Luigi Naldi

The AI-CHECK (Artificial Intelligence for CME Health E-learning Contents and Knowledge) study aimed to explore ChatGPT’s potential in continuing medical education (CME). In particular, its third phase investigated artificial intelligence (AI)’s ability to develop a CME course on acne in Italian and assessed participants’ satisfaction and attitudes toward AI-generated content. The e-learning course “Acne in the Age of ChatGPT” was hosted on the SAEPE platform and offered free to Italian physicians from May to September 2025. It included a handbook, two clinical vignettes, a knowledge questionnaire, and pre- and post-course surveys. ChatGPT-4 generated all educational materials following structured prompts. Content was reviewed by an expert dermatologist. A total of 556 physicians enrolled, with 95% completing the course. Most participants (74.4%) expressed support for AI use in medicine, while 79.8% recognized its value in developing CME materials. Overall satisfaction was excellent: over 95% rated the course as relevant, high-quality, and effective. This study demonstrates that ChatGPT can generate high-quality CME content with human supervision, ensuring accuracy and reliability. AI integration in CME shows promise for scalable, personalized, and innovative medical education.

Background: Skin diseases represent a major public health concern, affecting all demographics and imposing significant personal, economic and social burdens. Objectives: To assess the epidemiology and impact of skin diseases using the Global Burden of Disease (GBD) data, highlighting regional variations, knowledge gaps and areas for improvement. Methods: GBD data on prevalence, incidence, disability‐adjusted life years (DALYs) and years lived with disability (YLDs) for skin conditions from 2000 to 2021 were analysed. Results: According to GBD 2021, the global prevalence of skin and subcutaneous diseases, excluding sexually transmitted infections (STIs) and skin cancers, was 25.7%, ranking sixth among the most prevalent disease categories worldwide. These conditions accounted for a global YLDs rate of 495.1 per 100,000 person‐years, placing them eighth in terms of disease burden. Overall, the most prevalent conditions were STIs excluding HIV (16.8%), fungal infections (7.8%), and acne (3.0%). Squamous cell carcinoma, basal cell carcinoma and decubitus ulcers showed increasing prevalence, while atopic dermatitis, scabies and viral infections declined slightly. Acne, scabies and atopic dermatitis predominated in youth, whereas fungal infections, contact dermatitis and skin cancers predominated in older adults. Females showed higher prevalence of STIs, acne and atopic dermatitis. Infectious skin diseases were more common in Sub‐Saharan Africa and tropical regions, while chronic inflammatory and neoplastic diseases predominated in high‐income regions. Conclusions: Several limitations were identified. The GBD initiative only includes some dermatological diseases reflecting the difficulties in obtaining data because of the multitude of skin diseases with underreporting and misdiagnosis. There are significant gaps in data from low‐ and middle‐income countries, a lack of reliable trends and inadequate quality‐adjusted life year estimates considering cultural and social variations. The GBD initiative provides valuable insights but underestimates the true burden of skin diseases. Improved reporting and inclusion of neglected conditions are urgently needed through global collaboration among stakeholders.

Nemolizumab with background topical therapy (corticosteroids±calcineurin inhibitors) significantly improved skin lesions, itch and sleep in two global phase 3 trials (ARCADIA 1&2) in adolescents and adults with moderate-to-severe atopic dermatitis through week (W)16. Efficacy and safety of nemolizumab, combined with background topical therapy, were evaluated for an additional 32 weeks(W16–W48) with a focus on maintained skin responses in clinical responders (patients achieving Investigator’s Global Assessment [IGA] score of 0/1 [clear/almost clear] or ≥75% improvement in Eczema Area and Severity Index [EASI-75] at W16). At W16, clinical responders (N=507) to nemolizumab every 4 weeks (Q4W) were rerandomized (1 : 1 : 1) to receive nemolizumab 30 mg-Q4W/-Q8W/placebo (nemolizumab-withdrawal) subcutaneously with background topical therapy. At W48, 61.5% (strata-adjusted difference versusnemolizumab-withdrawal group [Δ], 11.8% [95% CI 1.3–22.3]) and 76.3% (Δ, 12.4% [95% CI 2.7– 22.0]) of patients in nemolizumab-Q4W; 60.4% (Δ, 10.7% [95% CI 0.3–21.0]) and 75.7% (Δ, 11.8% [95% CI 2.1–21.5]) in nemolizumab-Q8W; and 49.7% and 63.9% in nemolizumab-withdrawal group maintained response rate for IGA success and EASI-75, respectively. The percentage of patients who experienced ≥1 adverse events were similar across the groups (range: 53.5%–58.3%). Up to W48, skin responses were maintained without notable differences in the safety profile of both nemolizumab dosing regimens.

Abrocitinib, an oral selective Janus kinase 1 inhibitor, is approved for the treatment of moderate-tosevere atopic dermatitis (AD) in adults, but realworld evidence remains limited. We retrospectively collected data from 187 adult patients (mean age:37.9 years; 60.4% male), with AD and up to oneyear exposure to abrocitinib either 100 or 200 mg, between 2023 and 2025 across 22 Italiandermatology centres. Nearly half of patients were bio-experienced and 80 % had difficult-to-treat areas. Beginning at week 8, with either 100 (59%) or 200 mg, 52 % achieved Investigator Global Assessment 0/1, significantly increasing to 88 % at week 52. Similarly, Eczema Area and Severity Index (EASI) and itch had significantly higher 1-year response: increasing from 27.8% and 34.7%, respectively, at week 8, 79.6 % achieved EASI-90 and 73.5% reached Peak Pruritus-Numerical Rating Scale 0/1. Minimal disease activity, a stringent composite endpoint, was met by 18 % at week 8 and by 66 % at week 52. Safety was good, with mild adverse eventsmostly including abdominal pain, upper respiratory infections and headache. The 100 mg dose had a significantly better safety profile, with efficacy similar to 200 mg. These data confirm the efficacy and safety of abrocitinib, with early, progressive clinical benefitssustained through 1 year.

Focal facial dermal dysplasias (FFDDs) encompass four rare inherited disorders. FFDD types I, II, and III are characterized by bitemporal scar-like lesions present from birth, while FFDD IV is identified by analogous lesions localized in the periauricular area. Most FFDD IV cases show autosomal-recessive inheritance with mutations in the CYP26C1 gene. We describe three infants with bilateral, oval-shaped, hypopigmented preauricular lesions indicative of FFDD IV. It is crucial for physicians to recognize these rare conditions at an early stage to ensure proper diagnosis and to rule out associated malformations.

Skin cancer incidence is increasing worldwide. Clear communication between dermatologists and histopathologists, along with the possibility of sharing clinical images, is critically important. This survey aims to depict the level of communication between dermatologists and histopathologists in clinical practice in Italy. A group of histopathologists participating in monthly online meetings was recruited to participate in our survey. We collected information regarding dermatologists’ habits in providing or not providing clinical and dermatoscopic images of melanocytic/keratinocytic lesions. A total of 63 histopathologists agreed to participate. Fewer than 15% of histopathologists receive routine clinical or dermatoscopic images from dermatologists after the surgical excision of a melanocytic lesion, while clinical and dermoscopic images of non-melanoma skin cancers (NMSC) are sent in fewer than 10% of cases. Our survey revealed that, in Italy, the communication between dermatologists and pathologists is far from being optimal.

The large language model (LLM) ChatGPT can answer open-ended and complex questions, but its accuracy in providing reliable medical information requires a careful assessment. As part of the AI-CHECK (Artificial Intelligence for CME Health E-learning Contents and Knowledge) study, aimed at evaluating the potential of ChatGPT in continuous medical education (CME), we compared ChatGPT-generated educational content to the recommendations of the National Institute for Health and Care Excellence (NICE) guidelines on acne vulgaris. ChatGPT version 4 was exposed to a 23-item questionnaire developed by an experienced dermatologist. A panel of five dermatologists rated the answers positively in terms of “quality” (87.8%), “readability” (94.8%), “accuracy” (75.7%), “thoroughness” (85.2%), and “consistency” with guidelines (76.8%). The references provided by ChatGPT obtained positive ratings for “pertinence” (94.6%), “relevance” (91.2%), and “update” (62.3%). The internal reproducibility was adequate both for answers (93.5%) and references (67.4%). Answers related to issues of uncertainty and/or controversy in the scientific community scored the lowest. This study underscores the need to develop rigorous evaluation criteria for AI-generated medical content and for expert oversight to ensure accuracy and guideline adherence.