Painting CRISPR onto the Skin: Curing a Rare Genetic Skin Diseases with Topical Base Editing

“The real threshold is not a molecular number. It's whether patients and clinicians see a visible, meaningful improvement”Gaurav Sadhnani, CTO of Epithelica

ARCI is most commonly caused by mutations in TGM1, the gene encoding transglutaminase 1, an enzyme essential for cross-linking the outermost layer of the skin barrier. The most common mutation, a splice-site mutation called c.877-2A>G, accounts for about a third of all cases.

The study used a cytosine base editor, eTd-CBE, to restore the mutant nucleotide to its wild-type form. The key technical challenge was that the target sits immediately beside a 'bystander' base that must remain untouched. The team exploited eTd-CBE's narrow editing window to achieve up to 26% on-target editing in patient-derived keratinocytes, with no detectable bystander editing – a feat the authors describe as among the first demonstrations of therapeutic splice-site repair using a cytosine base editor.

Human skin blocks biomacromolecules from reaching the living cell layers below the stratum corneum. The team's solution was clinically approved laser microablation: short pulses from a commercially available dermatological laser create tiny, controlled pores that penetrate to a defined depth. The pores close again within 16 hours.

For Sadhnani, this laser step is a strategic asset rather than a bottleneck. »The procedure can be performed with commercially available lasers, which many dermatology centres are equipped with,« he says. »Dermatologists are already very familiar with laser-assisted skin treatments, so from a practical perspective, this fits naturally into existing clinical workflows.« He adds that covering the full adult skin surface area of 1.5 to 2 square metres may require multiple short sessions. Epithelica has filed a patent covering the core platform.

“The central open question is durability – specifically, whether we are editing the stem cell populations that sustain long-term tissue renewal”Sarah Hedtrich, the University of British Columbia and co-founder, CSO of Epithelica

The laser step also provides a safety feature: by targeting the non-vascularised epidermis, it limits the gene editor's exposure to the intended tissue. Using fluorescence imaging, qPCR, and DESI metabolic mass spectrometry imaging, the team found no evidence of systemic distribution of either the nanoparticles or the genetic cargo to any organ, even after repeated dosing.

In 3D ARCI disease models built from patient-derived cells, the team achieved a mean on-target gene correction of approximately 12%, with some samples reaching 24%. This translated into roughly 3% restoration of normal TG1 enzyme activity. Studies in other genodermatoses suggest that restoring as little as 5-10% of wild-type protein levels can alleviate severe symptoms, placing Epithelica's numbers well within the clinically relevant range.

Durability remains the central unknown

»The real threshold is not a molecular number,« Sadhnani argues. »It's whether patients and clinicians see a visible, meaningful improvement. In dermatology, clinical endpoints are often based on visual scoring and quality of life, because these diseases have a strong psychosocial component.«

The most consequential open question is whether the treatment can last. Single-cell RNA sequencing data showed that about 39% of transfected cells were basal keratinocytes – the layer containing the skin's stem and progenitor cells. If those cells are durably corrected, the edited gene should propagate through natural epidermal turnover, potentially offering a long-lasting cure.

Hedtrich is candid. »I don't think there is a single biological uncertainty that fundamentally blocks a first-in-human trial. The central open question is durability – specifically, whether we are editing the stem cell populations that sustain long-term tissue renewal.«

But she reframes the stakes: even if durability proves limited, »a treatment requiring repeat administration every few weeks that reliably reverses the clinical phenotype would already be a substantial improvement over existing options.«

“Our primary focus right now is ARCI,« Sadhnani says. »Prioritising a flagship asset that balances speed, regulatory pathway, and patient impact gives us the best foundation to then expand into additional indications”Gaurav Sadhnani, CTO of Epithelica

Comprehensive off-target analysis using four independent methods showed negligible off-target editing. In vivo toxicity studies in mice, conducted under a protocol agreed with Germany's Paul Ehrlich Institute, showed no deaths, no changes in body weight, no skin irritation, no significant immune cell infiltration, and no systemic cytokine elevation – even after repeated dosing. The lead LNP formulation triggered no significant immune activation in dendritic cells, showing minimal immune activation relative to comparator formulations, including an mRNA vaccine control used for benchmarking purposes.

Epithelica positions the therapy within existing regulatory frameworks

The therapy sits at an unusual crossroads: a gene-editing product applied topically, classified as an advanced therapy medicinal product under European law, yet one that stays in the skin.

Sadhnani sees this as an opportunity. »We don't think we're asking regulators to invent a completely new category,« he says, pointing to recent in vivo base editing milestones in the liver and retina. He highlights the FDA's proposed Plausible Mechanism Pathway, which contemplates approval of bespoke therapies when there is a clear biological mechanism and early clinical benefit. »Under this framework, mechanistic plausibility coupled with small-cohort evidence of effect can form the foundation for marketing authorisation.«

The company holds Orphan Drug Designation from both the FDA and EMA for ARCI, unlocking fast-track designation, smaller trial sizes, and priority review.

The paper frames the approach as a platform applicable to approximately 500 known genodermatoses. Different mutations would require different guide RNAs and potentially different base editors, but the delivery system and workflow would remain the same.

No other company has yet reported in situ base editing directly in human skin using a topical LNP approach. The closest clinical comparators remain ex vivo gene therapies in epidermolysis bullosa and viral gene augmentation strategies, such as Krystal Biotech's HSV-based platform. Those approaches require repeated application or surgical grafting, and none yet combine local delivery with permanent nucleotide correction. Whether Epithelica can sustain a first-mover advantage will depend not only on data, but on speed.

For now, Epithelica is deliberately focused. »Our primary focus right now is ARCI,« Sadhnani says. »Prioritising a flagship asset that balances speed, regulatory pathway, and patient impact gives us the best foundation to then expand into additional indications.«

“Deeper biological questions may remain, but unresolved curiosity is not the same as unresolved risk”Sarah Hedtrich, the University of British Columbia and co-founder, CSO of Epithelica

The deeper strategic question is whether Epithelica should ultimately be valued as a rare-disease asset with platform optionality, or as a delivery-platform company whose first indication happens to be ARCI. If the LNP–laser workflow proves reproducible and safe, the real long-term asset may be the delivery architecture itself rather than any single mutation it corrects. That distinction will shape both partnership dynamics and future capital raises.

Funding gates the path to patients

The most immediate bottleneck is capital. Epithelica is raising a seed round to fund GMP manufacturing and first-in-human preparations. »A major value inflection is a successful rescue in a human patient,« Sadhnani says. »But even before that, a key capital inflection point is the transition from preclinical work into the clinic.«

Hedtrich notes that the obstacles are no longer scientific. »The primary bottlenecks are non-biological: securing funding for a first-in-human trial and navigating the evolving IP landscape around CRISPR, which remains a practical challenge for early clinical translation, particularly in Europe.«

The CRISPR patent landscape remains fragmented, particularly for base editing technologies, where foundational rights are held by multiple institutions. Freedom-to-operate, sublicensing structures, and geographic patent variances could influence both the cost and the speed of clinical entry. For a small company, IP strategy becomes as operationally decisive as biology. Epithelica plans to remain independent through its seed stage but is open to partnering if it accelerates the path to patients.

The study acknowledges its own limitations. The data were generated using immortalised patient keratinocytes, not primary cells. The 3D skin models cannot fully replicate the complexity of in vivo human skin. Treating large body surface areas raises questions about cumulative immune responses. And the long-term kinetics of edited versus diseased cells remain unknown.

Hedtrich draws a distinction. »Deeper biological questions may remain, but unresolved curiosity is not the same as unresolved risk.«

For the roughly 1 to 7 in every 100,000 people living with moderate to severe ARCI, what matters is whether someone can move the science onto a laser-treated patch of skin and get it to patients. If Epithelica's preclinical data hold up in humans, that may be coming sooner than the field expected.

This article draws on the published study and separate interviews with both Gaurav Sadhnani and Sarah Hedtrich.