Four‐Dimensional Machine Learning Radiomics for the Pretreatment Assessment of Breast Cancer Pathologic Complete Response to Neoadjuvant Chemotherapy in Dynamic Contrast‐Enhanced MRI

Abstract

Background: Breast cancer response to neoadjuvant chemotherapy (NAC) is typically evaluated through the assessment of tumor size reduction after a few cycles of NAC. In case of treatment ineffectiveness, this results in the patient suffering potentially severe secondary effects without achieving any actual benefit. Purpose: To identify patients achieving pathologic complete response (pCR) after NAC by spatio-temporal radiomic analysis of dynamic contrast-enhanced (DCE) MRI images acquired before treatment. Study type: Single-center, retrospective. Population: A total of 251 DCE-MRI pretreatment images of breast cancer patients. Field strength/sequence: 1.5 T/3 T, T1-weighted DCE-MRI. Assessment: Tumor and peritumoral regions were segmented, and 348 radiomic features that quantify texture temporal variation, enhancement kinetics heterogeneity, and morphology were extracted. Based on subsets of features identified through forward selection, machine learning (ML) logistic regression models were trained separately with all images and stratifying on cancer molecular subtype and validated with leave-one-out cross-validation. Statistical tests: Feature significance was assessed using the Mann-Whitney U-test. Significance of the area under the receiver operating characteristics (ROC) curve (AUC) of the ML models was assessed using the associated 95% confidence interval (CI). Significance threshold was set to 0.05, adjusted with Bonferroni correction. Results: Nine features related to texture temporal variation and enhancement kinetics heterogeneity were significant in the discrimination of cases achieving pCR vs. non-pCR. The ML models achieved significant AUC of 0.707 (all cancers, n = 251, 59 pCR), 0.824 (luminal A, n = 107, 14 pCR), 0.823 (luminal B, n = 47, 15 pCR), 0.844 (HER2 enriched, n = 25, 11 pCR), 0.803 (triple negative, n = 72, 19 pCR). Data Conclusions: Differences in imaging phenotypes were found between complete and noncomplete responders. Furthermore, ML models trained per cancer subtype achieved high performance in classifying pCR vs. non-pCR cases. They may, therefore, have potential to help stratify patients according to the level of response predicted before treatment, pending further validation with larger prospective cohorts.

Pre- to post-contrast breast MRI synthesis for enhanced tumour segmentation

“…The reliance on whole-image quality metrics presents a limitation in the context of tasks focusing solely on the tumour region such as radiomicsbased tumour treatment response prediction. 3 By integrating tumour region-specific image quality metrics into evaluation frameworks, such as the herein presented Scaled Aggregate Measure (SAMe), the utility and quality of synthetic contrast injection can be complementarily analysed and correlated with clinical downstream task performance. Similarly, extending 2D-based image quality metrics to 3D can provide additional insights on quality and usefulness of synthetic DCE-MRI.…”

“…2 Beyond screening, DCE-MRI finds widespread use in breast cancer diagnosis and treatment, serving vital roles in monitoring, preoperative planning, treatment and neoadjuvant therapy response assessment, where the radiological response is asssessed through lesion size regress/progress. 3,4 However, the administration of gadolinium-based contrast agents is associated with a range of adverse risks and side effects. These include the deposition of residual substances and their bioaccumulation with unclear clinical significance and long-term consequences, [5][6][7][8][9] as well as the increased potential for nephrogenic systemic fibrosis.…”

An MRI‐based machine learning radiomics can predict short‐term response to neoadjuvant chemotherapy in patients with cervical squamous cell carcinoma: A multicenter study

“…Tian et al 29 extracted radiomic features from 277 patients receiving NACT for LACC, and the results indicated that the radiomics features could be used as effective predictors to help patients perform risk stratification and improve the selection of NACT. Marco et al 42 evaluated pathological complete responses to NACT in breast cancer using a machine learning radiomics approach in dynamic enhanced MRI. However, the clinical relevance of these studies is limited due to the relatively small sample size and the lack of validation in a multicenter collaborative setting.…”

Next generation immuno-oncology tumor profiling using a rapid, non-invasive, computational biophysics biomarker in early-stage breast cancer

“…Immunohistochemistry (IHC), although widely used to assess PD-L1 pathology, is nevertheless highly variable and dependent on experimental methodology (Patel and Kurzrock, 2015 ). An opportune moment therefore exists for employing innovative approaches incorporating mathematical modeling (Caballo et al, 2022 ; Howard et al, 2022 ), with the intent to develop novel biomarkers to determine tumor IO responsiveness, with a forthcoming need in ESBC (Franzoi et al, 2021 ).…”