JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Prediction model for individualized precision surgery in breast cancer patients with complete response on MRI and residual calcifications on mammography after neoadjuvant chemotherapy

  • 0Department of Radiology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea.
Cancer +

|

September 30, 2024

|

September 30, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Predicting residual breast cancer in calcifications after neoadjuvant chemotherapy (NAC) is crucial for surgical planning. This study identified molecular subtype and Ki-67 as key predictors, aiding in tailored treatment strategies for breast cancer patients.

Area Of Science

  • Oncology
  • Radiology
  • Breast Cancer Research

Background

  • Accurate assessment of residual carcinoma in suspicious calcifications post-neoadjuvant chemotherapy (NAC) is vital for surgical decision-making in breast cancer patients.
  • Differentiating between calcifications without residual carcinoma (ypCalc_0) and those with residual carcinoma (ypCalc_ca) impacts surgical approach.
  • This study focuses on developing a predictive model for patients with residual suspicious calcifications on mammography but complete response on MRI after NAC.

Purpose Of The Study

  • To identify factors predicting the absence (ypCalc_0) or presence (ypCalc_ca) of residual carcinoma in suspicious calcifications after NAC.
  • To develop and validate a prediction model for residual carcinoma in breast cancer patients with specific imaging findings post-NAC.

Main Methods

  • Retrospective analysis of breast cancer patients who received NAC and exhibited residual suspicious mammographic calcifications with complete MRI response.
  • Development and validation sets were used, spanning January 2019 to December 2022.
  • Multivariable logistic regression and decision tree analysis were employed to identify predictors and build the model.

Main Results

  • The study included 134 women in the development set and 146 in the validation set.
  • Molecular subtype and high Ki-67 were significant independent predictors of ypCalc_0 in the development set.
  • The final prediction model demonstrated good performance in the validation set (AUC 0.844), differentiating ypCalc_0 and ypCalc_ca based on hormone receptor (HR)/human epidermal growth factor receptor 2 (HER2) status and Ki-67 levels.

Conclusions

  • Minimized surgery may be suitable for HR-/HER2+ breast cancers with high Ki-67 showing residual calcifications post-NAC.
  • Complete excision is recommended for HR+/HER2- breast cancers, and for HR+/HER2+ or triple-negative (TN) breast cancers with low Ki-67.
  • The developed prediction model aids in tailoring surgical management based on tumor characteristics and response to NAC.

Keywords: