JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Individualized prediction of non-sentinel lymph node metastasis in Chinese breast cancer patients with ≥ 3 positive sentinel lymph nodes based on machine-learning algorithms

  • 0The Breast Center, Jieyang People's Hospital, Jieyang, Guangdong, 522000, People's Republic of China.
Bmc Cancer +

|

September 2, 2024

|

September 2, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

Machine learning accurately predicts non-sentinel lymph node metastasis in breast cancer (BC) patients with multiple positive sentinel lymph nodes (SLNs). This may allow avoiding axillary lymph node dissection (ALND), reducing complications and improving quality of life.

Area Of Science

  • Oncology
  • Medical Informatics
  • Surgical Oncology

Background

  • Axillary lymph node dissection (ALND) is standard for early-stage breast cancer (BC) with ≥3 positive sentinel lymph nodes (SLNs).
  • ALND can cause significant complications without always improving clinical outcomes.
  • Predicting non-sentinel lymph node (non-SLN) metastasis is crucial for treatment decisions.

Purpose Of The Study

  • Develop and validate machine learning (ML) models to predict non-SLN metastasis in Chinese BC patients with ≥3 positive SLNs.
  • Identify a potential strategy to omit ALND in selected patients.

Main Methods

  • Analyzed data from 2217 BC patients, focusing on 634 with positive SLNs.
  • Applied nine ML algorithms to predict non-SLN metastasis.
  • Evaluated model performance using ROC, precision-recall, and calibration curves; assessed clinical utility with Decision Curve Analysis (DCA).

Main Results

  • The Random Forest (RF) model demonstrated superior predictive performance (AUC 0.987 training, 0.828 validation, 0.870 external validation).
  • Key predictors included size of positive SLNs, tumor size, number of SLNs, and ER status.
  • The RF model showed robust predictive capabilities in external validation.

Conclusions

  • The developed RF model accurately predicts non-SLN metastasis in BC patients with ≥3 positive SLNs.
  • This predictive capability may allow avoiding ALND in selected patients, potentially supplemented with axillary radiotherapy.
  • This approach could decrease postoperative complications and enhance patient quality of life, warranting prospective clinical trial validation.

Keywords: