01 — Evidence
Evidence Summary
Luteolin's clinical record is limited to one small trial, while animal and cell studies build a much broader case across tumour-suppressive and host-protective mechanisms alike — none of it yet tested in a controlled human cancer trial.
Human
Clinical Record
Single Phase I safety trial
A small phase I trial (n=5) found daily oral luteolin safe over six months in men with prostate cancer under active surveillance, with an exploratory reduction in androgen-signalling markers.
No controlled trial has established that luteolin improves tumour control, progression, or survival — the only oncology trial was uncontrolled, with just five participants.
Animal
Preclinical Signal
Carcinogenesis, xenograft & host-protection models
Animal studies report reduced tumour-relevant signalling in a chemically induced colon-carcinogenesis model, reduced tumorigenicity in xenografts, and — separately — genuine host-protective findings in liver injury and cancer-cachexia models.
- Reduced inflammatory (COX-2/iNOS) and Wnt signalling in a colon carcinogenesis model
- Reduced tumorigenic potential in an esophageal xenograft, without significant toxicity
- Apoptosis via ER stress confirmed in a glioblastoma xenograft
- Preserved liver function and muscle mass in separate host-protection models
In Vitro
Cell Model Data
Broad mechanism panel; concentration-dependent
Luteolin has been studied across many cancer cell types, with several independent mechanistic threads — but the concentrations needed are far above what oral dosing is likely to achieve in free, unconjugated form.
- Direct STAT3 disruption via Hsp90 binding, and indirect suppression downstream of other mechanisms
- Suppressed FAK/PI3K/AKT, Wnt/β-catenin, and Ras/Raf/MEK/ERK growth signalling
- Intrinsic and extrinsic apoptosis induction, both independently confirmed
- NF-κB-dependent sensitisation to TNF-α-induced cell death
Human
Clinical Record
Human oncologic outcome data for luteolin remain limited to a single small phase I study: five men with diagnosed, low- to intermediate-risk prostate cancer under active surveillance took 50 mg of oral luteolin daily for six months, and the trial reported the regimen as safe over that period.[1] With an n of 5 and no comparator arm, this is a feasibility and tolerability signal, not efficacy evidence.
Continue reading — full research detail+
All five participants underwent a follow-up protocol biopsy after treatment: two showed a favourable response, one had stable disease, and two showed disease progression and subsequently underwent surgery.[1] Immunohistochemical analysis found decreased expression of the androgen receptor and NKX3.1 after treatment — but this biomarker change was reported specifically in noncancerous prostate lesions, not in the cancerous tissue itself, an important limitation on what it can be read to show.[1] A separate, non-oncology human data point exists but sits apart from the picture above: a randomised, double-blind, placebo-controlled trial of a combination nutraceutical (chlorogenic acid plus luteolin-7-glucoside, 150 mg/day, six months) in adults with metabolic syndrome reported improved cardiometabolic and hepatic markers over placebo, and a post-hoc subgroup analysis found the same benefit in the pre-obesity subset specifically.[26,27] This is a combination product with a small luteolin-glucoside content, in a non-oncology population with cardiometabolic endpoints — it doesn't add to the oncologic evidence picture, but it is the only other human luteolin dosing data identified.
Signal maturity: luteolin's only oncology trial is a small, uncontrolled safety study in men with diagnosed prostate cancer. No controlled trial has established that luteolin affects tumour control, progression, or survival — every finding below this point is preclinical.
Animal
Preclinical Signal
Six independent animal-model findings anchor luteolin's preclinical case, split between tumour-directed and host-protective effects. In an azoxymethane-induced mouse model of colon carcinogenesis — designed to mimic how colorectal cancer actually develops rather than implanting an existing tumour — oral luteolin reduced both inflammatory (iNOS, COX-2) signalling and Wnt/β-catenin pathway activity, alongside reduced cell proliferation.[13,14]
Continue reading — full research detail+
In a paclitaxel-resistant esophageal squamous cell carcinoma xenograft, luteolin reduced tumorigenic potential in nude mice without significant systemic toxicity, alongside in vitro suppression of the FAK/Src/PI3K/AKT signalling axis in the same resistant cell line.[4] In a nude-mouse glioblastoma xenograft, luteolin induced apoptosis via a lethal endoplasmic-reticulum stress cascade, with the antioxidant N-acetylcysteine reversing both the signalling and the resulting cell death — direct evidence the mechanism runs through reactive oxygen species.[18] In a CT26 colorectal-carcinoma xenograft, liposome-encapsulated luteolin outperformed free luteolin at the same dose, a formulation finding detailed further under Pharmacokinetics and Administration below.[6]
Separately, two animal findings are host-protective rather than tumour-suppressive, and are detailed in full under Protect in Pathway Interaction Profile below: in a mouse model of alcohol-induced liver injury, luteolin reduced hepatic damage markers via stabilisation of the transcription factor NRF2;[9] and in a Lewis lung cancer mouse cachexia model, luteolin preserved cardiac and skeletal muscle mass that would otherwise be lost to tumour-driven wasting.[20]
Signal maturity: animal findings span six distinct model contexts — AOM colon carcinogenesis, ESCC xenograft, glioblastoma xenograft, CT26 xenograft, ethanol-induced liver injury, and Lewis lung cachexia — and two different kinds of benefit, direct tumour suppression and host protection. No single mechanism has been corroborated across more than one model, and none has yet been tested in a human oncology trial.
In Vitro
Cell Model Data
Luteolin has been tested across a wide panel of human cancer cell lines — melanoma, esophageal, breast, glioblastoma, colon, gastric, cervical, and hepatocellular models all report independent mechanistic findings. Nearly all of this work used luteolin concentrations of 10–100 µM, generally as a sustained exposure — a range that mostly sits above what free luteolin is likely to reach and sustain in human plasma after standard oral dosing, given how rapidly it's conjugated (see Pharmacokinetics and Administration below).
Continue reading — full research detail+
Luteolin has been reported to bind directly to the chaperone protein Hsp90, disrupting its stabilising interaction with STAT3 and driving proteasome-dependent degradation of phosphorylated STAT3 — inducing apoptosis selectively in cancer cell lines (HeLa, HepG2) while producing only slight cytotoxicity in normal cell lines tested at the same concentrations.[11] A second, independent study found luteolin dephosphorylates STAT3 in gastric cancer cells via the phosphatase SHP-1, confirmed in a xenograft mouse model.[12]
Growth-signalling suppression has been reported across three separate cascades: FAK/PI3K/AKT in esophageal cancer cells (with molecular docking predicting direct interactions between luteolin and the FAK, Src, and AKT active sites — a computational prediction, not an empirical binding assay),[4] Wnt/β-catenin in an in vivo colon-carcinogenesis model,[13] and Ras/Raf/MEK/ERK in breast cancer cells via a microRNA-203-dependent mechanism.[16]
Both major apoptotic routes have independent primary-source support: luteolin's Hsp90/STAT3 mechanism above drives intrinsic (mitochondrial) apoptosis,[11] while a separate study found luteolin sensitises colorectal and cervical cancer cells to TNF-α-induced extrinsic apoptosis by inhibiting NF-κB-driven expression of the antiapoptotic genes A20 and c-IAP1.[17] The same NF-κB suppression underlies luteolin's Contain-relevant inflammatory signalling, detailed under Pathway Interaction Profile below.
Signal maturity: cell-model research offers the deepest mechanistic detail of any tier and the broadest cell-line panel tested for this compound, but it's also where the concentration gap matters most — nearly every reported effect requires sustained exposure at concentrations current oral dosing has not been shown to sustain, even where a brief, high-dose plasma peak can overlap the low end of that range.
Advertisement
02 — Pathways
Pathway Interaction Profile
Luteolin engages several distinct biological pathways relevant to tumour behaviour, grouped below by the functional role each one supports. This includes direct anti-tumour mechanisms and, further down, a separate set of pathways supporting the body's own resilience.
Luteolin's Contain classification rests on reported suppression of hypoxia-driven angiogenic signalling, the inflammatory signalling that primes surrounding tissue for a tumour, and the cellular transition that lets cancer cells detach and invade — corroborated in vivo for the inflammatory component specifically, which is what moves this role from partial to active.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Angiogenesis / VEGF / HIF-1α
Luteolin suppressed HIF-1α/VEGF-mediated epithelial-mesenchymal transition and angiogenic signalling in melanoma cells and endothelial cells at sub-cytotoxic concentrations, shifting EMT markers toward an epithelial phenotype via reduced phospho-AKT.[2] A related finding in M2-like tumour-associated macrophages under hypoxic mimicry showed 20 µM luteolin reduced HIF-1α and phosphorylated STAT3, lowering downstream VEGF and MMP-9 output.[3]
COX-2 / PGE₂
In an azoxymethane-induced mouse model of colon carcinogenesis, oral luteolin reduced expression of both inducible nitric oxide synthase and COX-2 relative to vehicle-treated controls — the finding that moves this role beyond in-vitro-only evidence.[14]
NF-κB / TNF-α / IL-6 inflammatory axis
Luteolin pretreatment inhibited TNF-α-induced NF-κB activation in colorectal and cervical cancer cells, suppressing the antiapoptotic genes A20 and c-IAP1 and thereby sensitising these cells to TNF-α-induced apoptosis — a finding that also supports the Attack classification below, via a different downstream readout of the same mechanism.[17]
Prevent Tumour Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
EMT & metastatic invasion
The same melanoma and endothelial-cell study above demonstrated direct suppression of EMT markers and reduced migratory and invasive behaviour in luteolin-treated cells.[2]
Luteolin's Starve classification is described as partial because, while two mechanistically distinct routes are each supported by a real primary finding, both are in-vitro only, at concentrations well above expected free-luteolin plasma exposure, with no animal-model corroboration identified for either.
Metabolic Flexibility Suppression
Research concerning metabolic adaptation and switching between fuel sources under pressure.
Autophagy & lysosomal system
In human hepatocellular carcinoma cells, luteolin (25–100 µM) increased autophagosome number, promoted LC3B-I-to-II conversion, and increased Beclin-1 expression; co-treatment with the autophagy inhibitor chloroquine reduced luteolin's apoptotic effect, indicating autophagy contributes mechanistically to — rather than merely accompanies — the resulting cell death.[19]
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
NRF2–GSH redox axis
In human colon cancer cells, luteolin decreased methylation of the NRF2 promoter — via inhibition of DNA methyltransferases and increased TET1 demethylase activity — increasing NRF2 expression; the resulting NRF2 protein interacted directly with the tumour suppressor p53, and this interaction was necessary for luteolin-induced apoptosis in this model, confirmed by knockdown experiments.[15] This is the tumour-context counterpart to a host-protective NRF2 finding detailed under Protect below — see the conflict-check discussion there.
Luteolin's Weaken classification reflects reported disruption of the growth and survival signalling tumour cells depend on to keep dividing, reached through three converging signalling cascades and a lethal stress response — corroborated beyond pure in-vitro data in two independent xenograft and carcinogenesis models.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
PI3K–AKT–mTOR
In paclitaxel-resistant esophageal squamous cell carcinoma cells, luteolin reduced phosphorylation along the FAK/Src/PI3K/AKT axis, and molecular docking predicted direct interactions between luteolin and the active sites of FAK, Src, and AKT (a computational prediction, not an empirical binding assay), decreasing proliferation and inducing cell-cycle arrest and apoptosis; in a nude-mouse xenograft of the same resistant line, luteolin reduced tumorigenic potential without significant systemic toxicity.[4]
Wnt / β-catenin
In an azoxymethane-induced mouse model of colon carcinogenesis, oral luteolin decreased active β-catenin, decreased phosphorylated (inactive) GSK-3β, and decreased cyclin D1 expression, reducing AOM-induced colonic cell proliferation — a second independent animal-model finding supporting this role.[13]
RAS–RAF–MEK–ERK (MAPK)
In breast cancer cells, luteolin reduced Ras and Raf expression and MEK/ERK phosphorylation; this effect, along with luteolin's broader anti-proliferative and anti-EMT activity, was substantially reversed by silencing microRNA-203, indicating MAPK suppression in this model runs through a microRNA-203-dependent mechanism.[16]
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
In two glioblastoma cell lines, luteolin increased intracellular reactive oxygen species, triggering phosphorylation of PERK and eIF2α and induction of ATF4, CHOP, and cleaved caspase-12 — a full canonical ER-stress apoptotic cascade — alongside mitochondrial dysfunction; the antioxidant N-acetylcysteine reversed both the signalling and the resulting apoptosis, and the anticancer effect was also confirmed in a nude-mouse xenograft.[18]
Luteolin's Attack classification is described as partial because, despite two mechanistically distinct and independently confirmed routes to direct tumour-cell killing, both remain in-vitro only, at concentrations well above expected free-luteolin plasma exposure, with no animal-model corroboration identified for either as a standalone killing mechanism. Worth noting: other luteolin-induced cell-death mechanisms described on this page — the ER-stress-driven apoptosis detailed under Weaken below, for instance — do have animal-model support. That finding sits under Weaken rather than Attack because of how its underlying mechanism is classified within this framework's structure, not because it's weaker evidence than the mechanisms below; a reader weighing luteolin's overall evidence for inducing tumour-cell death should read Weaken and Attack together rather than treating Attack's partial status as the full picture.
Direct Tumour-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Luteolin bound directly to Hsp90 — confirmed by molecular modelling and a surface plasmon resonance binding assay — blocking Hsp90's ATP-binding pocket and disrupting its stabilising association with STAT3. This promoted proteasome-dependent degradation of phosphorylated STAT3, inducing apoptosis in cervical and hepatocellular carcinoma cells at 20–60 µM while producing only slight cytotoxicity in normal cell lines tested at the same concentrations — a notable in-vitro selectivity signal, backed by an empirical binding assay rather than computational prediction alone.[11]
Extrinsic apoptosis (death receptors)
Luteolin pretreatment sensitised colorectal and cervical cancer cells to TNF-α-induced apoptosis; mechanistically, luteolin's inhibition of NF-κB suppressed the antiapoptotic genes A20 and c-IAP1, and also led to augmented and prolonged JNK activation, both contributing to enhanced death-receptor-pathway apoptotic signalling.[17]
Luteolin's Protect classification covers two distinct kinds of evidence, and neither is human clinical data. One is preclinical host-protection evidence relevant to oncology treatment — cachexia mitigation in a mouse model, and protection of cultured cells against a chemotherapy drug's toxicity — which has no defined pathway mechanism by nature and is detailed below rather than carrying a pathway card. The other is mechanism-based evidence that luteolin strengthens the body's own tissue resilience independent of any drug interaction, which does carry a real pathway card, set out further down.
Oncology Host-Status
Cachexia mitigation — in a Lewis lung cancer mouse model, oral luteolin preserved gastrocnemius and cardiac muscle mass relative to tumour-bearing vehicle controls. Mechanistically, luteolin reduced circulating TNF-α and IL-6, decreased expression of the muscle-wasting E3 ubiquitin ligase MuRF1 by inhibiting NF-κB activation, and reduced p38 MAPK expression — the first report of luteolin acting through this specific axis in a cachexia model.[20]
Toxicity reduction — in cultured auditory cells, luteolin protected against cisplatin-induced apoptosis by inducing heme oxygenase-1 via ERK activation; pharmacological HO inhibition and HO-1 antisense knockdown both abolished the protection, and luteolin reduced cisplatin-induced caspase-3 activation and mitochondrial dysfunction.[23] Cisplatin-induced ototoxicity is a well-documented, dose-limiting side effect of platinum-based chemotherapy in humans, but this finding does not establish prevention of ototoxicity in animals or people — it showed protection of a cultured cell model, not a clinical or even whole-animal outcome. It also doesn't establish that luteolin can protect normal auditory tissue without similarly protecting tumour cells from cisplatin, which would work against the drug's own purpose; that question was not addressed in this experiment. A complementary finding in the same cell line showed luteolin also protects against oxidative-stress-induced senescence more generally, via a distinct SIRT1/p53 mechanism.[24]
Hepatic Resilience & Clearance
Human and preclinical research on hepatic enzyme systems, bile-acid handling, and liver-related markers.
NRF2-mediated hepatoprotection
In a mouse model combining chronic and single-binge ethanol exposure, luteolin restored stability of the transcription factor NRF2, blocking nuclear accumulation of ACSS2 and associated histone H3 acetylation; this reduced hepatic lipogenesis, lowered plasma AST/ALT, and reduced hepatic oxidative-stress markers while restoring glutathione and catalase activity.[9]
A second, fully independent study found this same underlying mechanism — NRF2 stabilisation and activation — drives an apoptotic, tumour-suppressive outcome in a different tissue context: in human colon cancer cells, luteolin-induced NRF2 upregulation interacted with p53 to promote cancer-cell apoptosis, detailed under the NRF2–GSH redox axis pathway in Starve above.[15] The same mechanism produces genuine host benefit in normal liver tissue and genuine tumour suppression in a different cellular context — not a contradiction, but two favourable outcomes depending on where in the body it's happening.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Luteolin reduced inflammatory (COX-2) signalling in an animal model of colon cancer development — one of the few findings on this page confirmed beyond cell-culture evidence.
Prevent Tumour Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
Luteolin has been reported to suppress the cellular changes tumour cells use to detach and invade in a melanoma model, alongside reduced angiogenic markers.
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
Luteolin has been reported to increase NRF2 activity in colon cancer cells, driving apoptosis via a p53 interaction — a preclinical, cell-model finding.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
Luteolin has been reported to suppress growth signalling in esophageal cancer cells, confirmed in vivo — xenografted tumorigenicity was reduced without significant toxicity.
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
Luteolin has been shown to induce a lethal ER-stress cascade in glioblastoma cells, confirmed in a xenograft model.
Direct Tumour-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Reported via direct Hsp90 binding and STAT3 degradation, selective for cancer over normal cell lines at the same concentrations. Evidence is preclinical — in vitro concentration limitations apply.
Hepatic Resilience & Clearance
Human and preclinical research on hepatic enzyme systems, bile-acid handling, and liver-related markers.
NRF2-mediated hepatoprotection
Luteolin has been reported to reduce liver injury markers in an alcohol-induced liver damage animal model, via stabilisation of the antioxidant transcription factor NRF2.
Advertisement
03 — Pharmacokinetics
Pharmacokinetics and Administration
How luteolin moves through the body has more bearing on real-world use than the mechanistic evidence alone — most of what enters the bloodstream after an oral dose isn't luteolin at all, but the metabolites it's converted into within the first hour.
Absorption
Free luteolin has poor oral bioavailability, around 4%, and is rapidly converted to glucuronidated and methylated metabolites, whose activity generally differs from — and is often weaker than — the free compound. Natural co-occurring compounds and engineered delivery systems can meaningfully improve absorption.
The Concentration Gap
Most cell-model effects require concentrations of 10,000–100,000 nM. The highest free-luteolin plasma level confirmed for an oral dose is a brief peak around 19,200 nM that falls below detection within about an hour — overlapping the low end of the cell-study range, but only briefly, at a very high dose.
Clinical Dose Context
The only human oncology dosing data is 50 mg/day from a small prostate cancer trial. A separate, non-oncology trial used a 150 mg/day combination product with a much smaller luteolin-glucoside fraction.
Formulation Effects
A P-glycoprotein-targeting delivery system produced a roughly 29-fold increase in absorption; natural-extract co-administration produced a roughly 4-fold increase — tested independently, not head-to-head.
Metabolism
Metabolised via glucuronidation and methylation, with CYP1A1 and CYP1A2 bioactivation shown to increase genotoxicity in human cell studies — a mechanism distinct from luteolin's drug-interaction profile.
Co-Dosing Considerations
A potential interaction exists with statins: luteolin inhibits a liver transporter several statins depend on for uptake, in vitro, at a potency comparable to a known clinical interaction drug — but no human luteolin-statin data exists yet.
Absorption
Oral luteolin has poor native bioavailability. Following ingestion, it is rapidly and extensively metabolised to glucuronide conjugates and to methylated derivatives (chrysoeriol, diosmetin) via UDP-glucuronosyltransferases and catechol-O-methyltransferases; luteolin-3′-glucuronide showed the highest systemic exposure of the nine metabolites identified in rat plasma and bile, and unconjugated luteolin itself is described as hardly detectable in vivo.[5] Critically, these metabolites are not simply inactive byproducts: the same source describes luteolin as partially exerting its biological effects through its metabolites, and a separate line of research has found that luteolin-glucuronide retains measurable — generally weaker — biological activity of its own, and can be deconjugated back to free luteolin locally at sites of inflammation. Low measured plasma levels of the free aglycone therefore don't necessarily mean an absence of biologically relevant exposure; they mean the exposure is happening in a different chemical form whose own activity is less well characterised.[5]
Confirmed rat pharmacokinetic data quantify the scale of this constraint: absolute oral bioavailability of free luteolin was only 4.10% at a 50 mg/kg dose, and free-luteolin plasma concentrations fell below the assay's limit of quantification within one hour of oral dosing, despite a peak concentration of 5.5 µg/mL at 5 minutes — consistent with a large volume of distribution, high clearance, and probable enterohepatic recirculation.[21] A second, independent rat study at a lower oral dose (14.3 mg/kg) confirmed a peak plasma concentration of 1.97 µg/mL for pure luteolin; co-administering luteolin as part of a peanut hull extract rather than the pure compound more than quadrupled this figure, indicating co-occurring compounds in the natural extract matrix meaningfully improve intestinal absorption.[28]
The Concentration Gap
This is a constraint that should shape how everything in Evidence Summary above gets read, though the picture is more nuanced than a simple gap. Nearly every mechanistic finding described in this Brief used luteolin concentrations of 10–100 µM in cell culture, generally as a sustained exposure. Free luteolin is converted to conjugated metabolites very quickly after an oral dose, so most of that range sits well above what's achievable — but not all of it, as the table below shows.
| Benchmark | Concentration | Interpretation |
|---|---|---|
| In vitro effective range across pathway studies | 10,000–100,000 nM | The concentration range used throughout the mechanistic literature described above |
| Rat oral Cmax, pure luteolin (50 mg/kg dose, peak at 5 min) | ~19,200 nM | The highest free-luteolin plasma concentration confirmed in this Brief's primary literature — but transient, falling below quantification within about an hour[21] |
| Rat oral Cmax, pure luteolin (14.3 mg/kg dose) | ~6,900 nM | A second, independent oral PK study at a lower dose[28] |
These are rat, not human, figures, and the higher of the two peaks is brief: it occurs at a very high dose, lasts only minutes before falling below quantification, and reflects a single acute dose rather than the sustained exposure most cell studies actually apply. A transient rat peak overlapping the low end of the in vitro range doesn't reproduce the duration, human relevance, tissue concentration, or repeated exposure most mechanistic experiments use — and no comparable human free-luteolin PK dataset exists to confirm ordinary oral supplementation reaches or sustains concentrations anywhere near this range.
Clinical Dose Context
Human dosing data for luteolin is sparse and comes from two very different contexts that shouldn't be conflated: a small oncology safety trial using isolated luteolin, and a larger, non-oncology combination-product trial using a much smaller luteolin-glucoside fraction.
| Context | Dose | Source |
|---|---|---|
| Phase I prostate cancer safety trial (n=5) | 50 mg/day, 6 mo | Safe over 6 months; exploratory AR/NKX3.1 reduction[1] |
| Non-oncology combination trial (n=50 subgroup) | 150 mg/day Altilix® | ~3–6 mg/day luteolin-glucoside; cardiometabolic endpoints[26,27] |
| AOM colon carcinogenesis model | 1.2 mg/kg/day | Reduced COX-2/iNOS and Wnt pathway activity[13,14] |
| Rat PK reference doses | 14.3 & 50 mg/kg | Two independent oral/IV pharmacokinetic studies[21,28] |
Formulation Effects
Three independent delivery strategies have been shown to improve on luteolin's poor native absorption, each by a different mechanism — none of them tested against one another head-to-head, so the comparison below is across separate studies, not a single trial.
| Formulation | Mechanism | Study detail | Citation |
|---|---|---|---|
| Liposomal (Lipo-Lut) | Encapsulation improves solubility & delivery | Superior in-vivo antitumour effect vs. free luteolin, equal 50 mg/kg IV dose, CT26 xenograft | [6] |
| TPGS-based self-microemulsifying system | Nanodroplet solubilisation + P-gp efflux inhibition | ~29-fold AUC increase vs. unformulated luteolin, animal PK | [10] |
| Peanut hull extract co-administration | Natural co-occurring compounds aid absorption | ~4-fold Cmax/AUC increase vs. isolated pure luteolin, equal dose | [28] |
None of the three studies reported a change in luteolin's underlying mechanism of action — all three address delivery only, and no formulation-specific safety signal has been reported for any of them.
Metabolism
Glucuronidation and methylation are the two dominant, interdependent metabolic pathways, generating nine identified metabolites in rat plasma and bile; the two pathways compensate for one another when either is inhibited, with glucuronidation predominating overall.[5] Metabolic bioactivation is also directly relevant to luteolin's safety profile: in human lymphoblastoid cells individually expressing fourteen different human cytochrome P450 enzymes, luteolin's cytotoxicity and genotoxicity were significantly increased specifically in cells expressing CYP1A1 or CYP1A2, indicating that metabolism by these two enzymes converts luteolin to a more reactive, more toxic form in vitro.[22]
Co-Dosing Considerations
Luteolin itself — not only its circulating conjugates — inhibits the hepatic uptake transporter OATP1B1 in vitro, at a potency (IC50 0.85 µM) the reviewing literature compares to clarithromycin, a drug with clinically demonstrated statin-interaction risk in humans — though that comparison describes in-vitro potency only, not an equivalent level of clinical evidence, since no human luteolin-statin data exists.[8] A dedicated review of flavonoid-statin interactions names luteolin among the flavonoids with mechanistic potential to raise statin plasma levels via this route.[25] Combined with luteolin's independently confirmed CYP3A4 inhibition in vitro, this gives two separate plausible pharmacokinetic routes to elevated statin exposure, plus a caveated pharmacodynamic one detailed below — none of them confirmed in humans. Each row is flagged by the most cautious guidance its cited evidence supports.
Discuss whether to combine, separate, or avoid luteolin and a medication with your treating oncology team or physician.
| Flag | Interaction |
|---|---|
| Caution | Statins — luteolin inhibits OATP1B1 in vitro (IC50 0.85 µM), a hepatic uptake transporter several statins depend on to varying degrees, and also inhibits CYP3A4 in vitro at micromolar concentrations, most relevant to simvastatin and lovastatin, and to a lesser extent atorvastatin (pravastatin, rosuvastatin, and pitavastatin rely on different metabolic pathways, though transporter interactions may still be relevant).[7,8] A related compound, luteolin-7-glucoside, was separately shown to both repress HMG-CoA reductase gene expression and directly inhibit the enzyme in vitro — the same enzyme statins target — though only at high, non-physiologically-relevant concentrations per the study's own authors.[29] No human luteolin-statin pharmacokinetic study, case series, or clinical myopathy signal has been reported; the evidence here is in-vitro and mechanistic, creating a plausible interaction concern for high-dose luteolin supplements rather than an established one. |
| Caution | OATP1B1/OATP2B1 substrates generally — sulfate and glucuronide conjugates of luteolin, the dominant circulating forms in vivo, also potently inhibited these transporters in vitro; the same study found weak or no CYP2C9/CYP2C19/CYP3A4 inhibition by these conjugates specifically.[8] |
| Caution | CYP3A4/CYP3A5 substrates with a narrow therapeutic window — luteolin and the related flavone diosmetin inhibited CYP3A4- and CYP3A5-mediated midazolam metabolism in human liver microsomes at micromolar concentrations.[7] |
Advertisement
04 — Onset & Washout
Onset and Washout
Luteolin's timeline is defined almost entirely by how quickly it disappears: absorption is fast, but so is conversion to conjugated metabolites whose own activity is less well characterised — and how long sustained dosing needs to run before pathway-level effects would be expected to matter isn't yet established in the literature.
Immediate Onset
Free luteolin plasma concentrations peak within minutes of oral dosing but fall below quantification within about an hour, reflecting how rapidly the compound is conjugated.
Steady State
In a rat study using intravenous dosing, elimination half-lives were roughly 5–9 hours. The oral terminal phase couldn't be calculated because levels fell below detection too quickly — so this figure describes IV clearance, not confirmed oral kinetics.
Accumulated Effect
How many days of sustained dosing pathway-level effects require to emerge hasn't been quantified for this compound.
Dosing Pattern in Studies
Every animal study using luteolin for a pathway-level or host-protective effect used repeated, not single-dose, administration.
Washout
How long luteolin's influence can take to clear before it stops being a relevant factor for co-administered medications.
Based on intravenous rat half-lives of roughly 5–9 hours, one to two days is a rough pharmacokinetic estimate for clearing most circulating luteolin — not a confirmed human oral washout figure, since no such study exists.
Two Distinct Clocks
Luteolin's timeline data comes almost entirely from pharmacokinetic studies rather than from repeated-dosing studies that measured when a pathway-level effect actually emerged — so the direct-pharmacology clock (Clock A) is well characterised, while the downstream-phenotype clock (Clock B) can only be described qualitatively.
| Clock A — Direct Pharmacology | Clock B — Downstream Phenotype | |
|---|---|---|
| Latency | Fast — plasma peak within minutes | Not quantified — every animal study used repeated dosing |
| Persistence | Short after oral dosing — free compound below detection within ~1 hour; IV half-life (rat) ~5–9 hours | Not established for luteolin specifically |
| What it covers | Free-luteolin plasma exposure, acute CYP3A4/OATP1B1 interaction risk | Pathway-level changes across Contain, Starve, Weaken, and Attack detailed above; host-protective effects under Protect |
The direct-pharmacology clock (Clock A) is fast and shallow: free-luteolin plasma exposure peaks within minutes of a dose but falls below detection within about an hour, consistent with rapid conjugative metabolism. No single-dose study has measured how long a downstream pathway effect (Clock B) takes to emerge or fade for luteolin — every relevant animal study used repeated, multi-day or multi-week dosing protocols, confirming that sustained exposure is required without establishing the specific latency window.
Steady State and Accumulation
No steady-state accumulation data specific to repeated human or animal oral dosing exists for luteolin. A rat study measured elimination half-lives of roughly 5–9 hours after intravenous dosing; the oral terminal elimination phase could not be calculated in that same study because concentrations fell below the assay's detection limit too quickly. Applying the IV figure to oral, repeated dosing is a rough extrapolation, not a directly measured finding — on that basis alone, meaningful plasma accumulation across a normal once- or twice-daily interval seems unlikely, but this hasn't been confirmed.
Dosing Pattern in Studies
The mechanistic case for luteolin is broad (see Pathway Interaction Profile above), but nearly all of it is preclinical, and the concentration gap detailed under Pharmacokinetics and Administration above means retail-range oral dosing is unlikely to reach mechanistically meaningful free-luteolin exposure for most of the pathways described. The one human oncology trial used isolated luteolin at a low daily dose over six months, which is the closest this compound has to a studied dosing pattern — but it measured safety and exploratory biomarkers, not a tumour-relevant outcome.
For any application depending on the preclinical Contain, Starve, Weaken, or Attack pathways detailed above, see The Concentration Gap under Pharmacokinetics and Administration for what "meaningful" exposure would actually require, and how far current oral dosing sits from it.
Washout
Washout here is a rough pharmacokinetic extrapolation, not clinical guidance: a rat study measured elimination half-lives of roughly 5 to 9 hours for free luteolin and its conjugates, but only after intravenous administration — the oral terminal elimination phase couldn't be calculated in that study because concentrations fell below the assay's detection limit within about an hour. Applying the IV figure gives a rough one-to-two-day estimate for clearing most circulating compound, but no luteolin-specific human washout study exists to confirm it, and no tissue-accumulation pattern extending clearance beyond this estimate has been reported.
Advertisement
05 — Safety
Safety Profile
Luteolin's human safety record is thin rather than concerning — the single trial that exists reported no adverse events, but that trial was small, and several categories of safety data that would normally anchor this section simply haven't been established yet.
No adverse events in the one available trial — five patients, six months, no reported adverse events; too small to characterise broader tolerability.
Metabolic bioactivation and genotoxicity — an in-vitro mechanism, CYP1A1/CYP1A2-dependent, confirmed in human cell culture.
Potential statin interaction — luteolin inhibits a liver transporter and enzyme relevant to several statins' uptake and clearance, in vitro; no human data exists yet to confirm clinical relevance.
What's genuinely unknown — no dose-response adverse-effect data, regulatory signal, or established contraindication currently exists for this compound.
Adverse Effects in Human Trials
Human data is limited to one small trial and a weak indirect data point from a non-oncology combination-product study — neither constitutes a real adverse-effect frequency table, and this section doesn't pretend otherwise.
In the phase I prostate cancer trial, daily 50 mg oral luteolin for six months produced no reported adverse events in any of the five participants.[1] Separately, a randomised, double-blind, placebo-controlled trial of a chlorogenic-acid-and-luteolin-glucoside combination product enrolled 100 adults with metabolic syndrome for six months with monthly adherence checks and reported cardiometabolic benefit over placebo; a later post-hoc analysis of the 50-person pre-obesity subgroup from that same trial found the same benefit in that subset specifically, rather than representing a second, independent exposure cohort.[26,27] No explicit adverse-event frequency table was located in the available extracts of either paper. This is weak indirect evidence of tolerability, not a substitute for an actual adverse-event table.
Metabolic Bioactivation and Genotoxicity
Luteolin's own metabolism is a documented in-vitro genotoxicity mechanism, separate from its drug-interaction profile. In human lymphoblastoid cells individually expressing fourteen different human cytochrome P450 enzymes, luteolin produced concentration-dependent cytotoxicity, apoptosis, DNA damage, and chromosome damage; this toxicity was significantly increased specifically in cells expressing CYP1A1 or CYP1A2.[22] The same study found that these cells convert most luteolin to diosmetin, a less-toxic metabolite, within 24 hours — and that CYP1A1/CYP1A2 expression partially blocks this detoxifying conversion, which is the mechanistic basis for the increased toxicity in those specific cell lines. This is an in-vitro finding at concentrations not established as directly comparable to human oral exposure, but it is a real, independently confirmed mechanistic signal rather than a theoretical one.
What Remains Unestablished
No dose-dependent human adverse-effect frequency data, no regulatory signal (an FDA or EMA warning, a drug-induced liver injury registry entry), and no absolute contraindication (pregnancy or otherwise) currently exists in the published literature for luteolin. This isn't the same as luteolin being established as free of these risks — the specific studies that would establish them one way or the other don't appear to exist yet. The drug-interaction guidance covered under Pharmacokinetics and Administration's Co-Dosing Considerations above is the one area of real, mechanistically grounded safety-relevant evidence available, and isn't repeated here.
06 — Sourcing
Sourcing Guide
Formulation matters more than usual for luteolin specifically — free luteolin's poor native bioavailability means the choice between standard and enhanced-delivery forms genuinely changes how much reaches the bloodstream, as detailed under Pharmacokinetics and Administration above. Visit the Sourcing Guide for what's currently available to retail consumers.
Sourcing Guide06 — Literature
References
Last reviewed: July 2026