Advancing herbal medicine: Extent similarity assessment for a quantitative fingerprinting approach

Analysis of Herbal Medicinal Products (HMPs) presents unique challenges due to their complex nature and the often-unknown identity of their active constituents. The guideline of European Medicines Agency (EMA/HMPC/541422/2017 Corr.) requires a comprehensive and robust analytical method. To go beyond the limited scope of established analytical markers, chemometric methods have been applied to evaluate the quality of HMPs [ 1 ] [ 2 ] [ 3 ] [ 4 ]. Extent Similarity has been underestimated while it is a promising method for the comprehensive quantitative evaluation of fingerprinting data of HMP, especially for HMP categorized as “other extracts” as no active constituents are defined and the whole extract is considered as the active principle, but only proxy analytical markers are monitored. Objective: to develop and validate a quantitative fingerprinting method via HPLC-DAD, specifically utilizing the Extent Similarity method in the quality control of HMPs containing Licorice root fluid extract The fingerprinting method was applied to an exemplary herbal extract, with a focus on capturing the quality of the mixture in a holistic way. The method involves HPLC-DAD for fingerprint analysis and Extent Similarity method for a standardized quantitative chemometric analysis. Simulated data ([ Table 1 ]) mimics a sample diluted to 80% with minor variations. The results are shown in [ Table 2 ]. Table 1 simulated typical liquid chromatographic fingerprinting data of two samples. Sample B is 80% dilution from Sample A with minor variations in each peak with area below 500. Simulated Sample Peak 1 Peak 2 Peak 3 Peak 4 Peak 5 Peak 6 A (reference) 1 10 100 200 500 1000 B 0.7 7.5 80 155 400 800 Table 2 Similarity evaluation with different approaches on simulated typical liquid chromatographic fingerprinting data of two samples. Sample B is 80% dilution from Sample A. Method data evaluation Similarity results Voctor B Cosine similarity* 0-1.000 0.98 Correlation Coefficient* -1.000-1.000 0.99 Extent similarity* -∞-1.000 0.77 Adapted Similarity Score* 0-1.000 0.87 Euclidean distance** 0-∞ 231 * The closer the value to 1, the more similar the vectors are.** The closer the value to 0, the more similar the vectors are. The Extent Similarity method can perform a full analysis of an extract like that of licorice root. The fingerprint results assessed by the Extent Similarity method proved to be suitable for the quantitative quality control of HMPs. It demonstrated accuracy, sensitivity, and reliability in capturing the phytochemical constituents holistically. Publication History Article published online: 05 March 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany