Q-marker identification strategies in traditional Chinese medicines: a systematic review of research from 2020 to 2024

TL;DR

This paper reviews methods for identifying quality markers in traditional Chinese medicines from 2020 to 2024 and proposes a combined approach to improve consistency and biological relevance.

Contribution

A systematic evaluation of Q-marker identification strategies and a novel stepwise framework integrating multiple criteria for improved reliability and applicability.

Findings

Four main Q-marker identification strategies were identified and categorized based on their methodological approaches.

An integrated stepwise model combining bioavailability, biomarker relevance, correlation modeling, and multi-criteria scoring is recommended for improved Q-marker identification.

The proposed framework enhances biological relevance and practical applicability for quality assessment in multi-component herbal systems.

Abstract

The concept of “quality markers” (Q-markers) has emerged as a key solution to address limitations in the evaluation and standardization of traditional herbal medicines. Despite the introduction of various Q-marker identification strategies, methodological inconsistencies and a lack of standardization continue to pose challenges.

This review aims to systematically organize and evaluate Q-marker selection strategies published over the past 5 years and propose an optimal approach based on a comparative analysis of their strengths and limitations.

A comprehensive literature search was performed on the Web of Science and PubMed for studies published between January 2020 and December 2024 using keywords related to Q-marker identification in traditional prescriptions. After removing duplicates and screening for relevance, the eligible studies were systematically reviewed. Key information, including the prescription name, therapeutic targets, methodological steps for Q-marker selection, and the final identified Q-markers, was extracted and organized into summary tables. Based on the analysis, the advantages and limitations of each strategy were evaluated.

The studies were categorized into four representative strategies: [S1] mechanism-driven validation, which relies on network pharmacology and bioassays to align compounds with disease pathways (22 cases, 36.67%); [S2] profile–effect correlation modeling, which uses statistical and machine learning tools to link chemical composition with pharmacodynamic outcomes (24 cases, 40%); [S3] in silico preliminary filtering, which rapidly screens candidate compounds using computational predictions without experimental validation (8 cases, 13.33%); and [S4] multi-criteria decision frameworks, which integrate formulation hierarchy, efficacy, and chemical properties into composite scoring models (6 cases, 10.00%). The average number of Q-markers identified in each strategy was 7.23, 6.61, 8.25, and 7.5, respectively. While each strategy has unique analytical strengths, they often lack consistency and reproducibility when applied in isolation. To overcome this, we recommend a stepwise approach that integrates (1) compound selection based on bioavailability, (2) disease-relevant biomarker selection, (3) correlation modeling, and (4) a multi-criteria scoring framework based on TCM principles. This integrated model accounts for compound bioavailability, specificity, and formulation roles, enabling the identification of functionally relevant Q-markers, including low-abundance constituents.

This review can provide valuable insights to guide future research and development of traditional herbal medicines, particularly in the context of quality control and innovative drug discovery. The proposed framework improves biological relevance and practical applicability and may serve as a scalable model for the quality assessment of multi-component herbal systems and complex pharmacological formulations.

Flowchart depicting Q-marker strategies over time, from 2020 to 2024, dividing into four strategies: Network Pharmacology, Computational Analysis, Multi-dimensional Network, and a combination of strategies. Strengths and weaknesses are listed for each, with an evaluation matrix comparing experiment, TCM principles, complexity, strengths, and weaknesses. Strategy 2 combined with Strategy 4 is recommended for high reliability and practical application but noted for complexity and expertise requirements.

Figures