Personalized surgical therapy for advanced ovarian cancer: R0 resection after neoadjuvant chemotherapy is associated with decreased event-free survival compared with primary cytoreductive surgery

Abstract

Objectives: To determine differences in survival among patients who undergo complete gross resection (R0) after primary cytoreductive surgery (PDS) compared with interval cytoreductive surgery (IDS).Methods: Standard practice at our institution is to have all patients with advanced ovarian cancer (OC) undergo a triage algorithm to determine those who should receive PDS versus neoadjuvant chemotherapy (NACT) followed by IDS. Using a previously described algorithm, preoperatively defined parameters are analyzed using laparoscopy (LS) to determine who should be offered PDS. Patient outcomes were tracked prospectively. The Fisher exact test and Wilcoxon rank sum test were used to compare medians among patients with PDS versus IDS. Event-free survival (EFS) was defined as months from the date of diagnosis to the date of first recurrence, progressive disease, or death. Prognostic factors associated with EFS were evaluated with Cox proportional hazards regression.Results: Between April 2013 and September 2015, 206 patients with suspected advanced OC presented to our center. Of these, 119 were offered LS; 88% had serous histology; 132 (64%) patients had stage III disease; and 74 (36%) had stage IV disease, with more patients in the NACT group having stage IV disease (10% vs 51%, P < .001). Rates of R0 resection were similar between the PDS and IDS groups (87% [n = 65] vs 83% [n = 85], P = .521). Compared with patients who underwent PDS, those who had IDS had higher pretreatment CA-125 levels (675 vs 247, P < .001) and platelet counts (379 vs 303 × 109/L, P < .001). EFS was longer in the PDS group (20.8 months) compared with the IDS group (13.8 months, P = .002). Factors associated with decreased EFS on multivariate analysis include Charlson Comorbidity Index >3 (HR 2.20, 95% CI 1.30–3.72, P = .003), pretreatment CA-125 higher than 73 (HR 3.02, 95% CI 1.16–7.85, P = .032), pretreatment platelet counts greater than 306 × 109/L (HR 1.70, 95% CI 1.02–2.82, P = .041), and administration of NACT (HR 1.84, 95% CI 1.11–3.04, P = .019). Stage as a surrogate marker for tumor burden was not associated with EFS.Conclusions: Despite high rates of R0 cytoreduction at the time of IDS, patients undergoing NACT have decreased EFS than those undergoing PDS. Predicting those patients likely to achieve R0 resection in the upfront setting is important for maximizing patient outcomes. Objectives: To determine differences in survival among patients who undergo complete gross resection (R0) after primary cytoreductive surgery (PDS) compared with interval cytoreductive surgery (IDS). Methods: Standard practice at our institution is to have all patients with advanced ovarian cancer (OC) undergo a triage algorithm to determine those who should receive PDS versus neoadjuvant chemotherapy (NACT) followed by IDS. Using a previously described algorithm, preoperatively defined parameters are analyzed using laparoscopy (LS) to determine who should be offered PDS. Patient outcomes were tracked prospectively. The Fisher exact test and Wilcoxon rank sum test were used to compare medians among patients with PDS versus IDS. Event-free survival (EFS) was defined as months from the date of diagnosis to the date of first recurrence, progressive disease, or death. Prognostic factors associated with EFS were evaluated with Cox proportional hazards regression. Results: Between April 2013 and September 2015, 206 patients with suspected advanced OC presented to our center. Of these, 119 were offered LS; 88% had serous histology; 132 (64%) patients had stage III disease; and 74 (36%) had stage IV disease, with more patients in the NACT group having stage IV disease (10% vs 51%, P < .001). Rates of R0 resection were similar between the PDS and IDS groups (87% [n = 65] vs 83% [n = 85], P = .521). Compared with patients who underwent PDS, those who had IDS had higher pretreatment CA-125 levels (675 vs 247, P < .001) and platelet counts (379 vs 303 × 109/L, P < .001). EFS was longer in the PDS group (20.8 months) compared with the IDS group (13.8 months, P = .002). Factors associated with decreased EFS on multivariate analysis include Charlson Comorbidity Index >3 (HR 2.20, 95% CI 1.30–3.72, P = .003), pretreatment CA-125 higher than 73 (HR 3.02, 95% CI 1.16–7.85, P = .032), pretreatment platelet counts greater than 306 × 109/L (HR 1.70, 95% CI 1.02–2.82, P = .041), and administration of NACT (HR 1.84, 95% CI 1.11–3.04, P = .019). Stage as a surrogate marker for tumor burden was not associated with EFS. Conclusions: Despite high rates of R0 cytoreduction at the time of IDS, patients undergoing NACT have decreased EFS than those undergoing PDS. Predicting those patients likely to achieve R0 resection in the upfront setting is important for maximizing patient outcomes.