Cross Platform Blood Transcriptomics Identifies a Two-gene Classifier for Coronary Artery Disease Detection

---

Korkmaz B. E., Korkmaz S.

Namık Kemal Tıp Dergisi, cilt.14, sa.2, ss.160-171, 2026 (ESCI, TRDizin)

• Yayın Türü: Makale / Tam Makale
• Cilt numarası: 14 Sayı: 2
• Basım Tarihi: 2026
• Doi Numarası: 10.4274/nkmj.galenos.2026.65807
• Dergi Adı: Namık Kemal Tıp Dergisi
• Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), TR DİZİN (ULAKBİM)
• Sayfa Sayıları: ss.160-171
• Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
• Trakya Üniversitesi Adresli: Evet

Özet

Aim

Coronary artery disease (CAD) remains a major global health burden, and currently available blood biomarkers lack sensitivity for early detection. This study aimed to identify and validate circulating mRNA and long non-coding RNA (lncRNA) biomarkers of CAD and to develop a predictive transcriptomic model.

Materials and Methods

We analyzed plasma RNA-seq data from stable CAD patients and healthy controls (GSE208194) and validated findings in an independent peripheral blood microarray cohort (GSE113079). Differential expression was assessed using a threshold of |log2 fold-change| ≥1 and false-discovery rate <0.05. Enrichment analysis of gene ontology and Kyoto Encyclopedia of Genes and Genomes pathways was performed. A predictive model was constructed using logistic regression model-penalized logistic regression, with hyperparameters tuned within a nested cross-validation framework, and evaluated in the independent validation cohort.

Results

A total of 182 transcripts (177 mRNAs, 5 lncRNAs) were differentially expressed, with 91% down-regulated in CAD. Enrichment analysis revealed coordinated dysregulation of ribosomal biogenesis, cytoplasmic translation, mitochondrial oxidative phosphorylation, and p53/NF-κB inflammatory pathways. Cross-platform validation confirmed 85 transcripts, indicating a robust expression signature. The predictive model selected two genes, NEUROD2 and RPS27, achieving an external area under the curve of 0.820 in the validation cohort.

Conclusion

Blood transcriptomic profiling identifies a reproducible CAD-associated expression signature and supports a concise two-gene classifier reflecting inflammatory and metabolic stress-related pathways. These findings provide a framework for the further development of blood-based transcriptomic assays and warrant validation in larger, diverse populations to define clinical utility in cardiovascular risk assessment.