Utilization and Survival Impact of Definitive Hypofractionated Radiotherapy Compared to Conventionally Fractionated Radiotherapy in Stage I Non-Small Cell Lung Cancer

Abstract

The real-world impact of definitive hypofractionated radiotherapy (HFRT) is unclear for patients with stage I non-small cell lung cancer (NSCLC) not suitable for surgery, stereotactic body radiation therapy (SBRT), or chemoradiotherapy. HFRT is more convenient, is less expensive, and is amenable to BED10 (biologically effective dose assuming α/β=10) escalation compared to conventionally fractionated radiotherapy (CFRT). We tested our hypotheses that HFRT is being utilized differently among various subpopulations and is associated with similar overall survival (OS) compared to CFRT. This retrospective cohort analysis included patients in the National Cancer Database diagnosed with primary stage I (cT1-2aN0M0) NSCLC in 2008-2016 who underwent CFRT (70≤BED10<100 in ≥30 fractions), lower-dose HFRT (LD-HFRT, 70≤BED10<100 in 6-20 fractions), or higher-dose HFRT (HD-HFRT, 100≤BED10≤120 in 6-20 fractions). Patients who received surgery, SBRT in ≤5 fractions, chemotherapy, or immunotherapy were excluded, as were those who had radiotherapy at a location other than the reporting facility. Chi-square, ANOVA, and multivariable logistic regression (MVA) were used to assess associations among sociodemographic and clinicopathologic variables with fractionation. Overall survival (OS) was evaluated with Kaplan-Meier estimator, log-rank test, and multivariable Cox proportional hazards regression modeling. Our analysis included 3,515 patients with stage I NSCLC, among whom 2,510 (71%) received CFRT, 760 (22%) received LD-HFRT, and 245 (7%) received HD-HFRT. Median age was 75 years and 52.6% were female. On MVA, variables associated with receiving HFRT compared to CFRT included age>75 (odds ratio [OR] 1.43, 95% confidence interval [CI] 1.21-1.69, p<0.001), T1 disease (OR 1.67, CI 1.42-1.97, p<0.001), distance from residence to treatment facility ≥8 miles (OR 1.34, CI 1.13-1.59, p=0.001), treatment at an academic facility (OR 3.80, CI 3.14-4.61, p<0.001), and diagnosis year ≥2011 (OR 5.11, CI 4.22-6.20, p<0.001). HFRT was significantly associated with longer OS compared to CFRT on univariable analysis (median 29.4 [CI 26.6-32.7] vs. 27.2 [CI 25.8-28.4] months, log-rank p=0.03), but not on multivariable analysis (hazard ratio [HR] 0.93, CI 0.85-1.03, p=0.16). Multivariable results were similar when comparing LD-HFRT to CFRT (HR 0.92, CI 0.83-1.02, p=0.13) and HD-HFRT to CFRT (HR 0.93, CI 0.78-1.10, p=0.38). For stage I NSCLC, HFRT is increasing in utilization over time, is being employed differently among various subpopulations, and is associated with comparable OS compared to CFRT. These results suggest that HFRT may be a reasonable option for those who are not candidates for surgery, SBRT, or chemoradiotherapy. Given the inherent limitations of this retrospective hypothesis-generating study, including potential selection biases and a lack of data on comparative toxicities and tumor control, further investigation is needed.