Diagnostic yield of molecular testing among individuals with borderline, normal, or low HbA2 in β-thalassemia carrier screening: a systematic review and meta-analysis

Background HbA2 quantification is central to β-thalassemia carrier screening; however, individuals carrying HBB variants may have borderline, normal, or low HbA2 levels. Objective To estimate the molecular diagnostic yield of HBB testing among individuals with borderline, normal, or low HbA2 levels and to summarize additional molecular findings in this population. Methods We conducted a systematic review and single-arm meta-analysis of proportions involving studies that reported molecular testing results in relevant HbA2-defined populations. The primary outcome was the HBB variant detection rate. Logit-transformed proportions were pooled using random-effects models. Leave-one-out sensitivity analyses and exploratory assessments using funnel plots, Egger's regression test, Begg's rank correlation test, and the trim-and-fill method were also performed. All analyses were conducted using R version 4.6.0. Results Nineteen studies comprising 5,987 eligible individuals contributed to the primary meta-analysis; 1,023 individuals were HBB -positive. The pooled HBB variant detection rate was 15.0% (95% CI, 9.3%–23.3%), with substantial heterogeneity (I 2 = 97.4%) and a 95% prediction interval of 1.4%–67.9%. Between-subgroup differences were not statistically significant for screening setting (Q-between = 7.81, P = 0.0502) or molecular testing strategy (Q-between = 2.61, P = 0.271). Leave-one-out estimates ranged from 13.2% to 16.4%. Secondary analyses estimated detection rates of 31.1% (95% CI, 19.2%–46.3%) for any reported molecular defect and 17.6% (95% CI, 10.6%–27.8%) for α-thalassemia. HBD or δ-globin variants, KLF1 variants, and α-globin triplication were also identified. Exploratory tests did not provide clear evidence of small-study effects. Conclusions Molecular testing identified HBB variants in a clinically meaningful proportion of individuals with borderline, normal, or low HbA2 levels, although between-study heterogeneity was substantial. These findings support the use of molecular testing as a complementary, risk-adapted strategy for individuals with residual carrier risk rather than universal testing of all individuals with normal HbA2 levels. The pooled estimate should therefore be interpreted in the context of the clinical population rather than as a population-wide detection probability. Systematic Review Registration https://www.crd.york.ac.uk/prospero/ , identifier CRD420261434640.