Abstract
INTRODUCTION: It is accepted that patients with pancreatic cancer are best treated by a combination of surgical resection and systemic chemotherapy; the optimal sequence remains debated. There are ongoing trials on survival benefit of neoadjuvant therapy (NAT) compared to upfront surgery (UFS) in patients with resectable and borderline resectable pancreatic cancer; new and updated data must be analyzed for continued understanding of results. METHODS: A search of multiple databases was performed for randomized controlled trials (RCTs) comparing NAT with UFS for resectable and borderline resectable pancreatic cancer. The systematic review and meta-analysis were performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The primary outcome of interest was overall survival (OS) by intention-to-treat; subgroup analysis was also performed for resectable disease only. Meta-analyses were completed using random-effects models. RESULTS: There were 13,406 records screened for inclusion; 51 were retrieved for further review; 34 were assessed for eligibility. Ten RCTs were included with 1340 patients split almost equally between NAT and UFS. Gemcitabine-based NAT was used in all but two RCTs. There was a significant improvement in median OS for NAT compared to UFS (hazard ratio 0.78, 95% confidence interval 0.61-0.99; P = 0.04; I(2) = 43%). In the subgroup analysis of patients with resectable pancreatic cancer, there was no improvement in survival (hazard ratio 0.86, 95% confidence interval 0.59-1.26; P = 0.36; I(2) = 49%). CONCLUSIONS: In this meta-analysis of prospective RCTs assuming intention-to-treat, NAT was associated with improvement in OS relative to UFS in patients with resectable and borderline resectable pancreatic cancer. Given the heterogeneity of the NAT regimens studied in this meta-analysis and recruitment challenges, additional analyses are warranted to confirm these findings and to determine the optimal sequencing of treatment modalities.