GYNECOLOGICONCOLOGY x,200-206(1990)

Photodynamic Therapy of Choriocarcinoma Transplanted to the Hamster Cheek Pouch’

I. lntraperitoneal Photosensitization

ELY BRAND, M.D. ,2 HO-SUN CHOI, M.D., PH.D., KATIA KARALIS, M.D . ,* THANASSIS PAPAIOANNOU, M.Sc. ,* MICHAEL C. FISHBEIN, M.D. ,t GLENN BRAUNSTEIN, M.D.,* MACLYN E. WADE, M.D.,

LEO D. LAGASSE, M.D., AND WARREN S. GRUNDFEST, M.D.*

Division of Gynecologic Oncology, Departments of Obstetrics and Gynecology, *Medicine, $Surgery, and tPathology and The Laser Research Center, Cedars-Sinai Medical Center, Los Angeles, California 90048

Received January 12, 1989

Human choriocarcinoma (JEG3) cells were transplanted into the cheek pouch of hamsters and treated with photodynamic therapy. Twenty-four hours after intraperitoneal injection of the photosensitizer dihematoporphyrin ether (DHE), 20 tumors were illuminated with 100 J/cm* of 630-nm light from an argon pumped dye laser. Contralateral tumors served as controls. Di- hematoporphyrin ether alone had no effect on tumor growth, while laser light in the absence of DHE resulted in complete regression in 3 tumors (17%), and partial regression in 4 of 18 tumors (22%), possibly due to hyperthermia, P > 0.10. Using the combination of DHE plus light (photodynamic therapy) com- plete tumor regression was noted after a single treatment in 11 of 20 tumors (55%, mean tumor volume 279 mm3) and in 7 of 7 tumors (100%) after a second treatment. Two of 20 tumors were not retreated. Therefore, 18 of 20 tumors (90%) were grossly destroyed by one or two photodynamic treatments. Contralateral control tumors continued to grow to a median volume of 990 llUU3(/$= 26.30, P < 0.0001). Choriocarcinoma transplanted into the hamster cheek pouch is highly responsive to photody- namic therapy. 6 1990 Academic FW.ss, Inc.

Photodynamic therapy (PDT) produces tumor necrosis after red light activation of hematoporphyrins. Since he- matoporphyrins are retained in malignant cells longer than in normal cells, this enables selective cancer cell kill using very low energy red light with little toxicity. Although PDT of malignant tumors is gaining increasing acceptance [l-3], experience in treating gynecologic ma- lignancies is limited. Fewer than two dozen patients with

’ This work was supported in part by the Linda Reed Gynecologic Oncology Laser Research Fund.

2 To whom reprint requests should be addressed at Division of Gy- necologic Oncology, Department of Obstetrics and Gynecology, Uni- versity of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, CO 80262.

gynecologic cancers treated with PDT have been re- ported in the English literature. The response of gyne- cologic cancers to this modality has been encouraging. Complete responses have been seen in vaginal, cervical, metastatic endometrial cancer, and ovarian carcinoma metastatic to the vagina. Overall, complete responses occur in over 45% of tumors, while no response occurs in only 25% of patients treated [4-71.

Choricarcinoma is a tumor with established cell lines that secrete human chorionic gonadotropin (hCG) as an easily measured tumor marker. Although the effects of PDT on choriocarcinoma have been examined in vitro [8,9], this is the first study of PDT on a tumor of ex- traembryonic origin in vivo. Using a model of chorio- carcinoma developed by Hertz [lo] to study the effects of chemotherapy, we evaluated the effectiveness of PDT in tumors transplanted to the hamster cheek pouch. We report the gross and histologic parameters of tumor response.

MATERIAL AND METHODS

The JEG3 human choriocarcinoma cell line was ob- tained from tissue culture. Cells in confluent growth were trypsinized (0.02% trypsin) and suspended in tissue cul- ture medium. The cells were grown in Dulbecco’s mod- ified Eagle’s medium (Gibco Laboratories, Inc., Grand Island, NY) supplemented with 10% fetal calf serum, penicillin (500 units/ml), streptomycin (500 pg/ml), and 1% glutamine. Cells were diluted to 3-5 x lO’/ml and sterilely injected bilaterally into cheek pouches of 4- to 6-week-old female Syrian golden hamsters (Mesocricetus aurutus, Simonsen Laboratories, Gilroy, CA). The ani- mals were treated with dexamethasone (0.2 mg/kg) sub- cutaneously every other day to enhance tumor devel-

200 OO!W-8258190 $1.50 Copyright 6 1990 by Academic Press, Inc. AU rights of reproduction in any form reserved.

PHOTODYNAMIC THERAPY OF CHORIOCARCINOMA 201

I. GROWTH CONTROLS # TUMORS JEG-3 CELLS 10 HAMSTERS

0000 o”ooo -0 ’

II. LASER CONTROLS I I

10 HAMSTERS

0000

00 + 000

I

0000 00

000

17DHE CONTROLS

19

18

37

20 HAMSTERS

FIG. 1. Study design. Fifty-four tumors served as controls for growth, DHE and laser light without DHE. Twenty tumors were treated with 100 J/cm2 red (630 nm) light after sensitization with DHE.

opment. The hamsters were fed Purina Lab Chow and water ad libitum. Animal welfare was ensured in accor- dance with the guidelines of the Cedars-Sinai Animal Research Committee. The pouches were everted every l-3 days and tumor dimensions recorded in three per- pendicular axes. Tumor volume was calculated using an ellipsoid model (volume = 4/3 T x {a/2 x b/2 x c/2}, where a, b, and c are the largest perpendicular diameters).

The study comprised 40 animals equally divided be- tween treatment and controls as described in Fig. 1. A control series of 10 animals received no treatment in order to establish the growth pattern of the transplanted tumors. A second series of 10 animals (light controls) were treated with red laser light (630 nm) without ad- ministration of dihematoporphyrin ether (DHE, Photof- rin IIR, Photomedica Inc., Raritan, NJ). The temperature change during laser light delivery was monitored using a 0.008-in. hypodermic thermocouple probe (Model Hyp- 0, Omega Engineering, Stamford, CT). Twenty addi- tional hamsters were treated with photodynamic therapy to one cheek pouch tumor with the contralateral side serving as an internal procedural and DHE control. These animals received DHE injected intraperitoneally at 2 mg/kg. Prior to laser treatment animals were an- esthetized using intraperitoneal pentobarbital at 0.065 mdgm.

Twenty-four hours after DHE injection the larger of the two cheek pouch tumors was treated with 630-nm light emitted from an argon pumped dye laser (Coherent Lasers, Model CR599, Palo Alto, CA). Kiton Red 620 (Exciton Chemical Co., Dayton, OH) was used as the

fluorochrome in the dye laser. The laser and fiber outputs were measured by means of a power meter (Scientec Model 365, Boulder, CO). Each tumor was treated once or twice, depending on response, with 100 J/cm*.

Tumors larger than 5 mm were treated by parallel opposed antero-posterior and postero-anterior fields. Laser fiber outputs of 100-300 mW with a spot size of l-2 cm* for 13-17.5 min were used to illuminate the entire tumor uniformly within the laser beam. The spot size and therefore power density varied among animals to accommodate tumors of different size, but during each treatment these were constant. Therefore, the energy delivered during all treatments could be kept constant at 100 J/cm*; i.e., the product (W/cm*) x (set) was always equal to 100 J/cm*. Tumors not responsive to a single photodynamic treatment were retreated at 7 days in a fashion identical to the initial treatment.

Tumor volume was measured every 24-72 hr. Serum hCG production was measured in some animals by dou- ble antibody radioimmunoassay [ 111. In the absence of complete regression of gross tumor, the size of viable tumor was estimated on the basis of gross characteristics. After maximal gross necrosis the treated and control cheek pouches were excised, fixed in formalin, and ex- amined histologically. Animals were sacrificed using standard pentobarbital euthanasia solution 30-60 days postinjection. Statistical analysis was conducted using the chi-square statistic with Yates continuity correction, Student’s t test, and linear regression.

RESULTS

I. Choriocarcinoma Controls

Ten animals served as controls for tumor growth. Ten cheek pouches in five hamsters were inoculated with 4.6 x 10’ JEG-3 cells. One animal expired after adminis- tration of pentobarbital anesthesia on Day 9 with two 5- mm tumors. Six tumors exhibiting exponential growth were established in three hamsters. The fourth hamster developed two small tumors, one regressed to 1 mm3 by 32 days postinoculation and the other did not grow be- yond 1 mm3. The maximum tumor volume of the six enlarging tumors averaged 1945 mm3 at 30-32 days after inoculation.

Ten pouches in five additional animals were implanted with 0.5-mm3 tumors from a single donor choriocarci- noma growing in a hamster cheek pouch. By 7 days all 10 tumors were growing exponentially with a mean vol- ume of 2392 mm3 at 18 days post-transplantation. These were rapidly growing tumors that led to the death of four of five animals due to apparent sepsis or from tumor obstruction of the oropharynx. One animal demonstrated extensive necrosis of bilateral tumors after reaching a

202 BRAND ET AL.

TUMOR VOLUME O(Thou8ar& mm31

I

R 1L 2L 9R 3L 4R 4L 6R 6L 6R BL 7R IL 8R 8L 9R #

DAY 21 m MAXIMUM

FIG. 2. JEG3 controls. Tumor volume of the 17 untreated tumors at 21 days and at maximum size 21-86 days postinoculation with JEG- 3 cells. R, right; L, left pouch.

maximal volume of 5999 mm3 followed by partial regres- sion and autoamputation from the cheek pouch.

Thus, of the 20 control pouches in 10 animals, 19 developed tumors. Of the 17 evaluable tumors (omitting the anesthesia death and the l-mm tumor) 3 exhibited significant regression to less than 50 mm3 at the time of death or sacrifice (20-60 days). Thus the regression rate of the control tumors was 17.6% (3/17, Fig. 2). Because of this spontaneous reduction in volume, photodynamic therapy was performed unilaterally in a given hamster with the contralateral cheek pouch of each treated animal serving as a growth and DHE control.

II. Hematoporphyrin (DHE) Controls

In 20 hamsters inoculated with 3-5 x 10’ JEG-3 cells on Day 1, 37 of 40 implantations successfully developed viable, growing tumors (93% transplantation efficiency). Seventeen animals developed bilateral tumors with par- allel growth curves and 3 animals developed only uni- lateral tumors. No animal failed to develop at least a unilateral tumor. The pre- and postexperiment median tumor volume of the controls increased exponentially from 16 to 990 mm3 (Table 1). These tumors grew in the presence of intraperitoneal DHE which was administered for treatment of the contralateral tumor (Fig. 3). When compared with control animals not receiving DHE, there was no effect of DHE on tumor growth. Generally, mac- roscopic tumors were evident by 7-10 days postinocu- lation, although 4 of 37 tumors appeared 15-17 days postinjection. Partial tumor regression (>50% reduction in volume) spontaneously occurred after 25-35 days in 2 of the 17 control tumors (11.7%), which was not sig- nificantly different from controls not exposed to DHE. No tumor regressed completely in the DHE controls.

TABLE 1 Choriocarcinoma Tumor Volume before and after

Photodynamic Therapy

Volume (mm’)

Control Treated

1” 2b 1” 2h 3'

Hamster No. El E2 E3“ E4 E5 E6 E7 E8 E9

37 2096 44 0 6 990 99 440

16 5 24 0 44 1336 323 2078 21 1001 314 616

707 623 534 628 3 587 103 236 3 1698 110 402 2 314 85 0

Fl 14 F2 4 F3 0 F4 8 F.5 0 F6 27 F7 0 F8 13 Fsd 21 FlO 21 Fll 110

Mean volume (mm’) 62 Median volume (mm’) 16

1060 1886

99

1440

1319 15

245 442 892 990

37 0 8 0

37 0 29 0 57 0 37 0 16 660 42 275 44 0 26 251

103 0 104 279 43 0

n Tumor volume prior to initial therapy. b Tumor volume 7-10 days post-PDT. ’ Tumor volume 8 days after the second treatment. d DHE-exposed control tumors which regressed spontaneously.

2000

1600

1000

600

0

TUMOR VOLUME ha) INITIAL m FINAL

2boo-

1., 1 2 3 4 6 6 7 8 9 10 11 12 13 14 16 16 17

HAMSTER

FIG. 3. DHE controls. Tumor volume of 17 choriocarcinomas after intraperitoneal injection of DHE at the time of maximal growth (Days 6-17).

PHOTODYNAMIC THERAPY OF CHORIOCARCINOMA 203

The tumor doubling time was approximately 2.5 days in the exponential phase of growth.

III. Laser Controls

A series of 18 tumors in 10 animals were treated with 630-nm laser light at 100 J/cm* without DHE. Eleven tumors (61%) continued exponential growth from an in- itial volume of 28 mm3 at the time of laser treatment to 666 mm3 at 1 month. Three tumors in 2 animals com- pletely regressed (16.6%). Four tumors partially re- gressed (22.2%) from an initial volume of 51 mm3 to a stable volume of 17 mm3. Considerng the 15% regression rate (5/34) in the choriocarcinoma and hematoporphyrin (DHE) controls, regression after laser treatment alone (7/18) was not statistically significant (x2 = 2.59, P > 0.10). The mean temperature rise in the tumors treated by laser was 7.8”C and may be related to the regression trend observed.

IV. Photodynamic Therapy

Twenty hamsters underwent photodynamic therapy after establishing exponential tumor growth. Of the 20 treated tumors, 18 exhibited complete macroscopic regression after one or two laser treatments, as the 17 contralateral tumors persisted or enlarged (x2 = 26.30, P < 0.0001). Comparing the 18 tumors that completely regressed to the 7 tumors in the laser controls which regressed, photodynamic therapy was significantly more effective than laser light in the absence of DHE (x2 = 8.82, P < 0.005). Complete response was defined as disappearance of all macroscopic viable tumor. Histo- logic sections from treated tumors revealed extensive necrosis and intense inflammatory cell infiltrates. Oc- casional microscopic nests of intact cells of uncertain viability were observed, even in tumors that sponta- neously autoamputated from the cheek pouch. Tumor volume was plotted based on grossly viable tumor. For instance, the tumor shown in Fig. 4 exhibits complete necrosis and subsequently regressed. The choriocarci- noma growth curves in this animal are shown in Fig. 5.

To exclude the possibility that smaller tumors were selected for treatment we deliberately treated the larger tumor side. Mean tumor volume of the control side at the start of PDT was 62 mm3, mean tumor volume of the treated side was 104 mm3. After the first PDT treat- ment the mean tumor volume of the contralateral con- trols was 892 mm3 (median 990 mm3) versus 279 mm3 (median 0 mm3) for the treated side.

The 18 tumors that responded to PDT had a mean starting volume of 80 mm3 (46 mm3 for those responding to one treatment (n = 11) and 401 mm3 starting volume before the second treatment (n = 7)). There were 3 tumors larger than 300 mm3 at the start of the first PDT,

FIG. 4. Syrian golden hamster with DHE-exposed control tumor (right) and contralateral tumor demonstrating extensive necrosis (ar- row) 3 days after photodynamic therapy.

of which 1 responded completely (534 mm3) and 2 were nonresponders (314, 323 mm3). The tumor regression data were plotted as a function of time postlaser treat- ment (Fig. 6). Nonresponding tumors continued normal growth. Responding tumors began to show early necrosis by 36-72 hr after PDT, with complete regression by 8 days.

V. Repeat Photodynamic Treatment

After a single PDT treatment 9 tumors continued to grow. Thus, the response rate to a single treatment was 11/20 (55%). Two persistent tumors (nonresponders) were not retreated to look for any delayed regression beyond 2 weeks postlaser. These tumors continued to grow. One animal was sacrificed on Day 86 with hCG over 60,000 mIU/ml. At 1 week after the first treatment

-0 2 4 6 8 10 12 14 16 18 20 22 24

DAY

FIG. 5. Maximum tumor diameter of the treated and untreated sides of the animal in Fig. 4 as a function of postinoculation day. Hamster El: treated, right (0); control, left (+); laser (*).

204 BRAND ET AL.

TUMOR VOLUME (mm3) 500 I I 1

400 I+ n2

300 I n.7

-0 5 10 15 20

Days (Post Treatment) + SINGLE RX + RETRE_TED - NON-RESPONDERS

FIG. 6. Choriocarcinoma volume after photodynamic therapy. Eleven tumors regressed after a single treatment. Seven tumors treated a second time demonstrated prompt regression despite a mean volume 8 times that of the initially responding tumors.

7 animals were reinjected with DHE and exposed to laser light at 100 J/cm’ as in the first PDT treatment. All 7 of these retreated tumors regressed despite their large size. Therefore, 18 of 20 tumors (90%) completely regressed after one or two photodynamic therapy sessions.

The mean tumor volume at the initiation of the second laser treatment was 401 mm3 (median 402 mm’), signif- icantly larger than the initial treatment volumes, as ex- pected in growing tumors. By 48 hr post-treatment, the volume was smaller in 6 of 7 tumors. At 4 days post- treatment the mean volume was 70 mm3, and by 8 days post-treatment all 7 tumors were completely necrotic. Due to the small number of retreated tumors the differ- ence in response rate between one and two treatments was of borderline statistical significance (0.10 > P > 0.05), although in other models [12], and clinically [ll, multiple treatment with PDT improves tumor control.

VI. Pathology

The control choriocarcinoma exhibits large trophoblast cells which stain for hCG by immunoperoxidase tech- nique (Fig. 7). This is a highly vascular tumor. Many of the control tumors exhibit a focal inflammatory response consisting of polymorphonuclear and lymphocytic infil- trates. After administration of DHE no change is noted on light microscopy in the control tumors. Inflammatory infiltrates are more extensive after photodynamic ther- apy. Extensive hemorrhagic necrosis is seen following PDT with widespread degeneration of the choriocarci- noma cells. Only occasional nests of recognizable neo- plastic cells are apparent microscopically. hCG values correlate highly with tumor volume prior to PDT (r = 0.88, P < 0.01). Because each treated animal had an

internal control in the contralateral cheek pouch, post- treatment hCG could not be used to indicate response on the treated side.

DISCUSSION

A wide diversity of tumors derived from endoderm, ectoderm, and mesoderm are highly responsive to pho- todynamic therapy. Significant response rates have been observed in skin cancers (squamous cell carcinoma, mel- anoma, Kaposi’s sarcoma), intestinal tumors (esopha- geal, gastric, colon, and rectal carcinoma), breast cancer, brain tumors, and genito-urinary tract tumors (cervical, endometrial, vaginal, and bladder carcinoma) [l-7]. This study of human choriocarcinoma cells transplanted into the hamster cheek pouch adds a malignancy derived from extraembryonic tissues to the list of highly responsive tumors. Ninety percent of transplanted choriocarcinoma tumors were grossly eradicated after one or two pho- todynamic treatments.

The mechanism of selective retention of hematopor- phyrins by tumors remains unclear. Cell culture data indicate that the kinetics of hematoporphyrin uptake and release are identical in some normal and malignant cell lines [ 131, although others have found preferential uptake by malignant cells [14,15]. In vivo, the basis of selective toxicity appears to be the longer retention of hemato- porphyrins in malignant tumors compared to normal tis- sues [15,16]. Bugelski et al. [16] attribute this to the increased vascular permeability of tumors with dimin- ished lymphatic drainage as well as nonspecific binding to stromal elements. Macrophages in tumor stroma con- tain significant amounts of hematoporphyrin at 7 days compared to these cells in normal tissue which release porphyrins more rapidly. The hamster cheek pouch model is an excellent model for studying the mechanism of photodynamic therapy. Other workers have performed photodynamic therapy of hamster squamous cell carci- noma induced in the cheek pouch [17,18]. Our work demonstrates that heterotransplantation of human tu- mors transplanted to the buccal pouch is also a feasible model for PDT.

Some authors believe that PDT exerts its effects through the tumor vasculature. The rapid regression and hemorrhagic necrosis of choriocarcinoma in this study are consistent with an effect on the tumor vasculature by DHE when excited by 630-nm light. Tumors larger than 2 cm’ may be less responsive due to poor vascularity in necrotic areas, as well as the limited penetration of red light beyond 1.5 cm depth (-4% incident energy) [l]. Since cell death appears to be mediated by singlet oxygen production, hypoxic areas of tumors may be more responsive to treatment under conditions of higher tissue p0, [ 191. However, it is clear that choriocarcinoma

PHOTODYNAMIC THERAPY OF CHORIOCARCINOMA 205

FIG. 7. (a) DHE control tumor demonstrating broad sheets of viable trophoblast cells. (b) Choriocarcinoma after PDT with extensive inflammation and hemorrhagic necrosis. Asterisk reveals a thin band of recognizable neoplastic cells (hematoxylin-eosin stain {H&E} x 40). (c) Details of neoplastic cells which are predominantly mononuclear, with large pleomorphic nuclei and enlarged nucleoli (H&E x 160). (d) Immunohistochemical stain for hCG with moderate staining of most cells (curved arrow) and intense staining of occasional cells (straight arrow) (peroxidase-antiperoxidase x 160).

responds to PDT even in virro. Girotti and Hussa [8] studied a choriocarcinoma cell line, demonstrating large decreases in cell count and hCG production after low energy illumination (0.3 J/cm2) following incubation with hematoporphyrin derivative. Other workers have noted a shift of ploidy in choriocarcinoma cell cultures to a diploid pattern with reduction of aneuploid cell popu- lations after PDT [9].

The mechanism by which repeat treatment of initially nonresponsive, enlarging tumors results in tumor regres- sion is unclear. It may represent a dose-intensity phe- nomenon, since the second dose of DHE may increase the amount of singlet oxygen generated in the presence of DHE retained after the first treatment. Alternatively, repair of sublethal cell injury may be inhibited by the second treatment. This seems less likely since retreat- ment was delayed for 7 days, by which time sublethal injury should have been repaired. This model indicates

that tumors larger than 400 mm3 (approximately 1 cm diameter) are susceptible to PDT after multiple treat- ments. Further studies will examine the response of larger tumors to interstitial light and hematoporphyrin delivery.

Note added in proof. Intratumor injection of DHE results in complete response and hCG remission in 76% of tumors. Gynecol. Oncol. 34, 289 (1989).

REFERENCES

1. Dougherty, T. J. Photosensitization of malignant tumors, Semin. Surg. Oncol. 2, 24-37 (1986).

2. Prout, G. R., Lin, C.-W., Benson, R., Jr., Nseyo, U. O., Daly, J. J., Griffin, P. P., Tian, M., Lao, Y.-H., Mian, Y.-Z., Chen, X., Ren, F.-M., and Qiao, S.-J. Photodynamic therapy with hemato- porphyrin derivative in the treatment of superlicial transitional-cell carcinoma of the bladder, N. Engl. J. Med. 317, 1251-1255 (1987).

3. &huh, M., Nseyo, U. O., Potter, W. R., Dao, T. L., and Dough-

206 BRAND ET AL.

erty, T. J. Photodynamic therapy for palliation of locally recurrent breast carcinoma, J. Clin. Oncol. 5, 17661770 (1987).

4. Ward, B. G., Forbes, I. J., Cowled, P. A., McEvoy, M. M., and Cox, L. W. The treatment of vaginal recurrences of gynecologic malignancy with phototherapy following hematoporphyrin deriv- ative pretreatment, Amer. J. Obstet. Gynecol. 142,356-357 (1982).

5. Rettemnaier, M. A., Berman, M. L., DiSaia, P. J., Bums, R. G., and Bems, M. W. Photoradiation therapy of gynecologic malig- nancies, Gynecol. Oncol. 17, 200-206 (1984).

6. Soma, H., and Nutahara, S. Cancer of the female genitalia, in Lasers and hematoporphyrin derivative in cancer (Y. Hayata and T. J. Dougherty, Eds.), Igaku-Shoin, New York, p. 97 (1983).

7. McCaughan, J. S., Jr., Schellhas, H. F., Lomano, J., and Bethel, B. H. Photodynamic therapy of gynecologic neoplasms after pre- sensitization with hematoporphyrin derivative, Lasers Surg. Med. 5, 491-498 (1985).

8. Girotti, A. W., and Hussa, R. 0. Phototoxic effects of hemato- porphyrin derivative and its chromatographic fractions on hor- mone-producing human malignant trophoblast cells in vitro, Adv. Exp. Biol. Med. 191, 129-145 (1985).

9. Kubota, K., Kasai, T., and Iwasaki, H. Experimental studies on photoradiation therapy (PRT)-Effects of PRT on cultured human cancer cell lines, Nippon Sanka Fujinka Gakkai Zasshi 38, 863- 870 (1986).

10. Hertz, R. Choriocarcinoma of women maintained in serial passage in hamster and rat, Proc. Sot. Exp. Biol. Med. 102, 77-81 (1959).

Il. Vaitukaitis, J. L., Braunstein, G. D., and Ross, G. T. A radioim- munoassay which specifically measures human chorionic gonad-

otropin in the presence of human luteinizing hormone, Amer. J. Obstet. Gynecol. 113, 751-758 (1972).

12. Tochner, Z., Mitchell, J. B., Smith, P., Harrington, F., Glatstein, E., Russo, D., and Russo, A. Photodynamic therapy of ascites tumours within the peritoneal cavity, Brit. J. Cancer 53, 733-736 (1986).

13. Chang, C., and Dougherty, T. J. Photoradiation therapy: kinetics and thermodynamics of porphyrin uptake and loss in normal and malignant cells in culture, Abstr. Rad. Res. Sot. 74, 498 (1978).

14. Carrano, C. J., Tsutsui, M., and McConell, S. Tumor localizing agents: The transport of meso-tetra (p-sulfophenyl) porphine by Vero and HEp-2 cells in vitro, Chem. Biol. Interact. 21, 233-248 (1978).

15. Mossman, B. T., Gray, M. J., Silberman, L., and Lipson, R. L. Identification of neoplastic versus normal cells in human cervical cell culture, Obstet. Gynecol. 43, 635-639 (1974).

16. Bugelski, P. J., Porter, C. W., and Dougherty, T. J. Autoradi- ographic distribution of hematoporphyrin derivative in normal and tumor tissue of the mouse, Cancer Res. 41, 4606-4610 (1981).

17. Bums, R. A., Klaunig, J. E., Shulok, J. R., Davis, W. J., and Goldblatt, P. J. Tumor-localizing and photosensitizing properties of hematoporphyrin derivative in the hamster buccal pouch car- cinoma, Oral Surg. Oral Med. Pathol. 61, 368-372 (1986).

18. Berg, L. F., and Harris, D. M. Microscopic fluorescence in pho- todynamic therapy, Laryngoscope 96, 986-989 (1986).

19. Moan, J., and Sommer, S. Oxygen dependence of the photosen- sitizing effect of hematoporphyrin derivative in NHIK 3025 cells, Cancer Res. 45, 1608-1610 (1985).