Survival effect of tumor size and extrapancreatic extension in surgically resected pancreatic cancer: proposal for improved T classification.

The T classification for pancreatic cancer of the American Joint Committee on Cancer may be inaccurate owing to lack of consideration of tumor size in cases of extension beyond the pancreas. To examine the accuracy of American Joint Committee on Cancer staging and to determine the prognostic implication of combined tumor size and extrapancreatic extension, 6145 cases of pancreatic ductal adenocarcinomas from the Surveillance, Epidemiology, and End Results database were categorized according to tumor size and extension as follows: group 1 (G1, ≤2 cm and limited to the pancreas), G2 (>2 cm and limited to the pancreas), G3 (≤2 cm with extrapancreatic extension), and G4 (>2 cm with extrapancreatic extension). The median survival of G1, G2, G3, and G4 were 23, 15, 19, and 14 months, respectively (P < .001), and the survival time in G3 was closer to that of G2 than G4. To test the classification system for accuracy of prognosis, G3 was merged with G2. The survival discrimination of this new grouping was greater (overall comparison, P < .001; G1 versus G2 + G3, P < .001; G2 + G3 versus G4, P < .001; χ(2) = 92.043) than that of the current T-classification scheme (overall comparison, P < .001; G1 versus G2, P < .001; G2 versus G3 + G4, P = .048; χ(2) = 60.424). To better discriminate survival, patients with a tumor less than or equal to 2 cm extending beyond the pancreas should be downstaged from the current class T3 to class T2.