C-X-C motif chemokine receptor 4-directed PET signal in the arterial tree is not consistently linked to calcified plaque burden and cardiovascular risk

Abstract

Purpose: To establish the extent, distribution and frequency of in-vivo vessel wall [⁶⁸Ga]Ga-PentixaFor uptake and to determine its relationship with calcified atherosclerotic plaque burden (CAP) and cardiovascular risk factors (CVRF). Methods: 65 oncological patients undergoing [⁶⁸Ga]Ga-PentixaFor PET/CT were assessed. Radiotracer uptake (target-to-background ratio [TBR]) and CAP burden (including number of CAP sites, calcification circumference and thickness) in seven major vessel segments per patient were determined. We then investigated associations of vessel wall uptake with CAP burden, cardiovascular risk (CVRF and European Society of Cardiology [ESC] SCORE2/SCORE2-OP risk chart) and image noise (determined by coefficient of variation [CoV] from unaffected liver parenchyma). Results: We identified 1292 sites of high focal [⁶⁸Ga]Ga-PentixaFor uptake (PentixaFor+ sites) in the vessel wall in 65/65 (100%) patients, with concomitant calcification in 385/1292 (29.8%) sites. There were no significant associations between vessel wall uptake and CAP burden (number of PentixaFor+ sites: r ≤ 0.18, P ≥ 0.14; PentixaFor+ TBR: r ≤ 0.08, P ≥ 0.54). The number of PentixaFor+ sites showed a moderate correlation with cardiovascular risk (ESC SCORE2/SCORE2-OP, r = 0.30; number of CVRF, r = 0.26; P = 0.04, respectively), but failed to reach significance for PentixaFor+ TBR (r ≤ 0.18, P ≥ 0.22). In univariable regression analysis, body mass index (odds ratio [OR] 1.08, 95%-confidence interval [CI] 1.02-1.14) and CoV (OR, 1.07; CI, 1.05-1.10) were linked to TBR and the number of PentixaFor+ sites (P < 0.01, respectively), while injected activity was only associated with the latter imaging parameter (OR, 0.99; CI, 0.98-1.00; P = 0.04). In multivariable regression, injected activity (OR, 1.00; CI, 0.99-1.00) and CoV (OR, 1.06; CI, 1.06-1.07) remained significantly associated with the number of PentixaFor+ sites (P < 0.01, respectively). CoV, however, was the only parameter significantly linked to PentixaFor+ TBR on multivariable analysis (OR, 1.02; CI, 1.01-1.03; P < 0.01). Conclusion: On a visual and quantitative level, high focal [⁶⁸Ga]Ga-PentixaFor uptake in the arterial tree was not consistently linked to vessel wall calcification or cardiovascular risk. Image noise, however, may account for a substantial portion of apparent vessel wall uptake.

Keywords: C-X-C motif chemokine receptor 4; CXCR4; [68Ga]Ga-PentixaFor; atherosclerosis; cardiovascular risk factors; molecular imaging.

Figure 1

[⁶⁸Ga]Ga-PentixaFor PET/CT in a 89-year old man showing a fused parasagittal image reconstruction of the aortic arch and the descending thoracic aorta (A) and three paraaxial slices (B trough D) at levels indicated by the blue dotted line and arrows. In (B) arterial wall radiotracer uptake is partially co-localized with calcification, in (C) vessel wall calcification without radiotracer uptake is shown, and in (D) vessel wall radiotracer uptake without corresponding calcification and calcification without co-localized radiotracer uptake is visible.

Figure 2

Scatter plots with number of PentixaFor+ sites (A), target-to-background ratio (TBR) in those sites (B) and the number of calcified atherosclerotic plaques (CAP, C). Parameters are separated based on the number of cardiovascular risk factors (CVRF). Median is represented by the horizontal line. Numbers in every subgroup vary based on available PET uptake (number of Pentixafor+ sites and TBR) or CAP sites. ns = not significant

Figure 3

Scatter plots showing associations between the ESC SCORE/SCORE2-OP risk chart (A, C, E, G, I) and the number of cardiovascular risk factors (CVRF) (B, D, F, H, J) with the number of PentixaFor+ sites (A, B), the mean target-to-background-ratio (TBR) in PentixaFor+ sites (C, D), the number of calcified atherosclerotic plaques (CAP) (E, F), the cumulative calcification circumference (G, H), and the cumulative CAP thickness (I, J). There is a weak correlation between the number of PentixaFor+ sites and markers of cardiovascular risk (A, B), and a strong correlation between indicators of atherosclerotic calcification and cardiovascular risk (E trough J).

Figure 4

Scatter plots demonstrating associations between image noise determined by coefficient of variation (CoV) within normal liver parenchyma and injected activity (A), body mass index (BMI) (B), the number of PentixaFor+ sites (C), and PentixaFor+ site target-to-background ratio (TBR) (D). Lower injected activities and higher BMI were associated with higher image noise. In addition, more PentixaFor+ sites and higher TBR were associated with higher levels of image noise.