P2.02-015 Guideline Concordant Care is Associated with Better Survival for Patients with Stage III Non-Small Cell Lung Cancer

Abstract

Current evidence-based guideline-concordant care (GCC) is administration of platinum-based chemotherapy during thoracic radiotherapy (TRT) for locally advanced non-small cell lung cancer (NSCLC) patients with good performance status. This study evaluates factors associated with lack of GCC. Patients (pts) with unresected stage IIIA/IIIB NSCLC diagnosed from 2005 – 2013 and Charlson-Deyo Score 0 were identified from the National Cancer Data Base (NCDB). Primary outcomes measured were receipt of GCC, defined by administration of chemotherapy with TRT commencing within 2 weeks of each other and minimum TRT dose of 60 Gy, and overall survival (OS). Multivariable logistic regression (MLR) modeling was performed to identify variables associated with non-GCC. Cox proportional hazard modeling was utilized to examine OS. Patient characteristics (n=37,809) included: mean age 67.8 years; 55% male; 13% African American; 3.4% Hispanic, 3.6% ‘other’ race/ethnicity; 66% government-insured; mean tumor size 5.0 cm; 38% adenocarcinoma; 32% squamous cell carcinoma (SCC); 30% large cell/other histology. In total, 28% of pts received GCC. On MLR analysis, Hispanic pts were more likely to receive non-GCC (OR=1.34, p <0.001) compared to non-Hispanic pts. Uninsured pts were more likely to receive non-GCC (OR=1.57, p<0.001) compared to privately-insured pts. Patients treated in the western, southern, or northeastern U.S. were more likely to receive non-GCC (OR= 1.43, 1.45, 1.21, all p values <0.001) compared to pts treated in the Midwest. Adenocarcinoma and large-cell/other histological types were more likely to receive non-GCC (OR= 1.71, 1.39, both p<0.001) compared to SCC. For every one-year increase in age or 50-mile increase in distance to treatment facility, patients had a 4% or 3% increased odds of not receiving GCC (OR=1.04, 1.03; p<0.001, p = 0.003, respectively). On hazard modeling, those receiving non-GCC had higher death rates compared to those receiving GCC (HR=1.42, p<0.001). Survival rates were lower for Hispanics receiving non-GCC versus GCC (HR=1.24, p=0.034). Other groups with lower OS for non-GCC versus GCC included: the uninsured (HR=1.61, p<0.001), treatment in the western, southern, or northeastern US (HRs= 1.56, 1.40, 1.33, respectively, p<0.001), adenocarcinomas and large cell/other histologies (both HR=1.40, p<0.001). Socioeconomic factors, including Hispanic ethnicity, lack of insurance, geographic location, and distance from treatment facility are associated with receipt of non-GCC. Patient and disease specific factors including increasing age and adenocarcinoma histology are also associated with non-GCC. Future interventions could target these groups to improve provision of GCC.