A small acne trial offers a useful example of why the product behind a percentage matters. Over 28 days, investigators compared a modified 2% salicylic acid cream on one facial side with a benzoyl peroxide–adapalene gel on the other. Both sides showed reductions in acne lesions. That is evidence about two specific topical regimens under study conditions—not proof that every 2% salicylic acid product performs alike, and not a measurement of permanent pore shrinkage.
The study’s strongest feature was its within-person comparison: each participant contributed a treated side to both regimens. Its main constraints were equally important: only 31 people completed it, treatment was open-label, and the indexed abstract provides neither confidence intervals nor detailed adverse-event frequencies. This analysis uses the indexed primary report abstracts and available official FDA material; it does not claim examination of the full trial protocols, supplements, or subscription-only full article.
Thirty-one people with mild to moderate acne completed a randomized, open-label, single-center, split-face trial lasting 28 days.
Study facts at a glance
Study population: Thirty-one people with mild to moderate acne completed a randomized, open-label, single-center, split-face trial lasting 28 days. Source 3
Specific comparison: The trial compared topical 2% supramolecular salicylic acid cream with 5% benzoyl peroxide plus 0.1% adapalene gel, according to the title and methods. Source 3
Total lesion result: Reported total lesion reductions were 44.1% and 45.6%, respectively; the between-side comparison was not statistically significant. Source 3
Skin measurements: The abstract reported no statistically significant between-side differences in hydration, transepidermal water loss, brightness, or erythema indicators. Source 3
US label context: FDA’s OTC acne monograph lists salicylic acid at 0.5–2%; concentration alone does not establish compliance, product-specific approval, or equivalence to the studied cream. Source 4
One Person, Two Treated Sides
Investigators enrolled people with mild to moderate facial acne and randomly assigned the two topical treatments to opposite facial sides. The report describes a single-center, open-label study: treatment assignment was not concealed through blinding. Assessments occurred at baseline, day 14, and day 28. The abstract identifies 31 completers, but it does not disclose the total enrollment denominator or how many participants, if any, withdrew.
A split-face design reduces some differences that complicate comparisons between separate groups. Both facial sides belong to a person with the same age, general health, and broad environmental exposures. Random assignment helps address systematic side preference. Still, sides can differ at baseline, and the two sides are statistically related. They are not 62 independent patients; appropriate analysis must account for their pairing.
The design is therefore efficient for an early comparative study, but it cannot create a large population from a small one. Missing enrollment and withdrawal information also prevents assessment of whether completion favored people who tolerated treatment particularly well. The abstract’s findings most directly describe the participants who completed this short study, not an unrestricted acne population.
Read the underlying chart data: download CSV or download the source-labeled chart.
The formulations matter, including a concentration discrepancy
The salicylic acid arm used a 2% supramolecular preparation, described as a modified formulation, in a cream. The comparator was a gel containing both 5% benzoyl peroxide and adapalene. This was not salicylic acid versus benzoyl peroxide alone, nor was it a comparison of interchangeable products sharing one ingredient. The concentrations identify study arms; they are not instructions for combining products.
There is an internal reporting discrepancy worth preserving rather than silently correcting. The abstract’s objective mentions 0.01% adapalene, while its title and methods identify 0.1%; the objective also includes a parenthetical reference to 0.1%. This article uses 0.1% when describing the title-and-methods regimen, without claiming independent confirmation from the full paper.
A cream and gel can differ beyond their named active ingredients. The abstract does not allow separation of formulation effects from ingredient effects or characterization of the modified preparation’s behavior. The practical framework in our cream, cleanser, and liquid guide helps distinguish formats, but it does not turn an untested wash or liquid into the cream used here.
What the lesion reductions actually show
Investigators counted acne lesions and evaluated side effects in the targeted areas at each visit. The abstract does not explicitly designate a single primary endpoint, so total lesion reduction should not be promoted into a confirmed prespecified primary outcome. It reports day-28 changes for inflammatory lesions, noninflammatory lesions, and total lesions, alongside several instrumental skin measurements.
Inflammatory lesions included papules and pustules—the raised or pus-containing lesions counted in the report. Noninflammatory lesions represent a different acne category rather than a measurement of pore diameter. Across all three reported lesion categories, the numerical differences between regimens were small. The table presents those group-level reductions without assigning an unsupported mean or median label.
For total lesions, the numerical gap was 1.5 percentage points, favoring the comparator. That is a difference between two percentage reductions, not a 1.5% difference in the number of people cured. Neither regimen’s percentage establishes complete clearance, the fraction of participants who responded, or whether residual acne remained bothersome.
Reported day-28 lesion-count reductions in the split-face study
| Lesion category | 2% supramolecular salicylic acid cream | 5% benzoyl peroxide + 0.1% adapalene gel |
|---|---|---|
| Inflammatory lesions | 47.9% | 49.8% |
| Noninflammatory lesions | 43.1% | 42.7% |
| Total lesions | 44.1% | 45.6% |
Similar numbers are not proof of equivalence
The report states that the between-side lesion comparisons were not statistically significant, with all reported P values above 0.05. This means the analyses did not establish a difference under their statistical criteria. It does not establish that the treatments are equivalent, that either could replace the other for every patient, or that a meaningful difference has been ruled out.
Formal equivalence or noninferiority testing requires a specified acceptable difference and an analysis capable of testing that boundary. Those details are not available in the abstract. Confidence intervals are also absent. Without them, readers cannot see the range of treatment differences compatible with the data or judge how tightly the study estimated the apparent similarity.
As an illustration only, a 44.1% reduction applied to a hypothetical starting count of 100 lesions would leave 55.9 lesions in a mathematical calculation. That is not an observed patient result: real lesion counts are whole numbers, starting counts vary, and the abstract does not explain how individual changes were summarized. The example shows why a reduction percentage is not a clearance rate.
Barrier measurements answer a narrower question than comfort
The investigators measured skin water content and transepidermal water loss, often abbreviated TEWL. TEWL measures water passing through the skin into the surrounding air and is used to evaluate barrier function. They also measured a lightness indicator, L*, and an erythema-related indicator, a*, which captures a color dimension relevant to redness.
The abstract reports no statistically significant differences between facial sides for these measures. That is not the same as showing that hydration or water loss stayed unchanged from baseline. Two treatments could produce similar changes over time and still have no significant between-side difference. Numerical trajectories and confidence intervals would be needed to describe those changes reliably.
Instrument readings also do not substitute for complete symptom reporting. Although side effects were evaluated, the abstract supplies no detailed frequencies for burning, stinging, peeling, or treatment interruption. Thus, the findings do not establish an irritation-free regimen. Our irritation-versus-purging guide offers a vocabulary for describing symptoms rather than treating every reaction as evidence that an acne product is working.
Fewer acne lesions do not demonstrate smaller pores
This study counted lesions and measured selected skin properties; it did not report a pore-diameter endpoint, permanent pore closure, scarring improvement, or wrinkle reduction. A reduction in noninflammatory acne lesions can matter clinically without establishing any of those separate outcomes. Likewise, an instrumental lightness measurement is not a comprehensive test of complexion quality.
The distinction is especially relevant when acne evidence appears in marketing for cosmetic exfoliants. FDA discusses salicylic acid under the common cosmetic label beta hydroxy acid, or BHA, but that page is not a product-specific acne trial. Its discussion of historical industry safety work should not be treated as current FDA approval or a newly verified clinical benefit.
A useful interpretation is to match the claim to the measurement: counted lesions support discussion of lesion changes, while a permanent structural claim needs direct evidence about that structure and its durability. Our blackheads and whiteheads guide develops that distinction. Nothing in this trial measures how the skin looked after treatment ended.
Other salicylic acid trials tested different questions
Two other small randomized studies provide context without replicating this experiment. One compared a chloroxylenol–salicylic acid cream with benzoyl peroxide gel over 12 weeks. Another examined adding salicylic acid to a clindamycin–benzoyl peroxide regimen. Differences in additional ingredients, vehicles, follow-up, and study design prevent treating all three as repeated tests of salicylic acid alone.
In the 37-volunteer trial, reported week-12 inflammatory lesion reductions were 62% for the combination cream and 60% for benzoyl peroxide; noninflammatory reductions were 56% and 54%. Between-group lesion differences were not significant. Erythema and photosensitivity were significantly less frequent with the combination at week 12, but the abstract gives no event frequencies. Our analysis of that combination-cream trial examines the separate comparison.
The 49-patient add-on study reported significantly greater lesion reductions with added salicylic acid, but it supplies no effect sizes or confidence intervals in the abstract. Both regimens reduced outer-skin hydration, and the salicylic acid addition caused more mild to moderate transient dryness. That finding argues against a blanket claim that including salicylic acid always improves tolerability.
Three studies, three distinct comparisons
| Study design and size | Treatment question | Interpretive boundary |
|---|---|---|
| 28-day, randomized, open-label split-face study; 31 completers | Modified 2% salicylic acid cream versus benzoyl peroxide–adapalene gel | Short-term, paired comparison; nonsignificant differences do not establish equivalence |
| 12-week, randomized, double-blind study; 37 volunteers | 0.5% chloroxylenol + 2% salicylic acid cream versus 5% benzoyl peroxide gel | Combination treatment, not salicylic acid monotherapy; different vehicles |
| 12-week, randomized, single-blind study; 49 enrolled | Salicylic acid added to clindamycin–benzoyl peroxide versus the regimen without added salicylic acid | Add-on benefit reported without abstract effect sizes; more transient dryness |
US acne labels provide boundaries, not a bridge to this cream
FDA’s OTC topical acne monograph lists salicylic acid concentrations of 0.5–2% and benzoyl peroxide concentrations of 2.5–10%. Those ranges form only part of the applicable conditions. A product’s concentration alone does not establish full conformity, FDA product-specific approval, or clinical equivalence to a formulation tested elsewhere.
The monograph addresses acne-related uses, including blackheads and whiteheads, rather than permanent pore-size alteration. It also requires warnings concerning increased dryness and irritation when topical acne medicines are combined. Benzoyl peroxide has additional warnings, including fabric bleaching and precautions involving sunlight, sensitivity, and contact with eyes or mucous membranes.
These label distinctions are more actionable than matching a front-label percentage to a research headline. Our Drug Facts reading guide explains the document to inspect. The comparator here contained adapalene as well as benzoyl peroxide; the OTC salicylic acid concentration range does not validate every component or combination used in the study.
Rare serious reactions require a separate safety lens
A short trial with 31 completers cannot characterize uncommon serious reactions reliably. FDA’s safety communication distinguishes ordinary local irritation from rare serious hypersensitivity associated with some OTC acne products containing salicylic acid or benzoyl peroxide. The responsible ingredient was not established: active ingredients, inactive ingredients, or both could have contributed.
The historical FDA review identified 131 reports through January 28, 2013, including 58 hospitalizations and 50 reports categorized as anaphylactic. These are selected spontaneous reports, not an incidence study. Without a denominator showing how many people used the products, dividing those reports into a percentage risk for users would be misleading.
FDA advises stopping the product and seeking emergency care for throat tightness, breathing difficulty, faintness, or swelling of the eyes, face, lips, or tongue. A small-area sensitivity check does not guarantee that a later reaction will not occur. These warnings complement, rather than contradict, the trial’s limited measurements of redness and barrier function.
What stronger evidence would need to resolve
A larger study could clarify whether the small observed differences reflect genuinely close performance or imprecise estimation. Useful reporting would include enrollment and withdrawal counts, baseline lesion distributions, paired statistical methods, a clearly designated primary endpoint, and confidence intervals. Longer follow-up would address persistence of improvement and tolerability beyond four weeks; this trial does not answer those questions.
Detailed adverse-event counts and patient-reported outcomes would add information that instrument readings cannot supply. The accessible abstract also does not establish the sponsor or funding source, describe all formulation ingredients, or resolve the adapalene discrepancy. Those gaps should remain visible rather than being filled with assumptions about who funded the work or how products were manufactured.
Editorial interpretation: this evidence supports formulation-specific questions, not provider-level promises. Readers considering service agreements can use the CoreAge Rx review and Likeness Rx review to check documented product, payment, and renewal terms. Those documentary comparisons do not show that either provider supplies the studied formulation, offers condition-specific care, or reproduces these outcomes; nor do the results automatically transfer to compounded products, oral routes, or other formulations.
Questions about the evidence
Did salicylic acid work as well as the benzoyl peroxide–adapalene gel?
The reported lesion reductions were numerically close, and the study did not find statistically significant between-side differences. That supports further investigation of the modified cream. It does not establish formal equivalence or rule out a clinically relevant difference, particularly because confidence intervals and equivalence-testing methods are unavailable in the abstract.
Why are there 31 participants rather than 62?
Each of the 31 completers contributed two facial sides, one for each treatment. Those sides share the same person and are not independent participants. Counting them as 62 patients would exaggerate the population size and obscure the paired nature of the comparison.
Does the study establish that a standard 2% BHA liquid will produce the same result?
No. The study tested a modified supramolecular salicylic acid cream, not every product containing 2% salicylic acid. The active concentration does not describe the entire formulation, vehicle, or exposure. FDA’s cosmetic BHA background page also does not supply a clinical bridge from one formulation to another.
Were skin hydration and redness unchanged during treatment?
That cannot be concluded from the reported between-side comparisons. The abstract says the sides did not differ significantly on measured hydration, water loss, brightness, or erythema indicators. It does not provide enough numerical information to establish that either side remained unchanged from baseline or that participants experienced no irritation.
Does a 44.1% total lesion reduction mean 44.1% of participants cleared their acne?
No. Lesion-count reduction describes a change in counted lesions, not the proportion of people reaching clear skin. The abstract does not report a clearance rate. It also does not measure permanent pore shrinkage or show whether improvement persisted after the 28-day treatment period.
What this evidence supports
The study provides a useful pilot signal: a specific modified salicylic acid cream and a benzoyl peroxide–adapalene gel produced numerically similar short-term lesion reductions in 31 completers. Its within-person randomization strengthens the comparison, but its size, open-label design, short duration, and incomplete abstract reporting limit certainty. The most defensible conclusion is formulation-specific and outcome-specific. It supports larger, better-reported comparative trials—not universal interchangeability, a guarantee of comfort, or permanent pore-size claims.
Research access and limitations
Evidence access consisted of indexed primary study abstracts and the available official FDA documents. The split-face trial’s full text and supplements were not examined. Enrollment denominator, attrition, funding, detailed adverse-event frequencies, confidence intervals, and full statistical methods were unavailable in the accessed abstract. Its objective contains an adapalene concentration discrepancy relative to its title and methods.
Sources & further reading
Commercial sources establish what a seller publishes. They do not independently verify its results. Browse the source register.
- Comparison of chloroxylenol 0.5% plus salicylic acid 2% cream and benzoyl peroxide 5% gel in the treatment of acne vulgaris: a randomized double-blind study.Primary study abstract · checked 2026-10-21. Indexed primary study abstract
- Efficacy of the addition of salicylic acid to clindamycin and benzoyl peroxide combination for acne vulgaris.Primary study abstract · checked 2026-10-21. Indexed primary study abstract
- Efficacy and safety of 2% supramolecular salicylic acid compared with 5% benzoyl peroxide/0.1% adapalene in the acne treatment: a randomized, split-face, open-label, single-center study.Primary study abstract · checked 2026-10-21. Indexed primary study abstract
- FDA OTC monograph M006: topical acne drug productsOfficial FDA information · checked 2026-10-21. Official six-page administrative order, all sections read
- FDA safety communication: serious hypersensitivity to some OTC acne productsOfficial FDA information · checked 2026-10-21. Official FDA Spanish communication; entire safety notice and data summary read
- FDA: beta hydroxy acids in cosmeticsOfficial FDA information · checked 2026-10-21. Official FDA consumer background page, complete text read