GYNECOLOGIC ONCOLOGY 33, l-5 (1989)

Secondary Cytoreductive Surgery in Epithelial Ovarian Cancer: Nonresponders to First-Line Therapy’

MITCHELL MORRIS, M.D.,* DAVID M. GERSHENSON, M.D., AND J. TAYLOR WHARTON, M.D.

Department of Gynecology, The University of Texas, M. D. Anderson Cancer Center, Houston, Texas 77030

Received September 27, 1988

Thirty-three patients with progressive or stable epithelial ovarian cancer who underwent secondary tumor-reductive surgery at the University of Texas M. D. Anderson Cancer Center were ret- rospectively studied. All patients had primary cytoreductive surgery followed by chemotherapy. Lack of response to chemotherapy was clinically documented preoperatively. The tumors of 55% of the patients were secondarily cytoreduced to a greatest diameter of less than 2 cm. Sixty-six percent of the patients required bowel resection. Operative morbidity occurred in 24% of patients, mostly in those with bowel resection. Overall, median survival after sec- ondary surgery was 9.4 months. For patients with residual tumor less than 1 cm, median survival after secondary surgery was 19.5 months, compared with 8.3 months for patients with residual tumor of 1 cm or more (P < 0.004). Patients with an interval between primary cytoreductive effort and secondary surgery of less than 12 months survived a median of 7.3 months, compared with an 18.3-month median survival for patients with an interval of 12 months or more (P < 0.004). The present study provided no evidence that secondary surgery is of significant benefit in most patients who have progressive disease during chemotherapeutic treatment of epithelial ovarian cancer. o 1989 Academic PXSS, IUC.

INTRODUCTION

Cytoreductive surgery in epithelial cancer of the ovary is a well-established treatment approach. Numerous in- vestigators have shown improved survival rates in patients who have minimal residual tumor following the primary surgical effort [l-4]. But whether this result reflects the skill and aggressiveness of the surgeon or the biology of the tumor remains unresolved. Even more uncertain is the value of secondary cytoreductive efforts in patients in whom first-line therapy has failed. Published studies [5-81 of this issue have involved small numbers of patients with a wide variety of clinical courses.

’ Presented at the Thirty-Sixth Annual Clinical Meeting of the Amer- ican College of Obstetricians and Gynecologists.

’ Supported by American Cancer Society Award 86-259 and to whom correspondence should be addressed.

Patients who receive secondary cytoreductive surgery in epithelial ovarian cancer can be stratified into three major groups: (1) those who have undergone primary cytoreduction and while receiving first-line chemotherapy are clinically noted to have stable or progressive disease; (2) patients who have had a disease-free interval after primary therapy and then subsequently develop recurrent disease; and (3) those patients who have no clinical evi- dence of disease following chemotherapy and at the time of second-look laparotomy are found to have persistent disease.

The present study examines an 8-year experience at M. D. Anderson Cancer Center with secondary cyto- reductive surgery in patients of the first of these three categories--that is, in those in whom first-line therapy is definitely known to have failed before the end of chemotherapy. The second and third categories together will form the subject of a separate report from this institution.

PATIENTS AND METHODS

Between January 1, 1977, and December 3 1,1984,1572 patients presented to M. D. Anderson Cancer Center with malignant epithelial cancer of the ovary. Retrospective review identified 115 patients among these who underwent secondary cytoreductive surgery at this institution, all having received chemotherapy after the primary cyto- reductive effort. In 33 patients, whose cases constitute this report, secondary cytoreduction was prompted by the clinical finding of stable disease (2 patients) or pro- gressive disease (31 patients) before the completion of a planned course of chemotherapy. The remaining 82 pa- tients had secondary debulking at second-look laparotomy or recurrence.

Primary surgery had been at another institution in 30 of the study cases. Tumor volume and extent in these were determined by discussion of the findings with the

0090~8258/89 $1.50 Copyright 0 1989 by Academic Press, Inc.

All rights of reproduction in any form reserved.

2 MORRIS, GERSHENSON, AND WHARTON

outside surgeon, by review of the dictated operative sum- mary, and by pelvic examination on presentation to M. D. Anderson Cancer Center. The mean age at diagnosis of the 33 patients was 54 years (range 17-68). As one might expect, most of the cases (67%) were stage III by International Federation of Gynecology and Obstetrics criteria. Nine percent were stage IV, 18% stage I or II, and 6% unstaged. Six percent of the tumors were classified as grade 1, 24% as grade 2, and 58% as grade 3; the other 12% were not graded. None of the tumors were of low malignant potential. The histologic types were serous (42%), poorly differentiated (12%), mutinous (%), endometrioid (%), and clear cell (9%), with the remainder (18%) accounted for by tumors of mixed histology.

Accurate information regarding disease remaining after primary cytoreduction was available in 15 cases. There was no gross residual tumor in 5 of these, residual disease under 2 cm in diameter in 3, and 2 cm or more residual tumor or plaque in 7. Eighteen of the patients who had their surgery at an outside institution had insufficient documentation of residual tumor remaining. Sixteen pa- tients received single-agent chemotherapy, and 17 received combination chemotherapy. The single agents were all alkylating agents, and 7 of the combination regimens included cisplatin.

Thirty-three patients underwent secondary cytore- duction at M. D. Anderson Cancer Center. Tumor-re- ductive surgery, whenever technically feasible, was quite radical in an effort to reduce volume. The most common procedure was tumor debulking, indicating removal of tumor mass(es) without removal of adjacent structures (25 patients). Seventeen patients had segmental colon resection for metastatic disease. Ten patients underwent small bowel resection. Omentectomy was performed in 11 cases. Seven patients underwent hysterectomy, 4 per- itoneal stripping, and 3 genitourinary resection; 2 patients had a residual ovary removed. Pelvic exenteration was performed in 1 patient, splenectomy in 1, and pelvic node dissection in 1. The volume and extent of any tumor remaining after secondary surgery were in all cases care- fully documented, the former by measurement in centi- meters of the two greatest diameters and the latter by assessment of involvement of other organs or peritoneal surfaces. Following secondary surgery, 32 of the 33 pa- tients received second-line nonsurgical treatment; namely, chemotherapy in 29, irradiation in 2, and immunotherapy in 1.

Survival was calculated using the life table analysis of Berkson-Gage. Other statistical comparisons were made using the Lee-Desu statistic [91.

RESULTS

Secondary cytoreduction was a complete surgical suc- cess in 4 patients (i.e., there was no gross residual disease).

Three patients were left tumor less than 1 cm in diameter, and 11 with tumor greater than or equal to 1 cm but smaller than 2 cm. Ten patients had tumor of 2 cm or more. Five patients were left with plaque disease. Thus, optimal debulking (<2-cm residual) was accomplished in 55% of the patients. The most common sites of residual disease after secondary surgery were pelvic peritoneum (77%), diaphragm (45%), colon (42%), small bowel (32%), and retroperitoneal nodes (32%). The reasons for an in- ability to debulk to less than 2 cm were presence of retroperitoneal disease (7), excessive bulk (5), extensive bowel involvement (2), and poor medical risk for radical surgery (1). Of the 7 patients who had suboptimal primary cytoreduction, adequate reduction was achieved in 2 at the second effort. Six of the 8 patients with optimal primary reduction had less than 2 cm residual following secondary surgery.

The mean operative time was 266 min (range, 110- 450). Median blood loss was 1650 ml (range, 50-7000), and median intraoperative blood replacement was 3 units of blood (range, O-11). Median postoperative blood re- placement was 1 unit.

There were no intraoperative complications and there were no mortalities related to surgery. Eight patients suffered morbidity, including seven cases of prolonged ileus and seven cases of febrile morbidity requiring an- tibiotic therapy. Two patients had sepsis. Pelvic cellulitis, pneumonia, and mechanical small bowel obstruction oc- curred in one patient each. The patient with the most serious complication had disseminated intravascular co- agulation, sepsis, adult respiratory distress syndrome, and gastrointestinal bleeding.

Thirty-one percent of the patients who had bowel re- section suffered morbidity, whereas only 9% of those who did not have bowel resection had a morbid course. Those patients with optimal debulking had 22% morbidity, compared to 26% of those with suboptimal reduction.

At the time of analysis, three patients were alive. There was no clinical evidence of disease in two, at 100 and 93 months following secondary tumor reduction. The other surviving patient had indolent disease, at 92 months after secondary surgery. For the entire group, the median survival from diagnosis was 21.5 months. The median survival from secondary tumor reduction was 9.4 months.

Survival was affected by the amount of tumor remaining after secondary surgery. For the 18 patients with no residual or a residual of less than 2 cm, median survival from secondary tumor reduction was 12 months. If the residual disease was greater than or equal to 2 cm, patients survived a median 7.8 months (P < 0.03). If 1 cm is used as the dividing point (Fig. l), the median survival is 19.5 months (<l cm), versus 8.3 months (~1 cm) (P < 0.004).

Survival was not significantly affected by the type of first-line chemotherapy received (Fig. 2). The 16 patients who received single-agent therapy survived a median of

SECONDARY SURGERY IN OVARIAN CANCER 3

100-fyl-rc~

I\ \

-,801\ \ .P

01 . , , . : . , 0 10 20 30 40

Months

FIG. 1. Survival from the time of secondary tumor reduction based on residual tumor size. The squares indicate a maximum residual tumor diameter of less than 1 cm; the diamonds indicate residual disease of 1 cm or more (P < 0.004).

13 months from secondary cytoreduction. The 7 patients treated with a combination regimen that included cisplatin also survived a median of 13 months. The remaining 10 patients who received combination therapy without cis- platin survived a median of 9 months.

Response to chemotherapy following secondary cy- toreduction was also poor. Thirteen patients received second-line combination chemotherapy that contained cisplatin; 12 (92%) of these patients had progressive disease (the other patient was not evaluable). Fifteen patients were treated with combination regimens that did not con- tain cisplatin. Thirteen (87%) of these had progressive disease; 1 patient had a complete response, documented

SO-

L- I 10 20 30

Months

FIG. 2. Survival from the time of secondary tumor reduction based FIG. 3. Survival from the time of secondary tumor reduction ac- on the type of first-line chemotherapy. The closed squares indicate a cording to the interval between primary tumor reduction and secondary single-agent regimen (16 patients). The open squares indicate a com- tumor reduction. The squares indicate an interval of a year or less (20 bination regimen that included cisplatin (7 patients), and the diamonds patients); the diamonds indicate an interval of more than a year (13 a combination regimen without cisplatin (10 patients). patients).

with a second-look operation, and was without disease as of this report; the other was not evaluable. One patient received single-agent chemotherapy with melphalan and had a complete response. The 2 patients who received radiation therapy after secondary cytoreduction and the 1 who had immunotherapy, as well as 1 who had no further treatment, all died of diease.

The interval between primary and secondary surgery was a significant predictor of survival. When the 20 patients who underwent the operation within 12 months of the primary surgery are compared to the 13 with an operative interval of more than a year (Fig. 3), the median survivals are 7.3 months and 18.3 months, respectively (P < 0.004). This finding was independent of the ability to optimally reduce the tumor in these patients. Tumor was reduced to less than 2 cm in a similar proportion of the patients in either group.

Operative morbidity occurred in 7 of the 20 patients who underwent reoperation within a year compared with only 1 of the 13 who underwent reoperation after a year (P < 0.08).

DISCUSSION

The value of secondary cytoreductive surgery in ovarian cancer has been the subject of much speculation, yet few studies exist in the literature, and there are no reported prospective studies on this topic. The design of prospective trials is hampered by the small number of patients con- sidered for second surgery, the lack of a good second- line medical therapy, and the wide clinical variability within this patient population. Retrospective reviews, of which there are few, tend to have similar shortcomings.

“i-77--- 20 30

Months

4 MORRIS, GERSHENSON, AND WHARTON

Several reports describe “secondary” cytoreductive sur- gery, but many of the patients never underwent aggressive primary reduction, only chemotherapy [6,10,11].

Berek and co-workers [5], in a frequently cited review, analyzed 32 patients at the University of California at Los Angeles (UCLA) who underwent secondary reduction. They concluded that patients may benefit from secondary reduction when disease has only “partially progressed” on primary chemotherapy. As in our study, Berek found that residual disease of less than 1 cm and surgical interval of more than 12 months correlated with survival.

There is no attempt in the UCLA study to analyze patients according to their preoperative clinical course because the numbers were too small. In fact, most of their patients who did well were clinically without disease at the time of secondary surgery-that is, they underwent secondary reduction at second-look laparotomy. Patients who had progressive disease during chemotherapy were not separated from those with remission and subsequent recurrence.

The purpose of our study is to analyze the outcome of patients whose disease is not responding to first-line chemotherapy. In this group of patients, cytoreduction to less than 2 cm was possible at secondary surgery in the majority. However, only those who were left with less than 1 cm of tumor after second surgery had a sig- nificant survival advantage, and these were only 21% of the total group. The usual finding of the survival advantage in patients who have less than 2 cm of residual tumor after primary surgery may not be applicable to the sec- ondary cytoreductive procedure.

Of concern is the operative morbidity encountered in this group of patients, especially when a bowel resection is required. In one patient, no further therapy was ever given because of her debilitation following surgery. Patients who had a short interval between primary and secondary surgery had increased morbidity and shorter survival, suggesting that perhaps the biology of the tumor, and not our surgical ability, plays the major role. It should be noted that the overall median survival from the time of diagnosis in these patients was 21.5 months, which is similar to the overall median survival in other groups of patients with stage III ovarian cancer [11,12], a further suggestion that our surgical effort did not affect the course of the disease.

For one to justify the risks of secondary cytoreductive surgery, a reasonably effective second-line medical therapy must be available. When the patients in this study received first-line chemotherapy, only 21% received cisplatin. This encouraged the authors to attempt secondary surgery, since it was thought that an improved response to a platinum-containing combination might be seen. Unfor- tunately, 12 of the 13 patients subsequently treated with a cisplatin regimen had progressive disease. Cisplatin has

now been established as the optimal first-line chemo- therapeutic agent. Those patients who have progressive disease on this drug will usually be unresponsive to any other standard drug used in a second-line setting.

Recently, intraperitoneal chemotherapy has been pro- moted as an effective salvage modality for patients with minimal residual disease at second-look laparotomy [ 131. However, patients who do well with intraperitoneal treat- ment are primarily those who have demonstrated some response to first-line therapy and in whom there is no clinical evidence of disease at the time of second lapa- rotomy. In patients in whom disease is progressing on chemotherapy, there is little evidence that intraperitoneal chemotherapy is of benefit. While biologic response mod- ifiers and new chemotherapeutic agents have shown promise in some patients, as of yet none have shown significant efficacy in the second-line setting.

In patients who have unresponsive disease while re- ceiving first-line chemotherapy, there is little evidence that secondary cytoreduction is of any value. On the contrary, it is a morbid procedure that may result in prolonged hospitalization. If considered, this procedure should be carefully individualized to those patients who are eligible for second-line protocols and who have the capability to withstand the surgery.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

REFERENCES

Griffiths, C. T. Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma, Natl. Cancer Inst. Monogr. 42, 101-104 (1975). Hacker, N. F., Berek, J. S., Lagasse, L. D., Nieberg, R. K., and Elashoff, R. M. Primary cytoreductive surgery for epithelial ovarian cancer, Obsret. Gynecol. 61, 413-420 (1983).

Delgado, G., Oram, D. H., and Petrilli, E. S. Stage III epithelial ovarian cancer: The role of maximal surgical reduction, Gynecol. Oncol. 18, 293-298 (1984).

Munnell, E. W. The changing prognosis and treatment in cancer of the ovary. Amer. J. Obsfet. Gynecol. 100, 790-805 (1968).

Berek, J. S., Hacker, N. F., Lagasse, L. D., Nieberg, R. K., and Elashoff, R. M. Survival of patients following secondary cytore- ductive surgery in ovarian cancer, Obstet. Gynecol. 61, 189-193 (1983). Wiltshaw, E., Raju, K. S., and Dawson, I. The role of cytoreductive surgery in advanced carcinoma of the ovary: An analysis of primary and second surgery, Brit. J. Obstet. Gynaecol. 92,.522-527 (1985). Podratz, K. C., Malkasian, G. D., Jr., Hilton, J. F., Harris, E. A., and Gaffey, T. A. Second-look laparotomy in ovarian cancers: Evaluation of pathologic variables, Amer. J. Obstet. Gynecol. 152, 230-231 (1985). Maggino, T., Tredese, F., Valente, S., et al. Role of second look laparotomy in multidisciplinary treatment and in the follow up of advanced ovarian cancer, Eur. J. Gynaecol. Oncol. 4,26-29 (1983).

SPSS-X: Statistical package for the social sciences, 3rd ed. McGraw- Hill, New York (1988). Joyeux, H., Szawlowski, A. W., Saint-Aubert, B., Elashary, M. M., Solassol, C., and Pujol, H. Aggressive regional surgery for advanced ovarian carcinoma, Cancer 57, 142-147 (1986).

SECONDARY SURGERY IN OVARIAN CANCER 5

11. Young, R. C., Chabner, B. A., Hubbard, S. P., et al. Advanced phalan and c&platinum versus hexamethylmelamine, Adriamycin, ovarian adenocarcinoma. A prospective clinical trial of melphalan and cyclophosphamide in advanced ovarian cancer, Gynecol. Oncol. (L-PAM) versus combination chemotherapy, N. Engl. J. Med. 299, 15, 261-277 (1983). 1261-1266 (1978). 13. Howell, S. B., Zimm, S., Markman, M., et al. Long-term survival

of advanced refractory ovarian carcinoma patients with small-volume 12. Edwards, C. L., Herson, J., Gershenson, D. M., Copeland, L. J., disease treated with intraperitoneal chemotherapy, J. Clin. Oncol.

and Wharton, J. T. A prospective randomized clinical trial of mel- 5, 1607-1612 (1987).