Search

Simple Summary: Studies comparing performances of machine learning (ML) methods in building predictive models of toxicity in RT are rare. Thanks to the availability of a large cohort (n = 1314) of breast cancer patients homogeneously treated with tangential fields, different ML approaches could be compared. This work shows how more complex models typically achieve higher performances. At the same time, for this test case, the importance is given mainly by a few variables, and toxicity can be predicted by simpler models with similar performances. The availability of more individually characterizing features (here partially missing) is expected to have a likely much higher impact than the choice of the best-performing ML/DL approach. Purpose. Different ML models were compared to predict toxicity in RT on a large cohort (n = 1314). Methods. The endpoint was RTOG G2/G3 acute toxicity, resulting in 204/1314 patients with the event. The dataset, including 25 clinical, anatomical, and dosimetric features, was split into 984 for training and 330 for internal tests. The dataset was standardized; features with a high p-value at univariate LR and with Spearman ρ > 0.8 were excluded; synthesized data of the minority were generated to compensate for class imbalance. Twelve ML methods were considered. Model optimization and sequential backward selection were run to choose the best models with a parsimonious feature number. Finally, feature importance was derived for every model. Results. The model's performance was compared on a training–test dataset over different metrics: the best performance model was LightGBM. Logistic regression with three variables (LR3) selected via bootstrapping showed performances similar to the best-performing models. The AUC of test data is slightly above 0.65 for the best models (highest value: 0.662 with LightGBM). Conclusions. No model performed the best for all metrics: more complex ML models had better performances; however, models with just three features showed performances comparable to the best models using many (n = 13–19) features. [ABSTRACT FROM AUTHOR]

To evaluate the persistence of symptoms after radiotherapy (RT) for localised prostate cancer (PCa) and the association with quality of life (QOL). Prospective patient-reported outcome (PRO) from a multi-institutional study on PCa treated with radical RT (2010–2014) was analysed. Data was collected at baseline (BL) and follow-ups (FUPs) up to 5 years. Patients with BL and ≥3 late FUPs (≥6 months) were analysed. PRO was scored by means of the IPSS and ICIQ-SF (urinary), LENT-SOMA (gastrointestinal [GI]), and EORTC-C30 (pain, insomnia, fatigue, and QOL) questionnaires. Symptoms were defined 'persistent' if the median score over FUPs was ≥3 (urinary) or ≥2 (GI, pain, insomnia, and fatigue), and worse than BL. Different thresholds were chosen to have enough events for each symptom. QOL was linearly transformed on a continuous scale (0–100). Linear-mixed models were used to identify significant differences between groups with and without persistent symptoms including age, smoking status, previous abdominal surgery, and diabetes as confounders. Mean QOL differences between groups were evaluated longitudinally over FUPs. The analysis included 293 patients. Persistent urinary symptoms ranged from 2% (straining) to 12% (weak stream, and nocturia). Gastrointestinal symptoms ranged from 7% (rectal pain, and incontinence) to 30% (urgency). Proportions of pain, insomnia, and fatigue were 6, 13, and 18%. Significant QOL differences of small-to-medium clinical relevance were found for urinary incontinence, frequency, urgency, and nocturia. Among GI symptoms, rectal pain and incontinence showed small-to-medium differences. Fatigue was associated with the largest differences. The analysis showed that symptoms after RT for PCa occur with different persistence and their association with QOL varies in magnitude. A number of persistent urinary and GI symptoms showed differences in a comparable range. Urinary incontinence and frequency, rectal pain, and faecal incontinence more often had significant associations. Fatigue was also prevalent and associated with largely deteriorated QOL. [ABSTRACT FROM AUTHOR]