Why killing bacteria is not the answer: the science behind the next generation of acne care

“If the old tools clear my skin, why would killing bacteria be the wrong goal?”

Knowledge HubThe science12 min read

Download the full bookletPDF · 18 pages · Knowledge Hub booklet

There is a quiet irony at the heart of modern acne care. The treatments designed to clear the skin can, over time, degrade the very ecosystem that keeps it healthy. Antibiotics select for resistant strains and thin out protective microbes; benzoyl peroxide raises water loss and strips the barrier it is applied to protect; and almost every conventional active leaves the biofilm-encased bacteria that drive relapse untouched. The harder a product works to eradicate, the more collateral damage it tends to inflict — and the more reliably acne returns once treatment stops.

In one sentence

Conventional antibacterials work, but each carries an ecological cost — resistance, barrier damage, dysbiosis and an inflammatory burst when cells are ruptured — so the most defensible direction for acne care is to restore balance and physically remove persistence, not kill indiscriminately.

Short-term: potency has a price

The first twelve weeks of any regimen are where tolerance is won or lost — and where the trade-off between potency and skin cost becomes visible. Benzoyl peroxide carries the clearest short-term efficacy signal of any first-line topical, and the clearest irritancy burden: in a 12-week trial of 609 patients, drug-attributable skin exfoliation affected roughly one in five users, and a 10% formulation raises transepidermal water loss — the rate at which skin leaks moisture — about 1.8-fold.

Topical antibiotics feel gentle in week one, but the cost accrues invisibly as resistance and dysbiosis, which is precisely why guidelines now insist they never be used alone. Salicylic acid is better tolerated and microbiome-sparing, but its evidence base is thinner. Early selective, non-lytic approaches invert the usual trade-off — dismantling biofilm while sparing the barrier — though that evidence is still largely preclinical.

The agent with the strongest short-term clearance, benzoyl peroxide, also carries the heaviest barrier and microbiome cost. High efficacy and low collateral damage have, until now, pulled in opposite directions.

The long game: a five-decade trajectory

Resistance is the strongest long-term differentiator between treatment classes. In C. acnes, it climbed from roughly 20% of patients in the late 1970s to a peak of 64% by the late 1990s — and modern meta-analyses show macrolide resistance still rising, from about 10% in 2008 to 44% in 2024. Resistance can emerge within eight weeks of topical monotherapy and does not stay put: it spreads to household contacts and to other organisms.

Benzoyl peroxide is the important exception. Because it kills by non-specific oxidation rather than a single molecular target, C. acnes has not developed meaningful resistance to it — which is why every modern guideline pairs antibiotics with it. But resistance-neutral is not the same as consequence-free.

Antibiotic resistance in C. acnes has climbed for five decades. Benzoyl peroxide is the exception that proves the rule — oxidative killing leaves no resistance to select.

Recovery

It does not bounce back

After oral antibiotics, Pseudomonas blooms, Streptococcus stays elevated, and protective Lactobacillus remains depleted weeks after the course ends.

Relapse

“The rule,” not the exception

Biofilm shelters the bacteria conventional actives cannot reach, so the reservoir re-seeds within weeks — recurrence after non-isotretinoin therapy is expected.

The opening

Remove the structure

An enzyme that dismantles the biofilm scaffold without killing the cells disperses C. acnes biofilm in vivo — exposing it to immune clearance rather than a resistance arms race.

Remember this

Conventional therapy suppresses free-floating bacteria but cannot penetrate the biofilm sheltering C. acnes inside the follicle. So treatment clears the surface while the reservoir persists — and when therapy stops, recolonisation follows. Removing the structure beats escalating the dose.

Acne’s hidden public-health problem

It is tempting to treat acne prescribing as a private matter between clinician and patient. The antimicrobial-resistance data say otherwise. Bacterial resistance was associated with an estimated 4.95 million deaths in 2019, of which 1.27 million were directly attributable — a toll concentrated in lower-resource settings and forecast to climb toward 8.2 million associated deaths a year by 2050.

Dermatology is not a bystander: skin conditions account for around 8% of primary-care antibiotic prescriptions, and dermatologists prescribe more antibiotics per provider than any other specialty. C. acnes resistance is now a global phenomenon, dominated by macrolide resistance across Asia, with historical European surveys reaching as high as 94% in some countries.

The scale that reframes the question

1.27M

Deaths attributable to AMR, 2019

~64%

Peak C. acnes resistance recorded

~8%

Of primary-care antibiotics are for skin

What “antibiotic-free” really means

“Antibiotic-free” and “antiseptic-free” are powerful phrases, but they carry a nuance that decides whether a product is a cosmetic or a drug. Across the US, EU and Singapore, that line turns on intended use and mechanism, not ingredient labels. Benzoyl peroxide and salicylic acid are classed as over-the-counter acne drug actives — so a product containing them is a drug even when labelled “antibiotic-free.” The counter-intuitive consequence: antibiotic-free does not mean non-drug.

A second boundary matters for anything that influences microbes. The EU’s biocide rules capture products that control organisms “by means other than mere physical or mechanical action” — which is exactly the carve-out a physical biofilm-removal mechanism can occupy. A mechanism that disperses rather than kills, paired with disciplined “supports a balanced microbiome” language, occupies the clearest cosmetic pathway available. The binding constraint is that a cosmetic cannot make disease-treatment claims.

The strategic tension

The most recent translational review names regulatory clarity as a leading barrier for next-generation acne actives. The more powerfully a product demonstrates lesion reduction, the more it risks being pulled into drug classification — so the resolution is as much mechanistic as legal.

The lysis problem: when killing makes things worse

Killing a bacterium does not make its contents disappear — it releases them. When C. acnes is lysed, by antibiotics, oxidative biocides or lytic agents, it releases a bolus of porphyrins, cell-wall fragments and free fatty acids. Each is an established inflammatory trigger: porphyrins activate the NLRP3 ‘alarm’ inside skin cells, while cell-wall fragments are detected by the TLR2 ‘smoke detector’ that ignites the acne cytokine cascade.

Medicine has seen this before. The synchronised release of bacterial contents after killing has a name in other infections — the Jarisch–Herxheimer reaction — and the same TLR2 biology applies to the skin. For sensitive and pigmentation-prone skin, this inflammatory burst is not trivial: the cytokines that follow drive post-inflammatory hyperpigmentation, marks that can outlast the original lesion by months. A non-lytic approach keeps the cargo contained.

The new paradigm: restore, don’t eradicate

Assemble the evidence and a coherent direction emerges. Acne is a disorder of balance and persistence, not bacterial number: the load of C. acnes is similar in clear and acne-prone skin, and what differs is the loss of strain diversity, the entrenchment of biofilm and the inflammation that follows. A physical-removal paradigm answers each liability at once — it applies no selection pressure for resistance, spares the barrier and commensals, dismantles the relapse reservoir, and avoids the lysis-release burst.

This is a matter of direction, not displacement. Benzoyl peroxide remains a resistance-neutral workhorse, and the confirmatory long-term human trials for selective biofilm removal are still to be run. But across tolerability, resistance, stewardship, regulation and inflammatory safety, the weight of mechanism points the same way.

Two philosophies, five liabilities. Physically dispersing the biofilm addresses resistance, barrier, microbiome, relapse and inflammation in a single mechanism.

The old reflex	The modern science
Kill the bacterium	Restore balance and remove persistence
More killing is better	Killing has a barrier, ecology and resistance cost
“Antibiotic-free” means safe and simple	Mechanism and claims decide cosmetic vs drug
Eradicate C. acnes	Preserve diversity; dismantle the biofilm cleanly

The shift to remember

The question for the next generation of acne care is no longer how thoroughly can we kill? — but how intelligently can we restore?

Selected references

This explainer builds on a Consensus.ai synthesis comparing biofilm removal with conventional acne antibacterials, enriched and fact-checked against current peer-reviewed and regulatory sources (2015–2026). Full citations with DOIs are in the downloadable booklet, where each claim is graded as strong, moderate or preliminary. Statements describing selective biofilm-removal benefits reflect mechanistic and preclinical evidence — not claims of demonstrated clinical superiority. Educational content; not medical, regulatory or product advice.

All explainers Download the full briefing