TL;DR
The most-used anti-aging ingredient on earth was not designed to fight wrinkles. Retinoids reached the clinic as an acne treatment in the late 1960s, and the wrinkle benefit was a side effect that patients reported before researchers had a name for it. The discovery chain runs from vitamin A deficiency studies in the 1910s, through Albert Kligman’s lab in Philadelphia, to the FDA approval of tretinoin for photoaging in 1995.
If you ask a dermatologist which topical ingredient has the most evidence behind it, you will get retinoids most of the time. If you ask how that came to be, the story is less tidy than the marketing copy makes it look. Retinoids did not arrive on dermatology shelves through a planned anti-aging program. They arrived as an acne drug that, over a couple of decades of clinical use, kept doing things nobody had asked them to do.
I find that origin story matters, because it explains why retinoids still feel slightly mismatched with the products they live inside. They are a pharmacology success that got rebranded as a beauty ingredient. The friction in modern routines, the irritation, the slow ramp, the constant rephrasing of expectations, all of it traces back to that mismatch.
The vitamin A foundation
The chemistry behind retinoids starts with vitamin A. In the 1910s and 1920s, researchers including Elmer McCollum at Wisconsin established that animals deprived of a fat-soluble factor in their diet developed eye problems, growth failure, and characteristic skin changes (rough, scaly, follicular plugging). The factor was named vitamin A. By the mid-1930s, the structure was worked out, and synthetic vitamin A became available for nutritional research.
The skin observation mattered. Vitamin A deficiency produced something that looked, at the microscopic level, like a problem in how skin cells matured and shed. That observation sat in the literature for decades before anyone connected it to a topical treatment.
The acne accident
The clinical retinoid story begins with Albert Kligman, a dermatologist at the University of Pennsylvania, and his collaborator James Fulton. In the late 1960s the two were working on acne, and the working hypothesis was that abnormal follicular keratinization (the same pattern seen in vitamin A deficiency) was part of the disease. They reasoned that a topical retinoid might normalize follicular shedding and reduce comedone formation.
The first published clinical trial of tretinoin (all-trans retinoic acid) for acne appeared in JAMA in 1969. The result was substantial improvement in comedonal acne, with the now-familiar side effect profile (redness, peeling, sun sensitivity). The FDA approved tretinoin for acne in 1971 under the brand name Retin-A.
That was supposed to be the end of the story. It was the beginning.
The patient observation that built an industry
By the late 1970s, dermatologists prescribing Retin-A for acne in older adults started hearing the same comment from patients. The acne improved, but something else did too. Fine lines looked softer. Pigment patches faded. The skin looked, in the words of one of Kligman’s case notes, “better in ways the patient could not quite name.”
Kligman, who was already an unconventional thinker, took those reports seriously. Through the 1980s his lab and collaborators ran controlled trials specifically on photoaging endpoints. The pivotal study was a 1988 paper in JAMA by Weiss, Ellis, Headington, and Voorhees at Michigan, which showed that daily tretinoin produced measurable improvement in photodamaged skin over four months, including reductions in fine wrinkles and pigmentary mottling. That paper is the moment retinoids became an anti-aging story.
The FDA approved tretinoin specifically for photoaging in 1995, under the brand name Renova. By then the cosmetic industry had spent years circling the molecule, and the over-the-counter retinol market that followed was a direct response to a prescription drug that consumers wanted access to.
Where retinol came in
Retinol (vitamin A alcohol, the parent molecule) entered cosmetic formulations in the 1990s as a non-prescription workaround. The skin converts retinol to retinoic acid through a two-step enzymatic pathway, so a retinol product is, in effect, a slow and milder version of tretinoin. The conversion is inefficient, which is why retinol products usually need higher percentages than the prescription equivalent to produce comparable effects, and why the timelines are longer.
The same logic applies to the newer ester forms (retinyl palmitate, retinyl acetate) and to retinaldehyde, which sits one step closer to retinoic acid in the conversion chain. The cosmetic shelf is, in a sense, a graded ladder of how much enzymatic work the skin has to do before the active form arrives in the cell.
The contrarian H2: the science is older than the marketing makes it sound
Beauty marketing tends to present retinol as cutting-edge skincare innovation. That framing has always sat badly with me. Retinoic acid as a topical drug has been in clinical use for more than half a century. The 1988 Michigan study is older than most readers’ current retinol product. The pharmacology is mature; the formulations have improved; the basic story has not changed since the early 1990s.
The reason this matters is that newness implies provisional evidence, and retinoids are not provisional. They are one of the few skincare actives where the evidence base is decades deep and the mechanism is mapped down to the nuclear receptor level. The cosmetic industry’s tendency to package retinol as exciting and recent obscures the fact that the science underneath is unusually settled. The relevant question is which form, at which percentage, in which vehicle, not whether the category works.
The real numbers
The 1988 Weiss study in JAMA enrolled 30 patients with photodamaged facial and forearm skin and randomized them to tretinoin 0.1 percent cream or vehicle once daily for 16 weeks. Tretinoin produced statistically significant improvement on every measured endpoint, including a 50 percent or greater reduction in fine wrinkling in 14 of the 15 treated patients versus none in the vehicle group.
A 2006 meta-analysis published in the Archives of Dermatology by Samuel and colleagues synthesized 47 trials of topical retinoids for photoaging and concluded that the effect size for tretinoin on fine wrinkles was moderate but consistent, with effects visible from 12 weeks and continuing to develop over 6 to 12 months. The NIH’s PubChem and clinical pharmacology references put tretinoin’s binding affinity to retinoic acid receptors (RAR alpha, beta, and gamma) at the level that explains the broad transcriptional changes the drug produces.
What this means for your routine
A few practical implications fall out of the history. Retinoids work; the evidence is unusually deep. The ramp is real, because the pharmacology is real, and short-term irritation is the cost of the active form reaching its receptor at meaningful concentrations.
Prescription tretinoin remains the reference standard for both acne and photoaging. Cosmetic retinol products work too, more slowly and at higher applied percentages, because of the conversion step. Retinaldehyde is a reasonable middle option for sensitive skin. The ester forms (retinyl palmitate especially) are gentler still and produce smaller effects on a longer timeline.
For the broader thinking on how retinoids fit into a slow routine, see the retinol versus retinaldehyde primer, the retinol beginner guide, and the anti-aging actives explainer.
FAQ
Who actually discovered retinoids for skin? Albert Kligman and James Fulton at the University of Pennsylvania, working on acne in the late 1960s. The 1969 JAMA paper is the foundational clinical citation. The photoaging discovery came later, through Kligman’s continued work and the 1988 Michigan trial.
Was retinoic acid ever sold as a wrinkle drug from the start? No. The first FDA approval (1971) was for acne. The photoaging indication came in 1995, more than two decades later, after patient reports drove the formal studies.
Why is retinol so much weaker than tretinoin? Because retinol has to be enzymatically converted twice (to retinaldehyde, then to retinoic acid) before it can bind the receptor. Most of the applied dose never completes the conversion.
Is the newer plant-based bakuchiol a real retinoid alternative? Bakuchiol has some retinoid-like transcriptional effects in vitro and modest evidence in small clinical trials. It is gentler and worth considering for sensitive skin, but the evidence base is a fraction of retinoic acid’s.
Why does my retinol still sting after years of use? Receptor activation has a built-in inflammatory component that does not fully adapt. Most users stabilize at a low chronic level of mild sensitivity rather than reaching zero.
Tag hub: More on retinoids, retinol, and how to use them
Sources
Kligman AM, Fulton JE, Plewig G. Topical vitamin A acid in acne vulgaris. Archives of Dermatology 1969. Weiss JS, Ellis CN, Headington JT, Voorhees JJ. Topical tretinoin in the treatment of aging skin. JAMA 1988. Samuel M et al. Interventions for photodamaged skin. Archives of Dermatology 2006. FDA approval records for Retin-A (1971) and Renova (1995).