Purpose <p>To evaluate coronary artery calcium (CAC) scores in a large Turkish cohort using age- and sex-specific percentiles, and to assess the agreement between CAD-RADS plaque burden scores and percentile-based CAC classification.</p> Methods <p>A total of 4,887 clinically referred patients who underwent coronary CT angiography (CCTA) with non-contrast imaging were retrospectively evaluated. CAC scores were calculated using the Agatston method. Percentile reference values were generated according to age, sex, and number of cardiovascular risk factors. “Coronary age” was estimated by matching each individual’s CAC score to the 50th percentile value. The agreement between CAD-RADS P scores and percentile-based classification was assessed using Cohen’s kappa.</p> Results <p>Age, male sex, and traditional risk factors (hypertension, diabetes, hyperlipidemia, and smoking) were significantly associated with higher CAC scores (<i>p</i> &lt; 0.01). The likelihood of CAC &gt; 0 increased with each additional risk factor. Among individuals with low to intermediate cardiovascular risk, 22.3% had CAC scores above the 75th percentile. ROC analysis showed strong discriminatory ability for risk factor-based prediction of CAC &gt; 0 (AUC = 0.873). However, agreement between CAD-RADS P scores and percentile-based classification was poor (κ = 0.12, <i>p</i> &lt; 0.01).</p> Conclusion <p>Absolute CAC scores do not reliably reflect relative risk in younger or low-risk individuals. Percentile-based interpretation and coronary age estimation offer a more individualized approach, particularly valuable in low-to-intermediate risk populations.</p> Graphical abstract <p></p>