I expanded on the research, including Dr. Greger’s emphasis on the VATTC study risks, and got the following takeaways:
Tretinoin is, by a wide margin, the best-evidenced topical drug in dermatology for both acne and photoaging. The Griffiths NEJM 1993 collagen data, replicated in subsequent 2-year and 22-month studies, are real and clinically meaningful. For melasma in fixed triple combination, for acne, and as an adjunct to minoxidil in androgenetic alopecia, the evidence is genuinely strong. For striae rubrae, post-inflammatory hyperpigmentation, and pre-procedure priming, the evidence is moderate. For keratosis pilaris, warts, scarring, and most of the other off-label uses you’ll see on social media, evidence is thin and largely empirical. For chemoprevention of skin cancer, the largest RCT was negative.
If you’re evaluating it for yourself, four honest framing points:
1. It is a real drug with a real side-effect profile, not a cosmetic. Perhaps 1 in 7 users will quit because of irritation. Tolerability protocols (low concentration, infrequent dosing, dry skin, buffering, short-contact therapy) can significantly reduce that.
2. OTC retinaldehyde or adapalene 0.1% gel is a reasonable starting point if you want to test whether your skin tolerates retinoids before committing to a prescription, or if you want most of the benefit with fewer side effects. The Creidi 1998 head-to-head suggests retinaldehyde 0.05% can deliver tretinoin-comparable wrinkle reduction with about a third of the irritation.
3. The VATTC mortality finding deserves acknowledgment, not dismissal. Mainstream consensus is that it’s likely a statistical artifact in an elderly, comorbid, high-smoking population on extreme dosing — but it has not been mechanistically explained, and “probably nothing” is not the same as “definitely nothing.” For typical adult facial use at 0.025–0.05%, the absolute risk implied is very small. For long-term, large-surface, high-strength use in elderly smokers, the risk-benefit conversation should be explicit.
4. Visible benefit takes 3–6 months, peak benefit 12 months, and effects regress within months of stopping. This is a long-term commitment.
Tolerability and How to Minimize Irritation
About 85% of tretinoin users develop “retinoid dermatitis” (erythema, scaling, stinging, dryness) at some point; ~20% experience moderate-to-severe irritation, and ~15% discontinue treatment entirely for this reason. The phenomenon is called retinization and reflects compromised stratum corneum barrier function while the epidermis adapts. It is usually worst in weeks 2–6 and improves thereafter.
Evidence-based strategies to reduce irritation:
1. Start low and slow. Begin with 0.025% cream (cream > gel for tolerability) two or three nights per week, then increase frequency over 6–12 weeks as tolerated.
2. Apply to completely dry skin. Wait 15–20 minutes after washing — applying to damp skin meaningfully accelerates penetration and irritation.
3. Use a pea-sized amount for the whole face. More is not better.
4. Buffer with moisturizer. Either apply a bland moisturizer first (“buffering”), or moisturize ~20 minutes after the tretinoin (“sandwich method”). A Draelos Cutis 2006 RCT showed pre-treating with a barrier-supporting moisturizer containing niacinamide/panthenol/tocopheryl acetate significantly reduced retinization-related irritation without diminishing efficacy.
5. Short-contact therapy. Veraldi et al. 2013 showed that applying 0.05% tretinoin once daily for only 30 minutes and then washing it off achieved acne efficacy comparable to overnight use, with markedly less irritation (only 5.4% discontinuation for irritation versus the typical 15%).
6. Avoid same-night use of other actives that compromise the barrier — alpha-/beta-hydroxy acids, benzoyl peroxide (which also chemically degrades tretinoin if the formulation isn’t a stable microsphere), strong physical exfoliation. Niacinamide, hyaluronic acid, and ceramide products are compatible.
7. Rigorous daily SPF. Tretinoin thins the stratum corneum and is itself UV-degraded; sunscreen is non-negotiable.
Foundational Mechanism & Photoaging
Griffiths CE et al. “Restoration of collagen formation in photodamaged human skin by tretinoin.” N Engl J Med. 1993;329(8):530–535.
The landmark RCT behind the 80% collagen increase finding.
Restoration of collagen formation in photodamaged human skin by tretinoin (retinoic acid) - PubMed | Full text: https://www.nejm.org/doi/full/10.1056/NEJM199308193290803
Darlenski R et al. “Topical retinoids in the management of photodamaged skin: from theory to evidence-based practical approach.” Br J Dermatol. 2010;163(6):1157–1165.
Broad mechanism and clinical evidence review.
Novel approach to gene expression profiling in Sézary syndrome - PubMed
Systematic review of tretinoin RCTs for photoaging (Yin et al., Int Wound J. 2022)
Reviews 23 RCTs confirming efficacy through 24 months.
Topical tretinoin for treating photoaging: A systematic review of randomized controlled trials - PMC
The VATTC Mortality Signal — Most Important Safety Reading
Weinstock MA et al. “Topical tretinoin therapy and all-cause mortality.” Arch Dermatol. 2009;145(1):18–24.
The trial that was stopped early. Read the actual conclusions — the authors themselves call causation unlikely.
Topical tretinoin therapy and all-cause mortality - PubMed | ResearchGate PDF: https://www.researchgate.net/publication/23801055
Schilling LM, Dellavalle RP. “Dealing with unanticipated mortality in a large randomized clinical trial of topical tretinoin.” Arch Dermatol. 2009;145(1):76.
The editorial response calling for transparency rather than alarm.
Dealing with unanticipated mortality in a large randomized clinical trial of topical tretinoin - PubMed
Sangolli commentary — “Does topical tretinoin used for chemoprevention cause increased mortality?” J Cutan Aesthet Surg. 2009;2(2):101–102.
The most accessible plain-language analysis of why causation is implausible.
Does Topical Tretinoin Used for Chemoprevention Cause Increased Mortality? - PMC (free full text)
