A genome- and phenome-wide association study of plasma procalcitonin concentrations in individuals of European ancestry

TL;DR

This study identifies genetic variants linked to plasma procalcitonin levels and finds associations with traits related to calcium metabolism, bone health, and immune function.

Contribution

The study provides new insights into the genetic basis of procalcitonin and its links to various disease-related traits through genome-wide and phenome-wide analyses.

Findings

Four independent SNPs in three loci (CALCB, PBX4, PRDM15) were associated with plasma procalcitonin.

PRS for PCT was linked to calcium metabolism, vitamin D levels, bone fractures, and immune-related traits.

Genetically predicted PCT is associated with pathways involving bone health, kidney function, and type 2 diabetes.

Abstract

Procalcitonin (PCT) is a biomarker used to differentiate between viral and bacterial infections, though the underlying mechanisms are not yet fully understood. This study aimed to identify genetic variants associated with plasma PCT concentrations and explore the associations of genetically predicted PCT with a wide range of disease related traits in a PheWAS.

We conducted GWAS and meta-analysis using data from the MDCS (n = 4007), MPP (n = 5097), and PREVEND (n = 3344) cohorts. We used fine-mapping to prioritise likely causal variants and explored regulatory effects using eQTL data, summary-data-based Mendelian randomisation (SMR) and colocalisation. To validate the PCT findings, we conducted multi-trait analysis of GWAS (MTAG) combining our results with CALCA data from a large pQTL study. The polygenic risk score (PRS) for PCT was calculated in the UK Biobank (n = 457,418) based on the GWAS summary data, and associations between the PRS and 179 traits were assessed in a PheWAS.

We identified four independent significant SNPs in three loci associated with plasma PCT: CALCB (rs7119706, rs10832337), PBX4 (rs17217098), and PRDM15 (rs7277773). Fine-mapping prioritised 18 likely causal variants, including rs7119706 (near CALCB) and rs16930609 (mapped to CYP2R1) at the chromosome 11 locus. Our eQTL lookup identified significant results for 13 genes, but SMR and colocalisation analyses did not support their potentially causal effects on plasma PCT. The MTAG identified 28 additional significant SNPs across 14 loci. The PheWAS results revealed that PRS was associated with calcium metabolism-related traits, including calcium concentrations (p = 7.0 × 10−5), vitamin D concentrations (p = 2.0 × 10−219), and bone fractures (p = 6.5 × 10−4); metabolic traits, cardiovascular, renal, and liver function-related traits, and inflammation and immune-related traits.

Our findings suggest that genetically predicted PCT is associated with multiple pathways including calcium metabolism and immune function, and has potential clinical implications for bone health, kidney function, and type 2 diabetes.

China Scholarship Council (File no. 202006210041 to WZ and 201906010319 to SW, respectively).

Figures