The Lamp Is the Famous Risk. The Acrylate Is the Measured One.

In January 2023 a paper reported that a UV nail dryer damaged DNA and produced mutations in mammalian cells. It went everywhere, and the shorthand it produced — gel manicures cause cancer — has been sitting in the culture ever since.

The paper is real, careful and honestly titled. It also studied no people, and the exposure it used is not the exposure a hand receives.

What the 2023 study did

Three cell types were irradiated: primary mouse embryonic fibroblasts, human foreskin fibroblasts and human epidermal keratinocytes. Cultures in dishes, inside the lamp.

The device measured 7.5 milliwatts per square centimeter. Each session ran twenty minutes, which the authors calculate as nine joules per square centimeter delivered. Acute exposure meant two of those in a day. Chronic meant one on each of three consecutive days.

A single twenty-minute irradiation killed 20 to 30 percent of the cells. Three consecutive sessions killed 65 to 70 percent. The surviving cells showed reactive oxygen species, the oxidative DNA lesion 8-oxo-guanine, mitochondrial dysfunction, and a dose-dependent rise in a specific substitution pattern matching signatures previously attributed to reactive oxygen species.

All of that is measured and none of it is in dispute. What the paper does not show is a cancer, a person, or a dose anybody’s hand receives — a gel cure step runs a fraction of twenty minutes, and the authors’ own title claims only DNA damage and somatic mutations in mammalian cells. Twenty minutes in a sealed plate is not a manicure.

The replication that found the heat

A 2024 toxicology group set out to reproduce it, opening with the observation that the original report was surprising given the comparatively low level of UVA the skin receives during nail treatments.

Three things came out of it. UVA levels inside the unit were highly variable and depended on where in the box the sample sat. Prolonged exposure protocols did induce significant DNA damage, confirming the original. And the DNA damage only partially correlated with the UVA dose.

So they looked for what else was happening, and found that the unit’s internal temperature rose rapidly in use. Exposing human cells to those elevated temperatures acted synergistically with UVA to magnify both the cytotoxic and the genotoxic effect.

A culture dish sitting in a closed, heating box is not a hand resting in an open-sided lamp with blood flow through it. Part of the damage in the frightening result may have been the temperature, and a hand does not experience that confound.

A separate 2023 group tested the timing question directly: four minutes versus twenty, with and without SPF50. At four minutes, keratinocyte viability fell insignificantly. At twenty, it fell by 35 percent. Their conclusion is that typical manicure exposure time does not significantly affect keratinocyte viability.

The dose, in units you can picture

A Spanish group compared the emission spectrum of curing lamps against midday summer sun at mid-latitude, and worked out the biologically effective doses for erythema, non-melanoma skin cancer, DNA damage, photoimmunosuppression and pigmentation.

The total ultraviolet and high-energy visible dose per manicure session corresponded to three and a half to six minutes of noon sun in summer.

Their conclusion is that this represents a very low carcinogenic potential compared with sunlight — and, in the same sentence, a similar immunosuppressive potential, which is the part worth not skipping.

An earlier photobiological safety evaluation of six lamps from major US manufacturers placed them in the low to moderate risk groups at the hand position, with permissible daily exposure times running from about half an hour to over four hours, and found the total exposure from following the programmed cure times amounted to only a small fraction of the permissible daily occupational exposure.

And the mathematical modeling, from 2012, put it in the plainest terms available: for typical usage, tens or hundreds of thousands of women would need to use a lamp regularly for one to go on to develop a squamous cell carcinoma on the back of the hand as a direct consequence. That is a model, not a cohort, and I would rather say so than let it stand as epidemiology.

What the human evidence actually consists of

A 2025 review pulled together fourteen articles. Squamous cell carcinoma was described in five patients with a history of lamp use. Two in-vitro studies, disagreeing with each other. One registry analysis finding no melanoma cases associated with the lamps. Seven mathematical models, six of which calculated minimal to no risk.

Five patients, from case reports. That is the human evidence base, in total, and the reported interval between exposure and diagnosis in those cases ran from eleven to fifteen years.

A separate systematic review screened 2,331 articles and included nine, rating the risk of bias as high or unclear, and concluding that prolonged and repeated exposure may pose a low risk — while noting explicitly that the available evidence is weak.

Photoprotection for anyone using these lamps regularly and long-term is a reasonable, cheap precaution and the 2025 reviewers recommend it. Fingerless gloves or sunscreen on the backs of the hands cost nothing and settle the question for the individual. What is not supported is the sentence people have taken away.

The risk that is actually rising

While that argument has been running, something measurable has been happening, and hardly anybody outside dermatology clinics is talking about it.

2-hydroxyethyl methacrylate — HEMA — is the acrylate in gel and acrylic nail systems. It is a potent sensitizer, and once you are allergic to it you are allergic for life.

A seven-center European audit patch-tested 26,297 patients between 2016 and 2023. HEMA allergy from all sources ran at 2.82 percent in women against 0.34 percent in men. Nail-related contact allergy rose from 0.91 percent in 2016 to 1.98 percent in 2023 — though the number of contributing centers rose from two to seven over that period, so the trend is not strictly like for like, and the paper says so in its own numbers.

Their conclusion is blunt: EU legislation appears not to have had the intended impact on controlling allergic contact dermatitis from methacrylates in nail cosmetics.

A tertiary center in Israel tested 1,671 consecutive patients and found 135 positive — 8.1 percent, which the authors report as the highest frequency to date. In women, 11.0 percent against 1.0 in men. In women under thirty, 16.7 percent.

One in six. Of those with a clinically relevant reaction, nail cosmetics accounted for 95 percent, and eighteen of the twenty occupational cases were nail stylists.

A Spanish study across 24 centers tested 6,134 patients and found 4.3 percent positive, with occupational exposure carrying an odds ratio of 10.9 for a relevant reaction. A European study of 18,228 patients found that among those with allergic contact dermatitis from nail acrylates, 65 percent of the occupationally exposed developed it during their first year at work — and at least 11.7 percent had to leave their jobs.

And the detail nobody expects. In North American data covering 38,775 patch-tested patients, the primary site of nail-product contact dermatitis was the face in 43.0 percent of cases and the hands in 27.6. In the European series, the face was the only affected site in one in ten.

A rash on your cheeks and eyelids that you would never connect to your fingernails, because it is on the part of you your fingers touch. In the systematic review of gel polish side effects, the allergy developed on average thirty months after starting.

What is measured about the nails themselves

Less than you would think.

The claim that gel manicures thin the nail plate traces largely to a 2012 report of five subjects. Objective thickness was measured in one of them, by two methods that disagreed with each other — ultrasound showed 0.063 down to 0.050 centimeters, confocal microscopy 0.059 down to 0.030.

One person is the quantitative basis for most of the percentages in circulation.

The largest data is a self-questionnaire of 2,118 women, of whom 48.3 percent reported side effects while applying gel polish, around 20 percent while wearing it, and more than 75 percent after removing it — with nail plate changes significantly more frequent after removal than at any other stage, and associated with whether a professional did the procedure. Entirely self-reported, no examinations, but the removal signal is consistent and mechanically plausible.

On infection, the documented salon outbreak is not about gel at all. It is footbaths. An investigation into 110 customers of one salon with boils on their legs found rapidly growing mycobacteria, cultured the same organism from all ten footbaths, and identified shaving the legs before the pedicure as a risk factor with an odds ratio of 4.8. A follow-up swabbed 30 footbaths across 18 salons and found mycobacteria in 29 of them.

Do not shave your legs the morning of a pedicure. That one is well established and almost never said.

The people in the room all day

The exposures that matter most in a nail salon belong to whoever is working there.

Personal air monitoring in six Colorado salons found formaldehyde from 5.32 to 20.6 micrograms per cubic meter, toluene from 26.7 to 816, benzene from 3.13 to 51.8. One salon exceeded the national recommended exposure limit for formaldehyde. Seventy percent of the twenty workers surveyed reported at least one work-related health issue. The authors note that the aromatic concentrations were comparable to those measured in oil refinery and auto garage workers.

A study of 112 workers across 25 East Coast salons measured a mean of 39.45 parts per million of methyl methacrylate — a substance the US regulator has acted against in nail products, still being measured in the air years later.

Now the scale of the evidence. A 2026 scoping review of the entire nail salon workforce literature found 42 studies, 88 percent of them American, with sample sizes ranging from twenty to 150 workers. Inadequate ventilation in 71 percent. Only 17 percent examined any link between an exposure and a health outcome.

Twenty to 150 workers per study, for an industry of hundreds of thousands of people.

Two things cut against the alarm, and both belong in an honest account. A Polish screening study found 70 percent of manicurists reporting nasal symptoms on a questionnaire, 41 percent on interview, and 19 percent finally diagnosed with a work-related nasal irritant reaction — a threefold attenuation that is worth remembering every time a survey percentage is quoted. And the largest cancer study, linking 325,228 licensed California cosmetologists and manicurists to the cancer registry across 9,044 cancers, found lower overall incidence than the general female population: a rate ratio of 0.87 for manicurists. The authors flag that limited demographic data and the relative youth of the cohort may be shaping that result, and call for further follow-up. It is still the largest number in the field, and it points the other way.

The reproductive analysis from the same linked dataset found the clearer signals: gestational diabetes at an odds ratio of 1.28 for manicurists, rising to 1.59 in Vietnamese manicurists specifically, and placenta praevia at 1.46.

One correction about who these workers are

The nail salon workforce is nearly always described as Vietnamese women. I went looking for the enumeration behind that and there is no peer-reviewed national count of the US nail salon workforce by ethnicity.

The figure everyone uses — that 59 to 80 percent of California nail salons are run by Vietnamese women — appears in the introduction of a 2008 survey as background rather than as that paper’s own measurement. It is California-only, and it describes who runs salons rather than who works in them.

A 2022–23 survey of 146 registered salons and manicurists in New York and New Jersey found the largest group was Korean, at 52.1 percent, followed by Chinese at 26.

A California finding has been generalized to a country. Which matters, because the interventions that follow — translated training materials, outreach, inspection protocols — get designed around the wrong language and the wrong community. In that same New York and New Jersey survey, 79 percent of New York salons had mechanical ventilation against 52 percent in New Jersey, where no ventilation regulation exists, and 67.6 percent of manicurists reported symptoms related to chemical exposure.

The gap between those two states is a policy variable somebody could act on tomorrow.

Gloves, patch tests, and one thing about pedicures

Sunscreen or fingerless gloves before the lamp, because it costs nothing and it removes the question. Not because the cancer case is made — it is not — but because the immunosuppressive dose is comparable to a few minutes of noon sun, and a few minutes of noon sun several times a month for a decade is a real quantity.

Pay much closer attention to the allergy. If you develop dermatitis on your eyelids, cheeks or neck and cannot work out why, and you wear gel or acrylic nails, say so to whoever is assessing you and ask about HEMA. It is on the standard patch test series in several countries now, precisely because it was being missed.

And do not shave before a pedicure.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.