Implications of second-look laparotomy in the context of optimally resected stage III ovarian cancer: A non-randomized comparison using an explanatory analysis: A Gynecologic Oncology Group study

Objective. A non-randomized comparison of outcome in women undergoing second-look laparotomy (SLL) or clinical follow-up, after receiving six cycles of combination chemotherapy with paclitaxel plus either cisplatin or carboplatin, for optimally resected stage III ovarian cancer. Methods. Prior to chemotherapy randomization, patients chose whether or not to undergo SILL; this was a stratification factor to insure balance of treatment assignment. Any subsequent therapy was physician-directed. Explanatory analysis replaced intent-to-treat because of a higher likelihood of detecting SLL effect in the presence of noncompliance. Results. There were 393 patients (median age: 54) who Elected SLL and 399 (median age: 59) who Elected No SLL. The former group was more likely to have gross residual disease at initial surgery than the latter group (69% versus 60%, respectively). In the Elected SLL group, 59 (15%) patients subsequently refused surgery, in nine (2%) surgery was contraindicated, and 31 (8%) relapsed or died prior to the procedure. Cancer was found in 46% of 294 (75%) patients undergoing SLL. Since early failures (prior to SLL) do not address benefit, such patients (SLL: 32; No SLL: 33), defined as progression-free survival (PFS) < 6 months, were excluded from analysis. The adjusted relative risk of progression is 0.89 (95% confidence interval: 0.75, 1.07); the difference in median PFS is 1.0 month (SLL: 23.9 months; No SLL: 22.9 months). The survival rate curves are superimposable. Conclusion. In the context of a non-randomized comparison, the performance of a SLL was not associated with longer survival. (c) 2005 Elsevier Inc. All rights reserved.