Metabolomic Biomarker Panel for ACLF Progression Risk Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current ACLF prognostic tests lack accuracy and reliability, necessitating a more effective method for predicting short-term mortality risk and progression to ACLF in patients, which is crucial for timely intervention and resource allocation.

Innovation Solution

A biomarker system utilizing untargeted metabolomics analysis identifies specific biomarkers, combined through machine learning algorithms, to construct prediction models for short-term mortality and progression risk in ACLF and non-ACLF patients, leveraging LC-MS for clinical implementation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional scoring systems (MELD, MELD-Na, CTP) are used for ACLF prognosis prediction, then the evaluation process is simple and widely applicable, but the prediction accuracy is limited

Engineering Contradiction:
Improveprognosis prediction accuracyVSAvoidevaluation system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the prognostic evaluation into two distinct prediction models: one for ACLF patients (using 4 metabolites) and one for non-ACLF patients (using 3 metabolites). This segmentation allows each model to be optimized for its specific population, improving prediction accuracy while maintaining manageable complexity through targeted metabolite selection rather than comprehensive profiling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the evaluation parameters from traditional clinical scoring systems to metabolomic profiles. By measuring specific metabolite concentrations (such as alanine, aspartate, glutamate, and succinate) rather than relying on aggregated clinical scores, the system achieves higher prediction accuracy while providing more granular physiological insights into disease progression.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If comprehensive metabolomic profiling is performed to improve prediction accuracy, then prediction precision improves, but detection complexity and cost increase

Engineering Contradiction:
Improverisk prediction accuracyVSAvoidbiomarker detection complexity
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent extracts and focuses on specific key metabolites that are most discriminatory for ACLF prediction. Instead of measuring all possible metabolites, the system identifies and quantifies only the essential ones (alanine, aspartate, glutamate, succinate for ACLF; plus additional metabolites for non-ACLF progression), significantly reducing detection complexity while maintaining high prediction accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by selecting a subset of the most informative metabolites rather than performing comprehensive metabolomic profiling. This partial approach measures only what is necessary for accurate prediction, avoiding the excessive complexity and cost of full metabolomic analysis while still achieving superior performance to traditional scoring systems.

Inventive Principle:
Principle #16Partial or excessive action

3Reliability

If early identification of pre-ACLF patients is implemented, then intervention effectiveness improves, but diagnostic accuracy requirements increase

Engineering Contradiction:
Improveintervention timing reliabilityVSAvoiddiagnostic accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent enables preliminary action by predicting ACLF progression in non-ACLF patients before actual failure occurs. The prediction model identifies patients at high risk of developing ACLF within 28 days, allowing proactive intervention to be initiated early in the disease trajectory, thereby improving reliability of intervention timing while maintaining diagnostic accuracy through metabolite-based early detection.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The biomarker system provides accurate prediction models with AUCs of 0.80 for ACLF mortality and 0.85 for non-ACLF progression, enabling timely intervention and resource optimization.

Implementation Method 1

leveraging LC-MS for clinical implementation

Methodology Applied
Scientific EffectLiquid chromatography-mass spectrometry:

Data Source

PatentEP4692783A1Biomarker and system for predicting evolution into ACLF or death of ACLF patient, and uses thereof
Publication Date: 2026.02.11 SHANGHAI GANNING MEDICAL & TECHNOLOGY CO LTD
  • EP4692783A1 patent drawingFigure 1
  • EP4692783A1 patent drawingFigure 2
  • EP4692783A1 patent drawingFigure 3

AI summary

Provided are a biomarker and system for predicting progression to ACLF or mortality of an ACLF patient, and use thereof. Through untargeted metabolomics analysis of ACLF-related samples, a series of biomarkers capable of predicting a short-term mortality risk of the ACLF patient and a short-term progression risk to ACLF of a non-ACLF patient are found, and a biomarker combination is further screened therefrom, to construct a short-term mortality risk prediction model for ACLF patients and a short-term ACLF progression risk prediction model for non-ACLF patients, thereby achieving convenient and efficient prediction, seizing the golden window for intervention for high-risk groups, reducing economic burdens for low-risk patients, and meeting clinical needs.