Artificial Intelligence in Breast MRI Radiogenomics: Towards Accurate Prediction of Neoadjuvant Chemotherapy Responses

A three-sequence dynamic contrast enhanced abbreviated MRI protocol to evaluate response to breast cancer neoadjuvant chemotherapy

PURPOSE

To develop an ABP-MRI to evaluate response to NAC for invasive breast carcinoma.

STUDY TYPE

A single-center, cross-sectional study.

SUBJECTS

A consecutive series of 210 women with invasive breast carcinoma who underwent breast MRI after NAC between 2016 and 2020.

FIELD STRENGTH/SEQUENCE

1.5 T / Dynamic contrast-enhanced.

ASSESSMENT

MRI scans were independently reevaluated, with access to dynamic contrast-enhanced without contrast and to the first, second, and third post-contrast time (ABP-MRI 1-3).

STATISTICAL TESTS

The diagnostic performance of the ABP-MRIs and the Full protocol (FP-MRI) were analyzed. The Wilcoxon non-parametric test (p-value <0.050) was used to compare the capability in measuring the most extensive residual lesion.

RESULTS

The median age was 47 (24-80) years. ABP-MRI 1 showed higher specificity (84.6%; 77/91) but a higher probability of false-negatives (16.8%) and lower sensitivity (83.2%; 99/119) than ABP-MRI 2,3 and the FP-MRI, which were identical in specificity (81.3%; 74/91), probability of false-negatives (8.4%), and sensitivity (91.6%; 109/119). ABP-MRI 2 showed a mean underestimation of only 0.03 cm in the measurement of the longest axis of the residual lesion (p = 0.008) with an average reduction in the acquisition time of 75%, compared with the FP-MRI.

CONCLUSION

ABP-MRI 2 showed diagnostic performance equivalent to the FP-MRI with a 75% reduction in the acquisition time.

Radiomic Models Predict Tumor Microenvironment Using Artificial Intelligence-the Novel Biomarkers in Breast Cancer Immune Microenvironment

Breast cancer is the most common malignancy in women, and some subtypes are associated with a poor prognosis with a lack of efficacious therapy. Moreover, immunotherapy and the use of other novel antibody‒drug conjugates have been rapidly incorporated into the standard management of advanced breast cancer. To extract more benefit from these therapies, clarifying and monitoring the tumor microenvironment (TME) status is critical, but this is difficult to accomplish based on conventional approaches. Radiomics is a method wherein radiological image features are comprehensively collected and assessed to build connections with disease diagnosis, prognosis, therapy efficacy, the TME, etc In recent years, studies focused on predicting the TME using radiomics have increasingly emerged, most of which demonstrate meaningful results and show better capability than conventional methods in some aspects. Beyond predicting tumor-infiltrating lymphocytes, immunophenotypes, cytokines, infiltrating inflammatory factors, and other stromal components, radiomic models have the potential to provide a completely new approach to deciphering the TME and facilitating tumor management by physicians.