CUMULATIVE GENETIC RISK SCORE FOR PREDICTING MANIFEST PERSISTENT CYTOMEGALOVIRUS INFECTION, A FIVE-LOCUS MODEL IN THE FERGANA VALLEY POPULATION

Abstract

Objective: to develop and evaluate a cumulative genetic risk score based on five single nucleotide polymorphisms (SNPs) for predicting the manifest persistent form of cytomegalovirus infection (CMVI). Materials and methods. Genotyping for TNF-α (G308A), IL-10 (G>A), TLR4 (A896G), eNOS (G894T), and SOD2 (Ala16Val) was performed in 100 CMVI patients (Group I, n=42, latent; Group II, n=58, manifest persistent) and 80 healthy controls using real-time PCR with TaqMan probes (CFX96, Bio-Rad). A cumulative genetic risk score (0–10 points) was calculated by assigning 0 points for each protective homozygous genotype, 1 for heterozygous, and 2 for risk homozygous per locus. ROC analysis was used to evaluate diagnostic performance. Results. A high genetic score (≥6) was found in 65.5% of patients with the manifest form and only 16.7% with the latent form (χ²=24.12; p<0.001; OR=9.64; 95% CI 3.58–25.9). At the threshold of ≥6, sensitivity was 65.5%, specificity 83.3%, and AUC=0.78 (95% CI 0.69–0.87). Combining the genetic score with viral load (>10³ copies/mL) increased AUC to 0.85 (95% CI 0.77–0.93), specificity to 90.5%, and positive predictive value to 90.7%. No individual polymorphism achieved comparable diagnostic accuracy (individual AUC range: 0.61–0.68). Conclusion. The five-locus cumulative genetic risk score provides clinically meaningful stratification of CMVI patients. The combined model integrating genetic and virological data achieves high diagnostic accuracy and may serve as a practical tool for personalized management.