Systemic Therapy with Stereotactic Body Radiation Therapy (SBRT) for Early-Stage Non-Small Cell Lung Carcinoma (NSCLC): A Multi-institutional Analysis

Abstract

For patients with early-stage NSCLC, SBRT generally yields favorable local tumor control, but regional and distant failures occur in up to 20-30% of patients. Although adjuvant systemic therapy (ST) is often recommended for early-stage NSCLC with large tumors after surgery, there is a lack of evidence supporting the use of ST with SBRT. We used a multi-institutional database to evaluate the association between ST use and disease and survival outcomes for patients with early-stage NSCLC treated with SBRT. We conducted a retrospective cohort study using a multi-institutional, academic-community practice database, including consecutive patients with biopsy-proven T1-3N0M0 NSCLC treated with definitive SBRT from 2006 – 2015 at 114 sites. Patient cohorts were defined as those who received SBRT+ST or SBRT alone. Group characteristics were compared with Chi-square, Wilcoxon rank-sum, and logistic regression. Local, regional, and distant failure (LF, RF, and DF) were analyzed with multivariable competing risks regression with Fine and Gray’s proportional subhazard models. Progression-free and overall survival were analyzed with the Kaplan-Meier method and Cox regression. Competing risks regression was performed with 2:1 nearest-neighbor propensity score matching on clinical risk factors. Of 1,328 patients included, 54 (4.1%) received SBRT+ST. The most common ST regimen was a platinum doublet (n=38; 70.4%), followed by erlotinib (n=8; 14.8%), single agent chemotherapy (n=3; 5.6%), or unknown regimen (n=5; 9.3%). Compared with SBRT patients, SBRT+ST patients were younger (median age: 71 vs 78, P < 0.001), had larger tumors (>2 cm: 38.9% vs 21.5%, P = 0.003), and higher T-stage (T2-3: 42.6% vs 22.4%, P = 0.001). Median follow-up in living patients was 24 months. Compared with the SBRT cohort, the SBRT+ST cohort had significantly less DF (3.7% vs 13.0%, P = 0.04) and RF (0% vs 10.4%, P = 0.01), but not LF (7.4% vs 10.4%, P = 0.48). On multivariable analyses, SBRT+ST was independently associated with reduced DF (HR: 0.22, 95% CI: 0.05 - 0.88, P = 0.03) and overall failure (HR: 0.34, 95% CI: 0.15 - 0.76, P = 0.009) (RF not evaluable, as there were no events in the SBRT+ST cohort), with trend for improved progression-free (HR: 0.72, 95% CI: 0.49 – 1.06, P = 0.09), but not overall survival (HR: 0.78, 95% CI: 0.52 – 1.18, P = 0.24). After propensity score matching on size, T-stage, performance and smoking status, histology, and age, the SBRT+ST cohort had reduced DF, RF, and overall failure (each P < 0.05). This multi-institutional study shows improved regional, distant, and overall disease control in patients receiving adjuvant systemic therapy with SBRT for early stage NSCLC, despite the increased prevalence of larger and higher-stage tumors in this cohort. Prospective study is being planned to evaluate this hypothesis in high-risk subgroups.