Prognostic Scores for Liver Resection in Colorectal Metastases: Performance, Limitations, and Methodological...

Area of Science:

• Oncology
• Surgical Oncology
• Biostatistics

Background:

• Accurate survival prediction after resection of colorectal liver metastases is vital for treatment planning.
• Existing prognostic scores have questionable reliability, necessitating a comprehensive review.

Purpose of the Study:

• To systematically review and evaluate the performance of prognostic scores for colorectal liver metastases.
• To analyze the methodological approaches used in the external validation of these prognostic models.

Main Methods:

• A systematic literature review was performed using PubMed, Embase, and Cochrane Library (January 2015 - June 2024).
• Included studies focused on English-language publications with external validation of prognostic models.
• A random-effects meta-analysis was conducted to synthesize performance data.

Main Results:

• 48 prognostic scores were externally validated in 33,602 patients across 48 studies.
• Scores incorporating genetic, immunological, and radiomic data outperformed purely clinical scores (C-index 0.610-0.657 vs. 0.585).
• Models using perioperative data showed superior performance compared to preoperative models (C-index 0.671 vs. 0.600).

Conclusions:

• Current prognostic scores demonstrate low reliability (C-index ≤ 0.65) for predicting survival.
• Heterogeneity and variable quality of data limit the usability of existing prognostic tools.
• There is a critical need for developing novel prognostic tools and standardizing modeling and reporting.