Nomogram for Predicting Individual Survival After Recurrence of Advanced-Stage, High-Grade Ovarian Carcinoma

Abstract

Objective: To analyze clinical prognostic factors for survival after recurrence of high-grade, advanced-stage ovarian-peritoneal-tubal carcinoma and to develop a nomogram to predict individual survival after recurrence.

Methods: We retrospectively analyzed patients treated in multicenter Gynecologic Oncology Group protocols for stage III and IV ovarian-peritoneal-tubal carcinoma who underwent primary debulking surgery, received chemotherapy with paclitaxel and a platinum compound, and subsequently developed recurrence. Prognostic factors affecting survival were identified and used to develop a nomogram, which was both internally and externally validated.

Results: There were 4,739 patients included in this analysis, of whom, 84% had stage III and 16% had stage IV ovarian carcinoma. At a median follow-up of 88.8 months (95% CI 86.2-92.0 months), the vast majority of patients (89.4%) had died. The median survival after recurrence was 21.4 months (95% CI 20.5-21.9 months). Time to recurrence after initial chemotherapy, clear cell or mucinous histology, performance status, stage IV disease, and age were significant variables used to develop a nomogram for survival after recurrence, which had a concordance index of 0.67. The time to recurrence alone accounted for 85% of the prognostic information. Similar results were found for patients who underwent second look laparotomy and had a complete pathologic response or received intraperitoneal chemotherapy.

Conclusion: For individuals with advanced-stage ovarian carcinoma who recur after standard first-line therapy, estimated survivals after recurrence are closely related to the time to recurrence after chemotherapy and prognostic variables can be used to predict subsequent survival.

Clinical trial registration: ClinialTrials.gov, NCT00002568, NCT00837993, NCT00002717, NCT01074398, and NCT00011986.

Conflict of interest statement

Financial Disclosure

Melissa Geller has received research grants from Genentech Inc., FATE Therapeutics, Morphotek, and TESARO, Inc. The other authors did not report any potential conflicts of interest.

Figures

Figure 1.

Kaplan–Meier overall survival after recurrence. Figures below months indicate median survival: 21.4 months.

Figure 2.

Restricted cubic spline fit of effects of time to recurrence on overall survival. This figure shows the partial effects plot of log time to recurrence on the log hazard ratio. Note that the risk of death decreases with increasing time to recurrence more quickly for shorter intervals.

Figure 3.

Nomogram for predicting median survival time after recurrence. To use, find the patient’s recurrence interval on the recurrence axis, then draw a straight line upward to the points axis to determine how many points toward death the patient receives for her recurrence interval. Do this again for the other axes, each time drawing a straight line upward toward the points axis. Sum the points received for each predictor and find the sum on the total points axis. Draw a straight line down to the median-survival axis to find the patient’s median survival time after recurrence of ovarian cancer. FIGO, International Federation of Gynecology and Obstetrics.

Figure 4.

Calibration curve for the overall survival nomogram model. The dotted line represents an ideal nomogram and the solid line represents the observed nomogram. The vertical bars indicate 95% CIs and the open circles represent bias-corrected estimates.

Figure 5.

Kaplan–Meier overall survival after recurrence for the Gynecologic Oncology Group-218 validation cohort. Median survival: 25.4 months (95% CI, 23.0–28.3 months).