- - - - - - - - - - - - - - SYMPOSIUM ARTICLE - -

Role of chemotherapy in the future treatment of ovarian cancer ROBERT

F.

OZOLS

From the Department of Medical Oncology, Fox Chase Cancer Center, Philadelphia, PA, USA

Acta Obstet Gynecol Scand 1992;71 Suppl 155: 5~

Platinum-based chemotherapy has led to an improvement in complete response rates and duration of median remission, but has only given a modest improvement in overall survival in patients with advanced ovarian cancer. Chemotherapy will in the future focus upon: (I) improving the complete remission rate with new induction regimens; (2) identifying strategies capable of converting partial remission into complete remission; (3) preventing or delaying recurrences in patients who do achieve a complete remission; (4) identifying mechanisms of antineoplastic drug resistance and pharmacologic techniques capable of reversing drug resistance. Among the treatment approaches being utilized are high-dose chemotherapy with autologous bone marrow transplantation, development of new chemotherapeutic regimens which include Taxol and hexamethylmeiamine, and intraperitoneal chemotherapy. In addition, our understanding of the mechanisms of antineoplastic drug resistance has led to the development of novel therapeutic approaches. It has been demonstrated that resistance to platinum and alkylating agents is associatedwith both increased concentrations of cellular glutathione (GSH) as well as an increased capacity of tumor cells to repair damage to DNA. Inhibition of GSH biosynthesiswitli buthionine sulfoximine (BSO), a synthetic inhibitor of the enzyme gamma glutamyl cysteine synthetase. has led to the potentiation of alkylating agent activity in vitro and in vivo. A phase I trial of BSO plus melphalan is currently in progress and a trial of BSO plus carboplatin is planned. Inhibition of the DNA repair process with aphidicolin potentiates the cytotoxicity of cisplatin in drug-resistant tumor cells. Clinical trials of aphidicolinplus cisplatin await the completion of ongoing phase I trials of aphidicolin.

Chemotherapy has been the mainstay of treatment for most patients with advanced ovarian cancer (OC). Cytoreductive surgery is rarely, if ever, curative in patients with advanced disease and consequently postoperative therapy is indicated in virtually all patients. Epithelial OC is a drug-sensitive tumor with numerous agents having clinical activity. However, platinum complexes appear to be the most active group of compounds available for the treatment of this disease. Until recently the standard chemotherapeutic regimen in North America had been the two-drug combination of cisplatin plus cyclophosphamide (1). However, on the basis of a more favorable toxicity profile, carboplatin has essentially replaced cisplatin in the treatment of patients with advanced OC (2). Carboplatin is a cispla-

tin analog which is less nephrotoxic and neurotoxic, causes less nausea and vomiting, and consequently can be administered in an outpatient setting without the need for intensive hydration. The dose-limiting toxicity of carboplatin is myelosuppression, particularly thrombocytopenia (3). While carboplatin is less toxic than cisplatin, standard use of this drug has not led to an improved survival, compared with treatment with cisplatin. Carboplatin and cisplatin both share the same active intermediate; consequently there is marked crossresistance between these two drugs (3). Carboplatin plus cyclophosphamide will produce an objective response in about 80% of patients with advanced OC with 40-50% of patients achieving a complete clinical remission. However, median survival for paActa Obstet Gynecol Scand Suppl/55

56

R. F. Ozols

tients with platinum-treated chemotherapy has been in the range 18-30 months and, in most studies, 5-year survival rates have ranged between 20% and 25% (1). While platinum-based chemotherapy has led to higher response rates, and longer durations of remission, there has been only a modest improvement in survival compared with what could be expected with single-agent alkylating drugs. The vast majority of patients with advanced OC will die of their disease. In the next decade, clinical trials will be examining those factors which limit the current effectiveness of chemotherapy. It is obvious that the complete remission rate must be improved and new induction regimens will be evaluated. Furthermore, the importance of dose intensity will be prospectively evaluated, since the dose-limiting toxicity of carboplatin can be reduced with biologic and pharmacologic techniques. A substantial number of patients achieve a partial remission and strategies evaluating the ability to convert patients into a complete remission will also be evaluated. It is also apparent that the achievement of a complete remission does not ensure cure in patients with advanced Oc. Approximately 30-50% of patients will ultimately relapse after a complete remission (1). Consequently techniques which may be capable of preventing or delaying recurrences have the potential to markedly improve survival rates. Once patients do relapse, however, they are extremely unlikely to be cured. Salvage therapy regimens are not curative, due primarily to the development of broad cross-resistance which limits the effectiveness of chemotherapeutic agents from different pharmacologic classes.

Dose intensity of platinum compounds Retrospective studies have demonstrated that the dose intensity of cisplatin is an important factor in achieving optimum results. Levin & Hryniuk analysed the relationship between dose intensity of cisplatin and the complete remission rate in 33 published trials dealing with the chemotherapy of OC (4). They found a statistically significant correlation between the dose intensity of cisplatin and the ability to obtain a complete remission: however, such a relationship could not be demonstrated for doxyrubicin and alkylating agents. It should be pointed out that this was a retrospective analysis performed over a rather narrow dose range of cisplatin. However, a prospective evaluation of the importance of dose intensity of cisplatin has been limited by the peripheral neuropathy which has precluded dose escalations of cisplatin beyond 100 mg/rrr' in most situations. Carboplatin is more suitable for the evaluation of Acta Obstet Gynecol Scand Suppl /55

dose intensity than is cisplatin, because of the absence of neurotoxicity and nephrotoxicity. There are potential pharmacologic and biologic techniques which may decrease the dose limiting myelotoxicity of carboplatin and permit the evaluation of the importance of dose intensity over a much broader range. Shea et al. (5) have demonstrated that the dose of carboplatin can be escalated to 2000 mg/m' when administered together with autologous bone marrow transplantation before non-hematologic toxicity becomes dose limiting. In addition, clinical studies are in progress with peripheral stem cell transfusions which may decrease the dose limiting thrombocytopenia (6). The currently available cytokines for clinical use, Gm-CSF and G-CSF, can decrease the neutropenia associated with high-dose carboplatin but do not have a substantial effect on the thrombocytopenia. However, new cytokines are being developed, such as IL-3 and IL-I a which may have a stimulatory effect upon platelet precursors. An alternative way to increase the dose intensity of platinum compounds is to combine carboplatin and cisplatin (7). Numerous clinical trials are in progress, based upon the non-overlapping toxicities of these two agents. However, it remains to be determined whether one will gain an effective increase in dose intensity by using such techniques. Such studies may be facilitated by the use of pharmacologic agents which decrease the toxicities of platinum compounds. WR272I is a radiation protecting agent which has been demonstrated to protect against the myelotoxicity of cyclophosphamide. Studies are currently in progress to determine whether WR272I will also protect against the neurotoxicity of cisplatin and the myelosuppression of carboplatin (8, 9). New combination chemotherapy regimens will also be evaluated in the next decade, based upon the demonstration of activity of drugs in phase II trials. The most active agent identified in previously treated patients appears to be Taxol (10). This novel diterpene plant product which is extracted from the bark of western yew trees has been shown in phase II trials to be an extremely active agent in patients with Oc. In the Gynecologic Oncologic Group trial. Taxol was shown to have substantial activity in patients who had disease progression while taking cisplatin, or had a very short duration of remission to cisplatin (less than 6 months). Even in this highly unfavorable group of patients, there was a 7% complete remission rate as well as a 22% partial remission rate. The Gynecologic Oncology Group is currently randomizing patients with advanced OC to the two-drug regimen of Taxol plus cisplatin versus cisplatin plus cyclophosphamide. The dose limiting toxicity of Taxol has been myelosuppression and consequently there has been some reluctance to

Chemotherapy in ovarian cancer combine Taxol with carboplatin. However, it appears that Taxol plus carboplatin could be administered together with cytokines which are capable of reducing neutropenia and thrombocytopenia. Although hexamethylmelamine is not a new agent, it has only recently become generally available for clinical use. Hexamethylmelamine has been shown to have activity both in recurrent disease situations as well as in combination with other drugs in patients with previously untreated disease (11, 12). The H-CAP regimen as developed at Vanderbilt (12) has been shown to be particularly active in patients with small volume disease. In the long-term follow-up with the Vanderbilt study, the median survival of patients with small volume disease « 3 ern) treated with the H-CAP regimen is greater than 100 months. Furthermore, in a prospective randomized trial from the Mayo Clinic, the H-CAP regimen was statistically superior to the CP regimen in patients with small volume disease (defined as < 250 g of residual tumor) with regard to overall survival (13). Since prospective randomized trials have demonstrated that Adriamycin does not substantially enhance the efficacy of platinum-based chemotherapeutic regimens (14), it appears that the benefit of the H-CAP regimen may primarily relate to the addition of hexamethylmelamine. Combination chemotherapy regimens that combine carboplatin, cyclophosphamide and hexamethylmelamine are currently undergoing pilot trials in cooperative groups in the USA (15). Additional clinical trials are evaluating novel combinations which include carboplatin and ifosfamide (16). Ifosfamide too has been shown to have activity in previously treated patients with advanced OC, producing an overall response rate of 20% (17). The preliminary results suggested that carboplatinlifosfamide regimens, though active, will be associated with substantial myelosuppression which also may ameliorated by the use of cytokines.

Treatment of patients who achieve a partial remission While approximately 80% of patients will achieve a response to platinum-based chemotherapy, only half of the patients will achieve a complete remission. Improvements in survival have correlated with the ability to achieve a complete remission and consequently strategies are being tested which may be capable of converting patients who achieve a partial remission into a complete remission. One strategy is to use very high-dose chemotherapy with autologous bone marrow transplantation in this situation. Two clinical trials have demonstrated that high-dose che-

57

motherapy with autologous bone marrow transplantation is capable of achieving responses in heavily treated patients (5, 18). However, it appears that this form of treatment will be most effective in patients with small volume disease and when administered at a time when patients are not fully drug resistant. Consequently, clinical trials have recently been initiated in patients who have drug-sensitive tumor and who show only a partial response to 4-6 cycles of induction platinum-based chemotherapy. These patients have not yet manifested drug resistance and high-dose chemotherapy with autologous bone marrow transplantation may be capable of increasing the complete remission rate. An alternative way to increase the complete remission rate is to utilize intraperitoneal chemotherapy in patients who have little tumor volume remaining after induction intravenous chemotherapy. Clinical trials have demonstrated that approximately one-third of patients will show a negative third-look result after treatment with i.p. chemotherapy (19). However, there is no currently available data to demonstrate that i.p. chemotherapy has a curative potential in this situation. Intraperitoneal chemotherapy appears to be limited by several factors (20) including: 1) poor penetration of platinum into tumor nodules. It appears that platinum will penetrate only 1-2 mm into tumor masses and consequently this form of therapy will not be useful in patients with gross macroscopic disease. 2) Patients with advanced ovarian cancer frequently have retroperitoneal nodal involvement (30-40%) and this may represent a pharmacologic sanctuary from the high levels of drug which can be achieved in the peritoneal cavity via the intraperitoneal route. 3) Patients frequently have adhesions, which prevent adequate drug distribution throughout the peritoneal cavity. Consequently, while i.p. chemotherapy has been shown to be technically feasible in most patients with OC and leads to pharmacologic advantage, its overall role in the management of patients with advanced disease remains to be established, based upon the completion of ongoing clinical trials.

New studies in the treatment of patients who achieve a complete remission As noted, approximately 30-50% of patients who achieve a surgically confirmed complete remission ultimately relapse. Intense salvage therapy is ineffective in curing these patients; the development of an effective treatment capable of reducing the recurrence rate would lead to a substantial improvement in overall surgical. Table I lists experimental strategies which are currently being tested in patients who Acta Obstet Gynecol Scand Suppl 155

58

R. F. Ozols

Table I. Experimental approaches to prevention of recurrences from a complete remission. Intraperitoneal therapy Chemotherapy Immunotherapy p32

Systemic therapy Hexamethylmelamine Ifosfamide Non-cross-resistant combination High-dose chemotherapy with ABMT Whole abdominal irradiation

achieve a complete remission. In general, maintenance with consolidation therapy has not been proven useful in preventing recurrences in patients with solid tumors and who achieve a complete remission. However, OC may be a different situation, in as much as the primary reason why most patients relapse is that they have undetected residual disease at the completion of induction therapy. Even at second-look laparotomy, one is unable to adequately sample all potential sites of disease in the peritoneal cavity.

Pharmacologic reversal of drug resistance The major factor which limits the effectiveness of chemotherapy is the development of drug resistance. Not only do patients become resistant to individual drugs used in their induction regimen, they also exhibit broad cross-resistance which limits the effectiveness of salvage therapy. Cisplatin is, as noted, the most active agent in the treatment of patients with OC cancer and the mechanisms responsible for the development of resistance to cisplatin are undergoing extensive investigation. Table II lists some of the mechanisms that have been identified as playing a role in the development of resistance to platinum compounds. There is broad cross-resistance between alkylating agents, radiation, and cisplatin. One mechanism which may be responsible for the multi-drug resistance phenotype of OC relates to increased cellular levels of glutathione (GSH). It has been demonstrated in human OC cell lines that there is a correlation between drug resistance and GSH concentration (21). The lowest GSH levels are found in cell lines from previously untreated patients, while the highest concentrations are observed in cell lines established from drugresistant patients. Furthermore, drug-resistant variant cell lines established by incubation of a drugsensitive cell line, by means of increasing concentraActa Obstet Gynecol Scand Suppl /55

tions of either melphalan, cisplatin, or Adriamycin, have also increased the concentration of GSH. Reduction of GSH by incubating the cells with buthionine sulfoximine (BSO), a synthetic amino acid which is an irreversible inhibitor of the enzyme gamma glutamyl cysteine synthetase, reduces the GSH concentration and increases the cytotoxicity of platinum compounds and alkylating agents (22). Furthermore, in a nude mouse model of human OC in which the mice die of intra-abdominal carcinomatosis, it has been demonstrated that oral administration of BSO leads to an approximately 90% reduction in GSH concentration in the malignant tumor cells, the bone marrow, and the gastrointestinal mucosa. Reduction of GSH concentration markedly potentiates the cytotoxic effects of melphalan in this relevant in vivo model of human OC. Whereas melphalan alone has little impact upon survival, pretreatment with BSO leads to a 72% increase in survival, with long-term survival also observed in some of the animals (22). In non-tumor ovarian mice, BSO administration did not potentiate the toxicity of melphalan. Based upon these studies, a phase I trial of BSO plus melphalan is currently in progress (23). Preliminary results from this study have been reported and it has been demonstrated that GSH concentrations can be reduced in most cases, particularly in peripheral mononuclear cells as well as in the tumor cells of some patients (23). It is not certain whether or not BSO substantially increases the toxicity of melphalan in this patient population. It has also been demonstrated that both cisplatin and alkylating agent resistance to human OC cells is associated with increased DNA repair capacity (24). Inhibition of the DNA repair enzyme, DNA polymerase a, lowers the repair capacity and potentiates the cytotoxicity of platinum compounds in drug-resistant cells. Aphidicolin is currently undergoing clinical evaluation in Europe and trials of aphidicolin plus cisplatin are being planned in the USA (25). An alternative way to alter the detoxification pathways present in drug-resistant cells is to inhibit the glutathione S-transferase (GST) enzymes. It has been shown by Wang & Tew that the inhibition of GST restores sensitivity to certain alkylating agents in drug-resistant cell lines (26). Based upon these observations, clinical trials have been initiated of thiotepa together with ethacrynic acid, an inhibitor Table II. Mechanism of resistance to cisplatin. Alterations in intracellular drug level Inactivation by conjugation with GSH Binding to metallothionein Removal of DNA adducts from DNA

Chemotherapy in ovarian cancer Table III. Cor!elates of drug resistance in ovarian cancer. MDR-l Her 21neu oncogene DNA repair enzymes Polymerases Excision repair enzymes Glutathione Glutathione S-transferase Metallothionein

of GST, in drug-resistant patients at the Fox Chase Cancer Center. It is ultimately possible that combinations of modulators will have to be used, such as BSO together with aphidicolin in order to produce optimum results. Ovarian cancer is well suited for the study of mechanisms of drug resistance. It is a highly chemosensitive tumor and tissues are frequently available for analysis. A prospective study is currently in progress at the Fox Chase Cancer Center in which the tumors from previously untreated patients are assayed for a series of biochemical and molecular parameters (Table III), at the time of diagnosis. The patients then undergo treatment and, at the time of relapse, additional biopsies are assayed for the same factors. In this prospective manner, the specific mechanisms of drug resistance may be identified which will in turn lead to clinical trials aimed at reversing identified drug resistance in individual patients. The development of non-cross-resistant platinum compounds is also an area of active investigation. Phase I trials have recently begun with Ormaplatin (tetraplatin) based on laboratory observations of a lack of cross-resistance between this platinum complex and cisplatin in some drug-resistant human OC cell lines (26).

Conclusion Chemotherapy will continue to play an important role in the overall management of patients with ovarian cancer. New drugs are being developed which are not completely cross-resistant, such as Taxol and Ormaplatin. In addition, substantial progress has been made in identifying mechanisms of drug resistance associated with alkylating agents and platinum compounds. Clinical trials are already in progress with agents demonstrated in laboratory models to be capable of reversing the drug resistance phenotype. Thus, while there has been general disappointment with the end results of platinum-based chemotherapy in the last decade, there should be

59

considerable optimism regarding the identification of ways to improve the effectiveness of chemotherapy which may lead to an improvement in survival.

References 1. Ozols RF. Ovarian cancer. Sem Surg Oncol 1990; 6: 328-38. 2. Alberts DS, Mason-LikkilN, O'Toole RV, et at. Randomized phase III trial of chemoimmunotherapy in patients with previously untreated stages III and IV suboptimal disease ovarian cancer: A Southwest Oncology Group study. Gynecol Oncol 1989;32: 8-15. 3. Canetta R, Bragman K, Smaldone L, Rozencweig M. Carboplatin: Current status and future prospects. Cancer Treat Rev 1988; 15: 17-31. 4. Levin L, Hryniuk WM. Dose intensity analysis of chemotherapy regimens in ovarian carcinoma. J Clin Oncol 1978; 5: 5, 756-67. 5. Shea TC, Flaherty M, Elias A, et at. A phase I clinical and pharmacokinetic study of carboplatin and autologous bone marrow support. J Clin Oncol 1989; 7: 5, 651-61. 6. Mason JR, Mullen M, Bessent R, et at. Peripheral blood stem cells in patients receivingmultiple cyclesof high dose carboplatin and granulocyte macrophage colony stimulating factor. Proc Am Soc Clin Oncol 1991; 10: 108. 7. Gill I, Muggia F, Christian M, et at. Phase I study of carboplatin day I, and cisplatin, day 3. Proc Am Soc Clin Oncol 1990; 9: 74. 8. Mollman JE, Glover DJ, Hogan WM, Furman RE. Cisplatin neuropathy. Risk factors, prognosis and protection by WR2721. Cancer 1988; 61: 2192-95. 9. Spiritos M, Schuchter L, Glover D, et at. Phase I trial of carboplatin and WR2721. Proc Am Soc Clin Oncol 1991; 9: 67. 10. Thigpen T, Blessing J, Ball H et at. Phase II trial of Taxol as second-line therapy for ovarian carcinoma: A GOG study. Proc Am Soc Clin Oncol 1990;9: 156. 11. Manetta A, MacNeil C, Lyter JA, et at. Hexamethylmelamine as a single second-line agent in ovarian cancer. GynecolOncol 1990; 36: 93-96. 12. Hainsworth JD, Grosh WW, Burnett LS, et at. Advanced ovarian cancer: Long-term results of treatment with intensive cisplatin-based chemotherapy of brief duration. Ann Int Med 1988; 108: 2, 165-70. 13. Edmonson JH, Wieand HS, McCormickGW. Role of hexamethylmelamine in the treatment of ovarian cancer: Where is the needle in the haystack? J Natl Cancer Inst 1172. 14. Omura GA, Bundy BA, Berek JS, et at. Randomized trial of cyclophosphamide plus cisplatinwith or without doxorubicin in ovarian carcinoma: A GynecologicOncology Group study. J Clin Oncol 1989; 7: 4,457-65. 15. Hoosier Oncology Group: [study in progress). 16. Gallasher CJ, WiltshawE, Coleman RE, et at. A dose escalation study of carboplatin and ifosfamide in advanced ovarian cancer. Cancer Chemother Pharmacol 1989; 24: 54-57. Acta Obstet Gynecol Scand Suppl/55

60

R. F. Ozols

17. Sutton GP, Blessing lA, Photopulos G, et al. Phase II experience with ifosfamide/mesna in gynecologic malignancies: Preliminary report of Gynecologic Oncology Group studies. Semin Oncol 1989; 16: 1,68-72. 18. Shpall E, Clarke-Pearson D. Soper IT, et al. High dose alkylating agent chemotherapy with autologous bone marrow support in patients with Stage III-IV ovarian cancer. Gynecol Oncol 1990; 38: 386-91. 19. Reichman B, Markman M, Hakes T, et al. Intraperitoneal cisplatin and etoposide in the treatment of refractory/recurrent ovarian carcinoma. I Clin Oncol 1989; 7: 9, 1327-32. 20. Ozols RF. Intraperitoneal chemotherapy in ovarian cancer. Time's up. J Clin Oncol 1991; 9: 197-99. 21. Hamilton TC, Winker MA, Louie KG, et al. Augmentation of adriamycin, melphalan, and cisplatin cytotoxicity in drug-resistant and -sensitive human ovarian carcinoma cell lines by buthionine sulfoximine mediated glutathione depletion. Biochem Pharmacol 1985; 34: 2583-86. 22. Ozols RF, Louis KG, Plowman J, et al. Enhanced melphalan cytotoxicity in human ovarian cancer in vitro and in tumor-bearing nude mice by buthionine sulfoximine depletion of glutathione. Biochem Pharmacol 1987; 36: 147-53. 23. Hamilton TC, O'Dwyer P, Young R, et al. Phase I trial

Acta Obstet Gynecol Scand Suppl /55

24.

25.

26.

27.

of buthionine sulfoximine plus melphalan in patients with advanced cancer. Proc Am Soc Clin Oncol 1990; 9: 73. Lai GM, Ozols RF, Symth JF, et al. Enhanced DNA repair and resistance to cisplatin in human ovarian cancer. Biochem Pharmacol 1988; 37: 4597-600. Lai GM, Ozols RF, Young RC, Hamilton TC. Effect of glutathione on DNA repair in cisplatin-resistant human ovarian cancer cell lines. J Natl Cancer Inst 1989; 81: 535-39. . Wang AL, Tew KD. Increased glutathione Svtransferase in a cell line with acquired resistance to nitrogen mustards. Cancer Treat Rep 1985; 69: 677-82. Behrens BC, Hamilton TC, Masuda H, et al. Characterization of a cisplatin resistant human ovarian cancer cell line and its use in evaluation of platinum analogs. Cancer Res 1987; 47: 414-18.

Address for correspondence:

Robert F. Ozols, M.D., Ph.D. Department of Medical Oncology 7701 Burholme Avenue Philadelphia, PA 19111 USA