| Popis: |
Background/Objectives: Ages of 90.0 and 95.0 are widely considered as thresholds for longevity and extreme longevity, respectively. The goal of this study was to construct unweighted and weighted Genetic Longevity Scores (uGLS, wGLS) and to test their ability to predict a chance of survival beyond these ages. Methods: DNA samples of 314 unrelated elderly individuals (85.0+ years) were genotyped for 42 putative longevity SNPs in 27 genes, and all SNPs that had a p< ; 0.20 in univariate analyses were selected for testing in the multivariate logistic regression models. SNPs from the best model for each cut-off age at death were chosen for GLS calculation, with unweighted score representing a sum of longevity-related alleles, and weighted score a sum of each SNP’s value multiplied by its respective beta coefficient from the multivariate model. Results: GLS included nine SNPs (rs7412, rs50871, rs12206094, rs2267723, rs9536314, rs16847897, rs1800629, rs1042522, rs17202060) for the survival age of 90.0, and five SNPs (rs429358, rs12203592, rs4837525, rs6067484, rs1042522) for the age of 95.0. There was no significant sex difference in mean values of any GLS. All four GLSs (uGLS90.0, wGLS90.0, uGLS95.0, wGLS95.0) were positively correlated with age at death (p< ; 0.01). ROC curve analysis showed all four scores are predictive for reaching the longevity milestones. With area-under-curve of 0.690, weighted GLS90 was shown to be the most predictive. Conclusion: GLS based on nine longevity SNPs is predictive of survival to 90.0 years, and five SNP-based GLS is predictive of reaching 95.0 years in the elderly Croatian population. References: / Grants: CSF IP-01-2018-2497 (HECUBA). |
