Second-line (SL) chemotherapy containing oxaliplatin (OXA) in metastatic pancreatic cancer (PC): Whom should we trust?

Abstract

e15719 Background: After the intensification of first-line treatment in metastatic PC the number of patients fit and able to receive a SL treatment is increasing; despite these results, the choice of SL chemotherapy remains a challenging issue. The role of OXA in SL setting has given conflicting results in CONKO-003 and PANCREOX trials; the reasons of this inconsistency are difficult to understand. Methods: In order to detect a possible advantage from the use of oxaliplatin in SL, we pooled together the results of randomized phase II/III studies adding OXA to fluoropyrimidines after a gemcitabine-based first-line. Hazard ratios (HRs) for progression-free (PFS) and overall survival (OS) with 95% confidence intervals (CIs) were extracted, while the number of events was used for RR to determine odds ratio (OR) with corresponding 95%CIs. Fixed-effect and random-effect models were used as appropriate, on the basis of the evidence of a non-significant or significant heterogeneity among trials, respectively. The Review Manager software was used (version 5.3). Results: Three studies were identified; main characteristics and results are summarized in the table. No benefit in overall survival derived from the use of OXA (HR 1.03, 95% CI, 0.64-1.67; p=0.90) but significant heterogeneity was reported (p=0.003); the addition of OXA produced a significant benefit in terms of PFS (HR 0.81, 95% CI 0.68-0.97; p=0.02) with a higher chance of response (OR 1.81, 95% CI 1.01-3.24; P=0.05). Interestingly heterogeneity between trials was less evident in PFS and RR than in OS suggesting that this could be a confounding element in the interpretation of data. Conclusions: The reasons of conflicting results concerning the use of oxaliplatin doublets in SL setting are not clear. Given the recent advances in first-line setting, it would be interesting to identify an optimal sequential strategy, hopefully in a personalized approach. [Table: see text]