Key result
GDF-15 predicts 2-year CV death with ~0.83 AUC but shows modest diagnostic value for fCAD.
Why the study?
Rapid and accurate non-invasive diagnosis of functionally relevant CAD is an unmet clinical need. Although growth differentiation factor 15 (GDF-15) may be involved in CAD development, its diagnostic and prognostic value in suspected functionally relevant CAD required evaluation.
Does growth differentiation factor 15 (GDF-15) improve the diagnosis of functionally relevant coronary artery disease and predict all-cause and cardiovascular mortality in patients with suspected fCAD?
Cohort (n=3,379)
Blinded
Does growth differentiation factor 15 (GDF-15) improve the diagnosis of functionally relevant coronary artery disease and predict all-cause and cardiovascular mortality in patients with suspected fCAD?
In patients with suspected functionally relevant coronary artery disease, GDF-15 demonstrates limited diagnostic utility (AUC 0.598) for identifying ischemia, but provides high prognostic value for 2-year all-cause and cardiovascular mortality (AUC 0.815 and 0.834).
GDF-15 should not guide fCAD diagnosis; leaves open its incremental prognostic value in suspected coronary disease.
Introduction The rapid and accurate non-invasive diagnosis of significant and functionally relevant coronary artery disease (fCAD) is a crucial yet unmet clinical need. Growth differentiation factor 15 (GDF-15) may be involved in the development of CAD. Purpose We aimed to evaluate the diagnostic and prognostic value of GDF-15 in patients with suspected fCAD. Methods In this prospective cohort study, consecutive patients with suspected fCAD underwent myocardial perfusion imaging with single-photon-emission computed tomography, and coronary angiography plus fractional flow reserve, when available. The presence of fCAD was centrally adjudicated. GDF-15 concentrations were measured in a blinded fashion. Prognostic endpoints included all-cause mortality and cardiovascular death over five years. Results Among 3,379 patients eligible for this analysis (median age 68 years, 33% female, 44% known CAD), 1,160 patients (34%) had adjudicated fCAD. GDF-15 concentrations were significantly higher in patients with adjudicated fCAD compared to those without (1,325 ng/L (95%CI, 1,258-1,364) versus 1,041 ng/L (95%CI, 1,009-1,074), p<0.001). However, the diagnostic accuracy of GDF-15 for identifying fCAD was modest, with an area under the curve (AUC) of 0.598. When combined with clinical judgment, GDF-15 significantly increased the diagnostic accuracy (AUC 0.659). During a 5-year follow-up, 437 deaths (14%) occurred, of which 231 were cardiovascular. GDF-15 had very high prognostic accuracy for all-cause death and cardiovascular death (e.g. time-dependent AUC 0.815 and 0.834 at 2 years), at least comparable to high-sensitivity cardiac troponin (hs-cTn)T, and higher than hs-cTnI (Figures 1 and 2). Conclusion While GDF-15 has limited diagnostic utility, it exhibits high prognostic value in patients with suspected fCAD.
No takes yet. Share an insight, caveat, or question.
Nafaa et al. (2026) conducted a cohort in Suspected functionally relevant coronary artery disease (fCAD) (n=3,379). Growth differentiation factor 15 (GDF-15) vs. Clinical judgment, hs-cTnT, and hs-cTnI was evaluated on Diagnostic accuracy for functionally relevant CAD and prognostic accuracy for all-cause and cardiovascular mortality (AUC 0.598 (fCAD diagnosis); AUC 0.815 (all-cause death at 2 years); AUC 0.834 (CV death at 2 years), p=<0.001). GDF-15 showed modest diagnostic accuracy for fCAD (AUC 0.598) but high prognostic accuracy at 2 years for all-cause death (AUC 0.815) and cardiovascular death (AUC 0.834).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: