Association Between Radiation Therapy Dose and Survival in Patients With Stage I Non-Small Cell Lung Cancer Undergoing Stereotactic Body Radiation Therapy

Purpose/Objective(s): Although stereotactic body radiation therapy (SBRT) for early-stage non-small cell lung cancer (NSCLC) is associated with high local control rates, the optimal dose-fractionation regimen is unknown, as prospective and retrospective data argue for different dosing paradigms. Whether increasing stereotactic radiation therapy dose can compensate for larger tumor volume is an open question. We aimed to determine the comparative effectiveness of different SBRT dosing regimens using a large national database, focusing on the relative impact of dose as a function of tumor stage. Materials/Methods: Patients in the National Cancer Database with clinical T1-2 N0 NSCLC who were treated with SBRT were eligible for inclusion in this cohort. Patient survival data were only available for those with a minimum of 5 years of follow-up, and therefore the analysis was restricted to those diagnosed between 2003 and 2006. Median follow-up and overall survival (OS) were defined from the date of diagnosis to last contact. The biologically equivalent dose (BED) was calculated based on the linear quadratic formula using an alpha/beta ratio of 10. All patients received a minimum BED of 70 Gy. Highversus lower-dose (HD vs LD) SBRT was defined as BED above or below 150 Gy. Overall survival was estimated using Kaplan-Meier methods. Cox proportional hazard multivariable analysis (MVA) was used to estimate hazard ratios (HR), adjusting for age, sex, T stage, BED, facility type (academic vs community practice), insurance status, year of diagnosis, comorbidities, race, and histology. Results: A total of 498 patients (T1 Z 67%, T2 Z 33%) met inclusion criteria. The median follow-up was 3 years (range, 3 months-8.6 years) for the entire cohort and 5.6 years for surviving patients. The 5 most common dose fractionation schemes (% of cohort) utilized were 20 Gy x 3 (34%), 12 Gy x 4 (16%), 18 Gy x 3 (10%), 15 Gy x 3 (10%), and 10 Gy x 5 (2.6%). The median BED was 150 Gy (inter-quartile range 106 Gy 166 Gy). The 3and 5-year overall survival for the entire cohort was 50% and 30%. The 3 year OS for T1 vs T2 disease was 61% vs 30% (log rank p < 0.0001). The 3 year OS for patients who received HD vs LD was 55% vs 46% (log rank pZ 0.03). On subset analysis of the T1 cohort, there was no association between BED and 3-year OS (61% vs 60% with HD vs LD, p Z 0.9). Among the T2 cohort, patients receiving HD experienced superior 3 year OS (37% vs 24%, p Z 0.01). On MVA, the only factors independently prognostic for mortality were female gender (HR Z 0.76, p Z 0.01), T1 tumor (HR Z 0.51, p Z 0.0001), and HD (HR Z 0.78, p Z 0.01). Conclusions: This comparative effectiveness analysis of SBRT dose for patients with stage I NSCLC suggests higher doses are associated with a significant survival benefit, especially in patients with a larger burden of disease. Prospective data are needed to determine the optimal dose-fractionation regimen as a function of tumor volume. Author Disclosure: M. Koshy: None. R. Malik: None. D.J. Sher: None.