You are on page 1of 10


Secondary Cytoreductive Surgery for Patients with Relapsed Epithelial Ovarian Carcinoma: Who Benefits?
Rong-Yu Zang, M.D., Ph.D.1 Zi-Ting Li, M.D.1 Jie Tang, M.D.1 Xi Cheng, M.D.1 Shu-Mo Cai, M.D.1 Zhi-Yi Zhang, M.D.1 Nelson N. Teng, M.D., Ph.D.2

BACKGROUND. This study was performed to address patient selection criteria and
the role of secondary cytoreductive surgery (SCR) in patients with epithelial ovarian carcinoma (EOC) who had relapsed tumors after a progression-free interval 3 months. METHODS. One hundred seventeen patients with relapsed EOC after a clinical complete remission duration 3 months who underwent SCR were entered on this prospective trial. Survival curves were generated using the Kaplan–Meier method, and statistical comparisons were performed using log-rank tests, logistic stepwise regression analyses, and a Cox stepwise regression model.

Department of Gynecologic Oncology, Cancer Hospital of Fudan University (formerly Shanghai Medical University), Shanghai, China. Department of Obstetrics and Gynecology, Stanford University Medical Center, Stanford, California.


Supported in part by the New Star Project (Rong-Yu Zang) and by a grant from the Shanghai Health Bureau 99408 (Zi-Ting Li). The authors thank Yan Xing, Ph.D., at the M. D. Anderson Cancer Center for statistical assistance. Address for reprints: Rong-Yu Zang, M.D., Ph.D., Department of Gynecologic Oncology, Cancer Hospital of Fudan University, 399 Ling-Ling Road, Shanghai 200032, China; Fax: (011) 86 2164174774. Received September 16, 2003; revision received January 5, 2004; accepted January 6, 2004. © 2004 American Cancer Society DOI 10.1002/cncr.20106

RESULTS. The median patient age at the time of relapse was 53 years (range, 20 –78 years). The median survival was 22 months and the estimated 5-year survival rate for the entire cohort was 17.2%. Tumor was confined to a solitary site in 33 patients and to 2 sites in 84 patients. After they underwent SCR, 11 patients were rendered macroscopically disease free, 61 patients had residual disease that measured 1 cm in greatest dimension, and 45 patients had bulky intraabdominal residual disease. Survival was influenced by the extent of relapse disease (solitary site vs. multiple sites; P 0.0001), the size of residual disease after SCR (0 cm vs. 1 cm [P 0.1211], 1 cm vs. 1 cm [P 0.0002], and 0 cm vs. 1 cm [P 0.0011]), Eastern Cooperative Oncology Group performance status (0 vs. 1 [P 0.134], 1 vs. 2 [P 0.007], and 0 vs. 2 [P 0.0012]), and the number of cycles of salvage chemotherapy (1–2 cycles vs. 3–5 cycles [P 0.0144]; 1–2 cycles vs. 6 cycles [P 0.0001]; and 3–5 cycles vs. 6 cycles [P 0.0009]). The outcome of SCR was influenced by the extent of relapse disease (multiple sites [51.2%] vs. solitary sites [87.9%]; relative risk [RR] 9.1237; P 0.0002) and by the use of bowel resection (yes [60.9%] vs. no [37.5%]; RR 0.3828; P 0.0106). CONCLUSIONS. SCR was found to be safe for patients with relapsed EOC who achieved a clinical complete remission that lasted 3 months, with resectability similar to that of primary debulking surgery. Optimal surgical outcomes were achieved easily in patients who apparently had solitary tumor sites, with bowel resection making it possible to remove bulky tumors that involved the intestine. A survival benefit was provided by optimal SCR, particularly when surgery was supported by multiple courses of salvage chemotherapy. Cancer 2004;100: 1152– 61. © 2004 American Cancer Society. KEYWORDS: ovarian carcinoma, relapse, surgery, secondary cytoreductive surgery.


varian carcinoma represents 25% of all malignancies of the female genital tract but is the most common cause of death among women who develop gynecologic malignancies. The majority (nearly 75%) of patients with epithelial ovarian carcinoma (EOC) have advanced-stage disease at the time of diagnosis. The 5-year survival rate for patients with all stages of EOC is 53% but falls to 20 –31% for patients who present with advanced disease.1–3 Approximately 48 –

The authors of that study believed that consistently referring patients with apparent advanced EOC to expert centers for primary surgery may be the best means currently available for improving overall survival. 6) patient consent to undergo secondary surgery. maximal cytoreduction. A partial response required a reduction in tumor burden 50% that lasted 4 weeks and no appearance of new lesions. A responder was defined as any patient who exhibited a complete or partial response. and the SCR approach is restricted to a few large institutions. a widely quoted report from the M. rather than the dose intensity of platinum compound administered. Recent observations also indicate that SCR is feasible. Other criteria for inclusion were 1) a rising CA 125 level and/or radiographic or physical findings suggestive of relapse. 3) retrospective reviews are not convincing enough to advocate SCR as a standard approach for the treatment of patients with recurrent EOC. a meta-analysis summarizing data from 1989 to 1998 revealed that. for patients who underwent complete SCR.9 However. 2) evaluation of disease status within 2 weeks before enrolling onto the trial.SCR and Relapsed EOC/Zang et al. response was defined as the disappearance of all macroscopic disease for at least 4 weeks. and numerous issues remain: 1) Patients with recurrent EOC are highly selected for secondary surgery and.10 and other investigators11.8 Six years later.3 The unfavorable prognosis and diminished survival of patients who do not achieve complete remission with induction chemotherapy or who develop recurrent tumors after achieving a prior response and controversy regarding the management of recurrent EOC have led to the application of numerous active agents in the second-line setting. to date. D. 7) an Eastern Cooperative Oncology Group (ECOG) performance status 2. Those authors stated that. a responder was defined as a patient without any tumor for at least 8 weeks. 5) a surgical objective to remove all visible lesions in the pelvis and/or the abdomen. as distinguished from the conventional concept of recurrence. which was defined as tumor recurrence after a PFI 6 months after completing primary therapy. 2) the SCR surgical technique is more complicated compared with primary surgery. Even so. salvage chemotherapy is conducted easily with fewer restrictions on the physician. This clinical setting was categorized as relapse. Berek et al. and 8) the absence of medical contradictions to an extensive surgical procedure.4 – 6 Despite the introduction of new chemotherapeutic agents and the development of novel combinations. 1) which patients are suitable candidates for secondary cytoreductive surgery. and it is impossible to conduct such a trial. 1153 50% of women with EOC ultimately will develop a recurrent tumor and will require further treatment. For patients who underwent optimal cytoreduction. 3) an estimated survival 3 months.7 The situation is not as encouraging for patients who require secondary cytoreductive surgery (SCR). 4) the absence of hepatic parenchyma metastasis. and associated with significant prolongation of survival in selected patients with recurrent EOC. and 4) conversely. however. we conducted this prospective trial to address the following questions: When they develop recurrent EOC. that opinion is not unanimous.3 Randomized investigations examining the role of primary cytoreductive surgery are lacking. Anderson Cancer Center on a review of patients who had recurrent disease did not find any significant survival benefit from SCR. Definitions Response to salvage chemotherapy after SCR was assessed before each cycle for disease measurable on physical examination and as appropriate for tumors measured on medical imaging. Chest X-rays were repeated as necessary to monitor tumor response. and 2) which patients may benefit most from SCR? MATERIALS AND METHODS Eligibility Criteria Patients with EOC in this prospective trial had undergone primary cytoreductive surgery followed by platinum-based chemotherapy. the value of secondary tumor-reductive surgery for recurrent EOC was limited. All measurable disease sites were assessed us- .12 reported that a survival benefit was provided by successful secondary debulking surgery. inspired by the encouraging results available and our experience in a retrospective study. and no gold standard trials have been performed to date. which was defined as tumor recurrence after a progression-free interval (PFI) 3 months. a review of clinical outcomes in patients with recurrent EOC concluded that there has been little change in response duration and survival during the past 2 decades. Janicke et al. there are no standard selection criteria. were the first investigators to demonstrate that patients with recurrent EOC benefited from optimal SCR. was one of the most powerful determinants of cohort survival among patients with International Federation of Gynecology and Obstetrics (FIGO) Stage III or Stage IV EOC. A complete remission was defined as the disappearance of all measurable and assessable disease for at least 4 weeks. well tolerated. In 1983. during the platinum era. Strategies designed to improve these results continue to be investigated. more recently. However.13–17 However. in the absence of efficacious second-line medical therapy. They had established relapse.

All patients received either intraperitoneal or intravenous salvage chemotherapy. 14 patients (12%) were treated with topotecan. may affect the survival.18 chemotherapy. The median age at the time patients underwent secondary surgery was 53 years (range. 4 patients (3. all 79 patients were evaluated for response: Seven patients (8. Forty-six women (39. Compared with platinum-based. and performance status after surgery were examined using logistic stepwise regression analysis. The distributions of preoperative and postoperative factors in this cohort of patients are shown in Table 1. Serum CA 125 levels were determined every week after relapse was confirmed. Patients had mechanical and antibiotic bowel preparation prior to surgery whenever possible.0.8%) were treated with etoposide. Data were analyzed using SPSS soft package for Windows (version 9. Survival was measured from the day of either secondary surgery or chemotherapy before secondary surgery. The results of the Cox and logistic model analyses were reported with relative risks (RR) and 95% confidence intervals. 2004 / Volume 100 / Number 6 ing the same techniques that were used to measure disease sites at baseline. Factors that have an impact on the outcome of SCR. but surgical cytoreduction was discontinued if disease was encountered that could not be removed with existing techniques. Some patients with localized recurrences also received external beam radiation therapy. surgical outcome. Regarding the efficiency of second-line chemotherapy. 6 patients (5.1%) were treated with ifosfamide.3. second-line chemotherapy was defined as chemotherapy that included one of the following drugs or their combinations: topotecan. response to prior chemotherapy.5%) received 1 combinations of intravenous second-line salvage chemotherapy. 20 –78 years). Survival curves were generated using the Kaplan–Meier method and were compared using the log-rank test. and 73 patients (62. The resectability of relapsed lesions was assessed before surgery by two senior gynecologic surgeons. Cytoreductive surgery was defined as optimal if the greatest dimension of the largest residual tumor measured 1 cm and suboptimal if it measured 1 cm. The median follow-up was 16 months (range. After they recovered from surgery. or hexamethylmelamine. surgical findings. P 0. which was conducted at Cancer Hospital of Fudan University.4%) received 1–7 cycles of platinum-based intraperitoneal chemotherapy. etoposide. and 2 patients (1.05 was considered significant in these analyses. Platinum-based salvage chemotherapy was administered in 5 patients who were sensitive to platinum-based chemotherapy in primary therapy. PFI.. Statistical Considerations One hundred twenty-three eligible patients were targeted over a 4-year enrollment period with an additional 1-year of follow-up before the final analysis.3%) were treated with platin-docetaxel therapy. IL). 7 patients (6. 5– 84 months). when a Cox stepwise regression model is established. therefore.9%) achieved a clinical complete response. ifosfamide.4%) were treated with HCPT. and 49 patients (62%) achieved a partial clinical response. which was repeated every week. 15 patients (12. Other than second-line agents.4%) were treated with NVB.8%) were treated with paclitaxel. 43 patients (36. other treatment approaches. Chicago. In all. respectively.4 months. Survival was compared using a time-toevent methodology. Among 11 patients who underwent complete SCR. Therefore. patients received individualized salvage therapy based on initial treatment.7%) were treated with hexamethylmelamine.5%) were treated with platinum-based salvage RESULTS Patient Characteristics This trial. and anticipated ability to tolerate the toxicity of salvage chemotherapy. For economic reasons. opened in January 1998 and was closed to patient accrual in December 2001. docetaxel. 4 patients (3. This prospective study was initiated to determine the ability to surgically eliminate all visible disease in patients with relapsed EOC and the associated impact on survival. every variant that may have an effect on survival should be considered and entered into multivariate analysis. 5 patients (4.15. . hydroxycamptothecine (HCPT). 6 patients were treated with second-line salvage chemotherapy. The median PFI was 15. Life tables and the Kaplan–Meier method were used to obtain estimates of survival and median survival. in China. Seventy-nine patients (67. vinorelbine (NVB).1154 CANCER March 15.0%) were treated with gemcitabine. 117 patients were entered. first-line chemotherapy.3%) received some salvage chemotherapy before undergoing secondary surgery. 38 patients (32. Apart from SCR. paclitaxel cannot be used in the primary setting. Follow-up was available through December 2002. Treatment Approaches Chest X-rays and ultrasonic or computed tomography scans or magnetic resonance images of the pelvis and abdomen were obtained to evaluate disease status. it also is considered a second-line agent. gemcitabine. Among them. such as salvage chemotherapy and radiotherapy. SPSS Inc. the efficacy of second-line chemotherapy.

of patients No.0001 —b 26.003 25.0027 14.0 23.0 NS 26.0 22.0 16.5 0. b The median survival was not reached. SCR: secondary cytoreductive surgery.5 25.5 25.0 25.0 0.0 0. .0 26.5 NS 26.0 20.0 0.5 15.0 14.0 21.0 40.0181 NS NS NS FIGO: International Federation of Gynecology and Obstetrics.0 27.0 0.5 NS 21.0 16.5 20.0165 19.5 15.5 25.0 26.5 14.0 0.001 40.0001 —b 16.SCR and Relapsed EOC/Zang et al. PFI: progression-free interval.0 0. TABLE 1 Patient Characteristics Characteristic Age (yrs) 65 65 FIGO stage (1989) Stage I Stage II Stage III Stage IV Histology Serous Mucinous Adenocarcinoma Endometrioid Clear cell Mixed Other Grade Grade 1 Grade 2 Grade 3 Residual disease after primary surgery 1 cm 1 cm PFI (mos) 3–12 13–23 24 Recurrent ascites No Yes Extent of relapsed disease Solitary Multiple Chemotherapy cycles before SCR 0 2–8 ECOG performance status 0 1 2 Bowel resection or colostomy No Yes Residual disease after SCR 0 cm 1 cm 1 cm Second-line chemotherapy No Yes Efficacious second-line chemotherapy No/not treated Yes Cycles of salvage chemotherapy after SCR 1–2 3–5 6 Pelvic radiation No Yes Total No.0382 18.0 NS 7.5 22. of deaths Median survival (mos) 1155 P valuea NS 107 10 13 31 66 7 61 6 11 21 10 3 5 7 65 45 78 39 48 44 25 94 23 33 84 71 46 23 62 32 8 109 11 61 45 38 79 61 56 16 52 49 85 32 117 63 9 7 15 45 5 36 5 9 9 8 1 4 2 43 27 45 27 34 27 11 52 20 11 61 45 27 11 37 24 7 65 3 33 36 25 47 43 29 14 36 22 56 16 72 21. ECOG: Eastern Cooperative Oncology Group.0 0.5 30.5 26.017 Reference 0.5 24.5 17.0001 9.0 20.0 15.0 10.0507 NS NS 0.5 NS 14.5 0. NS: no statistical significance.0 26.5 27. a Log-rank test.

22 patients (18. intraperitoneal chemotherapy after SCR.7%) who underwent small bowel resection.1156 CANCER March 15. 1. 24 months: chi-square 6. 76. third. but multivariate analysis suggested that PFI did not influence prognosis (Tables 1 and 3).1 48. and response to second-line salvage chemotherapy influenced the probability of survival (Table 1). optimal SCR.4 7.2%) with multiple lesions (P 0. salvage chemotherapy after SCR.4 62. histology. Table 3 provides a summary of the factors that influenced survival.1 4. salvage chemotherapy before SCR. Log-rank analysis revealed that the factors extent of relapse disease. Among the patients who were treated with salvage chemotherapy before undergoing SCR. 13–23 months: chi-square 0. as described previously.7%) experienced complications. Nine patients (7. second. 0. and 6 patients (5. compared with patients who had a PFI of 3–12 months or 13–23 months.15 One hundred three patients (88%) underwent bowel resection. and 17 patients (14. SCR: secondary cytoreductive surgery.0 55. tumor grade. and 45 patients (38. and 3–12 months vs. patients who had a PFI 24 months experienced longer survival (3–12 months vs. and 8 patients had a solitary focus.7%) underwent peritoneal implantation ablation. and 1 patient each (0.0002). including 64 patients (54. 87. and fourth steps. an ECOG performance status of 0 –1.7 29. At secondary surgery. 11 patients (9.1%) underwent palliative colostomy: To eliminate selection bias. Patients who were treated with radiotherapy after SCR .3676].9%) underwent partial liver resection and urinary tract resection. residual disease after primary surgery. 3-year.9 79.6 27. and greater than six cycles of salvage chemotherapy after SCR were identified in the first. and the other patients were treated medically with success.0 8.4%) underwent excision of retained omental tissue. many surgical approaches have been used. and patients with those factors experienced prolonged survival (Figs. Kaplan–Meier tests revealed that.9 8.9%).0 30. 11 patients (9.2%. residual disease after SCR.1%). 61 patients (52.8%). retroperitoneal lymphadenectomy. Sixty-four patients (54.0 0. Of the 117 patients who underwent secondary surgery.5%) who underwent large bowel resection. 3 patients (2.4%) underwent retroperitoneal lymphadenectomy.0 10.81 [P 0.4 39. 1 patient underwent a distal pancreatectomy en bloc with splenectomy. 13–23 months vs. and the estimated 5-year survival rate was 17.12 [P 0. ascites at recurrence.3 61.7%). 24 months: chi-square 4. The median blood loss was 300 mL (range. PFI. The median operative time was 150 minutes (range. such as ileus (6 patients.2 30.0 59. 25 patients (21.9%) with a solitary focus and in 43 of 84 patients (51.3 69. 1– 4).4 30.2 0. ECOG performance status. Survival The median survival for the entire cohort was 22 months.1 35. 38 patients had multiple relapsed lesions. as identified in the Cox stepwise regression analysis. radiotherapy after SCR. Solitary focus.8 39. 30 – 420 minutes).7%) underwent SCR before they were treated with any salvage chemotherapy. Table 2 summarizes the stratified life table for 2-year.8 5. No patients suffered from mortality after surgery.4 21.5%) had bulky intraabdominal residual disease.4%) were rendered macroscopically disease free.5 51. those 9 patients were entered into the survival analyses. Optimal SCR was achieved in 29 of 33 patients (9 patients with microscopic residual disease and 20 patients with macroscopic residual disease. and whether second-line salvage chemotherapy was used after SCR did not appear to influence the probability of survival. To achieve optimal surgical results. FIGO stage.0424].6%) underwent open and close procedures.0142]).6%) underwent splenectomy.9 50.1%) had residual disease that measured 1 cm in greatest dimension. 5. 20 – 1500 mL). and 5-year survival.4 61.2 ECOG: Eastern Cooperative Oncology Group. One patient with a fistula required surgery.8 16.5 41. 3 patients (2. respectively. The factors age at relapse.2%) and to 2 sites in 84 patients (71.7 46.7 12.9 10.01 [P 0. it was found that tumor was confined to a solitary site in 33 patients (28.8 19.4 49. bowel resection.2 17. 2004 / Volume 100 / Number 6 TABLE 2 Life Table of Factors that Determined Survival after Relapse Survival rate (%) Factor Extent of relapsed disease Solitary Multiple lesions Residual disease 0 cm 1 cm 1 cm ECOG performance status before surgery 0 1 2 Salvage chemotherapy cycles after SCR 1–2 3–5 6 Total 2 yrs 3 yrs 5 yrs and 71 women (60. and fistula (1 patient.8%) who underwent modified posterior pelvic exenteration. ascites before primary surgery. Among the patients with unresectable disease. cutaneous wound infection (2 patients.

72. DISCUSSION SCR: the Best Choice for Patients with A Solitary Relapse With regard to SCR. the median survivals of patients who were treated with efficacious second-line chemotherapy and patients who received either nonefficacious or no second-line chemotherapy were 38.0 months. Among the patients who underwent optimal SCR.802 1157 Upper 6. 1 cm: chi-square 10.1707 2.5176 0.3579 0. and sur- gery for progressive disease compared with the issues raised by SCR for recurrent disease.18 Issues for debate are fewer in the settings of interval cytoreductive surgery. SCR: secondary cytoreductive surgery.0001 Relative risk 3.0001). More recently.0365 1. D.14 stated that complete cytoreduction was possible for the majority of patients with recurrent EOC and maximized survival if it was undertaken before salvage chemotherapy.189 0.5482 0.0011]).5934 2. Survival according to the size of residual disease after patients underwent secondary cytoreductive surgery for epithelial ovarian carcinoma (0 cm vs.75. 1 cm: chi-square 13.SCR and Relapsed EOC/Zang et al. SCR was found to be optimal in 18 of 25 patients (72%).6396 1.4 months for patients who had no visible residual disease after SCR compared with 19. The . debulking surgery at second-look laparotomy.1914 0. crosses: residual disease measuring 1 cm in greatest dimension. P 0.9291 SE: standard error.0042 0. TABLE 3 Factors that Affected Survival after Relapse: Determined by Cox Stepwise Regression Model 95% CI Factor Extent of relapsed disease Residual disease ECOG performance status Salvage chemotherapy post-SCR Beta coefficient 1. Triangles: no residual disease. ECOG: Eastern Cooperative Oncology Group.0018 0.0375 Wald chi-square 9. FIGURE 2.2061 15.234 0. FIGURE 1.0346 0. Anderson Cancer Center indicated that SCR for patients with recurrent ovarian carcinoma at an apparently solitary intraabdominal site resulted in optimal residual tumor in a high proportion of patients.1211].0856). The median survival in that study was 44.007). Further analyses were performed to identify the correlation between optimal SCR and efficacious second-line salvage chemotherapy.1183 0. 1 cm: chi-square 2. 95% CI: 95% confidence interval. 1 cm vs.1471 SE 0.0002]. solid line: patients with solitary lesions. and 0 cm vs.8632 Lower 1. Survival of patients with relapsed epithelial ovarian carcinoma (chi-square 17. Dashed line: patients with multiple lesions.4 [P 0. P 0. A large series of 106 patients reported by Eisenkop et al. an experience from the M.0598 1. plus signs: residual disease measuring 1 cm in greatest dimension. because it remains uncertain which patients with recurrent EOC are suitable for such surgery.4944 0.7 months and 20. In that study. although further analysis did not support the finding that this encouraging treatment approach was an independent determinant of prognosis.3 months for patients who had any macroscopic residual disease (P 0.5172 1. patients with EOC are grouped into four clinical settings.0165).4659 8. respectively (closely approximating statistical significance: chi-square 2.3665 P value 0.95. experienced lengthened survival (chi-square 5.55 [P 0.7641 4.59 [P 0. P 0.7302 0.

Seventy-two of 117 patients (61.2% in patients with multiple lesions. ECOG 2: chi-square 7. ECOG 1: chi-square 2.98 [P 0. but they will respond less to salvage chemotherapy compared with patients who have a PFI 6 months after completing front-line chemotherapy. because those patients are platinum nonresponders who remain a treatment dilemma. 24. survival assessments confirmed a statistically significant advantage for patients who had a solitary site of relapse.007].25 [P 0.0144]. However.57 [P 0.0012]). however.17.8% compared with 5. 3–5 cycles of chemotherapy. and 15 months. [P 0. patients with solitary lesions achieved optimal surgical outcomes more easily.76 [P 0. plus signs: 1–2 cycles. that is. with a 5-year survival rate of 49.0001). and should not be classified with platinum resistance. with a median survival of 40. 2004 / Volume 100 / Number 6 FIGURE 3. or 6 cycles of chemotherapy were 9. In particular. 6 cycles: chi-square 30. and there were significant difference between them (1–2 cycles vs.5%) who underwent optimal cytoreduction achieved a median survival 26. median survival was 56.0001).4% in patients who had multiple sites of relapse (P 0. Survival by chemotherapy after patients underwent secondary cytoreductive surgery for epithelial ovarian carcinoma. 1–2 cycles vs. patients with residual disease measuring 1 cm who achieve complete remission that lasts from 3 months to 6 months are for the most part platinum responders.9% in patients with a solitary focus and 51. there was a statistically significant. Triangles: 6 cycles. it is our understanding that they are quite different from patients who experience no remission at all during front-line chemotherapy. the resectability rate was 87. and 3–5 cycles vs.5 months. To our surprise.19 Little progress has been made in salvage second-line chemotherapy for these patients. crosses: 3–5 cycles. patients with a remission duration of 3– 6 months who developed relapsed disease also were included in this trial. 15.0 months. ECOG 2: chi-square 10. and 30. No investigators to date have advocated the use of SCR in the treatment of patients who have a PFI between 3 months and 6 months.98. respectively (ECOG 0 vs. To maximize the population who may benefit from SCR (in addition to patients with a PFI 6 months after completing first-line chemotherapy). 6 cycles: chi-square 10. when this factor was considered in the Cox stepwise regression model together with other variants that may influence survival.5 months in patients who underwent suboptimal surgery (P 0.16 The eligible patients were too few to reach a level of statistically significant difference. but a latent survival benefit may exist.0009]). This correlation remained significant after controlling for all other variables. positive correlation between the percentage of optimal SCR and median survival. respectively. 1 (crosses). This compelling evidence indicates that patients with who had a PFI of 3– 6 months and a PFI 6 months may be treated safely with surgery and may have a survival advantage with optimal SCR. The median survival of patients who received 1–2 cycles of chemotherapy. Survival by Eastern Cooperative Oncology Group (ECOG) performance status before patients underwent secondary cytoreductive surgery for epithelial ovarian carcinoma. In the current report.5 months.0001]. and 2 (plus signs).28. the extent of relapsed disease presented at the first step as the strongest determinant of survival. and ECOG 0 vs.9 months for patients who underwent optimal SCR and 25. [P 0.1158 CANCER March 15. whereas PFI was not found to be an independent determinant of survival (Table 3).0 months. 3–5 cycles: chi-square 5. FIGURE 4.134]. ECOG 1 vs.14.5 months compared with 14. There was no marked difference in survival between patients with PFI of 3–12 months and patients with a PFI of 13–23 months. There was a significant difference between patients who had an ECOG performance status of 0 (triangles). Compared with earlier . which may indicate that SCR is the best approach currently available for patients who have a solitary site of relapse. Furthermore.5 months.1 months for patients who underwent suboptimal SCR.

although there was no positive correlation between bowel resection and survival. On condition that the retroperitoneal cavity is not opened at primary surgery. Patients with relapsed EOC frequently respond to second-line treatments. SCR can be accomplished with significant but acceptable morbidity.10. examination and removal of multiple histologic specimens. such as bowel resection.5229 7.8. we identified other factors that influenced the outcome of patients who underwent SCR. distal pancreatectomy en bloc with splenectomy.9602 3. are applied in secondary surgery.8603 P value 0. relapsed EOC continues to be a therapeutic dilemma. once tumors recur. with a retroperitoneal approach used for pelvic peritoneum and rectosigmoid colon. bowel resection is an ideal approach to resect all bulky tumors that involve the intestine (RR 0.9%. studies. even if bulky retroperitoneal lymph nodes may adhere to critical retroperitoneal structures. 95% CI: 95% confidence interval. but the favorable results strongly indicate that curative bowel resection may be more efficacious compared with palliation in the treatment of patients with recurrent EOC.8333 0. It is interesting to note that another study at the Memorial Sloan-Kettering Cancer Center showed that 71% of patients with tumor-induced bowel obstruction who underwent palliative surgery had successful palliation (the ability to tolerate a regular or low-residue diet at least 60 days postoperatively). as distinguished the from conventional concept of a recurrence.5967 0. but the rate reported in the literature is approximately 1. and gynecologic surgery.14 although the latter rate lacks credibility. the technical success rates of SCR vary widely. a large volume of ascites with a prohibitive number of intestinal serosal implants resulting in small-volume residual miliary disease. from 37– 47%8.3828) (Table 4). The variant multiple relapse lesions was a very unfavorable factor (RR 9.3828 — Lower 2.20 Those observations did not address tumor resectability.SCR and Relapsed EOC/Zang et al.0051 Relative risk 9. Virtually all pelvic pathology is removable by en bloc resection of retained reproductive organs. urologic. The mortality risk is negligible. We believe these findings will greatly encourage gynecologic surgeons to train and improve their bowel resection techniques. In published series. to our knowledge. .3797 0. no other trial to date has advocated considering these patients for secondary surgery. visualization of all peritoneal surface. and 79% of patients were able to tolerate salvage chemotherapy. and the whole length of the small bowel en bloc resulting in tumor-wrapped intestine most often precluded any meaningful cytoreductive effort. Conversely. TABLE 4 Clinical Variants that Influenced the Outcome of Secondary Surgical Cytoreduction 95% CI Factor Extent of relapsed diseasea Bowel resection Constant Beta coefficient 2. Bulky disease that involved the mesentery and underlying structures. In addition. Surgical Techniques of SCR and Outcome Determinants Secondary tumor resection is possible technically in a significant proportion of patients who have tumors that are eradicated by primary surgery and first-line chemotherapy.21 Therefore. The techniques used in abdominal. partial liver resection. modified posterior pelvic exenteration.1237). Issues for Debate In this era of advanced medical technology.2109 0.1691 SE 0. it is of paramount importance for all clinicians to recognize that the primary objective of front-line therapy is to maximize the PFI.1237 0. and intraperitoneal or retroperitoneal tumor resection.3759 1.10. and urinary tract resection. but the impact on survival remains uncertain. they are removable with meticulous dissection because they rarely invade completely into a vessel wall. splenectomy.7% of patients experienced some degree of morbidity.0002 0. systematic retroperitoneal lymphadenectomy. a Relapse was defined as tumor recurrence after a progression-free interval (PFI) 3 months. colostomy. Preoperative imaging studies did not appear to predict unresectable disease.14 No postoperative mortality occurred.0106 0. with a PFI 6 months after the completion of primary therapy.12 to 83%. The SCR procedures include enterolysis.7999 — SE: standard error.14 In the current study.7303 6. With currently available surgical techniques.1304 Wald chi-square 13. which was lower compared with 24 – 63% of patients reported in the literature.1832 — 1159 Upper 29. the primary objective of salvage therapy is to maximize survival and quality of life. We determined that only 7.

Zhang ZY. Therefore. second-line chemotherapy. in selected patients. 1989. yet unproven. Secondary cytoreduction for recurrent ovarian cancer.85:278 –284. The second step is the development of more effective combination chemotherapy regimens that not only increase the overall response rate but also lead to an increased duration of response (remission inducement).20:2365–2369. but they converged after 3 years (data not shown).57: 61– 65. However. J Clin Oncol.53:5–26. 1997:1427–1539. Survival of patients following secondary cytoreductive surgery in ovarian cancer. and peritoneal carcinoma. CA Cancer J Clin. et al.23 It is increasingly apparent that a secondary surgical approach can be used safely and efficaciously and confers a survival advantage. Elashoff RM. Eisenkop SM. 1992. Zang RY. 16. Major strategies that include the following four steps are being tested in current clinical trials in an effort to improve the survival of patients with advanced EOC. Cancer. et al. Gynecol Oncol. the resectability approximates that of primary debulking surgery. Shahin MS. 1993. Friedman RL. J Clin Oncol. Zhang ZY. 2. 2003. we can offer patients a variety of different modalities to control disease. Montz FJ. and cisplatin in patients with recurrent ovarian cancer who responded to first-line platinum-based regimens. Hall J. Carbone A. cytoreductive surgery in ovarian carcinoma patients with a documented previously complete surgical response. 12. Gynecol Oncol. Nieberg RK. the current study showed that the 2 curves clearly separated within 3 years. which is of fundamental importance. The role of secondary cytoreductive surgery in the treatment of patients with recurrent epithelial ovarian carcinoma. Hellman S. The first step.11:434 – 439.70:2129 –2136. et al. may be the greatest benefit of bulky tumor resection. 2002. Copeland LJ. 1995. Murray T. Stringer CA. optimal SCR. Surgery and salvage chemotherapy for Chinese women with recurrent advanced epithelial ovarian carcinoma: a retrospective case-control study. Effect of cytoreductive surgery on survival of patients with recurrent epithelial ovarian cancer. A prospective study. Berek JS. in the current series. Munkarah A. Trimble EL.76:1606 –1614. Secondary cytoreductive surgery for recurrent epithelial ovarian cancer.61:189 –193. Cancer. LaGasse LD. Cantu MG. Ward E. 2004 / Volume 100 / Number 6 Between salvage chemotherapy and SCR there is not a better choice. Jemal A.22. J Clin Oncol. editors. Mandeli JP. Rosenberg SA. Hacker NF. 8. Gynecol Oncol. Eifel PJ. Enhancement of sensitivity to chemotherapy.81:237–241. Ovarian cancer. Gynecol Oncol. 2001. Radical surgical procedure improves survival time in patients with recurrent ovarian cancer. it is not surprising that. With respect to second-line salvage chemotherapy. Gynecol Oncol. et al. 2002. Rubin SC. and immunotherapy. 2001. REFERENCES 1. Bristow RE. Morris M. 2003. Obstet Gynecol. 4. Firedman RL. Wang H. The fourth step is establishing multidisciplinary treatments that include SCR and second-line chemotherapy. 17.13:419 – 427. et al.88:144–153. DeVita VT. Markman M. Spirtos NM. . Bowel resection is one valid approach to removing bulky tumors that involve the intestine. Li ZT. 2002. J Surg Oncol. Cancer. Secondary cytoreductive surgery for patients with recurrent epithelial ovarian carcinoma. and there are no data that are convincing enough to advocate SCR as a routine procedure. 2002. Secondary cytoreductive surgery for localized intra-abdominal recurrences in epithelial ovarian cancer. Cai SM. The currently available second-line agents are less successful in prolonging longterm survival. 1983.75:26 –30. Int J Gynecol Cancer. Secondary cytoreduction for ovarian cancer following cisplatin therapy. 13. particularly when it is supported by multiple courses of salvage chemotherapy. 9. et al. Parma G. 11.20:1248 –1259. Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. hormone therapy. 3. Zang RY. 2000. Phase II trial of weekly single-agent paclitaxel in platinum/paclitaxel-refractory ovarian cancer. Phase II evaluation of 3-day topotecan with cyclophosphamide in the treatment of recurrent ovarian cancer. Hanjani P.20:1232–1237. Dottino PR. Conclusions SCR is feasible and efficacious in patients with relapsed EOC and a good performance status who achieve a clinical complete remission that lasts 3 months. Holscher M. 15. 6. 1995. Vlamis V. with these objectives in mind. A survival benefit is provided by 14. Gershenson DM. 5. Kuhn W. Scarabelli C. Optimal surgical outcomes are achieved easily in patients with apparent solitary tumor sites. 2000. Ozols RF. Buda A. which is one reason why the overall prognosis of patients with advanced EOC remains pessimistic. Schwartz PE. Wharton JT. optimal SCR as well as multiple courses of salvage chemotherapy were found to improve survival significantly in patients with relapsed EOC. Li ZT. Segna RA. Cancer: principles and practice of oncology. is maximizing the possibility of cytoreduction at primary surgery. Tomacruz RS. Armstrong DK. fallopian tube carcinoma. 10. Nolte S. when controlling for factors in multivariant analysis. Spitz D. including SCR. J Clin Oncol. Janicke F. 7. Gallo A. The third step is the development of effective maintenance therapies that prevent or delay recurrences in patients who achieve a complete remission after standard chemotherapy (remission consolidation). and. 5th edition. 2003.83:504 –512.34: 334 –338. Vaccarello I. Philadelphia: Lippincott-Raven Publishers. Wolf JK. doxorubicin. Edwards CL.1160 CANCER March 15. In: Vincent T. et al. Levenback C. Randomized controlled trial of single-agent paclitaxel versus cyclophosphamide. Samuels A. Thun MJ. Cancer statistics. Ghafoor A. Eisenkop SM.

Semin Oncol. Management of recurrent ovarian cancer: evidence-based decisions. The current role of gemcitabine in ovarian cancer. 2000. 1996. 19. Karlan BY. 2001.89:306 –313. Future directions in the treatment of ovarian cancer. 2002. Venkatraman E. 26:798 – 804. Salom E.14:519 –527. Li ZT. 23. 2003. Ozols RF.28(2 Suppl 7):18 –24. Lagasse LD. Aghajanian C. Zhang ZY. Vaidya A. . Semin Oncol.29(1 Suppl 1):32– 42. Zang RY. Bristow RE. 18. Patient selection and review of the literature. Secondary surgery cytoreduction for advanced epithelial ovarian cancer. Pothuri B. Barakat RR.78:2049–2062. Mirhashemi R. Chi DS. Impact of secondary cytoreductive surgery on survival of patients with advanced epithelial ovarian cancer. 21. Ozols RF.SCR and Relapsed EOC/Zang et al. Almeida Z. 2002. Gynecol Oncol. Palliative surgery for bowel obstruction in re- 1161 current ovarian cancer: an updated series. 22. Cancer. et al. 20. Eur J Surg Oncol. Curr Opin Oncol.