Machine-learning prediction model for acute skin toxicity after breast radiation therapy using spectrophotometry

Abstract

PurposeRadiation-induced skin toxicity is a common and distressing side effect of breast radiation therapy (RT). We investigated the use of quantitative spectrophotometric markers as input parameters in supervised machine learning models to develop a predictive model for acute radiation toxicity.Methods and materialsOne hundred twenty-nine patients treated for adjuvant whole-breast radiotherapy were evaluated. Two spectrophotometer variables, i.e. the melanin (IM) and erythema (IE) indices, were used to quantitatively assess the skin physical changes. Measurements were performed at 4-time intervals: before RT, at the end of RT and 1 and 6 months after the end of RT. Together with clinical covariates, melanin and erythema indices were correlated with skin toxicity, evaluated using the Radiation Therapy Oncology Group (RTOG) guidelines. Binary group classes were labeled according to a RTOG cut-off score of ≥ 2. The patient’s dataset was randomly split into a training and testing set used for model development/validation and testing (75%/25% split). A 5-times repeated holdout cross-validation was performed. Three supervised machine learning models, including support vector machine (SVM), classification and regression tree analysis (CART) and logistic regression (LR), were employed for modeling and skin toxicity prediction purposes.ResultsThirty-four (26.4%) patients presented with adverse skin effects (RTOG ≥2) at the end of treatment. The two spectrophotometric variables at the beginning of RT (IM,T0 and IE,T0), together with the volumes of breast (PTV2) and boost surgical cavity (PTV1), the body mass index (BMI) and the dose fractionation scheme (FRAC) were found significantly associated with the RTOG score groups (p<0.05) in univariate analysis. The diagnostic performances measured by the area-under-curve (AUC) were 0.816, 0.734, 0.714, 0.691 and 0.664 for IM, IE, PTV2, PTV1 and BMI, respectively. Classification performances reported precision, recall and F1-values greater than 0.8 for all models. The SVM classifier using the RBF kernel had the best performance, with accuracy, precision, recall and F-score equal to 89.8%, 88.7%, 98.6% and 93.3%, respectively. CART analysis classified patients with IM,T0 ≥ 99 to be associated with RTOG ≥ 2 toxicity; subsequently, PTV1 and PTV2 played a significant role in increasing the classification rate. The CART model provided a very high diagnostic performance of AUC=0.959.ConclusionsSpectrophotometry is an objective and reliable tool able to assess radiation induced skin tissue injury. Using a machine learning approach, we were able to predict grade RTOG ≥2 skin toxicity in patients undergoing breast RT. This approach may prove useful for treatment management aiming to improve patient quality of life.

Comparing Performances of Predictive Models of Toxicity after Radiotherapy for Breast Cancer Using Different Machine Learning Approaches

“…Furthermore, some important variables are missing. In order to obtain better-performing models (say, AUC > 0.80), the availability of other potential predictors would be highly relevant: in the case of toxicity after breast WBI, more accurate DVH information, information on the genotype and/or the phenotype, densitometry characteristics of the breast may all importantly contribute to improving the accuracy of the resulting models, as seldom reported [ 13 , 23 , 24 , 25 ]. The availability of more individually characterizing features (here in part missing) is expected to have a likely much higher impact than the choice of the best-performing ML/DL approach.…”

“…To our knowledge, few studies comparing different modeling approaches have been conducted on breast toxicity. They are always on limited-size populations and/or consider only a few ML approaches [ 22 , 23 , 24 , 25 ]. In particular, these studies were performed on different outcomes (acute toxicity [ 22 , 25 ], acute desquamation [ 23 ], radiation-induced dermatitis [ 24 ]).…”

“…They are always on limited-size populations and/or consider only a few ML approaches [ 22 , 23 , 24 , 25 ]. In particular, these studies were performed on different outcomes (acute toxicity [ 22 , 25 ], acute desquamation [ 23 ], radiation-induced dermatitis [ 24 ]). The input variables used are diverse, going from spectrophotometric variables [ 25 ], baseline characteristics, comorbidities [ 23 ], and clinical variables [ 22 ], to radiomics features [ 24 ].…”

“…In particular, these studies were performed on different outcomes (acute toxicity [ 22 , 25 ], acute desquamation [ 23 ], radiation-induced dermatitis [ 24 ]). The input variables used are diverse, going from spectrophotometric variables [ 25 ], baseline characteristics, comorbidities [ 23 ], and clinical variables [ 22 ], to radiomics features [ 24 ]. Some considered small [ 24 , 25 ] or large [ 22 , 23 ] cohorts.…”

“…The input variables used are diverse, going from spectrophotometric variables [ 25 ], baseline characteristics, comorbidities [ 23 ], and clinical variables [ 22 ], to radiomics features [ 24 ]. Some considered small [ 24 , 25 ] or large [ 22 , 23 ] cohorts. Their performances differ depending on the features used.…”