Outcomes and Prognostic Factors in Patients Receiving Adjuvant Therapy for Resected Exocrine Pancreas Cancer

Abstract

Following resection of non-metastatic exocrine pancreas cancer the optimal adjuvant treatment modality is uncertain. While a randomized trial has shown benefit of chemotherapy compared to observation, the role and timing of chemoradiation (CRT) is less clear. For more insight into the role of radiotherapy we reviewed outcomes and prognostic factors of patients treated with CRT, chemotherapy before and after CRT (C-CRT) and chemotherapy alone (CA). We reviewed 1,051 consecutive patients with non-metastatic exocrine pancreas carcinoma who underwent resection (R0 or R1) at our institution between March 1985 and January 2011. The study cohort includes 668 patients who received adjuvant therapy with > 6 months follow-up. Adjuvant treatment was immediate CRT (group A, n = 336, 50%), C-CRT (group B, n = 237, 36%), or CA (group C, n = 95, 14%). CRT most often consisted of 50.4 Gy in 28 fractions with concurrent 5-FU. C-CRT most frequently consisted of 2 cycles of gemcitabine before and after CRT. The most common CA regimen was 6 cycles of gemcitabine. Overall survival (OS) was estimated using the Kaplan-Meier method. Univariate and multivariate analysis (MVA) were performed using Cox proportional hazards regression models incorporating age, gender, pathologic features, and TNM stage. Median patient age was 65 years (range, 29-88 years). Median follow-up was 8 years (range, 0.5-25.8 years). Tumor location was pancreas head (81%), body (5%), tail (6.5%), or unspecified (7.5%). Tumor grade was I (1%), II (15 %), III (71 %), or IV (13 %). T stage was T1 (7.5%), T2 (23%), T3 (67%), or T4 (2.5%). N stage was N0 (39%) or N1 (61%). Extent of resection was R0 (82%) or R1 (18%). Group C had a higher mean age, group B had a higher T and N stage, and group A and B had a higher rate of R1 resections. ECOG performance status was similar between the groups. Median survival for the entire cohort was 25.8 months and 22.2, 32.4, and 22.2 months for groups A, B, and C, respectively. Three-year and 5-year OS was 33% and 23% for group A, 46% and 28% for group B, and 31% and 12% for group C (p = 0.02). Group B had improved OS relative to group A (p = 0.0001, HR = 0.67, 95% CI = 0.55-0.82), and group C (p = 0.009, HR = 0.67, 95% CI = 0.51-0.9). Other predictors of OS on MVA include grade ≥ 3 (p = 0.02, HR = 1.3, 95% CI = 1.05-1.7), T stage ≥ T3 (p = 0.02, HR = 1.25, 95% CI = 1.04-1.5), stage N1 (p = 0.0006, HR = 1.4, 95% CI = 1.15-1.7), and R1 status (p = 0.0002, HR = 1.55, 95% CI = 1.2-1.9). This large, single institution retrospective study suggests a survival benefit with C-CRT compared to CRT or CA, despite a higher frequency of negative prognostic factors in this cohort. These results are hypothesis generating and should be validated in a randomized clinical trial.