Treatment at high-volume facilities and academic centers in relation to overall survival in patients with locally advanced pancreatic cancer: A National Cancer Database analysis

Abstract

392 Background: Numerous studies have shown that treatment at a high volume facility (HVF) for patients (pts) with pancreatic cancer is associated with improved outcomes, particularly with pancreatectomy. In fact, a recent study showed that pts undergoing a pancreatectomy at an academic center (AC) is independently associated with improved outcomes. However, the role of chemotherapy (CT) and radiation (RT) in the treatment of locally advanced pancreatic cancer (LAPC) at HVF and AC, to our knowledge, has not been studied. Herein, we investigate the benefit of treatment at HVF and AC compared to low volume facilities (LVF) and non-academic centers (NAC) with CT or chemoradiation (CRT) in pts with LAPC. Methods: The National Cancer Database (NCDB) was utilized to identify LAPC patients treated at all facility types. All patients were treated with CT or CRT. Univariate (UVA) and multivariate (MVA) Cox regression were performed to identify the impact of HVF and AC on overall survival (OS) when compared to LVF and NAC, respectively. HVF was defined as the top 5% of facilities by number of pts treated. Results: From 2004 – 2014, a total of 10139 pts were identified. The median age was 66 years (range 22-90) with median follow up of 48.8 months (46-52.1 months); 49.9% were male and 50.1% female. All pts had clinical stage 3/T4 disease irrespective of nodal metastases. Of these, 4779 pts were treated at an AC and 5260 were treated at a NAC and 588 were treated at HVF and 9551 were treated at LVF. On UVA, age, high median income, high education level, comorbidities, and recent year of diagnosis were associated with improved OS. ACs were associated with improved OS when compared to non-AC (HR 0.92 95% CI 0.88 – 0.96, p = 0.004), as were HVF when compared to LVH (HR 0.84 95% CI 0.76 – 0.92, p < 0.001). Odds ratio for undergoing surgical resection at HVF and AC was 1.68 and 1.37 (p < 0.001), respectively, when compared to LVF and NAC. Conclusions: The treatment of LAPC patients with CT or CRT at an AC led to significantly improved rate of surgical resection and OS. In the absence of prospective data, these results support the referral of pts with LAPC to HVF and/or AC for evaluation and treatment.