Neoadjuvant chemotherapy and radiation for patients with locally unresectable pancreatic adenocarcinoma: feasibility, efficacy, and survival.

ABSTRACT

BACKGROUND: We evaluated the feasibility and efficacy of neoadjuvant chemotherapy and radiation for patients with locally unresectable pancreatic cancer. MATERIALS AND METHODS: From October 2000 to August 2006, 245 patients with pancreatic adenocarcinoma underwent surgical exploration at our institution. Of these, 78 patients (32%) had undergone neoadjuvant therapy for initially unresectable disease, whereas the remaining patients (serving as the control group) were explored at presentation (n=167). All neoadjuvant patients received gemcitabine-based chemotherapy, often in conjunction with docetaxal and capecitabine in a regimen called GTX (81%). Seventy-five percent of neoadjuvant patients also received preoperative abdominal radiation (5,040 rad). RESULTS: Neoadjuvant patients were younger than control-group patients (60.8 vs 66.2 years, respectively, p<0.002). Seventy-six percent of neoadjuvant patients were resected as compared to 83% of control patients (NS). Concomitant vascular resection was required in 76% of neoadjuvant patients but only 20% of NS (p<0.01). Complications were more frequent in the neoadjuvant group (44.1 vs 30.9%, p<0.05), and mortality was higher (10.2 vs 2.9%, p<0.03). Among the neoadjuvant patients, all but one of the deaths were in patients that underwent arterial reconstruction. Mortality for patients undergoing a standard pancreatectomy without vascular resection was 0.8% in this series. Of patients resected, negative margins were achieved in 84.7% of neoadjuvant patients and 72.7% of NS. Within the cohort of neoadjuvant patients, radiation significantly increased the complication rate (13.3 vs 54.6%, p<0.006), but did not affect median survival (512 vs 729 days, NS). Median survival for patients who received neoadjuvant therapy (503 days) was longer than NS that were found to be unresectable at surgery (192 days, p<0.001) and equivalent to NS that were resected (498 days). CONCLUSIONS: Resection rate, margin status, and median survivals were equivalent when neoadjuvant patients were compared to patients considered resectable by traditional criteria, demonstrating equal efficacy. Surgical resection with venous reconstruction following neoadjuvant therapy for patients with locally advanced pancreatic cancer can be performed with acceptable morbidity and mortality. This approach extended the boundaries of surgical resection and greatly increased median survival for the "inoperable" patient with advanced pancreatic cancer.