Alpha Arbutin vs Kojic Acid vs Glutathione: What the Evidence Shows

Alpha arbutin vs kojic acid is a closer contest than labels suggest. Both target tyrosinase, the enzyme that starts melanin production, and on the human enzyme arbutin and kojic acid both test weak. Glutathione acts partly on the same step, and its topical evidence is thin. Below: how each works, what human studies measured, the caps Europe now sets, and how all three compare with tranexamic acid and niacinamide.

How tyrosinase brighteners work, and why the enzyme source matters

Melanin production starts with tyrosinase. A 2018 paper in the Journal of Investigative Dermatology calls it the rate-limiting enzyme of melanin production. That makes it the obvious target for fading dark spots.

The same paper names a problem. Most known tyrosinase inhibitors were found by testing them on mushroom tyrosinase, and most of them lack clinical efficacy. So the team screened 50,000 compounds against recombinant human tyrosinase instead.

The results reset expectations. Hydroquinone and arbutin inhibited the human enzyme only weakly. The dose needed to halve its activity sat in the millimolar range. Kojic acid needed more than 500 micromoles per liter. The best hit, a compound called thiamidol, needed about 1.1. Lower numbers mean a stronger inhibitor.

A commentary in the same journal summed up the lesson in its title: save mushrooms for the pizza. Keep that in mind whenever a brightener's pitch rests on enzyme tests.

Tyrosinase is not the only lever. Niacinamide had no effect on mushroom tyrosinase in one key study. Yet it cut the transfer of pigment packages to skin cells by 35 to 68 percent in a coculture model. Tranexamic acid is thought to interfere with signaling between pigment cells and surrounding skin cells, through the plasminogen and plasmin system. We cover it in tranexamic acid in skincare.

Alpha arbutin vs kojic acid vs glutathione at a glance

	Alpha arbutin	Kojic acid	Glutathione
What it is	Hydroquinone bound to glucose, made by chemical or enzymatic synthesis	A compound made by fungi, mainly Aspergillus species	A three amino acid antioxidant your body makes
How it is thought to work	Tyrosinase inhibition	Tyrosinase inhibition, binding the enzyme's copper	Tyrosinase inhibition, plus a shift toward lighter pheomelanin
Key human evidence	One 30-person split-face pilot, combined with kojic acid	Small melasma trials, mostly combined with hydroquinone or glycolic acid	One 30-woman split-face trial at 2 percent
EU limit in face products	2 percent in face creams	1 percent in face and hand products	Not covered by the 2024 EU amendment
Main caution	Can release hydroquinone under some conditions	Contact allergy	Limited safety data

Alpha arbutin: where the "10 times stronger" claim comes from

Arbutin comes in two forms. Both are glycosylated hydroquinones, meaning hydroquinone with a glucose attached. Beta arbutin, usually just called arbutin, occurs in plants such as bearberry, wheat and pear. Alpha arbutin is made by chemical and enzymatic methods.

Alpha arbutin is often sold as ten times stronger. Even a 2018 review repeats that it is over 10 times more effective than arbutin. The figure traces to a 1995 enzyme study. There, alpha arbutin inhibited tyrosinase from mouse melanoma 10 times as strongly as beta arbutin. The same study found it did not inhibit mushroom tyrosinase at all. The famous number describes a mouse enzyme in a test tube.

Human cell data exist too. In cultured human melanoma cells, 0.5 millimolar alpha arbutin cut melanin to 76 percent of untreated levels. In a three-dimensional human skin model, it cut melanin to 40 percent of control without harming cell viability.

Human trials are scarce. The most direct one is a 2025 split-face pilot in 30 people with melasma. One side got a cream with 5 percent alpha arbutin and 2 percent kojic acid for 12 weeks. The other side got triple combination cream, the reference treatment for melasma. Changes in melanin index and severity score did not differ between sides. Physician global scores improved only on the triple combination side. That side also relapsed more after stopping, and it caused more redness and stinging (2025 pilot trial).

Two caveats matter. The trial mixed alpha arbutin with kojic acid, so neither gets clean credit. And its 5 percent alpha arbutin is more than double the 2 percent cap the European Union now sets for face creams.

Lab results on arbutin are inconsistent too. A 2022 study set out to explain contradictions in the literature. On mushroom tyrosinase, both arbutins inhibited one enzyme activity while activating the other. Test design can flip the answer.

The hydroquinone question

Arbutin is hydroquinone with a sugar attached. The obvious question is whether the sugar comes off. Sometimes it can.

In a lab study, common skin bacteria, including Staphylococcus epidermidis, hydrolyzed arbutin to hydroquinone. A 2016 stability study found both arbutins released hydroquinone under strong hydrolytic conditions. Their stability in finished products depended on formulation type and pH.

Regulators took note. The European Union's 2024 amendment caps alpha arbutin at 2 percent in face creams and 0.5 percent in body lotions. It caps arbutin at 7 percent in face creams. For both, hydroquinone must stay as low as possible, no higher than unavoidable trace levels. Noncompliant products could not be placed on the EU market from February 2025.

A 2021 review lands in the same place. It describes dermatitis from arbutin as rare but urges caution, because hydroquinone may be generated during product use. For why hydroquinone itself is handled so carefully, see tranexamic acid vs hydroquinone.

Kojic acid: fungal origin, copper and contact allergy

Kojic acid is made by fungi, primarily Aspergillus species. It inhibits tyrosinase partly by binding the copper at the enzyme's active site. On the human enzyme, as noted above, it tested weak.

Its human trials are old, small, and usually paired with something stronger. In 1996, 39 patients used kojic acid with glycolic acid on one side of the face. The other side got hydroquinone with glycolic acid. About half responded equally. Kojic acid did better in 28 percent and hydroquinone in 21 percent, a difference that was not statistically significant. The kojic side was more irritating.

In 1999, 40 Chinese women with melasma used a gel of 10 percent glycolic acid and 2 percent hydroquinone on both sides. Adding 2 percent kojic acid to one side helped. More than half the melasma cleared in 60 percent of patients on the kojic side versus 47.5 percent without it.

A 2023 meta-analysis pooled 45 studies and 2,359 patients for efficacy. On a standardized scale, kojic acid lowered melasma severity scores by 0.9, against 1.3 for hydroquinone alone. Pooled irritation for kojic acid was 5.3 percent. The authors concluded that non-hydroquinone agents may be considered as alternatives, with zinc sulfate the exception.

Allergy is the known downside. In a 1995 Japanese series, 8 of 220 women with suspected cosmetic dermatitis had used kojic acid products. Five of those eight reacted to kojic acid on patch testing, after 1 to 12 months of use. The authors judged it to have high sensitizing potential. The EU now caps it at 1 percent in face and hand products.

Glutathione: why the topical evidence is thin

Glutathione is a tripeptide of cysteine, glycine and glutamate, and a major antioxidant your body makes. Its lightening effect is attributed to direct and indirect tyrosinase inhibition. It is also credited with a switch from dark eumelanin toward lighter pheomelanin production.

The best topical trial is small. Thirty healthy women aged 30 to 50 applied a 2 percent oxidized glutathione lotion to one side of the face. Placebo went on the other side, twice daily for 10 weeks. The treated side had a significantly lower melanin index by the end. It also scored better on hydration and smoothness (2014 split-face trial). Note the form. This trial used oxidized glutathione, not the reduced form usually discussed.

Beyond that, the evidence runs out quickly. A 2025 systematic review of topical glutathione screened 446 articles and found only five clinical trials to include. It called the research limited, and the safety data limited too. A second 2025 review found 0.5 percent topical glutathione beat 0.1 percent and placebo. It also described lightening from topical and oral forms as unsustainable. An earlier 2019 review called the overall evidence inconclusive.

Intravenous glutathione drips are a separate issue. A 2016 review found no evidence of their efficacy. It also noted a public warning from the Philippine Food and Drug Administration over adverse effects (2016 review). The 2025 review calls intravenous glutathione contraindicated. That is a medical procedure, not skincare, and it sits outside this guide.

How they compare with tranexamic acid and niacinamide

A 2019 systematic review of 35 randomized melasma trials gave strong recommendations to three options. They were cysteamine, triple combination therapy and tranexamic acid. Neither arbutin nor kojic acid was among them. The review said natural compounds carry low risk, but need more research on efficacy, formulation and concentration.

A 2018 review of natural pigment ingredients reached a similar view. It found few clinical trials, mostly short, with open questions about long-term efficacy and safety.

Different mechanisms also mean different jobs. Tyrosinase inhibitors act on pigment production. Niacinamide acts on pigment transfer. Tranexamic acid acts on signaling between cell types.

If you are choosing among the three:

• Alpha arbutin has the mildest reported safety record, with dermatitis described as rare. The EU cap of 2 percent in face creams is a sensible ceiling to look for.
• Kojic acid has more trial history, mostly in combinations. Its main risk is contact allergy, so patch test for several days. Stay at or below 1 percent in face products.
• Glutathione has the thinnest topical evidence of the three. Treat it as optional.

Whichever you pick, wear daily broad-spectrum sunscreen. Melasma also behaves differently from ordinary spots. Our guides to how to fade dark spots and melasma vs hyperpigmentation cover the order that works. For spreading or recurring patches, see a board-certified dermatologist.

Frequently asked questions

Is alpha arbutin better than kojic acid?

We found no trial comparing them head to head. Kojic acid has more clinical history, mostly in combination with hydroquinone or glycolic acid, and a clearer allergy signal. Alpha arbutin has less human data and a milder safety record. On the human enzyme, arbutin and kojic acid both tested weak.

Can you use alpha arbutin and kojic acid together?

The one recent split-face trial used both at once, at 5 percent and 2 percent. Combining them makes it harder to tell which one caused a reaction. Keep each within the EU limits, 2 percent for alpha arbutin in face creams and 1 percent for kojic acid. Add one new product at a time.

Does alpha arbutin turn into hydroquinone on skin?

It can under some conditions. Both arbutins release hydroquinone under strong hydrolytic conditions in the lab, and one study showed skin bacteria can split regular arbutin. Stability depends on the formulation and its pH. EU rules require hydroquinone in arbutin products to stay at unavoidable trace levels.

Does topical glutathione lighten skin?

Possibly, a little. One 30-woman trial of 2 percent oxidized glutathione lowered the melanin index against placebo over 10 weeks. A 2025 systematic review found only five topical clinical trials and called the research and safety data limited. Treat it as unproven rather than disproven.

Which is gentlest for sensitive skin?

Arbutin has the mildest reported profile, with dermatitis described as rare. Kojic acid causes contact allergy in some users, and it was more irritating than hydroquinone in one comparison. Glutathione's topical safety data are too limited to rank. Patch test any of them for several days first.

How long do these brighteners take to work?

The trials above ran 10 to 12 weeks, some with follow-up after stopping. Pigment fades slowly, so judge any brightener after about three months of consistent use. Daily sunscreen has to be part of that test, or the result means little.