Search

Human humoral immune responses to SARS-CoV-2 vaccines exhibit substantial inter-individual variability and have been linked to vaccine efficacy. To elucidate the underlying mechanism behind this variability, we conducted a genome-wide association study (GWAS) on the anti-spike IgG serostatus of UK Biobank participants who were previously uninfected by SARS-CoV-2 and had received either the first dose (n = 54,066) or the second dose (n = 46,232) of COVID-19 vaccines. Our analysis revealed significant genome-wide associations between the IgG antibody serostatus following the initial vaccine and human leukocyte antigen (HLA) class II alleles. Specifically, the HLA-DRB1∗13:02 allele (MAF = 4.0%, OR = 0.75, p = 2.34e-16) demonstrated the most statistically significant protective effect against IgG seronegativity. This protective effect was driven by an alteration from arginine (Arg) to glutamic acid (Glu) at position 71 on HLA-DRβ1 (p = 1.88e-25), leading to a change in the electrostatic potential of pocket 4 of the peptide binding groove. Notably, the impact of HLA alleles on IgG responses was cell type specific, and we observed a shared genetic predisposition between IgG status and susceptibility/severity of COVID-19. These results were replicated within independent cohorts where IgG serostatus was assayed by two different antibody serology tests. Our findings provide insights into the biological mechanism underlying individual variation in responses to COVID-19 vaccines and highlight the need to consider the influence of constitutive genetics when designing vaccination strategies for optimizing protection and control of infectious disease across diverse populations.

Objective: Anterior uveitis is a common extra‐articular manifestation of axial spondyloarthritis (AxSpA). We set to evaluate the risk of anterior uveitis (AU) with biologics and synthetic disease‐modifying drugs in AxSpA. Methods: We conducted a systematic review and meta‐analysis to identify phase II/III double‐blinded randomized controlled trials of anti–tumor necrosis factor (TNF) monoclonal antibodies (mAb), anti–interleukin‐17 (anti–IL‐17), and Janus kinase inhibitors (JAKi) in AxSpA. Patient‐exposure years (PEY) were calculated using the per‐protocol approach. Incidence rate (IR) of AU/100 person‐years were calculated by treatment group using the random effects approach. Network meta‐analysis (NMA) was used to estimate risk of AU in treatment groups, expressed as IR ratios (IRRs). Bias was assessed using the Cochrane Risk of Bias‐2 tool. Results: Forty‐four trials were included: 17 anti‐TNF mAb (1,004 PEY), 9 etanercept (180 PEY), 13 anti–IL‐17 (1,834 PEY), and 6 JAKi (331 PEY). The IR of AU were as follows for anti‐TNF mAb: 4.1, 95% confidence interval (CI) 0–8.5; etanercept: 5.4, 95% CI 0–16.0; anti–IL‐17: 2.8, 95% CI 1.6–4.1; JAKi: 1.5, 95% CI 0.0–3.0; and placebo: 10.8, 95% CI 7.4–14.1. In NMA, IRRs of treatments compared with placebo were as follows for anti‐TNF mAb: 0.32, 95% CI 0.10–1.04; etanercept 0.42, 95% CI 0.08–2.38; anti–IL‐17: 0.43, 95% CI 0.19–0.98; and JAKi: 0.32, 95% CI 0.06–1.67. Comparisons between anti‐TNF mAb, anti–IL‐17, and JAKi did not demonstrate any significant difference in AU risk. Using the surface under the cumulative ranking curve approach to rank AU risk, anti‐TNF mAbs were associated with the lowest risk followed by JAKi, anti–IL‐17, and etanercept. All treatments were ranked superior to placebo. Conclusion: Anti‐TNF mAbs, JAKi, and anti–IL‐17 appear protective against AU events in individuals with AxSpA, with no significant differences in risk of AU between treatments. [ABSTRACT FROM AUTHOR]

Objectives To evaluate trust-level performance in time to initiation of DMARD therapy in patients with early inflammatory arthritis (EIA), with identification of the change in performance trajectories over time and investigation of trust characteristics associated with this change. Methods We included 130 trusts from the UK contributing to the National Early Inflammatory Arthritis Audit (NEIAA) from 2018 to 2020. The primary outcome was days from referral to initiation of DMARD therapy in patients with EIA. Latent class growth mixture models were applied to identify distinct groups of trusts with similar trajectories of performance change over time. We used mixed effects linear and multinomial logistic regression models to evaluate the association between delay in treatment and trust-level characteristics. Results The mean time to DMARD initiation was 53 days (s. d. 18), with an average 0.3-day decrease with each month over time. Four latent trajectories were identified in our cohort, with >77% of individual trusts showing ongoing improvements in decreasing treatment waiting times. Prior to separating by latent class, time to DMARD initiation was shorter in trusts with higher rheumatology staffing, a local EIA treatment pathway and those with access to musculoskeletal ultrasound. Trusts with more nurses in the rheumatology department were less likely to be in the worst performance group [odds ratio 0.69 (95% CI 0.49, 0.93)]. Conclusion In this cohort study, we observed a reduction in treatment waiting time over time. Trusts with better staffed and improved EIA clinical structure are likely to initiate definitive treatment earlier in patients with EIA. [ABSTRACT FROM AUTHOR]

To address the current objective of short‐term traffic flow prediction on roads, this paper aims to predict not only the future traffic flow on roads, but also the average speed of future roads. Specifically, the authors propose a short‐term traffic flow prediction method based on the Convolutional Neural Network—Gate Recurrent Unit—Multi Task Learning‐time‐separated attention mechanism (AST‐MTL‐CNN‐GRU). Firstly, the authors add a time‐separated attention mechanism that can effectively generate attention matrices independently for different locations and times of the day. Secondly, the three main parameters of traffic flow, average speed and occupancy are combined in pairs as input data while the multi‐task learning is applied to predict the future traffic flow and average speed on the road. Finally, the CNNs are used to extract the spatial features and the GRUs are used to extract temporal features. Compared with the baseline models (diffusion CNNs (DCRNN), 1D‐CNN‐long short‐term memory (1DCNN‐LSTM), convolutional LSTM (ConvLSTM), graph WaveNet and ST‐ResNet), the authors' proposed model has high performance in terms of accuracy, stability and robustness. A large number of ablation experiments are conducted based on five‐fold cross validation using the Quanzhou city dataset and the California dataset. The experimental results show that the combination of average speed and occupancy for input has higher prediction accuracy. In addition, the authors' proposed model can outperform the other models being compared in terms of accuracy, stability and robustness even with noise added. [ABSTRACT FROM AUTHOR]