Effect of preoperative chemoradiotherapy on surgical margin status of resected adenocarcinoma of the head of the pancreas

We examined the effect of preoperative chemoradiotherapy on the ability to obtain pathologically negative resection margins in patients undergoing pan creaticoduodenectomy for adenocarcinonma of the head of the pancreas. Betwe en 1987 and 2000, 100 patients underwent Whipple resection with curative in tent for primary adenocarcinoma of the head of the pancreas. Pathologic ass essment of six margins (proximal and distal superior mesenteric artery: pro ximal and distal superior mesenteric vein, pancreas, retroperitoneum, commo n bile duct, and hepatic artery) was undertaken bq either frozen section (p ancreas and common duct) or permanent section. A margin was considered posi tive if tumor was present less than 1 mm from the inked specimen. Il Margin s noted to be positive on frozen section were resected whenever possible. O f the 100 patients treated, 47 (47%) undern ent postoperative radiation and chemotherapy (group I) and 53 (53%) received preoperative chemoradiotherap y (group IT) with either 5-fluorouracil (32 patients) or gemcitabine (ZI pa tients). Patient demographics and operative parameters were similar in the two groups, with the exception of preoperative tumor size (CT scan), which Mas greater an group II (P <0.001), and number of previous operations, whic h a as greater in group II (P <0.0001). Statistical analysis of the number of negative surgical margins clear of tumor uas performed using Fisher's ex act test. All patients (100%) had six margins assessed for microscopic invo lvement with tumor. In the preoperative therapy group, 5 (7.5%) of 53 patie nts had more than one positive margin, whereas 21 (44.7%) of 47 patients wi thout preoperative therapy had more than one margin with disease extension (P <0.001). Additionally, only 11 (25.6%) of the 47 patients without preope rative therapy had six negative margins vs. 27 (50.9%) of 53 in the group r eceiving preoperative therapy (P = 0.013). Survival analysis reveals a sign ificant increase in survival in margin-negative patients (P = 0.02). Simila rly; a strong trend toward improved disease-free and overall survival seen in patients with a single positive margin vs, multiple margins. Overall, we find a negative impact on survival with an increasing number of positive m argins (P = 0.025, hazard ratio 1.3). When stratified for individual margin status, survival was decreased in patients with positive superior mesenter ic artery (P = 0.06) and vein (P = 0.04) margins. However, this has not yet resulted in a significant increase in disease-free or overall survival for patients receiving preoperative therapy (P = 0.07).