Photodynamic therapy of human choriocarcinoma transplanted to the hamster cheek pouch

GYNECOLOGIC

ONCOLOGY

34,

289-293 (1989)

Photodynamic Therapy of Human Choriocarcinoma Transplanted to the Hamster Cheek Pouch II. Intra-lesional Photosensitization’** ELY BRAND,

M.D., *,3 HO-SUN CHOI, M.D., PH.D., *-4 GLENN D. BRAUNSTEIN, M.D.,‘F WARREN S. GRUNDFJEST, M.D.,S AND LEO D. LAGASSE, M.D.*

Division of Gynecologic Oncology, Departments of *Obstetrics and Gynecology, tMedicine, and $Surgery, and The Laser Research Center, Cedars-Sinai Medical Center, Los Angeles, California 90048 Received November 23, 1988

Photodynamic therapy (PDT) uses light-activated compounds, such as hematoporphyrhrs, to produce cytotoxic effects after illumination. Human choriocarcinoma cells were transplanted into the hamster cheek pouch to study PDT. The transplanted choriocarcinoma secretes human chorionic gonadotropin (hCG) in proportion to tumor volume. Red light (630 nm) from an argonpumped dye laser (100-200 J/cm’) was used to illuminate tumors sensitized with dihematoporphyrin ether (DHE). Previous work has demonstrated complete regression (CR) of 90% of tumors (18/20) after one or two PDT sessions,while contralateral cheek pouch tumors continued to grow despite httraperltoneal DHE. Neither DHE nor laser light alone resulted in sign&ant CRs. In this study we evaluated intratumoral injection of DHE followed in 2 hr by laser treatment. In all tumors, localization of DHE was demonstrated by induced fluorescence with ultraviolet light or He:Cd laser. After a single treatment, 14 of 38 tumors (37%) completely regressed (hCG < mIU/ml); 4 tumors regressed grossly with low-level hCG [partial regression (PR)]. After repeat treatment there were 10 additional CRs in 19 rapidly enlarging tumors. After a third treatment 3 CRs and 3 PRs were achieved in 6 tumors. Becauseof large volumes, 2 of 3 progressing tumors failed to fluoresce uniformly after intratumoral DHE and were treated after intraperitoneal DHE injection; both completely responded. Overall, 29 of 38 tumors (76%) completely responded to PDT, and 7 partially responded (18%) with no gross tumor ’ Presented at the annual meeting of the Society of Gynecologic Oncologists, Maui, Hawaii, February 5-9, 1989. 2 This work was supported in part by a grant from the Linda Reed Gynecologic Oncology Laser Research Fund. ’ Present address: Division of Gynecologic Oncology, Box B198, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, CO 80262. To whom all correspondence should be addressed. 4 Present address: Department of Obstetrics and Gynecology, Chonnam University Medical School, Kwangtu, South Korea.

remaining in 5 of the 7. Only 5% of tumors (2/38) were nonresponders. Photodynamlc therapy results in gross elimination of 90% of tumors (52/58) in this model after intraperltoneal or intratumoral DHE sensitization (P < 0.0001). DHE in choriocarcinomas is easily detected and may enable detection of occult foci of malignancy. Chorlocarcinoma transplanted into the hamster cheek pouch is highly responsive to photodynamic therapy. Clinical trials of PDT in gynecologlc cancers are warranted to confhm the high response rates observed in refractory nongyAcademic Press, Inc. necologic cancers. 0 1989

The use of light as a cytotoxic agent was first reported by Raab in 1900, when he killed paramecia with sunlight after exposure to the vital dye acridine orange [l]. Since that time a variety of phototoxic compounds activated by light in the visible spectrum have been described. Among the most widely studied are the porphyrin derivatives , particularly hematoporphyrins . Policard demonstrated that these compounds are retained by malignant tumors using ultraviolet fluorescence [2]. Subsequently, Lipson and colleagues diagnosed cancer of the cervix, vagina, peritoneum, and other sites by fluorescence of tumors pretreated with hematoporphyrin derivative [3]. Currently, dihematoporphyrin ether (DHE), the active component of hematoporphrin derivative 141, is being extensively studied in phase III trials of photodynamic therapy in a variety of cancers [5]. Photodynamic therapy (PDT) uses DHE to sensitize tumors to light. On exposure to red light (625-635 nm) DHE undergoes excitation to a triplet electron spin state which, through intersystem crossing, produces singlet oxygen. The oxygen radicals produced in the DHE-contaming tumor tissue then oxidize cell and mitochondrial membranes, resulting in cell death [6]. Relatively little

289 OWO-8258/89$1.50

Copyright 0 1989by AcademicPress,Inc. All rights of reproductionin say form reserved.

BRAND ET AL.

290

attention has been directed to PDT treatment frequency and intralesional photosensitization. This study examines these parameters in hamster cheek pouch xenografts of a human choriocarcinoma cell line. MATERIALS AND METHODS Human choriocarcinoma cells were obtained from the JEG-3 cell line. The method of tumor innoculation has been previously described [7]. Two million cells were injected bilaterally into each of 40 cheek pouches in 20 female Syrian golden hamsters (Mesocricetus auralus, Simonsen Laboratories, Gilroy, CA). Macroscopic tumors were evident by 6-l 1 days. Tumor volume (cubic millimeters) was calculated from the largest perpendicular radii [4/37r(r, x r2 x r3)]. Animals were fed water and Purina rat chow ad libitum. The study was approved by the Cedars-Sinai animal research committee. PDT was performed 2 hr after direct intratumoral injection of 1.0 mg DHE (Photofrin II, Quadra. Logics Technologies, Vancouver, British Columbia). Hamsters were anesthetized with 0.5 mg/kg intraperitoneal pentobarbital. An argon-pumped tunable dye laser (Model MDS 90, Meditech GmbH, Hamburg, West Germany) was used as the source of red (630 nm) light. A total of 200 J/cm2 were delivered at a mean output of 200 mW using a spot size of 0.3-l .5 cm*, through a 600~pm quartz fiber. DHE uptake by the tumors was documented by red fluorescence elicited by excitation with blue light (432 nm) from a He:Cd laser (Omnichrome, Omega Engineering, Stamford, Conn.) as well as an ordinary Wood’s lamp (Fig. 1). In three large tumors uniform uptake was not observed and these animals received 25 mg/kg intraperitoneal DHE followed by PDT.

FIG. 1. This hamster demonstrates bilateral choriocarcinoma with hCG levels greater than 1200 mIU/ml prior to photodynamic therapy.

Tumors macroscopically evident at 1 week after PDT were treated a second time with red light. If fluorescence indicated the persistence of DHE at 7 days, the animals were not reinjected with DHE. Serum human chorionic gonadotropin (hCG) production was measured by double-antibody radioimmunoassay every 7 days [S]. Complete regression (CR) was defined as absence of all visible tumor with serum hCG less than 5 mIU/ml. Partial response (PR) was defined as greater than 50% volume reduction, or absence of gross tumor with persistent serum hCG. Statistical analysis was done using the x2 statistic with Yates’ continuity correction and Student’s t test for comparison of mean tumor volumes. Discordant responses in a given animal were evaluated retrospectively on the basis of hCG after further treatment. For example, if the right tumor regressed grossly and the left tumor continued to grow, the animal still produced hCG. After a second treatment, only to the left side, the tumor regressed and serum hCG then became undetectable. Therefore, the right tumor was considered a single-treatment CR, and the left a single-treatment NR and repeat-treatment CR. RESULTS Thirty-eight tumors developed in the 40 cheek pouches (Fig. 1). After the first PDT treatment 14 (37%) completely regressed with undetectable hCG levels in six animals (12 tumors) exhibiting bilateral CRs. Of the 19 tumors that were exposed to a second treatment, 10 (53%) completely regressed. The mean tumor volume was 420 mm3, similar to re-treatment after intraperitoneal DHE exposure (401 mm3) [7]. Serum hCG was undetectable in 8 tumors with bilateral regression. Of 6 tumors treated a third time, there were 3 CRs and 3 PRs from a mean treatment size of 246 mm3. Thus, 27 of 38 (71%) tumors were CRs after one to three PDT sessions. Repeat treatment eradicated tumors significantly larger (357 mm3 mean tumor volume) than CRs after a single treatment (60 mm3, P < 0.01). There were seven PRs (mean hCG, 222 mIU/ml) at 6 weeks (Fig. 2). In five of these, however, there was no gross tumor remaining, but merely low-level hCG production. When followed out to 12 weeks, four of the five animals had negative hCGs, with one animal demonstrating residual trophoblast and an hCG of 31 mIU/ml. As hCG correlates highly with pretreatment tumor volume in this model (r = 0.88) [7], these animals were considered partial responders, even if one cheek pouch had no visible tumor and the contralateral side had evident tumor, since serum hCG production could not be localized to either side. Four tumors continued progressive growth (11%) (hCG > 1200 mIU/ml). Of the 40 tumors, only one still produced hCG at 12 weeks. In this

PHOTODYNAMIC

291

THERAPY OF HUMAN CHORIOCARCINOMA TUMOR SIZE (mm31

-+-

HCG (mlUlml)

FIG. 2. Partial tumor necrosis after photodynamic therapy. After a second PDT treatment the tumor completely regressed.

case the left tumor was considered a CR on the basis of cheek pouch histology demonstrating no residual trophoblast; the right tumor was a partial responder because of the persistent hCG with residual trophoblast seen. Tumor response is summarized in Table 1. Light microscopy of tumors after PDT reveals extensive hemorrhagic necrosis, presumably as a result of the disruption of endothelial cells in tumor blood vessels [5]. Therefore, hCG measurements are a more sensitive indicator of viable tumor than gross size. This is demonstrated in Fig. 3, in which hCG levels decline despite transient tumor enlargement after treatment. Because of the large size of three tumors failing two and three treatments (mean 578 mm3), uniform intralesional DHE uptake was not possible as monitored by fluorescence. Therefore, they were treated a third and fourth time after intraperitoneal injection of DHE with 1 Responseto Photodynamic Therapy by Treatment Frequency and Tumor Size TABLE

Number of tumors”

Number of treatments

Complete regression

Partial regression

No regression

1 2 3

14 (60) 10 (420) 3 (149)

4 (255) 0 3 (176)

1 (161) 3 (337) 0

27

7b

46

Total

a Numbers in parentheses represent the mean tumor volume (mm? at the beginning of treatment in each group. b Two nonresponding tumors and one partially responding tumor were treated after intraperitoneal DHE sensitization (mean tumor volume 394 mm’) with two tumors completely responding.

0 + PDT

10

20

30

DAY

FIG. 3. Intratumor sensitization in choriocarcinoma. Correlation of hCG values with tumor volume after a single treatment in one hamster. Because of hemorrhage and edema the tumor enlarged briefly, producing low levels of hCG, prior to complete regression.

one PR (volume 165 mm3, hCG > 1200 mIU/ml) two CRs.

and

DISCUSSION The hamster cheek pouch is an excellent model for studying laser treatment of transplanted tumors. In this study of photodynamic therapy 34 of 38 tumors (90%) grossly regressed after one to four treatments. Seventyone percent of tumors completely responded using gross examination plus gonadotropin measurements. In an earlier study [7], contralateral cheek pouch tumors served as controls; therefore, hCG values were not used to confirm complete response, as in the present study in which both sides in a given animal were treated. Even if one takes into account the spontaneous regression rate of lo-20% [7] by x2 analysis comparing PDT and no treatment, the 90% regression rate remains highly significant, P < 0.0001. Regression after DHE alone without laser light occurs in 12% of tumors (not statistically different from the spontaneous regression rate), and regression after red light treatment without DHE in

292

BRAND

22%. The latter may result from hyperthermia, since laser treatment is associated with a 7.W mean temperature elevation [7]. However, laser light alone did not result in statistically significant tumor regression compared with DHE alone and untreated controls [7]. Although intraperitoneal DHE sensitization is as effective as intralesional sensitization in this model (34/38 tumors grossly eradicated compared with 18/20 in earlier work [7]), local injection of DHE allows the theoretical advantage of avoiding systemic toxicity. In addition, this study demonstrates that large tumors refractory to intralesional injection can be eliminated after intraperitoneal DHE followed by PDT. Multiple treatments enable tumor control in tumors failing single treatment without additional toxicity. The presence of DHE in the tumors can be confirmed by He:Cd laser-induced fluorescence (Fig. 4) to avoid repetitive DHE administration prior to laser. In clinical trials the major toxicity of PDT is skin photosensitivity, which can last between 4 and 8 weeks [5]. Further work needs to demonstrate that this does not occur after local DHE injection. Patients enrolled in a trial of PDT in malignant brain tumors, after intraarterial or intratumoral DHE, demonstrated only minor skin photosensitivity [9]. Similarly, there was no skin phototox-

ET AL.

icity in 10 patients with colorectal and other carcinomas treated by CT-guided intraluminal sensitization at Fox Chase Cancer Center [lo]. Clinical experience using PDT in gynecology remains limited, with only 36 patients reported in the literature. Despite this small number, complete responses have been reported in 20 patients with a variety of primary and metastatic tumors to the vulva, vagina, cervix, and endometrium [ Ill. This study demonstrates that human choriocarcinoma takes up locally injected DHE and responds to PDT. The use of the He:Cd laser to detect tumor fluorescence after DHE administration may allow for improved diagnosis and possible treatment if 630-nm light can be delivered to the site of refractory disease, for instance, choriocarcinoma in the brain or liver. The treatment of highrisk choriocarcinoma is associated with significant morbidity and mortality. Approximately one-third of patients are refractory to conventional chemotherapy and radiotherapy [12]. Photodynamic therapy with little toxicity should be technically feasible in gestational trophoblastic disease. For instance, laser light could be delivered using quartz fibers passed through a hysteroscope after molar evacuation, in cases of invasive moles and in uterine choriocarcinoma resistant to chemotherapy. Clinical

FIG. 4. The presence of DHE in the tumor is indicated by the dark areas in this photograph (produced by red fluorescence on excitation with blue light from a He:Cd laser). Normal tissue and tumor not injected with DHE appear lighter because the photograph was taken through an orange filter.

PHOTODYNAMIC

THERAPY OF HUMAN CHORIOCARCINOMA

trials of PDT are justified by the high success rate noted in this animal model. REFERENCES 1. Raab, 0. Uber die Wirkung lluorescirender Stoffe auf Infusorien, Z. Bid. 39, 524-546 (1900). 2. Policard, A. Etudes sur les aspects offer& par des tumeur experimentales examinee a la lumiere de Woods, C.R. Sot. Eiol. 91, 1423-1427 (1924). 3. Lipson, R. L., Baldes, E. J., and Gray, M. J. Hematoporphyrin derivative for the detection and management of cancer, Cancer 20, 2255-2257 (1967). 4. Dougherty, T. J., Boyle, D. G., Weishaupt, K. R., Henderson, B. A., Potter, W. R., Bellnier, D. A., and Wityk, K. E. Photoradiation therapy-Clinical and drug advances, in Porphyrin photosensitization (D. Kessel and T. J. Dougherty, Eds.), Plenum, New York (1983). 5. Manyak, M. J., Russo, A. R., Smith, P., and Glatstein, E. Photodynamic therapy, J. Clin. Oncol. 6, 380-391 (1988). 6. Moan, J. Porphyrin-sensitized photodynamic inactivation of cells: A review, Lasers Med. Sci. 1, 5-12 (1986).

293

7. Brand, E., Choi, H. S., Karalis, K., Papaioannou, T., Fishbein, M. C., Braunstein, G. D., Wade, M. E., Lagasse, L. D., and Grundfest, W. S. Photodynamic therapy of choriocarcinoma transplanted into the hamster cheek pouch. Gynecol. Oncol. (in press). 8. Vaitukaitis, J. L., Braunstein, G. D., and Ross, G. T. A radioimmunoassay which specifically measures human chorionic gonadotropin in the presence of human luteinizing hormone, Amer. J. Obstet. Gynecol. 113, 751-758 (1972). 9. Kostron, H., Fritsch, E., and Plangger, C. Photodynamic treatment of malignant brain tumors: A phase I/II trial, in Proceedings, 2nd Biennial Meeting, International Photodynamic Association, London, England, July 1988, Lasers Med. Sci., Abstract 41. 10. Gatenby, R. A., Hartz, W. H., Engstrom, P. F., Rosenblum, J. S., Hammond, N. D., Kessler, H. B., Moldofsky, P. J., Clair, M. R., Unger, E., and Broder, G. J. C-T guided laser therapy in resistant human tumors: Phase I clinical trials, Radiology 163, 172-175 (1987). 11. Brand, E. Photodynamic therapy of cancer, Contemp. ObjGyn 33, 30-50, (1989). 12. DuBeshter, B., Berkowitz, R. S., Goldstein, D. P., Cramer, D. W., and Bernstein, M. R. Metastatic gestational trophoblastic disease: Experience at the New England Trophoblastic Disease Center, 1965 to 1985, Obstet. Gynecol. 69, 390-395 (1987).