The objective was to assess whether an early response to neoadjuvant chemotherapy in women with advanced ovarian cancer may predict short‐ and long‐term clinical outcome.
This is a retrospective study of all women with stage III‐IV tubo‐ovarian cancer treated with neoadjuvant chemotherapy at a single center in Montreal between 2003 and 2014. Logistic regression models were used to evaluate the association between cancer antigen 125 (CA‐125) levels during neoadjuvant chemotherapy and debulking success. Cox proportional hazard models were used to estimate hazard ratios and their respective 95% CI for death and recurrence. Harrell's concordance indices were calculated to evaluate which variables best predicted the chemotherapy‐free interval and overall survival in our population.
In all, 105 women were included. Following the first, second, and third cycles of neoadjuvant chemotherapy, CA‐125 levels had a median reduction of 43.2%, 85.4%, and 92.9%, respectively, compared with CA‐125 levels at diagnosis. As early as the second cycle, CA‐125 was associated with overall survival (hazard ratio 1.03, 95% CI 1.01‐1.05, per 50 U/mL increment). By the third cycle, CA‐125 did not only predict overall survival (hazard ratio 1.04, 95% CI 1.01‐1.08), but it predicted overall survival better than the success of debulking surgery (Harrell's concordance index 0.646 vs 0.616). Both absolute CA‐125 levels and relative reduction in CA‐125 levels after 2 and 3 cycles predicted the chance to achieve complete debulking (P < .05).
Reduction of CA‐125 levels during neoadjuvant chemotherapy provides an early predictive tool that strongly correlates with successful cytoreductive surgery and long‐term clinical outcome in women with advanced high‐grade serous and endometrioid ovarian cancer.