal PS diseased tissue to normal tissue ratio can increase the selectivity of PDT. A clinical pharmacokinetic study of PSD-007 in tumor patients showed that PSD-007 was at its highest blood concentration shortly after intravenous infusion and that the blood concentration dropped quickly with time. For example, at 6 min after infusion, the blood concentration of PSD-007 was found to be 24.62 ␮g/ml, but it dropped to 9.89 ␮g/ml 2 h post-infusion [16]. On the other hand, the concentration of PSD-007 in other tissues, such as muscle and tumors, slowly increased after administration and reached an apex at 6 h after administration in tumor-bearing mice [17]. Li et al. [18] found that while Photofrin was mainly localized in blood vessels shortly after injection (less than 30 min), it was mostly concentrated in the skin and tumors in tumor-bearing mice after 6 h. When given at an early time point (<30 min), PDT caused significant destruction of the tumor blood vessels. Conversely, when PDT was administered at the later time point (>6 h), it was the tumor cells and overlying skin that were the tissues severely damaged. Kurohane et al. [8] showed that while BPD-MA-mediated PDT caused the complete blockage of blood flow in the neovasculature when the irradiation occurred 15 min after injection (15-min PDT), it did not inhibit blood flow when the irradiation occurred 3 h after the injection (3-h PDT). The 15-min PDT was found to strongly suppress tumor growth, perhaps through damaging endothelial cells in the tumor neovasculature. This is in contrast with the 3-h PDT, which is believed to have been directly cytotoxic to the tumor cells. Hence, in traditional PDT, the DLI often lasts from 24 to 72 h because that is the time frame in which PS has its best tumor to normal tissue ratio. On the other hand, in V-PDT, the DLI is usually shorter than 30 min [8]. It was revealed that the plasma concentration of PS is at

115

its peak shortly after intravenous administration and that PS is quickly taken up by endothelial cells after it is infused [19]. When the irradiation is initiated at such an early time point, V-PDT caused selective damage on vascular endothelial cells that resulted in blood flow stasis, followed by thrombosis, vascular occlusion, and eventually, the destruction of the abnormal microvasculature [20,21]. In our initial study on the subject, we used intravital microscopic observation to determine that the rat mesenteric microcirculation had differential sensitivity to PDT, with capillaries being the most sensitive part of the microvasculature. It was found that the stasis time of the capillaries was between 6 and 16 min at different types of PS-mediated PDT [22]. In a chicken comb microcirculation model, it was demonstrated that a DLI of zero (0-time PDT) led to PDT causing the most selective damage (vascular tissue damaged, peripheral tissue spared) [23]. In V-PDT, a relatively lower power density [11] or energy density [24] is used compared to that used in traditional PDT for tumor eradication [25]. However, for GI mucosal vascular lesions caused by the abnormal dilatation of capillaries, the proper PDT parameters have not been determined. Ido [26] reported that PDT administered to treat early esophageal carcinoma unintentionally cured coexisting esophageal varices in one patient. In this study, a 630-nm laser was utilized for illumination for 30 min with a power density of 360 J/cm2 . Li et al. [14] utilized PDT with the parameters of a laser power density of 150 mW/cm2 and an irradiation time of 40 min to successfully treat esophageal varices in 14 patients. A study on rabbit auricular veins demonstrated that at a power density of 300 mW/cm2 , laser illumination at 15 min post-PS injection had more significant photodynamic effects than illumination at 5 or 10 min [27]. V-PDT at 50—100 mW/cm2 and 90—540 J/cm2 has been shown to safely and efficiently treat PWS [28]. Based on the above studies, a wide range of power densities (60—150 mW/cm2 ) and doses (54—270 J/cm2 ) were tried in this study, and the preliminary results showed that a PDT protocol that used such a dose and power density could efficiently block a dilated GI vessel. As this study had a very small sample size, it is hard to draw any solid conclusions about the optimal power density and light dose to use in V-PDT. Our study showed that such a V-PDT regimen had good hemostatic effects. The active bleeding of the irradiation area was stopped, and the color of the dilated vessel became darker by the end of V-PDT treatment. These findings might be related to the occlusion of the target vessel during the irradiation, which suggests that the dilated GI vessels are very sensitive to V-PDT. At 6 months after the last V-PDT administration, endoscopic examination revealed that the dilated vessels had almost disappeared in the majority of the seven patients and that the residual dilated vessels showed no signs of overt or occult bleeding. All patients showed significant improvement in regard to their anemia and no longer required transfusions to maintain a stable hemoglobin level. Longer follow-ups (7—27 months) also showed no signs indicative of the recurrence of bleeding. Together, these results indicate that V-PDT can quickly induce persistent hemostasis.

116 As mentioned above, the therapeutic selectivity of V-PDT is based on transient differences in the PS vascular tissue to normal tissue concentration ratio. PS is at its greatest vascular tissue to normal tissue ratio shortly after intravenous administration, but this ratio decreases significantly over time [29]. Thus, V-PDT might cause non-selective damage to the surrounding normal tissue if the irradiation occurs more than a few hours after PS administration. Therefore, in this study, the V-PDT irradiation time of a treatment session did not last longer than 2 h. This V-PDT protocol caused significant damage to the abnormal vascular lesions but in the majority of patients led to only temporary edema in the overlying intestinal mucosa, which disappeared shortly after V-PDT. In two patients, a small number of superficial ulcerations formed in the mucosa, but these ulcerations were completely healed without evidence of scarring within 4 weeks after V-PDT. Furthermore, no serious adverse events occurred during the 7—27 months of follow-up after the last V-PDT. These results confirmed that this V-PDT regimen spared the peripheral tissue and thus had good selectivity for vascular tissue. In conclusion, this preliminary study indicates that V-PDT is a highly selective, safe, minimally invasive, well tolerated, and effective modality for the control of bleeding in GI mucosal vascular lesions. Studies with larger sample sizes are warranted to further confirm the efficacy and safety of V-PDT in the treatment of bleeding GI mucosal vascular lesions. Furthermore, the treatment parameters of V-PDT also need to be further optimized.

Acknowledgments This work was supported by the National High Technology Research and Development Program of China (No. 2008AA030117), the National Natural Science Foundation of China (No. 60878055), and the Key Program of National Natural Science Foundation of China (No. 61036014).

References [1] Liberski SM, McGarrity TJ, Hartle RJ, Varano V, Reynolds D. The watermelon stomach: long-term outcome in patients treated with Nd:YAG laser therapy. Gastrointest Endosc 1994;40:584—7. [2] Pavey DA, Craig PI. Endoscopic therapy for upper-GI vascular ectasias. Gastrointest Endosc 2004;59:233—8. [3] Jensen DM, Machicado GA, Cheng S, Jensen ME, Jutabha R. A randomized prospective study of endoscopic bipolar electrocoagulation and heater probe treatment of chronic rectal bleeding from radiation telangiectasia. Gastrointest Endosc 1997;45:20—5. [4] Naga M, Esmat S, Naguib M, Sedrak H. Long-term effect of argon plasma coagulation (APC) in the treatment of gastric antral vascular ectasia (GAVE). Arab J Gastroenterol 2011;12:40—3. [5] Ben-Soussan E, Antonietti M, Savoye G, Herve S, Ducrotté P, Lerebours E. Argon plasma coagulation in the treatment of hemorrhagic radiation proctitis is efficient but requires a perfect colonic cleansing to be safe. Eur J Gastroenterol Hepatol 2004;16:1315—8. [6] Sharman WM, Allen CM, van Lier JE. Photodynamic therapeutics: basic principles and clinical applications. Drug Discov Today 1999;4:507—17.

H. Qiu et al. [7] Seshadri M, Spernyak JA, Mazurchuk R, et al. Tumor vascular response to photodynamic therapy and the antivascular agent 5,6-dimethylxanthenone-4-acetic acid: implications for combination therapy. Clin Cancer Res 2005;11:4241—50. [8] Kurohane K, Tominaga A, Sato K, North JR, Namba Y, Oku N. Photodynamic therapy targeted to tumor-induced angiogenic vessels. Cancer Lett 2001;167:49—56. [9] Chen B, Pogue BW, Hoopes PJ, Hasan T. Vascular and cellular targeting for photodynamic therapy. Crit Rev Eukaryot Gene Expr 2006;16:279—305. [10] Menezes da Silva FA, Newman EL. Time-dependent photodynamic damage to blood vessels: correlation with serum photosensitizer levels. Photochem Photobiol 1995;61:414—6. [11] Gu Y, Li JH, Jiang YP, et al. A clinic study of photodynamic therapy for port wine stains: a report of 40 cases. Chin J Laser Med Surg 1992;1:6—10 [in Chinese]. [12] Verteporfin In Photodynamic Therapy Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in agerelated macular degeneration: two-year results of a randomized clinical trial including lesions with occult with no classic choroidal neovascularization — verteporfin in photodynamic therapy report 2. Am J Ophthalmol 2001;131:541—60. [13] Schmidt-Erfurth U, Hasan T. Mechanisms of action of photodynamic therapy with verteporfin for the treatment of age-related macular degeneration. Surv Ophthalmol 2000;45:195—214. [14] Li CZ, Cheng LF, Wang ZQ, Gu Y. Attempt of photodynamic therapy on esophageal varices. Lasers Med Sci 2009;24:167—71. [15] Xu DY. Research and development of photodynamic therapy photosensitizers in China. Photodiagnosis Photodyn Ther 2007;4:13—25. [16] Xu YZ, Chen SL, Chen YD, Yin XS, Zhang H. Study on clinical pharmacokinetics of the new photosensitizer photocarcinorin for tumor localization and photoradiation therapy. Acad J Sec Mil Med Univ 1986;7:82—6 [in Chinese]. [17] Xu DY. Principle, photosensitizers and introduction of clinical application. In: Tumor photodynamic therapy. Beijing: Chinese Medicinal Science and Technique Press; 1996. p. 149 [in Chinese]. [18] Li LB, Luo RC. Effect of drug—light interval on the mode of action of Photofrin photodynamic therapy in a mouse tumor model. Lasers Med Sci 2009;24:597—603. [19] Liu FG, Gu Y, Fu QT, Pan YM, Li JH. Absorptive characteristics of HMME and HpD in chicken comb skin and vascular endothelial cells. Chin J Laser Med Surg 2001;10:9—12 [in Chinese]. [20] Fingar VH. Vascular effects of photodynamic therapy. J Clin Laser Med Surg 1996;14:323—8. [21] Schmidt-Erfurth U, Hasan T, Schomacker K, Flotte T, Birngrube R. In vivo uptake of liposomal benzoporphyrin derivative and photothrombosis in experimental corneal neovascularization. Lasers Surg Med 1995;17:178—88. [22] Zhang L, Gu Y, Liu FG, Zhao JC, Jin XY. Study of mesenteric microvascular sensibility to PDT effects in rat. Chin J Laser Med Surg 2004;13:208—11 [in Chinese]. [23] Liu FG, Gu Y, Liu HL, et al. An experimental study on the comparison of photodynamic effects of hematoporphyrin monomethyl ether and hematoporphyrin derivative. Chin J Laser Med Surg 2001;10:69—73 [in Chinese]. [24] Gu Y, Huang NY, Liang J, Pan YM, Liu FG. Clinical study of 1949 cases of port wine stains treated with vascular photodynamic therapy (Gu’s PDT). Ann Dermatol Venereol 2007;134:241—4 [in French]. [25] Wang KK. Role of photodynamic therapy for the upper gut. Tech Gastrointest Endosc 2010;12:35—9. [26] Ido K, Yamamoto H, Kawamoto C, Koiwai H, Yoshida Y, Kimura K. Esophageal varices obliterated by photodynamic therapy for coexisting early esophageal carcinoma. Gastrointest Endosc 1997;45(5):420—3.

Vascular-targeted photodynamic therapy for bleeding gastrointestinal mucosal vascular lesions [27] Nagamine N, Ido K, Saihuku K, et al. Photodynamic effects on rabbit auricular veins after photosensitization with porfimer sodium: implications of the results with respect to the treatment of esophageal varices with photodynamic therapy. Gastrointest Endosc 2002;55:420—4. [28] Qiu HX, Gu Y, Wang Y, Huang NY. Twenty years of clinical experience with a new modality of vascular-targeted photodynamic

117

therapy for port wine stains at Chinese PLA General Hospital. Dermatol Surg 2011;37:1603—10. [29] Hamblin MR, Rajadhyaksha M, Momma T, Soukos NS, Hasan T. In vivo fluorescence imaging of the transport of charged chlorin e6 conjugates in a rat orthotopic prostate tumour. Br J Cancer 1999;81:261—8.