Systems-based CXCR4-directed imaging of inflammation in patients after myocardial infarction: predictive value for adverse renal outcomes

647 Objectives: In cardiovascular disease, there is tight interaction between the heart and kidneys, which is the foundation of cardiorenal syndrome (CRS), contributes to adverse outcome, and forms a basis for targeted drug intervention. We speculated that whole-body molecular imaging provides insights into inflammatory heart-kidney interaction after myocardial infarction (MI) and predicts adverse renal outcome. Methods: CXCR4-directed PET datasets of 96 acute MI patients following revascularization (median 3.4 days) were evaluated for Ga-68-pentixafor tracer retention. Glomerular filtration rate (GFR) was measured at baseline and follow-up. Renal event (RE) was defined as rapid renal disease progression (GFR loss >5 ml/min/1.73m2 within one year). Results: CXCR4 signal derived from renal parenchyma correlated with the MI territory signal (R=0.25, P<0.05) and remote myocardium signal (R=0.39, P<0.0001), suggesting an inflammatory link between heart and kidneys. At time of PET, GFR did not correlate with CXCR4 signal in any organ (R≤0.04, n.s.). During follow-up, GFR was available in 48/96 subjects (50%) and declined from 83.3±18.2 to 79.8±19.2 ml/min/1.73m2. In 16/48 (33.3%), RE occurred mean 80.5 days post-MI. GFR at baseline (AUC=0.49, p=0.47) and PET-derived signal from the infarct (AUC=0.55, p=0.29) and kidneys (AUC=0.58, p=0.16) failed to be significant for RE prediction in a ROC analysis. Subjects with RE, however, had elevated baseline CXCR4 signal in remote myocardium (SUVpeak, 1.6 ± 0.38 vs 1.37 ± 0.32 for no RE; p=0.02). Ga-68-pentixafor uptake in remote myocardium reached significance in ROC analysis (AUC=0.68, p=0.02) and a cutoff of 1.475 yielded a sensitivity of 75% and specificity of 66%. For patients above the cutoff, the probability for RE was 81.9% (Hazard Ratio, 4.53; p=0.04). Using the ROC-derived cutoff, Kaplan-Meier analysis revealed a significant distinction between high-risk vs. low-risk individuals (p<0.05). Conclusions: After acute MI, inflammatory molecular signatures appear to contribute to CRS development, with a particular role for non-directly injured, remote myocardial involvement. Cardiac Ga-68-pentixafor PET may identify risk of adverse renal outcome after MI, thereby suggesting a potential future role for imaging-guided, renoprotective strategies early after the acute event. Acknowledgement. This project was supported by PRACTIS - Clinician Scientist Program, funded by the German Research Foundation (DFG, ME 3696/3-1) and the clinical research group KFO311.