Impact of resection margin status on survival in pancreatic cancer patients after neoadjuvant treatment and pancreatoduodenectomy.

BACKGROUND: Resection margin status has been recognized as an independent prognostic factor on overall survival in pancreatic cancer patients undergoing surgical resection. However, its impact after neoadjuvant treatment remains uncertain. METHODS: We analyzed 305 patients with resectable or borderline resectable pancreatic cancer treated with neoadjuvant therapy and pancreatoduodenectomy at 3 tertiary referral centers between 2010 and 2017. Positive resection margin was defined as 1 or more cancer cells at any margin. Overall survival was measured from the date of surgery until death or last follow-up. RESULTS: One hundred and seventy-eight patients received neoadjuvant chemotherapy and 127 received neoadjuvant chemoradiotherapy. The median overall survival was 29.8 months. The 1-, 3-, and 5-year overall survival rates were 79.2%, 44.0%, and 23.5%, respectively. Negative margin was achieved in 275 (90.2%) patients. Negative margin resection patients had a significantly longer overall survival than positive resection margin patients (31.3 vs 16.3 months, P < .001). In univariate analyses, overall survival was associated with age, margin status, histologic grade, ypT, number of positive lymph nodes, perineural invasion, treatment effect, postoperative carbohydrate antigen 19-9, and adjuvant therapy. Positive margin resection, poorly differentiated carcinoma, treatment effect score of 3, postoperative carbohydrate antigen 19-9 of 37 U/mL or higher, and lack of adjuvant therapy were predictive of poor overall survival in multivariate Cox regression analysis. CONCLUSION: Margin status was an independent predictor of overall survival in patients treated with neoadjuvant therapy and pancreatoduodenectomy, supporting the use of a negative margin resection as a surrogate of adequate oncological resection in this setting. Our findings may also have significant implications for patient stratification in future randomized trials.