Can Radiomics Predict Pathologic Complete Response After Neoadjuvant Chemoradiotherapy for Rectal Cancer? A...

Area of Science:

• Oncology
• Radiology
• Medical Imaging

Background:

• Rectal cancer treatment is evolving with neoadjuvant therapies and non-operative management.
• Assessing treatment response is crucial for guiding patient management.
• Pathologic complete response (pCR) is a key endpoint, often correlated with clinical response.
• Magnetic resonance imaging (MRI) plays a vital role in evaluating response.

Purpose of the Study:

• To systematically review and perform a meta-analysis of MRI-based radiomics for predicting pCR in rectal cancer.
• To assess the diagnostic accuracy of radiomics models in predicting pCR.

Main Methods:

• Systematic review and meta-analysis of 29 studies.
• Inclusion of data from 4486 patients.
• Evaluation of radiomics models using baseline imaging, delta radiomic data, and post-neoadjuvant imaging.
• Analysis of radiomics-only and radiomics-combined-with-clinical-data models.

Main Results:

• Radiomics models showed good diagnostic accuracy for predicting pCR.
• Radiomics combined with clinical data achieved an area under the curve (AUC) of 0.88 for baseline imaging.
• Delta radiomic data achieved an AUC of 0.86.
• Post-neoadjuvant radiomics models ranged from 0.75 to 0.83 AUC.

Conclusions:

• MRI-based radiomics shows promise in predicting pathologic complete response in rectal cancer.
• These prediction models can support individualized treatment decisions.
• Further integration of radiomics may enhance precision medicine approaches in rectal cancer care.