Milk thistle extract shows antifibrotic promise but fails clinical translation

Silymarin, the flavonolignan-rich extract from milk thistle (Silybum marianum), has accumulated a large body of preclinical evidence suggesting it can interfere with fibrotic processes in the liver, kidney, lung, peritoneum, and skin. Yet a new critical review published in Frontiers in Pharmacology concludes that this laboratory activity has not translated into a proven antifibrotic therapy for humans. The review, written by Zhe Jin and colleagues at Jingjiang People’s Hospital Affiliated to Yangzhou University, systematically separates what the experimental literature supports from what clinical trials have actually demonstrated, and finds a wide gap between the two.

The authors conducted a structured, non-systematic evidence-mapping search of PubMed/MEDLINE, Web of Science, and Scopus covering English-language literature through December 2025, followed by a targeted update through 9 August 2026 that focused on fibrosis, pharmacokinetics, drug delivery, ferroptosis, succinate metabolism, and the protein ANXA6. Because the work was a critical narrative review rather than a formal systematic review, no PRISMA flow diagram was used, and evidence was weighted qualitatively rather than scored. Human randomized studies and trials with fibrosis-specific structural endpoints were given greater translational weight than studies relying on biochemical surrogates.

A central problem the review identifies is chemical heterogeneity. Silymarin is not a single molecule but a multicomponent preparation containing several flavonolignans, chiefly silybin (also called silibinin), along with isosilybin, silychristin, silydianin, and minor constituents such as taxifolin. Analysis of more than 40 commercial milk-thistle extracts showed that the ratios of selected flavonolignans are remarkably consistent between products, but overall composition can still fluctuate depending on processing methods, plant chemotype, and source material. The authors argue that this variation should be treated as a matter of mechanistic rigor rather than a mere manufacturing detail, since crude extracts, purified silibinin, phospholipid complexes, and nanocarrier formulations are not pharmacologically interchangeable.

Pharmacokinetics compound the problem. Conventional oral silymarin produces low and variable systemic exposure, and the individual flavonolignans behave differently in the bloodstream. In healthy volunteers given a standardized milk-thistle extract, six primary flavonolignans were detectable in plasma, with free exposures ranked silybin A highest, followed by silybin B, isosilybin B, isosilybin A, silychristin, and silydianin. After single doses of 175 to 525 milligrams, average peak concentrations of free silybin A ranged from roughly 107 to 299 nanograms per milliliter, while free silybin B ranged from about 31 to 121 nanograms per milliliter. At steady state with 175 milligrams three times daily, free silybin A averaged around 135 nanograms per milliliter. Most circulating material exists as glucuronide and sulfate conjugates rather than free compound, and biliary excretion with enterohepatic circulation shapes disposition further.

Formulation can dramatically change these numbers. In a crossover study of 16 healthy subjects, a micellar silymarin preparation produced peak exposure approximately 19-fold higher and total systemic exposure over 24 hours approximately 11-fold higher than an unformulated comparator. Phosphatidylcholine complexation similarly increases silybin exposure, and experimental platforms including nanoliposomes, liver-targeted nanocolloids, redox nanoparticles, and sustained-release matrices can alter release, delivery, and pharmacological activity. The review stresses, however, that a pharmacokinetic advantage is not antifibrotic efficacy: increased exposure or preclinical tissue targeting does not by itself establish clinical benefit.

On the mechanistic side, the preclinical literature is broad. Silymarin-containing interventions have been shown to modulate TGF-β/Smad signaling, a canonical profibrotic axis governing fibroblast activation and collagen production. In thioacetamide-induced hepatic fibrosis, comparisons of silymarin, caffeine, and their combination revealed decreases in TGF-β1, CTGF, α-SMA, and LPAR1-related markers, particularly with the combination regimen. Silymarin also reduced CTGF, hepatic hydroxyproline, and histological fibrosis in carbon tetrachloride-treated rats. In renal models, nanoliposomal silymarin co-suppressed TGF-β/Smad and JAK2/STAT3/SOCS1 signaling in diabetic nephropathy, and oral silymarin in a 2024 thioacetamide model improved renal function and reduced interstitial fibrosis, collagen I, and CTGF. In cultured human skin fibroblasts, silibinin shifted matrix regulation toward lower type I collagen production with reduced Smad2/3 activation, and purified silibinin inhibits the proliferation and migration of LX-2 hepatic stellate cells.

Recent work has extended the mechanistic picture. In a 2025 MASH model, purified silybin improved the mitochondrial lipid environment and complex II activity in hepatocytes, reduced succinate accumulation and release, and made hepatocyte-conditioned medium less stimulatory to stellate cells, supporting an upstream metabolic mechanism in that specific model. In renal fibrosis, a silybin-FTH1 interaction was identified, with loss- and gain-of-function experiments showing that FTH1 deletion worsened ferroptotic damage and fibrosis while overexpression was protective. In pulmonary fibrosis, silibinin physically interacted with ANXA6 and modified endoplasmic reticulum and oxidative stress responses along with PI3K-AKT-mTOR signaling. The authors caution that these target-level findings require independent validation, direct in vivo demonstration of target engagement, and confirmation that the mechanisms operate at clinically achievable concentrations, which currently cannot be verified because human target-tissue exposure data are lacking.

The clinical record is far less convincing. An early randomized cirrhosis trial reported a survival benefit, more apparent in alcoholic cirrhosis and Child-Pugh A disease, but a later multicenter randomized trial in the same disease found no survival advantage. In chronic hepatitis C, randomized treatment with silymarin was well tolerated but produced no significant improvement in HCV RNA or aminotransferase levels. Observational data from the HALT-C cohort showed that self-reported silymarin use correlated with less histological progression from fibrosis to cirrhosis but not with clinical outcomes. In biopsy-proven NASH, a randomized trial of oral silymarin 700 milligrams three times daily for 48 weeks missed its primary endpoint of a 30 percent reduction in NAFLD activity score, though a secondary fibrosis-related signal was observed and requires confirmation. The SyNCH trial of high-dose silymarin in non-cirrhotic NASH showed acceptable safety but no significant effect on its primary histological endpoint. A 24-week randomized MASLD trial reported improved liver-stiffness measurements, which the review classifies as a non-invasive signal rather than proof of scar regression given the absence of paired histology. A 2025 Cochrane review covering 17 trials and 2,069 people concluded that the overall risks and benefits were unclear, with low and very low certainty evidence and substantial clinical heterogeneity.

Renal clinical findings are similarly mixed. A three-month trial of silymarin added to renin-angiotensin system inhibitors in diabetic nephropathy reported reductions in albuminuria and selected inflammatory and oxidative-stress markers, but these are surrogate measures that cannot prove reversal of structural renal fibrosis. In contrast, a subsequent randomized trial of 102 normotensive patients with type 2 diabetes and proteinuria found that silymarin as an adjunct to optimized RAS blockade did not significantly affect the primary composite outcome, renal outcomes, or proteinuria over two years. The review also distinguishes carefully between categories of fibrosis evidence: antifibrogenic effects during ongoing injury, effects on established fibrosis, and true regression of pre-existing scar are not equivalent outcomes, and changes in oxidative stress or collagen markers alone do not demonstrate regression.

The authors’ overall conclusion is that silymarin possesses biologically plausible multitarget antifibrotic activity, strongest for prevention or attenuation of fibro-inflammatory processes in preclinical models, less consistent for effects on established fibrosis, and currently insufficient to support any claim that it reproducibly reverses fibrosis in humans. Notably, in one carbon tetrachloride withdrawal model neither silymarin nor silibinin produced a significant fibrolytic effect, while a later high-dose study reported greater histological restoration, illustrating the discordance in regression data. The translational gap is attributed to multiple factors: product heterogeneity, limited oral exposure, incomplete standardization, and endpoint selection that often relied on biochemical or virological surrogates rather than direct fibrosis measures. Future development, the review argues, should use chemically defined preparations, exposure-informed dosing with pharmacokinetic measurements, validated target engagement, and fibrosis-centered endpoints such as histological stage, elastography, and collagen-turnover measures. On current evidence, silymarin remains an investigational adjunctive candidate rather than an established antifibrotic treatment, and the authors suggest it may ultimately be more plausible as an adjunctive fibro-inflammatory modulator combined with metabolic or anti-inflammatory strategies than as stand-alone monotherapy for advanced fibrosis.

Subject of Research: The antifibrotic potential of the milk thistle extract silymarin, spanning its phytochemistry, mechanisms, preclinical evidence, and limited clinical translation.

Article Title: Phytochemical basis, multitarget pharmacology, and translational perspectives of silymarin: emphasis on its antifibrotic effects

Article References: Jin, Z., Zhu, J., Lu, B., Li, Y., & Chen, Y. (2026). Phytochemical basis, multitarget pharmacology, and translational perspectives of silymarin: emphasis on its antifibrotic effects. Frontiers in Pharmacology, 17, Article 2001949. https://doi.org/10.3389/fphar.2026.2001949

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.2001949

Keywords: silymarin, milk thistle, fibrosis, silybin, TGF-beta, pharmacokinetics, liver fibrosis, NASH, drug delivery, flavonolignans, preclinical models, clinical trials