Background: Pharmacogenomic variation contributes to interindividual differences in drug response, yet population-specific pharmacogenomic profiles remain insufficiently characterized in the Western Balkans (WB). This study aimed to comprehensively characterize pharmacogenomic variability in the WB populations using whole-genome sequencing (WGS), assess population differences in pharmacogenomic profiles, evaluate concordance among star allele calling tools, and identify rare and previously unreported variants in pharmacogenes. Methods: WGS data from 250 healthy individuals from five WB populations were analyzed across 88 pharmacogenes. Star alleles were assigned using Aldy4, PharmCAT, PyPGx, and Stargazer. Principal component analysis (PCA) was used to assess population structure, and pharmacogenomic allele and phenotype frequencies were compared within five European subpopulations. Unreported pharmacovariants were annotated using Ensembl VEP, gnomAD, ClinVar, and ClinPGx, with functional impact assessed using multiple in silico prediction tools. Results: A total of 357 star alleles were identified, with high tool concordance except for complex genes such as CYP2D6. WB populations clustered within the European genetic space, with limited subpopulation differentiation. Significant differences were observed within WB and compared with European populations, including clinically relevant genes CYP2B6, CYP2C19, RYR1, and VKORC1. Among 33 clinically relevant pharmacogenes, 22,638 SNVs were identified, including six previously unreported variants. Conclusions: The WB populations show substantial pharmacogenomic similarity to other European populations, while also exhibiting clinically relevant population-specific differences. Comprehensive WGS-based profiling revealed both established and previously unreported pharmacogenomic variation, highlighting the importance of including underrepresented populations in precision medicine research.

Background/Objectives: Acute myeloid leukemia is a genetically diverse hematological malignancy where patient outcomes vary significantly. Long non-coding RNA (lncRNA) GAS5 acts as a tumor suppressor and is frequently downregulated in various cancers, as well as in AML. In the current study, we aimed to explore the effects of GAS5 promoter variants on its expression levels in AML patients, their prognostic significance, and to investigate their functional effects. Methods: The GAS5 promoter region containing rs55829688 and rs145204276 was sequenced in 75 AML patients. Statistical analyses were performed to assess their associations with GAS5 expression and outcomes. An in vitro functional study in K562 cells evaluated the effects of these variants on the transcriptional activity of constructs containing each variant. In silico analysis was used to predict changes to transcription factor binding sites. Results: Patients carrying the rs55829688 TC/CC genotype exhibited lower GAS5 expression and were more frequently categorized into the adverse risk group. In intermediate-risk patients, this genotype trended toward lower overall survival and higher bone marrow blast percentages. In vitro, the construct harboring the rs55829688 C allele showed a two-fold decrease in reporter gene activity compared to the construct bearing both wild type alleles. In silico analysis identified RUNX3 as the most likely transcription factor affected by this variant. The variant rs145204276 was considered for the first time in AML; however, no significant clinical associations or transcriptional effects were found. Conclusions: Taken together, our findings provide evidence that the rs55829688 promoter variant reduces GAS5 expression in AML and could potentially be a prognostic marker.