Prognostic factors for survival and resection in patients with initial nonresectable locally advanced pancreatic cancer treated with chemoradiotherapy.

Abstract

Controversies regarding the optimal therapy for patients with locally advanced pancreatic cancer (LAPC) exist. Although the prognosis as a whole remains dismal, subgroups are known to benefit from intensive therapy, including chemoradiotherapy (CRT). We describe the results in 178 patients treated from 2001 to 2010 and have developed a prognostic model for both survival and the possibility of a subsequent resection in these patients. From 2001 until 2010, 178 consecutive patients with LAPC were treated and included in the present study, with CRT consisting of 50 Gy in 27 fractions combined with tegafur-uracil(UFT)/folinic acid(FA). The median survival from diagnosis was 11.5 months. Adverse events of Grade 3 or above were seen in 36% of the patients. Ninety-three percent of the patients completed all fractions. A Cox regression model for survival demonstrated resection (hazard ratio [HR] 0.12; 95% confidence interval [CI], 0.1-0.3) and pre-CRT gemcitabine-based therapy (HR 0.57; 95% CI, 0.4-0.9) as being associated with a favorable outcome, increasing gross tumor volume (HR 1.14; 95% CI, 1.0-1.3) was associated with shorter survival. A logistic regression model showed Stage III disease (odds ratio [OR] 0.16; 95% CI, 0.0-1.1) and abnormal hemoglobin (OR 0.26; 95% CI, 0.0-1.2) as being associated with lower odds of resection. This study confirms the favorable prognosis for patients receiving gemcitabine therapy before CRT and the poor prognosis associated with increasing tumor volume. In addition, CRT in patients with abnormal hemoglobin and Stage III disease rarely induced tumor shrinkage allowing subsequent resection.