Benefit of Adjuvant Chemoradiation Therapy for Ampullary Adenocarcinoma: The Johns Hopkins Hospital - Mayo Clinic Collaborative Study

Abstract

The role of adjuvant chemoradiation therapy (CRT) for ampullary adenocarcinoma (AA) is unknown. We combined the experience of two institutions to examine the efficacy of adjuvant CRT. Patients who underwent curative surgery for AA at the Johns Hopkins Hospital (JHH, n = 290; 1992-2007) and at the Mayo Clinic (n = 130; 1977-2005) were reviewed. Patients who died within 60 days of surgery, <60 days follow-up, and had metastatic disease at surgery were excluded. Those missing information on T-stage, tumor size, margin status, node status, and histologic grade were also excluded. The final cohort included 186 patients (n = 104 JHH, n = 82 Mayo). Most patients received 5-FU based CRT with conformal RT. Median overall survival (mOS) was 39.9 months for all patients and did not significantly differ by institution (p = 0.350). Patients who received CRT (n = 66) vs. no CRT (n = 120) tended to be younger (median age 63.3 vs. 71.3 yrs, p < 0.001), more likely stage T3/4 (57.6% vs. 30.8%, p < 0.001), node positive disease (72.7% vs. 30.0%, p < 0.001), margin positive disease (4.6% vs. 0.0%, p = 0.019), and from JHH (71.2% vs. 47.5%, p = 0.002). For all patients, 58.6% were male, 54.3% had histologic grade 3 disease with a median tumor size of 2.2 cm, and those who received CRT did not differ by these factors. Five-year OS was 39.1% for the entire group of patients (37.2% for the surgery-only group and 42.1% for the adjuvant CRT group, log-rank test p = 0.84). Of those who died (n = 79), 53.2% had liver mets. On univariate analysis, adverse prognostic factors for OS included T stage (T3/4 vs. T1/T2, RR = 1.86, p = 0.002), node positive status (RR = 3.18, p < 0.001, median OS was 23.0 months for node-positive vs. 79.4 months for node-negative), and histological grade (3 vs. 1/2, RR = 1.69, p = 0.011). Age > = 75, tumor size >=3 cm, gender, and margin positivity were not associated with OS. Adjuvant CRT was not significantly associated with OS (mOS 40.1 mos no CRT vs. 39.9 mos adjuvant CRT, RR = 0.96, p = 0.84). Adjusting for institution, compared to no CRT use of adjuvant CRT improved OS among node positive (RR = 0.47, p = 0.004; mOS 32.1 vs. 15.7 mos) and node negative disease (RR = 0.45, p = 0.11; mOS 103.2 vs. 61.6 mos). MVA adjusting for institution, age, size, gender, T-stage, node status, and grade demonstrated adjuvant CRT was significantly associated with overall survival (RR = 0.40, p < 0.001). Patients with resected ampullary adenocarcinoma benefit from 5-FU based CRT, particularly among node positive patients. Adjustment for confounders suggests adjuvant CRT improves survival, regardless of node status.