AHAs vs Retinoids: They Are Not the Same Thing

CLINICAL · 17 June 2026

TL;DR

AHAs and retinoids are both described as anti-ageing actives. They are not interchangeable. They work through different mechanisms, at different skin depths, and with different timelines — and understanding the distinction is what determines how to use them effectively.

Alpha hydroxy acids: surface exfoliants

AHAs — including glycolic acid, lactic acid, mandelic acid, and malic acid — are water-soluble acids that work primarily in the upper layers of the epidermis. Their mechanism is keratolytic: they loosen the intercellular bonds that hold dead corneocytes together at the skin surface, accelerating their shedding. The result is faster surface cell turnover, improved texture, brighter appearance, and better penetration of subsequent actives.

AHA efficacy is pH-dependent, typically requiring a formulation pH of approximately 3.5–4 for meaningful exfoliation. Glycolic acid, with the smallest molecular weight, penetrates most readily; mandelic acid penetrates most slowly and tends to be gentler; lactic acid is intermediate and also functions as a humectant.

The effects of OTC-strength AHAs are predominantly epidermal. Higher-strength glycolic acid formulations can stimulate fibroblast activity and dermal glycosaminoglycan production, but the evidence for dermal remodelling is considerably weaker than that for retinoids and should not be overstated.

Retinoids: nuclear receptor actives

Retinoids — the family of vitamin A derivatives including retinol, retinaldehyde, adapalene, and prescription tretinoin — work through a fundamentally different mechanism. They bind to nuclear retinoic acid receptors (RARs) and retinoid X receptors (RXRs), altering gene transcription in keratinocytes and fibroblasts. The result is a cascade of changes: increased collagen synthesis, suppression of MMP activity, accelerated epidermal cell turnover, and normalisation of disordered keratinisation.

This receptor-mediated mechanism is what makes retinoids genuinely different from surface exfoliants. They produce structural change in the dermis. Prescription tretinoin remains the most evidence-based topical ingredient for treating photoageing because of these effects.

Over-the-counter retinol requires conversion to retinoic acid in the skin (retinol → retinaldehyde → retinoic acid), which reduces potency but also reduces irritation. Adapalene — now available without prescription in the United States and a small number of other markets — is selectively active at RARβ and RARγ subtypes, is photostable, and is generally less irritating than tretinoin. It is increasingly used as an entry retinoid in modern dermatology practice. Prescription tretinoin is already in active form and produces the strongest effects, with corresponding greater irritation potential during initiation.

Retinoid dermatitis — dryness, peeling, and redness during the early weeks of use — is common and reflects genuine barrier perturbation during the adjustment phase. It is not a sign of damage but a predictable response to accelerated cell turnover.

Using both: irritation, not chemistry

The reason to be thoughtful about using AHAs and retinoids together is irritation, not chemical incompatibility. The idea that AHAs “neutralise” retinoids through pH is considerably weaker than the clinical evidence suggests, and it is possible to combine glycolic acid with tretinoin or adapalene effectively once skin tolerance is established. Modern retinoid formulations are more stable than older preparations.

The practical issue is that using both together compounds barrier perturbation and increases the risk of irritation, dryness, and sensitivity — particularly when starting out. Alternating nights is a sensible strategy to reduce this risk, especially for beginners and people with sensitive skin, but it is not a strict rule. As tolerance builds, some people use both in the same routine without difficulty.

Retinoids are traditionally used at night, partly by convention and partly because older formulations degraded in UV. Photostability and photosensitivity are two different things worth separating here. Adapalene is chemically photostable and has no clinical evidence of photosensitising the skin, making it the most straightforward candidate for daytime use. Tazarotene is also chemically photostable, but at least one clinical study found that pretreatment with tazarotene lowered the skin’s threshold for UVB-induced erythema — meaning it can increase the skin’s own sensitivity to sunlight even though the molecule itself doesn’t degrade in light. Tretinoin’s UV sensitivity is most pronounced in older, non-micronized formulations specifically; morning use is not a universal contraindication, particularly with newer micronized formulations, but it isn’t risk-free either. AHAs in the morning increase photosensitivity and are generally better suited to evening use. Daily broad-spectrum sunscreen is advisable regardless of which active is used or when it’s applied.

On HRT

There is no established evidence that oestrogen itself increases sensitivity to retinoids. What is clinically recognised — and directly documented in trans women, not just extrapolated from cisgender populations — is that HRT often reduces sebum production, and some people experience increased dryness during early transition. These changes can make retinoid initiation less comfortable than it might otherwise be. Introducing retinoids gradually in this context — starting with lower concentrations or less frequent application — tends to improve tolerability. This is consistent with standard dermatological advice for anyone new to retinoids, applied with appropriate attention to the changing skin environment of early HRT.

⚠️ Clinical note: Alternating AHAs and retinoids is a practical strategy to reduce irritation, not a chemical rule. Many people tolerate both in the same routine once skin is adapted. The HRT-specific guidance here is based on clinical observation of reduced sebum and dryness directly documented in transgender women, rather than purely extrapolated from cisgender data.


Sources

More Clinical

View all →