Development and internal validation of a radiomics-clinical combined model for predicting axillary pathological complete response in clinically node-positive breast cancer patients after neoadjuvant chemotherapy - Takeaways - MDSpire

Development and internal validation of a radiomics-clinical combined model for predicting axillary pathological complete response in clinically node-positive breast cancer patients after neoadjuvant chemotherapy

  • By

  • Weitao Yan

  • Wenxuan Lu

  • Ying Dai

  • Xiangchao Meng

  • Kai Feng

  • July 10, 2026

  • 0 min

Share

  • 1

    A combined radiomics-clinical model was developed to predict axillary pathological complete response (apCR) in clinically node-positive breast cancer patients.

  • 2

    The study enrolled 386 patients, with a training cohort of 270 and a validation cohort of 116, to assess the model's performance.

  • 3

    The combined model achieved a validation AUC of 0.703, outperforming the radiomics-only model but not significantly better than the clinical-only model.

  • 4

    The overall apCR rate among participants was 43.5%, with risk stratification showing varying rates across different tertiles.

  • 5

    The model demonstrated moderate discriminatory ability, but its current misclassification rate limits direct clinical application for surgical de-escalation.

Original Source(s)

Development and internal validation of a radiomics-clinical combined model for predicting axillary pathological complete response in clinically node-positive breast cancer patients after neoadjuvant chemotherapy

  • by Weitao Yan, Wenxuan Lu, Ying Dai, Xiangchao Meng, Kai Feng

  • July 10, 2026

Related Content

Ferroptosis in chemotherapy resistance and resensitization in breast cancer: a systematic review of preclinical evidence and translational implications

Multimodal imaging features of primary breast epithelial–myoepithelial carcinoma: a case report and literature review

Efficacy and safety of traditional Chinese medicine decoctions in breast cancer treatment: a network meta-analysis