Table 3.

Model performance for baseline features predicting distressing or intolerable side effects treatment outcome.

Model	AUC (95% CI)	Accuracy	Sensitivity	Specificity	R2
Clinical predictors	.618 (.587, .649)	.577	.541	.611	.030
Clinical predictors + elastic net SNPs	.608 (.577, .639)	.583	.547	.617	.024

Clinical predictors model includes sociodemographic and pre-treatment symptom variables only. The model in the second row adds the SNPs identified by the elastic net feature selection to the clinical predictors model. For threshold-dependent metrics (accuracy, sensitivity, specificity), a probability threshold of 0.5 was used for classification.