Sortilin as a Biomarker for Cardiovascular Disease Revisited

Peter Loof Møller; Palle D Rohde; Simon Winther; Peter Breining; Louise Nissen; Anders Nykjaer; Morten Bøttcher; Mette Nyegaard; Mads Kjolby

doi:10.3389/fcvm.2021.652584

Sortilin as a Biomarker for Cardiovascular Disease Revisited

Front Cardiovasc Med. 2021 Apr 16:8:652584. doi: 10.3389/fcvm.2021.652584. eCollection 2021.

Authors

Peter Loof Møller¹, Palle D Rohde², Simon Winther³, Peter Breining^{1

4}, Louise Nissen³, Anders Nykjaer^{1

4}, Morten Bøttcher³, Mette Nyegaard^{1

5}, Mads Kjolby^{1

4

6

7}

Affiliations

¹ Department of Biomedicine, Aarhus University, Aarhus, Denmark.
² Department of Chemistry and Bioscience, Aalborg University, Aalborg, Denmark.
³ Department of Cardiology, Gødstrup Hospital, NIDO, Herning, Denmark.
⁴ PROMEMO and DANDRITE, Aarhus University, Aarhus, Denmark.
⁵ Department of Health Science and Technology, Aalborg University, Aalborg, Denmark.
⁶ Department of Clinical Pharmacology, Aarhus University Hospital, Aarhus, Denmark.
⁷ Steno Diabetes Center Aarhus, Aarhus University Hospital, Aarhus, Denmark.

Abstract

Genetic variants in the genomic region containing SORT1 (encoding the protein sortilin) are strongly associated with cholesterol levels and the risk of coronary artery disease (CAD). Circulating sortilin has therefore been proposed as a potential biomarker for cardiovascular disease. Multiple studies have reported association between plasma sortilin levels and cardiovascular outcomes. However, the findings are not consistent across studies, and most studies have small sample sizes. The aim of this study was to evaluate sortilin as a biomarker for CAD in a well-characterized cohort with symptoms suggestive of CAD. In total, we enrolled 1,173 patients with suspected stable CAD referred to coronary computed tomography angiography. Sortilin was measured in plasma using two different technologies for quantifying circulating sortilin: a custom-made enzyme-linked immunosorbent assay (ELISA) and OLINK Cardiovascular Panel II. We found a relative poor correlation between the two methods (correlation coefficient = 0.21). In addition, genotyping and whole-genome sequencing were performed on all patients. By whole-genome regression analysis of sortilin levels measured with ELISA and OLINK, two independent cis protein quantitative trait loci (pQTL) on chromosome 1p13.3 were identified, with one of them being a well-established risk locus for CAD. Incorporating rare genetic variants from whole-genome sequence data did not identify any additional pQTLs for plasma sortilin. None of the traditional CAD risk factors, such as sex, age, smoking, and statin use, were associated with plasma sortilin levels. Furthermore, there was no association between circulating sortilin levels and coronary artery calcium score (CACS) or disease severity. Sortilin did not improve discrimination of obstructive CAD, when added to a clinical pretest probability (PTP) model for CAD. Overall, our results indicate that studies using different methodologies for measuring circulating sortilin should be compared with caution. In conclusion, the well-known SORT1 risk locus for CAD is linked to lower sortilin levels in circulation, measured with ELISA; however, the effect sizes are too small for sortilin to be a useful biomarker for CAD in a clinical setting of low- to intermediate-risk chest-pain patients.

Keywords: Dan-NICAD; OLINK; SORT1; cardiovascular disease; pQTL; protein biomarkers; sortilin.