To analyze the association between a disintegrin and metalloproteinase 33 (ADAM33) gene polymorphisms and asthma susceptibility, explore the molecular mechanism by which they mediate airway remodeling, and evaluate the impact of these polymorphisms on the efficacy of allergen immunotherapy (AIT).

PubMed, Embase, Web of Science, Cochrane Library, CNKI, Wanfang, SinoMed, and ClinicalTrials.gov were searched from inception to December 31, 2025, supplemented by manual reference tracing. Inclusion criteria were established according to the PICOS principle: asthma patients and healthy controls, with reported ADAM33 genotype frequencies, airway remodeling phenotypes, or AIT efficacy. Data were extracted, risk of bias was assessed, and case-control matching and genotype quality control were performed. Random-effects model analyses were conducted using Stata 17 (metan, metareg, metainf) and R 4.3 (metafor, forestplot); heterogeneity was assessed by Cochran Q, I^2, tau2, and H^2; leave-one-out sensitivity analysis was performed; and mixed-effects meta-regression was used to explore sources of heterogeneity, reporting regression coefficient beta (95%CI), R^2, and residual I². Bioinformatics annotation (HaploReg, RegulomeDB, GTEx, STRING, miRBase) and eQTL/chromatin interaction analyses were also performed.

A total of 2 847 records were initially retrieved, with 1926 remaining after deduplication; 1 523 reviews and irrelevant studies were excluded, leaving 403, and after full-text screening, 47 studies (43 case-control, 4 prospective) were finally included, comprising 23, 816 asthma cases and 28, 942 healthy controls, including 28 from Asia, 14 from Europe, 4 from North America, and 1 from South America. The pooled OR of the rs3918396 (S1) allele model (A vs. G) was 1.18 (95%CI 1.09~1.28, I²=38%); after removing any single study, the OR ranged from 1.14 to 1.22, with I² decreasing to 19%~29%. rs2280090 (T2), rs2280091 (T1), and rs511898 (F+ 1) were significant in Asian populations and childhood asthma. rs2280090 was associated with annual FEV₁ decline and airway wall thickening. Meta-regression suggested that ethnicity (Asian vs. others) was the main source of heterogeneity (beta=0.21, 95%CI 0.04~0.38, partial R²=31.2%, residual I²=19%). Carriers of ADAM33 risk alleles showed a smaller reduction in asthma exacerbation risk during house dust mite sublingual immunotherapy (P<0.05); risk genotypes were associated with poorer asthma control, lower FEV₁%, and higher sADAM33 levels.

ADAM33 gene polymorphisms, especially the highly linked rs3918396-rs2280090-rs2280091, participate in airway remodeling through soluble ADAM33(sADAM33)-mediated protease activity, airway smooth muscle/fibroblast proliferation, and Th2 inflammation, and represent potential genetic markers for asthma susceptibility and differential AIT efficacy, providing potential molecular targets for personalized asthma treatment.