Improving outcomes in non-small cell lung cancer; population analysis of radical radiotherapy.

Abstract

Regional utilisation of radical radiotherapy (RT) in non-small cell lung cancer (NSCLC) was used to define optimal utilisation to improve outcome and as a surrogate for evidence of RT efficacy. 65,412 NSCLC cases diagnosed in England 2012-13 were linked to comprehensive national radiotherapy dataset, hospital admissions and the Office of National Statistics. Geographical variation in utilisation was determined using a multivariate binary logistic regression analysis after adjusting for age, stage, deprivation, comorbidity and other radical treatment and the effect of radical RT utilisation on survival was investigated. Survival was adjusted for dependent and independent variables and the effect of differing levels of utilisation was assessed by the log likelihood test. 17.6% cases potentially eligible for radical RT (stages 0-III) received radiotherapy with radical intent. Utilisation of radical RT had an impact on survival (p < 0.00001). Adjusting for all prognostic and treatment variables counties with lowest utilisation (≤15%) had the worst survival (HR = 1.13). The highest utilisation quintile counties (≥25%) had worse survival compared to counties with lower utilisation (≈20%) (p < 0.0001). Analysis of stages II&III showed the same pattern; increase in utilisation from 20% to ≥25% resulting in a 3% drop in 2-year population survival (p = 0.001). The utilisation of radical RT has a significant impact on NSCLC population survival. Improvement in survival of NSCLC population can be achieved by offering radical RT to a larger proportion of patients while avoiding excessive use. Geographical variation in RT utilisation provides indirect evidence of survival benefit of radical radiotherapy.