Analysis of prognostic factors in patients with advanced pancreatic cancer: Subgroup analysis of a randomized phase III trial comparing single-agent gemcitabine to the gemcitabine plus cisplatin combination

Abstract

4105 Background: The combination of gemcitabine (Gem) and cisplatin (Cis) has shown promising activity in several phase II and III trials in the treatment of patients with advanced pancreatic cancer. While in a previously reported phase III trial (Heinemann V et al, Proc ASCO 22: # 1003) the Gem+Cis regimen (arm A) showed only a trend towards a superior overall-survival (OS) as compared to single-agent Gem (arm B), there might be subgroups of patients that more clearly benefit from combination therapy. Methods: This phase III trial enrolled 195 patients with histologically proven advanced pancreatic adenocarcinoma to receive either Gem+Cis or Gem alone. Subgroup analyses were performed to evaluate the impact of treatment, Karnofsky performance status (KPS) and other prognostic factors on OS and progression-free survival (PFS). Treatment differences between the subgroups were analysed by the Kaplan-Meier method, the log-rank test and the Cox model. Results: Median OS was 225 days (d) in arm A and 181 d in arm B, respectively (p=0,15). Data of 160 patients were available for this subgroup analysis. In univariate as well as multivariate analysis, KPS determined at the time of randomisation was the strongest prognostic factor for median OS (KPS 70–80%: 143 d vs. KPS 90–100%: 270 d, p=0,0006) and for PFS (KPS 70–80%: 86 d vs. KPS 90–100%: 162 d, p=0,0043). In the subgroup with a good KPS (90–100%) median OS was 322 d for patients treated with Gem+Cis and 206 d for treatment with Gem alone (p=0,051). Median OS for the subgroup Gem/KPS 70–80% was 143 d and for Gem+Cis/KPS 70–80% 147 d, respectively (p=0,64). Conclusion: In this randomized phase III trial, patients with a good KPS (90–100%) appeared to benefit from the more intensive combination therapy Gem+Cis and a had a considerably longer median OS as compared to patients treated with gemcitabine alone. By contrast, in patients with a poor KPS (70–80%) no survival benefit was observed when Gem alone was compared to the combination therapy with Gem+Cis. This observation should be included into the design of future trials on combination therapy of pancreatic cancer. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Lilly Germany Lilly Germany