Does practice make perfect for pancreatic adenocarcinoma? Higher volume centers delivering chemoradiation are associated with improved surgical and survival outcomes

Abstract

Background: Chemoradiation delivery prior to surgical resection for pancreatic cancer theoretically may improve local control and thereby increase patient survival. Radiation techniques and concomitant chemotherapy regimens vary between centers. Facilities that have a higher patient volume receiving neoadjuvant chemoradiation may have superior radiation planning and delivery as well as supportive care. This study addresses the impact of volume of neoadjuvant chemoradiation on patient outcome. Methods: Patients with localized pancreatic adenocarcinoma who received neoadjuvant chemoradiation and surgery were identified in the National Cancer Database (NCDB) from 2006 to 2015. Facilities were stratified into low- and high-volume based on number of patients receiving neoadjuvant chemoradiation (cut point = 12). Surgical margin status, lymph node status, readmissions, mortality, and survival were compared between patients at low-volume and high-volume centers as well as patients referred elsewhere for neoadjuvant radiation. Fisher’s exact test was used to compare categorical variables between groups. Other than survival time, continuous variables were categorized. Survival was compared using Kaplan-Meier plots and the log-rank test. Cox proportional hazard regression was used to model survival time. Results: Of the 933 patients identified, 312 (33 %) received treatment at high-volume facilities and 621 (67%) received treatment at low-volume facilities. Over time, a larger portion of patients were treated at low-volume centers (74.10% in 2014–2015 vs 54.65% in 2006–2007, p = 0.05). Among those treated at high-volume centers, more patients were female (54.17% vs 46.86%, p = 0.037) and patients were treated at an earlier stage of disease (high-volume: stage I 25.32%, stage II 64.42%, stage III 10.26%; low-volume: stage I 21.7%, stage II 56.20%, stage III 22.06%; p < 0.05). The percentage of patients age 65 years or greater, burden of co-morbidities, and local income, education and population density levels did not differ from high- and low-volume centers. Patients treated at high-volume chemoradiation facilities were more likely to have negative margins following resection (91.50% vs 84.78%, p < 0.05). There was no difference in the number of lymph nodes examined, the frequency of node positivity, or the rates of complete pathologic response. Patients treated at high-volume centers had no difference in 30- or 90-day mortality rates. Post-operatively, patients at high-volume centers were more likely to undergo adjuvant therapy (33.65% vs 20.93%, p < 0.05). The median survival of patients treated at high-volume centers was 31.84 months compared to 24.71 months at low-volume centers (p < 0.05). Conclusion: Most patients receiving neoadjuvant chemoradiation therapy do not receive care at facilities frequently providing this service. Patients treated at high-volume centers have increased rates of negative margins at the time of resection as well as improved overall survival. The volume-outcome relationship that has been demonstrated in pancreatic surgery may translate to neoadjuvant chemoradiation, adding even further complexity to the challenge of operationalizing this observation in our current health care environment.