This year 2022 was an eventful year for researchers publishing using products and services from Olink. There were some 315 publications in 2022 alone, and we recently celebrated our 1000th publication1.
Below is just a small highlighted selection of the many outstanding Olink publications of 2022, a short description of the problem or challenge the researchers set out to solve, and links are at the bottom to access the respective journal article. This brief list covers a wide range of applications of Olink technology, from stratification of patient samples for more efficient clinical trials (Bowman et al.) to determining the mechanism-of-action for a popular diabetes medication which also reduces risk of heart failure (Zanaad et al.) to combining genomics with proteomics to obtain pQTLs to identify new drug targets (Henry et al.)
From determining a predictive risk score for severe COVID that also suggests novel therapies based upon existing FDA-approved on-market drugs (Al Nesf et al.) to finding novel biomarkers to distinguish Alzhemiers’ patients with mild cognitive impairment versus non-Alzhemier’s patients with dementia (del Campo et al.), it is likely there is a reference or two below that matches your existing research goals.
Stratification of Patient Samples for Efficient Clinical Trials
Bowman et al. 20222
The progressive fibrosing form of interstitial lung disease (ILD) is a devastating and frequently fatal condition and may develop from any one of a number of different ILD variants. There are limited markers available for ongoing progressive ILD, but no predictive markers to identify at risk ILD patients, which would severely complicate any trials for preventive medication.
Using samples from patients with several different forms of ILD, the Olink Explore 384 Inflammation I Panel and a machine learning model, they identified a 12-protein signature (AGER,CST7, CXCL10, DPP10, FASLG, ITGB6, KRT19, MEPE, PLAUR, PNPT1, TNFSF11, and WFIKKN2) that identified individuals at high risk to develop the progressive form of ILD2. This had a negative predictive value of 0.91, meaning that <10% of patients without the high risk signature would develop progressive disease 1 year after blood sampling.
They calculated that by using this signature as a pre-stratification criterion, the number of ILD patients required for an effective clinical trial would be reduced by 80%.
Determining Mechanism of Action of Empagliflozin (Jardiance®)
Zannad et al. 20223
Sodium glucose co-transporter 2 (SGLT2) inhibitors improve cardiovascular outcomes in diverse patient populations, but their mechanism of action requires further study.
Over 1250 circulating proteins were measured at baseline, week 12 and week 52 in 1134 patients from EMPEROR-Reduced and EMPEROR-Preserved clinical trials using the Olink® Explore 1536 platform3. Statistical and bioinformatical analyses identified differentially expressed proteins (empagliflozin vs placebo), which were then linked to demonstrated biological actions in the heart and kidneys.4
The most common biological action of differentially-expressed proteins appeared to be the promotion of autophagic flux (the degradation of proteins within individual cells) in the heart, kidney or endothelium, a feature of 6 proteins. Other effects of differentially-expressed proteins on the heart included the reduction of oxidative stress, inhibition of inflammation and fibrosis, and the enhancement of mitochondrial health and energy, repair and regenerative capacity. The actions of differentially expressed proteins in the kidney involved promotion of autophagy, integrity and regeneration, suppression of renal inflammation and fibrosis, and modulation of renal tubular sodium reabsorption.
Empowering Genomics with Proteomics in Heart Failure Research
Henry et al. 20224
Heart failure (HF) is a highly prevalent disorder for which disease mechanisms are incompletely understood. The discovery of disease-associated proteins with causal genetic evidence provides an opportunity to identify new therapeutic targets.
The researchers investigated the observational and causal associations of 90 cardiovascular proteins, which were measured using Olink protein biomarker assays. They estimated the associations of 90 cardiovascular proteins with incident heart failure by means of a fixed-effect meta-analysis of four population-based studies, comprising a total of 3,019 participants with 732 HF events.
The causal effects of HF-associated proteins were then investigated by Mendelian randomization (MR), using cis-protein quantitative loci genetic instruments identified from genome-wide association studies (GWAS) in over 30,000 individuals.
The researchers discovered eight proteins had evidence of a causal association with HF that was robust to multiverse sensitivity analysis: higher CSF-1 (macrophage colony-stimulating factor 1), Gal-3 (galectin-3) and KIM-1 (kidney injury molecule 1) were positively associated with risk of HF, whereas higher ADM (adrenomedullin), CHI3L1 (chitinase-3-like protein 1), CTSL1 (cathepsin L1), FGF-23 (fibroblast growth factor 23) and MMP-12 (Matrix metalloproteinase-12) were protective. Therapeutics targeting ADM and Gal-3 are currently under evaluation in clinical trials, and all the remaining proteins were considered druggable, except KIM-1.
Determining a Predictive Risk Score for Severe COVID-19
Al-Nesf et al. 20225
Using a set of Olink Target 96 panels, 893 plasma proteins were profiled in 50 severe, 50 mild-moderate COVID-19 patients, and 50 healthy controls. This research shows 375 proteins are differentially expressed in the plasma of severe COVID-19 patients. From these 375 differentially expressed proteins the researchers derived a 12-protein “COVID-19 molecular severity score” (CLEC4C, PTX3, TNC, SMOC1, HGF, IL1RL1, IL6, AREG, KRT19, TNFRSF10B, IL18R1, MSTN) that had exceptionally high predictive value for severe disease: 100% specificity and 98% sensitivity (AUC= 0.999)5.
Based on the plasma proteomics and clinical lab tests, they also report a 12-plasma protein signature and a model of seven routine clinical tests that validate in an independent cohort as early risk predictors of COVID-19 severity and patient survival. These differentially expressed plasma proteins are implicated in the pathogenesis of COVID-19 and present targets for candidate drugs to prevent or treat severe complications.
A Broad Alzheimer’s Disease Screen for Biomarker Discovery and Potential Translation
del Campo et al. 20226
Researchers analyzed CSF samples from patients with AD and other types of dementia, initially using 11 Olink Target 96 panels across 797 patient and control samples. They included groups of patients with mild cognitive impairment with Alzheimer’s pathological changes (MCI(Aβ+)), AD dementia and patients with other non-AD dementias and cognitively normal controls. They were able to identify >100 CSF proteins dysregulated in MCI(Aβ+) or AD compared to controls or non-AD dementias. These proteins were associated with diverse biological processes such as cellular remodeling, lipid metabolism, protein clearance, vascular function, synaptic disfunction and immunity.
Using data-driven modeling, they identified an 8-protein signature (ABL1, SDC4, CLEC5A,MMP-10, ITGB2, TREM1, THBD and SPON2) that stratified MCI(Aβ+) and AD from controls with an extremely high AUC of 0.966. In addition, validation in a completely independent cohort maintained an impressive AUC of 0.94.
The same signature could also discriminate between AD and non-AD dementia with a still impressive, but lower AUC of 0.8, leading them to separately look for a better signature specifically for this type of stratification. This produced a 9-protein signature (ABL1, THOP1, ENO2, DDC, ITGB2, GZMB, MMP7, VEGFR-3 and PTK7) that improved the AUC to 0.87, and when this was applied to more stringently diagnosed samples only from patients with genetically or pathologically confirmed non-AD dementia, this score improved to 0.96.
For more information about this research paper, please visit our related news article.7
RNA-Seq Confirming a Novel Cardiac Biomarker for Pulmonary Arterial Hypertension
Boucherat et al. 20228
Clinical outcomes of patients with pulmonary arterial hypertension (PAH) are tightly linked to right ventricular (RV) function, but the mechanisms and proteins involved in RV remodeling are poorly understood. This study combined RNA-Seq and Mass Spectrometry proteomic analysis of healthy and diseased RV tissue samples from PAH patients along with plasma proteomics using the Olink Explore Cardiometabolic 384 panel.
The tissue-based multiomics identified a panel of proteins mainly related to cardiac fibrosis that were differentially regulated in diseased RV tissue. The Olink analysis showed an expected large increase in NT-proBNP, NPPB and cardiac troponin (TNNI3) in patients with PAH, and the NT-proBNP data were cross-validated with clinical measurements taken at the time of sampling.8
These data correlated very tightly (r= 0.9368), “demonstrating the accuracy of the Olink method” in the author’s words. Overall, 62 proteins were upregulated in PAH patients and when this data was integrated with the RV tissue RNA-Seq and MS data, 5 fibrosis-related proteins were identified as upregulated in PAH/RV damage according to all 3 methods – LTBP-2, COL6A3, COL18A1, TNC and CA1.
The researchers point out the strength of combining multiomic tissue-specific and systemic analysis, identifying accessible blood-based markers that can be directly linked to changes in the diseased tissues. “Using a multi-omics and multi-tissue-based prioritization strategy, we not only confirmed some biomarkers previously characterized but also identified new ones with strong predictive value and that represent potential targets directed at supporting or improving RV function in PAH”.
Small-sample Tissue RNA-Seq Combined with Circulating Proteome Analysis in Systemic Sclerosis
Clark et al. 20229
Skin systemic sclerosis is currently assessed by an invasive and error-prone method called the modified Rodnan skin score (mRSS). More reliable info can be obtained by RNA measurements of skin biopsies, but the desired clinical aim would be for a minimally-invasive blood biomarker that accurately reflects the biology of the disease site.
Here researchers combined RNA-Seq from skin biopsies to look firstly at correlating protein changes in interstitial fluid from skin blisters, before extrapolating to plasma protein measurements. Using 13 Olink Target 96 panels they identified key analytes present in both RNA transcripts and proteins in the interstitial fluid that correlate with degree of skin involvement in systemic sclerosis, and correlate with protein levels in the blood.
Multivariate analysis confirmed four of these analytes were statistically significant and independently correlated to skin involvement (OL4A1, COMP, TNC, SPON1).9 Each of these candidate markers may reflect distinct facets of skin pathogenesis and so add substantial value to previous studies independently linking these proteins to the current mRSS method of assessment. This is a promising plasma protein signature that could be used for assessment of skin severity, case stratification, and as a potential outcome measure for clinical trials and practice.
Novel Predictive Markers of anti-TNF Response in Rheumatoid Arthritis Patients
Prasad et al. 202210
Rheumatoid arthritis (RA) is a chronic autoimmune condition, characterized by joint pain, damage and disability, which can be addressed in a high proportion of patients by timely use of targeted biologic treatments. The researchers, looking for predictive biomarkers of anti-TNFα treatment response in >100 rheumatoid arthritis (RA) patients, used four Olink Target 96 Panels to take baseline protein measurements in conjunction with the start of therapy and then followed the clinical responses over 6 months.
Initial proteomics analysis (PCA) indicated two distinct clusters indicative of RA endotypes with marked differences in gender balance and disease severity.10 This was a novel finding that could have significance for better understanding and treatment of RA, but the protein profile was not predictive for anti-TNFα treatment response.
They then divided the patients into clinically assessed responder and non-responder groups and applied machine learning algorithms to identify patterns associated with drug response. This produced a 17-protein ML classifier they called “ATRPred” (Anti-TNF Treatment Response Predictor): REN, IL13, ARNT, HAOX1, DPP10, SPON1, GDNF, TRAILR2, HRT19, FCRL6, hOSCAR, TNFS13B, RARRES2, MMP1, CXCL1, PRKCO, MCP2. This classifier had an impressive AUC accuracy of 0.86 to predict responders to therapy. The authors conclude that their ATRPred classifier “may aid clinicians in deciding about putting an RA patient under anti-TNF therapy. This will help in saving the treatment cost as well as preventing nonresponsive patients to go through refractory condition of the disease leading to poor quality of life.”
References:
- Olink Proteomics AB. 2022 November 3. Olink announces milestone achievement of 1,000 peer-reviewed articles citing use of PEA technology [Press release] https://investors.olink.com/news-releases/news-release-details/olink-announces-milestone-achievement-1000-peer-reviewed and also https://olink.com/news/celebrating-1000-olink-publications/
- Bowman WS and Oldham JM et al. Proteomic biomarkers of progressive fibrosing interstitial lung disease: a multicentre cohort analysis. Lancet Respir Med. (2022) 10(6):593-602. doi:10.1016/S2213-2600(21)00503-8
- Zannad F, Ferreira JP et al. Effect of Empagliflozin on Circulating Proteomics in Heart Failure: Mechanistic Insights from the EMPEROR Program. (2022) European Heart Journal, DOI: 10.1093/eurheartj/ehac495
- Henry A and HERMES and SCALLOP Consortia, et al. Therapeutic Targets for Heart Failure Identified Using Proteomics and Mendelian Randomization. Circulation. (2022) 145(16):1205-1217. doi:10.1161/CIRCULATIONAHA.121.056663
- Al-Nesf MAY and Jares A-E et al. Prognostic tools and candidate drugs based on plasma proteomics of patients with severe COVID-19 complications. Nat Commun. (2022) 13(1):946. doi:10.1038/s41467-022-28639-4
- del Campo and Teunissen C.E. et al. CSF proteome profiling across the Alzheimer’s disease spectrum reflects the multifactorial nature of the disease and identifies specific biomarker panels. Nat Aging 2, 1040–1053 (2022). https://www.nature.com/articles/s43587-022-00300-1
- News article: Profiling of Alzheimer’s Disease CSF reveals specific biomarkers with potential diagnostic utility
- Boucherat O. and Bonnet S et al. Identification of LTBP-2 as a plasma biomarker for right ventricular dysfunction in human pulmonary arterial hypertension. Nat Cardiovasc Res 1, 748–760 (2022). Doi:10.1038/s44161-022-00113-w
- Clark KEN and Denton CP et al. Integrated analysis of dermal blister fluid proteomics and genome-wide skin gene expression in systemic sclerosis: an observational study. Lancet Rheumatol. (2022) 4(7):e507-e516. doi:10.1016/S2665-9913(22)00094-7
- Prasad B and Shukla P, et al. (2022) ATRPred: A machine learning based tool for clinical decision making of anti-TNF treatment in rheumatoid arthritis patients. PLOS Comput Biol. (2022) 18(7): e1010204. https://doi.org/10.1371/journal.pcbi.1010204
