Gemcitabine mono-therapy versus gemcitabine plus targeted therapy in advanced pancreatic cancer: a meta-analysis of randomized phase III trials

Abstract

Background: Prognosis of advanced pancreatic cancer is dismal and the novel targeted therapies, albeit successfully used to treat many advanced tumors, have shown modest results. We performed a meta-analysis in order to quantify the effect size on survival of adding targeted therapy to single agent gemcitabine. Methods: Randomized phase III trials comparing gemcitabine mono-therapy versus gemcitabine plus a targeted agent in first-line treatment of advanced pancreatic cancer designed on survival as primary outcome were selected. Search was done through Medline and the registry of the NIH. Keywords used for searching were ‘pancreas’, ‘pancreatic’, ‘gemcitabine’. Study quality was assessed with MERGE criteria. Findings were depicted in classical Forest plots. Publication bias was evaluated by the construction of funnel plot. Results: Nine studies met the meta-analysis inclusion criteria including 4564 patients. The target therapies were: erlotinib, cetuximab, rigosertib, elpamotide, bevacizumab, aflibercept, axitinib, masitinib and ganitumab. There was no statistically significant heterogeneity among the nine trials (p = 0.77). The hazard ratio (HR) of the pooled analysis was 0.998 (CI 95%: 0.932–1.068). Subgroup meta-analysis was also performed in anti-EGFR and anti-angiogenesis trials: the pooled HR were 0.94 (CI 95%: 0.705–1.175) and 1.055 (CI 95%: 0.913–1.197), respectively. Conclusions: The present meta-analysis does not show significant improvements in survival for targeted drugs in advanced pancreatic cancer. The possible reason of these results could be linked to the biology of pancreatic cancer as well as to the absence of predictive factors.