702 Background: Pancreatic cancer continues to have a dismal prognosis despite improvements in surgical care. Resection status and lymph node status continues to be key factors impacting prognosis. There remains significant debate on the utility of increasing nodal harvest in patients undergoing surgery, while others report an improvement in survival. We sought to evaluate the impact of lymph node positivity to lymph nodes resected in patients with pancreatic cancer undergoing resection. Methods: Utilizing the National Cancer Database, we identified patients diagnosed with pancreatic cancer who underwent neoadjuvant chemotherapy (NCT) or neoadjuvant chemoradiation (NCR) followed by resection. We then stratified by nodal status and calculated a LN+/LNR (lymph node resected) ratio. Pearson’s Chi-square test was used to compare categorical variables. Survival analysis was estimated by the Kaplan-Meier method and associated log-rank test. Propensity score matching (PSM) was completed for variables of age, sex, tumor size, LN removed, LN positive, pathogenic T- stage and N-stage, grade, 30- and 90-day mortality, R0 resection, facility volume, response, adjuvant therapy, and LNR. Significance was considered at p-value ≤0.05. Results: We identified 4117 patients with a median age of 65 (25-90). There were 2323 patients who underwent NCT and 1794 who underwent NCR. The median LN resected in the NCT was 18 vs 15 in the NCR, p<0.001. However, there were more LN+ in the NCT vs NCR, 2.1 vs 0.9, p<0.001. The R0 resection rates were lower in the NCT vs NCR 81% vs 86%, p<0.001. Additionally, complete response rates were higher amongst the NCR, 9.2% vs 4.7%, p<0.001. Lymph node ratios were consistently higher in the NCT cohorts among all ratio strata, p<0.001. Patients with pN+ disease had a median and 5-year survival of 26.4mo and 17% compared to their pN- counterparts at 38.8mo and 36%, p<0.001. Patients treated with NCT demonstrating a LNR of 0.01-0.1 had median and 5-year survival of 32.8 and 27%, LNR 0.11-0.4 (25.5mo and 12%), and LNR>0.4 (19.1mo and 10%), p<0.001. Patients treated with NCR demonstrating a LNR of 0.01-0.1 had median and 5-year survival of 29.1 and 21%, LNR 0.11-0.4 (25.8mo and 17%), and LNR>0.4 (21.3mo and 5%), p<0.001. PSM confirmed a higher LNR in the NCT cohort compared to NCR, p<0.001, however survival was also longer in these patients among the LNR 0.01-0.1, and 0.11-0.4, p<0.001. Conclusions: Lymph node status in patients with pancreatic cancer continues to be a significant prognostic indicator of survival. Moreover, lymph node ratio may be a more accurate predicter of survival in patients undergoing neoadjuvant chemotherapy and neoadjuvant chemoradiation.
Read full abstract