Predicting radiation pneumonitis in locally advanced stage II–III non-small cell lung cancer using machine learning. (April 2019)

• Record Type: Journal Article
• Authors: Luna, José Marcio; Chao, Hann-Hsiang; Diffenderfer, Eric S.; Valdes, Gilmer; Chinniah, Chidambaram; Ma, Grace; Cengel, Keith A.; Solberg, Timothy D.; Berman, Abigail T.; Simone, Charles B.
• Is Part Of: Radiotherapy and oncology. Volume 133(2019)
• Page Start: 106
• Page End: 112
• Publication Date: 2019-04
• Subjects: Radiation pneumonitis -- Non-small cell lung cancer -- Machine learning -- Random forest -- RUSBoost -- CART -- Support vector machines -- Logistic regression; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2019.01.003
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: Among an extensive set of 32 clinical and dosimetric features, Lung V20, mean lung dose, lung V10 and lung V5 are the best individual predictors of radiation pneumonitis in stage II–III LA-NSCLC. The combined predictive performance of radiation pneumonitis predictors such as maximum esophagus dose, lung V20, mean lung dose, pack-year, lung V5 and lung V10 improves the performance of individual predictors up to a 24.6% improvement rate using random forest. Lung V20, maximum esophagus dose and mean lung dose are consistently selected as the most important predictors of radiation pneumonitis by the machine learning algorithms, random forest, RUSBoost and CART. Abstract: Background and purpose: Radiation pneumonitis (RP) is a radiotherapy dose-limiting toxicity for locally advanced non-small cell lung cancer (LA-NSCLC). Prior studies have proposed relevant dosimetric constraints to limit this toxicity. Using machine learning algorithms, we performed analyses of contributing factors in the development of RP to uncover previously unidentified criteria and elucidate the relative importance of individual factors. Materials and methods: We evaluated 32 clinical features per patient in a cohort of 203 stage II–III LA-NSCLC patients treated with definitive chemoradiation to a median dose of 66.6 Gy in 1.8 Gy daily fractions at our institution from 2008 to 2016. Of this cohort, 17.7% of patients developed grade ≥2 RP. Univariate analysis was performed using trained decision stumps to individually analyze statistically significant predictors of RP and perform feature selection. Applying Random Forest, we performed multivariate analysis to assess the combined performance of important predictors of RP. Results: On univariate analysis, lung V20, lung mean, lung V10 and lung V5 were found to be significant RP predictors with the greatest balance of specificity and sensitivity. On multivariate analysis, Random Forest (AUC = 0.66, p = 0.0005) identified esophagus max (20.5%), lung V20 (16.4%), lung mean (15.7%) and pack-year (14.9%) as the most common primary differentiators of RP. Conclusions: We highlight Random Forest as an accurate machine learning method to identify known and new predictors of symptomatic RP. Furthermore, this analysis confirms the importance of lung V20, lung mean and pack-year as predictors of RP while also introducing esophagus max as an important RP predictor.