Patterns of failure after pancreaticoduodenectomy for ampullary carcinoma.

The clinical courses of 41 patients with ampullary carcinoma were retrospectively reviewed to determine patterns of failure after resection. The five year actuarial local control and overall survival rates of 29 patients undergoing only pancreaticoduodenectomy were 69 and 55 percent, respectively. For 12 patients with "low risk" pathologic features (tumors limited to the ampulla or duodenum, well or moderately well-differentiated histologic factors, uninvolved lymph nodes or resection margins), the five year actuarial local control and survival rate was 100 and 80 percent, respectively. Adjuvant treatment may be unnecessary for this favorable subset of patients. On the other hand, the five year actuarial local control and survival after pancreaticoduodenectomy of 17 patients with "high risk" pathologic features (tumors invasive of the pancreas, poorly differentiated histologic findings, involved lymph nodes or resection margins) was only 50 and 38 percent, respectively (p < 0.05). In 12 patients at "high risk" who also received postoperative radiation therapy after pancreaticoduodenectomy, there was a trend toward better local control (83 percent), but there was no improvement in survival. Distant metastases (liver, peritoneum and pleura) were the dominant factor in determining outcome in this group. Therefore, we propose a trial of preoperative irradiation in hopes of enhancing these outcomes by reducing the risk of dissemination of cancer cells during surgical resection, especially among the 70 percent of patients with high risk pathologic features.

Duke Scholars

Author Christopher G. Willett Radiation Oncology