Intraperitoneal carboplatin in the treatment of minimal residual ovarian cancer

Intraperitoneal carboplatin in the treatment of minimal residual ovarian cancer

GYNJXOLOGIC ONCOLOGY 36, 306-311 (1990) lntraperitoneal Carboplatin in the Treatment of Minimal Residual Ovarian Cancer PER PFEWFER,* OLAV BENNEDB...

647KB Sizes 0 Downloads 240 Views

GYNJXOLOGIC

ONCOLOGY

36,

306-311 (1990)

lntraperitoneal Carboplatin in the Treatment of Minimal Residual Ovarian Cancer PER PFEWFER,* OLAV BENNEDB&x,t AND KAMMA BERTELSEN* *Department of Oncology R, Odense University Hospital, Denmark, and TDepartment of Oncology, Aalborg Hospital, Denmark Received July 31, 1989

Thirty-one ovarian cancer patients with minimal residual disease after intravenous cisplatin combination chemotherapy were treated with intraperitoneal carboplatin (IP-Ca). The dose of IPCa was escalated from 150 to 350 mg/m’. Twenty-two patients received at least three courses of IP-Ca and were evaluable for efficacy. Third-look laparotomy was done in 10 patients. Two patients obtained a complete pathological response (CPR) lasting 33+ and 41+ months, respectively; 8 patients had minimal residual disease. Median survival for all patients was 14 -I- months. All patients were eligible for toxicity. Maximum tolerable dose in these heavily pretreated patients was 300 mg/m’ IP-Ca. The doselimiting toxicity was thrombocytopenia; in 27% of the patients who received 350 mg/m’ IP-Ca, WHO grade 4 thrombocytopenia was seen. No patient had WHO grade 4 lenkopenia. Subjective side effects consisted of mild nausea and vomiting (WHO). In conclusion, IP-Ca can produce CPR in heavily pretreated patients with only minor side effects. o two Academic PW, tn~. INTRODUCTION

The use of cisplatin-containing regimens in advanced ovarian cancer results in a pathologic complete response rate of about 25% [l-4]. However, at second-look laparotomy (SL) residual tumor with nodules measuring less than 1 to 2 cm, so called minimal residual disease, will be observed in U-20% of the patients U-31. Second-line intravenous (iv) chemotherapy after cisplatin-containing regimens has only rarely led to complete response. In contrast,

intraperitoneal

UP) chemo-

therapy has produced complete response in as high as 30% of patients with minimal residual disease [1,4,51. Because ovarian cancer remains confined to the peri-

toneal cavity for long periods and because intraperitoneal instillation of drugs offers the potential of exposing the surface of ovarian cancer nodules to extraordinarily high concentrations of drugs, the concept of this administration is attractive [4,61. Persistent

neurotoxicity

and nephrotoxicity

are how-

ever important limiting factors to the prolonged use of cisplatin. This makes carboplatin, a second generation platinum derivative with low neuro- and nephrotoxic potential, of interest. Therefore we decided to investigate this treatment modality. We report the results of intraperitoneal carboplatin (IP-Ca) in 31 patients with minimal residual disease. MATERIAL

Between April 1985and October 1988,31 patients with histologically verified ovarian cancer entered the present study. Patients who had received at least six courses of polychemotherapy with cis-platinum-containing regimens were eligible. Additional inclusiob criteria were minimal macroscopic (visible nodules <2 cm) or microscopic (detectable but not visible tumor) residual disease at second-look laparotomy. Therapy with IP-Ca had to be initiated within 4 weeks from second-look laparotomy. Exclusion criteria included (1) performance status >2; (2) age >69 years; (3) WBC <4 x log/liter and platelets
306 0090-8258/90

$1.50

Copyright 8 19!Xl by Academic Press, Inc. A~I rights of reproduction in any form reserved.

AND METHODS

CARBOPLATIN

IN THE TREATMENT

Carboplatin was diluted in 2 liters of normal saline at room temperature to ensure an equal distribution [7-91. For the administration of IP-Ca totally implantable portals or “single-dose” drug delivery systems were used. There were major problems with the totally implantable systems and therefore new single-use delivery systems (Veress needle for laparoscopic procedures) were used [ 10,111. We planned to drain the peritoneal cavity after 4 hr but after the first 22 instillations (seven patients) of carboplatin only a median of 500 ml was removed. Hematological toxicity was independent of the amount of drained fluid from the peritoneal cavity and the patients had minor discomfort during the draining procedure. Therefore we decided to remove the single-use system immediately after instillation of IP-Ca. During the next 24-48 hr the fluid (carboplatin and saline) was absorbed [1,41. Therapeutic efficacy was evaluated in terms of response, duration of response, and survival. A pathologic complete response was defined as disappearance of all known macroscopical tumors with negative biopsies and negative peritoneal washings. Progressive disease (PD) was defined as tumor nodules ~2 cm at relook laparotomy, clinical detectable tumor during IP-Ca, or development of ascites or hydrothorax containing malignant cells. No change (NC) was defined as minimal or microscopic residual disease at relook laparotomy. Duration of response for patients with complete pathologic response was defined as time elapsing between documented response and clinically verified relapse, death, or March 1989, whichever was reached first. Survival was defined as the time from first treatment with IP-Ca until death or March 1989, whichever was reached first. All patients who received at least one course of IP-Ca were considered eligible for toxicity. All patients who received at least three courses of IP-Ca were considered evaluable for efficacy and survival. Adverse effects were graded according to WHO [12]. No prophylactic antiemetics were administered as routine. Survival was estimated by the Kaplan-Meier method [13]. All median values are followed by range. RESULTS The characteristics of the 31 patients are summarized in Table 1. All 31 patients received at least 1 course of IP-Ca and were eligible for toxicity. They received a total of 135 courses of treatment. Median age was 56 years (36 to 67 years) at the time of first IP-Ca. Treatment was started a median of 14 months (9 to 49 months) after the primary treatment. At the time of primary diagnosis 2 patients had stage I or II disease, 28 had stage III disease, and 1 patient had stage IV disease according to

OF RESIDUAL

307

CANCER

TABLE 1 Characteristics of 31 Patients Evaluable for Toxicity Characteristic

No. of patients

Eligible for toxicity Total number of IP-chemotherapy Median age (years): Prior therapy Median time since primary treatment (months): CAP CaC-P PJAC Median No. of courses: Reasons for discontinuation of treatment Problem’s with the delivery system Nephrotoxicity Progressive disease Early death Dose of IP-Ca cl50 mg/m2 150 mg/m’ 200 mg/m* 250 mg/m’ 300 mg/m’ 350 mg/m*

31 135 56 (36-67) 14 (9-49) 26 3 2 10 (6-12) 6 1 I 1 21 19 44 21 17 13

Note. Median numbers are followed by range in brackets. CAP, cyclophosphamide, adriamycin, &-platinum; CaC-P, carboplatin, cyclophosphamide, &-platinum; P-JAC, c&platinum, carboplatin, Adriamycin, cyclophosphamide; IP-Ca, intraperitoneal carboplatin.

FIG0 (International Federation of Gynecologists and Obstetricians). Prior treatment consisted of cisplatin combination chemotherapy in all cases. Twenty-six patients had prior treatment with a median of 10 (6 to 12) courses of PAC (cisplatin, Adriamycin, and cyclophosphamide), 3 patients were treated with CaC-P (carboplatin, cyclophosphamide, and cisplatin) and 2 patients received P-JAC (cisplatin, carboplatin, Adriamycin, and cyclophosphamide). In 7 patients IP-Ca was administered using totally implantable systems and in 24 patients single-use delivery systems were used. Nine patients received only one or two courses of treatment and were thus not evaluable for treatment efficacy. The reasons for cessation of IP-Ca were problems with the delivery system (six patients), early death (one patient), nephrotoxicity (one patient), and progressive disease (one patient). So, 22 patients were evaluable for treatment efficacy (Table 2). They received a median of 6 (3 to 12) courses of IP-Ca. Median dose of carboplatin was 200 mg/m’ (125 to 350 mg/m*). Ten patients had a second-look laparotomy performed. Two patients, who obtained a pathologic complete response (both were treated with 200 mg/m’ carboplatin IP, 6 and 12 courses, respectively),

308

PFEIFFER,

Characteristics

TABLE 2 of 22 Patients Evaluable

Characteristic Evaluable for efficacy Median age (years): Disease at start of IP-Ca Macroscopic disease Microscopic disease Positive cytology only Starting dose of IP-Ca 150 mg/m’ 200 mg/m’ 250 mg/m” 300 mg/m’ 350 mg/m* Median number of courses: Median cumulative dose bdm’): Results Complete response No change Not evaluable Progressive disease Median survival (months):

BENNEDBRK,

for Efficacy

AND BERTELSEN

iooi=Y

No. of patients 22 51 (36-67) 8 12 2 2 11 3 2 4 6 (3-12) 1200 (600-3600)

60 I 6040 -

I

30 2010 0

I 0

I1

I1 12

1

I

I I 24 lknths

I

I

I 36

I

I,, 46

FIG. 1. Kaplan-Meier estimate of survival in 31 ovarian cancer patients with minima1 residual disease treated with intraperitoneal carboplatin. 2 12 9 12

13+ (5+ to 41+)

Note. Median numbers are followed by range in brackets. IP-Ca, intraperitoneal carboplatin.

are alive without sign of recurrent disease at 33 and 41 months, respectively. Eight patients had NC; however, 2 patients had disappearance of macroscopic residual disease. Twelve patients had no second-look laparotomy. Two patients had clinical PD after three and five courses of IP-Ca; they survived 6 and 16 months, respectively. Ten patients did not want to have a SL performed and were considered nonevaluable for response. However, 1 patient had palpable tumor above the vaginal top and obtained a clinically complete response. She was alive without signs of recurrent disease at 20 months. Two patients died after 10 and 14 months, respectively; 2 are known to have PD at 10 and 12 months; and 5 patients are alive without recurrent disease at median 9 + months (5+ to 23 + months). Median survival for all patients was 14+ months (4 to 45+ months). Thirteen patients died after median 14 months (4 to 28 months). The survival curve for all patients is seen in Fig. 1. Side effects due to the administration of IP chemotherapy has been described elsewhere [lo,1 II. No patients had major complications such as perforation of the bowel or bladder or peritonitis. Toxicity due to carboplatin was generally mild. The dose-limiting toxicity was thrombocytopenia (Fig. 2), 27% (95% confidence intervals: 7 to 60%) of patients who received 350 mg/m’ IP-Ca had WHO grade 4 thrombocytopenia. However, two of five patients who started

treatment with IP-Ca 350 mg/m’ had WHO grade 4 thrombocytopenia. No patient had WHO grade 4 leukopenia. No episodes of septicemia or severe hemorrhage were noted and no treatment-related deaths were seen. Subjective side effects consisted of mild nausea and vomiting (Table 3). One patient had treatment ceased after two courses of IP-Ca due to a slight compromised renal function; she also experienced a slight aggravation of cisplatin-induced neuropathy. DISCUSSION The prognosis of advanced epithelial ovarian cancer is poor with an overall 5-year survival rate of 15-20%. Median survival for all patients is around 24 months [ 1,3]. The use of cisplatin-containing regimens in advanced ovarian cancer induces complete clinical responses in about 50% of patients; however, only 25% achieve a complete pathologic response. Five-year survival for patients in complete pathologic response is about 70%. For patients with microscopic disease, 5-year survival variations from 20 to 70% have been published [14,15], but for minimal macroscopic disease it is only 20%. When treatment after second look is considered, it is thus necessary to distinguish between patients with microscopic disease and patients with minimal macroscopic disease. Some authors claim it to be at random whether a patient is classified as complete pathologic response or as microscopic disease at SL. Neijt et al. [I61 found a comparable survival for patients with complete pathologic response and microscopic disease. In contrast a Danish study showed significantly shorter survival for patients with residual disease compared to patients with complete pathologic response [15]. The opinion that the majority

CARBOPLATIN

IN THE TREATMENT

&ado&2 1 1

OF RESIDUAL

CANCER

309

Gmde3

///I4

W of patients

I se249

-349

Dose mg/m2 FIG.

2.

Hematological

toxicity during treatment with intraperitoneal

of patients classified as complete pathologic response really are microscopic disease is supported by the high frequency of recurrences in long-term survivors. After 5 to 10 years of observation 60 to 70% of the patients will haxe relapsed [ 171. Many authors argue therefore that not only patients with minimal macroscopic disease and microscopic disease but also patients with complete pathologic response should receive consolidation treatment after SL. Further iv chemotherapy, whole abdominal irradiation, and IP installation of radioactive or cytotoxic drugs have been tested as consolidation treatment, but at present, an improvement in survival rate has not been demonstrated. An Italian multicenter study group compared whole abdominal irradiation with iv chemotherapy as consolidation treatment in patients with minimal residual disease at SL [ 181. Overall survival was not superior to that of untreated patients in other series. The Danish Ovarian Cancer Group (DACOVA) has recently shown that iv chemotherapy after second-look in heavily pretreated patients has only marginal effect. Of 15 patients, TABLE

3

Subjective Side Effects Nausea and vomiting <200 200 250 300 350

mg’/m* mg/m2 mg/m* mg/m’ mg/m*

WHO grade, median (range) 0 (0

to

3)

1 (0 to 3) 1 (0 to 2) 2 (0

to 2)

1 (0 to 3)

carboplatin. Column 1: thrombocytopenia;

column 2: leukopenia.

treated with six courses of CAP before SL, only 1 patient had objective response at third-look laparotomy after further treatment with six courses of CAP [191. IP chemotherapy has been tested in the Dutch ovarian cancer group and has been shown to be technically amenable. At third-look laparotomy a rate of complete pathologic response of 30% was found. However, the impact on long-term survival could not be determined [4]. Ovarian carcinoma characteristically remains confined to the peritoneal cavity. The administration of IP chemotherapy has therefore biologic and pharmacokinetic advantages in the treatment of ovarian cancer patients. Peripheral neuropathy and nephrotoxicity is an important limiting factor to the prolonged use of cisplatin. This makes carboplatin, a second generation platinum derivate with a low neuro and nephrotoxic potential, of interest. Carboplatin has a higher molecular weight and a higher aqueous solubility than cisplatin; therefore it is cleared more slowly from the peritoneal cavity [20]. Peak peritoneal fluid concentration and area under the concentration versus time curve for IP-administered carboplatin confirms the increased exposure to the drug [20]. In addition, carboplatin is more slowly and less avidly bound to plasma proteins than cisplatin. This accounts for many of the pharmacokinetic differences between cisplatin and carboplatin. Plasma half-life of total platinum is shorter and a larger percentage is excreted in the urine when injected as carboplatin, compared with cisplatin [21]. Penetration of drugs from the peritoneal cavity into the tumor is restricted to a few millimeters. It is therefore

310

PFEIFFER.

BENNEDBzEK,

important that effective plasma concentrations are reached if tumor nodules are larger than a few millimeters. DeGregorio et al. [20] and Elferink et al. [22] found that plasma pharmacokinetics after IP administration of carboplatin was comparable to pharmacokinetics after iv administration. This means that in addition to exposure of the periphery of a tumor to high local drug concentrations, well perfused areas of tumor will also be exposed to drug concentrations equivalent to that obtained via direct iv administration. McVie et al. [l] withdrew IP fluid after a dwell time of 4 hr, and in contrast to our results, they found relative lack of myelotoxicity after IP treatment with 200-300 mg/m* carboplatin in pretreated patients. This is in accordance with the results of Elferink et al. [22], who showed that maximum tolerated dose (MTD) of IP carboplatin, when IP fluid is removed after 4 hr, may be up to 1.5 times higher than the MTD, when no attempt to drain the fluid is made. When fluid is instilled into the peritoneal cavity, approximately 2 liters is necessary [7-91 to assure that the whole peritoneal surface is reached. This amount can be instilled with only minor discomfort to the patient. Adhesions appear in a large number of patients and this seems to be one of the critical points to overcome. Tumor cells or clusters are often located in such adherences. The diffusion of chemotherapeutics into such clusters or lumps of malignant cells is impossible to predict. Parts of the peritoneal cavity may also be inaccessible because of these adherences. At present there is no sufficient method to ensure that instilled fluid and drug reach the whole peritoneal surface. X-ray pictures, CT scans, and ultrasonic control of fluid distribution are of limited value, yet they are the only methods available [ 1,4,23]. Numerous publications have described the use of different chemotherapeutic intraperitoneal regimens [6,24271; however, only a few authors have discussed the delivery systems employed [8,25,28,29]. Among the peritoneal access devices the totally implantable systems (Vascular-Access-Port or Port-a-Cath) seem to be superior to the Tenckhoff catheters [28]. At the Odense University Hospital the results for the totally implantable system were not comparable to the results reported by others. Therefore we started to use the Veress needle (Karl Storz GmbH&Co., Tuttlingen, West Germany) as a single-use intraperitoneal catheter for intraperitoneal chemotherapy [IO], without complications. Therefore it was possible to administer IP-Ca with only minor complications. At Aalborg Hospital an ultrasonic device was used for guiding the insertion of a pigtail catheter into the peritoneal cavity [l I]. This single-use system was also used without any significant complication.

AND BERTELSEN

CONCLUSION It is possible to obtain complete pathological response by the use of intraperitoneal carboplatin in previously intensively treated ovarian cancer patients with minimal residual disease. The treatment was well tolerated; the dose-limiting factor was thrombocytopenia. In these very heavily pretreated patients, where no attempt was made to drain the fluid, the maximum tolerated dose was 300 mg/m” carboplatin. Whether the conversion of microscopic disease to complete pathologic response has any effect on survival remains to be settled. Further studies, which distinguish between microscopic disease and minimal macroscopic disease, are needed. Randomized studies comparing IP chemotherapy and observation of patients with microscopic disease and complete pathologic response after iv chemotherapy are needed to demonstrate whether IP chemotherapy after SL has any influence on survival. REFERENCES 1. McVie, J. G., Ten Bokkel Huinink, W. W., Dubbelman, R., Franklin, H., v. d., Vijgh, W., and Klein, I. Phase I study and pharmacokinetics of intraperitoneal carboplatin, Cancer Treat. Rev. 12(Suppl. A), 35-41 (1985). 2. Ho, A. G., Beller, U., Speyer, J. L., Colombo, J. L., Wemz, J., and Beckman, E. M. A reassessment of the role of second-look laparotomy in advanced ovarian cancer, J. Clin. Oncol. 5, 13161321 (1987). 3. Bertelsen, K., Jacobsen, A., Andersen, J. E., Ahrons, S., Hjortkjaer Pedersen, P., Kjzer, H., Arffmann, E., Bichel, P., Boestofte, E., Strojer Christophersen, I., Gregersen, E., Kern Hansen, E., Holund, B., Jacobsen, M., Kaalund Jensen, H., Love Jepsen, F., Larsen, G., Sandberg Nielsen, E., Nyland, M., Olsen, J., Pandura, J., Rank, F., Sell, A., and Sogaard, H. A randomized study of cyclophosphamide and cis-platinum with or without doxorubicin in advanced ovarian carcinoma, Gynecol. Oncol. 28, 161169 (1987). 4. McVie, J. G. Intraperitoneal chemotherapy experience in the Netherlands Cancer Institute, Bit. J. Cancer 56, 702-703 (1987). 5. Ten Bokkel Huinink, W. W., Dubbelman, R., Aartsen, E., Franklin, H., and McVie, J. G. Experiental and clinical results with intraperitoneal cisplatin, Semin. Oncol. lZ(Supp1. 4), 43-46 (1985). 6. Zakris, E. L., Dewhirst, J. E., Riviere, J. E., Hoopes, P. J., Page, R. L., and Oleson, J. R. Pharmacokinetics and toxicity of intraperitoneal cisplatin combined with regional hyperthermia, J. Clin. Oncol. 5, 1613-1620 (1987). Dunnick, N. R., Jones, R. B., Doppman, J. L., Speyer, J., and Myers, C. E. Intraperitoneal contrast infusion for assessment of intraperitoneal fluid dynamics, Am. .I. Roentgen01 133, 221-223 (1979). Myers, C. E., and Collins, J. M. Pharmacology of intraperitoneal chemotherapy, Cancer Invest. l(5), 395-407 (1983)., Rosenshein, N., Blake, D., McIntyre, P. A., Parmley, T., Natarajan, T. K., Dvornicky, J., and Nickoloff, E. The effect of volume on the distribution of substances instilled into the peritoneal cavity, Gynecol. Oncol. 6, 106-110 (1978).

CARBOPLATIN

IN THE TREATMENT

10. Pfeiffer, P., Asmussen, L., Kvist-Poulsen, H., Bertelsen, K. Intraperitoneal chemotherapy: Introduction of a new “single use” delivery system. A preliminary report, Gynecol. Oncol., 35, 4749 (1989). 11. Bennedbak, O., Bartholdy, N. J., Fisker, P., and Panduro, J. Percutaneous catherisation of the peritoneal cavity guided by ultrasonography in the treatment of intraabdominal carcinoma of the ovary, Eur. J. Gyn. Oncol., in press. 12. WHO. WHO handbook for reporting results of cancer treatments, WHO Offset Publ. No. 48. WHO Geneva (1979). 13. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations, J. Amer. Stat. Assoc. 53, 457-481 (1958). 14. Copeland, L. J., and Gershenson, D. M. Ovarian cancer recurrences in patients with no macroscopic tumor at second-look laparotomy, Obstet. Gynecol. 68, 873 (1986). 15. Bertelsen, K., Hansen, M. K., Pedersen, P. H., Larsen, G., Nyland, M., Jacobsen, M., and Andersen, J. E. The prognostic and therapeutic value of second-look laparotomy in advanced ovarian cancer, Brit. J. Obstet. Gynecol. 95, 1231-1236 (1988). 16. Neijt, J. P., Ten Bokkel Huinink, W. W., v. d. Burg, M. E. L., v. Oesterom, A. T., Willemse, P. H. B., Heintz, A. P. M., v. Lent, M., Trinbos, J. B., Bouma, J., Vermorke, J. B., and v. Houwelingen, J. C. Randomized trial comparing two combination chemotherapy regimens (CHAP-5 v CP) in advanced ovarian carcinoma, J. Clin. Oncol. 5, 1157-1168 (1987). 17. Louie, K. G., Ozols, R. F., Myers, C. E., Ostchega, Y., Jenkins, J., Howser, D., and Young, R. C. Long-term results of a cisplatincontaining combination chemotherapy regimen for the treatment of advanced ovarian carcinoma, J. C/in. Oncol. 4, 1579-1585 (1986). 18. Mangioni, C., Epis, A., Vassena, L., Scalambrino, S., Placa, F., Colombo, A., and Marsoni, S. Radiotherapy (RT) versus chemotherapy (CH) as second line treatment of minimal residual disease (MRD) in advanced epithelial ovarian cancer (EOC), in Proceedings, Int. Gynecol. Cancer Sot. Session V-3 (1987) (Abstract). 19. Bertelsen, K., Jakobsen, A., Kern Hansen, M., Larsen, G., Nyland, M., Panduro, J., Pedersen, P. H., Strayer, I., and Andersen,

20.

21. 22.

23.

24.

25.

26.

27.

28.

29.

OF RESIDUAL

CANCER

311

J. E. A randomized trial of six cycles versus twelve cycles of cyclophosphamide, adriamycin and c&platinum (cap) in advanced ovarian cancer, Proc. ASCO 8, 583 (1989) (Abstract). DeGregorio, M. W., Lum, B., Holleran, W. M., Wilbur, B. J., and Sikic, B. I. Preliminary observations of intraperitoneal carboplatin pharmacokinetics during a phase I study of the Northern California Oncology Group. Cancer Chemother. Pharmacol. 18, 235-238 (1986). Van Echo, D. A., Egorin, M. J., and Aisner, J. The pharmacology of carboplatin, Semin. Oncol. lB(Supp1. 5), l-6 (1989). Elferink, F., van der Vijgh, W. J. F., Ten Bokkel Huinink, W. W., Dubbelman, R., and McVie, J. G. Pharmacokinetics of carboplatin after intraperitoneal administration, Cancer Ckemother. Pkarmacol. 21, 57-60 (1988). Markman, M. Intraperitoneal chemotherapy as treatment of ovarian carcinoma: Why, How and When? Obstet. Gynecol. Surv. 42, 533-539 (1987). Howell, S. B., Pfeifle, C. L., Wung, W. E., Ohlsen, R. A., Lucas, W. E., Yon, J. L., and Green, M. Intraperitoneal cisplatin with systemic thiosulfate protection, Ann. Intern. Med. 97, 845-851 (1982). Runowicz, C. D., Dottino, P. R., Shafir, M. K., Mark, M. A., and Cohen, J. J. Catheter complications associated with intraperitoneal chemotherapy, GynecoI. Oncol. 24, 41-50 (1986). Walton, L. A. Intraperitoneal chemotherapy in gynecologic malignancies, in Chemotherapy of gynecologic cancer (G. Deppe, Ed.), A. R. Liss, New York, pp. 321-330 (1984). Howell, S. B., Zimm, S., Markman, M., Abramson, I. S., Cleary, S., Lucas, W. E., and Weiss, R. J. Long-term survival of advanced ovarian carcinoma patients with small-volume disease treated with intraperitoneal chemotherapy, J. Clin. Oncol. 5, 1607-1612 (1987). Pfeifle, C. E., Howell, S. B., Markman, M., and Lucas, W. Totally implantable system for peritoneal access, J. Clin. Oncol. 11, 12771280 (1984). Leehey, D. J., Daugirdas, J. T., Ing, T. S, Geis, W. P., and Giacchino, J. L. The case against the “temporary” intraperitoneal dialysis catheter, Int. J. Artif. Organs 5, 334-335 (1982).