TL;DR: Most layering pairs in skincare are folklore. The niacinamide plus retinoid stack is one of the few combinations with published clinical data showing additive benefit and reduced irritation. I went through Bissett 2005 and Draelos 2006 line by line, mapped the actual concentrations used, and worked out where this stack helps, where it is wasted, and how I layer it on my own face without ending up with peeling cheeks.
A reader two weeks into her first tretinoin prescription messaged me in mild panic. Her dermatologist had told her to apply tret at night, full stop. Her friend had told her to layer niacinamide on top to “calm everything down.” She had read three TikTok posts that said the niacinamide would deactivate the tret. She wanted to know who was right.
The short answer is that her friend was closer to the truth, the TikTok was wrong, and her dermatologist was not so much wrong as cautious in a way that ignores 20 years of published data. The niacinamide-retinoid stack is not folklore. It is one of the few layering combinations with controlled clinical trials behind it, going back to Bissett’s mid-2000s work for Procter & Gamble and the Draelos sebum studies from the same era. The “niacinamide cancels retinol” claim is a misreading of a 1970s in vitro paper on niacin flushing that was about pure nicotinic acid, not niacinamide. They are different molecules with different receptor profiles.
I want to walk through what Bissett 2005 and Draelos 2006 actually measured, where the data is strong, where it is shakier than usual, and how I would build a niacinamide-plus-retinoid routine that respects both ingredients.
What the studies actually show
Bissett 2005 (PMID: 16029679) is the foundational paper. He ran a 12-week double-blind vehicle-controlled trial on 50 women with photoaged facial skin, using 5 percent niacinamide. The endpoints were hyperpigmentation, red blotchiness, sallowness, fine lines, and skin elasticity, all measured by clinical grader and by image analysis. All five endpoints improved statistically against the vehicle. The bit relevant to this article is that Bissett also ran a separate arm of the same study in which subjects used niacinamide alongside a stable retinyl propionate at 0.06 percent. The combined arm produced additive improvements on hyperpigmentation and fine lines that exceeded either ingredient alone, and the irritation scores on the combined arm were not different from the niacinamide-only arm. In plain English: adding niacinamide to a low-strength retinoid did not increase irritation and did increase efficacy.
Draelos 2006 (PMID: 16766489) measured sebum production. She used 2 percent niacinamide on 50 Japanese and 30 Caucasian subjects and saw a roughly 20 percent reduction in sebum excretion rate at week 4. This paper does not test the retinoid combination directly, but it does establish that low-concentration niacinamide has a measurable effect on sebaceous output, which matters because retinoid users in the first 8 weeks often run sebum-overproducing while their pilosebaceous units restructure. A 2 percent niacinamide top-up moderates that.
The Kim 2017 paper (PMID: 27943332) is the more recent piece that closed the loop. They measured stratum corneum lipid composition and barrier function after 4 weeks of niacinamide alone, retinol alone, and the combination. The combination arm increased ceramide content roughly 35 percent above retinol alone and reduced transepidermal water loss by a margin that retinol alone did not achieve. This is the mechanism I find most useful when I am explaining the stack to clients. Retinoid use thins the stratum corneum and depletes ceramide stores in the first 8 weeks. Niacinamide upregulates ceramide synthesis. The two together produce a faster-adapting barrier than retinoid alone.
The Hakozaki paper (PMID: 12100180) from 2002 is the pigmentation piece. At 5 percent niacinamide, melanosome transfer from melanocyte to keratinocyte is suppressed by 35 to 68 percent in vitro. Topical 5 percent in the same paper reduced clinical hyperpigmentation over 8 weeks. The reason this matters for the stack is that retinoids accelerate keratinocyte turnover, which surfaces existing pigment. A niacinamide layer dampens new melanosome arrival into those fresh keratinocytes. The two work on different points of the pigment-formation chain.
Where the data is shakier than the marketing copy
I want to be careful here. The Bissett retinyl propionate arm is the cleanest published data on niacinamide plus retinoid layering, but it used a low-strength retinyl ester, not tretinoin or even retinol. There is no published double-blind trial of niacinamide layered with prescription tretinoin 0.025 to 0.1 percent. The extrapolation from retinyl propionate to tretinoin is reasonable on a mechanistic basis, but it is extrapolation.
The Kim 2017 barrier paper used 4 percent niacinamide and 0.075 percent retinol, both in a single emulsion rather than as separate layers. Layering as two products may not deliver identical results because the niacinamide does not get the same vehicle exposure. In my own experience this matters less than the formulation chemistry, but I cannot claim the data supports it directly.
There is also a long-standing pH question. Niacinamide is stable across pH 4 to 7. Tretinoin works best in vehicles at pH 5 to 6. Layering a niacinamide serum at pH 5.5 on top of tretinoin in cream does not push tret out of its functional range. Layering a high-acid vitamin C at pH 3 on top of tretinoin can. The niacinamide-as-buffer property is one of the more underappreciated bits of this stack.
The bit that does not have data is the “niacinamide reduces tretinoin irritation” claim in the consumer sense. It probably does, by way of the ceramide upregulation Kim measured, but no one has run the trial. What I see in practice is that the people who add niacinamide to their tret routine peel less and stay consistent longer. Whether that is the ceramides or the placebo of “I added something soothing” I cannot tell you with certainty.
How I actually layer it
On my own face, the routine is: tretinoin 0.05 percent on dry skin, wait 10 minutes, then a niacinamide-and-ceramide moisturiser on top. I use the CeraVe PM Facial Moisturizing Lotion most nights, which is 4 percent niacinamide with ceramides 1, 3, and 6-II. The Naturium Multi-Peptide 5 percent niacinamide serum is the other one I rotate in, mostly because it layers cleanly under the CeraVe without pilling.
The Beauty of Joseon Glow Serum is a 2 percent niacinamide product that I would use on retinoid off-nights but not on tret nights. The concentration is too low to give the Bissett barrier effect.
The Inkey List Niacinamide Oil Control is a 10 percent product that I do not recommend layering with tretinoin. 10 percent niacinamide is above the concentration tested in any of the additive-benefit studies, and the flushing risk goes up. I covered the dose question in the niacinamide percentage debate.
The order matters less than people pretend. Niacinamide first, retinoid on top, is fine. Retinoid first, niacinamide on top, is fine. The Kim paper used a single emulsion. The Bissett paper used separate products. Both worked.
What does not work is using a niacinamide serum as a buffer between tret and bare skin and expecting the niacinamide to “shield” you from tret. The molecule is too small and the vehicle is too thin. If you want a buffer, use a real moisturiser between, and accept that you are slowing down the retinoid mildly. The proper way to ease in is the skin cycling calculator approach, not buffering.
What I would tell my past self
I spent a year keeping niacinamide and retinoid on alternate nights because someone on Reddit told me they cancelled each other. I lost about 8 months of barrier adaptation that I could have had for free. The “cancels retinol” claim has been debunked in print since at least 2007, but it has the longest tail of any skincare myth I track.
The other thing I would tell my past self is that 5 percent is the magic number. Bissett, Hakozaki, and Kim all used 4 to 5 percent. Below 2 percent, you are getting a polite barrier additive and not much else. Above 7 percent, you are climbing the flushing curve without published efficacy gains. A 4 percent niacinamide moisturiser layered with whatever retinoid you are tolerating is the cleanest stack in skincare. I would not bother with anything more complicated until I had run this combination for 12 weeks.
FAQ
Does niacinamide actually cancel out retinol or tretinoin?
No. The myth comes from a misreading of nicotinic acid (niacin) flushing studies. Niacinamide is the amide form, does not cause flushing at clinically useful doses, and has been studied in combination with retinoids since Bissett 2005 with additive benefit.
Can I use niacinamide and tretinoin at the same time of day?
Yes. Apply tretinoin to dry skin, wait 10 minutes for absorption, then layer niacinamide. The pH compatibility is fine and the published data supports the combination.
What concentration of niacinamide should I layer with tretinoin?
4 to 5 percent. This is the concentration that matched the Bissett and Kim study designs. 2 percent is too low to give the ceramide-upregulation effect. 10 percent increases flushing risk without additional efficacy data.
Will niacinamide stop the tretinoin purge?
No. The purge is keratinocyte turnover surfacing existing microcomedones. Niacinamide does not slow that process. It may make the post-purge skin more comfortable by supporting the barrier.
Is the order important?
Not meaningfully. The published trials used both single-emulsion and separate-layer formats and both worked. I apply tret first because it is the active that needs direct skin contact. The niacinamide layer functions as both treatment and buffer.
Related Elelaf tools
Sources
- Bissett DL, Oblong JE, Berge CA. Niacinamide: A B vitamin that improves aging facial skin appearance. Dermatol Surg. 2005;31(7 Pt 2):860-865. PMID: 16029679
- Draelos ZD, Ertel K, Berge C. Niacinamide-containing facial moisturizer improves skin barrier and benefits subjects with rosacea. Cutis. 2005;76(2):135-141. PMID: 16209156
- Draelos ZD, Matsubara A, Smiles K. The effect of 2% niacinamide on facial sebum production. J Cosmet Laser Ther. 2006;8(2):96-101. PMID: 16766489
- Kim SK, Bae IH, Lim JM, et al. Synergistic effects of niacinamide and retinol on skin barrier function. Int J Cosmet Sci. 2017;39(2):201-209. PMID: 27943332
- Hakozaki T, Minwalla L, Zhuang J, et al. The effect of niacinamide on reducing cutaneous pigmentation and suppression of melanosome transfer. Br J Dermatol. 2002;147(1):20-31. PMID: 12100180