Evidence Hub

Unveiling the efficacy predictors and potential mechanisms of Semen Cuscutae against osteoporosis via machine learning and meta-analysis: a preclinical study

Dong B, Tang R, Wan D, Yuan H, Leng C, Wang R, Li F, He J et al.

Frontiers in immunology, 2026

Abstract

Purpose: To systematically evaluate the anti-osteoporotic efficacy and underlying mechanisms of Semen Cuscutae and to explore potential experimental factors associated with treatment efficacy using an integration of meta-analysis and machine learning (ML).

Methods: A comprehensive search of eight databases (from inception to December 1, 2025) was conducted to identify randomized controlled trials of Semen Cuscutae in osteoporotic animal models. Meta-analysis was performed to assess Bone Mineral Density (BMD), bone histomorphometry, and inflammatory/metabolic markers. Additionally, an exploratory XGBoost model interpreted via SHapley Additive exPlanations (SHAP) was constructed to quantitatively rank ten experimental variables potentially associated with therapeutic efficacy.

Results: Fifteen studies were included in the final analysis. Meta-analysis demonstrated that Semen Cuscutae intervention significantly increased femoral and lumbar spine BMD, improved trabecular micro-architecture (increased BV/TV and Tb.Th; decreased Tb.Sp), and enhanced biomechanical strength. Mechanistically, Semen Cuscutae remodeled the osteoimmune microenvironment by suppressing pro-inflammatory cytokines (IL-6) and bone resorption markers (TRACP, CTX, RANKL), while ALP showed an increasing trend and osteogenic or reparative factors, including OPG, IGF, and TGF-β, were significantly upregulated. Exploratory ML analysis ranked "Species," "Dose," and "Treatment Duration" as the variables most…

How strong is the evidence?

Evidence Hub reads this as a meta-analysis, done in animal and computer models (in silico), not in people. Its evidence level is 5 of 5 (Very low · case / narrative / animal / in vitro / in silico). Case reports, narrative reviews, and animal, cell or computer-model studies give the weakest evidence for human health.

How the five levels of evidence work. The level is assigned automatically from the abstract and can be wrong, so check the paper.

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free