- Email: [email protected]
Does Debulking Surgery Improve Survival in Biologically Aggressive Ovarian Carcinoma?
GYNECOLOGIC ONCOLOGY ARTICLE NO.
67, 208–214 (1997)
GO974839
Does Debulking Surgery Improve Survival in Biologically Aggressive Ovarian Carcinoma? T. Le, M.D.,* G. V. Krepart, M.D.,† R. J. Lotocki, M.D.,† and M. S. Heywood, M.D.† *Department of Obstetrics, Gynecology, and Reproductive Sciences, Division of Gynecologic Oncology, University of Saskatchewan, Saskatoon, Saskatchewan S7N 0W8, Canada; and †Department of Obstetrics, Gynecology, and Reproductive Sciences, Division of Gynecologic Oncology, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada Received March 25, 1997
Aggressive tumor reduction surgery has been widely used in patients with advanced stage epithelial ovarian carcinoma before initiation of cytotoxic chemotherapy. No randomized controlled trial has been carried out to confirm the benefits of such procedures. To examine the role of cytoreductive surgery in the management of stage 2 and 3 patients with epithelial ovarian carcinoma treated with postoperative adjuvant platinum-based chemotherapy, survival analysis was carried out on patients with initial microscopic disease documented on staging laparotomies, patients with large volume of disease at time of exploration and tumor reduced to microscopic residuals, and patients who were suboptimally debulked with more than 2-cm residual disease. Twentyfour, 81, and 191 patients were identified from a computerized data base, respectively. Kaplan–Meier survival estimates showed that 62% with initial microscopic residual are alive with no evidence of disease at 5 years and 56% of patients left with microscopic residuals after tumor reduction are alive and well at 5 years. There was no statistical significant difference between these two groups. The groups are equivalent with respect to known adverse prognostic factors. In contrast, 5-year survival in the suboptimal debulked group was significantly lower at 15%. Debulking surgery to achieve microscopic residual disease improved the prognosis in patients with initial large volume of disease. Survival was similar to survival in patients with microscopic disease at time of exploration. The beneficial effect may be attributed to the removal of chemoresistant clones in bulky tumors. Tumor reduction surgery remains important in the management of advanced stage epithelial ovarian carcinoma. q 1997 Academic Press
INTRODUCTION
Epithelial ovarian carcinoma is the leading cause of death among all female genital tract cancers with an estimated 26,700 new cases and an associated 14,800 deaths in the United States in 1996 [1]. About three-quarters of all patients with invasive epithelial ovarian carcinoma will have intraperitoneal metastasis at the time of exploratory laparotomy. Optimal treatment of these patients has been widely accepted
Gyn 4839
/
6d24$$$141
MATERIALS AND METHODS
A retrospective chart review from 1976 to 1996 of all ovarian neoplasms treated at the Manitoba Cancer Treatment
208
0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
as being initial aggressive tumor resection followed by combination platinum-based cytotoxic chemotherapy. The goal of surgical debulking is to leave the smallest volume of residual disease possible. Patients with large amount of disease at the conclusion of surgery have a significantly shortened progression-free interval, as well as overall survival [2– 5]. Unfortunately, there has been no prospective randomized controlled trial to conclusively prove the benefits of radical tumor debulking procedures over initial treatment with chemotherapy. The role of tumor biology in influencing the eventual prognosis of patients with advanced stage epithelial ovarian carcinoma has only recently been recognized. Factors other than amount of residual disease are important in predicting survival. Hoskin et al. [6], in a recent reanalysis of GOG protocol 52, found a differential survival in patients with optimal surgical debulking (õ1-cm residual) favoring patients with initial small volume diseases and those with less than 20 residual nodules. The benefit of tumor reduction surgery in improving survival of advanced ovarian cancer had been questioned [15–18]. In this report we examine the benefit of debulking surgery by comparing patients with initial intra-abdominal microscopic disease documented on staging laparotomies, patients with large macroscopic disease resected to microscopic residuals, and patients with suboptimal debulking (ú2-cm residuals). It was hypothesized that patients with large-volume residuals have more aggressive tumor characteristics that preclude extensive resection and associated poorer prognosis. Those patients debulked to microscopic residual disease were assumed to have more favorable tumor biology with those presenting initially with microscopic diseases having the most indolent growth and the best prognosis.
10-20-97 01:02:04
goas
AP: GYN
209
DEBULKING SURGERY IN OVARIAN CARCINOMA
and Research Foundation was carried out. Patients’ demographic data, treatment modalities, tumor characteristics, and survival statuses were abstracted from initial histories, operative reports, pathology reports, and follow-up clinic notes. These data were entered into a computerized data base for subsequent analysis. All surgical procedures were performed by the same gynecologic oncologists (G.V.K., R.L., M.H.) and all specimens were reviewed by gynecological pathologists. Follow-ups were recorded from the time of treatment initiation until time of death or time of last available clinic note. From the data base, patients with epithelial ovarian cancer, excluding borderline and sarcoma histologies, were identified. Patients with surgically documented stage 1 or 4 and those not treated by adjuvant postoperative platinum-based chemotherapy were excluded. Patients with surgical stages 2 and 3 were selected and divided into three groups according to the initial and residual disease status. Group 1 consisted of patients with microscopic metastatic disease documented on staging laparotomies performed for a clinically stage 1 ovarian carcinoma. Our staging procedures routinely involve a midline abdominal incision, washings from both paracolic gutter and pelvis, careful palpation and inspection of upper abdomen and pelvis, total abdominal hysterectomy and bilateral salpingoophorectomy, total omentectomy, pelvic and para-aortic node samplings, and biopsies of adhesions, as well as multiple random peritoneal biopsies covering all serosal surfaces and diaphragm. Usually a total of 30 to 40 samples are obtained. Patients are classified as stage 3 on the basis of positive upper peritoneal biopsies, microscopic omental deposits, or involvement of the retroperitoneal lymph nodes. In about 4% of our positive staging laparotomies, microscopic positive nodes were the only involved site. Stage 2 patients have only positive biopsies of the pelvic peritoneum with negative omentum and negative nodes. Group 2 consisted of patients with initial large (ú2 cm) intra-abdominal metastasis most of whom were completely resected after TAH/BSO/omentectomy. About 5% of all patients in this group underwent bowel resections as part of the debulking procedures to achieve microscopic residual status. Group 3 consisted of patients having initial gross peritoneal carcinomatosis, but residual status was suboptimal (ú2 cm). Extensive debulkings with upper abdominal procedures were not routinely carried out once optimal residual disease status was deemed not achievable. Postoperatively, all patients were treated with combination of cis-platinum (50 mg/m2) and cyclophosphamide (600 mg/m2) every 4 weeks for a maximum of nine courses or until there was objective evidence of disease progression on physical examination and serial CA-125 measurements.
AID
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
Various salvage chemotherapeutic regimens were used after failure of first-line chemotherapy. Survival figures were computed using the product limit estimate described by Kaplan and Meier [7]. The log rank test was used for testing differences in survival among patient groups [8]. RESULTS
A total of 512 patients were selected (104 stage 2 and 408 stage 3) from 716 patients with epithelial ovarian carcinoma. Of note, patients were classified as having stage 2, in the optimally debulked group, on the basis of dense adhesions to surrounding structures or the presence of gross tumor spread in the pelvis and histologically negative omentum with no gross disease in the upper abdomen. We do not routinely perform random upper peritoneal biopsies nor retroperitoneal lymph node sampling in this patient population. One hundred seventy patients were excluded because macroscopic residual disease was less than 2 cm. Twentyfive of 130 patients with microscopic residuals and 21 of 212 patients with ú2-cm residuals who were not treated with platinum-based chemotherapy for various reasons were also excluded. Two hundred ninety-six patients were evaluated as per inclusion criteria. Twenty-four patients were in group 1, 81 in group 2, and 191 in group 3. Patients’ characteristics are presented in Table 1. Patients in group 3 were significantly older than those in the other groups. There are no significant differences between groups 1 and 2 regarding histopathological subtypes, grades, and stage distributions. Patients in the suboptimal debulked group (group 3) were more likely to have serous or anaplastic histologies, with an associated higher percentage of grade 3 tumors than those who had optimal debulkings. Multivariate analysis demonstrated the size of residual diseases as the single most important independent variable influencing survival across the three study groups. Table 2 shows the recurrence patterns and duration of response to salvage chemotherapy across the three study groups. There was no statistical significant difference in recurrence rate between groups 1 and 2 (42% vs 46%) with significantly more patients in group 3 suffering disease relapses. Survival curves of groups 1 and 2 are also statistically similar (P Å 0.6788). Five-year survival was 62% in group 1, 56% in group 2, and 15% in group 3 (Fig. 1). All surviving patients showed no clinical evidence of disease with normal CA-125 levels at 5 years. Further survival analysis within group 2 showed a significant difference favoring stage 2 patients over stage 3 patients even though all were left with microscopic residuals (Fig. 2). It is likely that some patients with true stage 1 were included in the stage 2 group, accounting for the favorable
goas
AP: GYN
210
LE ET AL.
TABLE 1 Patient and Tumor Characteristics in Staged and Debulked Study Groups Staging laparotomies (N Å 24) Mean age Histologies Serous Mucinous Endometrioids Clear cells Anaplastic Grade 1 2 3 Stage distribution in each group 2 3
Debulking microscopic res (N Å 81)
Debulking ú2-cm res. (N Å 191)
56.6
58.8
61.8
66.7% 8% 12.5% 4.8% 8%
42% 8.4% 26% 15% 8.6%
64.4% 2.6% 9% 5% 19%
9.5% 52.4% 38.1%
20% 36% 44%
2.2% 16% 81.8%
55.6% (N Å 45) a 44.4% (N Å 36)
37.5% (N Å 9) 62.5% (N Å 15)
0.5% (N Å 1) 99.5% (N Å 190)
a Stage 2 patients with complete resection of macroscopic disease (TAH / BSO / omentectomy) with no attempt made no document extra pelvic spread by multiple biopsies or retroperitoneal node biopsies.
difference in survival. After excluding stage 2 patients from all study groups, however, similar survival was still demonstrable between groups 1 and 2 (Fig. 3). Because of the apparent high percentages of clear cell and anaplastic histologies in groups 2 and 3, respectively, further analysis was carried out with respect to these subtypes. Fiveyear survival estimates were 71% for clear-cell patients compared to 53% for the rest of group 2 patients. In group 3, 5-year survival was 8% for anaplastic types compared to 16% for the other types combined. Interestingly, the patterns of recurrence were similar for groups 2 and 3 with 92% of relapses occurring intra-abdominally and 8% detected distantly. In contrast, 30% of group 1 recurrences were in a distant location. Significantly, more patients in group 1 than in group 2 or 3 had documented recurrences more than 5 years after diagnosis (10% vs 1%). Once disease recurrence was documented, there was no
significant difference in duration of responses to further salvage therapies across the study groups. DISCUSSION
About three-quarters of all patients with epithelial ovarian carcinoma have intra-abdominal metastasis. Primary cytoreductive surgery followed by combination cytotoxic chemotherapy has become the standard of care in these patients. Numerous reports have shown retrospectively a favorable impact of small residual tumor volume on the rates of complete clinical responses, negative second-look laparotomies, progression-free interval, and overall survival [2–5]. Various definitions of ‘‘optimal residual disease’’ exist but it is now generally accepted to be the presence of no tumor nodules that are more than 1 cm in diameter after surgery. Using this definition, about one-third of patients with advanced
TABLE 2 Recurrence Patterns and Responses to Salvage Therapy
Recurrences Intra-abdominal Distant Time to recurence õ2 years 2–5 years ú5 years Response to salvage therapy õ6 months 6 months–1 years ú1 year
AID
Gyn 4839
/
Staging laparotomies (N Å 24)
Debulking micro. residual (N Å 81)
Debulking ú2-cm residual (N Å 191)
42% (N Å 10) 70% (N Å 7) 30% (N Å 3)
46% (N Å 37) 92% (N Å 34) 8% (N Å 3)
86% (N Å 164) 92% (N Å 151) 8% (N Å 13)
90% 0% 10%
73% 27% 0%
96% 3% 1%
30% 30% 40%
36% 18% 46%
40% 31% 29%
6d24$$$142
10-20-97 01:02:04
goas
AP: GYN
DEBULKING SURGERY IN OVARIAN CARCINOMA
FIG. 1.
Survival of stage 2/3 patients with different residual diseases treated with adjuvant platinum-based chemotherapy. N Å 296 patients.
epithelial ovarian carcinoma could be debulked to optimal status [9]. The theoretical benefits of tumor reduction surgery include improvement in patient’s sense of well being; stimulating remaining tumor cells into active division, enhancing susceptibility to chemotherapy cell kills; improving delivery
FIG. 2.
AID
211
of chemotherapy by removing necrotic tumor masses; and decreasing the chance of dealing with chemoresistant cells. In 1979, Goldie and Coldman derived a mathematical model for relating the drug sensitivities of tumors to their spontaneous mutation rate [13]. The mathematical model predicts
Survival of stage 2 vs stage 3 patients with tumor reduced to microscopic residual disease treated with adjuvant platinum-based chemotherapy.
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
goas
AP: GYN
212
LE ET AL.
FIG. 3.
Survival of stage 3 patients with different residual diseases treated with adjuvant platinum-based chemotherapy. N Å 241 patients.
that the curability of a tumor (or the probability of finding no drug-resistant cells) decreases precipitously as the tumor approaches a critical size as determined by its intrinsic rate of spontaneous mutation. The clinical implications are that smaller tumors are easier to cure with chemotherapy and the best moment for effective treatment is as soon after detection as possible [14]. No randomized trial, however, had been carried out to confirm the theoretical benefit of tumor reduction. Questions have been raised as to whether patients who are optimally debulked have a less aggressive tumor biology that allows such a procedure or whether those patients would have done just as well if treated initially with neoadjuvant chemotherapy followed by surgical debulkings [18]. In this paper, we compare the outcomes of similarly treated patient groups with different initial disease volumes and residual status. Patients with initial microscopic disease presumably have the most favorable tumor characteristics and prognosis when compared to those with macroscopic disease, as predicted by the Goldie–Coldman hypothesis. In our study population, older age, serous/anaplastic histologies, and high-grade tumors acted as high-risk biologic markers in predicting suboptimal debulking, and residual disease more than 2 cm predicts significantly compromised survival (group 3). This is consistent with a recent report by Makar et al. [10]. Patient groups treated with staging laparotomies and optimal debulkings to microscopic residuals are similar in terms of stage, grade, and histologic subtype distributions. Survival analysis did not show a difference in patients
AID
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
presenting with initial microscopic disease compared to those with presumed poorer prognosis tumor with initial large macroscopic disease tumor reduced down to microscopic residuals (Fig. 1, P Å 0.6788). As expected, patients left with more than 2-cm residuals had significantly poorer survival than the other two groups (P õ 0.0001). It is possible that tumor reduction surgery to microscopic residuals reduces the number of chemotherapy-resistant clonogenic malignant cells to improve the prognosis in this subgroup of patients as predicted by the Goldie–Coldman hypothesis [13, 14]. Microscopic residual status is an important prognostic factor. Recent analysis at our institution showed similar survivals in patients with stage 3 epithelial ovarian carcinoma with õ2-cm and ú2-cm residuals. A significantly prolonged survival occurred only in those with microscopic residuals (Fig. 4). Similarities in locations and times of recurrences support the initial assumption of equal biologic behavior in the optimally debulked and suboptimally debulked patients. In contrast, many patients with initial microscopic disease tend to have distant and late (ú5 years) recurrences. Similar responses to salvage therapies after failure of primary chemotherapy were observed across study groups likely because of the emergence and presence of chemoresistant clones. The effect of tumor biology has recently been suggested as a prognostic factor. Farias-Eisner et al. [11] reported on 78 patients with stage 3 disease with minimal (õ5 mm) residual disease. Tumor grade, distribution of residual disease, and extent of carcinomatosis were found to indepen-
goas
AP: GYN
DEBULKING SURGERY IN OVARIAN CARCINOMA
Stage 3 patients treated with platinum-based chemotherapy as per residual disease status. N Å 330 patients.
FIG. 4.
dently affect survival despite optimal debulking surgery in all patients. Hoskin et al. [6] reported a significant difference in survival for ‘‘optimal’’ stage 3 epithelial ovarian cancer treated with platinum and cyclophosphamide with or without doxorubicin, favoring those patients with initial abdominal disease of 1 cm or less over those cytoreduced to 1 cm or less. The number of residual disease nodules (ú20) was also an adverse prognostic factor. It has been proposed that patients with large amount of disease at exploration may have had their disease for a longer time or may have a more aggressive biology conferring a poorer survival. The authors questioned the role of debulking surgery in improving prognosis in this patient population. As acknowledged by the authors, however, since all patients had macroscopic initial disease and all were treated initially with tumor reduction, the benefit of surgery could never be fully assessed. Despite being a retrospective study, which may contain some confounding variables and biases, our study confirms the benefits of cytoreductive surgery in improving survival in patients with macroscopic disease completely resected. The results from our analysis, as well as those from the recent EORCT phase 3 trial of interventional surgical debulking in patients with epithelial ovarian carcinoma [12], support the continuing use of aggressive tumor debulking procedures in the management of patients with advanced stage ovarian carcinoma. A definite randomized controlled trial studying cytoreductive surgery followed by chemotherapy versus neoadjuvant chemotherapy followed by tumor
AID
Gyn 4839
213
/
6d24$$$143
10-20-97 01:02:04
debulkings is necessary to prove the benefits of aggressive tumor resection. REFERENCES 1. American Cancer Society: Cancer facts and figures—1996. American Cancer Society, 1996 2. Griffiths CT: Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma. Natl Cancer Inst Monogr 42:101–104, 1975 3. Hacker NF, Berek JS, Lagasse LD, Niebrig RK, Elashoff RM: Primary cytoreductive surgery for epithelial ovarian cancer. Obstet Gynecol 61:413–420, 1983 4. Delgado G, Oram DH, Petrilli ES: Stage 3 ovarian cancer: The role of maximal surgical cytoreduction. Gynecol Oncol 18:293, 1984 5. Allen DG, Heintz AP, Touw FW: A meta-analysis of residual disease and survival in stage 3 and 4 carcinoma of the ovary. Eur J Gynecol Oncol 16(5):349–356, 1995 6. Hoskin WJ, Bundy BN, Thigpen TJ, Omura GA: The influence of cytoreductive surgery on recurrence-free interval and survival in small volume stage 3 epithelial ovarian cancer: A Gynecologic Oncology Group study. Gynecol Oncol 47:159–166, 1992 7. Kaplan EL, Meier P: Non parametric estimation from incomplete observation. J Am Stat Assoc 53:457–481, 1958 8. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163–170, 1966 9. Rubin SC: Surgery for ovarian cancer. Hematol Oncol Clin North Am 6(4):851–865, 1992 10. Makar AP, Baekelandt M, Trope CG, Kristensen GB: The prognostic significance of residual disease, FIGO substage, tumour histology, and grade in patients with FIGO stage 3 ovarian cancer. Gynecol Oncol 56:175–180, 1995
goas
AP: GYN
214
LE ET AL.
11. Farias-Eisner R, Teng F, Oliveira M, Leuchter R, Karlan B, Lagasse LD, Berek JS: The influence of tumor grade, distribution, and extent of carcinomatosis in minimal residual stage 3 epithelial ovarian cancer after optimal primary cytoreductive surgery. Gynecol Oncol 55:108– 110, 1994 12. Van der Berg MEJ, Van Lent M, et al.: The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. N Engl J Med 332:629, 1995 13. Goldie JH, Coldman AJ: A mathematical model for relating the drug sensitivity of tumours to their spontaneous mutation rate. Cancer Treat Rep 63:1727–1733, 1979 14. Goldie JH, Coldman AJ: The genetic origin of drug resistance in neoplasms: Implications for systemic therapy. Cancer Res 44:3643–3653, 1984
AID
Gyn 4839
/
6d24$$$143
10-20-97 01:02:04
15. Chambers JT, Chambers SK, Voynick IM, Schwartz PE: Neoadjuvant chemotherapy in stage X ovarian carcinoma. Gynecol Oncol 37:327– 331, 1990 16. Schwartz PE, Chambers JT, Makuch R: Neoadjuvant chemotherapy for advanced ovarian cancer. Gynecol Oncol 53:33–37, 1994 17. Surwitt E, Childers J, Atlas I, Nour M, Hatch K, Hallum A, Alberts D: Neoadjuvant chemotherapy for advanced ovarian cancer. Int J Gynecol Cancer 6:356–361, 1996 18. Schartz PE, Chambers JT, Kohorn EI, Rutherford TJ, Schubert W, Zelterman D: Neoadjuvant chemotherapy for advanced ovarian cancer: Long term survival. Proceedings of the Society of Pelvic Surgeons Annual Meeting, London, September 16, 1996, p 31
goas
AP: GYN
67, 208–214 (1997)
GO974839
Does Debulking Surgery Improve Survival in Biologically Aggressive Ovarian Carcinoma? T. Le, M.D.,* G. V. Krepart, M.D.,† R. J. Lotocki, M.D.,† and M. S. Heywood, M.D.† *Department of Obstetrics, Gynecology, and Reproductive Sciences, Division of Gynecologic Oncology, University of Saskatchewan, Saskatoon, Saskatchewan S7N 0W8, Canada; and †Department of Obstetrics, Gynecology, and Reproductive Sciences, Division of Gynecologic Oncology, University of Manitoba, Winnipeg, Manitoba R3T 2N2, Canada Received March 25, 1997
Aggressive tumor reduction surgery has been widely used in patients with advanced stage epithelial ovarian carcinoma before initiation of cytotoxic chemotherapy. No randomized controlled trial has been carried out to confirm the benefits of such procedures. To examine the role of cytoreductive surgery in the management of stage 2 and 3 patients with epithelial ovarian carcinoma treated with postoperative adjuvant platinum-based chemotherapy, survival analysis was carried out on patients with initial microscopic disease documented on staging laparotomies, patients with large volume of disease at time of exploration and tumor reduced to microscopic residuals, and patients who were suboptimally debulked with more than 2-cm residual disease. Twentyfour, 81, and 191 patients were identified from a computerized data base, respectively. Kaplan–Meier survival estimates showed that 62% with initial microscopic residual are alive with no evidence of disease at 5 years and 56% of patients left with microscopic residuals after tumor reduction are alive and well at 5 years. There was no statistical significant difference between these two groups. The groups are equivalent with respect to known adverse prognostic factors. In contrast, 5-year survival in the suboptimal debulked group was significantly lower at 15%. Debulking surgery to achieve microscopic residual disease improved the prognosis in patients with initial large volume of disease. Survival was similar to survival in patients with microscopic disease at time of exploration. The beneficial effect may be attributed to the removal of chemoresistant clones in bulky tumors. Tumor reduction surgery remains important in the management of advanced stage epithelial ovarian carcinoma. q 1997 Academic Press
INTRODUCTION
Epithelial ovarian carcinoma is the leading cause of death among all female genital tract cancers with an estimated 26,700 new cases and an associated 14,800 deaths in the United States in 1996 [1]. About three-quarters of all patients with invasive epithelial ovarian carcinoma will have intraperitoneal metastasis at the time of exploratory laparotomy. Optimal treatment of these patients has been widely accepted
Gyn 4839
/
6d24$$$141
MATERIALS AND METHODS
A retrospective chart review from 1976 to 1996 of all ovarian neoplasms treated at the Manitoba Cancer Treatment
208
0090-8258/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
as being initial aggressive tumor resection followed by combination platinum-based cytotoxic chemotherapy. The goal of surgical debulking is to leave the smallest volume of residual disease possible. Patients with large amount of disease at the conclusion of surgery have a significantly shortened progression-free interval, as well as overall survival [2– 5]. Unfortunately, there has been no prospective randomized controlled trial to conclusively prove the benefits of radical tumor debulking procedures over initial treatment with chemotherapy. The role of tumor biology in influencing the eventual prognosis of patients with advanced stage epithelial ovarian carcinoma has only recently been recognized. Factors other than amount of residual disease are important in predicting survival. Hoskin et al. [6], in a recent reanalysis of GOG protocol 52, found a differential survival in patients with optimal surgical debulking (õ1-cm residual) favoring patients with initial small volume diseases and those with less than 20 residual nodules. The benefit of tumor reduction surgery in improving survival of advanced ovarian cancer had been questioned [15–18]. In this report we examine the benefit of debulking surgery by comparing patients with initial intra-abdominal microscopic disease documented on staging laparotomies, patients with large macroscopic disease resected to microscopic residuals, and patients with suboptimal debulking (ú2-cm residuals). It was hypothesized that patients with large-volume residuals have more aggressive tumor characteristics that preclude extensive resection and associated poorer prognosis. Those patients debulked to microscopic residual disease were assumed to have more favorable tumor biology with those presenting initially with microscopic diseases having the most indolent growth and the best prognosis.
10-20-97 01:02:04
goas
AP: GYN
209
DEBULKING SURGERY IN OVARIAN CARCINOMA
and Research Foundation was carried out. Patients’ demographic data, treatment modalities, tumor characteristics, and survival statuses were abstracted from initial histories, operative reports, pathology reports, and follow-up clinic notes. These data were entered into a computerized data base for subsequent analysis. All surgical procedures were performed by the same gynecologic oncologists (G.V.K., R.L., M.H.) and all specimens were reviewed by gynecological pathologists. Follow-ups were recorded from the time of treatment initiation until time of death or time of last available clinic note. From the data base, patients with epithelial ovarian cancer, excluding borderline and sarcoma histologies, were identified. Patients with surgically documented stage 1 or 4 and those not treated by adjuvant postoperative platinum-based chemotherapy were excluded. Patients with surgical stages 2 and 3 were selected and divided into three groups according to the initial and residual disease status. Group 1 consisted of patients with microscopic metastatic disease documented on staging laparotomies performed for a clinically stage 1 ovarian carcinoma. Our staging procedures routinely involve a midline abdominal incision, washings from both paracolic gutter and pelvis, careful palpation and inspection of upper abdomen and pelvis, total abdominal hysterectomy and bilateral salpingoophorectomy, total omentectomy, pelvic and para-aortic node samplings, and biopsies of adhesions, as well as multiple random peritoneal biopsies covering all serosal surfaces and diaphragm. Usually a total of 30 to 40 samples are obtained. Patients are classified as stage 3 on the basis of positive upper peritoneal biopsies, microscopic omental deposits, or involvement of the retroperitoneal lymph nodes. In about 4% of our positive staging laparotomies, microscopic positive nodes were the only involved site. Stage 2 patients have only positive biopsies of the pelvic peritoneum with negative omentum and negative nodes. Group 2 consisted of patients with initial large (ú2 cm) intra-abdominal metastasis most of whom were completely resected after TAH/BSO/omentectomy. About 5% of all patients in this group underwent bowel resections as part of the debulking procedures to achieve microscopic residual status. Group 3 consisted of patients having initial gross peritoneal carcinomatosis, but residual status was suboptimal (ú2 cm). Extensive debulkings with upper abdominal procedures were not routinely carried out once optimal residual disease status was deemed not achievable. Postoperatively, all patients were treated with combination of cis-platinum (50 mg/m2) and cyclophosphamide (600 mg/m2) every 4 weeks for a maximum of nine courses or until there was objective evidence of disease progression on physical examination and serial CA-125 measurements.
AID
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
Various salvage chemotherapeutic regimens were used after failure of first-line chemotherapy. Survival figures were computed using the product limit estimate described by Kaplan and Meier [7]. The log rank test was used for testing differences in survival among patient groups [8]. RESULTS
A total of 512 patients were selected (104 stage 2 and 408 stage 3) from 716 patients with epithelial ovarian carcinoma. Of note, patients were classified as having stage 2, in the optimally debulked group, on the basis of dense adhesions to surrounding structures or the presence of gross tumor spread in the pelvis and histologically negative omentum with no gross disease in the upper abdomen. We do not routinely perform random upper peritoneal biopsies nor retroperitoneal lymph node sampling in this patient population. One hundred seventy patients were excluded because macroscopic residual disease was less than 2 cm. Twentyfive of 130 patients with microscopic residuals and 21 of 212 patients with ú2-cm residuals who were not treated with platinum-based chemotherapy for various reasons were also excluded. Two hundred ninety-six patients were evaluated as per inclusion criteria. Twenty-four patients were in group 1, 81 in group 2, and 191 in group 3. Patients’ characteristics are presented in Table 1. Patients in group 3 were significantly older than those in the other groups. There are no significant differences between groups 1 and 2 regarding histopathological subtypes, grades, and stage distributions. Patients in the suboptimal debulked group (group 3) were more likely to have serous or anaplastic histologies, with an associated higher percentage of grade 3 tumors than those who had optimal debulkings. Multivariate analysis demonstrated the size of residual diseases as the single most important independent variable influencing survival across the three study groups. Table 2 shows the recurrence patterns and duration of response to salvage chemotherapy across the three study groups. There was no statistical significant difference in recurrence rate between groups 1 and 2 (42% vs 46%) with significantly more patients in group 3 suffering disease relapses. Survival curves of groups 1 and 2 are also statistically similar (P Å 0.6788). Five-year survival was 62% in group 1, 56% in group 2, and 15% in group 3 (Fig. 1). All surviving patients showed no clinical evidence of disease with normal CA-125 levels at 5 years. Further survival analysis within group 2 showed a significant difference favoring stage 2 patients over stage 3 patients even though all were left with microscopic residuals (Fig. 2). It is likely that some patients with true stage 1 were included in the stage 2 group, accounting for the favorable
goas
AP: GYN
210
LE ET AL.
TABLE 1 Patient and Tumor Characteristics in Staged and Debulked Study Groups Staging laparotomies (N Å 24) Mean age Histologies Serous Mucinous Endometrioids Clear cells Anaplastic Grade 1 2 3 Stage distribution in each group 2 3
Debulking microscopic res (N Å 81)
Debulking ú2-cm res. (N Å 191)
56.6
58.8
61.8
66.7% 8% 12.5% 4.8% 8%
42% 8.4% 26% 15% 8.6%
64.4% 2.6% 9% 5% 19%
9.5% 52.4% 38.1%
20% 36% 44%
2.2% 16% 81.8%
55.6% (N Å 45) a 44.4% (N Å 36)
37.5% (N Å 9) 62.5% (N Å 15)
0.5% (N Å 1) 99.5% (N Å 190)
a Stage 2 patients with complete resection of macroscopic disease (TAH / BSO / omentectomy) with no attempt made no document extra pelvic spread by multiple biopsies or retroperitoneal node biopsies.
difference in survival. After excluding stage 2 patients from all study groups, however, similar survival was still demonstrable between groups 1 and 2 (Fig. 3). Because of the apparent high percentages of clear cell and anaplastic histologies in groups 2 and 3, respectively, further analysis was carried out with respect to these subtypes. Fiveyear survival estimates were 71% for clear-cell patients compared to 53% for the rest of group 2 patients. In group 3, 5-year survival was 8% for anaplastic types compared to 16% for the other types combined. Interestingly, the patterns of recurrence were similar for groups 2 and 3 with 92% of relapses occurring intra-abdominally and 8% detected distantly. In contrast, 30% of group 1 recurrences were in a distant location. Significantly, more patients in group 1 than in group 2 or 3 had documented recurrences more than 5 years after diagnosis (10% vs 1%). Once disease recurrence was documented, there was no
significant difference in duration of responses to further salvage therapies across the study groups. DISCUSSION
About three-quarters of all patients with epithelial ovarian carcinoma have intra-abdominal metastasis. Primary cytoreductive surgery followed by combination cytotoxic chemotherapy has become the standard of care in these patients. Numerous reports have shown retrospectively a favorable impact of small residual tumor volume on the rates of complete clinical responses, negative second-look laparotomies, progression-free interval, and overall survival [2–5]. Various definitions of ‘‘optimal residual disease’’ exist but it is now generally accepted to be the presence of no tumor nodules that are more than 1 cm in diameter after surgery. Using this definition, about one-third of patients with advanced
TABLE 2 Recurrence Patterns and Responses to Salvage Therapy
Recurrences Intra-abdominal Distant Time to recurence õ2 years 2–5 years ú5 years Response to salvage therapy õ6 months 6 months–1 years ú1 year
AID
Gyn 4839
/
Staging laparotomies (N Å 24)
Debulking micro. residual (N Å 81)
Debulking ú2-cm residual (N Å 191)
42% (N Å 10) 70% (N Å 7) 30% (N Å 3)
46% (N Å 37) 92% (N Å 34) 8% (N Å 3)
86% (N Å 164) 92% (N Å 151) 8% (N Å 13)
90% 0% 10%
73% 27% 0%
96% 3% 1%
30% 30% 40%
36% 18% 46%
40% 31% 29%
6d24$$$142
10-20-97 01:02:04
goas
AP: GYN
DEBULKING SURGERY IN OVARIAN CARCINOMA
FIG. 1.
Survival of stage 2/3 patients with different residual diseases treated with adjuvant platinum-based chemotherapy. N Å 296 patients.
epithelial ovarian carcinoma could be debulked to optimal status [9]. The theoretical benefits of tumor reduction surgery include improvement in patient’s sense of well being; stimulating remaining tumor cells into active division, enhancing susceptibility to chemotherapy cell kills; improving delivery
FIG. 2.
AID
211
of chemotherapy by removing necrotic tumor masses; and decreasing the chance of dealing with chemoresistant cells. In 1979, Goldie and Coldman derived a mathematical model for relating the drug sensitivities of tumors to their spontaneous mutation rate [13]. The mathematical model predicts
Survival of stage 2 vs stage 3 patients with tumor reduced to microscopic residual disease treated with adjuvant platinum-based chemotherapy.
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
goas
AP: GYN
212
LE ET AL.
FIG. 3.
Survival of stage 3 patients with different residual diseases treated with adjuvant platinum-based chemotherapy. N Å 241 patients.
that the curability of a tumor (or the probability of finding no drug-resistant cells) decreases precipitously as the tumor approaches a critical size as determined by its intrinsic rate of spontaneous mutation. The clinical implications are that smaller tumors are easier to cure with chemotherapy and the best moment for effective treatment is as soon after detection as possible [14]. No randomized trial, however, had been carried out to confirm the theoretical benefit of tumor reduction. Questions have been raised as to whether patients who are optimally debulked have a less aggressive tumor biology that allows such a procedure or whether those patients would have done just as well if treated initially with neoadjuvant chemotherapy followed by surgical debulkings [18]. In this paper, we compare the outcomes of similarly treated patient groups with different initial disease volumes and residual status. Patients with initial microscopic disease presumably have the most favorable tumor characteristics and prognosis when compared to those with macroscopic disease, as predicted by the Goldie–Coldman hypothesis. In our study population, older age, serous/anaplastic histologies, and high-grade tumors acted as high-risk biologic markers in predicting suboptimal debulking, and residual disease more than 2 cm predicts significantly compromised survival (group 3). This is consistent with a recent report by Makar et al. [10]. Patient groups treated with staging laparotomies and optimal debulkings to microscopic residuals are similar in terms of stage, grade, and histologic subtype distributions. Survival analysis did not show a difference in patients
AID
Gyn 4839
/
6d24$$$142
10-20-97 01:02:04
presenting with initial microscopic disease compared to those with presumed poorer prognosis tumor with initial large macroscopic disease tumor reduced down to microscopic residuals (Fig. 1, P Å 0.6788). As expected, patients left with more than 2-cm residuals had significantly poorer survival than the other two groups (P õ 0.0001). It is possible that tumor reduction surgery to microscopic residuals reduces the number of chemotherapy-resistant clonogenic malignant cells to improve the prognosis in this subgroup of patients as predicted by the Goldie–Coldman hypothesis [13, 14]. Microscopic residual status is an important prognostic factor. Recent analysis at our institution showed similar survivals in patients with stage 3 epithelial ovarian carcinoma with õ2-cm and ú2-cm residuals. A significantly prolonged survival occurred only in those with microscopic residuals (Fig. 4). Similarities in locations and times of recurrences support the initial assumption of equal biologic behavior in the optimally debulked and suboptimally debulked patients. In contrast, many patients with initial microscopic disease tend to have distant and late (ú5 years) recurrences. Similar responses to salvage therapies after failure of primary chemotherapy were observed across study groups likely because of the emergence and presence of chemoresistant clones. The effect of tumor biology has recently been suggested as a prognostic factor. Farias-Eisner et al. [11] reported on 78 patients with stage 3 disease with minimal (õ5 mm) residual disease. Tumor grade, distribution of residual disease, and extent of carcinomatosis were found to indepen-
goas
AP: GYN
DEBULKING SURGERY IN OVARIAN CARCINOMA
Stage 3 patients treated with platinum-based chemotherapy as per residual disease status. N Å 330 patients.
FIG. 4.
dently affect survival despite optimal debulking surgery in all patients. Hoskin et al. [6] reported a significant difference in survival for ‘‘optimal’’ stage 3 epithelial ovarian cancer treated with platinum and cyclophosphamide with or without doxorubicin, favoring those patients with initial abdominal disease of 1 cm or less over those cytoreduced to 1 cm or less. The number of residual disease nodules (ú20) was also an adverse prognostic factor. It has been proposed that patients with large amount of disease at exploration may have had their disease for a longer time or may have a more aggressive biology conferring a poorer survival. The authors questioned the role of debulking surgery in improving prognosis in this patient population. As acknowledged by the authors, however, since all patients had macroscopic initial disease and all were treated initially with tumor reduction, the benefit of surgery could never be fully assessed. Despite being a retrospective study, which may contain some confounding variables and biases, our study confirms the benefits of cytoreductive surgery in improving survival in patients with macroscopic disease completely resected. The results from our analysis, as well as those from the recent EORCT phase 3 trial of interventional surgical debulking in patients with epithelial ovarian carcinoma [12], support the continuing use of aggressive tumor debulking procedures in the management of patients with advanced stage ovarian carcinoma. A definite randomized controlled trial studying cytoreductive surgery followed by chemotherapy versus neoadjuvant chemotherapy followed by tumor
AID
Gyn 4839
213
/
6d24$$$143
10-20-97 01:02:04
debulkings is necessary to prove the benefits of aggressive tumor resection. REFERENCES 1. American Cancer Society: Cancer facts and figures—1996. American Cancer Society, 1996 2. Griffiths CT: Surgical resection of tumor bulk in the primary treatment of ovarian carcinoma. Natl Cancer Inst Monogr 42:101–104, 1975 3. Hacker NF, Berek JS, Lagasse LD, Niebrig RK, Elashoff RM: Primary cytoreductive surgery for epithelial ovarian cancer. Obstet Gynecol 61:413–420, 1983 4. Delgado G, Oram DH, Petrilli ES: Stage 3 ovarian cancer: The role of maximal surgical cytoreduction. Gynecol Oncol 18:293, 1984 5. Allen DG, Heintz AP, Touw FW: A meta-analysis of residual disease and survival in stage 3 and 4 carcinoma of the ovary. Eur J Gynecol Oncol 16(5):349–356, 1995 6. Hoskin WJ, Bundy BN, Thigpen TJ, Omura GA: The influence of cytoreductive surgery on recurrence-free interval and survival in small volume stage 3 epithelial ovarian cancer: A Gynecologic Oncology Group study. Gynecol Oncol 47:159–166, 1992 7. Kaplan EL, Meier P: Non parametric estimation from incomplete observation. J Am Stat Assoc 53:457–481, 1958 8. Mantel N: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 50:163–170, 1966 9. Rubin SC: Surgery for ovarian cancer. Hematol Oncol Clin North Am 6(4):851–865, 1992 10. Makar AP, Baekelandt M, Trope CG, Kristensen GB: The prognostic significance of residual disease, FIGO substage, tumour histology, and grade in patients with FIGO stage 3 ovarian cancer. Gynecol Oncol 56:175–180, 1995
goas
AP: GYN
214
LE ET AL.
11. Farias-Eisner R, Teng F, Oliveira M, Leuchter R, Karlan B, Lagasse LD, Berek JS: The influence of tumor grade, distribution, and extent of carcinomatosis in minimal residual stage 3 epithelial ovarian cancer after optimal primary cytoreductive surgery. Gynecol Oncol 55:108– 110, 1994 12. Van der Berg MEJ, Van Lent M, et al.: The effect of debulking surgery after induction chemotherapy on the prognosis in advanced epithelial ovarian cancer. N Engl J Med 332:629, 1995 13. Goldie JH, Coldman AJ: A mathematical model for relating the drug sensitivity of tumours to their spontaneous mutation rate. Cancer Treat Rep 63:1727–1733, 1979 14. Goldie JH, Coldman AJ: The genetic origin of drug resistance in neoplasms: Implications for systemic therapy. Cancer Res 44:3643–3653, 1984
AID
Gyn 4839
/
6d24$$$143
10-20-97 01:02:04
15. Chambers JT, Chambers SK, Voynick IM, Schwartz PE: Neoadjuvant chemotherapy in stage X ovarian carcinoma. Gynecol Oncol 37:327– 331, 1990 16. Schwartz PE, Chambers JT, Makuch R: Neoadjuvant chemotherapy for advanced ovarian cancer. Gynecol Oncol 53:33–37, 1994 17. Surwitt E, Childers J, Atlas I, Nour M, Hatch K, Hallum A, Alberts D: Neoadjuvant chemotherapy for advanced ovarian cancer. Int J Gynecol Cancer 6:356–361, 1996 18. Schartz PE, Chambers JT, Kohorn EI, Rutherford TJ, Schubert W, Zelterman D: Neoadjuvant chemotherapy for advanced ovarian cancer: Long term survival. Proceedings of the Society of Pelvic Surgeons Annual Meeting, London, September 16, 1996, p 31
goas
AP: GYN