Follow us
pubmed meta image 2
🧑🏼‍💻 Research - November 23, 2024

Using Machine Learning Models to Predict Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer.

🌟 Stay Updated!
Join Dr. Ailexa’s channels to receive the latest insights in health and AI.

⚡ Quick Summary

This study explored the use of machine learning (ML) models to predict pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer patients. The findings indicate that the random forest model achieved a high AUC of 0.795, demonstrating the potential of ML in enhancing treatment outcomes.

🔍 Key Details

  • 📊 Dataset: 499 breast cancer patients treated with NAC
  • 🧩 Features used: 11 clinical features
  • ⚙️ Technology: Five different ML models, including random forest
  • 🏆 Performance: Highest AUC of 0.795 in the KKH cohort

🔑 Key Takeaways

  • 📊 Neoadjuvant chemotherapy is increasingly utilized in breast cancer treatment.
  • 💡 Machine learning models can effectively predict pCR to NAC.
  • 🏆 Random forest models showed the best performance with an AUC of 0.795.
  • 🔍 Imputation of missing data improved predictive values significantly.
  • 👩‍🔬 Key predictors included estrogen receptor intensity, HER2 intensity, and age at diagnosis.
  • 🌍 Study conducted at National Cancer Centre Singapore and KK Hospital.
  • 🆔 ClinicalTrials.gov Identifier: Not specified in the study.

📚 Background

Neoadjuvant chemotherapy (NAC) is a critical treatment strategy for breast cancer, aimed at reducing tumor size before surgery. However, predicting the response to NAC remains a challenge. The integration of machine learning into clinical practice offers a promising avenue for enhancing predictive accuracy, potentially leading to more personalized treatment plans.

🗒️ Study

This study involved a cohort of 499 patients treated with NAC across two medical centers in Singapore from January 2014 to December 2017. Researchers employed eleven clinical features to train five different ML models, focusing on the prediction of pCR. The study also evaluated methods for handling missing data, including listwise deletion and imputation.

📈 Results

The study found that 24.6% of patients in the NCCS training set, 24.7% in the NCCS testing set, and 24.8% in the KKH testing set achieved pCR. The random forest model and the imputed model both achieved the highest AUCs of 0.794 and 0.795, respectively, indicating strong predictive capabilities. Notably, the imputed model also demonstrated a higher positive predictive value (PPV) of 98.2% compared to 95.1% for the base model.

🌍 Impact and Implications

The implications of this study are significant for the field of oncology. By utilizing machine learning to predict treatment responses, healthcare providers can tailor therapies more effectively, potentially improving patient outcomes. This approach not only enhances the understanding of individual patient responses but also paves the way for more personalized medicine in breast cancer treatment.

🔮 Conclusion

This research highlights the transformative potential of machine learning in predicting pCR after NAC in breast cancer patients. The findings suggest that employing ML models, particularly the random forest approach, can lead to improved predictive accuracy and better clinical decision-making. Continued exploration in this area is essential for advancing treatment strategies and enhancing patient care.

💬 Your comments

What are your thoughts on the use of machine learning in predicting treatment responses for breast cancer? We would love to hear your insights! 💬 Leave your comments below or connect with us on social media:

Using Machine Learning Models to Predict Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer.

Abstract

PURPOSE: Neoadjuvant chemotherapy (NAC) is increasingly used in breast cancer. Predictive modeling is useful in predicting pathologic complete response (pCR) to NAC. We test machine learning (ML) models to predict pCR in breast cancer and explore methods of handling missing data.
METHODS: Four hundred and ninety-nine patients with breast cancer treated with NAC in two centers in Singapore (National Cancer Centre Singapore [NCCS] and KK Hospital) between January 2014 and December 2017 were included. Eleven clinical features were used to train five different ML models. Listwise deletion and imputation were evaluated on handling missing data. Model performance was evaluated by AUC and calibration (Brier score). Feature importance from the best performing model in the external testing data set was calculated using Shapley additive explanations.
RESULTS: Seventy-two (24.6%), 18 (24.7%), and 31 (24.8%) patients attained pCR in NCCS training, NCCS testing, and KK Women’s and Children’s Hospital (KKH) testing data sets, respectively. The random forest (RF) base and imputed models have the highest AUCs in the KKH cohort of 0.794 (95% CI, 0.709 to 0.873) and 0.795 (95% CI, 0.706 to 0.871), respectively, and were the best calibrated with the lowest Brier score. No statistically significant difference was noted between AUCs of the base and imputed models in all data sets. The imputed model had a larger positive predictive value (PPV; 98.2% v 95.1%) and negative predictive value (NPV; 96.7% v 90.0%) than the base model in the KKH data set. Estrogen receptor intensity, human epidermal growth factor 2 intensity, and age at diagnosis were the three most important predictors.
CONCLUSION: ML, particularly RF, demonstrates reasonable accuracy in pCR prediction after NAC. Imputing missing fields in the data can improve the PPV and NPV of the pCR prediction model.

Author: [‘Rahadian RE’, ‘Tan HQ’, ‘Ho BS’, ‘Kumaran A’, ‘Villanueva A’, ‘Sng J’, ‘Tan RSYC’, ‘Tan TJY’, ‘Tan VKM’, ‘Tan BKT’, ‘Lim GH’, ‘Cai Y’, ‘Nei WL’, ‘Wong FY’]

Journal: JCO Clin Cancer Inform

Citation: Rahadian RE, et al. Using Machine Learning Models to Predict Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer. Using Machine Learning Models to Predict Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer. 2024; 8:e2400071. doi: 10.1200/CCI.24.00071

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on whatsapp
WhatsApp

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.