Evaluation of the MDACC clinical classification system for pancreatic cancer patients in an European multicenter cohort

Background: The MDACC group recommends to extend the current borderline classification for pancreatic cancer into three groups: type A patients with resectable/borderline tumor anatomy, type B with resectable/borderline resectable tumor anatomy and clinical findings suspicious for extrapancreatic disease and type C with borderline resectable and marginal performance status/severe pre-existing comorbidity profile or age>80. This study intents to evaluate the proposed borderline classification system in a multicenter patient cohort without neoadjuvant treatment. Methods: Evaluation was based on a multicenter database of pancreatic cancer patients undergoing surgery from 2005 to 2016 (n ¼ 1020). Complications were classified based on the Clavien-Dindo classification. c2-test, KaplaneMeier estimator and Cox regression hazard model were used for statistical analysis. Results: Most patients (55.1%) were assigned as type A patients, followed by type C (35.8%) and type B patients (9.1%). Neither the complication rate, nor the mortality rate revealed a correlation to any subgroup. Type B patients had a significant worse progression free (p < 0.001) and overall survival (p ¼ 0.005). Type B classificationwas identified as an independent prognostic marker for progression free survival (p ¼ 0.005, HR 1.47). Conclusion: The evaluation of the proposed classification in a cohort without neoadjuvant treatment did not justify an additional medical borderline subgroup. A new subgroup based on prognostic borderline patients might be the main target group for neoadjuvant protocols in future.