Dose Escalation after RTOG 0617: Recent Trends and Predictors of Radiation Doses >70 Gy in Stage III Non-Small Cell Lung Cancer from the National Cancer Database

Abstract

Unresectable stage III non-small cell lung cancer (NSCLC) is treated with definitive chemoradiotherapy with high rates of local recurrence. RTOG 0617 was a tumor dose-escalation (DE) trial that unexpectedly reported inferior overall survival with increasing radiation (RT) dose. The DE arm was closed in 2011 and final results were published in 2015. National Comprehensive Cancer Center guidelines currently suggest doses of 60 to 70 Gy. The purpose of this study was to assess the recent use of DE RT in the US. We identified patients in the National Cancer Database (NCDB) with unresectable clinical stage III NSCLC undergoing definitive chemoradiotherapy (multi-agent chemotherapy concurrent with RT to doses ≥ 60 Gy but < 80 Gy) between diagnosis years 2004 and 2016. RT dose (in conventional fractionation) was defined as either standard dose (SD; 60-70 Gy) or dose escalated (DE; > 70 Gy) and analyzed over time using Pearson’s Chi-squared test. Confidence intervals (CI) around the estimated annual percentage of patients treated with DE RT were calculated using the Clopper-Pearson method. Predictors of DE RT receipt were determined using Chi-squared tests and multivariable logistic regression analysis. The Kaplan-Meier method was used to estimate median survival (MS). From 2004 to 2010 the percentage of patients receiving DE RT increased steadily from 7.1% (95% CI 6.0-8.2%) to 12.2% (95% CI 11.0-13.3%). Subsequently, this number declined from 10.2% (95% CI 9.2-11.2%) in 2011 to 1.7% (95% CI 1.3-2.0%) in 2016 (p < 0.001). In 2016 (n = 4,725 patients), the most common radiation dose was 60 Gy (50.9%) with 66 Gy second most common (17.7%). Only 7.9% of patients received a dose greater than 66.6 Gy. On univariable analysis, receipt of DE RT in the era from 2012 to 2016 was associated with treatment at non-academic centers (4.1% vs 1.9% academic; p < 0.001), white race (vs. other; p = 0.040), and cT3-4 (vs. cT1-2; p = 0.001). Private insurance/Medicare (vs. Medicaid/uninsured; p = 0.082), Charlson-Deyo score ≥ 1 (vs. 0; p = 0.153), upper lobe primary (vs. middle or lower; p = 0.124), and cN0-1 (vs. cN2-3; p = 0.098) trended towards significance in predicting DE RT and were included in the multivariable model. Variables remaining significant for predicting DE RT receipt on multivariable logistic regression (correcting for trend over diagnosis year) included non-academic center (OR 2.38, 95% CI 1.91-2.96, p < 0.001) and cT3-4 (OR 1.28, 95% CI 1.09-1.50, p = 0.003). After a median follow up of 36.2 months, the median survival was 22.1 months in the SD group and 19.7 months in the DE group (log-rank, p = 0.090). Since the closure of the RTOG 0617 DE RT arm in 2011, the use of doses over 70 Gy has progressively declined and 60 Gy in conventional fractionation has become most commonly used. DE RT is rare in the modern era but more likely used for tumors with advanced T stage or at non-academic centers.