July 23, 2019
The Food and Drug Administration almost never tests products itself. But in May, the Journal of the American Medical Association published the results of a randomized trial, conducted by F.D.A. researchers, to determine whether the chemicals in four commercially available sunscreens are absorbed through the skin into the bloodstream. Four times daily, subjects were coated in one of the formulas in an amount determined to be the maximum a person might use: two milligrams per square centimeter of skin over 75 percent of the body. Later, blood samples were drawn and analyzed. All of the sunscreen chemicals were detected in concentrations that exceeded an F.D.A. threshold past which manufacturers are required to do further toxicology tests. “People who use sunscreens very reasonably presume they have been tested and are safe and effective,” says Kanade Shinkai, a dermatologist at the University of California, San Francisco, and an author of an editorial accompanying the JAMA study. “And we don’t really have that evidence.”
Legally, the U.S. regards sunscreen as a drug, meaning a substance “intended for use in the diagnosis, cure, mitigation, treatment or prevention of disease” — in this case, sunburn and skin cancer — and/or one that affects “the structure or any function of the body.” Until 1962, drugs could be sold in the U.S. without any data to support claims of their efficacy. But that year, reports that a sedative called thalidomide had caused severe birth defects in thousands of babies in Western Europe led to the Kefauver-Harris Amendment to the Federal Food, Drug and Cosmetic Act, which requires drug makers to satisfy the F.D.A. that their products are safe and effective before they go on sale.
But more than 100,000 over-the-counter drug products were already on the market, including sunscreens, each of which, under the new law, needed review. To streamline the process, in 1972 the F.D.A. sorted them into therapeutic categories (antacids, for example) as assigned “monographs,” which included lists of active ingredients. If publicly available data demonstrated that these ingredients were generally safe and effective, they could be used in current and future products under conditions specified in the monograph without further review.
Almost 50 years later, a third of the monographs, including the one for sunscreens, have not been finalized; hundreds of over-the-counter drugs currently on sale have not yet been determined to be safe and effective. (The agency blames the delay on an antiquated system and a lack of funding.) Sunscreens are unique, however, in that the way we use them has changed significantly. Decades ago, the sunscreens that Americans were dabbing on their noses were often mineral concoctions, like zinc oxide and titanium dioxide, that sat on the skin in a thick white cream and physically blocked the sun’s rays. But as awareness grew that ultraviolet rays can cause skin cancer even without burning the skin, public health experts began to advise that people wear sunscreen daily on all exposed areas of the body. (Hats, long-sleeves and avoiding prolonged sun exposure are also recommended.) This substantially increased the usage of less messy, chemical sunscreens — which contain molecules, or “filters,” that can absorb a “broad spectrum” of ultraviolet light — including by pregnant women and children as young as 6 months.
Initially, it was assumed these chemicals, like mineral sunscreens, stayed on the surface of the skin. Then, in 1997, a study published in The Lancet demonstrated that after subjects applied sunscreen, the UV filter oxybenzone was present in their urine; in 2008, a national health survey by the Centers for Disease Control and Prevention found oxybenzone in 97 percent of urine samples; a 2010 study of nursing mothers in a Swiss hospital reported that 85 percent had UV filters in their breast milk. (Because UV filters are in everything from makeup to shampoo to patio furniture, sunscreen was probably not the only source.) This widespread use has raised environmental concerns: Hawaii and Key West, Fla., have recently banned sunscreen ingredients, including oxybenzone, that studies have suggested may be damaging coral reefs. Yet despite sunscreen’s extensive use over decades, there has never been any indication that sunscreen chemicals are harmful to humans.
While drugs — like other products and health recommendations — can be tested for years in clinical trials of hundreds, or even thousands, of people, this doesn’t always predict how they will affect millions of people after decades of use. In the general population, dangerous side effects can remain invisible in the absence of large, long-term studies. In 2002, some six million women were using hormone-replacement drugs to relieve menopause symptoms — one chemical was substituted for another, innocuously, it seemed — when a large federal study showed that after five years, the drugs increased the risk of breast cancer, heart attack and blood clots.
So, the lack of negative evidence alone doesn’t prove that sunscreens are safe. Animal studies have raised the possibility that some UV filters, including oxybenzone, may disrupt the endocrine system, which can adversely affect reproduction, development and immunity. Even if those filters only slightly elevated the same risks in people, which no evidence suggests they do, says Dr. Robert Califf, a former F.D.A. Commissioner and professor at the Duke University School of Medicine and the other author of the JAMA editorial, “small differences in a big population can be very important, and it would be very hard to see it. People have trouble getting pregnant, men are sterile — these things happen every day, so you don’t think back, Gee, it must be the sunscreen. And it’s probably not.”
But population-wide studies of sunscreens are especially difficult to perform because so many variables are involved. In the U.S., 14 active ingredients in the sunscreen monograph are available to be combined any number of ways. People use many formulas, apply them in varying amounts using different methods (e.g., sprays and lotions) and engage in a wide range of activities while wearing them. This is partly why, Shinkai says, basic research is lacking: “We don’t actually know what the proper dose is to prevent skin cancer and whether that is different for different agents or even different combinations of agents that are used in sunscreen.” (The American Academy of Dermatologists recommends one ounce for most people.) More information could be lifesaving: For instance, diagnoses of melanoma, the most lethal type of skin cancer, are increasing. And though there is strong evidence that sunscreen use prevents skin cancer, experts disagree whether the available data shows that current sunscreen formulas and application methods protect against melanoma specifically. There might be better practices we aren’t aware of.
There may also be more effective sunscreens unavailable to Americans. In Europe, newer formulations that are broader-spectrum have been in use for years, but the F.D.A. won’t consider them without more data, leaving the U.S. marketplace with 1970s-era ingredients that, except for mineral varieties, the F.D.A. has not deemed safe and effective, either, but that it also can’t remove without eliminating a valuable form of skin-cancer protection. “The public needs to know, Should I continue to use sunscreen?” says Henry Lim, former president of the American Academy of Dermatology and a dermatologist at Henry Ford Hospital in Detroit. “And that is becoming a challenge.”
The F.D.A. is maintaining its official position that Americans should keep using sunscreen. But by showing that chemicals are being absorbed, the agency has effectively forced manufacturers to provide additional data, which they argue is unreasonably exhaustive, by November or risk having their products pulled off U.S. shelves. (The F.D.A. will likely grant them an extension.)
All of which raises the question of how deeply the unknown risks of a medically beneficial drug can and should be explored. “Science doesn’t hold still, and we keep learning things,” Dr. Janet Woodcock, director of the F.D.A. Center for Drug Evaluation and Research, points out. That is, all of us are taking part in toxicology studies, whether we like it — or know it — or not.