Mills and colleagues published a paper in 1986 (Br J Dermatol) demonstrating that benzoyl peroxide at 2.5% concentration produced clinical efficacy indistinguishable from 10% concentration in acne treatment. Their work measured comedone reduction, inflammatory lesion count, and patient-rated outcomes at 4, 8, and 12 weeks. The 2.5% and 10% groups landed within statistical noise of each other.
This is one of the most important papers in OTC acne pharmacology, and one of the least-cited. Forty years later, the dominant US retail benzoyl peroxide formulations are 5% and 10%. The 2.5% formulations are a minority of shelf space. The marketing logic favors higher numbers; the clinical logic, particularly for Fitzpatrick IV-VI skin, favors lower.
I’ll walk through the paradox at the core of benzoyl peroxide in SoC acne care, the concentration data, the SoC-specific protocols, and the alternatives when BP isn’t the right tool.
The paradox in SoC acne treatment
Benzoyl peroxide kills C. acnes (the bacterial driver of inflammatory acne) and oxidizes follicular content. Both actions reduce inflammatory acne. Both actions also irritate the surrounding skin, and in Fitzpatrick IV-VI patients, irritation triggers the PAR-2 mediated melanosome transfer cascade. The skin types most prone to inflammatory acne overlap substantially with the skin types most prone to inflammation-induced PIH.
Clinically, this creates the dominant SoC acne treatment failure pattern: BP clears the active acne over 8-12 weeks while simultaneously leaving the patient with hyperpigmentation that takes 6-12 months to fade. The patient sees clear skin and dark marks; the treatment looks like a partial success. From the patient’s perspective, the acne went down but the cosmetic problem — the dark marks — is worse.
The mechanism explanation matters because it points to the protocol fix. The PIH isn’t a flaw in BP; it’s a dose-dependent response to irritation, and reducing irritation reduces PIH without proportionally reducing efficacy.
What the 1986 evidence showed
Mills and colleagues randomized acne patients across 2.5%, 5%, and 10% BP formulations. The efficacy curves at 12 weeks were essentially superimposable. The 10% formulation produced meaningfully more erythema, desquamation, and stinging — but no more reduction in lesion count, no faster onset, and no better long-term clearance.
Subsequent work has replicated this. Lower-concentration BP formulations achieve the bacterial-load reduction needed for acne control because the concentration of BP isn’t the rate-limiting step. The dose-response curve flattens above ~2-3%. Higher concentrations recruit more irritation without recruiting more anti-acne effect.
For Fitzpatrick I-III skin, the 10% irritation is tolerable, and the cosmetic consequence of additional irritation is short-lived erythema. For Fitzpatrick IV-VI, the additional irritation creates PIH that outlasts the acne treatment course by half a year.
Concentration thresholds for SoC
The clinical guidance I use:
Starting concentration: 2.5%, wash-off (cleanser format), not leave-on.
Frequency: every other day for 2 weeks, daily after if tolerance is established.
Escalation: 2.5% leave-on is the next step, not 5%. Most patients never need to go above 2.5%.
Substitution: if 2.5% triggers visible PIH or persistent erythema at week 4, switch to alternatives rather than reducing frequency further.
The 5% and 10% formulations are reasonable for Fitzpatrick I-III patients who tolerate them. For Fitzpatrick IV-VI, they’re rarely the right tool.
Short-contact therapy
For patients who need BP but can’t tolerate leave-on application:
Apply BP wash to wet skin, lather, leave on for 60-90 seconds, rinse thoroughly. This delivers enough contact time for bacterial reduction (C. acnes is killed within seconds at therapeutic BP concentrations) without sustained skin contact that drives irritation.
Wash-off short-contact protocols have efficacy data comparable to leave-on for inflammatory acne. The trade-off is comedonal acne — the longer contact time of leave-on helps follicular penetration, which matters more for comedones than for inflammatory lesions. Patients with predominantly inflammatory acne (papules, pustules) get most of the benefit from short-contact. Patients with predominantly comedonal acne (blackheads, closed comedones) may need leave-on or a different active entirely — adapalene is more effective for comedones.
The PIH already-present case
If PIH has already appeared from BP treatment, the protocol pivots:
Discontinue BP, or reduce to 2-3x weekly maintenance at lowest concentration. Add a PIH-targeted topical: kojic acid 1-2%, alpha-arbutin 2%, or topical azelaic acid 10-20% (which doubles as comedolytic and antibacterial, replacing some BP function).
Photograph the PIH at the moment of pivot. The fade timeline for SoC PIH is 6-12 months even with appropriate treatment. Photographs are the only reliable progress measure — daily mirror inspection is essentially noise.
Continue active sun protection. UV exposure to active PIH meaningfully extends the fade timeline. Mineral SPF or tinted formulations with iron oxide for visible-light protection.
When to substitute BP entirely
Several actives perform comparably to BP for inflammatory acne in SoC with lower PIH risk.
Topical azelaic acid (15-20%): antibacterial, anti-inflammatory, and mildly comedolytic. Lower irritation profile than BP at equivalent efficacy in head-to-head trials. Particularly useful in SoC because it has documented tyrosinase inhibition — treating acne and PIH simultaneously rather than sequentially. Alexis and colleagues (J Drugs Dermatol 2014) reviewed the SoC-specific evidence.
Topical clindamycin: prescription only, but the antibacterial mechanism delivers most of BP’s anti-inflammatory effect without the oxidative irritation. Resistance concerns mean it’s usually combined with BP at lower concentrations in commercial products (clindamycin + 2.5% BP gel).
Adapalene 0.1%: technically a retinoid, but adapalene’s irritation profile in SoC is meaningfully lower than tretinoin. For comedonal acne specifically, adapalene outperforms BP and has a gentler PIH-risk profile.
Dapsone 5-7.5% gel: prescription, but increasingly used in SoC inflammatory acne with PIH concerns. Stein Gold and colleagues (J Drugs Dermatol 2017) documented good SoC outcomes.
Clascoterone 1% cream: the newest entry, an androgen receptor antagonist. Strong evidence for hormonal acne with low irritation profile. Expensive and prescription-only, but worth knowing.
A working SoC acne protocol
For mild-to-moderate inflammatory acne in Fitzpatrick IV-VI:
Morning: gentle non-foaming cleanser, niacinamide 4% serum, mineral SPF.
Evening: BP 2.5% wash (60-second short contact), gentle moisturizer with ceramides. Three nights per week initially, ramp to nightly over 4 weeks if tolerated.
Week 8 reassessment: if active acne is controlled and no PIH has appeared, maintain. If PIH has appeared, substitute azelaic acid 15-20% for BP and consider adapalene 0.1% as an additional comedolytic.
Week 16 reassessment: if PIH is the dominant remaining issue, the protocol shifts to PIH-fading actives (tranexamic acid 3%, kojic acid 2%, alpha-arbutin 2%) while maintaining acne control with the lowest-irritation active that holds the result.
FAQ
*Is benzoyl peroxide safe for dark skin?*
Yes, at appropriate concentrations and protocols. The clinical evidence has been clear since 1986 that 2.5% is as effective as 10% with meaningfully less irritation. The risk in Fitzpatrick IV-VI skin is irritation-induced PIH, which is dose-dependent — using the lowest effective concentration minimizes risk without compromising acne control.
*Why does benzoyl peroxide cause dark marks?*
BP’s bacterial-killing action irritates surrounding skin. In Fitzpatrick IV-VI patients, irritation triggers PAR-2 mediated melanosome transfer and α-MSH signaling — the same pathway driving most iatrogenic PIH in darker skin tones. The dark marks aren’t from BP itself; they’re from the inflammation BP causes.
*What’s the best benzoyl peroxide concentration for Fitzpatrick V?*
2.5% is the evidence-supported starting concentration. Higher concentrations add irritation without adding efficacy. Most Fitzpatrick IV-VI patients never need to escalate above 2.5%.
*What’s an alternative to BP for SoC acne?*
Azelaic acid 15-20% has the strongest evidence for SoC inflammatory acne with lower PIH risk. Adapalene 0.1% is the preferred alternative for comedonal acne. Dapsone, clascoterone, and clindamycin are prescription options for refractory cases.
References
- Mills OH Jr, Kligman AM, Pochi P, Comite H. Comparing 2.5%, 5%, and 10% benzoyl peroxide on inflammatory acne vulgaris. Int J Dermatol. 1986. PubMed.
- Davis EC, Callender VD. Postinflammatory hyperpigmentation: a review of the epidemiology, clinical features, and treatment options in skin of color. J Clin Aesthet Dermatol. 2010. PubMed.
- Alexis AF, Lamb A. Acne treatment in skin of color. J Drugs Dermatol. 2014. PubMed.
Khabir Uddin is the founder of Elelaf Journal. This article is editorial, not medical advice. For refractory acne or active PIH that hasn’t responded to OTC protocols within 12 weeks, see a board-certified dermatologist with experience in skin of color.