JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

A colorectal liver metastasis prediction model based on the combination of lipoprotein-associated phospholipase A2 and serum biomarker levels

  • 0Department of Clinical Laboratory, Xiangya Hospital, Central South University, Changsha 410008, Hunan, China.
Clinica Chimica Acta; International Journal of Clinical Chemistry +

|

January 18, 2025

|

January 18, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

Serum lipoprotein-associated phospholipase A2 (Lp-PLA2) shows strong predictive value for colorectal liver metastasis (CRLM). A machine learning model combining Lp-PLA2 with standard markers effectively predicts CRLM, aiding early detection.

Area Of Science

  • Biochemistry
  • Oncology
  • Medical Diagnostics

Background

  • Colorectal liver metastasis (CRLM) poses a significant clinical challenge.
  • Early and accurate detection of CRLM is crucial for effective patient management.
  • Biomarkers for CRLM prediction require further investigation.

Purpose Of The Study

  • To evaluate the predictive capability of serum lipoprotein-associated phospholipase A2 (Lp-PLA2) in patients with colorectal liver metastasis (CRLM).
  • To develop and validate a machine learning (ML) model for CRLM prediction using Lp-PLA2 and conventional serological markers.

Main Methods

  • Serum Lp-PLA2 levels were measured in 507 participants (162 healthy controls, 186 non-CRLM patients, 159 CRLM patients).
  • A prediction model was constructed using Random Forest ML algorithm, integrating Lp-PLA2 with standard laboratory parameters (ALB, GLB, ALT, LDH, TC).
  • Model performance was assessed using AUC, sensitivity, specificity, PPV, and NPV.

Main Results

  • Serum Lp-PLA2 levels were significantly higher in CRLM patients compared to healthy controls and non-CRLM patients (P < 0.0001).
  • The Random Forest model achieved an AUC of 0.918, demonstrating high predictive accuracy.
  • The model incorporating Lp-PLA2 achieved a sensitivity of 0.823 and specificity of 0.889.

Conclusions

  • Serum Lp-PLA2 is a valuable biomarker for predicting CRLM.
  • The developed Random Forest model integrating Lp-PLA2 and conventional markers offers robust CRLM prediction.
  • This approach shows promise for improving the early detection of colorectal liver metastasis.

Keywords: