- Email: [email protected]
Prognostic implications of large volume residual disease in patients with advanced stage epithelial ovarian cancer
GYNECOLOGIC
ONCOLOGY
27, 220-225 (1987)
Prognostic Implications of Large Volume Residual Disease in Patients with Advanced Stage Epithelial Ovarian Cancer H. H. GALLION, M.D.,* J. R. VAN NAGELL, M.D.,* E. S. DONALDSON, M.D. ,* M. B. HANSON, M.D.,* AND R. J. KRYSCIO, PH.D.? Departments of Obstetrics and Gynecology, *Division of Gynecologic Oncology and tdtatistics, University of Kentucky Medical Center, Lexington, Kentucky 40536 Received April 8, 1986 Thirty-two patients with Stage III or IV epithelial ovarian cancer and residual tumor volumes in excess of 2 cm in diameter after initial debulking were treated with platinum and cyclophosphamide chemotherapy. Twenty-seven patients (84%) received at least six courses of chemotherapy. Six patients developed grade 3 or 4 hematologic toxicity and one patient died with granulocytopenia and sepsis. The actuarial survival of the total group of patients was 78% at 12 months, 27% at 24 months, and 11% at 36 month. Of patients with residual disease 2-4 cm in diameter, 82% were alive 12 months after diagnosis and 46% were alive at 24 months. In contrast, patients with > 4 cm residual disease had a 12month survival of 73% and a 24-month survival of only 7%. The size of residual tumor after surgery remains a very important prognostic factor in patients treated with platinumbased combination chemotherapy. Reoperation with futher tumor debulking should be considered in ovarian cancer patients referred for chemotherapy with large volume residual disease to maximize response to combination chemotherapy. o 1987 Academic press, IIK.
INTRODUCTION Ovarian cancer remains the leading cause of death from gynecologic cancer in the United States. It has been estimated that in 1986 over 11,000 women will die of this disease [l]. Despite these statistics, several significant advances have been made in the treatment of ovarian cancer during the past decade. More accurate pretherapy evaluation, and surgical staging have allowed proper individualization of therapy, and combination platinum-based chemotherapy regimens have shown increased activity against epithelial ovarian malignancies [2,3]. Although stage of disease, tumor grade, and volume of residual disease following surgical debulking have all been shown to be important prognostic factors in ovarian cancer, several recent investigations have indicated that combination chemotherapy may be effective even in the presence of large volume tumors [4-61. The present investigation was undertaken to determine the efficacy of combination platinum and cyclophosphamide (CP) chemotherapy in patients with advanced stage epithelial ovarian cancer who have large volume residual tumors after surgical debulking. 220 0090-8258187 $1SO Copyright All rights
0 1987 by Academic Press, Inc. of reproduction in any form reserved.
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
221
MATERIALS AND METHODS Subjects for this investigation were 32 patients with Stage III or IV epithelial ovarian cancer and residual tumor volumes in excess of 2 cm in diameter who were treated at the University of Kentucky Medical Center from 1979 to 1984. Primary surgery included total abdominal hysterectomy, bilateral salpingooophorectomy, omentectomy, appendectomy, and tumor debulking in 30 patients. Eight patients also underwent bowel and peritoneal resections to achieve maximal tumor reduction. In two patients, extensive surgery could not be performed and only tumor biopsies were obtained. Estimations of the location and size of all tumor masses were made in each case at the completion of surgery. No patient with residual disease < 2 cm in diameter was included in this study. Histologic review of the cell type and grade of each tumor was performed in all cases. A l-day course of combination chemotherapy consisting of intravenous cyclophosphamide (1000 mg/*) and &-platinum (50 mg/m*) was routinely started in the immediate postoperative period. Prior to the administration of chemotherapy, patients were hydrated and diuresis was established with intravenous furosemide (40 mg) and manmtol (12.5 g). Cyclophosphamide was then given as an intravenous bolus followed by a 30-min infusion of cis-platinum. Hydration was continued for a minimum of 2 hr following chemotherapy. The first course of chemotherapy was given in the hospital and subsequent courses were given on an outpatient basis. Minimum laboratory criteria for administration of chemotherapy included a white blood cell count of at least 3000/mm3, a platelet count of 100,000/mm3, a hematocrit of 30%, and a serum creatinine d 1.5 mg/dl. The dose of cisplatinum was modified for renal impairment according to the following schedule: serum creatinine < 1.1 mg/dl-total dose; serum creatinine 1.1-1.3 mg/dl, 25% dose reduction; serum creatinine > 1.3-1.5 mg/dl, 50% dose reduction; serum creatinine > 1.5 mg/dl, 100% dose reduction. Chemotherapy was repeated at 4week intervals for a maximum of 12 courses. Second-look laparotomy was performed in those patients with no clinical evidence of disease following 12 months of chemotherapy. After completion of treatment, patients were examined every 3 months for 1 year and every 6 months thereafter. Intravenous pyelography, CT scanning, and pelvic ultrasonography were performed when indicated in selected cases. RESULTS The clinical characteristics of patients studied are presented in Table 1. The mean age of the patients was 56 years and the mean gravidity 3. Twenty-eight TABLE 1 PATIENT CHARACTERISTICS
Age (years) Gravidity Height (in.) Weight (Ibs)
Mean
Range
56 3 64 145
30-79 o-11 60-69 77-237
222
GALLION
ET AL.
patients had FIG0 Stage III ovarian caner and four patients had Stage IV disease. Histologic parameters of the tumors are illustrated in Table 2. The most common cell type was serous cystadenocarcinoma, followed by mutinous cystadenocarcinema and endometrioid carcinoma. Sixty-six percent of all tumors were poorly differentiated. Following surgical debulking, all patients had residual tumors in excess of 2 cm in diameter and 47% had tumors larger than 4 cm in diameter. Twenty-seven patients received at least 6 courses of chemotherapy, and 16 (50%) received lo-12 courses. Reasons for cessation of chemotherapy included rapid disease progression in 11 patients and patient refusal in 3 cases. Toxicity associated with combination chemotherapy was graded according to COG criteria (Table 3). All patients in this study experienced nausea during or after chemotherapy. However, these symptoms were well controlled by antiemetic therapy. Six patients experienced grade 3 or 4 hematologic toxicity and one patient died with granuloctypenia and sepsis after her third course of chemotherapy. Renal toxicity was minimal and consisted of mild elevation of serum creatinine in six patients. Grade 1 neurotoxicity consisting of mild paresthesias was noted in two patients. One patient developed acute myelogenous leukemia after 12 months of chemotherapy and died with no evidence of disease 34 months after treatment. At the completion of chemotherapy, eight patients had no clinical evidence of disease and underwent second-look laparotomy. Three of these patients had grossly visible disease at the time of second-look surgery and all died of disease within 12 months. Two patients had microscopic disease at second-look laparotomy and both are alive and well without evidence of disease 21 and 23 months after surgery. Three patients had no evidence of disease at the time of second-look laparotomy. Two of these patients developed recurrent disease 15 and 33 months later and eventually died of disease. One patient remains alive and well with no evidence of disease 47 months after negative second-look surgery, No patient with residual disease in excess of 4 cm in diameter after initial tumor debulking had a negative second-look laparotomy. Actuarial survivals were determined by the Kaplan-Meier method [7]. The actuarial survival of the total group of patients was 78% at 12 months, 27% at 24 months, and 11% at 36 months (Fig. 1). Of patients with residual disease 24 cm in diameter, 82% were alive 12 months after diagnosis and 46% were alive TABLE 2 HISTOL~CIC
CHARACTERISTICS
OF TUMORS STUDIED
(N = 32) Cell type Serous cystadenocarcinoma Mutinous cystadenocarcinoma Endometrioid carcinoma Clear cell carcinoma Undifferentiated carcinoma Grade 1 2 3
No. 22 3 3 2 2 3 8 21
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
223
TABLE 3 CHEMOTHERAPY TOXICITY No. Myelotoxicity WBC (mm3) 3000-3999 2000-2999 1000-1999N-2SN Neurotoxicity None Mild paresthesias
Orll 0
II 6
12 I8 sufwv~~
IIll 24
4 3 3 3 10 4 3 1 2 0 0 1 0 11 18 3 0 26 6 30 2
11
3036424854606672 TIME
(hi0~~ki.9
FIG. 1. Actuarial survival of ovarian cancer patients with > 2 cm residual disease treated with CP chemotherapy (n = 32).
224
GALLION
ET AL.
at 24 months. In contrast, patients with residual disease in excess of 4 cm in diameter had a 12 months survival of 73% and only one patient was alive 24 months after initial surgery (Fig. 2). The survival curves for these two groups are significantly different when compared by the Mantel-Cox statistic (P < 0.04) [8]. Seventeen patients had ascites. All of these patients experienced progressive disease and their 2-year survival was only 35%. DISCUSSION
Due to the lack of early symptoms and effective screening methods, the majority of ovarian cancer patients present with Stage III or IV disease. Although stage of disease, tumor differentiation, and residual tumor volume have all been shown to be prognostically significant in ovarian cancer patients treated with single alkylating agent therapy, it was hoped that residual tumor volume would be less important in patients treated with platinum based combination chemotherapy. In an initial clinical trial from the Mt. Sinai Medical Center, Cohen and coworkers reported that the response rate of epithelial ovarian cancer to combination platinum chemotherapy was independent of tumor size [4]. Specifically, 40% of patients with maximum gross residual tumors 6 cm in diameter had no disease at second-look compared to 35% negative second-look findings in patients with < 2 cm residual disease. In contrast, Steiner and colleagues reported that the 2-year survival of ovarian cancer patients treated with PA chemotherapy was 81% when residual disease was less than 2 cm in diameter compared to only 34% with larger volume residual disease [9]. A question of fundamental importance in this study concerns the activity of CP chemotherapy in epithelial ovarian cancer. In a recent randomized study from Italy comparing different platinum combinations in 392 untreated patients with Stages III and IV ovarian cancer, Mangioni and co-workers concluded that the combination of CP was equally as effective as PAC and was associated with
0
6
12 I8
24
SURVIVAL
30
1 36 42 48 54 60 66 72 TIME (MONTHS)
FIG. 2. Actuarial survival of ovarian cancer patients treated with CP chemotherapy to the volume of residual disease (E’ < 0.04).
according
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
225
significantly less toxicity (10). Patients were all treated with six courses of CP chemotherapy in essentially the same doses as that in the present study. Since 27 of the 32 patients treated at the University of Kentucky received at least six courses of CP therapy, it is reasonable to assume that they were given an opportunity to respond to an effective regimen of platinum based chemotherapy. The findings of the present investigation indicate that the use of CP chemotherapy was associated with a relatively high short-term survival (78% at 12 months) in patients with advanced stage ovarian cancer and residual tumor volumes in excess of 2 cm. Unfortunately, the duration of survival was limited, and only 27% of the patients were alive 2 years after diagnosis. The survival of patients with residual disease > 4 cm in diameter was significantly lower than that of patients with smaller amounts of residual disease, and only one patient survived more than 24 months. These data confirm the previous observations of Smith and Day from the M.D. Anderson Hospital who reported a 5-year survival of only 8% in patients with Stage III epithelial ovarian cancer and residual disease 3-6 cm in diameter [ 111. These findings indicate that the size of residual tumor following surgery remains an extremely important prognostic variable in patients treated with platinum based chemotherapy, and emphasizes the need for aggressive tumor debulking. Patients referred for chemotherapy with large volumes of residual disease should be considered for reoperation with further tumor debulking to maximize response to combination chemotherapy. REFERENCES 1. Silverberg, E., and Lubera, J. Cancer statistics, 1986, CA Cancer J. C/in. 36, 9-25 (1986). 2. Griffiths, C. T., Parker, L. M., and Fuller, A. F. Role of cytoreductive surgical treatment in the management of advanced ovarian cancer, Cancer Treat. Rep. 63, 235-240 (1979). 3. Piver, M. S. Ovarian carcinoma-a decade of progress, Cancer 54, 2706-2715 (1984). 4. Cohen, C. J., Goldberg, J. D., Holland, J. F., Bruckner, H. W., Deppe, G., Gusberg, S. B., Wallach, R. C., Kabakow, B., and Rodin, A. Improved therapy with cisplatin regimens for patients with ovarian carcinoma (F.I.G.O. stages III and IV) as measured by surgical endstaging (second-look operation), Amer. J. Obstet. Gynecol. 145, 955-967 (1983). 5. Decker, D. J., Fleming, T. R., Malkasian, G. D., Webb, M. J., Jefferies, J. A., and Edmonson, J. H. Cyclophosphamide plus c&platinum in combination: treatment program for stage III or IV ovarian carcinoma, Obstet. Gynecol. 60, 481-487 (1982). 6. Gershenson, D. M., Wharton, J. T., Herson, J., Edwards, C. L., and Rutledge. F. Single agent c&platinum therapy for advanced ovarian cancer, Obstet. Gynecol. 58, 487-496, 1981. 7. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations, J. Amer. Statist. Assoc. 53, 457-481 (1958). 8. Mantel, N. Evaluation of survival data and two new rank order statistics arising in its consideration, Canad. Chem. Rep. 50, 163-170 (1966). 9. Steiner, M., Rubinov, R., Borovik, R., Cohen, Y., and Robinson, E. Multimodal approach (surgery, chemotherapy, and radiotherapy) in the treatment of advanced ovarian carcinoma, Cancer 55, 2748-2752 (1985). 10. Mangioni, C., Colombo, N., Marsoni, S., Boris, G., Billoni, C., Epis, A., Sessa, C., Valesecchi, Bianchi, U. A., and Pecorelli, S. Treatment of advanced ovarian cancer (OC): a randomized trial comparing different platinum (P) combinations, Gynecol. Oncol. 23, 252 (1986). 11. Smith, J. P., and Day, T. G. Review of ovarian cancer at the University of Texas Systems Cancer Center, M.D. Anderson Hospital and Tumor Institute, Amer. J. Obstet. Gynecol. 135, 984-993 ( 1979).
ONCOLOGY
27, 220-225 (1987)
Prognostic Implications of Large Volume Residual Disease in Patients with Advanced Stage Epithelial Ovarian Cancer H. H. GALLION, M.D.,* J. R. VAN NAGELL, M.D.,* E. S. DONALDSON, M.D. ,* M. B. HANSON, M.D.,* AND R. J. KRYSCIO, PH.D.? Departments of Obstetrics and Gynecology, *Division of Gynecologic Oncology and tdtatistics, University of Kentucky Medical Center, Lexington, Kentucky 40536 Received April 8, 1986 Thirty-two patients with Stage III or IV epithelial ovarian cancer and residual tumor volumes in excess of 2 cm in diameter after initial debulking were treated with platinum and cyclophosphamide chemotherapy. Twenty-seven patients (84%) received at least six courses of chemotherapy. Six patients developed grade 3 or 4 hematologic toxicity and one patient died with granulocytopenia and sepsis. The actuarial survival of the total group of patients was 78% at 12 months, 27% at 24 months, and 11% at 36 month. Of patients with residual disease 2-4 cm in diameter, 82% were alive 12 months after diagnosis and 46% were alive at 24 months. In contrast, patients with > 4 cm residual disease had a 12month survival of 73% and a 24-month survival of only 7%. The size of residual tumor after surgery remains a very important prognostic factor in patients treated with platinumbased combination chemotherapy. Reoperation with futher tumor debulking should be considered in ovarian cancer patients referred for chemotherapy with large volume residual disease to maximize response to combination chemotherapy. o 1987 Academic press, IIK.
INTRODUCTION Ovarian cancer remains the leading cause of death from gynecologic cancer in the United States. It has been estimated that in 1986 over 11,000 women will die of this disease [l]. Despite these statistics, several significant advances have been made in the treatment of ovarian cancer during the past decade. More accurate pretherapy evaluation, and surgical staging have allowed proper individualization of therapy, and combination platinum-based chemotherapy regimens have shown increased activity against epithelial ovarian malignancies [2,3]. Although stage of disease, tumor grade, and volume of residual disease following surgical debulking have all been shown to be important prognostic factors in ovarian cancer, several recent investigations have indicated that combination chemotherapy may be effective even in the presence of large volume tumors [4-61. The present investigation was undertaken to determine the efficacy of combination platinum and cyclophosphamide (CP) chemotherapy in patients with advanced stage epithelial ovarian cancer who have large volume residual tumors after surgical debulking. 220 0090-8258187 $1SO Copyright All rights
0 1987 by Academic Press, Inc. of reproduction in any form reserved.
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
221
MATERIALS AND METHODS Subjects for this investigation were 32 patients with Stage III or IV epithelial ovarian cancer and residual tumor volumes in excess of 2 cm in diameter who were treated at the University of Kentucky Medical Center from 1979 to 1984. Primary surgery included total abdominal hysterectomy, bilateral salpingooophorectomy, omentectomy, appendectomy, and tumor debulking in 30 patients. Eight patients also underwent bowel and peritoneal resections to achieve maximal tumor reduction. In two patients, extensive surgery could not be performed and only tumor biopsies were obtained. Estimations of the location and size of all tumor masses were made in each case at the completion of surgery. No patient with residual disease < 2 cm in diameter was included in this study. Histologic review of the cell type and grade of each tumor was performed in all cases. A l-day course of combination chemotherapy consisting of intravenous cyclophosphamide (1000 mg/*) and &-platinum (50 mg/m*) was routinely started in the immediate postoperative period. Prior to the administration of chemotherapy, patients were hydrated and diuresis was established with intravenous furosemide (40 mg) and manmtol (12.5 g). Cyclophosphamide was then given as an intravenous bolus followed by a 30-min infusion of cis-platinum. Hydration was continued for a minimum of 2 hr following chemotherapy. The first course of chemotherapy was given in the hospital and subsequent courses were given on an outpatient basis. Minimum laboratory criteria for administration of chemotherapy included a white blood cell count of at least 3000/mm3, a platelet count of 100,000/mm3, a hematocrit of 30%, and a serum creatinine d 1.5 mg/dl. The dose of cisplatinum was modified for renal impairment according to the following schedule: serum creatinine < 1.1 mg/dl-total dose; serum creatinine 1.1-1.3 mg/dl, 25% dose reduction; serum creatinine > 1.3-1.5 mg/dl, 50% dose reduction; serum creatinine > 1.5 mg/dl, 100% dose reduction. Chemotherapy was repeated at 4week intervals for a maximum of 12 courses. Second-look laparotomy was performed in those patients with no clinical evidence of disease following 12 months of chemotherapy. After completion of treatment, patients were examined every 3 months for 1 year and every 6 months thereafter. Intravenous pyelography, CT scanning, and pelvic ultrasonography were performed when indicated in selected cases. RESULTS The clinical characteristics of patients studied are presented in Table 1. The mean age of the patients was 56 years and the mean gravidity 3. Twenty-eight TABLE 1 PATIENT CHARACTERISTICS
Age (years) Gravidity Height (in.) Weight (Ibs)
Mean
Range
56 3 64 145
30-79 o-11 60-69 77-237
222
GALLION
ET AL.
patients had FIG0 Stage III ovarian caner and four patients had Stage IV disease. Histologic parameters of the tumors are illustrated in Table 2. The most common cell type was serous cystadenocarcinoma, followed by mutinous cystadenocarcinema and endometrioid carcinoma. Sixty-six percent of all tumors were poorly differentiated. Following surgical debulking, all patients had residual tumors in excess of 2 cm in diameter and 47% had tumors larger than 4 cm in diameter. Twenty-seven patients received at least 6 courses of chemotherapy, and 16 (50%) received lo-12 courses. Reasons for cessation of chemotherapy included rapid disease progression in 11 patients and patient refusal in 3 cases. Toxicity associated with combination chemotherapy was graded according to COG criteria (Table 3). All patients in this study experienced nausea during or after chemotherapy. However, these symptoms were well controlled by antiemetic therapy. Six patients experienced grade 3 or 4 hematologic toxicity and one patient died with granuloctypenia and sepsis after her third course of chemotherapy. Renal toxicity was minimal and consisted of mild elevation of serum creatinine in six patients. Grade 1 neurotoxicity consisting of mild paresthesias was noted in two patients. One patient developed acute myelogenous leukemia after 12 months of chemotherapy and died with no evidence of disease 34 months after treatment. At the completion of chemotherapy, eight patients had no clinical evidence of disease and underwent second-look laparotomy. Three of these patients had grossly visible disease at the time of second-look surgery and all died of disease within 12 months. Two patients had microscopic disease at second-look laparotomy and both are alive and well without evidence of disease 21 and 23 months after surgery. Three patients had no evidence of disease at the time of second-look laparotomy. Two of these patients developed recurrent disease 15 and 33 months later and eventually died of disease. One patient remains alive and well with no evidence of disease 47 months after negative second-look surgery, No patient with residual disease in excess of 4 cm in diameter after initial tumor debulking had a negative second-look laparotomy. Actuarial survivals were determined by the Kaplan-Meier method [7]. The actuarial survival of the total group of patients was 78% at 12 months, 27% at 24 months, and 11% at 36 months (Fig. 1). Of patients with residual disease 24 cm in diameter, 82% were alive 12 months after diagnosis and 46% were alive TABLE 2 HISTOL~CIC
CHARACTERISTICS
OF TUMORS STUDIED
(N = 32) Cell type Serous cystadenocarcinoma Mutinous cystadenocarcinoma Endometrioid carcinoma Clear cell carcinoma Undifferentiated carcinoma Grade 1 2 3
No. 22 3 3 2 2 3 8 21
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
223
TABLE 3 CHEMOTHERAPY TOXICITY No. Myelotoxicity WBC (mm3) 3000-3999 2000-2999 1000-1999
Orll 0
II 6
12 I8 sufwv~~
IIll 24
4 3 3 3 10 4 3 1 2 0 0 1 0 11 18 3 0 26 6 30 2
11
3036424854606672 TIME
(hi0~~ki.9
FIG. 1. Actuarial survival of ovarian cancer patients with > 2 cm residual disease treated with CP chemotherapy (n = 32).
224
GALLION
ET AL.
at 24 months. In contrast, patients with residual disease in excess of 4 cm in diameter had a 12 months survival of 73% and only one patient was alive 24 months after initial surgery (Fig. 2). The survival curves for these two groups are significantly different when compared by the Mantel-Cox statistic (P < 0.04) [8]. Seventeen patients had ascites. All of these patients experienced progressive disease and their 2-year survival was only 35%. DISCUSSION
Due to the lack of early symptoms and effective screening methods, the majority of ovarian cancer patients present with Stage III or IV disease. Although stage of disease, tumor differentiation, and residual tumor volume have all been shown to be prognostically significant in ovarian cancer patients treated with single alkylating agent therapy, it was hoped that residual tumor volume would be less important in patients treated with platinum based combination chemotherapy. In an initial clinical trial from the Mt. Sinai Medical Center, Cohen and coworkers reported that the response rate of epithelial ovarian cancer to combination platinum chemotherapy was independent of tumor size [4]. Specifically, 40% of patients with maximum gross residual tumors 6 cm in diameter had no disease at second-look compared to 35% negative second-look findings in patients with < 2 cm residual disease. In contrast, Steiner and colleagues reported that the 2-year survival of ovarian cancer patients treated with PA chemotherapy was 81% when residual disease was less than 2 cm in diameter compared to only 34% with larger volume residual disease [9]. A question of fundamental importance in this study concerns the activity of CP chemotherapy in epithelial ovarian cancer. In a recent randomized study from Italy comparing different platinum combinations in 392 untreated patients with Stages III and IV ovarian cancer, Mangioni and co-workers concluded that the combination of CP was equally as effective as PAC and was associated with
0
6
12 I8
24
SURVIVAL
30
1 36 42 48 54 60 66 72 TIME (MONTHS)
FIG. 2. Actuarial survival of ovarian cancer patients treated with CP chemotherapy to the volume of residual disease (E’ < 0.04).
according
PROGNOSTIC IMPLICATIONS
OF RESIDUAL
DISEASE
225
significantly less toxicity (10). Patients were all treated with six courses of CP chemotherapy in essentially the same doses as that in the present study. Since 27 of the 32 patients treated at the University of Kentucky received at least six courses of CP therapy, it is reasonable to assume that they were given an opportunity to respond to an effective regimen of platinum based chemotherapy. The findings of the present investigation indicate that the use of CP chemotherapy was associated with a relatively high short-term survival (78% at 12 months) in patients with advanced stage ovarian cancer and residual tumor volumes in excess of 2 cm. Unfortunately, the duration of survival was limited, and only 27% of the patients were alive 2 years after diagnosis. The survival of patients with residual disease > 4 cm in diameter was significantly lower than that of patients with smaller amounts of residual disease, and only one patient survived more than 24 months. These data confirm the previous observations of Smith and Day from the M.D. Anderson Hospital who reported a 5-year survival of only 8% in patients with Stage III epithelial ovarian cancer and residual disease 3-6 cm in diameter [ 111. These findings indicate that the size of residual tumor following surgery remains an extremely important prognostic variable in patients treated with platinum based chemotherapy, and emphasizes the need for aggressive tumor debulking. Patients referred for chemotherapy with large volumes of residual disease should be considered for reoperation with further tumor debulking to maximize response to combination chemotherapy. REFERENCES 1. Silverberg, E., and Lubera, J. Cancer statistics, 1986, CA Cancer J. C/in. 36, 9-25 (1986). 2. Griffiths, C. T., Parker, L. M., and Fuller, A. F. Role of cytoreductive surgical treatment in the management of advanced ovarian cancer, Cancer Treat. Rep. 63, 235-240 (1979). 3. Piver, M. S. Ovarian carcinoma-a decade of progress, Cancer 54, 2706-2715 (1984). 4. Cohen, C. J., Goldberg, J. D., Holland, J. F., Bruckner, H. W., Deppe, G., Gusberg, S. B., Wallach, R. C., Kabakow, B., and Rodin, A. Improved therapy with cisplatin regimens for patients with ovarian carcinoma (F.I.G.O. stages III and IV) as measured by surgical endstaging (second-look operation), Amer. J. Obstet. Gynecol. 145, 955-967 (1983). 5. Decker, D. J., Fleming, T. R., Malkasian, G. D., Webb, M. J., Jefferies, J. A., and Edmonson, J. H. Cyclophosphamide plus c&platinum in combination: treatment program for stage III or IV ovarian carcinoma, Obstet. Gynecol. 60, 481-487 (1982). 6. Gershenson, D. M., Wharton, J. T., Herson, J., Edwards, C. L., and Rutledge. F. Single agent c&platinum therapy for advanced ovarian cancer, Obstet. Gynecol. 58, 487-496, 1981. 7. Kaplan, E. L., and Meier, P. Nonparametric estimation from incomplete observations, J. Amer. Statist. Assoc. 53, 457-481 (1958). 8. Mantel, N. Evaluation of survival data and two new rank order statistics arising in its consideration, Canad. Chem. Rep. 50, 163-170 (1966). 9. Steiner, M., Rubinov, R., Borovik, R., Cohen, Y., and Robinson, E. Multimodal approach (surgery, chemotherapy, and radiotherapy) in the treatment of advanced ovarian carcinoma, Cancer 55, 2748-2752 (1985). 10. Mangioni, C., Colombo, N., Marsoni, S., Boris, G., Billoni, C., Epis, A., Sessa, C., Valesecchi, Bianchi, U. A., and Pecorelli, S. Treatment of advanced ovarian cancer (OC): a randomized trial comparing different platinum (P) combinations, Gynecol. Oncol. 23, 252 (1986). 11. Smith, J. P., and Day, T. G. Review of ovarian cancer at the University of Texas Systems Cancer Center, M.D. Anderson Hospital and Tumor Institute, Amer. J. Obstet. Gynecol. 135, 984-993 ( 1979).