01 — Evidence
Evidence Summary
Sulforaphane's evidence divides cleanly by what was measured. In people, the endpoints are biomarkers and cancer-prevention chemistry, not tumor shrinkage; the tumor-directed mechanisms are broad but preclinical, and — unusually for a dietary compound — several were reproduced with oral dosing in animals. The three tiers below are read together because each sets the limits of the others.
Human
Clinical Record
Biomarker and prevention; no tumor-response proof
Sulforaphane's human oncology record is biomarker- and prevention-based, not tumor-regression. A single-arm phase II and a randomized prostate-cancer trial both missed their primary PSA endpoints while showing exploratory slowing of PSA kinetics. Controlled trials have since shown lower tissue proliferation — in bronchial and prostate tumor tissue — and confirmed that sulforaphane reaches human prostate tissue, but none demonstrates a change in cancer outcome. Its firmest host-side signal: randomized broccoli-sprout trials raised excretion of carcinogen-detoxification products.
Animal
Preclinical Signal
Broad, multi-cancer, several oral
The tumor-directed case lives here, and it matters that several models used oral or dietary dosing rather than injection. Sulforaphane suppressed tumor growth and preneoplastic change across several types.
- Chemoprevention of intestinal polyposis with dietary dosing in a colorectal model
- Anti-angiogenic and pro-apoptotic effects in prostate and pancreatic tumors
- Epigenetic (HDAC) inhibition reaching tissue after a single oral dose
- Reduced cancer-stem-like fraction in breast and pancreatic models
In Vitro
Cell Model Data
Wide footprint; exposure gap
A wide, convergent mechanistic footprint across prostate, colon, bladder, breast, endometrial, and liver cells — cell-cycle arrest, mitochondrial apoptosis, redox stress, HDAC inhibition, and angiogenic suppression. The governing caveat is exposure: most effects appear at concentrations above what oral human dosing achieves in blood.
- Dual-context redox: antioxidant genes in normal cells, oxidative stress in tumor cells
- G2/M cell-cycle arrest and Bax/Bak-dependent mitochondrial apoptosis
- Wnt and ERK effects that differ in direction between models
- Effective range mostly above achievable plasma concentrations
Human
Clinical Record
There is no controlled human evidence that sulforaphane shrinks an established tumor, and the two trials that tested it most directly missed their primary marks. A single-arm phase II of sulforaphane-rich broccoli-sprout extract (200 µmol/day) in recurrent prostate cancer had only 1 of 20 men reach a ≥50% PSA decline — missing its primary endpoint; on-treatment PSA doubling time lengthened from 6.1 to 9.6 months, but this compared each man to his own pre-treatment kinetics with no control arm, so it cannot attribute the change to sulforaphane.[1] A double-blind randomized trial of stabilized free sulforaphane (60 mg/day) after prostatectomy likewise missed its primary log-PSA-slope endpoint, while secondary PSA measures favored sulforaphane — a smaller mean rise and a longer doubling time, a hypothesis-generating result rather than a positive trial.[2] That is the honest headline for tumor-directed human evidence.
Continue reading — full research detail+
The rest of the human record separates into tissue delivery, tissue pharmacodynamics, and prevention biomarkers — several positive, all surrogate. Delivery is now established: in a randomized four-week trial, men taking a glucoraphanin supplement had significantly more sulforaphane in prostate tissue than controls, in both the peripheral and transition zones (on the order of 0.7 nmol/g), confirming the compound reaches the gland — though whether that concentration is therapeutically sufficient is unknown.[43] On pharmacodynamics, the strongest controlled result is a 12-month randomized trial in former smokers at high lung-cancer risk: sulforaphane (95 µmol/day) did not change bronchial histopathology, caspase-3, or TUNEL, but the bronchial Ki-67 proliferation index fell 20% on sulforaphane while rising 65% on placebo (P=0.014), with a larger difference at high staining intensity and a bioavailability-dependent effect.[44] A 2026 double-blind pre-prostatectomy trial of a broccoli supplement reported lower tumor c-Myc, ACC1, FASN, and Ki-67 in the active arm, though serum and tumor free-fatty-acid readouts were unchanged,[45] and the year-long ESCAPE trial found dose-dependent attenuation of oncogenic transcriptional changes in the prostate of men on active surveillance without establishing reduced clinical progression.[46] The breast-biopsy trial sits in the same tier — within-arm falls in Ki-67 and HDAC3 that were not confirmed by the adjusted placebo comparison.[3] Alongside these are the gastric mucosal-antioxidant,[4] buccal-NQO1,[5] and feasibility-limited pancreatic[6] signals. The chemoprevention detox-excretion trials are detailed under Protect below.
Signal maturity: human evidence is real and, in tissue-proliferation terms, now includes a controlled bronchial Ki-67 reduction and demonstrated prostate-tissue delivery — but every endpoint is a surrogate (proliferation index, tissue biomarker, detox excretion, PSA kinetics), not cancer incidence, progression, or survival. The correct reading: delivery to human tissue is established; therapeutically sufficient exposure and any clinical anticancer outcome are not.
Animal
Preclinical Signal
Sulforaphane's tumor-directed case is carried by animal models, and unlike many dietary compounds several used oral or dietary dosing rather than injection. Dietary sulforaphane chemoprevented intestinal polyposis in ApcMin/+ mice, reducing polyp number and size with lower proliferation, higher apoptosis, and suppressed p-Akt, p-ERK, and p-JNK in the adenomas;[7] oral sulforaphane (5.6 µmol three times weekly) retarded PC-3 prostate xenograft growth by over half with caspase-mediated apoptosis;[8] and 20 mg/kg reduced orthotopic pancreatic cancer-stem-cell tumor growth about 45%.[16]
Continue reading — full research detail+
The mechanisms were dissected, not merely observed. A single oral 10 µmol dose inhibited histone deacetylase in mouse colonic mucosa within six hours, with histone acetylation at the P21 and bax promoters[19] — an epigenetic effect reaching tissue after realistic oral dosing. Dietary broccoli-sprout sulforaphane prevented estrogen-receptor-negative mammary tumors through epigenetic remodeling, with the strongest effect from maternal exposure,[9] and the pancreatic cancer-stem-cell model lowered Zeb-1, Nanog, and Oct-4 while raising E-cadherin via Sonic-hedgehog/GLI blockade.[16]
Signal maturity: animal evidence is the strongest tumor-directed tier and, importantly, includes oral and dietary dosing that reproduces several effects in vivo — the translational anchor the in-vitro tier lacks. The limit is that no animal efficacy has yet been matched by a positive human tumor outcome.
In Vitro
Cell Model Data
Sulforaphane's cell-level footprint is wide and convergent — G2/M arrest, Bax/Bak-dependent mitochondrial apoptosis, NRF2-driven oxidative stress with glutathione depletion, HDAC inhibition, and suppression of PI3K–AKT–mTOR, NF-κB, and HIF-1α/VEGF — across prostate, colon, bladder, breast, endometrial, and hepatocellular lines. The governing caveat is exposure: most mechanistic work uses 10–40 µM, while human plasma sulforaphane and its metabolites sit in the nanomolar-to-roughly-1-µM range.
Continue reading — full research detail+
Two features deserve stating plainly. First, the redox mechanism is genuinely dual-context — the same NRF2 axis that induces antioxidant and detoxification genes in normal cells drives reactive oxygen species and depletes glutathione inside tumor cells, an effect abolished by antioxidants.[17,18] Second, direction is not always clean: sulforaphane's effect on ERK is pro-apoptotic through ERK activation in prostate and glioblastoma cells,[27,28] yet dietary dosing suppressed ERK in colorectal adenomas in vivo,[7] and its in-vitro Wnt inhibition was not reproduced as a β-catenin change in the same colorectal animal model.[25,7] The closest-to-achievable effect is on breast cancer-stem cells at 1–5 µM.[26] A more unsettling wrinkle: in T24 bladder cells a low dose of sulforaphane (2.5 µM) increased proliferation and migration while doses above 10 µM inhibited growth, a biphasic effect tied to Nrf2-mediated glutathione handling.[47] That matters because achievable human exposure sits near the low end of this range — so the concentration gap is not only "too little to work" but, in at least one model, an exposure window where the biological direction can differ from the high-dose anticancer effect.
Signal maturity: in-vitro evidence is broad and convergent but gated by the concentration gap — most effects require concentrations above achievable plasma sulforaphane. Read every mechanism in this tier through the Pharmacokinetics and Administration section below, and note where a model's direction of effect was not reproduced in vivo.
Advertisement
02 — Pathways
Pathway Interaction Profile
Sulforaphane engages a wide range of tumor-relevant pathways, but each tumor-directed role below is classified partial because the concentrations that drive its effects in the lab were not shown to be achieved at a human tumor. Sulforaphane does reach human prostate tissue, and several effects are corroborated in oral-dose animal models — but whether the levels attained are therapeutically sufficient is unknown, and in at least one model the low end of the achievable range acts in the opposite direction (see Evidence above). Protect is the exception, and the only role with human trial evidence.
Sulforaphane's Contain classification rests on in-vivo-corroborated anti-angiogenic and EMT evidence across more than one tumor type — a suppressed VEGF/HIF-1α program and restored E-cadherin confirmed in living tumors. It is read as partial because those exposures were reached in animal models and cell systems, not shown at the systemic human tumor.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Angiogenesis / VEGF / HIF-1α
Sulforaphane has been reported to suppress the hypoxia-driven HIF-1α/VEGF angiogenic program, its most reproducible Contain node: it inhibited HUVEC tube formation and hepatocellular-carcinoma-stimulated angiogenesis through STAT3/HIF-1α/VEGF, with reduced intratumoral HIF-1α and VEGF in an in-vivo chick chorioallantoic-membrane assay,[12] blocked hypoxia-induced HIF-1α and VEGF concentration-dependently in HCT116 colon and AGS gastric cells,[13] and inhibited HIF-1α synthesis via JNK/ERK — not AKT — in DU145 prostate and tongue squamous cells.[14]
NF-κB / TNF-α / IL-6 inflammatory axis
In PC-3 and LNCaP prostate cells sulforaphane inhibited nuclear translocation of p65-NF-κB and downregulated the inhibitor-of-apoptosis proteins cIAP1, cIAP2, and XIAP, activating Apaf-1/Bax-dependent apoptosis;[10] the broader TNF-α/IL-6 arm of this axis was demonstrated only in a non-cancer in-vivo model, so the oncology claim is limited to the NF-κB and IAP nodes actually shown in tumor cells.[11]
Prevent Tumor Cell Shedding
Research concerning invasion and escape from existing lesions (EMT and ECM breach).
EMT & metastatic invasion
Sulforaphane has been reported to restore E-cadherin and lower the EMT transcription factors Snail and ZEB1 — cutting attachment, invasion, and migration in T24 bladder cells through COX-2/MMP-2,9 and the miR-200c/ZEB1 axis,[15] and reducing Zeb-1 while raising E-cadherin in an orthotopic pancreatic cancer-stem-cell model where 20 mg/kg cut tumor growth about 45% via Sonic-hedgehog/GLI blockade, with Nanog and Oct-4 also lowered.[16]
Sulforaphane's Starve classification is narrow and preclinical: a single redox mechanism in which growth-suppressive concentrations have been reported to raise oxidative stress inside tumor cells and deplete their glutathione buffer. The same axis is protective on the host side — a genuine selective pattern surfaced under Protect below — but the tumor-directed half is in-vitro.
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
NRF2–GSH redox axis
At growth-suppressive concentrations sulforaphane has been reported to generate reactive oxygen species, disrupt the mitochondrial membrane potential, and rapidly deplete glutathione in PC-3 and DU145 prostate cells, triggering an apoptosis abolished by N-acetylcysteine and catalase — establishing ROS as the initiating event.[17] The same NRF2 axis runs the other way in normal tissue, inducing NQO1 and MRP1 detoxification in untransformed colon cells while regulating the antioxidant-response element differently in cancer cells,[18] cross-referenced under Protect's Host-Selective Redox Buffering below. This dual pattern is real in cell systems but is not a settled tumor-selective rule: NRF2 activation can also protect malignant cells and may be undesirable in established, NRF2-high tumors,[43] and the low end of the achievable exposure range promoted rather than suppressed growth in one bladder model.[47] Whether the redox effect is reliably tumor-selective in people is not established.
Sulforaphane's Weaken classification is its broadest tumor-directed role. Its best-supported node is HDAC/epigenetic inhibition — demonstrated in vivo and, unusually, in human blood cells — alongside PI3K–AKT–mTOR suppression and a consistent G2/M arrest; the Wnt and MAPK effects are real but genuinely mixed between models and read more cautiously. It is partial for the same reason as the others: broad mechanism, systemic human exposure not shown to reach it.
Metabolic Weakening
Research concerning tumour metabolic competence and adaptive capacity over time.
Epigenetic regulation & transcriptional control
Sulforaphane is a histone-deacetylase inhibitor, and this is its strongest-supported Weaken node: a single oral 10 µmol dose inhibited HDAC in mouse colonic mucosa within six hours with increased histone H3/H4 acetylation and acetylation at the P21 and bax promoters, and elevated acetyl-histones were also detected in human peripheral blood mononuclear cells;[19] in BPH-1, LnCaP, and PC-3 prostate cells 15 µM sulforaphane reduced HDAC activity 30–40% and raised p21 and Bax protein with cell-cycle arrest and apoptosis[20] — the one Weaken mechanism here with a human pharmacodynamic readout, and the basis for sulforaphane's characterization as a dietary HDAC inhibitor that pairs epigenetic derepression with phase-II enzyme induction.[42]
Expansion Suppression
Research concerning proliferation, cell-cycle progression, and the capacity of lesions to add durable mass.
PI3K–AKT–mTOR (signaling)
Sulforaphane inhibited AKT and mTOR in endometrial Ishikawa/HEC-1 cells with G2/M arrest, intrinsic apoptosis, and EMT reversal, and Ishikawa xenograft growth inhibition exceeded the paclitaxel comparator arm without toxicity;[21] dietary sulforaphane suppressed phospho-Akt in ApcMin/+ intestinal adenomas in vivo,[7] and sulforaphane combined with acetazolamide downregulated PI3K/Akt/mTOR with p21 induction in bronchial-carcinoid cells and xenografts — a combination, not sulforaphane alone.[22]
Cell cycle checkpoints (CDK4/6–RB–E2F, G1/S, G2/M)
Sulforaphane has been reported to produce a consistent G2/M arrest: in HT29 colon cells through cdc2 kinase with cyclin B1 and p21 induction — roscovitine abolished both the arrest and the apoptosis[24] — and in osteosarcoma LM8/MG63 cells, where 20 µM caused complete growth inhibition with p53-independent p21 induction and caspase-3 cleavage, while intraperitoneal sulforaphane reduced LM8 xenografts to under 30% of controls.[23]
Wnt / β-catenin
Sulforaphane blocked Wnt/β-catenin signaling downstream of β-catenin degradation in SW480, DLD1, and HCT116 colorectal cells, inhibiting β-catenin reporters and repressing target genes (LEF1/TCF4 overexpression rescued growth),[25] and at 1–5 µM suppressed the Wnt/β-catenin self-renewal program in breast cancer-stem cells, with 50 mg/kg reducing the ALDH-positive fraction over 50% in xenografts[26] — but dietary sulforaphane left β-catenin and cyclin-D1 unchanged in ApcMin/+ adenomas in vivo, so the in-vitro inhibition is not corroborated by the in-vivo Wnt biomarker and is read cautiously.[7]
RAS–RAF–MEK–ERK (MAPK)
Sulforaphane's ERK effect is direction-dependent rather than uniform: in PC-3 prostate cells it activated ERK1/2 and JNK1/2 to drive apoptosis — dominant-negative ERK2/JNK1 blocked cell death and reversed Bcl-2 downregulation[27] — and its cysteine conjugate (sulforaphane-cysteine) drove glioblastoma apoptosis through sustained ERK1/2 activation,[28] yet dietary sulforaphane suppressed phospho-ERK in ApcMin/+ colorectal adenomas in vivo,[7] a mixed axis where the direction of the ERK effect depends on the model.
Sulforaphane's Attack classification covers direct tumor-cell death by the intrinsic mitochondrial route, corroborated in cell and mouse-embryonic-fibroblast genetic models. The apoptotic mechanism is well-worked; the human-relevant caveat is again systemic exposure, so the role is read as partial.
Direct Tumor-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Sulforaphane's apoptosis is Bax/Bak-dependent — it increased Bax and Bak, drove Bax conformational change and mitochondrial translocation with cytochrome-c and Smac/DIABLO release, activated caspase-9 then caspase-3, raised Apaf-1 and lowered XIAP, and single- or double-Bax/Bak-knockout cells resisted;[29] in UM-UC-3 bladder cells the mitochondria were the primary isothiocyanate target, with membrane-potential collapse and cytochrome-c release preferentially activating caspase-9.[30]
Sulforaphane's Protect classification has the strongest human evidence of any role here — but the human endpoints are chemoprevention and detoxification biomarkers (carcinogen-conjugate excretion, phase-II enzyme induction, tissue proliferation markers), not demonstrated reductions in cancer incidence, treatment toxicity, symptoms, or survival. Oncology Host-Status covers that biomarker evidence, detailed below rather than carrying pathway cards by design. Disease-Resilience covers the mechanism-based host-selective redox finding — a dual-context pattern in which the same NRF2 chemistry induces host detoxification while, at higher concentrations, stressing tumor cells; it is mechanistically striking, but its tumor-selectivity is not established in people.
Oncology Host-Status
Chemoprevention — carcinogen detoxification (lead signal) — in a randomized trial of 291 adults in Qidong, China, a broccoli-sprout beverage (600 µmol glucoraphanin plus 40 µmol sulforaphane daily for 12 weeks) significantly increased urinary excretion of the detoxification conjugates of the airborne carcinogens benzene (61%) and acrolein (23%), with the benzene effect larger in GSTT1-positive individuals;[39] a 50-person crossover confirmed 20–50% increases in these detox conjugates.[40] These measure increased urinary excretion of conjugated carcinogen metabolites, consistent with enhanced detoxification and elimination of those exposures — not a reduction in cancer incidence.
Phase-II enzyme induction — a pilot induced the carcinogen-detoxification gene NQO1 in buccal mucosa,[5] and the breast-biopsy trial lowered tissue Ki-67 and HDAC3 within the treated group[3] — pharmacodynamic biomarker evidence rather than an outcome.
Host-Selective Redox Buffering
Studies evaluating whether normal host tissues can be protected from oxidative stress selectively.
NRF2-driven host protection — the same axis that taxes the tumor
Sulforaphane has been reported to activate NRF2 — through a moderate rise in reactive oxygen species that also mobilizes the lysosomal regulator TFEB — to induce HO-1, lysosomal, and phase-II detoxification genes that protect normal cells from oxidative stress,[41,18] while at higher concentrations the same NRF2–glutathione chemistry can run the other way inside tumor cells, raising ROS and depleting glutathione (the mechanism carried under Starve above).[17] The host-detoxification arm is the better-supported half — it is matched in humans by the chemoprevention detox-excretion trials under Oncology Host-Status. The tumor-stressing arm is preclinical and concentration-dependent, and NRF2 activation is genuinely double-edged in cancer: it can also support tumor-cell survival and is described as undesirable in established, NRF2-high tumors.[43] This is a biologically striking one-mechanism, two-context pattern, not an established tumor-selective benefit in people.
Block Seeding & Niche Formation
Research concerning formation of supportive pre-metastatic niches at distant sites.
Sulforaphane has been reported to suppress the hypoxia-driven HIF-1α/VEGF angiogenic program, confirmed in a hepatocellular in-vivo model and across colon, gastric, and prostate cells — its most reproducible Contain-role finding, though at exposures oral dosing was not shown to reach at a systemic tumor.
Redox Buffering Taxation (Controlled)
Research concerning tumour-cell redox buffering and vulnerability to oxidative pressure, separate from host redox protection.
Sulforaphane has been reported to selectively raise oxidative stress and deplete glutathione inside tumor cells, taxing their antioxidant buffer, while the same axis protects host tissue on the other side. A genuinely selective mechanism, but the tumor-directed half is preclinical.
Metabolic Weakening
Research concerning tumour metabolic competence and adaptive capacity over time.
Epigenetic regulation & transcriptional control
Sulforaphane is a histone-deacetylase inhibitor — its best-supported Weaken mechanism, shown after a single oral dose in animal tissue and, unusually, in human blood cells — derepressing p21 and Bax to arrest the cell cycle.
Direct Tumor-Directed Killing
Research concerning regulated tumour-cell death (apoptosis, ferroptosis, necroptosis).
Intrinsic apoptosis (mitochondrial / Bcl-2)
Sulforaphane has been reported to drive Bax/Bak-dependent mitochondrial apoptosis — cytochrome-c release and caspase-9/-3 activation — confirmed as the primary isothiocyanate mechanism in prostate and bladder cell models.
Host-Selective Redox Buffering
Studies evaluating whether normal host tissues can be protected from oxidative stress selectively.
One redox mechanism, two sides
Sulforaphane has been reported to protect normal cells from oxidative stress by inducing detoxification genes through NRF2, while at higher concentrations the same chemistry can stress tumor cells — a striking dual-context pattern, with the host-detox arm matched by human chemoprevention trials. NRF2 activation is double-edged in cancer, so tumor-selectivity is not established in people.
Advertisement
03 — Pharmacokinetics
Pharmacokinetics and Administration
Sulforaphane's central pharmacokinetic fact is a conversion problem, not a solubility one: the active isothiocyanate has to be released from its glucoraphanin precursor by the enzyme myrosinase, and how much reaches the blood depends on whether that enzyme is active in the food or supplement. What does circulate is cleared the same day, and it sits below the concentrations most cell mechanisms require.
Absorption
Sulforaphane appears in plasma within hours, but only if myrosinase is available to release it from glucoraphanin: ingested isothiocyanate is about six-fold more bioavailable than its glucosinolate precursor, and thorough chewing raises delivery further.
The Conversion Gap
Bioavailability tracks myrosinase, not the label dose. Fresh myrosinase-active sprouts gave roughly 10× more sulforaphane than commercially frozen broccoli, and air-dried sprouts 74% urinary recovery versus 19% from a glucoraphanin powder lacking the enzyme.
Concentration Gap
Systemic sulforaphane in humans peaks at about 1 µM and is transient, while most cell mechanisms use 10–40 µM. Sulforaphane does reach prostate tissue (~0.7 nmol/g), but whether that level is therapeutically sufficient is unknown.
Clinical Dose Context
Human doses are heterogeneous — from tens of µmol of preformed sulforaphane to several hundred µmol of glucoraphanin or free sulforaphane daily. Nominal micromole doses are not interchangeable, because conversion and exposure vary by formulation.
Metabolism
Sulforaphane is cleared by the mercapturic-acid pathway — glutathione then cysteine and N-acetylcysteine conjugates — the same electrophile chemistry that activates NRF2. It interconverts in vivo with its analogue erucin; no consistent CYP interaction is established.
Co-Dosing Considerations
No human drug-interaction study for sulforaphane exists. The consideration that matters most in oncology is active cancer treatment — sulforaphane's NRF2 effect can cut both ways and there are no controlled combination data; discuss timing with the oncology team.
Absorption
Sulforaphane's absorption is governed by conversion, not solubility. The molecule does not exist pre-formed in intact broccoli — it is released from the storage glucosinolate glucoraphanin by the enzyme myrosinase, supplied either by the plant (in raw or minimally processed sprouts) or by gut microbiota. Ingested isothiocyanate is roughly six-fold more bioavailable than its glucosinolate precursor — urinary dithiocarbamate recovery of 88.9 versus 13.1 µmol after equal doses — and thorough chewing raised recovery further, to 42.4 versus 28.8 µmol.[34] What circulates appears within hours and is cleared the same day.
The Concentration Gap
This is the figure that governs how the entire Pathway Interaction Profile above should be read. Most of sulforaphane's mechanistic cell work uses 10–40 µM of free compound, whereas systemic sulforaphane in humans peaks at roughly 1 µM after a glucoraphanin-rich dose and is largely gone within a day — a gap of one to two orders of magnitude, widened further because much of what circulates is present as mercapturic-acid conjugates (glutathione, cysteine, and N-acetylcysteine forms) whose biological equivalence to free sulforaphane is not established.[43,31] The gap is partly offset in specific cases: the breast cancer-stem-cell effect occurred at 1–5 µM,[26] and several oral or dietary animal models reproduced effects in vivo.[7,8,9,16] It is also no longer accurate to say sulforaphane never reaches human tumor tissue — a randomized trial detected it in prostate tissue after four weeks of glucoraphanin, on the order of 0.7 nmol/g.[43] What remains unestablished is whether that tissue level is therapeutically sufficient: twice-daily oral dosing showed no dose-response for the molecular targets HO-1, HDAC activity, or p21 in humans.[35]
| Benchmark | Concentration | Interpretation |
|---|---|---|
| Typical in-vitro mechanistic range | 10–40 µM | free sulforaphane; prostate, colon, and osteosarcoma studies[17,20,23] |
| Breast cancer-stem-cell effect (lowest) | 1–5 µM | the mechanism closest to achievable human exposure[26] |
| Peak systemic sulforaphane in humans | ~1 µM | transient after glucoraphanin-rich intake; largely undetectable by 24 h[43] |
| Human prostate tissue (4-wk glucoraphanin) | ~0.7 nmol/g | delivery demonstrated; therapeutic sufficiency unknown[43] |
Clinical Dose Context
The doses studied are highly heterogeneous — from tens of micromoles of preformed sulforaphane to several hundred micromoles of glucoraphanin or free sulforaphane per day — set by study purpose rather than a defined anti-tumor target. Nominal micromole doses are not interchangeable, because glucoraphanin must still be converted to sulforaphane and that conversion varies with myrosinase, processing, and gut microbiota. Split dosing matters: twice-daily administration sustained higher later-timepoint plasma metabolite levels than once-daily.[35]
| Context | Dose | Source |
|---|---|---|
| Recurrent-prostate phase II (SFN-rich extract) | 200 µmol/day SFN | PSA-doubling-time signal, primary endpoint missed[1] |
| Post-prostatectomy RCT (stabilized free SFN) | 60 mg/day | the most predictable form — no conversion step[2] |
| Absorption / dosing-schedule study | 200 µmol/day (single vs split) | sprouts ~3× higher plasma metabolites than extract[35] |
| Advanced-pancreatic pilot (POUDER) | ~508 µmol SFN + 411 µmol glucoraphanin/day | 15 capsules; high pill burden, 72% dropout[6] |
| Qidong chemoprevention beverage | 40 µmol SFN + 600 µmol glucoraphanin/day | carcinogen-detox-excretion trial[39] |
Formulation Effects
Formulation is the dominant efficacy lever, and it is about myrosinase, not solubility. Myrosinase-active whole sprouts or seeds delivered three-to-four-fold more sulforaphane than an equimolar glucoraphanin supplement lacking active myrosinase,[33] and commercial freezing-blanching or a glucoraphanin-only powder cut delivered sulforaphane several-fold.[36,37] Preformed stabilized sulforaphane removes the conversion step entirely and gives the most predictable exposure — the form used in the post-prostatectomy trial.[2] Where a product provides glucoraphanin without active myrosinase, co-ingesting a myrosinase source such as fresh sprouts rescues conversion.[37]
Metabolism
Sulforaphane is cleared through the mercapturic-acid pathway — glutathione conjugation followed by cysteinylglycine, cysteine, and N-acetylcysteine metabolites — which is mechanistically the same electrophile chemistry that underlies its NRF2 activation.[34,31] Sulforaphane and its reduced analogue erucin interconvert in vivo.[32] A GSTP1 polymorphism did not affect its metabolism or excretion,[32] and a GSTT1 effect appeared only for the benzene detoxification conjugate;[39] no consistent CYP-mediated interaction profile is established.
Co-Dosing Considerations
No controlled human drug-interaction study for sulforaphane was identified, so the flags below are mechanistic prudence rather than documented clinical interactions. Each row is flagged by the most cautious guidance its cited evidence supports.
Discuss whether to combine, separate, or avoid sulforaphane and a medication with your treating oncology team or physician.
| Flag | Interaction |
|---|---|
| Caution | Active anticancer chemotherapy or radiotherapy — sulforaphane potentiated doxorubicin and cisplatin in some preclinical models, while NRF2 induction can enhance stress tolerance and drug resistance in others; the direction depends on the cancer, drug, concentration, and schedule, and there are no controlled human combination data. This is the interaction to raise with the oncology team, not to self-manage.[49] |
| Caution | Other strong electrophilic, NRF2-activating, or redox-active agents — sulforaphane engages the NRF2 and glutathione systems, so stacking several such agents shares that engagement. This is a mechanistic consideration, not a clinically-documented interaction.[17,18] |
| Monitor | Drugs cleared by phase-II detoxification enzymes — sustained induction of these enzymes could in principle alter how a co-administered drug is handled, though no consistent human CYP or phase-II interaction has been demonstrated for sulforaphane; do not infer safety from the absence of reported interactions.[38] |
Advertisement
04 — Onset & Washout
Onset and Washout
Sulforaphane produces two effects on two different clocks: a brief pharmacological exposure that clears within the day, and the slow chemoprevention and biomarker benefits that only emerged over sustained dosing. The distinction matters for reading the trials and for timing discussions with a clinician.
Immediate Onset
Oral sulforaphane appears in the blood within hours and is cleared the same day through mercapturic-acid conjugation and urinary excretion, so direct pharmacological exposure is genuinely brief.
Steady State
With rapid same-day clearance a large plasma reservoir would not be expected to build across doses; split twice-daily dosing kept later-timepoint levels higher than once-daily.
Accumulated Effect
The human outcomes were measured over sustained dosing — up to a year in the tissue trials — though some effects (detox excretion) appear within days. What longer trials establish is persistence during continued use, not a delayed onset.
Dosing Pattern in Studies
The trials used steady daily dosing, once or split. This describes how sulforaphane was studied, not a recommended regimen.
Washout
How long sulforaphane's influence can take to clear before it stops being a relevant factor.
No compound-specific tissue-retention washout data was identified. Given rapid same-day plasma clearance, plasma washout is rapid — but any interaction consideration should be raised with the care team as soon as sulforaphane use begins, not held until a window closes, and any washout decision before a procedure or new medication defers to the treating team.
Two Distinct Clocks
Reading sulforaphane's onset as a single number invites the wrong question. The direct-pharmacology clock (Clock A) is fast and brief: oral sulforaphane appears in plasma within hours and is cleared the same day through mercapturic-acid conjugation, circulating largely as glutathione, cysteine, and N-acetylcysteine metabolites rather than free compound.[31,34] The downstream-phenotype clock (Clock B) is where the human evidence lives, and it is slow — the carcinogen-detoxification, gastric, and biomarker outcomes emerged over sustained dosing, and detox-conjugate excretion was maintained across 12 weeks of continued intake.[39,3] The two clocks do not confirm each other: fast plasma clearance does not mean a correspondingly fast — or slow — biological effect. Some pharmacodynamic effects (detox-conjugate excretion) appear within days and are sustained during dosing; others (tissue biomarker change) were measured over months. Different studies measured different outcomes over different intervals, so "onset" here is study-specific rather than a single delayed clock.
| Clock A — Direct Pharmacology | Clock B — Downstream Phenotype | |
|---|---|---|
| Latency | Fast — appears within hours | Days to weeks (chemoprevention and biomarker trials) |
| Persistence | Short — cleared the same day | Sustained across a full dosing course |
| What it covers | Brief direct plasma exposure | Chemoprevention and biomarker outcomes in Evidence Summary and the Protect findings above |
Steady State and Accumulation
Substantial accumulation of plasma sulforaphane would not be expected from same-day clearance, so each dose behaves closer to a single exposure than a building reservoir — though this is a pharmacokinetic inference, and formulation-specific steady-state data remain limited. Twice-daily dosing sustained higher later-timepoint metabolite levels than once-daily.[35] Consistency of a myrosinase-active source and schedule, not any single dose, is what determines whether useful exposure is achieved at all.
Dosing Pattern in Studies
The trials reporting benefit used steady daily dosing — once or split — over weeks, including the 12-week carcinogen-detoxification trial. This describes how sulforaphane was studied, not a recommended regimen.[39,35]
Washout
No compound-specific washout window was identified for sulforaphane. Given rapid same-day plasma clearance, plasma washout is rapid — but sulforaphane has been detected in prostate tissue after weeks of dosing,[43] and post-discontinuation tissue clearance, along with how quickly NRF2 transcription and enzyme induction resolve, has not been characterized; plasma disappearance does not define tissue pharmacodynamics. The practical guidance is close to the reverse of a fixed window: because any interaction consideration is present while the compound is being taken, it should be raised with the care team as soon as sulforaphane use begins rather than managed by a washout before a new medication, and any decision before a procedure defers to the treating team.
Advertisement
05 — Safety
Safety Profile
Sulforaphane and broccoli-sprout extracts are well tolerated in human trials, with a benign profile dominated by mild gastrointestinal effects. The thyroid concern sometimes raised for concentrated cruciferous intake has been examined directly and not borne out, and the longest controlled trial ran a full year without a severe treatment-related event.
Gastrointestinal disturbance — bloating, nausea, and digestive discomfort are the most common effects, generally mild and dose-related, and most pronounced at high capsule doses.
No thyroid signal on direct testing — a 12-week randomized trial found no effect on thyroid hormones or autoimmunity, and a 7-day phase I found no thyroid or liver laboratory abnormality; the goitrogen concern was not borne out at studied doses.
Pill-burden tolerability — at therapeutic intent the practical limit is intake: a 15-capsule daily regimen in advanced cancer was difficult to sustain, with a high dropout rate.
Long-term data limited; no pregnancy data — the longest controlled trial ran 12 months with no severe treatment-related event, but multi-year and pregnancy safety data are lacking, and safety is not interchangeable across formulations.
Adverse Effects in Human Trials
The human safety record is reassuring and dominated by mild events, though it should be read by formulation and duration rather than pooled. In a small placebo-controlled phase I in healthy volunteers (21 doses over 7 days), a 32-test hematology/chemistry panel plus transaminases and thyroid function (TSH, T3, free T4) showed no significant or consistent abnormality[38] — reassuring, but a short study. The thyroid concern specifically raised for cruciferous compounds has since been tested directly: a 12-week randomized trial of a sulforaphane/glucoraphanin beverage in 45 women found no effect on thyroid hormones or autoimmune status.[48] The longest controlled exposure — a 12-month randomized lung-chemoprevention trial in former smokers — reported no severe treatment-related adverse event.[44] The most common adverse effects are gastrointestinal: at the high capsule burden used in advanced pancreatic cancer (15 capsules per day, ~508 µmol sulforaphane), digestive problems, nausea, and emesis increased and intake was difficult for some patients, though self-care capacity was not severely affected — a tolerability, not a toxicity, ceiling.[6] Two gaps stay honestly open: multi-year and pregnancy safety data are lacking, and "no serious signal" across these mostly small studies does not establish equivalent safety across every form — whole sprouts, glucoraphanin powder, myrosinase-treated extract, free stabilized sulforaphane, and very-high-dose regimens are not one interchangeable exposure. Sulforaphane's drug-interaction picture is uncharacterized — no controlled human interaction study exists — and is set out under Co-Dosing Considerations in Pharmacokinetics and Administration above rather than repeated here.
06 — Sourcing
Sourcing Guide
With sulforaphane, the single biggest factor in whether a product delivers anything close to the exposures the research above describes is myrosinase — the enzyme that converts glucoraphanin to active sulforaphane. Myrosinase-active or preformed-stabilized forms have the strongest case; a glucoraphanin powder without active myrosinase can deliver a fraction of the label dose. Our Sourcing Guide offers a curated list of products available on the retail market, alongside brand quality and accessibility.
Sulforaphane Sourcing Guide07 — Literature
References
Last reviewed: August 2026