Consensus in determining the resectability of locally progressed pancreatic ductal adenocarcinoma - results of the Conko-007 multicenter trial

Background: One critical step in the therapy of patients with localized pancreatic cancer is the determination of local resectability. The decision between primary surgery versus upfront local or systemic cancer therapy seems especially to differ between pancreatic cancer centers. In our cohort study, we analyzed the independent judgement of resectability of five experienced high volume pancreatic surgeons in 200 consecutive patients with borderline resectable or locally advanced pancreatic cancer. Methods: Pretherapeutic CT or MRI scans of 200 consecutive patients with borderline resectable or locally advanced pancreatic cancer were evaluated by 5 independent pancreatic surgeons. Resectability and the degree of abutment of the tumor to the venous and arterial structures adjacent to the pancreas were reported. Interrater reliability and dispersion indices were compared. Results: One hundred ninety-four CT scans and 6 MRI scans were evaluated and all parameters were evaluated by all surgeons in 133 (66.5%) cases. Low agreement was observed for tumor infiltration of venous structures (kappa = 0.265 and kappa = 0.285) while good agreement was achieved for the abutment of the tumor to arterial structures (interrater reliability celiac trunk kappa = 0.708 P < 0.001). In patients with vascular tumor contact indicating locally advanced disease, surgeons highly agreed on unresectability, but in patients with vascular tumor abutment consistent with borderline resectable disease, the judgement of resectability was less uniform (dispersion index locally advanced vs. borderline resectable p < 0.05). Conclusion: Excellent agreement between surgeons exists in determining the presence of arterial abutment and locally advanced pancreatic cancer. The determination of resectability in borderline resectable patients is influenced by additional subjective factors.