Intraperitoneal chemotherapy for advanced ovarian cancer

Intraperitoneal chemotherapy for advanced ovarian cancer

Intraperitoneal chemotherapy for advanced ovarian cancer William E. Lucas, M.D., Maurie Markman, M.D., and Stephen B. Howell, M.D. San Diego, Califor...

546KB Sizes 2 Downloads 197 Views

Intraperitoneal chemotherapy for advanced ovarian cancer William E. Lucas, M.D., Maurie Markman, M.D., and Stephen B. Howell, M.D.

San Diego, California Treatment of patients with advanced ovarian cancer who have failure of first-line chemotherapy is rarely effective. Preliminary pharmacokinetic and phase II clinical studies established the feasibility of delivering relatively high concentrations of cisplatin intraperitoneally via a semipermanent catheter, while using intravenous sodium thiosulfate as a neutralizing agent to decrease the nephrotoxicity of cisplatin. Sixty patients with advanced ovarian cancer, all of whom had failure of first-line chemotherapy (including cisplatin in 56 of 60), were treated with high-dose intraperitoneal cisplatin in combination with doxorubicin and/or cytarabine. Of the 46 patients evaluable for response, 19 (42%) showed an objective response, most often (12/19) disappearance of malignant ascites. No serious drug-associated morbidity was observed aside from three cases of intestinal obstruction which may have been due in part to druginduced adhesions. It is felt that prospective studies to compare the efficacy of intraperitoneal chemotherapy with other forms of "salvage" therapy, as well as its use as initial chemotherapy for advanced ovarian cancer, need to be done. (AM J OBSTET GYNECOL 1985;152:474-8.)

Key words: Ovarian cancer, intraperitoneal chemotherapy Treatment of ovarian cancer that persists after aggressive operation and systemic chemotherapy is often ineffective. Alternative systemic chemotherapeutic regimens are rarely successful and abdominal irradiation can be expecterl to benefit only those patients with minimal residual disease, often at the cost of substantial morbidity. In a search for better strategies to control persistent or recurrent ovarian cancer, a system for delivering high concentrations of specific chemotherapeutic agents intraperitoneally via a totally implanted peritoneal access catheter has been developed.' Preliminary pharmacokinetic studies with cisplatin,2 one of the most effective drugs currently available to treat ovarian cancer, showed that intraperitoneal administration could achieve local levels many times higher than those achievable in plasma. It was also shown that concurrent systemic administration of sodium thiosulfate could prevent nephrotoxicity. A preliminary phase II clinical trial of intraperitoneal cisplatin in combination with

From the Division of Gynecologic Oncology, Department of Reproductive Medicine, and the Division of Hematology/Oncology, Department of Internal Medicine, University of California (San Diego). Supported by Grants CA 23100 and CA 35309 from the National Cancer Institute and from the General Clinical Research Center of University of California (San Diego) Medical Center. Dr. Howell is a Clayton Foundation investigator. This work was conducted in part by the Cwyton Foundation, California Division. Presented at the Fifty-first Annual Meeting of the Pacific Coast Obstetrical and Gynecological Society, Tucson, Arizona, October 2127, 1984. Reprint requests: William E. Lucas, M.D., Department of Reproductive Medicine, UCSD Medical Center, 225 Dickinson, San Diego, CA 92103.

474

doxorubicin and cytarabine suggested an objective response rate of 35% to 45%.' The present report describes the results achieved with intraperitoneal chemotherapy in a larger group of patients with refractory ovanan cancer.

Methods and clinical material The study population comprised 60 consecutive patients with refractory ovarian cancer. All patients signed an informed consent approved by the Human Subjects Committee, University of California (San Diego) School of Medicine. The major characteristics of these cancers in terms of stage, grade, and histologic type and the amount of disease present at the time intraperitoneal therapy was begun are as shown in Tables I to IV. Fifty-six of 60 patients had received multiple courses of cisplatin and cyclophosphamide and 53 patients had received doxorubicin in addition. Four patients had received an alkylating agent alone or in combination with 5-fluorouracil, methotrexate, and hexamethylmelamine. At the time of referral for intraperitoneal therapy, 42 patients had ascites and/or tumor masses known to be >2 em in diameter. Eighteen patients had tumor masses <2 em in diameter or positive cytologic results only. In addition to having failure of front-line chemotherapy, patients had to have recovered from any toxicities attributable to prior therapy and have a life expectancy of >2 months. Values for serum creatinine had to be ,.,; 1.5 mg/dl, bilirubin <3.0 mg/dl, white blood cell count ;;.3000/mm', and platelet count ;;.IOO,OOO/mm3 • All patients either had a semipermanent catheter

Intraperitoneal chemotherapy for ovarian cancer

Volume 152 Number 4

Table I. Initial stage of disease

I

Stage IC liB III IV

475

Table III. Results of histologic examination No. of patients

Type

No.

1 3 44 12

Serous cystadenocarcinoma Solid adenocarcinoma Endometrioid Mucinous

39 13 4 4

Table II. Tumor grade Grade

No. of patients

1 (well differentiated) 2 (moderately differentiated) 3 (poorly differentiated)

6 20 34

(Tenckhoff or Toronto Western Hospital) attached to a subcutaneously implanted reservoir (Port-a-Cath) placed into the peritoneal cavity prior to beginning the treatment program or were treated by the percutaneous placement of a peritoneal dialysis catheter with each treatment course. Access to the reservoir of the semipermanent catheters was via a 19-gauge Huber needle inserted percutaneously after careful antiseptic preparation of the overlying skin. Port-a-Cath peritoneal portals and 19-gauge Huber needles were purchased from Pharmacia Nu Tech, Piscataway, New Jersey; Tenckhoff catheters were purchased from Cobe Laboratories, Inc, Lakewood, Colorado; Toronto Western Hospital catheters were purchased from Accurate Surgical Instruments Co., Toronto, Ontario, Canada. The first 15 patients were treated with three drugs, including cisplatin at a dose of 100 or 200 mg/m 2 along with sodium thiosulfate, cytarabine at a dose of 500 mg/2 L (I0- 3 mol/L), and doxorubicin at a dose of 20 or 2 mg/2 L treatment volume. The 20 mg dose of doxorubicin was abandoned early in the trial because of an unacceptable level of abdominal pain. The next 45 patients were treated with cisplatin, either 100 or 200 mg/m 2 , with cytarabine given at a dose of2000 mg (4 X I0- 3 mol/L) in the 2 treatment volume. All patients received intravenous sodium thiosulfate, 4 gm/m 2 as an initial bolus and 12 gm/m 2 as a 6-hour infusion beginning at the initiation of the chemotherapy instillation. The drugs were administered as rapidly as possible and remained in the peritoneal cavity for 4 hours, at which time an attempt was made to remove as much of the fluid as possible. All patients received overnight hydration prior to treatment. Therapy was repeated at 21- to 28-day intervals, depending on the patient's recovery from any toxicity of therapy. An intensive antiemetic regimen was used as part of this treatment program. Unfortunately, in spite of this antiemetic program, nausea and vomiting were quite significant.

Table IV. Amount of disease present at start of intraperitoneal chemotherapy Aggregate masses <2 em in diameter, no ascites Aggregate masses >2 em and/or ascites* *Three patients with ascites masses <2 em in diameter.

18 42

had aggregate

tumor

Table V. Favorable response to intraperitoneal chemotherapy Duration

Type of response Disappearance of ascites Decrease in size of masses Conversion of cytologic results, positive to negative Pathologic complete response

(mo)

12

4

2

3-7+ 3-20+ 4-8+ 9

Results Of the 60 patients, 14 are not evaluable for response at this time. At the time of insertion of the intraperitoneal catheter in these patients the presence of persistent cancer had been documented by biopsy or cytologic testing, but no disease was detectable by noninvasive techniques. To date there is no clinical evidence of active disease in this subgroup with follow-up ranging from 4 to 19 months (average 11 months) since the beginning of intraperitoneal chemotherapy. Among the 45 evaluable patients, 19 (42%) showed an evaluable response to intraperitoneal chemotherapy (Table V). Responses related to initial stage of disease, grade of disease, and amount of residual disease are shown in Tables VI to VIII. A total of 248 courses of intraperitoneal chemotherapy are evaluable for toxicity. Only six courses (2.5%) were associated with a serum creatinine rise to > 1.5 mg/ 100 dl. In each case the patient had been heavily pretreated with cisplatin and in all cases the value returned to baseline within several weeks. White blood cell counts dropped below 2000/mm 3 following only four courses, with platelet nadirs of <75,000/mm 3 being observed following 5% of courses. In no case did a patient require platelet transfusions or hospitalization for complications of leukopenia. Two patients developed culture-proved bacterial peritonitis. They both recovered completely with catheter removal. Thirteen courses of therapy were asso-

476

Lucas, Markman, and Howell

June 15, 1985 Am J Obstet Gynecol

Table VI. Tumor grade and response to intraperitoneal chemotherapy Grade

1 2

2

3

6 11

Total

19

Not evaluable

No response

Response

4

9 18 27

5 5

14

Table VII. Amount of residual tumor and response to intraperitoneal chemotherapy

Masses <2 em, no ascites Masses >2 em and/or ascites Total

Response

No response

2 17 19

23 27

Not evaluable

12

4

2

14

Table VIII. Initial stage of disease and response to intraperitoneal chemotherapy Stage

Response

No response

Not evaluable

IC liB Ill IV

2 14

21

1 1 9

Total

19

27

14

3

ciated with fever without any evidence of infection. Significant local pain, defined as lasting longer than 72 hours or requiring narcotic analgesia, developed following 60% of treatment courses with the 20 mg dose of doxorubicin, in 20% of courses with the 2 mg doxorubicin dose, and in fewer than 10% of courses when doxorubicin was not included in the treatment regimen. Adhesions following intraperitoneal chemotherapy have resulted in 55% of the catheters becoming oneway valves within 6 months. Three patients, two responders and one nonresponder, have required operation for intestinal obstruction due to adhesions. Two of these three patients had received intraperitoneal doxorubicin.

Comment Pharmacologic modeling has suggested that direct intracavitary administration of certain chemotherapeutic agents by dialysis may result in a much greater drug exposure for the tumor than intravenous administration of the same drugs.• Specifically, drugs with low clearance rates from the peritoneal cavity and high clearance rates from the systemic circulation will demonstrate this phenomenon. Two major potential benefits may result from such a therapeutic approach to malignancies localized to body cavities. First, if the efficacy of a particular agent were concentration-dependent, then direct intracavitary therapy resulting in increased local drug concentration might improve antitumor response. Second, if

6

3

a drug would be metabolized into a nontoxic form prior to entry into the systemic circulation during passage through the liver or upon entry into the plasma by the administration of a neutralizing agent, toxicity might be reduced. Several drugs have been investigated for intraperitoneal administration. Cisplatin has been delivered safely in a dose of up to 270 mg/m 2 with simultaneous intravenous sodium thiosulfate. 1 Thiosulfate was selected as a neutralizing agent on the basis of its ability to protect against cisplatin-induced renal toxicity in a murine model." It is thought to react covalently with cisplatin, with the resulting complex not being toxic or possessing antitumor activity. 5 Not only has a pharmacokinetic advantage for peritoneal cavity drug exposure been demonstrated, but the systemic delivery of active cisplatin is increased twofold (at an intraperitoneal dose of 270 mg/m 2 ) over that observed with an intravenous dose of 100 mg/m 2 • 1 Following in vitro demonstration that ovarian carcinoma is sensitive to concentrations of cytarabine achievable during intraperitoneal chemotherapy, the safety and efficacy of cytarabine in vivo, when administered by 5-day dialysis exchange, was demonstrated. 6 In this report we have described our initial experience with combination intracavitary chemotherapy in patients with refractory ovarian carcinoma. The response rate of 42% in this high-risk group of patients, all of whom had received intensive systemic chemotherapy, appears to be at least as favorable as alternative salvage regimens. It is noteworthy that 17 of 40 evalu-

Intraperitoneal chemotherapy for ovarian cancer

Volume 152 Number 4

able patients (42%) with ascites and/or residual tumor masses >2 em in diameter showed a favorable objective response. This latter group rarely benefits from abdominal irradiation or alternative systemic chemotherapy regimens after prior intensive treatment with cisplatin. It is also noteworthy that response rate does not appear to be related to tumor grade (T;;tble VI). Fourteen patients known at the time of initiation of intraperitoneal chemotherapy to have residual cancer, not evaluable by noninvasive techniques, have been followed up from 4 to 19 months without clinical evidence of recurrent disease. Since this is the same group of patients who are most likely to benefit from whole abdominal irradiation, long-term follow-up should yield preliminary information regarding the relative merits of these two forms of "rescue" therapy for systemic chemotherapy failures. An unresolved problem is the occurrence of adhesions following intraperitoneal chemotherapy. Approximately half of the catheters have become one-way valves within 6 months of placement, permitting introduction of fluid but not removal. Instillation of 32% dextran 70 at the time of initial catheter placement and after each course of chemotherapy has not been very effective in preventing adhesions. There is some evidence that nonsteroidal anti-inflammatory drugs such as ibuprofen may be of some value in preventing adhesions.7 The theory on which this therapy is proposed is that since arachidonic acid metabolites mediate the acute fibroproliferative inflammatory response following surgical trauma, high doses of an anti-arachidonic acid agent should suppress this response. It is to be hoped that appropriate clinical trials will fulfill this premise. Our current protocol includes the use of ibuprofen, 600 mg three times a day, starting at the time of catheter placement. Since this was a pilot study, future studies are needed to define the optimal drug combinations, concentrations, and schedules for intraperitoneal chemotherapy. Eventually randomized controlled clinical trials comparing the intraperitoneal route to systemic administration as the initial mode of chemotherapy for ovarian cancer will need to be done to determine whether the pharmacokinetic advantage achieved by the use of this form of therapy can be translated into increased response rates and longer periods of disease-free survival. REFERENCES I. Pfeifle CE, Howell SB, Markman M, Lucas WE. Totally

implantable system for peritoneal access. J Clin Oncol 1984;2:1277. 2. Howell SB, Pfeifle CE, Wung WE, et a!. intraperitoneal cisplatin with systemic thiosulfate protection. Ann Intern Med 1982;97:845. 3. Markman M, Green MR, Pfeifle CE, Lucas WE, Yon JL, Howell SB. Combination intracavitary chemotherapy in pa-

4.

5.

6. 7.

477

tients with stage III-IV ovarian carcinoma failing standard treatment regimens. Proc Am Soc Clin Oncoll983;2:147. Dedrick RL, Myers CE, Bungay PM, DeVita VT. Pharmacokinetic rationale for peritoneal drug administration in the treatment of ovarian cancer. Cancer Treat Rep 1978;62: 1. Howell SB, Taetle R. The effect of sodium thiosulfate on cis-dichlorodiammineplatinum. II. Toxicity and antitumor activity in the L 210 leukemia. Cancer Treat Rep 1980; 64:611. King ME, Pfeifle CE, Howell SB. Intraperitoneal cytosine arabinoside in ovarian carcinoma. J Clin Oncoll984;2:662. Nashimura K, Nakamura PM, diZerega GS. Ibuprofen inhibition of postsurgical adhesion formation: a time and dose response biochemical evaluation in rabbits. J Surg Res 1984;36: 115. Discussion

C. HILL, San Francisco, California. I am particularly pleased to be able to discuss this paper, not only because Dr. Lucas and his colleagues are pioneering new methods of chemotherapy in the treatment of advanced ovarian carcinoma but also because it allows me to acknowledge publicly my appreciation to him for allowing me to spend a day earlier this year in observing and learning his technique in establishing a delivery system for the administration of prolonged intraperitoneal chemotherapy. Finding residual disease in the peritoneal cavity following what appears to be a complete remission in response to systemic chemotherapy is an extremely disappointing experience to the clinician and is particularly devastating to the patient who has hopes of a cure. Despite excellent immediate results from surgical debulking and systemic chemotherapy, the long-term results are less than anticipated, and there is little evidence to date that the 5-year survival rates in ovarian epithelial malignancies have been significantly influenced by these methods. Most patients with ovarian malignancy still present initially with widespread peritoneal metastases and, unless all the tumor is removed or destroyed, will die of the disease within a few years. The age-adjusted death rate for ovarian cancer has increased from 3/100,000 female population in 1930 to 13 in 1975.' Although the immediate survival and the quality of life are improved by treatment that takes into consideration such factors as tumor volume, histologic grade, and location of metastases, the odds that a patient with Stage III disease will be living in 5 years are still < 10%. Ovarian cancer tends to remain localized to the peritoneal cavity. Because of this, attempts have been made to increase the local concentration of chemotherapeutic agents by intraperitoneal administration. Twenty years ago, thiotepa was being introduced into the peritoneal cavity, but it became evident that this particular agent was not activated until it had been absorbed into the bloodstream and carried to the liver. However, much has been learned about the pharmacokinetics of these agents since that time. In 1978 Jones et al! reported that there was a tenfold to thirtyfold difference in concentration between intraperitoneal and plasma levels DR. EDWARD

478

Lucas, Markman, and Howell

when methotrexate was administered intra peritoneally. Studies since that time have indicated there are distinct advantages to intraperitoneal administration of cisplatin, Adriamycin, and 5-fluorouracil with greater tumor exposure to the drugs and less exposure of the peripheral tissues and therefore less toxicity. Dr. Lucas' report today shows that effective combination chemotherapy with cisplatin and cytarabine can be administered to patients with disease that is refractory to systemic therapy and, furthermore, that this can be accomplished with relative safety and acceptable levels of toxicity. He has found that Adriamycin in a 20 mg dose injected into the peritoneal cavity, even though it is dissolved in a large volume of fluid, cannot be used because of severe abdominal pain. He has demonstrated that bacterial contamination of the peritoneal cavity and extensive adhesion formation with possible bowel obstruction are complications that should be watched for and may be preventable. Finally, he has pointed out that this is an experimental approach with much more work to be done in order to determine whether the intraperitoneal mode of administration can be improved in patients with persistent disease and if it might be advantageous as first-line therapy. Several questions come to mind. First, would you describe the antiemetic program used in this study? Second, with reference to the 46 evaluable patients, was there any relationship between the histologic grade of the tumor or the length and type of previous systemic chemotherapy and the presence or lack of a response to intraperitoneal therapy? Finally, in the 19 patients in whom a response was noted, can you give us some idea as to the mean duration of the responses? I wish to thank the essayist for sharing with us his experience and his insights. Only by continuing and supporting this type of clinical investigation can significant advances be made in the management of the most dread of all gynecologic disorders-ovarian cancer. REFERENCES 1. Malkasian GD, Cupps RE, Earle JD. Management of local and regional ovarian carcinoma. In: Ozols RF, Young RC, eds. Principles of cancer treatment. New York: McGrawHill, 1982:Chap 52. 2. Jones RB, Myers CE, Guarino AM, et al. High volume intraperitoneal chemotherapy ("belly bath") for ovarian cancer. Cancer Chemother Pharmacoll978;1:161.

DR. PAUL MoRROW, Los Angeles, California. This is one of the most promising techniques that has come along in treating ovarian cancer. I have a question about your results because 12 of these 17 responders were measured in terms of control of ascites. As you know, there are a number of agents that can be put into the peritoneal cavity that are not known to be antitumor drugs, for example, tetracycline, which can control as-

June 15, 1985 Am J Obstet Gynecol

cites. In the Gynecologic Oncology Group for example, control of serous effusion is no longer used as an indication of response. I wonder if you would comment about that. DR. JACK STAPLETON, Reno, Nevada. I have two questions. One is in reference to the possibility of using intraperitoneal baths as primary therapy. Do you think that at the conclusion of operation it is best to leave in an intraperitoneal tube or catheter for just an initial bath and subsequently place the patient on a regimen of systemic therapy? Second, did you notice any increase in complications in patients who received intraperitoneal chromic phosphate? DR. LuCAS (Closing). I appreciate Dr. Hill's comments. Our antiemetic regimen has been the popular "shotgun" approach with Decadron, droperidol, metoclopramide, hypnosis, deconditioning, and anything else that might possibly work, but it is basically three agents in combination: Decadron, droperidol, and metoclopramide. Dr. Hill asked about the type of previous chemotherapy. That is a very significant point to readdress because 56 of60 patients had previously been treated with multiple course of cisplatin. You might expect a very low response rate to intraperitoneal cisplatin, so any response rate would tend to be looked at favorably. Again, this is a preliminary pilot study. The duration of response rates cannot be determined with accuracy because we have not reached our end points in some groups, particularly the patients who had minimally evaluable disease. To answer Dr. Morrow's question with regard to ascites, I think that ascites can be treated in a number of different ways. The question is what is the. response rate with other methods compared to our method? Sixteen of these patients had little or no evaluable disease. The rest of these patients were considered end-stage failures. We had patients coming in from a number of areas and some of them were on the border of having obstruction. It was a subjective evaluation to judge whether they had 2 months' life expectancy. I can think of a number of patients who had failure of first-line chemotherapy, second-line chemotherapy, and irradiation and responded for quite a long period of time. I think that ascites control should be looked at very carefully in terms of validity of being an objective criterion. To answer Dr. Stapleton, I think the application of this technique as primary therapy is very logical and needs to be explored. Perhaps a compilation of data from around the country from various groups who are using this or similar techniques needs to be completed before we embark on its use as primary therapy. We do have a protocol that is approved by the Human Use Committee for front-line chemotherapy and has a very logical place in the treatment after surgical debulking of the tumor. The second questionwe do not have any patients who have had previous chromic phosphate but we would not hesitate to use it.