Multi-Facility vs. Single-Facility Concurrent Chemoradiation for Lung Cancer

Abstract

Coordination of definitive concurrent chemoradiotherapy (CCRT) for locally advanced lung cancer can be challenging, especially if the therapies are delivered at different facilities. We hypothesized that patients with non-small cell lung cancer (NSCLC) or small cell lung cancer (SCLC) who undergo chemotherapy and radiotherapy (RT) in multiple facilities (MF) compared to a single facility (SF) will have different sociodemographic/clinical characteristics but similar overall survival (OS).We examined 2 cohorts of patients diagnosed in 2010-16 in the National Cancer Database: (1) stage III(N2+) NSCLC patients who received CCRT (60-70 Gy in 1.8-2 Gy fractions) and (2) stage III SCLC patients who received CCRT (45 Gy in 30 fractions or 60-70 Gy in 30-35 fractions). CCRT was defined as definitive multi-agent chemotherapy and external beam RT initiated within 30 days of each other. All patients received their entire course of RT at the reporting facility. Patients who received surgery or immunotherapy were excluded. We used univariable (UVA) and multivariable (MVA) analyses to assess potential associations of MF treatment with sociodemographic/clinical factors and OS. Odds ratios (OR) and hazard ratios (HR) with 95% confidence intervals (CI) are reported. Subset analyses were performed to evaluate the impact of diagnosis year (< 2014 vs. ≥2014) and facility volume (low vs. high, defined by 75th percentile) on OS in both cohorts.Our NSCLC cohort included 9279 patients; 32.4% received MF treatment. Our SCLC cohort included 2683 patients; 31.4% received MF treatment. Characteristics significantly associated with MF treatment on MVA for both cohorts included higher median household income, earlier year of diagnosis, absence of comorbidities, treatment at a community cancer program, and treatment at a facility in the South or Midwest. NSCLC patients receiving MF treatment were less likely to travel > 20 miles for RT (OR 0.80 [CI 0.70-0.90], P < 0.001) on UVA, while no such association was found in SCLC. For NSCLC, we found no significant difference in OS between patients who received MF vs. SF treatment on UVA (23.6 [CI 22.3-24.9] vs. 23.4 [CI 22.6-24.3] months, P = 0.55) or MVA (HR 1.00 [CI 0.95-1.05], P = 0.99). For SCLC, we similarly found no significant difference in OS between patients who received MF vs. SF treatment on UVA (21.7 [CI 20.4-24.0] vs. 22.1 [CI 21.2-23.4] months, P = 0.98) or MVA (HR 0.96 [CI 0.87-1.06], P = 0.44). Subset analyses also yielded no significant differences.Receiving definitive CCRT in MF vs. SF does not appear to significantly affect OS in locally advanced NSCLC or SCLC, although NSCLC patients receiving MF care are less likely to travel farther for RT. While outcomes of local control and toxicity have yet to be explored, traveling longer distances to receive lung cancer CCRT at the same facility may not be necessary to optimize survival.