Why Is American Sunscreen Still Stuck on a 1999 List?

On June 9, 2026, the Food and Drug Administration announced that it had added a new active ingredient to the over-the-counter sunscreen monograph.⁷ The ingredient is bemotrizinol, a photostable organic filter that absorbs across both the UVB and UVA ranges. The final order, OTC000039, was issued on June 10 and took effect on August 9, 2026.⁴ It was announced in the Federal Register the same day under docket FDA-2025-N-6494.⁵ It is the first new sunscreen active ingredient the agency has permitted since the late 1990s.⁷

That is a real change and worth saying plainly. It is also narrower than most of the coverage suggested. The bigger question the FDA opened in 2019, which is whether the filters Americans have actually been buying for decades still meet the agency's own safety standard, remains open. Nothing about it was resolved in June.

What actually changed in American sunscreen this summer?

One ingredient was added to a list of sixteen. The amended monograph, printed in full inside the final order, now permits seventeen sunscreen active ingredients: aminobenzoic acid, avobenzone, bemotrizinol, cinoxate, dioxybenzone, ensulizole, homosalate, meradimate, octinoxate, octisalate, octocrylene, oxybenzone, padimate O, sulisobenzone, titanium dioxide, trolamine salicylate and zinc oxide, each capped at a stated concentration.⁴

Two names on that list are the tell. Aminobenzoic acid, usually called PABA, is still permitted at up to 15 percent, and trolamine salicylate at up to 12 percent.⁴ The FDA has now twice proposed striking both. In 2019 it proposed that neither is generally recognized as safe and effective, citing allergic and photoallergic reactions and cross-sensitization for PABA, and bleeding risk from anticoagulant effects and salicylate toxicity for trolamine salicylate.² It proposed the same thing again in 2021.¹ Neither proposal has been finalized. The two ingredients the agency has said it wants off the list are still legally on it, and most of the filters sold everywhere else in the world are still not.

Where does the FDA sunscreen monograph actually stand right now?

The popular version of this story gets the mechanism wrong. Sunscreen in the United States is regulated as a drug, not a cosmetic, and most sunscreen reaches the market without any product-specific approval. It is legal because it conforms to a monograph, which is a standing rulebook of permitted ingredients, concentrations, dosage forms, tests and label language.

The operative rulebook is still the 1999 final monograph. In 2019 the FDA proposed to rewrite it through ordinary notice-and-comment rulemaking.² Before that rulemaking could finish, Congress rebuilt the entire process. The Coronavirus Aid, Relief, and Economic Security Act, enacted March 27, 2020, replaced monograph rulemaking with an administrative order process under section 505G of the Food, Drug, and Cosmetic Act, and it deemed the existing sunscreen requirements to be a final order on the day it passed.⁴ That deemed final order is Monograph M020. It is why the 1999 list is still the law.

The 2019 proposal was then reissued in the new format as Proposed Administrative Order OTC000008, signed September 24, 2021, with the notice of availability published three days later.¹³ If finalized, it would replace the deemed final order in its entirety.³ It has not been finalized. The June 2026 bemotrizinol order states this directly in a footnote, noting that it does not address OTC000008 and that if OTC000008 is ever finalized, all products marketed under M020, bemotrizinol products included, would have to meet the new conditions.⁴

So the honest status line, as of today, is this. One new filter is in. The full rewrite has been sitting as a proposed order for almost five years. The FDA's own nonbinding forecast of monograph activities it intends to initiate over the following three years, posted in September 2025, does not list a sunscreen item.¹⁶

Which filters did the FDA say it could not vouch for?

This part gets quoted wrong constantly. In 2019 the FDA sorted the sixteen legacy ingredients into three buckets. Two were proposed as generally recognized as safe and effective at up to 25 percent: zinc oxide and titanium dioxide. Two were proposed as not safe and effective: PABA and trolamine salicylate. The remaining twelve were proposed as lacking sufficient data for a positive determination: cinoxate, dioxybenzone, ensulizole, homosalate, meradimate, octinoxate, octisalate, octocrylene, oxybenzone, padimate O, sulisobenzone and avobenzone.² The 2021 order carried the same three buckets forward.¹

The FDA anticipated exactly the misreading that followed. Its 2021 order says the proposal "does not represent a conclusion by FDA that the sunscreen active ingredients included in the 1999 Final Monograph, but proposed here as needing additional data, are unsafe for use in sunscreens." The agency said it wanted more information because conditions had changed, specifically because people now use far more sunscreen than they did when these ingredients were first evaluated.¹

A gap in a file is not a finding. Those are different things, and an accountant learns the difference early, because a missing schedule and an adverse finding get filed in very different drawers.

Does absorption into the bloodstream mean a filter is harmful?

No, and it is the claim in this subject most likely to be mangled.

The FDA ran two of its own maximal usage trials and published both in JAMA. The 2019 trial enrolled 24 healthy volunteers who had one of four commercial products applied at 2 milligrams per square centimeter to 75 percent of body surface area, four times a day for four days, with 30 blood draws over a week. Geometric mean maximum plasma concentrations of oxybenzone reached 209.6, 194.9 and 169.3 nanograms per milliliter in three of the products, and every ingredient tested crossed 0.5 nanograms per milliliter after the first day.⁹ The 2020 trial repeated the design with 48 participants and six ingredients: avobenzone, oxybenzone, octocrylene, homosalate, octisalate and octinoxate. All six exceeded the same threshold on day one after a single application, with oxybenzone reaching 258.1 nanograms per milliliter from a lotion.¹⁰

The 0.5 nanogram per milliliter figure is a paperwork trigger rather than a danger line, and that is the part that gets dropped. The FDA's own order explains that if a properly conducted maximal usage trial keeps steady-state blood levels under 0.5 nanograms per milliliter and the toxicology package shows nothing else concerning, the agency will generally waive the systemic carcinogenicity study and some of the developmental and reproductive toxicity work that would otherwise be required. The number was chosen as roughly the highest plasma level below which the carcinogenic risk of any unknown compound would be under one in 100,000 after a single dose, following the threshold of toxicological concern approach.¹ Crossing it does not say a compound is dangerous. It says the shortcut is off the table and the long-form studies are now owed.

Both JAMA papers end with the same sentence. The findings "do not indicate that individuals should refrain from the use of sunscreen."⁹¹⁰

The protective evidence is not in doubt in the way the absorption headlines implied. In the Nambour randomized trial in Queensland, adults assigned to daily sunscreen developed squamous cell carcinomas at a rate ratio of 0.61 against the comparison group, with a confidence interval of 0.46 to 0.81, though basal cell carcinoma was not reduced.¹² Ten years after the trial ended, follow-up of the same population found three invasive melanomas in the daily sunscreen group against eleven in the discretionary group, a hazard ratio of 0.27.¹³ Those are small absolute counts and the melanoma result should be read with that in mind, but the direction is consistent and it points one way.

Is American sunscreen actually weaker against ultraviolet A?

American labels barely describe UVA at all. Under the current monograph, "Broad Spectrum" is a pass or fail result on a single measurement called critical wavelength, and a product passes if its mean critical wavelength is 370 nanometers or greater.⁴ There is no number on the front of the bottle that tells a buyer how much UVA protection they are getting, only whether the product cleared one bar.

The FDA does not think that bar is doing enough work. Its 2021 order says the evidence linking UVA exposure to skin cancer has grown substantially, and that products can pass the current test while still having uneven protection across the spectrum. In the agency's own framing, a consumer using such a product may successfully avoid sunburn while accumulating a large UVA dose.¹ The proposed fix has three parts: require broad spectrum performance for every product labeled SPF 15 or above, add a requirement that broad spectrum products achieve a UVA I to total UV ratio of 0.7 or higher, and cap labeled SPF at 60+ while permitting formulation up to SPF 80.¹ None of that is in force. It is all still in the proposed order.

Comparative testing points the same direction, with real limits. One in vitro study measured twenty commercially available United States products against both standards. Nineteen of twenty met the American critical wavelength requirement. Eleven of twenty met the European expectation that the UVA protection factor be at least one third of the SPF.¹¹ Twenty products is a small sample, the testing was in vitro rather than on skin, and two of the five authors worked for UV filter manufacturers, one of them the company that later filed the bemotrizinol request. Read it as a signal, not a census.

Does Europe simply regulate these ingredients more loosely?

It does not, and this is where the folk version of the story falls apart. Europe has approved more filters and has also been harder on the old ones. The European Union's Scientific Committee on Consumer Safety reviewed benzophenone-3, which Americans know as oxybenzone, and concluded that use as a UV filter at up to 6 percent in body cream, propellant spray or pump spray is not safe for the consumer, while identifying 2.2 percent in those product types as a concentration it considers safe.¹⁴ The American monograph still permits oxybenzone at up to 6 percent.⁴

Europe has moved in both directions, adding filters and tightening limits on the ones it already allowed. The American list has done the first of those once, this June, and the second not at all.

Which filters are still missing from American shelves?

Eight ingredients entered the American queue through time and extent applications, some of them more than twenty years ago. In late 2014 and early 2015 the FDA issued proposed sunscreen orders on all eight and tentatively found in every case that the data were insufficient.⁸ The eight were bemotrizinol, bisoctrizole, drometrizole trisiloxane, octyl triazone, amiloxate, diethylhexyl butamido triazone, ecamsule and enzacamene. BASF sponsored the first three of those and L'Oreal sponsored drometrizole trisiloxane and ecamsule. For amiloxate and enzacamene, the FDA reported no contact from the sponsor since 2003. For diethylhexyl butamido triazone, none since 2010. Ecamsule was already sold in the United States, but only inside four products approved under new drug applications rather than through the monograph.⁸

Bemotrizinol has now moved, though not through that queue and not through that sponsor. The other seven have not moved. The FDA's own record describes these as ingredients with long marketing histories outside the United States, and the review of bemotrizinol notes that products containing it are sold on every continent except Antarctica.⁶

Who was supposed to pay for the safety data?

The regulatory coverage keeps circling this without landing on it. The FDA never refused these filters. It asked for a data package and waited. In its 2018 report to Congress it said so in flat language: the agency relies on industry to submit the data needed to support a determination, none of the additional data requested had been received, and the statute imposes no deadline on industry for submitting it.⁸

Follow the money on both sides of that sentence. On the agency side, the report disclosed roughly 8.4 million dollars of costs over a two-year period, about 56 employees touching the work and 12.76 person-years consumed, against a baseline where appropriations funded only 18 full-time equivalents for all over-the-counter monograph review across roughly 88 rulemakings. The report also notes that the FDA was not permitted to use user fee money for monograph work at all.⁸

On the industry side the arithmetic is worse, and it is the actual reason nothing moved. A monograph ingredient is not property. Once an ingredient is listed, any manufacturer whose product conforms to the monograph may use it. The sponsor pays for the maximal usage trial, the carcinogenicity study, the developmental and reproductive toxicity work and the years of regulatory time, and what it buys is an entry on a public list that its competitors may use the following morning. The category it is spending into is commodity priced sun care. I have written enough capital requests to recognize the shape: a project with a real cost, a long horizon, a public benefit and no mechanism by which the payer keeps the return is not a project anyone approves. That is a missing property right, not a scientific failure.

What finally moved bemotrizinol through the system?

Two things changed, and both came from the CARES Act rather than from any new science about sunscreen.

• Somebody funds the review. The act created a user fee program for over-the-counter monograph drugs, so a company can file an order request and pay for the agency time, with the Tier 1 request fee set at 559,777 dollars for fiscal year 2025.¹⁵
• Somebody keeps the upside. Section 505G of the Food, Drug, and Cosmetic Act gives the requestor 18 months of exclusivity following the effective date, during which only that company and its licensees, assignees or successors may market products using the change.⁴

With those two pieces in place, the timeline stops looking like a decades-long stall. DSM Nutritional Products submitted its request on September 23, 2024. The FDA issued a proposed order on December 12, 2025, took comments through January 26, 2026, and issued the final order on June 10, 2026, roughly seven months from proposal to final.⁷⁴

The data behaved, which is why it was a good test case. The sponsor's maximal usage trial found bemotrizinol is not readily absorbed even with repeat application, with most measurements below the 0.1 nanogram per milliliter limit of quantitation and overall mean absorption consistently under 0.5 nanograms per milliliter, so the carcinogenicity and reproductive toxicity studies that absorption would have triggered were not required.⁶ The FDA also noted something the press releases did not. The clinical program was small by ordinary drug standards, with 484 subjects across the usage trial and three dermal safety studies plus 47 more in the SPF trials, and the agency accepted it partly because of the extensive postmarket history in other countries.⁶ That is a reasonable judgment. It is also a reminder that the American file was thin precisely because nobody had been paying to fill it.

What does any of this change at the drugstore?

In the short run, very little. The order took effect on August 9, 2026, and a permitted ingredient is not a product. Formulations have to be built, SPF tested, labeled and manufactured, and for 18 months the marketing right runs to one company and whoever it licenses.⁴ Everything on the shelf today is still built from the same sixteen ingredients, still labeled with a pass or fail broad spectrum mark that carries no UVA number, and still governed by a rulebook written in 1999.

None of that is a reason to skip sunscreen, and the evidence does not support treating it as one. The randomized trial evidence that regular use reduces squamous cell carcinoma, and the follow-up evidence pointing to reduced invasive melanoma, both stand.¹²¹³ The FDA has not found the current filters unsafe, has said so explicitly, and has kept them legal throughout.¹ The argument here is that American buyers are entitled to better options and better labels, not that the current options are hazardous.

What would actually unstick the rest of the list?

Two separate levers, and confusing them is why this looks unsolvable.

The first lever is per ingredient and it belongs to industry. For each of the seven filters still waiting, someone has to file a request, pay the fee, fund the studies and take the 18 month window as their return. Bemotrizinol proved that the path works when a sponsor with an interest walks it. It proved nothing about whether 18 months of exclusivity is worth the cost of a full data package for the next ingredient down the list, and that calculation, not any scientific dispute, is what determines who goes next.

The second lever is the rewrite, and it belongs to the FDA alone. Finalizing OTC000008 would resolve the status of the twelve legacy ingredients, remove PABA and trolamine salicylate, require broad spectrum protection at every SPF 15 and above, and add a real UVA I requirement to the broad spectrum test.¹ No sponsor has to fund that. No exclusivity window is needed. It has been pending since September 2021, and the agency's own published forecast gives no date.¹⁶

One filter got through because the law finally let someone own the result for a year and a half. The rest of the list is waiting on the same arithmetic, and the rewrite is waiting on nothing but a signature.