Higher Per-Center Utilization of Surgery Confers Superior Survival in Stage II NSCLC
David C Becerra, Soraya L Voigt, Michael S Mulvihill, Joshua Watson, Morgan L Cox, Babatunde A Yerokun, Paul J Speicher, Jacob A Klapper, Thomas A D'Amico, Matthew G Hartwig
Duke University, Durham, NC
BACKGROUND: Lobar resection is the gold standard therapy for patients with stage II non-small cell lung cancer (NSCLC). Consensus based guidelines recommend curative intent resection for this disease based on demonstrated superior survival. Despite this, significant variability exists in the utilization of surgery. This study tested the hypothesis that center-based variation in the rate of utilization of surgery impacts survival in early stage NSCLC. Disparities in the adherence to guideline-concordant delivery of surgical therapy may represent a modifiable behavior to improve oncologic outcomes.
METHODS: We queried the National Cancer Database for adult patients with clinical stage II NSCLC from 2004-2014. Patient, tumor, and center level characteristics were analyzed. Mixed-effects multivariable models were subsequently developed to establish the per-center adjusted rate of surgery. Patients were stratified into quartiles based on the treating center's adjusted rate of surgery. Survival was estimated using the Kaplan-Meier method and assessed using the log-rank test. Multivariable Cox proportional hazard models were then developed to estimate the effect of adjusted rates of surgery on overall survival after accounting for patient, tumor, and center-level factors.
RESULTS: A total of 54,758 patients met inclusion criteria. Of the entire cohort, 44.8% (N=24,569) of patients underwent surgical resection. Patients treated at centers with the highest rate of surgery were more likely to receive surgical resection (60.7% [N=11,057] in highest v 20.9% [N=1,795] in lowest quartile) and to be treated at a highest-volume center (68.1% [N=12,419] in highest v 34.9% [N=2,997] in lowest quartile) (both p<0.001). Across utilization cohorts, patients were similar in age (mean 68.2 years in highest versus 69.6 years in lowest) and Charleson-Deyo Comorbidity Score ≥2 (14.3% [N=2,604] in highest v 14.5% [N=1,247] in lowest quartile (p>0.05)). Five-year overall survival was significantly higher in those patients treated at centers with the highest utilization rate of surgery (31.9% [N=1,755 at risk] v 18.6% [N=545 at risk], p<0.001). Following adjustment for patient, tumor, and center level factors, an adjusted rate of surgery in the lowest 25th percentile was associated with lower survival (adjusted hazard ratio of 1.34, 95% CI 1.29-1.39, p<0.001).
CONCLUSIONS: Treatment at a center with a higher adjusted rate of surgery confers a significant survival advantage for patients with stage II NSCLC, even after adjustment for hospital volume, surgical approach, and other confounders. Targeted efforts to improve adherence to evidence-based guidelines for the provision of surgery in early stage NSCLC at centers with low rates of utilization of surgery may represent a meaningful opportunity to improve outcomes in this disease.
Back to 2018 Program