Elevated levels of PD-L1 are associated with rs72705424 and increased risk of incident cardiovascular disease in the UK Biobank

Abstract

Immune checkpoint (IC) inhibitors have revolutionised the cancer field by targeting immunomodulatory properties of tumours. However, these targeted therapies are also known for their cardiotoxic effect. Immune checkpoint ligand, programmed death-ligand 1 (PD-L1), specifically, has been implicated in cardiac pathophysiology. Despite this, little is known about the impact of genetic variations on IC signalling and their contribution to cardiovascular disease (CVD). Therefore, we performed a large-scale biobank study examining the effect of a single nucleotide polymorphism (SNP) on protein expression and CVD outcomes. First, 490 541 participants in the UK Biobank were genotyped using the pre-processed whole-genome sequencing files. Carriers of the PD-L1 (CD274) SNP were defined by having either one or two copies of the minor allele of the intronic variant rs72705424 (chr9: 5456083G>A) with a population allele frequency of 2%. The baseline plasma levels were determined through normalised protein expression values. After removing individuals with familial relations and of non-caucasian ethnicity, a 1:1 age- and sex-matched cohort (n=2430) of carriers and non-carriers was created. Associations between rs72705424 and PD-L1 plasma levels at baseline were assessed. Incident CVD (arrhythmia, coronary artery disease, cardiac inflammation and heart failure) was ascertained through the associated first CVD hospital diagnoses and dates. A cox proportional hazard model was fitted to evaluate the association between the scaled plasma PD-L1 levels and incident CVD, adjusted for age, sex, body mass index and c-reactive protein levels. Carriers of the PD-L1 rs72705424 SNP exhibited a significant increase of baseline circulating PD-L1 levels compared to age- and sex-matched non-carriers (0.15 ± 0.39 vs. 0.15 ± 0.37, p = 6.36 × 10^-12) Additionally, among all individuals with PD-L1 plasma measurements without CVD at baseline, elevated PD-L1 levels were associated with an increased risk of incident CVD, with a hazard ratio (HR) of 1.25 (confidence interval (CI): 1.22-1.27, p = 1.03 × 10^-112). After model adjustment, we observed a HR of 1.12 per standard deviation increase in PD-L1 (95% CI: 1.10-1.15, p = 1.30 × 10^-24). These findings indicate that the variant rs72705424 is associated with higher circulating PD-L1 levels and that elevated PD-L1 increases the risk of incident CVD. Collectively, these results highlight how genetic variation can affect circulating factors and the subsequent CVD risks, underscoring the relevance of further investigation.