Objective: Intraperitoneal (IP) chemotherapy has gained enthusiasm in the treatment of ovarian cancer. Despite having a better survival advantage than intravenous (IV) chemotherapy, IP chemotherapy still poses significant morbidity and complications. Identifying the subset of patients who could best benefit from IP chemotherapy, and those who would least benefit from this treatment, thus avoiding potential complications, is critical.
Methods: Between January 2001 and December 2007, 367 patients with stage III epithelial ovarian cancer underwent randomized trial for IP/IV chemotherapy were recruited to construct a nomogram, which is a graphical representation of Cox proportional hazards model adopting six weighted risk factors including age, CA125, IP/IV delivery, stage, histology, and upper abdominal metastases. The nomogram was internally validated for discrimination and calibration. The concordance index was used for quantifying the predictive ability of overall survival with bootstrapping to correct for bias.
Results: The cycles of completed IP chemotherapy had an impact on overall survival (> or =5 vs. < or =4 cycles, P=0.02). A nomogram for predicting median survival and 5-year survival probability was constructed with a concordance index of 0.72. Upper abdominal tumor metastases (P<0.001) and colon resection (P=0.02) predicted increased chances for early discontinuation of IP chemotherapy.
Conclusions: At least five IP cycles are needed to achieve better survival. Nomogram can help to identify the subset of patients who can least benefit from IP chemotherapy, thus avoiding potential IP complications and help to facilitate discussion between patient and physician, risk stratification, and help to guide clinical care.