Upfront surgery versus neoadjuvant therapy for resectable pancreatic cancer: systematic review and Bayesian network meta-analysis

Alison Bradley, Robert Van Der Meer

Research output: Contribution to journalArticlepeer-review

17 Citations (Scopus)
13 Downloads (Pure)


Current treatment recommendations for resectable pancreatic cancer support upfront resection and adjuvant therapy. Randomized controlled trials offering comparison with the emerging neoadjuvant approach are lacking.

This review aims to compare both treatment strategies for resectable pancreatic cancer. PubMed, MEDLINE, Embase, Cochrane Database and Cochrane Databases were searched for studies comparing neoadjuvant and surgery-first with adjuvant therapy for resectable pancreatic cancer. A Bayesian network meta-analysis was conducted using the Markov chain Monte Carlo method. Cochrane Collaboration’s risk of bias, ROBINS-I and GRADE tools were used to assess quality and risk of bias of included trials.

9 studies compared neoadjuvant therapy and surgery-first with adjuvant therapy (n = 22,285). Aggregate rate (AR) of R0 resection for neoadjuvant therapy was 0.8008 (0.3636–0.9144) versus 0.7515 (0.2026–0.8611) odds ratio (O.R.) 1.27 (95% CI 0.60–1.96). 1-year survival AR for neoadjuvant therapy was 0.7969 (0.6061–0.9500) versus 0.7481 (0.4848–0.8500) O.R. 1.38 (95% CI 0.69–2.96). 2-year survival AR for neoadjuvant therapy was 0.5178 (0.3000–0.5970) versus 0.5131 (0.2727–0.5346) O.R. 1.26 (95% CI 0.94–1.74). 5-year AR survival for neoadjuvant therapy was 0.2069 (0.0323–0.3300) versus 0.1783 (0.0606–0.2300) O.R. 1.19 (95% CI 0.65–1.73).

In conclusion neoadjuvant therapy may offer benefit over surgery-first and adjuvant therapy. However, further randomized controlled trials are needed.
Original languageEnglish
Article number4354
Number of pages7
JournalScientific Reports
Issue number1
Publication statusPublished - 13 Mar 2019


  • pancreatic cancer
  • neoadjuvant therapy
  • Bayesian network model
  • surgery

Cite this