NEW YORK — Using a multiomics and multi-trait approach, an international team of researchers has prioritized 30 potential therapeutic targets for dyslipidemia, which they believe are more likely to succeed in clinical trials than other drug targets.
In a study published in Cell Reports Medicine on Wednesday, scientists at the Samsung Advanced Institute for Health Sciences & Technology in South Korea and elsewhere investigated over 15,000 genes to find those that control both lipid levels and lipid-related traits, using multi-trait transcriptome-wide Mendelian randomization (MR) analysis.
"We introduce a genetic-driven approach based on causal inferences that can inform drug target prioritization, repurposing, and adverse effects of using lipid-lowering agents," the authors wrote, adding that drug targets with genetic support are severalfold more likely to succeed in clinical trials than other targets.
Although dyslipidemia — the imbalance of different types of lipids in the blood — is among the leading causes of death worldwide and a significant risk factor for cardiovascular-related death and conditions, the number of drugs available for patients who do not do well on statins is limited.
For their study, the researchers analyzed data from 276,249 participants in the UK Biobank with confirmed European ancestry and available genotyping data.
Their two-part analysis started with a phenome-wide association study (PheWAS) for three types of lipids: low-density lipoprotein cholesterol (LDL), high-density lipoprotein cholesterol (HDL), and triglycerides. To do this, the researchers created a polygenic risk score to predict each participant's lipid levels. Using the PRS and PheCodes, they categorized 62 lipid-driven phenotypes. This was followed by an MR analysis, which helped them link 19 phenotypes with LDL alone, 21 with HDL alone, 11 with triglycerides alone, and 11 with at least two lipids. One of the findings was that LDL cholesterol-lowering agents could be associated with an increased risk of cholelithiasis.
Next, they conducted transcriptome-wide MR for lipids, using expression data for more than 15,000 genes from five tissues, and classified the causal genes they found into four categories. Similarly, they performed transcriptome-wide MR for the 11 phenotypes previously linked to several lipids.
From there, using protein quantitative trait loci (pQTL) and protein-protein interaction data, they identified several pleiotropic genes associated with multiple lipids and cardiovascular traits.
After further drug target prioritization analysis, the researchers shortlisted 30 candidate drug target genes for dyslipidemia. Of note, 20 percent of these targets already had drugs for dyslipidemia either approved or under investigation.
Of the targets, they identified SORT1, CELSR2, and PSRC1 as top candidates because of their high level of pleiotropy. SORT1, which codes sortilin and is not yet under consideration for dyslipidemia, has been extensively studied experimentally, they wrote, and could be a target for therapeutic interventions, possibly using base editing.
The multiomics approach has several advantages over genome-wide association studies (GWAS) to identify drug targets, as GWAS data cannot reliably pinpoint causal variants or genes, the authors noted. Moreover, the targets prioritized in the study have a 22-fold higher likelihood of being approved or under investigation in clinical trials than GWAS-curated targets, they added.
While the new findings may guide clinicians to understand how targeted therapy for a specific lipid would affect patients, the study has a few limitations, they added, including that not all of the 30 prioritized drug targets may be druggable, suggesting further research.