Leibovit-Reiben, Zachary

Importance Hypopigmented mycosis fungoides (HMF) is uncommon in adults, and data on clinicopathologic features and long-term outcomes remain sparse. Objective To investigate the clinicopathologic features and outcomes of adults with HMF in a large multicenter cohort. Design, Setting, and Participants This retrospective multicenter cohort study conducted from January 2011 to October 2023 at 6 US tertiary referral centers and included adults (18 years or older at diagnosis) with biopsy-confirmed HMF and hypopigmentation documented at diagnosis, with a median (range) follow-up of 39 (1-347) months. Data were analyzed from February to December 2025. Exposures Clinical variant, immunophenotype predominance, and polymerase chain reaction–based T-cell receptor ( TCR ) gene rearrangement in peripheral blood and/or skin. Main Outcomes and Measures Best overall response, defined as complete response, partial response, stable disease, or progressive disease, and progression to a higher stage. Results The cohort included 224 adults with HMF (median [range] age, 44 [18-74] years), of whom 164 (73%) were women (60 men [27%]), 5 (2%) were Asian individuals, 1 (0.4%) was a Hispanic/Latino individual, 1 (0.4%) was a Middle Eastern individual, 19 (8%) were White individuals, and 186 (83%) were Black individuals; 219 (98%) presented with early-stage disease. Most had hypopigmented lesions only (159 [71%]), whereas 65 (29%) exhibited mixed-variant mycosis fungoides; outcomes were similar between groups. Among 141 patients with available immunophenotyping, 88 (63%) had CD8-positive predominance. Among 207 treated patients, most (179 [90%]) received skin-directed therapy alone. Among 198 evaluable patients, the best overall response was complete response for 52 (26%), partial response for 101 (51%), stable disease for 34 (17%), and progressive disease for 11 (6%). Over a median (range) follow-up of 39 (1-347) months, 14 of 224 patients (6%) experienced progression to a higher stage, including 5 of 219 patients (2%) with early-stage disease at baseline who experienced progression to advanced-stage disease. Peripheral blood TCR monoclonality was detected in 23 of 85 tested patients (27%) and was associated with poorer treatment response but not with disease progression. Conclusions and Relevance The finding of this cohort study suggest that adult HMF demonstrated a generally indolent course, with favorable response to skin-directed therapy, irrespective of immunophenotype or mixed-variant presentation. Peripheral blood TCR monoclonality was associated with lower treatment response but not disease progression.

Abstract Lichen Planopilaris (LPP) is a lymphocyte-mediated scarring alopecia characterized by progressive follicular destruction and fibrosis. In this clinical trial, patients with biopsy proven LPP were treated with deucravacitinib (an oral inhibitor of tyrosine kinase 2 (TYK2)) 6 mg BID for 24 weeks (NCT-06091956). Bulk and single-cell RNA sequencing was performed on paired pre- and post-treatment scalp biopsies from baseline and week 4. Patients (N=10) demonstrated improvements in PGA (88.9%, p=0.008), LPPAI (−2.3 points, SD 1.1, p=0.002) and Skindex-16 (−21.0 points, SD 22.1, p=0.014) scores at week 24. Bulk transcriptomic analysis of untreated LPP revealed upregulation of type I Interferon (IFN)-stimulated genes and pathways related to inflammation, immune activation, keratinization, and extracellular matrix remodelling, with downregulation of immune and inflammatory pathways following treatment. Single-cell RNA-seq of LPP was characterized by enrichment of CD8+GZMK+ T cells which showed downregulation of T-cell receptor signaling as well as antiviral pathways with treatment. Basal keratinocytes exhibited reduced cytokine and interferon signaling and decreased communication with NK cells following treatment. CCL19+ fibroblasts were prominent in untreated disease was attenuated after treatment, with downregulation of type I IFN signaling. Selective TYK2 inhibition with deucravacitinib effectively suppresses these inflammatory circuits in LPP and represents a promising therapeutic strategy.

BACKGROUND There is a scarcity of artificial intelligence models trained on frozen pathology. One way to expand the clinical utility of models trained on permanent pathology is by applying them to frozen sections and fine-tune based on weaknesses. OBJECTIVE To qualitatively evaluate a deep learning model trained on permanent pathology to classify squamous cell carcinoma on Mohs surgery frozen sections to learn model shortcomings and inform retraining and fine-tuning. MATERIALS AND METHODS The authors trained a model for classification of tumor on 746 skin biopsy slides and tested it on 15 Mohs surgery frozen sections. The authors estimated performance metrics and compared the regions of interest generated by the model with the original H&E slides. RESULTS The model achieved an AUC-ROC of 0.985 and 0.796 for tumor classification in permanent pathology and in frozen sections, respectively. Regions of interest for frozen sections with scarce tumor areas were inaccurate, focusing on normal tissue for slides classified as false negative, or highlighting structures different from tumor (e.g., inflammation, muscle, and nerves) for slides classified as true positive. CONCLUSION Deep anatomical structures more commonly present in Mohs frozen pathology might represent data out-of-distribution for models trained on permanent pathology, potentially leading to inadequate model outputs.

ImportanceThere exists substantial heterogeneity in outcomes within T stages for patients with cutaneous squamous cell carcinoma (cSCC).ObjectiveTo determine whether a customized generative pretrained transformer model, trained on a comprehensive dataset with more than 1 trillion parameters and equipped with relevant focused context and retrieval augmented generation (RAG), could excel in aggregating and interpreting vast quantities of data to develop a novel class-based risk stratification system that outperforms the current standards.Design, Setting, and ParticipantsTo build the RAG knowledge base, a systematic review of the literature was conducted that addressed risk factors for poor outcomes in cSCC. Using the RAG-enabled generative pretrained transformer (GPT) model, we developed a novel class-based risk stratification system that assigned point values for risk factors, culminating in a GPT-based prognostication system called the artificial intelligence–derived risk score (AIRIS). The system’s performance was validated on a combined prospective and retrospective cohort of 2379 primary cSCC tumors (1996-2023) with at least 36 months of follow-up, against Brigham and Women’s Hospital (BWH) and American Joint Committee on Cancer Staging Manual, eighth edition (AJCC8) systems in stratifying risk for locoregional recurrence (LR), nodal metastasis (NM), distant metastasis (DM), and disease-specific death (DSD).Main Outcomes and MeasuresPerformance metrics evaluated included distinctiveness, homogeneity, and monotonicity, as defined by the AJCC8, as well as sensitivity, specificity, positive predictive value, negative predictive value, accuracy, the area under the receiver operating characteristic curve, and concordance.ResultsThe median age at diagnosis was 73 (IQR, 64-81) years, with 38.5% female patients and 61.5% male patients. The AIRIS prognostication system demonstrated superior sensitivity across all outcomes (LR, 49.1%; NM, 73.7%; DM, 82.5%; and DSD, 72.2%) and the highest area under the receiver operating characteristic curve values (LR, 0.69; NM, 0.81; DM, 0.85; and DSD, 0.80), indicating significantly enhanced discriminative capability compared with the BWH and AJCC8 systems. While all systems were comparably distinctive, the AIRIS prognostication system consistently demonstrated the lowest proportion of tumors exhibiting poor outcomes in low-risk categories, suggesting its improved homogeneity and monotonicity.Conclusions and RelevanceThe results of this diagnostic study suggest that the AIRIS system outperforms the existing BWH and AJCC8 prognostication systems, potentially providing a more effective tool for predicting poor outcomes in cSCC. This study illustrates the potential of large language models in refining prognostic tools, offering implications for treating patients with cancer.

BACKGROUND Cutaneous lichen planus (LP) is a recalcitrant, difficult-to-treat, inflammatory skin disease characterized by pruritic, flat-topped, violaceous papules on the skin. Baricitinib is an oral Janus kinase (JAK) 1/2 inhibitor that interrupts the signaling pathway of IFN-γ, a cytokine implicated in the pathogenesis of LP. METHODS In this phase II trial, 12 patients with cutaneous LP received 2 mg daily baricitinib for 16 weeks, accompanied by in-depth spatial, single-cell, and bulk transcriptomic profiling of pre- and posttreatment samples. RESULTS An early and sustained clinical response was seen, with 83.3% of patients responsive at week 16. Our molecular data identified a unique, oligoclonal IFN-γ, CD8 + , and CXCL13 + cytotoxic T cell population in LP skin and demonstrated a rapid decrease in IFN signature within 2 weeks of treatment, most prominently in the basal layer of the epidermis. CONCLUSION This study demonstrates the efficacy and molecular mechanisms of JAK inhibition in LP. TRIAL REGISTRATION NCT05188521 FUNDING Eli Lilly, Appignani Benefactor Funds, 5P30AR075043, Mayo Clinic Clinical Trials Stimulus Funds.

Abstract Background Prior studies have established an association between a history of abuse and more severe migraine presentation. Objectives This cross‐sectional, observational study of a clinic‐based migraine population used validated measures to elucidate migraine‐specific and migraine‐related burdens among patients with a history of abuse. Methods Patients with migraine ( n = 866) from the American Registry for Migraine Research self‐reported if they had a history of emotional, physical, and/or sexual abuse and completed questionnaires assessing migraine‐related burden: Migraine Disability Assessment, Subjective Cognitive Impairment Scale for Migraine Attacks, Work Productivity and Activity Impairment, Patient‐Reported Outcomes Measurement Information System Pain Interference, Patient Health Questionnaire‐2, and Generalized Anxiety Disorder‐7. Migraine‐related burden in patients with versus without a history of abuse was compared. Subsequently, a mediation analysis evaluated the impact of depression and anxiety symptoms in the relationship between abuse history and migraine burden. Results A history of abuse was reported by 36.5% ( n = 316/866) of participants. After controlling for patient age, sex, years lived with headache, and headache frequency, a history of abuse was significantly associated with more severe migraine‐related disability. The combined burden of depression and anxiety symptoms mediated the relationship. Conclusion A history of abuse is associated with greater migraine‐related disability. Future studies should determine if identification and management of the psychological and physical sequelae of abuse reduce migraine burden.

Abstract Objective To investigate the impact of having headaches prior to traumatic brain injury (TBI) on headache features and long‐term patient health outcomes. Background and Methods This was an exploratory analysis of patients with TBI who were enrolled in the American Registry for Migraine Research (ARMR), a multicenter, prospective, longitudinal patient registry composed of patients with International Classification of Headache Disorders, 3rd edition (ICHD‐3)–defined headache diagnoses. The ARMR study enrolled 2,707 patients between February 1, 2016 and May 6, 2020, 565 of whom qualified for this analysis. Those with headaches prior to their TBI were compared to those without headaches prior to their TBI for ICHD‐3 diagnoses, headache frequency and intensity, headache‐related disability (Migraine Disability Assessment score), symptoms of anxiety (General Anxiety Disorder [GAD‐7]), depression (two items from Patient Health Questionnaire‐9), post‐traumatic stress disorder (PTSD), cutaneous allodynia (12‐item Allodynia Symptom Checklist [ASC‐12]), cognitive dysfunction (Migraine Attacks Subjective Cognitive Impairments Scale [Mig‐SCog]), pain interference (Patient‐Reported Outcomes Measurement Information System‐Pain Interference), and work productivity (Work Productivity and Activity Impairment). Results Among 565 participants with TBI, 350 had headaches prior to their TBI. Those with pre‐TBI headaches were less likely to receive a diagnosis of post‐traumatic headache (PTH; 14/350 [4.0%] vs. 21/215 [9.8%], p = 0.006), even though 25.7% reported new or worsening headaches within 7 days of their TBI. Those with pre‐TBI headaches had higher ASC‐12 scores (2.4 ± 3.5 vs. 1.8 ± 3.4, p = 0.030), Mig‐SCog scores (9.3 ± 4.7 vs. 8 ± 4.9, p = 0.004), and GAD‐7 scores (6.9 ± 5.1 vs. 6.2 ± 5.4, p = 0.039), and were more likely to have a migraine diagnosis (335/350 [95.7%] vs. 192/215 [89.3%], p = 0.003). Conclusions Those with headaches prior to TBI are less likely to receive a diagnosis of PTH. They have more severe symptoms of cutaneous allodynia, cognitive impairment, and generalized anxiety. This analysis suggests that pre‐TBI headaches might impact post‐TBI headache diagnoses and associated features.