Prognostic indicators for survival after resection of pancreatic adenocarcinoma

From October 1983 to October 1990, 799 patients with the diagnosis of pancreatic adenocarcinoma were admitted to Memorial Sloan-Kettering Cancer Center, and their records were entered into a prospective database. Curative resection was possible in 146 patients (18%), with a 30-day operative mortality of 3.4%. Median follow-up of survivors in the resection group was 28 months. Actuarial 5-year survival in patients who did not undergo resection was 0% (n = 653), compared with 24% in the patients who had resection (p < 0.00001). Of 52 patients available for 5-year follow-up, 10 (19%) were alive at 5 years. Tumors with lymph node involvement, poor histologic tumor differentiation, and tumor size greater than 2.5 cm were predictors of a significantly worse survival in both univariate and multivariate analysis. Other factors, including blood transfusion, blood loss, and operative time, had no significant effect on survival. The majority of patients with pancreatic adenocarcinoma have unresectable disease with a poor prognosis. For patients who undergo curative resection, their prognosis appears to be determined by the biology of the tumor rather than factors involved in the resection.