Search

Background: Myasthenia gravis (MG) is an autoimmune disease characterized by muscle weakness and fatigability. The fluctuating nature of the disease course impedes the clinical management. Objective: The purpose of the study was to establish and validate a machine learning (ML)–based model for predicting the short-term clinical outcome in MG patients with different antibody types. Methods: We studied 890 MG patients who had regular follow-ups at 11 tertiary centers in China from 1 January 2015 to 31 July 2021 (653 patients for derivation and 237 for validation). The short-term outcome was the modified post-intervention status (PIS) at a 6-month visit. A two-step variable screening was used to determine the factors for model construction and 14 ML algorithms were used for model optimisation. Results: The derivation cohort included 653 patients from Huashan hospital [age 44.24 (17.22) years, female 57.6%, generalized MG 73.5%], and the validation cohort included 237 patients from 10 independent centers [age 44.24 (17.22) years, female 55.0%, generalized MG 81.2%]. The ML model identified patients who were improved with an area under the receiver operating characteristic curve (AUC) of 0.91 [0.89–0.93], ‘Unchanged’ 0.89 [0.87–0.91], and ‘Worse’ 0.89 [0.85–0.92] in the derivation cohort, whereas identified patients who were improved with an AUC of 0.84 [0.79–0.89], ‘Unchanged’ 0.74 [0.67–0.82], and ‘Worse’ 0.79 [0.70–0.88] in the validation cohort. Both datasets presented a good calibration ability by fitting the expectation slopes. The model is finally explained by 25 simple predictors and transferred to a feasible web tool for an initial assessment. Conclusion: The explainable, ML-based predictive model can aid in forecasting the short-term outcome for MG with good accuracy in clinical practice.

Background and objectivesMyasthenia gravis (MG) is a T cell-driven, autoantibody-mediated disorder affecting transmission in neuromuscular junctions. The associations between the peripheral T cells and MG have been extensively studied. However, they are mainly of observational nature, thus limiting our understanding of the effect of inflammatory biomarkers on MG risk. With large data sets now available, we used Mendelian randomization (MR) analysis to investigate whether the biomarkers on T cells are causally associated with MG and further validate the relationships.MethodsWe performed a two-sample MR analysis using genetic data from one genome-wide association study (GWAS) for 210 extensive T-cell traits in 3,757 general population individuals and the largest GWAS for MG currently available (1,873 patients versus 36,370 age/gender-matched controls) from US and Italy. Then the biomarkers of interest were validated separately in two GWASs for MG in FIN biobank (232 patients versus 217,056 controls) and UK biobank (152 patients versus 386,631 controls).ResultsIn the first analysis, three T-cell traits were identified to be causally protective for MG risk: 1) CD8 on terminally differentiated CD8+ T cells (OR [95% CI] = 0.71 [0.59, 0.86], P = 5.62e-04, adjusted P =2.81e-02); 2) CD4+ regulatory T proportion in T cells (OR [95% CI] = 0.44 [0.26, 0.72], P = 1.30e-03, adjusted P =2.81e-02); 3) HVEM expression on total T cells (OR [95% CI] = 0.67 [0.52, 0.86], P = 1.61e-03, adjusted P =2.81e-02) and other eight T-cell subtypes (e.g., naïve CD4+ T cells). In particular, HVEM is a novel immune checkpoint on T cells that has never been linked to MG before. The SNPs on the TNFRSF14 per se further support a more direct link between the HVEM and MG. The validation analysis replicated these results in both FIN and UK biobanks. Both datasets showed a concordant protective trend supporting the findings, albeit not significant.ConclusionThis study highlighted the role of HVEM on T cells as a novel molecular-modified factor for MG risk and validated the causality between T cells and MG. These findings may advance our understanding of MG’s immunopathology and facilitate the future development of predictive disease-relevant biomarkers.

Objective: To determine prevalence and factors associated with neuropathic pain symptoms in a multiethnic cohort of Malaysian adult diabetic patients. Methods: This was aprospective cross-sectional observational study of hospital-based diabetic outpatients in Malaysia. Subjects were interviewed for their demographic data and medical history. The painDETECT questionnaire was used to screen for neuropathic pain symptoms and pain intensity was assessed using the numeric pain rating scale (NPRS). Neuropathy symptoms and signs were assessed using the Neuropathy Symptom Score (NSS) and Neuropathy Disability Score (NDS). Results: Of 242 patients, 140 (58%) were women, with a mean age of 61 ± 11.4 years (range 21 to 81). Ninety nine(40.9%) were Malay, 64 (26.4%) Chinese, 76 (31.4%) Indian and three (1.2%) were Eurasian. Mean duration of diabetes was 15.9± 9.8 years (range 1 to 53) and 232 (95.9%) patients had Type II diabetes. Peripheral neuropathy,based on NSS and NDS criteria, was found in 83 (34.3%). Thirteen (5.4%) patients were found to likely have neuropathic pain symptoms and this was independently associated with peripheral neuropathy ((OR) = 3.40, 95% confidence interval (CI): 1.04, 11.14) and Indian ethnicity (OR = 5.44, 95% CI: 1.50, 19.57)). Patients with neuropathic pain had higher average pain intensity scores. Conclusions: The prevalence of neuropathic pain symptoms in a Malaysian DM patient cohort was low and was associated with the severity of neuropathy symptoms and Indian ethnicity. The causes for ethnic differences are unknown and could be due socio-cultural or physiological differences in neuropathic pain perception. [ABSTRACT FROM AUTHOR]