Dosimetric predictors of patient-reported toxicity after prostate stereotactic body radiotherapy: Analysis of full range of the dose–volume histogram using ensemble machine learning. (July 2020)

• Record Type: Journal Article
• Authors: Pan, Xiaoying; Levin-Epstein, Rebecca; Huang, Jiahao; Ruan, Dan; King, Christopher R.; Kishan, Amar U.; Steinberg, Michael L.; Qi, X. Sharon
• Is Part Of: Radiotherapy and oncology. Volume 148(2020)
• Page Start: 181
• Page End: 188
• Publication Date: 2020-07
• Subjects: Ensemble machine learning -- Prostate stereotactic body radiotherapy (SBRT) -- Patient-reported outcomes -- Health-related quality of life (HRQOL) -- Toxicity -- Outcome-driven treatment planning; 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.2020.04.013
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: Machine learning methods were used to assess dosimetric associations with patient-reported outcomes. Full range of dosimetric quantities demonstrated predictive ability for patient-reported toxicities after prostate SBRT. Dosimetric predictors with patient-reported toxicities were identified using advanced ML methods. Predictive DVH metrics can be employed to refine planning guidelines to further reduce toxicity. Outcome-driven treatment planning may further improve treatment outcome. Abstract: Background and purpose: This study aims to evaluate the associations between dosimetric parameters and patient-reported outcomes, and to identify latent dosimetric parameters that most correlate with acute and subacute patient-reported urinary and rectal toxicity after prostate stereotactic body radiotherapy (SBRT) using machine learning methods. Materials and methods: Eighty-six patients who underwent prostate SBRT (40 Gy in 5 fractions) were included. Patient-reported health-related quality of life (HRQOL) outcomes were derived from bowel and bladder symptom scores on the Expanded Prostate Cancer Index Composite (EPIC-26) at 3 and 12 months post-SBRT. We utilized ensemble machine learning (ML) to interrogate the entire dose–volume histogram (DVH) to evaluate relationships between dose–volume parameters and HRQOL changes. The latent predictive dosimetric parameters that were most associated with HRQOL changes in urinary and rectal function were thus identified. An external cohort of 26 prostate SBRT patients was acquired to further test the predictive models. Results: Bladder dose–volume metrics strongly predicted patient-reported urinary irritative and incontinence symptoms (area under the curves [AUCs] of 0.79 and 0.87, respectively) at 12 months. Maximum bladder dose, bladder V102.5%, bladder volume, and conformity indices (V50/VPTV and V100/VPTV) were most predictive of HRQOL changes in both urinary domains. No strong rectal toxicity dosimetric association was identified (AUC = 0.64). Conclusion: We demonstrated the application of advanced ML methods to identify a set of dosimetric variables that most highly correlated with patient-reported urinary HRQOL. DVH quantities identified with these methods may be used to achieve outcome-driven planning objectives to further reduce patient-reported toxicity with prostate SBRT.