Caffeic acid

Pharmacology & Research

Caffeic acid is a hydroxycinnamic acid — a phenolic acid found throughout the plant kingdom and, despite the name, unrelated to caffeine. The single most important thing to know about it is that plants rarely carry much of the free molecule: it occurs overwhelmingly as esters — chlorogenic acid (caffeoylquinic acid), rosmarinic acid, chicoric and caftaric acid, and CAPE (caffeic acid phenethyl ester). A great deal of the “caffeic acid” literature is really about one of those esters, which behave and dose very differently. The free acid’s own best-substantiated action is direct antioxidant chemistry, backed by a clean, reproducible finding that it is a selective inhibitor of 5-lipoxygenase ^{2Reference 2Koshihara Y et al. · 1984Caffeic acid is a selective inhibitor for leukotriene biosynthesisView study →}. But its pharmacokinetics are limiting — oral caffeic acid is largely a pro-metabolite, rapidly methylated to ferulic acid and conjugated ^{4Reference 4Omar MH et al. · 2012AnimalAbsorption, disposition, metabolism, and excretion of [3-¹⁴C]caffeic acid in ratsView study →} — so most in-vitro potency does not carry through to the body. Human data on the isolate amount to a single small topical trial.

1. Antioxidant (property)

This is the best-substantiated action. The catechol (ortho-dihydroxy) ring plus the α,β-unsaturated side chain make caffeic acid an efficient electron/hydrogen donor: it scavenges DPPH and superoxide and protects cultured endothelial cells and monocytes from oxidative stress at low micromolar concentrations ^{1Reference 1Zapolska-Downar D et al. · 2002Protective properties of artichoke (Cynara scolymus) against oxidative stress induced in cultured endothelial cells and monocytesView study →}.

Gap: almost all of this is in-vitro or ex-vivo. In the body the free acid is rapidly methylated and conjugated (see Pharmacokinetics), so the circulating “antioxidant” is largely ferulic acid and conjugates, not caffeic acid itself ^{4Reference 4Omar MH et al. · 2012AnimalAbsorption, disposition, metabolism, and excretion of [3-¹⁴C]caffeic acid in ratsView study →}.

2. Anti-inflammatory (5-LOX)

A classic, reproducible biochemical result: caffeic acid selectively inhibits leukotriene biosynthesis via 5-lipoxygenase without blocking cyclooxygenase ^{2Reference 2Koshihara Y et al. · 1984Caffeic acid is a selective inhibitor for leukotriene biosynthesisView study →}. Broader Nrf2-activating and NF-κB-suppressing anti-inflammatory themes are often attached to caffeic acid, but the strongest such data are for the esters (CAPE, caffeic acid methyl ester), not the free acid.

Gap: no controlled human anti-inflammatory data for the isolate, and the whole-animal potency of the free acid is unproven ^{2Reference 2Koshihara Y et al. · 1984Caffeic acid is a selective inhibitor for leukotriene biosynthesisView study →}.

3. Recurrent mouth ulcers (topical)

The only human trial of isolated caffeic acid. In a randomised, double-blind, placebo-controlled study (n ≈ 47), caffeic acid mucoadhesive buccal tablets reduced lesion size and pain in minor recurrent aphthous stomatitis over a 7-day follow-up ^{3Reference 3Salehi M et al. · 2024RCTEvaluation of caffeic acid mucoadhesive tablets on minor recurrent aphthous stomatitis: a randomized, double-blind, placebo-controlled clinical trialView study →}.

Gap: small, single-centre, and local delivery — a tablet held against the ulcer bypasses the absorption/metabolism problem entirely, so it says nothing about systemic oral dosing ^{3Reference 3Salehi M et al. · 2024RCTEvaluation of caffeic acid mucoadhesive tablets on minor recurrent aphthous stomatitis: a randomized, double-blind, placebo-controlled clinical trialView study →}.

4. Antidiabetic / metabolic

In streptozotocin-diabetic rats, caffeic acid lowered plasma glucose, proposed to act through β-endorphin release on α1A-adrenoceptors to enhance glucose uptake ^{5Reference 5Hsu FL et al. · 2003AnimalRelease of β-endorphin by caffeic acid to lower plasma glucose in streptozotocin-induced diabetic ratsView study →}; in a high-fat-diet/STZ model it attenuated diabetic kidney disease by restoring autophagy over 12 weeks ^{6Reference 6Matboli M et al. · 2017AnimalCaffeic acid attenuates diabetic kidney disease via modulation of autophagy in a high-fat diet/streptozotocin-induced diabetic ratView study →}.

Gap: rodent only, with disparate and unreconciled mechanisms; the AMPK-type metabolic effects seen in cell work are largely reported for caffeic acid esters rather than the free acid ^{5,6Reference 5Hsu FL et al. · 2003AnimalRelease of β-endorphin by caffeic acid to lower plasma glucose in streptozotocin-induced diabetic ratsView study →Reference 6Matboli M et al. · 2017AnimalCaffeic acid attenuates diabetic kidney disease via modulation of autophagy in a high-fat diet/streptozotocin-induced diabetic ratView study →}.

5. Antidepressant-like

Isolated caffeic acid (~4 mg/kg i.p.) reduced immobility in the forced-swim test with anxiolytic-like effects in mice ^{7Reference 7Nogueira Neto JD et al. · 2020AnimalAnxiolytic and antidepressant-like effects of Annona coriacea (Mart.) and caffeic acid in miceView study →}.

Gap: preclinical only. The antidepressant result frequently cited for pau d’arco (NMDA/nitric-oxide pathway) comes from a whole Tabebuia extract, not the isolate ^{8Reference 8Freitas AE et al. · 2013AnimalNMDA receptors and the L-arginine–nitric-oxide–cGMP pathway are implicated in the antidepressant-like action of the ethanolic extract from Tabebuia avellanedae in miceView study →}, and should not be read as isolate evidence.

6. Antimicrobial / antiviral

Caffeic acid ranks among the phenolics predicted to drive yerba maté’s activity against MRSA in a computational contribution analysis ^{9Reference 9Rempe CS et al. · 2015Computational ranking of yerba maté small molecules based on their predicted contribution to antibacterial activity against MRSAView study →}, and it appears in anti-herpes work — though the herpes data are for a Melissa extract, not the isolated acid ^{10Reference 10Astani A et al. · 2014Attachment and penetration of acyclovir-resistant herpes simplex virus are inhibited by Melissa officinalis extractView study →}.

Gap: weak evidence tier (screens, extracts, in-silico), no isolate efficacy in vivo ^{9,10Reference 9Rempe CS et al. · 2015Computational ranking of yerba maté small molecules based on their predicted contribution to antibacterial activity against MRSAView study →Reference 10Astani A et al. · 2014Attachment and penetration of acyclovir-resistant herpes simplex virus are inhibited by Melissa officinalis extractView study →}.

Mechanisms

Target / pathway	Effect	Relevant to
Direct ROS/radical quenching (catechol)	scavenges DPPH, superoxide, peroxyl; protects cells	antioxidant
5-lipoxygenase / leukotriene synthesis	selective inhibition (COX spared)	anti-inflammatory
Nrf2 → HO-1/SOD/CAT; NF-κB → TNF-α, iNOS, COX-2	antioxidant induction / anti-inflammatory — strongest for the esters (CAPE), weaker for the free acid	anti-inflammatory (ester-weighted)
β-endorphin release → α1A-adrenoceptor	↑ glucose uptake, ↓ plasma glucose (rat)	antidiabetic
Autophagy reactivation (renal)	restores autophagosomes, ↓ albuminuria (rat)	diabetic nephropathy

Pharmacokinetics

Caffeic acid’s pharmacokinetics are load-bearing and unfavourable. It is absorbed in the small intestine (the best-absorbed of the common hydroxycinnamics in cell-uptake comparisons), but overall oral bioavailability is low and systemic exposure to the intact molecule is small ^{4Reference 4Omar MH et al. · 2012AnimalAbsorption, disposition, metabolism, and excretion of [3-¹⁴C]caffeic acid in ratsView study →}. It undergoes rapid, extensive first-pass and gut-wall metabolism: the preferential route is O-methylation to ferulic acid (COMT), alongside sulfation and glucuronidation — so a substantial fraction is effectively delivered onward as ferulic acid, and ferulic/isoferulic/dihydroferulic acids serve as the urinary biomarkers of intake ^{4Reference 4Omar MH et al. · 2012AnimalAbsorption, disposition, metabolism, and excretion of [3-¹⁴C]caffeic acid in ratsView study →}. Dietary “caffeic acid” is mostly eaten as chlorogenic acid, which is poorly absorbed intact and largely hydrolysed by colonic microbiota to free caffeic acid before further breakdown — so plasma kinetics after coffee or plant intake differ from a pure caffeic acid dose. The practical consequence: oral caffeic acid behaves as a pro-metabolite, and much of any in-vivo activity is really that of ferulic acid and conjugates ^{4Reference 4Omar MH et al. · 2012AnimalAbsorption, disposition, metabolism, and excretion of [3-¹⁴C]caffeic acid in ratsView study →}.

Clinical trials

Human trials of isolated caffeic acid amount to one — a small placebo-controlled RCT of caffeic acid buccal tablets for mouth ulcers, delivered topically ^{3Reference 3Salehi M et al. · 2024RCTEvaluation of caffeic acid mucoadhesive tablets on minor recurrent aphthous stomatitis: a randomized, double-blind, placebo-controlled clinical trialView study →}. Everything else in the human literature is coffee, whole-plant, or ester (chlorogenic acid, rosmarinic acid) — there is no human RCT of orally-dosed isolated caffeic acid for a systemic indication.

Last checked: July 2026.

Toxicity & Safety

At the levels supplied by food and herbs, caffeic acid is well tolerated — it is present in most fruit, vegetables, coffee and many herbs in this database, and no acute human toxicity or overdose syndrome is described for the isolate.

The caveat that must accompany that reassurance is a high-dose rodent carcinogenicity signal. In lifetime feeding studies at 2.0% of the diet — a very high dose — caffeic acid produced forestomach squamous-cell papillomas and carcinomas at high incidence in rats, with lower-incidence kidney changes ^{11Reference 11Hagiwara A et al. · 1991AnimalForestomach and kidney carcinogenicity of caffeic acid in F344 rats and C57BL/6N × C3H/HeN F1 miceView study →}, and a medium-term multi-organ rat model confirmed carcinogenic/co-carcinogenic activity in forestomach and kidney ^{12Reference 12Hirose M et al. · 1998AnimalCarcinogenicity of antioxidants BHA, caffeic acid, sesamol, 4-methoxyphenol and catechol at low doses in a rat medium-term multi-organ modelView study →}. On that animal evidence, IARC classified caffeic acid as Group 2B (“possibly carcinogenic to humans”) in 1993. Three points put this in context: (1) the tumours arise in the rodent forestomach — an organ humans do not have — and are thought to reflect local irritation and hyperplasia at a non-physiological 2% dietary load; (2) the exposures are orders of magnitude above any dietary intake; and (3) genotoxicity testing is largely negative — an in-vivo Pig-a mutation study found no signal for a mutagenic mechanism ^{13Reference 13Bhalli JA et al. · 2019Caffeic acid genotoxicity: correlation of the Pig-a assay with regulatory genetic-toxicology in-vivo endpointsView study →}, consistent with a non-genotoxic, high-dose local effect. The signal is real and worth stating, but it does not translate into a demonstrated human cancer risk at dietary or herbal exposure — and it should never be inverted into an “anticancer” claim (that literature is for the ester CAPE, a different molecule).

No well-characterised human drug interactions exist for the isolate; as a COMT substrate it competes for methylation, and 5-LOX inhibition is a theoretical additive with anti-leukotriene or antiplatelet agents — both speculative, with no clinical reports.

Dosage

There is no established human oral dose for isolated caffeic acid, and nothing here is a recommendation. The only human dosing on record is topical — caffeic acid mucoadhesive buccal tablets applied to mouth ulcers ^{3Reference 3Salehi M et al. · 2024RCTEvaluation of caffeic acid mucoadhesive tablets on minor recurrent aphthous stomatitis: a randomized, double-blind, placebo-controlled clinical trialView study →}, which does not generalise to systemic use. For scale only, preclinical work used ~4 mg/kg i.p. in mice ^{7Reference 7Nogueira Neto JD et al. · 2020AnimalAnxiolytic and antidepressant-like effects of Annona coriacea (Mart.) and caffeic acid in miceView study →} and daily oral dosing over weeks in rats ^{5,6Reference 5Hsu FL et al. · 2003AnimalRelease of β-endorphin by caffeic acid to lower plasma glucose in streptozotocin-induced diabetic ratsView study →Reference 6Matboli M et al. · 2017AnimalCaffeic acid attenuates diabetic kidney disease via modulation of autophagy in a high-fat diet/streptozotocin-induced diabetic ratView study →}; the 2% chronic dietary exposure is the carcinogenicity-signal range and is a hazard dose, not a use dose ^{11Reference 11Hagiwara A et al. · 1991AnimalForestomach and kidney carcinogenicity of caffeic acid in F344 rats and C57BL/6N × C3H/HeN F1 miceView study →}. In practice caffeic acid is obtained through diet and whole herbs; there is no evidence base supporting isolated high-dose supplementation.

References