Radiation Dose, Technique and Use of Brain Radiation on Overall Survival in Patients with Limited-Stage Small Cell Lung Cancer

Abstract

<h3>Purpose/Objective(s)</h3> The standard of care for limited-stage small cell lung cancer (LS-SCLC) currently involves concurrent chemoradiation +/- prophylactic cranial irradiation (PCI). Previous research has presented conflicting data regarding the optimal radiation therapy (RT) technique, RT dosage, and impact of brain RT on long term overall survival (OS). This study aimed to explore prognostic factors of OS in LS-SCLC cancer patients using a population-based cohort. <h3>Materials/Methods</h3> Data on all patients diagnosed with LS-SCLC in Ontario, Canada from 2005-2017 was obtained using the Institute for Clinical Evaluative Sciences (IC/ES) Data and Analytic Services. Eligible patients received curative intent chemoradiation, and OS was estimated from a landmark date of 6 months post-diagnosis to avoid survivor bias. Kaplan-Meier methods and Cox regression were used to estimate OS and explore differences between RT dose (40-49, 50-59, 60+ Gy), RT technique (3DCRT, IMRT, VMAT), and use of brain RT (PCI – 25 Gy, whole brain RT – 20 or 30 Gy, other brain RT, no brain RT). Multivariate analyses adjusted for sex, age, income quintile, Charlson comorbidity index, disease stage, prior history of cardiac disease, brain RT dose, and laterality. <h3>Results</h3> A total of 1360 LS-SCLC patients who received chemoradiation were included. Median OS (95% CI) was 18.1 months (16.6-19.4). Higher RT dose had significantly improved OS when compared to 40-49 Gy on univariate analysis (HR=0.83, 0.71-0.95, 60+ Gy), however, this did not persist when adjusting for other prognostic factors on multivariate analysis (HR=0.90, 0.77-1.06, 60+ Gy). The use of whole brain RT was associated with worse OS on both univariate (HR=1.47, 1.24-1.74) and multivariate (HR=1.42, 1.18-1.70) analyses. OS (95% CI) at 5-years for RT dose were 40-49 Gy: 16% (13-19%), 50-59 Gy: 23% (18-29), and 60+ Gy: 20% (16-25). For RT technique, OS (95% CI) at 5-years were 3DCRT: 19% (13-22), IMRT: 18% (13-23), and VMAT: 12% (4-25). Multivariate analyses demonstrated that male sex (HR=1.36, 1.20-1.55), more advanced stage disease (HR=1.35/1.94/3.08/1.67 for stage 2/3/4/Unknown vs 1), and older age were prognostic for worse OS. <h3>Conclusion</h3> Analyses of population-based data demonstrated that male sex, older age, and more advanced disease stage were notable predictors of OS in LS-SCLC. RT dose, when adjusting for other important prognostic variables did not predict for improved OS. However, use of whole brain RT (versus PCI or no PCI) was associated with worse OS. Use of PCI versus no PCI did not influence survival in this cohort, nor did RT technique.