A priori prediction of tumour response to neoadjuvant chemotherapy in breast cancer patients using quantitative CT and machine learning

Abstract

Response to neoadjuvant chemotherapy (nAc) has demonstrated a high correlation to survival in locally advanced breast cancer (LABC) patients. An early prediction of responsiveness to NAC could facilitate treatment adjustments on an individual patient basis that would be expected to improve treatment outcomes and patient survival. This study investigated, for the first time, the efficacy of quantitative computed tomography (qct) parametric imaging to characterize intra-tumour heterogeneity and its application in predicting tumour response to NAC in LABC patients. Textural analyses were performed on CT images acquired from 72 patients before the start of chemotherapy to determine quantitative features of intra-tumour heterogeneity. the best feature subset for response prediction was selected through a sequential feature selection with bootstrap 0.632 + area under the receiver operating characteristic (Roc) curve (AUC 0.632+) as a performance criterion. Several classifiers were evaluated for response prediction using the selected feature subset. Amongst the applied classifiers an Adaboost decision tree provided the best results with cross-validated AUC 0.632+ , accuracy, sensitivity and specificity of 0.89, 84%, 80% and 88%, respectively. The promising results obtained in this study demonstrate the potential of the proposed biomarkers to be used as predictors of LABC tumour response to NAC prior to the start of treatment.

A priori prediction of response in multicentre locally advanced breast cancer (LABC) patients using quantitative ultrasound and derivative texture methods

“…In their study, univariate parameter of homogeneity of HbO 2 map achieved the best 86% sensitivity, 89% specificity, and 88% accuracy from 37 LABC patients undergoing NAC. Recently, Moghadas-Dasterdji et al used quantitative textural features extracted from computerized tomography (CT) images for developing a priori response predictive model using different machine learning classification algorithms [ 56 ]. Their study included 72 LABC patients undergoing NAC and obtained the best classification results of 80% sensitivity, 88% specificity, 84% accuracy, and 0.89 AUC 0.632+ using an Adaboost decision tree (DT) classification algorithm [ 56 ].…”

“…Recently, Moghadas-Dasterdji et al used quantitative textural features extracted from computerized tomography (CT) images for developing a priori response predictive model using different machine learning classification algorithms [ 56 ]. Their study included 72 LABC patients undergoing NAC and obtained the best classification results of 80% sensitivity, 88% specificity, 84% accuracy, and 0.89 AUC 0.632+ using an Adaboost decision tree (DT) classification algorithm [ 56 ]. The response prediction results of these different imaging modalities are comparable with the results presented in this study.…”

MRI texture features from tumor core and margin in the prediction of response to neoadjuvant chemotherapy in patients with locally advanced breast cancer

“…The response was classified using a modified RECIST score, using a combination of post-NAC imaging and histopathological specimen, which served as the gold standard [ 26 ]. Patients were classified into binary treatment response groups: responders and non-responders.…”

Multi-Dimensional Scaling Analysis of Key Regulatory Genes in Prostate Cancer Using the TCGA Database

“…Even when studies revealed a large preponderance of specific clusters, such as a molecular taxonomy in which 74% of tumors falling into one of seven subtypes defined by specific gene fusions or mutations [ 20 ], patient stratification and outcome remain elusive [ 21 ]. Many challenges remain to integrate epidemiological studies with molecular investigations and clinical analyses to gain fundamental insights into how environmental, dietary, and lifestyle influences contribute to the development of PC, and thus a priori models selecting target genes involved in macroscopic functions could shed new light on this complex disease, as previous studies on different types of cancers have revealed [ 22 ].…”