Resection of localized pancreatic head ductal adenocarcinoma (LPHDA) has a limited impact on survival. Mechanisms of improvement provided by preoperative chemoradiation therapy (CRT) remain under debate. This study analyzes the outcome of patients treated for LPHDA to delineate the benefits of CRT. Among 87 patients with LPHDA, 17 had a pancreaticoduodenectomy alone (group I). Thirty-nine with initially resectable cancers received CRT with 5-fluorouracil-based chemotherapy (group II). Thirty-one with initially unresectable cancers were similarly treated by CRT (group III). Patients in groups II and III were restaged after completion of CRT. In patients with resectable disease, resection was planned. Patients in groups I and II were statistically comparable in terms of age, sex, and pretherapeutic stage. Median survival and 2-year overall survival in group I were 13.7 months and 31%, respectively. In group II, 23 patients (59%) had a pancreaticoduodenectomy (group IIa) and 16 patients (41%) did not have resection (group IIb). Median survival and 2-year overall survival were as follows: group IIa, 26.6 months and 51%; and group IIb, 6.1 months and 0%, respectively. In group IIa, pathologic examination revealed eight major responses (35%) including two sterilized specimens, and none of the patients had locoregional recurrence. In group III, none of the patients had resection, and median survival was 8 months with one 2-year survivor. Patient selection appears to play a major role with regard to results achieved with preoperative CRT followed by pancreaticoduodenectomy. However, a high histologic response rate and excellent local control can also be achieved.