Abstract Long-term survivors after pancreatic resection for PDAC are rare, constituting a specific subset of patients that remains poorly understood. This study aims to identify the clinic-pathological, molecular, and therapeutic factors for predicting long-term survival (LTS). A systematic review and random-effect meta-analysis were conducted. Inclusion criteria were PDAC histology, resected patients, studies reporting risk factors, and comparing two groups. The primary endpoint was to evaluate predictive factors for LTS in patients with PDAC who underwent surgery. Results were reported with the Mantel–Haenszel random effects model using Risk Ratio (RR) or Mean Difference (MD). Meta-regression analysis was used to clarify heterogeneity. Nineteen studies, involving a total of 5412 patients, were included: 1097 (20,3%) in group LTS and 4334 (79,7%) in group STS (short-term survivors). These factors were associated to LTS: small size (RR 1.53, 95% IC 1.14; 2.05); T1-T2 stage (RR 1.07, 95% IC 1.03; 1.11); N0 (RR 1.82, 95% IC 1.60; 2.09); AJCC Stage I (RR 2.28 95% IC 1.87; 2.79); low-grade G1-2 (RR 1.21, 95% IC 1.09; 1.34); R0 resection (RR 1.11, 95% IC 1.08; 1.13); low levels of CEA (MD − 4.41, 95% IC − 6.23; − 2.59) and Ca 19.9 (MD − 66.4, 95% IC: − 71.9; − 60.9); absence of perineural invasion (RR 0.93, 95% IC: 0.90; 0.96), lymph-vascular invasion (RR 0.87, 95% IC: 0.83; 0.91), venous invasion (RR 0.63, 95% IC: 0.48; 0.83) and perioperative transfusions (RR 0.56, 95% IC: 0.40; 0.79). Several factors are associated with an LTS. They can be considered reliable indicators for predicting tumor progression.
D’Ambra et al. (Mon,) studied this question.