Absolute and Relative Risk of New-Onset Psoriasis Associated With Tumor Necrosis Factor-α Inhibitor Treatment in Patients With Immune-Mediated Inflammatory Diseases: A Danish Nationwide Cohort Study

Exploring the Shared Diagnostic Genes in IBD and Psoriasis through Bioinformatics and Experimental Assays

Background: Inflammatory bowel disease (IBD) is a persistent, non-specific inflammation affecting the intestines. Psoriasis is a long-lasting inflammatory disorder of the skin. There is a comorbidity correlation between IBD and psoriasis, but the specific pathogenesis of the comorbidity is unclear. Materials and methods: In this study, we analyzed datasets sourced from the Gene Expression Omnibus (GEO) database, and identified shared genes of IBD and psoriasis through differential expression analysis and weighted gene co-expression network analysis (WGCNA). Then three machine learning algorithms were applied to identify shared diagnostic genes. Next, the validation of shared diagnostic genes was evaluated with ROC curves, with the AUC determined. Subsequently, single sample gene set enrichment analysis (ssGSEA) and immune infiltration analysis were conducted. Furthermore, we obtained potential drugs such as securinine in the Drug Signature Database (DsigDB) and 7 traditional Chinese medicines in the Coremine database, which might have therapeutic effects on the comorbidity of IBD and psoriasis. Finally, we confirmed the expression of the shared diagnostic gene in colitis and psoriasis mice tissues through RT-PCR, Western blot and immunohistochemistry (IHC) methods. Results: The results showed that AQP9 had the highest diagnostic value for two diseases. AQP9 had AUC values of 93.681% for UC, 89.629% for CD,and 78.689% for psoriasis in the internal validation datasets. In the external validation datasets, AQP9 had AUC values of 90.394% for UC, 93.909% for CD,and 82.906% for psoriasis. Immune infiltration analysis and ssGSEA revealed that AQP9 might impact the disease process of IBD and psoriasis by participating in the NF-kappaB signaling pathway, and modulating immune cell differentiation. Furthermore, the expression levels of AQP9 were consistently validated, showing upregulation in IBD and downregulation in psoriasis, compared to the control group. Conclusions: This study revealed the shared diagnostic genes and potential mechanisms of the comorbidity of IBD and psoriasis, providing new directions for future research on exploring the comorbidity mechanisms and treatment targets.

Algorithmic Identification of Treatment-Emergent Adverse Events From Clinical Notes Using Large Language Models: A Pilot Study in Inflammatory Bowel Disease

Outpatient clinical notes are a rich source of information regarding drug safety. However, data in these notes are currently underutilized for pharmacovigilance due to methodological limitations in text mining. Large language models (LLMs) like Bidirectional Encoder Representations from Transformers (BERT) have shown progress in a range of natural language processing tasks but have not yet been evaluated on adverse event (AE) detection. We adapted a new clinical LLM, University of California - San Francisco (UCSF)-BERT, to identify serious AEs (SAEs) occurring after treatment with a non-steroid immunosuppressant for inflammatory bowel disease (IBD). We compared this model to other language models that have previously been applied to AE detection. We annotated 928 outpatient IBD notes corresponding to 928 individual patients with IBD for all SAE-associated hospitalizations occurring after treatment with a non-steroid immunosuppressant. These notes contained 703 SAEs in total, the most common of which was failure of intended efficacy. Out of eight candidate models, UCSF-BERT achieved the highest numerical performance on identifying drug-SAE pairs from this corpus (accuracy 88-92%, macro F1 61-68%), with 5-10% greater accuracy than previously published models. UCSF-BERT was significantly superior at identifying hospitalization events emergent to medication use (P < 0.01). LLMs like UCSF-BERT achieve numerically superior accuracy on the challenging task of SAE detection from clinical notes compared with prior methods. Future work is needed to adapt this methodology to improve model performance and evaluation using multicenter data and newer architectures like Generative pre-trained transformer (GPT). Our findings support the potential value of using large language models to enhance pharmacovigilance.

Palmoplantar Pustulosis: A Systematic Review of Risk Factors and Therapies

Palmoplantar pustulosis (PPP) is a chronic, relapsing, inflammatory disease that can occur alone or in association with arthritis. There is still controversy about whether it should be separated from psoriasis or classified as pustular psoriasis. Furthermore, drug-induced paradoxical PPP is a special variant of PPP that differs from classic PPP in several ways. Treatment of PPP is still challenging, and there are a number of treatment-resistant cases. This review summarizes the risk factors for the development of PPP and the currently available treatment modalities. Female sex, smokers or ex-smokers, obesity, thyroid dysfunction, and treatment with a tumor necrosis factor (TNF)-α inhibitor have been identified as risk factors for the disease's development, severity, and course. Topical treatments and phototherapy are effective for some patients and are used as a first-line or adjuvant treatment modality. Conventional treatments including retinoids and fumaric acid show good effects and can increase the efficacy of treatment with psoralen + ultraviolet light therapy (PUVA). Ciclosporin is fast acting, but relapse mostly occurs immediately after cessation. TNF-α inhibitors are efficient, and an even better response can be achieved with IL-17 and IL-23 blockers as well as apremilast. The effect of Janus kinase inhibitors seems to be promising according to case reports, but further investigations with larger cohorts are needed.

Algorithmic identification of treatment-emergent adverse events from clinical notes using large language models: a pilot study in inflammatory bowel disease

Background and Aims

Outpatient clinical notes are a rich source of information regarding drug safety. However, data in these notes are currently underutilized for pharmacovigilance due to methodological limitations in text mining. Large language models (LLM) like BERT have shown progress in a range of natural language processing tasks but have not yet been evaluated on adverse event detection.

Methods

We adapted a new clinical LLM, UCSF BERT, to identify serious adverse events (SAEs) occurring after treatment with a non-steroid immunosuppressant for inflammatory bowel disease (IBD). We compared this model to other language models that have previously been applied to AE detection.

Results

We annotated 928 outpatient IBD notes corresponding to 928 individual IBD patients for all SAE-associated hospitalizations occurring after treatment with a non-steroid immunosuppressant. These notes contained 703 SAEs in total, the most common of which was failure of intended efficacy. Out of 8 candidate models, UCSF BERT achieved the highest numerical performance on identifying drug-SAE pairs from this corpus (accuracy 88-92%, macro F1 61-68%), with 5-10% greater accuracy than previously published models. UCSF BERT was significantly superior at identifying hospitalization events emergent to medication use (p < 0.01).

Conclusions

LLMs like UCSF BERT achieve numerically superior accuracy on the challenging task of SAE detection from clinical notes compared to prior methods. Future work is needed to adapt this methodology to improve model performance and evaluation using multi-center data and newer architectures like GPT. Our findings support the potential value of using large language models to enhance pharmacovigilance.