Two studies show the “critical role” that large biobanks may play in the process of discovering drug targets for ischemic heart disease (IHD) and heart failure. The studies highlight both the strengths and limitations of proteomic and multiomic analyses of large biobanks to identify potential drug targets, according to an editorial commenting on both papers. These papers and editorial were published Monday online and in the Nov. 14 issue of the Journal of the American College of Cardiology. Plasma proteomics analysis Mohsen Mazidi, PhD, and Neil Wright, PhD, both of the University of Oxford, England, and colleagues, performed a nested case-cohort study in the China Kadoorie Biobank (CKB), which yielded 1,971 incident cases of ischemic heart disease (IHD) and 2,001 subjects in a subcohort. The IHD cases were randomly selected from a sample of such cases that had genome-wide association studies, no history of cardiovascular disease, and were not statin users. The subcohort subjects were randomly chosen from 69,353 genotyped participants; they were not genetically related to each other, and they also had no history of cardiovascular disease or statin use. Cis-protein quantitative loci for proteins from the Chinese subjects, and those from the U.K. Biobank (UKB), were assessed for causal relevance for IHD in observational analyses. The investigators then replicated the observational associations of proteomics with IHD using Mendelian randomization with instrumental variables discovered in the CKB and replicated in the UKB. The combined analyses of proteomic and genomic data in Chinese adults strongly supported causal relevance of 13 proteins for IHD. Four of those proteins (FURIN, F24, ASGR1 and MMP3) were further replicated in European adults. The importance of several of the 13 proteins was confirmed for cardiovascular disease or related traits, as was their potential as drug development targets for IHD. However, of those 13 proteins, seven yielded no evidence of drug development, while of the remaining six, only one was associated with drug development for IHD. FURIN, which is highly expressed in endothelial cells, was identified as a potential novel target, and MMP3 (matrix metalloproteinase-3) was identified as a potential repurposing target, for IHD. Multiomics analysis Wouter Ouwerkerk, PhD, and Joao P. Belo Pereira, PhD, both of Amsterdam University Medical Centers, and colleagues, used machine learning to integrate genetic, transcriptomic and proteomic data from 2,516 patients with heart failure from the BIOSTAT-CHF study (index cohort). They validated their results with an independent cohort of 1,738 patients (validation cohort). At a median 21-month follow-up, 26% of patients in the index cohort and 32% of patients in the validation cohort died. In index cohort, compared to patients who survived, those who died were older (73.0 ± 11.2 years vs. 68.0 ± 11.9 years; p<0.0001) and had a higher New York Heart Association (NYHA) functional class (Class III-IV 73% vs. 57%; p<0.0001) and more comorbidities. The validation cohort showed a similar pattern. The authors’ systems biology multiomics approach identified four major pathways for the progression of heart failure: PI3K-Akt, MAPK, and Ras signaling pathways, and EGFR tyrosine kinase inhibitor resistance. These pathways are associated with decreased activation of the ERBB2 receptor, which can be modified by neuregulin. ‘Substantial work still needed’ In an accompanying editorial, W.H. Wilson Tang, MD, of the Cleveland Clinic, and Wolfgang Koenig, MD, of the German Center for Cardiovascular Research, said the “papers underscore the potential of multiomics data analyses from well-curated large cardiovascular biobanks to expedite cardiovascular drug development.” The editorialists also noted several limitations of these kinds of analyses. Large biobanks and population studies are often from nonrandomized subjects, which limits the generalizability of findings from analyses of these datasets. Assay measurement processes may vary, and complex data integration and analysis requires the consideration of multiple factors. Those seeking to link genetics with proteomics must show that genetic variants affect the levels of the specific protein they are analyzing and that they are not associated with confounding factors that could influence the outcome, the editorialists wrote. Given these limitations, the expert commenters expressed excitement about the potential for the studied techniques to help discover more potential cardiovascular drug targets. They concluded that “large biobanks may play a critical role in facilitating and accelerating the process of drug discovery by using multiomics technologies combined with novel analytical tools to identify new pathways, but substantial work is still needed for the validation of technologies, standardization of data analyses, and integration of proteomics with other molecular and phenotypic data.” “In addition, we should not forget that even then we are only generating hypotheses and need further in vitro and in vivo experiments to elucidate mechanisms underlying the reported associations and subsequent human studies to determine actual efficacy in a patient population,” Tang and Koenig wrote. Sources: Mazidi M, Wright N, Yao P, et al. Plasma Proteomics to Identify Drug Targets for Ischemic Heart Disease. J Am Coll Cardiol. 2023;82:1906–1920. Ouwerkerk W, Belo Pereira JP, Maasland T, et al. Multiomics Analysis Provides Novel Pathways Related to Progression of Heart Failure. J Am Coll Cardiol. 2023;82:1921–1931. Tang WHW, Koenig W. Multiomics Insights to Accelerate Drug Development: Will They Hold Their Promises? J Am Coll Cardiol. 2023;82:1932–1935. Image Credit: sizsus – stock.adobe.com