This week in Dermatology

This AudioScholar briefing is generated by artificial intelligence for healthcare professionals and trainees. It is not medical advice.

This week: a personalised neoantigen therapy added to pembrolizumab extends recurrence-free survival in resected melanoma, first-trimester doxycycline is linked to more pregnancy loss, and an oral TYK2 and JAK1 inhibitor improves cutaneous sarcoidosis in a small placebo-controlled trial.

Welcome to This Week in Dermatology. We're covering 10 papers this week: 3 in depth, 4 in brief, 3 quick signals.

In the New England Journal of Medicine, Long and colleagues report a phase 3, double-blind trial of intismeran autogene, an individualised neoantigen therapy, added to pembrolizumab after complete resection of stage IIB to IV cutaneous melanoma [1]. In total, 1137 patients were randomised two to one to intismeran or placebo, given intramuscularly every three weeks for nine doses, and every patient received pembrolizumab every six weeks for nine cycles. At a median follow-up of about two years, recurrence-free survival at 24 months was about 82 percent with the combination against 70 percent with pembrolizumab alone. Distant-metastasis-free survival also improved, to about 89 percent against 80 percent. The benefit was not apparent in the first six months; after that point the risk of recurrence was cut by more than half. Grade 3 or higher adverse events occurred in just under a quarter of patients on the combination and about a fifth on pembrolizumab alone. Eight patients in the combination group died from adverse events, five of them before receiving intismeran. The main limitation is follow-up: two years is short for melanoma, and the abstract reports no overall survival. This is the first phase 3 evidence that an individualised neoantigen therapy adds to adjuvant checkpoint inhibition in high-risk melanoma.

Winterstein and colleagues, writing in Annals of Internal Medicine, used two large United States claims databases, MarketScan commercial and Medicaid, for a retrospective active-comparator cohort study of first-trimester oral doxycycline [2]. Doxycycline-exposed pregnancies were propensity-matched to comparators: amoxicillin or a macrolide in more than 218,000 pregnancies with respiratory tract infection, and topical clindamycin in about 6,600 pregnancies with acne. For respiratory infection, doxycycline was associated with roughly 40 to 50 percent higher risk of pregnancy loss than amoxicillin, consistent across both databases, with similar results against macrolides. The acne comparison, the one closest to dermatology practice, was split. Risk was about half again higher than with topical clindamycin in the commercial database, but there was no statistically significant difference in Medicaid. Where associations were significant, the absolute excess was 7 to 11 percentage points. The authors acknowledge possible residual confounding and misclassification of exposure and outcome, and congenital malformations were not studied. The acne cohort is also far smaller than the infection cohorts. The study adds large, replicated observational data on a drug commonly prescribed to women of reproductive age, and the authors frame it as information for weighing risks and benefits.

In JAMA Dermatology, Rosenbach and colleagues report the phase 2 portion of BEACON, a global, multicentre, double-blind, placebo-controlled trial of brepocitinib, an oral inhibitor of TYK2 and JAK1, in adults with biopsy-proven active cutaneous sarcoidosis, a disease with no approved therapy [3]. Thirty-one adults were randomised to 45 milligrams, 15 milligrams or placebo daily for 16 weeks, with systemic glucocorticoids stopped per protocol. On the activity score, both doses improved by about 22 points, against less than one point with placebo. Stricter end points favoured the higher dose: about seven in ten patients on 45 milligrams reached clear or almost clear on the investigator global assessment, while no placebo patient did. Every patient on the higher dose had at least a 10-point reduction. Responses extended to lupus pernio, and skin-related quality of life improved. There were no deaths or serious adverse events. The limitation is size: 31 participants over 16 weeks say little about durability or uncommon harms, and this is an interim analysis. It is the first randomised, placebo-controlled trial to show efficacy for a targeted therapy in this disease, and phase 3 is under way.

Now, four in brief.

Staying with Janus kinase inhibition, Mostaghimi and colleagues in the Journal of Investigative Dermatology report a Bayesian network meta-analysis of seven phase 3 or 4 trials, about 3,700 patients with severe alopecia areata, comparing baricitinib, ritlecitinib, deuruxolitinib and upadacitinib [4]. The estimated probability of reaching a SALT score of 20 or less at week 24 was highest for upadacitinib 30 milligrams, at about 54 percent, followed by upadacitinib 15 milligrams, deuruxolitinib, baricitinib 4 milligrams, ritlecitinib and baricitinib 2 milligrams, against under 3 percent with placebo. These are indirect comparisons across trials with different designs, upadacitinib is the one agent outside the FDA-approved group, and head-to-head trials are still lacking.

In the Journal of the American Academy of Dermatology, Xu and colleagues ran a single-centre, randomised, double-blind, placebo-controlled trial of 0.2 percent sirolimus gel twice daily in 65 patients with hyperplastic rosacea [5]. After 12 weeks, about 72 percent of patients on sirolimus reached an investigator global assessment of 1 or less, against a third on placebo, and erythema success was more than twice as frequent. An artificial intelligence estimate of protrusion volume also fell more with sirolimus. There were no serious adverse events. The authors call these exploratory signals: the sample is small, follow-up short, the volume algorithm unvalidated, and systemic absorption and quality of life were not assessed.

Muha and colleagues, in JAMA Dermatology, used linked administrative data from Quebec to follow about 6,200 patients with morphea and 8,100 with systemic sclerosis [6]. Cancer incidence was about a quarter higher than in the general population with morphea and just over a third higher with systemic sclerosis. Haematologic cancer risk more than doubled in systemic sclerosis but was not significantly raised in morphea. Breast and female reproductive cancers predominated in morphea, lung cancer in systemic sclerosis, and younger patients carried the highest relative risk. Immunosuppressant use was associated with lower cancer risk, an association open to confounding in claims-based data, and whether screening helps remains untested.

In the Journal of the European Academy of Dermatology and Venereology, Karakioulaki and colleagues describe 38 patients with dystrophic epidermolysis bullosa treated with beremagene geperpavec, the topical COL7A1 gene therapy, in a multicentre observational study [7]. Wound surface area fell significantly across healing wounds, and haemoglobin and body mass index rose. Patients with persistent wounds had lower body mass index, higher leucocyte counts and less frequent systemic biologic therapy. Anti-collagen VII antibodies increased during treatment, while other antibody titres stayed stable and did not predict response. Most patients reported less pain, itch and dressing time. With no control group and a small cohort, these are real-world associations.

Three quick signals.

In Clinical and Experimental Dermatology, Gundelwein and colleagues followed about 7,200 patients with severe atopic dermatitis treated with dupilumab for at least six months [8]. About one in eight patients stopped for a year or more without switching, and most of those patients stayed off systemic therapy throughout follow-up. This is descriptive cohort data, and reasons for stopping are not reported.

Ajmani and colleagues, in the Journal of the American Academy of Dermatology, open a two-part continuing education review of primary cicatricial alopecias [9]. Part one updates classification by predominant immune infiltrate, whether lymphocytic, neutrophilic, mixed or nonspecific, and covers trichoscopy, biopsy technique and digital pathology. It is a narrative review without new data.

The second part, also by Ajmani and colleagues in the Journal of the American Academy of Dermatology, reviews treatment stratified by infiltrate pattern [10]. No therapy is FDA-approved, so all use is off label. Emerging options include metformin, JAK inhibitors and biologics targeting IL-17 and IL-23, with minoxidil, platelet-rich plasma, light therapy and narrowband UVB as adjuncts. Evidence for most agents remains limited.

Stepping back across this week's papers, the largest datasets were observational, drawn from claims and registries, while most of the randomised evidence came from small or short trials. Only the melanoma trial combined scale with randomisation, and even there follow-up was brief and survival was not reported. Where the observational work did split across databases, the disagreement fell in the population most relevant to dermatology, which leaves the signal less settled than its sample size suggests.

If you only have time for one paper this week, make it the doxycycline and pregnancy loss cohort [2]. It reopens the question of early-pregnancy safety for a drug widely prescribed to women of reproductive age, and its acne results disagree between databases, so that question is not yet settled.

That's your roundup for This Week in Dermatology. The full transcript and references are available on the episode page. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And a quick rating on Apple Podcasts or Spotify helps other physicians discover the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.