Association of HLA-DRB1 alleles with rheumatoid arthritis in Mediterranean populations: A meta-analysis with regional differences, advanced heterogeneity and evidence quality assessment

ObjectiveThis meta-analysis evaluates the association between HLA-DRB1 alleles and rheumatoid arthritis (RA) susceptibility in Mediterranean populations, incorporating regional subgroup analyses, heterogeneity assessment, and evidence quality grading. MethodsA systematic search was conducted across 11 major databases and grey literature sources until December 31, 2023. Eligible studies reported HLA-DRB1 allele frequencies in RA patients and controls from Mediterranean countries. Pooled odds ratios were calculated using random-effects models. Heterogeneity was assessed using Cochranes Q, I{superscript 2}, prediction intervals, cumulative distribution function and meta-regression. Risk of bias was evaluated using the Risk of Bias in Non-randomized Studies--of Exposures (ROBINS-E) tool. Publication bias was examined using funnel-plots, the trim-and-fill method, and Ruckers-test. Sensitivity analyses excluded studies with very-high risk of bias and performed leave-one-out diagnostics. Post hoc statistical power analyses were performed for each allele comparison. Evidence certainty was evaluated using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework. Registered in PROSPERO (CRD42025641117). ResultsForty-eight studies from 13 Mediterranean countries (6,536 RA cases, 10,211 controls) were included. HLA-DRB1*01, *04, *09, and *10 were associated with increased RA risk, with *04 showing the highest pooled effect size (OR = 2.412, 95%CI [2.204-2.641]). HLA-DRB1*03, *07, *08, *11, *12, *13, and *14 were protective, with *13 having the lowest odds ratio (OR = 0.580, 95%CI [0.527-0.638]). Subgroup analyses identified additional risk variants (*01:01, *04:01, *04:04, *04:05, *04:08, and *10:01) and one protective allele (*04:03). No significant regional differences were found between Southern Europe and the Middle East and North Africa. Heterogeneity ranged from moderate to high (I{superscript 2} = 37.9%- 61.1%). Most allele comparisons exhibited sufficient power, although some associations (*03, *08, *09, and *04:03) were unstable in sensitivity analyses. Overall certainty of evidence was rated low to very low. ConclusionsHLA-DRB1*04 and *10 were strongly associated with RA risk, while *13 and *07 were protective. Although HLA-DRB1*09 was significantly associated with RA, it was unstable and supported by very low quality evidence. Despite similar patterns across regions, heterogeneity and study limitations underscore the need for better-controlled, population-specific genetic studies.