Genetic Susceptibility Within the Immune System Beyond Conventional Risk Factors: An Integrative Analysis in a Kazakh Cohort

Background/Objectives: Genetic factors are known to influence overall immune vulnerability to diseases, such as tuberculosis (TB); however, isolating the genetic component can be challenging as social and environmental causes, such as smoking, alcohol consumption, and unemployment, also drive disease risk. To reduce heterogeneity related to these factors, we performed a targeted immunogenetic association study of TB susceptibility in a Kazakh case–control cohort selected among individuals without the typical socioeconomic and nutritional risk factors linked to TB. Methods: We performed genotyping using an immune-focused targeted array; after quality control at sample and variant level, removal of ancestry-based outliers, and adjustment for population stratification, the final dataset included 164 participants (84 cases, 80 controls) and 181,595 variants. We performed four complementary analyses: single-variant association testing using logistic regression with Firth correction where required adjusted for sex and genetic ancestry, gene-based aggregation testing, functional annotation, and polygenic risk scoring using existing multi-ancestry TB genetic resources. Results: While no variant reached formal significance, the top nominal signal, rs1560036 (OR = 2.98, p = 2.4 × 10−5), lies within CPLX2, which ranked among the top genes in gene-based analysis, a result driven largely by the same variant. A nominally associated variant in the classical MHC region, rs9263600, is separately reported in GTEx as a cis-eQTL for HLA-B and HLA-C. Similarly, polygenic risk scores showed directionally consistent but statistically non-significant associations with case status (Asia-base AUC = 0.560; Europe-base AUC = 0.537). Conclusions: Although our study was not adequately powered to detect small-to-moderate genetic effects, it offers an important perspective on TB genetics in a population underrepresented in existing research. These findings provide an openly available dataset and candidate signals for replication in larger, independent cohorts.