Can radiomics from dynamic contrast-enhanced MRI effectively predict response to neoadjuvant chemotherapy in breast cancer?: A meta-analysis.

[OBJECTIVE] This study aimed to evaluate the diagnostic accuracy of radiomics based on dynamic contrast-enhanced MRI (DCE-MRI) in predicting pathological complete response (pCR) in breast cancer patients undergoing neoadjuvant chemotherapy (NAC).

[MATERIALS AND METHODS] Relevant studies published up until September 2023, were searched in the PubMed, Web of Science, and The Cochrane Library databases, and screened based on inclusion criteria. The diagnostic performance of radiomics was evaluated using pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR).

[RESULTS] The analysis included 7 studies with a total of 1272 cases. Pooled estimates suggested overall diagnostic accuracy of radiomics in detecting pCR were determined: sensitivity, 81% (95% CI, 73-87%); specificity, 74% (95% CI, 54-88%); PLR, 3.1 (95% CI, 1.6-6.1); NLR, 0.26 (95% CI, 0.17-0.39); DOR, 12 (95% CI, 5-31); and AUC, 0.83 (95% CI, 0.80-0.86). By MRI field, studies using 3.0 T showed slightly lower sensitivity compared to those using 1.5 T/3.0 T, but notably higher specificity in comparison, and PLR and DOR were higher when using 3.0 T than when using 1.5 T/3.0 T. By image acquisition time, combining pre-NAC with other time points showed better DOR and PLR performance than using pre-NAC alone.

[CONCLUSIONS] The radiomics analysis based on Dynamic Contrast-Enhanced MRI (DCE-MRI) demonstrated substantial predictive efficacy for achieving pathological complete response (pCR) following neoadjuvant chemotherapy in breast cancer. Consequently, we advocate for the integration of this tool as a supplementary resource to inform and enhance clinical decision-making processes.