Sulforaphane Bioavailability: Why Most Supplements Never Become Sulforaphane, and How to Check Yours

Bioavailability, for sulforaphane specifically, is not a general absorption score. It is one measurement: the share of an administered dose recovered as sulforaphane and its mercapturic-acid metabolites in urine over the 24 hours after the dose. That endpoint is the field's working convention, applied consistently across independent research groups^1,2,4,6.

Sulforaphane arrives fast and leaves fast. In four healthy volunteers given 200 µmol of broccoli sprout isothiocyanates, peak plasma concentrations of 0.943–2.27 µmol/L appeared roughly one hour after the dose, and elimination followed first-order kinetics with a half-life of 1.77 ± 0.13 hours⁶.

In one line: sulforaphane bioavailability is the percentage of a dose that reaches circulation as sulforaphane and is recovered as measurable metabolites within 24 hours, not the amount of glucoraphanin the capsule contained^1,6.

Read next: what sulforaphane is

Glucoraphanin is the storage form. It is a stable glucosinolate found in broccoli, broccoli sprouts, and nearly every broccoli-derived supplement on the shelf, and on its own it does very little, because intact glucosinolates are far less bioavailable than the isothiocyanates they become after hydrolysis¹⁰. Myrosinase is the enzyme that performs that hydrolysis, cutting glucoraphanin into sulforaphane. Think of the two as a match and the strip on the side of the box: kept apart, nothing happens; brought together, the reaction is immediate.

In an intact plant the two sit in separate cell compartments; chewing, chopping or crushing is what brings them together, and a supplement has to do that job for you.

When active myrosinase is present, human bioavailability of sulforaphane reached about 40–41.8% of a 100 µmol dose; when it was absent, the same measurement fell to about 9.4–11.5%¹. A second randomised crossover, in 50 adults, drew the line even more sharply: about 70% of the dose recovered from a pre-hydrolysed sulforaphane beverage versus about 5% from a glucoraphanin-rich one².

A glucoraphanin-only product is not a dead end — your gut bacteria do some of the conversion, just inconsistently, across a roughly thirty-fold range between people⁵.

Supplying the enzyme alongside the precursor changes the outcome: adding 1 g of powdered brown mustard to cooked broccoli increased sulforaphane bioavailability up to roughly eight-fold in a randomised crossover of 12 adults³, and a four-subject pilot using air-dried broccoli sprout powder as the donor points the same way¹¹.

A February 2026 randomised crossover in 16 adults, published in Scientific Reports, found sulforaphane bioavailability of 39.8% when a glucoraphanin-rich broccoli seed extract was paired with mustard-seed myrosinase, against 18.6% for the same extract alone¹². The person-to-person range was wide in both arms, 5.3–76.9% and 2.2–38.1%¹². Three qualifiers travel with those figures: the design was single-dose, so it says nothing about daily use over weeks; the sample was 16 people; and the study was primarily funded by the company that makes the glucoraphanin ingredient it tested, with one author employed there and one a paid consultant¹².

Sulforaphane supplements come in three structurally different forms, and the difference decides which step has to happen inside you and which one already happened in the factory.

Glucoraphanin only. Glucoraphanin-only supplements supply the precursor with no enzyme alongside it. Conversion then depends entirely on your gut bacteria, which vary enormously between people⁵, and this is the form that produced the low end of the bioavailability range in human testing^1,2.

Glucoraphanin plus a myrosinase source. Glucoraphanin paired with a myrosinase source supplies the precursor and the active enzyme together, usually as broccoli sprout material or mustard seed. This pairing has the strongest supporting human data of the three: mustard powder raised bioavailability up to roughly eight-fold in a randomised crossover³, and preparations containing active myrosinase were three- to four-fold more bioavailable than those without¹.

Stabilised, pre-converted sulforaphane. Stabilised sulforaphane is sulforaphane itself, preserved in the product rather than generated in your gut. What does not exist is a head-to-head human trial of a stabilised sulforaphane product against a glucoraphanin-plus-enzyme product; the nearest study puts a stabilised sulforaphane through a placebo-controlled pharmacokinetic trial with no glucoraphanin arm to compare it against¹³. Anyone ranking those two against each other is reasoning, not citing.

Synthetic sulforaphane is a fourth category, used in research. No human trial has compared it with plant-derived sulforaphane in the same subjects, or in a matched cohort under one protocol, so this page makes no claim about the two¹⁴.

Capsules, powders, and liquids are delivery formats, not conversion strategies, and the human literature reflects that: no trial has compared them head to head for sulforaphane bioavailability⁷.

Fresh broccoli sprouts deliver sulforaphane better than most supplements, and that deserves saying before the limits are named. In a crossover of 12 subjects, fresh sprouts produced peak plasma sulforaphane seven-fold higher and 24-hour urinary excretion five-fold higher than a broccoli supplement taken by the same people¹⁵.

Broccoli sprouts sit inside a wider family: cruciferous vegetables such as kale, cabbage, Brussels sprouts and mature broccoli all carry glucosinolates and their own myrosinase. We hold no verified comparison of glucoraphanin content across them, so treat them as one category rather than as interchangeable doses.

The limit is heat. In a human feeding study, raw broccoli gave 37% bioavailability against 3.4% for cooked, and peak plasma time slid from 1.6 hours to 6 hours as bacterial conversion replaced the plant enzyme⁴.

The second limit is that you cannot verify what you ate: a tray of sprouts carries no panel, and we hold no verified source for how much glucoraphanin content varies from tray to tray, so treat that as unknown. Food gets the enzyme right and the dose approximate; a formulated product can, in principle, get both right — if it includes the enzyme.

Read next: fresh sprouts versus a supplement

Myrosinase, the enzyme that turns glucoraphanin into sulforaphane, stops working once broccoli gets hot enough.

Light heat can help. In lab tests on broccoli tissue, warming florets or sprouts to 60 °C before chopping raised the sulforaphane formed, because it switched off a second protein that otherwise turns the reaction toward sulforaphane nitrile instead. Heating florets to 70 °C and above lowered sulforaphane¹⁶. Steaming for 1 to 3 minutes gave more sulforaphane than boiling or microwaving in three of four broccoli varieties tested⁸.

Mustard restores formation after boiling. Boiled broccoli formed no significant sulforaphane in a lab test; stirring in powdered mustard seed afterwards brought formation back, because mustard carries its own myrosinase¹⁷. In 12 people, brown mustard powder stirred into cooked broccoli raised sulforaphane bioavailability³.

Frozen broccoli has lost its enzyme. Commercial frozen broccoli is blanched before freezing. When 18 people ate soup made from fresh or from frozen broccoli, sulforaphane bioavailability was about tenfold higher from the fresh⁹. In lab work, thawed frozen broccoli formed sulforaphane only when a myrosinase source was added, and 0.25% daikon radish was enough¹⁸.

Chop, then wait. In a lab stir-fry test, chopped broccoli left to stand for 90 minutes before stir-frying had 2.8 times the sulforaphane of broccoli stir-fried straight after chopping¹⁹. Ninety minutes is the only resting time that study measured.

Chewing. Sprouts chewed thoroughly led to more isothiocyanate metabolites in urine than sprouts swallowed whole: 42.4 against 28.8 µmol in a small human study¹⁰.

A capsule is not cooked at home, so what matters is whether its label names an active-myrosinase ingredient: the label section below shows where to find it.

In one line: boiling and commercial freezing remove the enzyme that makes sulforaphane; brief steaming, resting chopped broccoli before cooking, chewing well, and adding mustard or daikon each led to more sulforaphane.

Seven factors are routinely described as absorption levers. Four are supported by human data, one is in-vitro only, and two are not levers at all — they are explanations for why results differ.

1. Active myrosinase in the product. Active myrosinase in the product is the largest measured effect on this page. Adding a myrosinase source to a glucoraphanin source raised bioavailability up to roughly eight-fold³, and preparations containing active myrosinase reached three to four times the bioavailability of those without¹. A four-subject pilot using sprout powder as the enzyme donor points the same way¹¹.

2. Taking it with food, or with fat. Taking sulforaphane with food, or with fat, is the most repeated tip in the category and the least supported. No fed-versus-fasted human trial for sulforaphane was located in the search behind this page's sources²⁰; the advice is inference from general absorption principles. The closest human study put broccoli sprouts into protein, lipid and fibre gels in five subjects and found the gels lowered bioavailability rather than raising it — explorative, in the authors' own word, and not statistically significant for sulforaphane²⁰. Insufficient data. Take it with a meal if that is what keeps you consistent, not because absorption has been shown to depend on it.

3. Heat exposure. Heat exposure is an established lever, and it applies to the raw material as much as to your cooking. Raw versus cooked broccoli produced 37% versus 3.4% bioavailability in a human feeding study⁴.

4. A dose that clears the studied threshold. A dose that clears the studied threshold is the fourth lever, and one trial measured it directly. In a randomised placebo-controlled trial of 170 people, median 24-hour urinary sulforaphane metabolite output was 24.6 µmol at the full dose, 10.3 µmol at half dose, and 4.3 µmol at one-fifth dose. Only the full-dose group showed a statistically significant change in the detoxification biomarker the trial tracked; the half-dose and one-fifth-dose groups did not²¹.

5. Daily consistency. Daily consistency is not an absorption lever we can support. We have no source showing that sulforaphane absorption improves with repeated dosing, and no support for the claim that the body adapts its absorption to a schedule. Treat consistency as what makes a routine assessable over weeks rather than as an absorption mechanism.

6. Vitamin C as a cofactor. Vitamin C acts as a myrosinase cofactor in vitro, and only in vitro: crystal structures show ascorbate binding the enzyme's active site and substituting for its catalytic base²². That is structural biochemistry, not a feeding study, and no human trial of added vitamin C on sulforaphane bioavailability exists in our ledger. It belongs in a formula as a reasoned inclusion, not a promised boost.

7. Your gut bacteria. Your gut bacteria are not a lever you can pull; they are the main reason two people report different experiences on the same product⁵.

The evidence base for sulforaphane bioavailability is small, consistent in direction, and specific about what it does not cover.

Dose timing for sulforaphane is where confident advice most outruns the evidence. What is established is the shape of a single dose: plasma sulforaphane peaks around one hour after intake and clears with a half-life of 1.77 ± 0.13 hours⁶. A dose is a short pulse, not a plateau you top up.

When the enzyme is destroyed and gut bacteria have to do the conversion instead, peak plasma time moved from 1.6 hours to 6 hours⁴. Once it arrived, it cleared at a similar pace either way: excretion half-lives were 2.6 hours after raw broccoli and 2.4 hours after cooked, in eight men⁴.

What does not exist is any human trial comparing morning against evening dosing for sulforaphane⁷. Recommendations to take it at breakfast because digestion is more active then are reasoning from general circadian principles, not findings from a sulforaphane trial. Insufficient data. Take it at the time you will actually keep.

A supplement label answers the bioavailability question in about four lines, if you know which four.

Does an ingredient carry active myrosinase? A label should name active myrosinase as an ingredient line rather than describe it with a processing adjective. "Cold-pressed" and "low-temperature" describe a method nobody outside the factory can verify, while an enzyme named on the panel is a stated fact you can hold the maker to. This is the difference the human data is largest on^1,3.

Is the headline number precursor or product? Glucoraphanin and sulforaphane are not interchangeable units, and a panel that gives one while implying the other is doing the reader's arithmetic for them.

Does the dose reach what has been studied? The dose on the panel is worth comparing against the one trial on this page with a dose-response arm, which found the full dose distinguishable from placebo and the fractional doses not²¹.

And what cannot be read off any label: no label tells you whether the contents match the claim. We searched for a peer-reviewed analytical survey comparing label claims against measured content across commercial sulforaphane products and did not find one²³. Third-party testing documents are the only route, and they are a company-by-company matter, not a category fact.

Sproutly's own panel, as a worked example: 700 mg broccoli seed extract standardised to 13% glucoraphanin, 200 mg broccoli sprout extract carrying active myrosinase, 100 mg mustard seed extract, and 50 mg vitamin C, in a two-capsule serving, 30 servings per pouch (label). Seven hundred milligrams at 13% is 91 mg of glucoraphanin nominally in the serving, arithmetic off the panel rather than a measurement. The panel also states "Up to 30mg sulforaphane per serving"; that is the label's claim, and it is not a bioavailability figure. The sprout extract is where the active myrosinase sits, so the conversion step has a named source rather than an implied one. Mustard seed is listed alongside it; the panel assigns it no mechanistic role, and the human evidence for mustard as a myrosinase source is a 12-person crossover using mustard powder with food³, not a claim about this capsule. Vitamin C is included as a cofactor on in-vitro structural evidence²².

Read next: what independent testing showed

Read next: how to choose a sulforaphane supplement

Individual variation in sulforaphane conversion is large, measured, and mostly bacterial. When 23 adults ate the same 200 g of cooked broccoli, delivering roughly 120 µmol of glucosinolates, 24-hour urinary isothiocyanate output ranged from 1.2 to 34.7 µmol — about 1% to 29% of the dose⁵.

High excreters averaged 23.4 ± 7.5 µmol per 24 hours against 5.6 ± 2.7 µmol for low excreters, and when their own faecal bacteria were incubated with glucoraphanin outside the body, the high excreters' bacteria degraded 13.3 ± 5.9% of it in 48 hours versus 5.6 ± 4.8% for the low excreters⁵. The difference between people was reproducible in a dish.

If a product relies on your bacteria to do the conversion, your result is a lottery someone else already ran differently. We hold no source attributing this variation to age, genetics, or medication use — those explanations circulate, but not with evidence we can cite here.

Five habits reduce how much sulforaphane you actually absorb, each mapping to a lever above.

Buying the precursor without the enzyme. Buying glucoraphanin with no enzyme source hands the conversion step to bacteria that may or may not perform it⁵; preparations with active myrosinase were three- to four-fold more bioavailable¹.

Cooking the sprouts you are relying on. Heat is what removes myrosinase activity⁴, so if sprouts are your enzyme source, cooking them removes the reason you were eating them.

Taking a fraction of a studied dose. Only the full dose was distinguishable from placebo in the one trial with dose arms²¹. A smaller dose is not a slower version of the same thing.

Reading the front of the pack instead of the panel. The front carries claims; the panel carries ingredient lines and amounts. No peer-reviewed survey verifies label claims against measured content across this category²³.

Judging a product by taste, or by a feeling on a timeline. No source in this ledger connects a peppery taste, or any other subjective sensation, to a measured absorption figure.

The 2019 Am J Clin Nutr trial of 170 adults is the reason absorption is worth optimising: only its full-dose group produced a statistically significant change in the detoxification biomarker being measured, while the two fractional-dose groups did not²¹. The dose was load-bearing. A product that converts poorly is not delivering a smaller version of the studied intake; it may not be reaching the studied intake at all.

Within what that evidence supports: sulforaphane supports the body's natural detoxification processes, at the full studied dose rather than at any dose. It is also studied as an activator of the Nrf2 pathway, which regulates the body's own antioxidant and detoxification enzymes²⁴ — the basis for the softer statement that it supports the body's antioxidant defences. That one stays soft on purpose: it is mechanism, not a clinical result, and the one human trial built to test Nrf2 target-gene induction directly was null on its primary endpoint²⁴. Both are statements about supporting normal function, not claims about any condition.

Read next: the body's natural detoxification processes

Glucoraphanin is not sulforaphane, and the enzyme that connects them is the step the supplement category most often leaves out: with active myrosinase, human bioavailability measured about 40–41.8%; without it, about 9.4–11.5%¹. Format and dose timing are unmeasured in humans⁷, and taking it with food has never been tested head to head against taking it fasted²⁰, so neither belongs anywhere near the top of your list. Check the panel for a named active-myrosinase ingredient, check that the dose is not a fraction of what was studied²¹, and give any routine weeks rather than days before you judge it.

1. Fahey JW, et al. 2015. PLoS One. Crossover, n=5 core subjects, augmented to n=22 for the supplement comparison. Bioavailability ~9.4–11.5% of dose without active myrosinase; ~40–41.8% with it (100 µmol dose). PMID 26524341 · https://doi.org/10.1371/journal.pone.0140963
2. Egner PA, et al. 2011. Cancer Prev Res. Randomised crossover, n=50 (47 completed). ~70% of dose recovered in 24-h urine from the pre-hydrolysed sulforaphane beverage vs ~5% from the glucoraphanin-rich beverage; median 12-h excretion 93.9 vs 19.3 µmol. PMID 21372038 · https://doi.org/10.1158/1940-6207.CAPR-10-0296 (cited for bioavailability figures only)
3. Okunade O, et al. 2018. Mol Nutr Food Res. Randomised crossover, n=12. Cooked broccoli plus 1 g powdered brown mustard raised sulforaphane bioavailability up to ~8-fold, with urinary SFN-NAC up ~4-fold. PMID 29806738 · https://doi.org/10.1002/mnfr.201700980
4. Vermeulen M, et al. 2008. J Agric Food Chem. Human feeding study, raw vs cooked broccoli. Bioavailability 37% raw vs 3.4% cooked; peak plasma time 1.6 h vs 6 h. PMID 18950181 · https://doi.org/10.1021/jf801989e
5. Li F, Hullar MAJ, Beresford SAA, Lampe JW. 2011. Br J Nutr. Feeding study plus ex vivo faecal incubation, n=23 (10 selected). 24-h urinary ITC output 1.2–34.7 µmol on 200 g cooked broccoli; high vs low excreters 23.4 ± 7.5 vs 5.6 ± 2.7 µmol (p<0.01); ex vivo degradation 13.3 ± 5.9% vs 5.6 ± 4.8% (p=0.05). PMID 21342607 · https://doi.org/10.1017/S0007114511000274
6. Ye L, et al. 2002. Clin Chim Acta. n=4 healthy volunteers, 200 µmol broccoli sprout isothiocyanates. Cmax 0.943–2.27 µmol/L at ~1 h; first-order elimination half-life 1.77 ± 0.13 h; 24-h urinary excretion used as the bioavailability endpoint. PMID 11750273
7. Evidence gap. No human trial comparing capsule vs powder vs liquid delivery formats for sulforaphane bioavailability was located, and no sulforaphane-specific morning-vs-evening dosing trial was located. Sources recommending a dosing time are supplement commentary reasoning from general circadian principles, not trial findings. *Scope: this reference covers those two absences only* — the capsule/powder/liquid comparison and morning-vs-evening dosing. It is not cited for any other missing comparison on this page.
8. Wang GC, Farnham M, Jeffery EH. 2012. J Agric Food Chem. Lab study, four broccoli cultivars. Steaming for 1.0–3.0 min increased sulforaphane yield in three of the four; boiling and microwaving lost sulforaphane within 1 min. PMID 22471240 · https://doi.org/10.1021/jf2050284
9. Saha S, et al. 2012. Mol Nutr Food Res. Human study, n=18. Sulforaphane bioavailability about tenfold higher from soup made with fresh broccoli than with commercial frozen broccoli; the blanching before freezing destroyed myrosinase. PMID 23109475 · https://doi.org/10.1002/mnfr.201200225
10. Shapiro TA, et al. 2001. Cancer Epidemiol Biomarkers Prev. Human excretion study. Isothiocyanates are substantially more bioavailable than intact glucosinolates; cited here for mechanism only, not for any percentage figure. PMID 11352861 (background only) Since October 2026, also cited for one figure: chewing versus swallowing sprouts whole, 42.4 ± 7.5 vs 28.8 ± 2.6 µmol dithiocarbamates excreted (P = 0.049).
11. Cramer JM, Jeffery EH. 2011. Nutrition and Cancer. Pilot, n=4, four conditions per subject. Myrosinase-rich air-dried broccoli sprout powder combined with glucoraphanin-rich broccoli powder enhanced and accelerated sulforaphane absorption vs powder alone. https://doi.org/10.1080/01635581.2011.523495 (pilot strength, n=4; PMID not independently confirmed)
12. Mastaloudis A, Holcomb L, Fahey JW, et al. Sci Rep. 2026 Feb 15;16:9162. Randomised, double-blind, single-dose crossover, n=16 (9 female, 7 male). Sulforaphane bioavailability 39.8 ± 3.1% (range 5.3–76.9%) for a glucoraphanin-rich broccoli seed extract taken with mustard-seed myrosinase vs 18.6 ± 3.1% (range 2.2–38.1%) for the extract alone. PMID 41692762 · https://doi.org/10.1038/s41598-026-39389-4. Funding statement: primarily funded by Brassica Protection Products LLC; one author employed by the funder, one a consultant to it.
13. Evidence gap. No human trial was located that compares a stabilised or pre-formed sulforaphane product head to head against a glucoraphanin-plus-myrosinase product in the same study. Searches covered stabilised and cyclodextrin-complexed sulforaphane, comparative bioavailability, and head-to-head designs against glucoraphanin or myrosinase. Closest study found: the SFX-01 (enteric-coated, cyclodextrin-stabilised sulforaphane) Phase 1 randomised, double-blind, dose-escalation trial in healthy male participants, PMID 39520658 · https://pubmed.ncbi.nlm.nih.gov/39520658/. It does not qualify because its only comparator is placebo — the trial carries no glucoraphanin-plus-myrosinase arm and no sprout- or seed-extract arm, so it establishes its own pharmacokinetics rather than a comparison.
14. Evidence gap. No human trial was located that doses the same subjects, or a matched cohort under one protocol, with both a synthetic and a plant-derived sulforaphane preparation and compares plasma or urinary recovery. Searches covered synthetic versus natural sulforaphane bioavailability and the racemic-versus-single-enantiomer framing. Closest study found: the same SFX-01 Phase 1 trial, PMID 39520658, which is relevant because SFX-01 is synthetic racemic sulforaphane while plant-derived sulforaphane is the single R-(−) enantiomer. It does not qualify because it has no plant-derived comparator arm; setting its numbers beside those from the plant-derived studies here would be a cross-study comparison confounded by population, dose and assay, not a controlled head-to-head.
15. Clarke JD, et al. 2011. Pharmacological Research. Crossover, n=12 (plus 4 controls), one-month washout. Fresh broccoli sprouts produced 7-fold higher peak plasma sulforaphane and 5-fold higher 24-h urinary excretion than a broccoli supplement in the same subjects. PMID 21816223 · https://doi.org/10.1016/j.phrs.2011.07.005
16. Matusheski NV, Juvik JA, Jeffery EH. 2004. Phytochemistry. Lab study, broccoli florets and sprouts. Heating to 60 °C before homogenisation increased sulforaphane and decreased sulforaphane nitrile; 70 °C and above decreased both in florets. PMID 15184012 · https://doi.org/10.1016/j.phytochem.2004.04.013
17. Ghawi SK, Methven L, Niranjan K. 2013. Food Chem. Lab study. Boiling prevented any significant sulforaphane formation; adding powdered mustard seed to the heat-processed broccoli significantly increased it. PMID 23411305 · https://doi.org/10.1016/j.foodchem.2012.10.119
18. Dosz EB, Jeffery EH. 2013. J Food Sci. Lab study. Blanching at 86 °C or higher inactivated myrosinase; thawed frozen broccoli formed sulforaphane only when a myrosinase source was added; 0.25% daikon radish supported formation. PMID 23915112 · https://doi.org/10.1111/1750-3841.12221
19. Wu Y, et al. 2018. J Agric Food Chem. Lab study. Letting chopped broccoli hydrolyse for 90 min before stir-frying increased sulforaphane 2.8-fold versus stir-frying straight away; myrosinase activity fell 80% after 3 min of stir-frying. PMID 29357241 · https://doi.org/10.1021/acs.jafc.7b05913
20. Oliviero T, et al. 2018. Mol Nutr Food Res. Crossover, n=5, sprouts incorporated into protein/lipid/fibre gels. Gels lowered bioavailability of preformed isothiocyanates versus sprouts alone; not statistically significant for sulforaphane specifically; described by the authors as explorative. PMID 29532635 · https://doi.org/10.1002/mnfr.201700837 (explorative, small). *No fed-versus-fasted human trial was located in the same search*; this entry is the citation for that absence as well as for the food-matrix finding.
21. Chen J-G, et al. 2019. Am J Clin Nutr. Randomised, placebo-controlled, n=170 across placebo and three dose levels. Median 24-h urinary sulforaphane metabolites 24.6 µmol (full dose), 10.3 µmol (½ dose), 4.3 µmol (⅕ dose); only the full-dose group reached statistical significance on the measured detoxification biomarker (+63.2%). PMID 31268126 · https://doi.org/10.1093/ajcn/nqz122 (cited for dose-response and bioavailability figures only)
22. Burmeister WP, et al. 2000. J Biol Chem. X-ray crystallography, *in vitro / structural biochemistry — not a human or feeding study*. Ascorbate binds the myrosinase active site and substitutes for the enzyme's catalytic base, acting as a cofactor for the hydrolysis step. PMID 10978344 · https://doi.org/10.1074/jbc.M006796200
23. Evidence gap. No peer-reviewed analytical survey comparing label claims against measured glucoraphanin or sulforaphane content across commercial products was located. Trade-press and seller-published "independent lab" comparisons are marketing content and are not cited here.
24. Yagishita Y, Fahey JW, Dinkova-Kostova AT, Kensler TW. 2019. Molecules. *Review article, not a new trial: synthesises in-vitro, animal and human-biomarker evidence that sulforaphane activates the Keap1–Nrf2–ARE pathway, which upregulates phase 2 antioxidant and detoxification enzymes including the glutathione S-transferases and NQO1. PMID 31590459 · https://doi.org/10.3390/molecules24193593 (cited for mechanism only — not for any clinical outcome, and not as evidence of an individual response at a given dose). Caveat carried into the body copy:* the one human RCT purpose-built to test Nrf2 target-gene induction by oral sulforaphane was null on its primary endpoints (PMID 27832073); that trial ran in a clinical population and is cited here for its null methodological point alone, with its clinical context deliberately not carried onto this page.