01 — Evidence
Evidence Summary
Quercetin's anti-tumor mechanism set is broad and partly confirmed in animals, but its only human oncology trial is a single Phase I that also revealed a toxicity ceiling — while the strongest human evidence sits on the host side, in blood-pressure, inflammatory-marker and senolytic data.
Human
Clinical Record
One Phase I oncology trial; host-physiology RCTs
The only interventional oncology trial is a 1996 Phase I intravenous dose-escalation, which was dose-limited by kidney toxicity and produced only anecdotal tumor-marker responses. What has real human evidence is host physiology and host aging.
- Modest blood-pressure reduction across randomized-trial meta-analyses
- Lower C-reactive protein and improved MASLD liver enzymes (general, not oncology-specific)
- Senescent-cell clearance shown for dasatinib+quercetin (the combination, not quercetin alone)
Animal
Preclinical Signal
Xenograft, chemoprevention & hematologic models
Multiple in-vivo models corroborate the tumor-directed case — including an oral (dietary) chemoprevention readout, which is the strongest kind for a poorly-absorbed compound.
- Reduced NF-κB signaling and growth in an NSCLC xenograft
- Dietary quercetin cut colon aberrant crypt foci (oral chemoprevention)
- Suppressed STAT3-driven melanoma metastasis; reduced hexokinase-2 in a liver xenograft
In Vitro
Cell Model Data
Broad mechanism panel; concentration-limited
Deep and broad across many tumor types, but nearly all of it used concentrations far above what oral dosing achieves — and the metabolites that actually circulate lose the antiproliferative activity.
- Suppressed PI3K/AKT/mTOR, MAPK/ERK, STAT3 and Wnt/β-catenin signaling
- Bax-dependent apoptosis and ferroptosis, both tumor-selective
- Effects generally at 25–100 µM, above achievable free-quercetin plasma levels
Human
Clinical Record
Quercetin's human oncology outcome evidence is genuinely thin, and honesty about that is the point. The only interventional oncology trial is a 1996 Phase I intravenous dose-escalation: quercetin given as an IV bolus from 60 to 1700 mg/m² produced dose-limiting nephrotoxicity at 1700 mg/m² (recommended Phase II dose 1400 mg/m²), inhibited lymphocyte tyrosine-kinase phosphorylation in 9 of 11 patients, and yielded two anecdotal marker responses.[1]
Continue reading — full research detail+
In that trial a cisplatin-refractory ovarian patient's CA-125 fell from 295 to 55 U/mL and one hepatoma patient's AFP fell — a feasibility and pharmacodynamic signal, not efficacy, and a clear demonstration that quercetin has a real toxicity ceiling of its own.[1] Chemoprevention epidemiology does not rescue the picture: in the Women's Health Study (38,408 women, 11.5 years), higher total-flavonoid and quercetin intake showed no association with total or site-specific cancer (relative risk ~0.97, P-trend 0.72).[29] Where human evidence is real is general host physiology, not proven oncology support. Meta-analyses of randomized trials show quercetin modestly lowers blood pressure (systolic −3.04 mmHg, greater at ≥500 mg/day)[30] and C-reactive protein (with LDL reduction but no effect on IL-6/TNF, and high heterogeneity across trials),[31] and improves ALT/AST/GGT in metabolic-dysfunction-associated steatotic liver disease[32] — genuine human outcomes, but none of them shown to improve treatment tolerance or cancer-specific outcomes.
A prominent host-aging thread is the dasatinib+quercetin (D+Q) senolytic combination — but the evidence is for the combination, and quercetin's independent contribution is unknown, since dasatinib is itself a potent senolytic. Small open-label pilots reported that D+Q altered senescence-related endpoints: 14 idiopathic-pulmonary-fibrosis patients improved walking distance, gait speed and chair-stands,[33] and 9 diabetic-kidney-disease patients showed reduced adipose senescent-cell burden and circulating senescence-associated secretory factors (IL-1α, IL-6, MMP-9, MMP-12).[34] These were small, non-randomized, non-cancer pilots; a subsequent Phase 2 randomized trial of D+Q in postmenopausal women did not meet its primary bone-resorption endpoint, with benefit appearing only in an exploratory high-senescent-burden subgroup.[60] This is emerging, combination-specific evidence — not established benefit for quercetin alone.
Signal maturity: there is no controlled human oncology efficacy trial for quercetin — only a Phase I that also defined its own nephrotoxic ceiling. The strongest human evidence is host-directed (blood pressure, inflammatory markers, MASLD liver enzymes, and the senolytic program now entering cancer-survivor trials), while every tumor-directed finding below this point is preclinical.
Animal
Preclinical Signal
The tumor-directed case rests on a useful set of in-vivo models. Oral/intraperitoneal quercetin inhibited an HCC827 non-small-cell lung cancer xenograft while suppressing Src/Fn14/NF-κB signaling (colony formation 22% vs 35%, invasion 28.8% vs 65.2%; Src overexpression reversed the effect),[2] and reduced aberrant crypt foci in a dietary (2%, oral) azoxymethane colon-carcinogenesis model — an oral in-vivo chemoprevention readout, though at a very high dietary exposure and one plausibly driven by high local gut-lumen levels rather than systemic exposure.[3]
Continue reading — full research detail+
Quercetin suppressed A375 melanoma xenograft growth and B16F10 lung metastasis by inhibiting STAT3 and its targets Mcl-1, MMP-2, MMP-9 and VEGF,[9] and reduced hexokinase-2, Ki-67 and tumor volume in an SMMC-7721 hepatocellular xenograft (50 mg/kg).[21] In a gastric xenograft, quercetin plus low-dose irinotecan lowered epithelial-mesenchymal-transition and angiogenic markers (Twist1, ITGβ6, COX-2, VEGF-A, VEGFR2),[5] and gold-nanoparticle-conjugated quercetin — a delivery formulation, not oral aglycone — impeded a DMBA-induced mammary carcinoma with reduced EMT and angiogenesis.[4] Quercetin inhibited in-vivo chorioallantoic-membrane angiogenesis by roughly 85%.[6] In a hematologic model, oral quercetin reduced WEHI-3 leukemia burden while promoting macrophage phagocytosis and NK-cell activity,[54] and quercetin reduced CT-26 colon and MCF-7 breast tumor volumes in mice alongside broad in-vitro apoptosis.[55]
Signal maturity: in-vivo findings span lung, colon (dietary chemoprevention), melanoma (metastasis), liver, gastric, breast and hematologic models, with a well-characterized set of mechanisms. But several used intraperitoneal or high dietary dosing rather than ordinary oral supplementation, and none has been tested in a human oncology trial.
In Vitro
Cell Model Data
This is the deepest and broadest tier, and the one the concentration gap most constrains. Across colorectal, breast, lung, prostate, liver, gastric, ovarian and cervical models, quercetin has been reported to suppress NF-κB, PI3K/AKT/mTOR, MAPK/ERK, JAK/STAT3 and Wnt/β-catenin signaling, arrest the cell cycle, induce endoplasmic-reticulum-stress apoptosis, restrain hexokinase-2-driven glycolysis, suppress tumor NRF2, and trigger both Bax-dependent apoptosis and ferroptosis.[11,13,15,17,18,19,22,23,26,56]
Continue reading — full research detail+
Nearly all of this work used quercetin at 25–100 µM (the NRF2 studies at 155–233 µM), generally as sustained exposure — a range roughly two to three orders of magnitude above the submicromolar free aglycone that standard oral dosing achieves. The gap is not only quantitative: in cell assays, glucuronidation of quercetin removed most of its direct antiproliferative activity, so the form that predominantly circulates is generally much less directly active than the free aglycone these experiments tested.[40] That activity varies by metabolite (some conjugates retain activity, and inflamed tissue can locally deconjugate quercetin back toward the aglycone), so this is a strong exposure caveat rather than proof of total inactivity — but it is the central reason the tumor-directed roles are read cautiously (see Pharmacokinetics and Administration below).
Several of the strongest findings are also combination studies — quercetin with gemcitabine, 5-fluorouracil, cisplatin, aconitine or vitamin C — where its contribution is chemosensitization rather than standalone killing.[11,12,16,20,24]
Signal maturity: cell-model research offers the deepest mechanistic detail and the broadest tumor-type panel for this compound, but it is also where the concentration gap matters most — most reported effects require sustained exposure at concentrations oral dosing has not been shown to reach or sustain in active, unconjugated form.
Advertisement
02 — Pathways
Pathway Interaction Profile
Quercetin engages many pathways relevant to tumor behavior, grouped below by the functional role each one supports. Direct anti-tumor mechanisms come first; further down, a separate set of pathways supports the body's own resilience — and, running through all of them, the caveat that oral exposure rarely reaches the concentrations these mechanisms need.
Quercetin's Contain classification is the broadest of its tumor-directed roles — its mechanisms recur in whole animals, not only in dishes — but it is held at partial by the same constraint that runs through the whole page: the animal readouts used intraperitoneal dosing or high (2% dietary) exposures, and ordinary oral supplementation is not shown to reach the concentrations these effects require. The evidence is real; its translation to oral use is not established.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
NF-κB / TNF-α / IL-6 inflammatory axis
Quercetin (100 µM) suppressed Src-mediated Fn14/NF-κB signaling in non-small-cell lung cancer cells (reduced nuclear p65, phospho-IκBα, phospho-IKKβ), cutting colony formation (22% vs 35%) and invasion (28.8% vs 65.2%); an HCC827 xenograft confirmed tumor-growth inhibition in vivo, and Src overexpression reversed every effect.[2] Separately, oral 2% dietary quercetin reduced azoxymethane-induced aberrant crypt foci in rats — an inflammation-linked, oral, in-vivo chemoprevention readout.[3]
Angiogenesis / VEGF / HIF-1α
Quercetin inhibited angiogenesis by ~84.7% in an in-vivo chorioallantoic-membrane assay, with computational docking to VEGF, HIF-1α and VEGFR2.[6] Gold-nanoparticle-conjugated quercetin — a delivery formulation, not the oral aglycone — reduced VEGFR-2 and suppressed capillary tube formation and new-vessel growth.[4]
Prevent Tumor Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
EMT & metastatic invasion
In a gastric-cancer xenograft, quercetin with low-dose irinotecan lowered Twist1, ITGβ6, COX-2, VEGF-A and VEGFR2 and reduced Tie2⁺ monocytes;[5] gold-nanoparticle quercetin reversed EMT markers (reduced vimentin, N-cadherin, Snail, Slug and Twist; raised E-cadherin) and inhibited migration and invasion in a DMBA rat mammary model.[4]
Prevent Dormant Reactivation
Research concerning wake-up signalling and reactivation of dormant disseminated tumour cells.
Cancer stemness (CD44 / ALDH / Nanog/Sox2)
Quercetin suppressed the CD44⁺/CD24⁻ breast cancer-stem-cell population and its behavior by inhibiting PI3K/Akt/mTOR,[7] and dietary quercetin targeted pancreatic cancer stem cells, inhibiting self-renewal and sphere formation.[8]
Quercetin's Starve classification is partial: one route has a genuine in-vivo anchor, but the other two are double-edged — a pro-oxidant redox effect only at very high concentrations, and an autophagy response that is actually pro-survival and can blunt quercetin's own killing.
Glucose Axis Pressure
Research concerning glycolytic ATP production and glycolytic intermediates used by cancer cells.
Aerobic glycolysis (Warburg effect)
The strongest Starve evidence: quercetin dose-dependently reduced glucose uptake and lactate production and downregulated hexokinase-2 via the Akt-mTOR pathway — hexokinase-2 overexpression rescued growth and glycolysis — and an SMMC-7721 hepatocellular xenograft (50 mg/kg) confirmed reduced tumor volume, hexokinase-2, Ki-67 and phospho-Akt/phospho-mTOR in vivo.[21] A second route suppressed glycolysis through the G3BP1/YWHAZ axis in oral squamous carcinoma.[22]
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
NRF2–GSH redox axis
Quercetin has been reported to behave as a tumor-cell pro-oxidant here, suppressing NRF2 and antioxidant-defense genes to sensitize cells to oxidative stress — but at half-maximal concentrations of 155–233 µM, and often only in combination with vitamin C, a severe concentration gap.[23,24] This is the tumor-directed counterpart to quercetin's host-side redox activity: the same NRF2 axis it suppresses in tumor cells is the one it can support in host tissue (see Protect below).
Metabolic Flexibility Suppression
Research concerning metabolic adaptation and switching between fuel sources under pressure.
Autophagy & lysosomal system — resistance signal
This card documents a counter-signal, not a starving effect. Quercetin induced autophagy in gastric cancer cells via Akt-mTOR and HIF-1α, but it was protective (pro-survival) autophagy — the tumor cells used it to survive quercetin exposure, and inhibiting autophagy actually potentiated quercetin's apoptosis.[25] So this is a resistance mechanism rather than evidence that quercetin suppresses metabolic flexibility; it argues for pairing quercetin with an autophagy inhibitor and does not count as positive Starve evidence.
Quercetin's Weaken classification reflects reported disruption of the growth and survival signaling tumor cells depend on, reached through several converging cascades and a lethal stress response. One pathway (STAT3) carries animal corroboration; the rest are high-concentration cell-culture findings, several of them combination experiments — so the role is held at partial, with the melanoma xenograft as a key animal-corroborated pathway rather than proof that oral dosing reproduces these effects systemically.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
JAK/STAT (STAT3)
The role's in-vivo anchor: quercetin reduced STAT3 phosphorylation and nuclear localization and its targets Mcl-1, MMP-2, MMP-9 and VEGF, suppressing A375 melanoma xenograft growth and B16F10 lung metastasis (constitutively-active STAT3 partly rescued the cells).[9] The source reports the xenograft and metastasis outcomes but does not specify the dosing route, so this anchor is read with the same oral-exposure caveat that governs the whole tumor-directed profile — the unambiguously oral in-vivo corroboration sits under Contain's dietary chemoprevention above.[3] A second route runs through Axl/IL-6/STAT3 in glioblastoma, reported to induce apoptosis while sparing ~85% of normal astrocytes.[10]
PI3K–AKT–mTOR
Quercetin (6.25–50 µM) downregulated RAGE and phospho-PI3K/AKT/mTOR, raising the Bax/Bcl-2 ratio and caspase-3 and enhancing gemcitabine cytotoxicity in parental and gemcitabine-resistant pancreatic cells;[11] in cisplatin-resistant ovarian cells it suppressed the antioxidant system (SOD2, catalase, GPX1, HO-1, Nrf2) alongside PI3K/Akt/mTOR at 100 µM.[12]
RAS–RAF–MEK–ERK (MAPK)
Quercetin induced caspase-dependent apoptosis with Bcl-2 downregulation and inhibition of both PI3K/Akt and ERK in HepG2 hepatoma cells,[13] and — with midkine knockdown — reduced prostate cancer-stem-cell survival and migration through PI3K/AKT and MAPK/ERK.[14]
Wnt / β-catenin
Quercetin inhibited β-catenin/Tcf transcriptional activity in SW480 colon cells and in cells expressing a degradation-resistant mutant β-catenin, placing its action downstream of the APC–Axin–GSK3β destruction complex,[15] and potentiated 5-fluorouracil in colon cells by targeting Wnt/β-catenin via miR-27a.[16]
Cell cycle checkpoints (CDK4/6–RB–E2F, G1/S, G2/M)
Quercetin caused G2/M accumulation and apoptosis in MCF-7 breast cells,[17] and stabilized wild-type p53 to drive G1 arrest and apoptosis in HepG2 hepatoma cells.[18] In vitro.
Attrition Pressure
Research concerning cellular stress vulnerability and net tumour-cell attrition under sustained conditions.
ER stress & unfolded protein response (UPR)
Quercetin downregulated HSP70, and simultaneous suppression of the chaperones GRP78 and HSP70 enhanced its UPR-mediated pro-apoptotic effect in breast cells;[19] quercetin plus aconitine synergistically induced ER-stress apoptosis and downregulated MDR1 in cervical HeLa cells.[20] In vitro.
Quercetin's Attack classification is partial: it has two mechanistically distinct routes to direct cell killing — mitochondrial apoptosis and ferroptosis. Selective toxicity was reported in particular paired cancer/normal-cell experiments, but broad tumor selectivity across tissues and doses is not established; the in-vivo support is limited and non-oral, and the effective concentrations (~100 µM) sit far above achievable oral free-quercetin exposure.
Direct Tumor-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Quercetin (100 µM) reduced Akt and Bad phosphorylation, dissociated Bcl-xL from Bax, and drove Bax mitochondrial translocation, cytochrome-c release and caspase-3/8/9 and PARP cleavage — and was tumor-selective, as Bax-null colon cells resisted it while normal prostate epithelial cells survived the same dose.[26] A parallel mitochondrial/caspase-3 route was confirmed in triple-negative MDA-MB-231 breast cells,[27] and quercetin reduced CT-26 colon and MCF-7 breast tumor volumes in mice (route unspecified) alongside apoptosis across nine cell lines.[55] Selective, with limited in-vivo support, but the effective concentrations sit above achievable oral free-quercetin exposure.
Ferroptosis (execution / cell death)
Quercetin induced ferroptosis in gastric cancer cells by binding SLC1A5 and blocking NRF2 nuclear translocation, lowering the xCT/GPX4 antioxidant axis and raising lipid peroxidation, intracellular iron and p-Camk2/p-DRP1 signaling.[56] In vitro plus molecular docking, with no xenograft — and double-edged, like quercetin's redox behavior generally: a ferroptosis inducer in the tumor-cell context here, whereas the same molecule buffers oxidative stress in host tissue (see Protect below).
Quercetin's Protect classification is the role with the most genuine human evidence, across two sub-scopes — though its cancer-specific relevance is largely inferred rather than demonstrated. One is host-status evidence relevant to oncology — the dasatinib+quercetin (combination) senolytic program, chemoprevention data, and the single human oncology trial — reported below as cited findings without a pathway card. The other is mechanism-based host-resilience evidence (inflammatory, cardiovascular and hepatic) from randomized human trials, which carries real cards further down. None of it has yet been shown to improve cancer treatment tolerance or cancer-specific outcomes.
Oncology Host-Status
Host-aging and therapy-induced senescence (combination-specific) — quercetin is the “Q” in the senolytic combination dasatinib+quercetin (D+Q), the most active thread in its cancer-adjacent pipeline — but the evidence is for the combination, and quercetin's independent contribution is unknown, since dasatinib is itself a potent senolytic. Small open-label pilots reported that D+Q improved physical function in 14 idiopathic-pulmonary-fibrosis patients[33] and reduced adipose senescent-cell burden and circulating senescence-associated secretory factors (IL-1α, IL-6, MMP-9, MMP-12) in 9 diabetic-kidney-disease patients.[34] These were small, non-randomized, non-cancer pilots, and a subsequent Phase 2 randomized trial of D+Q in postmenopausal women did not meet its primary endpoint, with benefit only in an exploratory high-senescent-burden subgroup.[60] Senescent-cell clearance is nonetheless relevant to cancer-survivor frailty and therapy-accelerated aging, which is why D+Q has moved into registered oncology trials (in adult survivors of childhood cancer, and with CAR-T therapy in myeloma) — none of which has yet reported efficacy.
Chemoprevention (mixed) — a green-tea plus quercetin combination roughly doubled prostate-xenograft growth inhibition versus green tea alone (45% vs 21%) while raising non-methylated green-tea polyphenols and tumor apoptosis,[28] but the largest chemoprevention epidemiology, the Women's Health Study, found no cancer-incidence benefit from dietary quercetin or total flavonoids.[29]
The one human oncology trial — the 1996 Phase I intravenous quercetin trial is the sole interventional oncology dataset: it inhibited lymphocyte tyrosine kinase and produced two anecdotal marker responses, and was dose-limited by nephrotoxicity, so it is as much a safety-ceiling finding as an efficacy signal.[1]
Inflammatory Regulation
Human and preclinical research on systemic inflammatory regulation, distinct from immune-cell surveillance and organ-specific injury.
Systemic inflammatory reduction
A meta-analysis of 16 randomized trials found quercetin lowered C-reactive protein and total and LDL cholesterol, though not IL-6 or TNF-α, and with high heterogeneity across the pooled trials;[31] an open-label trial in 60 severe-COVID patients found 1000 mg/day quercetin (with antivirals) shortened discharge time and lowered alkaline phosphatase, C-reactive protein and lactate dehydrogenase — a non-oncology anti-inflammatory data point.[35]
NF-κB modulation is one plausible contributor to these anti-inflammatory effects, and is also quercetin's tumor-directed Contain mechanism (see the NF-κB axis above) — a potential shared mechanism, host-beneficial and tumor-directed in different settings. The human biomarker trials measured C-reactive protein, however; they did not establish that quercetin lowered it specifically through NF-κB, so the mechanistic link is an interpretation rather than a demonstrated causal bridge.
Other Organ-System Reserve
Research concerning renal, cardiac, pulmonary, and other non-hepatic organ reserve under stress.
Cardiovascular reserve (blood pressure)
A meta-analysis of 7 randomized trials (587 patients) found quercetin lowered systolic (−3.04 mmHg) and diastolic (−2.63 mmHg) blood pressure, with a larger effect at ≥500 mg/day.[30] The circulating quercetin conjugates that dominate in vivo lack direct vasorelaxant activity but prevent endothelial dysfunction under oxidative stress, a plausible mechanistic basis and an independent host benefit.[42]
Hepatic Resilience & Clearance
Human and preclinical research on hepatic enzyme systems, bile-acid handling, and liver-related markers.
Liver-enzyme improvement in fatty liver disease
A meta-analysis of 7 randomized trials (540 patients) found quercetin improved ALT, AST, GGT and direct bilirubin in metabolic-dysfunction-associated steatotic liver disease — an independent host-resilience benefit rather than an adverse-effect finding.[32]
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
NF-κB / TNF-α / IL-6 inflammatory axis
Quercetin reduced NF-κB inflammatory signaling and tumor growth in a lung cancer xenograft, and cut colon aberrant crypt foci in a high (2%) dietary model — animal-corroborated, though ordinary oral supplementation is not shown to reach these exposures.
Prevent Tumor Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
Quercetin suppressed the cellular changes tumor cells use to detach and invade, reported in gastric and breast models alongside reduced angiogenic markers.
Prevent Dormant Reactivation
Research concerning wake-up signalling and reactivation of dormant disseminated tumour cells.
Cancer stemness (CD44 / ALDH / Nanog/Sox2)
Quercetin reduced the cancer-stem-cell compartment in breast and pancreatic models by suppressing PI3K/Akt/mTOR — a preclinical, cell-model finding.
Glucose Axis Pressure
Research concerning glycolytic ATP production and glycolytic intermediates used by cancer cells.
Aerobic glycolysis (Warburg effect)
Quercetin reduced glucose uptake and hexokinase-2 to restrain tumor glycolysis, confirmed in a liver cancer xenograft — the strongest Starve evidence, though the other redox and autophagy routes are double-edged.
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
Quercetin suppressed STAT3 signaling and its metastatic targets, confirmed in a melanoma xenograft with reduced lung metastasis — the role's key animal-corroborated pathway, though systemic oral translation is unproven.
Quercetin suppressed PI3K/AKT/mTOR growth signaling across pancreatic and ovarian models, enhancing chemotherapy sensitivity — a cell-model finding.
Direct Tumor-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Quercetin triggered Bax-dependent mitochondrial apoptosis, selective for cancer over normal cells at the same concentration — but in vitro at high concentrations, with only limited in-vivo support.
Inflammatory Regulation
Human and preclinical research on systemic inflammatory regulation, distinct from immune-cell surveillance and organ-specific injury.
Systemic inflammatory reduction
Quercetin lowered C-reactive protein in a meta-analysis of randomized trials — the same NF-κB mechanism that drives its tumor-directed Contain activity, benefiting host and tumor settings differently.
Advertisement
03 — Pharmacokinetics
Pharmacokinetics and Administration
How quercetin moves through the body matters more than the mechanistic evidence alone — because most of what enters the bloodstream after an oral dose isn't free quercetin at all, but conjugated metabolites that lose the antiproliferative activity, and because delivery form changes exposure more than milligram dose does.
Absorption
Standard oral quercetin has low, variable bioavailability — around 2–4% absolute in animal studies — and is rapidly converted so that only conjugated metabolites, not free aglycone, circulate. Food matrix and glycoside form change exposure roughly 5-fold.
The Concentration Gap
Mechanistic studies used 25–100 µM; free quercetin after oral dosing is generally submicromolar and transient. And in cell assays glucuronidation removes most of the direct antiproliferative activity, so the circulating form is generally much less active against tumors (though activity varies by metabolite).
Clinical Dose Context
Human oral trials used hundreds of mg to ~1000 mg/day, tolerated to 2000 mg/day. The only oncology dosing was intravenous, capped by kidney toxicity rather than efficacy.
Formulation Effects
EMIQ raised absolute bioavailability to 35% (vs 2% for aglycone) in rats and ~3× isoquercitrin in humans; quercetin phytosome gave up to ~20× higher plasma quercetin than the unformulated compound.
Metabolism
Rapid phase-II conjugation — glucuronidation, sulfation and methylation to isorhamnetin — dominates. Some conjugates keep host-side (vascular) activity even though the anticancer activity is lost.
Co-Dosing Considerations
Human studies show quercetin raises exposure to P-glycoprotein, CYP2C9 and CYP2E1 substrates. The key oncology flag: quercetin can chemically inactivate bortezomib — avoid without oncologist approval.
Absorption
Standard oral quercetin has low, highly variable bioavailability. It is rapidly and extensively metabolized in the enterocyte and liver, so that only conjugated metabolites — quercetin glucuronides and sulfates, and methylated isorhamnetin — circulate; free aglycone is barely detectable in plasma.[36] The sugar moiety governs exposure: quercetin-4′-glucoside and onion quercetin (100 mg) reached a plasma peak of about 2.1–2.3 µg/mL at ~0.7 h, whereas rutin (a rhamnoglucoside, 200 mg) reached only 0.3–0.6 µg/mL at 4–7 h.[36] Absolute bioavailability is low across species — about 4% in dogs, with more than 80% of circulating flavonols conjugated.[48]
Food matrix matters as much as dose: an onion-skin extract gave 4.8× higher exposure than pure quercetin dihydrate,[37] and a quercetin-enriched cereal bar gave 5× higher exposure than a powder-filled capsule.[38]
The Concentration Gap
This is the fact that should reshape how the entire Evidence Summary is read, and for quercetin it is unusually stark. Mechanistic studies used quercetin at 25–100 µM (the NRF2 work at 155–233 µM), whereas free aglycone after conventional oral dosing is generally reported in the submicromolar range and is transient. One single-dose estimate (from a Hypericum extract) put the free-quercetin peak near 47 ng/mL (~0.16 µM);[39] the exact figure varies with dose, food matrix, glycoside form and assay, but a systemic exposure gap of roughly two to three orders of magnitude remains.
| Benchmark | Concentration | Interpretation |
|---|---|---|
| In vitro effective range across pathway studies | 25–100 µM | The concentration range used throughout the mechanistic literature above (NRF2 work higher still, 155–233 µM) |
| Human free-quercetin plasma peak (one single-dose estimate) | ~0.16 µM | About 47 ng/mL from a Hypericum-extract study — formulation-specific, but broadly representative of the submicromolar, transient free-aglycone range[39] |
| Active form actually circulating | Conjugates | In cell assays glucuronidation removed most direct antiproliferative activity, so circulating conjugates are generally much less active against tumor cells (activity varies by metabolite)[40] |
The gap is not merely quantitative. In cell assays, glucuronidation — the dominant conjugation route — removed most of quercetin's antiproliferative activity in cancer cells; methylation to isorhamnetin reduced but did not eliminate it, and the 4′-sulfate retained meaningful activity.[40] Activity therefore varies by metabolite, and inflamed tissue can locally deconjugate quercetin back toward the aglycone, so this is a strong exposure caveat rather than proof of total inactivity — but the practical upshot is that the form predominantly circulating is generally much less active than the free aglycone the mechanistic literature tested, which is why the tumor-directed roles are read cautiously.
One exception cuts the other way: because quercetin is poorly absorbed, a large fraction stays in the gut lumen, so local colorectal exposure can be high even when plasma exposure is low — which makes a local effect in the colon (as in the dietary aberrant-crypt-foci study) more biologically plausible than systemic treatment of distant tumors.
Clinical Dose Context
Human studies have used single and repeated oral doses in the hundreds-of-mg to ~1000 mg/day range, with blood levels strongly formulation-dependent, and tolerability studies have gone to 2000 mg/day.[36,52] The only oncology dosing is the intravenous route, where the ceiling was set not by efficacy but by renal dose-limiting toxicity.
| Context | Dose | Source |
|---|---|---|
| Phase I oncology (intravenous) | up to 1700 mg/m² | Dose-limiting nephrotoxicity; recommended dose 1400 mg/m²[1] |
| Senolytic combination (with dasatinib) | 1000–1250 mg/day | Short intermittent pulses (e.g. 3 days/week)[33,34] |
| Host-physiology trials (oral) | ~500–1000 mg/day | Blood-pressure, CRP and MASLD endpoints[30,31,32] |
| Tolerability ceiling (oral) | 2000 mg/day | Safely tolerated over one week in COPD patients[52] |
Formulation Effects
Because free quercetin absorbs so poorly, delivery form changes exposure more than milligram dose does — and this is where standard quercetin, EMIQ and phytosome diverge. EMIQ (enzymatically modified isoquercitrin, also called α-glycosyl isoquercitrin or AGIQ) is a solubilized glycoside that intestinal enzymes readily hydrolyze to quercetin; quercetin phytosome (Quercefit) is a food-grade lecithin complex. These forms are designed to improve delivery rather than introduce a new active compound; whether they reproduce the same metabolite and tissue-exposure profile has not been fully established, and higher total plasma exposure is not the same as higher active exposure at a tumor.
| Formulation | Improvement | Study detail | Citation |
|---|---|---|---|
| Standard quercetin | Absolute bioavailability ~2% (rat) | Reference point; only conjugates circulate | [44] |
| EMIQ (AGIQ) | 35% BA (rat); ~3× isoquercitrin (human) | ~17× quercetin on animal true-bioavailability; food additive in Japan, US GRAS | [43,44,45] |
| Quercetin phytosome (Quercefit) | up to ~20× (human) | Lecithin complex; randomized crossover, n=12; no notable side effects | [46] |
| Micellar (chrysin-quercetin-rutin) | >2× AUC (human) | Crossover, n=16; PK endpoint was chrysin, quercetin co-encapsulated | [47] |
The EMIQ figures must be read carefully: the ~17×-versus-quercetin number is animal true-bioavailability (35% vs 2%), whereas the human figure is roughly 3× isoquercitrin exposure.[45] The phytosome human data comes from a small (n=12), proprietary-formulation crossover — suitable for sourcing context, not proof of therapeutic superiority. No qualifying human quercetin-specific pharmacokinetic data was identified for LipoMicel, self-emulsifying, cyclodextrin or nanoparticle forms, so no fold-figure is stated for those. Crucially, none of these formulations has been shown to reach the 25–100 µM mechanistic range, deliver more free aglycone to a tumor, or improve any clinical oncology outcome — and a form that raises systemic exposure could equally raise interaction potential (see Co-Dosing).
Metabolism
Quercetin undergoes rapid phase-II conjugation — glucuronidation, sulfation and methylation — generating many circulating species. In rats fed quercetin glucosides, a mixed conjugate (isorhamnetin-7-O-glucuronide-4′-O-sulfate) was identified as a major plasma metabolite, with mixed glucuronide-sulfates dominating;[41] human plasma likewise contains multiple glucuronidated, sulfated and methylated conjugates, though the dominant profile varies with source, dose and analytical method. These metabolites are not simply inactive byproducts: quercetin glucuronide and sulfate conjugates lack direct vasorelaxant activity but prevent endothelial dysfunction under oxidative stress, so some host-side activity survives conjugation[42] — even as the anticancer antiproliferative activity does not.[40] Low free-aglycone plasma levels therefore mean the exposure is happening in a different chemical form, whose host activity is retained but whose tumor-directed activity is largely not.
Co-Dosing Considerations
Quercetin measurably modulates drug-transporter and drug-metabolizing systems in humans, and its interactions are drug-specific — they can increase, decrease, or otherwise alter a co-administered drug's effect. Most are moderate human pharmacokinetic changes, but one is potentially harmful and oncology-specific: quercetin can chemically inactivate boronic-acid proteasome inhibitors such as bortezomib. Each row is flagged by the most cautious guidance its cited evidence supports; preclinical synergy with a chemotherapy does not establish clinical benefit, and a formulation that raises exposure could raise interaction potential too.
Discuss whether to combine, separate, or avoid quercetin and a medication with your treating oncology team or physician.
| Flag | Interaction |
|---|---|
| Avoid | Bortezomib and other boronic-acid proteasome inhibitors (e.g. ixazomib, carfilzomib) — quercetin chemically reacts with bortezomib's boronic-acid group and blocked bortezomib-induced apoptosis in multiple myeloma and malignant B-cell models in a dose-dependent way (an effect reversed by adding boric acid).[57] The human clinical magnitude is unknown, but concurrent high-dose quercetin cannot be assumed safe with these drugs and should be used only with explicit approval from the treating oncologist. |
| Caution | P-glycoprotein substrates (e.g. fexofenadine, digoxin) — quercetin 500 mg three times daily raised fexofenadine exposure by 55% and peak level by 68% in healthy volunteers by inhibiting P-glycoprotein efflux.[49] For narrow-therapeutic-index P-gp substrates such as digoxin the human data is indirect (quercetin short-term inhibits and long-term induces P-glycoprotein and CYP3A4 in vitro), so this is a precautionary caution rather than an established interaction.[51] |
| Caution | CYP2C9 substrates (e.g. diclofenac, and NSAIDs/oral hypoglycemics cleared this way) — quercetin 500 mg twice daily for 10 days raised diclofenac exposure and reduced its 4-hydroxy metabolite in healthy volunteers, consistent with CYP2C9 inhibition.[58] |
| Caution | CYP2E1 substrates (e.g. chlorzoxazone) — quercetin 500 mg twice daily raised chlorzoxazone exposure by 69% and peak level by 48% in healthy subjects.[50] |
| Caution | Warfarin and other anticoagulants — quercetin and its metabolites strongly displaced warfarin from serum albumin in vitro (while being only weak CYP2C9 inhibitors, so this is an albumin-binding rather than a metabolic interaction).[59] A clinically meaningful interaction is not established, but warfarin's narrow therapeutic index justifies INR monitoring and clinician review with high-dose quercetin. |
| Monitor | Cytotoxic chemotherapy and targeted agents generally — interactions are drug-specific and can raise, lower, or otherwise alter exposure and effect. Preclinical work shows chemosensitization with some agents (gemcitabine, 5-fluorouracil, cisplatin),[11,12,16] but antagonism with others (bortezomib, above), so preclinical synergy does not establish clinical benefit and any concurrent regimen should be assessed against the exact drugs with the oncology team. |
Advertisement
04 — Onset & Washout
Onset and Washout
Quercetin's timeline splits into two clocks: free quercetin appears and clears quickly but at low levels, while the effects that actually have human evidence — blood pressure, inflammation, senescent-cell clearance — accrue over weeks or arrive as intermittent pulses.
Immediate Onset
Quercetin metabolites may peak within about an hour for readily absorbed glucosides, but later for less-absorbed forms such as rutin; absolute free-aglycone levels stay low (tens of ng/mL) and are rapidly conjugated — there is no fast, high pharmacologic peak of the active free molecule.
Steady State
The circulating conjugates have a long terminal half-life (roughly 11–24 hours depending on form), with a characteristic double-peak reflecting enterohepatic recirculation.
Accumulated Effect
Host-physiology endpoints (blood pressure, CRP, liver enzymes) were measured over weeks to months of daily dosing in the trials that showed them.
Dosing Pattern in Studies
Host-physiology trials used steady daily oral dosing; the senolytic combination used intermittent pulses (e.g. 3 days/week); oncology exposure was intravenous. This describes how quercetin was studied, not a recommended regimen.
Washout
How long quercetin's influence takes to clear before it stops being a relevant factor for co-administered medications.
Most circulating metabolites are expected to decline over several days based on the terminal half-life of quercetin conjugates, but no human study has established a washout period for medication interactions or procedures — and transporter and enzyme effects need not clear at the same rate as plasma metabolites, so timing defers to the care team and the specific interacting drug.
Two Distinct Clocks
Quercetin's timeline separates cleanly into a fast pharmacokinetic clock and a slow effect clock. The direct-pharmacology clock (Clock A) governs the free molecule and its transporter/enzyme interactions; the downstream-phenotype clock (Clock B) governs the host-physiology and senolytic effects that actually carry human evidence.
| Clock A — Direct Pharmacology | Clock B — Downstream Phenotype | |
|---|---|---|
| Latency | Fast — free-quercetin plasma peak within ~1 hour | Slow — host-physiology endpoints measured over weeks to months |
| Persistence | Conjugate terminal half-life ~11–24 hours; double-peak from enterohepatic recirculation | Senolytic effect delivered in intermittent pulses; senescent-cell clearance can persist after the drug clears |
| What it covers | Free-quercetin exposure; P-glycoprotein and CYP interaction risk | Blood pressure, inflammatory markers, liver enzymes, senescent-cell clearance |
The direct-pharmacology clock (Clock A) is fast and shallow: free quercetin peaks within an hour but at low levels, and is rapidly conjugated. Clock B is where quercetin's human evidence lives — and it runs on the order of weeks (host physiology) or as deliberate intermittent pulses (the senolytic “hit-and-run” logic, where senescent-cell clearance outlasts the drug).
Steady State and Accumulation
Circulating quercetin is present almost entirely as conjugated metabolites with a long terminal half-life (~11–24 hours depending on the glycoside form and food matrix), and a double-peak plasma profile reflects enterohepatic recirculation.[36,39] Meaningful accumulation of the active free aglycone across a normal once- or twice-daily interval is unlikely given how quickly it is conjugated, but the conjugate pool can carry over between doses; no dedicated repeated-dose steady-state study quantifies this for a specific formulation.
Dosing Pattern in Studies
Human trials used steady daily oral dosing (hundreds of mg to ~1000 mg/day) for host-physiology endpoints, intermittent pulsed dosing (e.g. 3 days/week, 1000–1250 mg/day with dasatinib) for the senolytic combination, and the intravenous route for the one oncology trial.[30,33,34] This describes how quercetin was studied, not a recommended regimen. For any application depending on the preclinical tumor-directed pathways above, see The Concentration Gap under Pharmacokinetics and Administration for how far ordinary oral dosing sits from the concentrations those mechanisms require.
Washout
There is no validated washout period for quercetin. As a pharmacokinetic extrapolation, most circulating compound would likely clear within a few days of stopping, given the terminal half-life of circulating conjugates (reported around 11–24 hours, though values vary with the analyte measured, formulation and dosing pattern, and one repeated-dose study reported a shorter apparent half-life). But no quercetin-specific human washout study exists; transporter and enzyme effects need not clear at the same rate as plasma metabolites; and high-exposure formulations may behave differently. Because quercetin measurably inhibits P-glycoprotein, CYP2C9 and CYP2E1 in humans, any washout decision before a procedure or a new medication defers to the care team and the specific interacting drug, and interaction risk is raised with that team as soon as supplementation begins.
Advertisement
05 — Safety
Safety Profile
Quercetin is well tolerated at ordinary oral supplement doses, with a reassuring liver profile — but it carries one genuine, route-specific toxicity signal at high intravenous exposure, and its drug-interaction activity is the more practically important consideration.
Tolerated to 2000 mg/day in a one-week study — a small placebo-controlled dose-escalation found no drug-related severe adverse events, but this does not establish long-term safety at that dose.
Reassuring liver profile — the NCBI LiverTox monograph assigns quercetin likelihood score E (an unlikely cause of clinically apparent liver injury), with no reported enzyme elevations or injury cases at typical doses.
Route-specific kidney toxicity — intravenous quercetin was dose-limited by nephrotoxicity in the Phase I trial; this applies to high parenteral exposure, not ordinary oral supplement doses.
Drug interactions are the bigger consideration — quercetin modulates P-glycoprotein and CYP enzymes and can chemically inactivate bortezomib; interaction detail is covered under Co-Dosing, not repeated here.
Adverse Effects in Human Trials
Quercetin is well tolerated at ordinary oral supplement doses. A small placebo-controlled dose-escalation in COPD patients found oral quercetin tolerated up to 2000 mg/day over one week, with no drug-related severe adverse events on blood counts or metabolic panels and only mild gastroesophageal reflux at placebo-equivalent rates — but the trial's own authors noted it did not establish long-term safety.[52] Reported effects elsewhere are mild and mostly gastrointestinal — nausea, abdominal discomfort, headache. Long-term human safety data at ≥1000 mg/day remain limited, particularly beyond about 12 weeks; given the parenteral renal-toxicity signal and quercetin's clearance interactions, a separate caution is reasonable in pre-existing chronic kidney disease. No human pregnancy or lactation safety data was identified; this defers to a clinician rather than an assertion of safety.
Route-Specific Renal Toxicity and Liver Profile
The one genuine dose-related toxicity signal is renal and route-specific: in the Phase I trial, intravenous quercetin was dose-limited by nephrotoxicity at 1700 mg/m² — a caution that applies to high parenteral exposure, not to ordinary oral supplement doses.[1] On the liver, the evidence is reassuring: the NCBI LiverTox monograph assigns quercetin a likelihood score of E (an unlikely cause of clinically apparent liver injury), with no reported serum-enzyme elevations or drug-induced-liver-injury cases at typical doses.[53] Consistent with that, quercetin improves liver enzymes in fatty liver disease — a host benefit detailed under Protect, not an adverse effect.
Drug-Interaction Activity (see Co-Dosing)
The most practically important safety-relevant property of quercetin is not intrinsic toxicity but its documented effect on drug-transporter and drug-metabolizing systems: in healthy volunteers it raised the exposure of P-glycoprotein (fexofenadine), CYP2C9 (diclofenac) and CYP2E1 (chlorzoxazone) substrates, it can chemically inactivate the proteasome inhibitor bortezomib (a potential loss of that drug's anticancer effect), and it can displace warfarin from serum albumin. This interaction guidance — including which flags warrant Avoid versus Caution — is set out in full under Pharmacokinetics and Administration's Co-Dosing Considerations above and is not repeated here.
06 — Sourcing
Sourcing Guide
Formulation matters more than usual for quercetin specifically — standard quercetin's poor absorption means the choice between standard, EMIQ and phytosome 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 Guide07 — Literature
References
Last reviewed: July 2026