October 14, 2026 · Science

What the Protein Scores Added Beyond the Measured Numbers

Multi-protein scores for triglyceride-rich lipoproteins and lipoprotein(a) predicted incident coronary artery disease independently of their measured blood concentrations. In a primary prevention cohort of 35,269 UK Biobank participants, researchers tracked 1,599 coronary artery disease events across time using Cox proportional hazards regression.

A blood draw for apolipoprotein B provides particle counts, but the proteins carried on these particles convey biological information beyond the count. For triglyceride-rich lipoproteins, each standard deviation increase in the multi-protein score carried a hazard ratio of 1.16, with a 95 percent confidence interval of 1.08 to 1.24. The protein score for lipoprotein(a) yielded a hazard ratio of 1.09 per standard deviation, with a 95 percent confidence interval of 1.03 to 1.15.

The multi-protein score for low-density lipoprotein showed no independent association with coronary events once researchers adjusted for measured lipoprotein concentration. That score produced a hazard ratio of 0.95 per standard deviation, with a 95 percent confidence interval of 0.89 to 1.02. Because the confidence interval crosses 1.00, the low-density lipoprotein protein signature added no predictive information beyond the standard particle number.

Adjusting for measured concentration means the statistical model accounted for the physical amount of circulating particles reported on a standard lab panel. If the particle count explained the entirety of disease risk, all three protein scores would have lost their statistical significance. Instead, the protein signatures for triglyceride-rich lipoproteins and lipoprotein(a) continued to predict coronary events after adjustment. The proteins carried risk information that the measured panel numbers did not.

The short version

A blood test report for lipoprotein(a) gives you a single concentration number. A study published in the European Heart Journal by researchers at Chalmers University of Technology shows that this number does not capture the full biological picture. In 35,269 UK Biobank participants, multi-protein signatures for triglyceride-rich lipoproteins (TRL) and lipoprotein(a) [Lp(a)] predicted incident coronary artery disease (CAD) even after adjusting for the measured particle levels. The hazard ratio was 1.16 per standard deviation for the TRL score and 1.09 for the Lp(a) score across 1,599 coronary events. The score for low-density lipoprotein (LDL) showed no association once adjusted for LDL concentration, with a hazard ratio of 0.95. Both TRL and Lp(a) signatures shared 36 proteins that cluster in inflammatory pathways and recruit immune cells to arterial walls. This work explains why TRL and Lp(a) carry higher per-particle risk than LDL. It does not provide a new clinical test. The researchers state directly that their analysis did not test an intervention or establish a test for medical use. You cannot buy this protein panel, and you should not change your current care plan because of it.

The 36 Proteins TRL and Lp(a) Share

Triglyceride-rich lipoproteins and lipoprotein(a) shared 36 distinct plasma proteins enriched in inflammatory pathways, separating both particle classes from low-density lipoprotein. Across 2,918 tested proteins, the research team identified 30 proteins associated with low-density lipoprotein, 471 associated with triglyceride-rich lipoproteins, and 53 associated with lipoprotein(a).

The protein signatures for triglyceride-rich lipoproteins and lipoprotein(a) were distinct from the low-density lipoprotein pattern yet overlapped substantially with each other. Both particle classes carried proteins that attract immune cells, which could contribute to inflammation and to structural changes in blood vessel walls. In contrast, the low-density lipoprotein protein pattern was mainly associated with the buildup of cholesterol in arterial tissue.

Proteins That Attract Immune Cells

Our guide to lipoprotein(a) details how apolipoprotein(a) carries oxidized phospholipids that provoke vascular inflammation and behaves as an acute-phase reactant. This proteomics study expands that framework by identifying 36 specific shared proteins that link lipoprotein(a) directly to the inflammatory pathways seen on triglyceride-rich lipoproteins.

In earlier work by the same research group, counting harmful particles was the most effective way to predict future heart disease risk. That work established that triglyceride-rich lipoproteins and lipoprotein(a) are more harmful per particle than low-density lipoprotein, positioning them as the primary drivers of remaining risk after cholesterol is lowered.

Why the LDL Null Result Is the Easiest Thing to Misread

The null hazard ratio for the low-density lipoprotein protein score does not indicate that low-density lipoprotein is harmless. With a hazard ratio of 0.95 and a 95 percent confidence interval of 0.89 to 1.02, the low-density lipoprotein protein score was not associated with coronary events after adjusting for particle concentration.

Low-density lipoprotein is the most abundant atherogenic particle in circulation and primarily acts by depositing cholesterol directly into artery walls. Senior author Clemens Wittenbecher from Chalmers University of Technology explained the clinical context: "Lowering LDL cholesterol remains a cornerstone of prevention."

Residual Risk After LDL Lowering

When medical treatment successfully reduces low-density lipoprotein cholesterol to a healthy level, patients can still experience cardiovascular events. That lingering vulnerability is known as residual cardiovascular risk, and it is largely propelled by triglyceride-rich lipoproteins and lipoprotein(a). These particles carry proteins that recruit immune cells, and their per-particle atherogenicity exceeds that of low-density lipoprotein.

How the Study Was Built, and What That Buys

The investigators applied a dual-method concordance protocol to evaluate 2,918 plasma proteins across 35,269 UK Biobank participants. The analysis evaluated plasma proteins associated with lipoprotein concentrations using multivariable-adjusted regression alongside one-sample Mendelian randomization.

Testing 2,918 proteins creates a substantial multiple-comparison problem where false positive associations can easily arise. To prevent statistical artifacts, the authors carried forward only proteins that met a Bonferroni-adjusted significance threshold of P < 0.05 in both observational regression and Mendelian randomization. Those validated proteins were compiled into multi-protein scores and evaluated against 1,599 incident coronary events.

The research team subsequently confirmed their protein scores in the Multi-Ethnic Study of Atherosclerosis, which included 5,915 participants. They also conducted high-dimensional mediation analysis to assess how proteomic alterations explained lipoprotein-associated coronary risk. While this design rigorously filters confounding, the analysis remains observational and genetic rather than an interventional trial.

Proteins change dynamically with diet, stress, aging, and disease, providing a real-time snapshot of human physiology. However, establishing an association between a protein score and coronary disease does not prove that pharmaceutically modifying those proteins will prevent arterial damage. You can read more about how we evaluate observational studies versus randomized trials in our research methodology guide.

What Is Actually on Your Panel Today

Commercial clinical laboratories do not offer multi-protein scores for lipoprotein risk, leaving standard blood tests as your operational tools. In their published paper in the European Heart Journal, the researchers stated directly that the study did not establish a protein test for clinical use or test a new medical treatment.

You can already order validated tests for the primary particles examined in this research. Our comparative breakdown of lipoprotein(a) vs apolipoprotein B explains how these measurements interact during clinical risk assessment.

Routine Blood Tests You Can Order Today

Triglyceride-rich lipoproteins cannot be counted directly on standard lipid profiles. Instead, routine clinical practice monitors their presence through non-HDL cholesterol and the triglyceride-HDL ratio. At Longevity Benchmark, we evaluate testing protocols by what clinical trials substantiate rather than prospective research prototypes.

Our broader guide to cardiovascular biomarkers outlines how these conventional lipid measurements guide preventative therapy. Until validated proteomic assays enter clinical practice, these routine markers remain the primary basis for medical decision-making.

What Would Change This Read

A randomized controlled trial showing that therapeutically targeting these specific protein pathways reduces heart attacks would change how clinicians interpret this research. Observational associations and Mendelian randomization point to biological mechanisms, but only clinical trials confirm that modifying a target improves health outcomes.

A standardized clinical assay measuring these protein scores would also be required before doctors could incorporate them into patient care. The study authors noted that future research must establish whether measuring or targeting these biological processes improves prevention alongside existing medical treatments.

We previously reported on a different lipoprotein(a) investigation examining plaque progression on carotid ultrasound in the Suita cohort, available at our Suita cohort coverage. That study tracked physical plaque growth in Japanese adults, whereas this new European Heart Journal study analyzed circulating plasma proteomics in European and multi-ethnic cohorts. Discuss your apolipoprotein B and lipoprotein(a) results with your physician to evaluate how residual vascular risk fits into your overall cardiovascular prevention plan.

Sources

  • European Heart Journal, doi.org. Source of the study design, 2,918 proteins, UK Biobank cohort (n = 35,269), 1,599 CAD events, MESA replication (n = 5,915), concordant protein counts (30 LDL, 471 TRL, 53 Lp(a)), 36 shared proteins, hazard ratios, confidence intervals, and the statement that no clinical test or treatment was established (accessed October 14, 2026).
  • MedicalXpress, medicalxpress.com. Institutional reporting on the Chalmers University study, lead author Jakub Morze, senior author Clemens Wittenbecher, immune-cell recruitment mechanisms, vessel wall structural changes, and the quote affirming LDL reduction as a cornerstone of prevention (accessed October 14, 2026).
  • Chalmers University of Technology, www.chalmers.se. Institutional affiliation for lead author Jakub Morze and senior author Clemens Wittenbecher in the precision medicine and diagnostics research group (accessed October 14, 2026).

Frequently Asked Questions

Is lipoprotein(a) an inflammatory marker?

Lipoprotein(a) carries inflammatory properties, but it is classified as an apolipoprotein B-containing lipoprotein particle rather than a standalone inflammatory marker. Our biomarker guide notes that apolipoprotein(a) carries oxidized phospholipids that provoke vascular inflammation and behaves as an acute-phase reactant. In this European Heart Journal study, the 53 proteins associated with lipoprotein(a) were enriched in inflammatory pathways and included proteins that recruit immune cells to arterial walls. These immune-recruiting proteins contribute to inflammation and vessel wall structural changes. While lipoprotein(a) reflects and promotes inflammatory activity, it remains an atherogenic particle that delivers cholesterol to artery walls.

Does this study mean LDL cholesterol does not matter?

No, this study does not suggest that low-density lipoprotein is unimportant. The lack of association for the LDL protein score occurred because the analysis adjusted for measured LDL concentration. Low-density lipoprotein is the most abundant cholesterol-carrying particle in blood and is mainly associated with cholesterol buildup in artery walls. Senior author Clemens Wittenbecher noted that lowering LDL cholesterol remains a cornerstone of prevention. The findings show that after LDL concentration is accounted for, the LDL-associated proteins do not add extra risk information, whereas protein scores for TRL and Lp(a) continue to predict coronary disease events.

Can I get my lipoprotein protein signature tested?

You cannot order a lipoprotein protein score from any clinical laboratory. The authors explicitly stated that their study did not establish a protein test for clinical use or evaluate a new treatment. The 2,918-protein panel was an academic research tool used to evaluate biological pathways in the UK Biobank and MESA cohorts. Routine clinical testing remains limited to standard lipid panels, apolipoprotein B, and lipoprotein(a) particle concentrations. Patients looking to assess residual cardiovascular risk should focus on established blood biomarkers rather than proteomic signatures.

What are triglyceride-rich lipoproteins, and are they on a standard lipid panel?

Triglyceride-rich lipoproteins are apolipoprotein B-containing particles that carry triglycerides and cholesterol throughout the bloodstream. Standard lipid panels do not report a direct count of triglyceride-rich lipoprotein particles. Instead, clinicians evaluate their presence using routine surrogates such as non-HDL cholesterol and the triglyceride-HDL ratio. Research from the Chalmers University team previously showed that triglyceride-rich lipoproteins and lipoprotein(a) are more harmful per particle than low-density lipoprotein. They represent major sources of residual cardiovascular risk after low-density lipoprotein cholesterol is brought to target levels.

Why were triglyceride-rich lipoproteins and Lp(a) so similar to each other?

Triglyceride-rich lipoproteins and lipoprotein(a) shared a set of 36 concordant plasma proteins that were distinct from the low-density lipoprotein signature. Both particle classes showed strong enrichment in biological pathways related to vascular inflammation. These shared proteins are involved in recruiting immune cells, which contributes to arterial wall inflammation and structural changes in vessel tissue. In earlier work, both particle types demonstrated higher per-particle atherogenicity than low-density lipoprotein. Their shared proteomic characteristics help explain why both contribute to cardiovascular disease independently of traditional cholesterol measurements.

Does a high Lp(a) result mean something different now?

A high lipoprotein(a) result carries the same clinical interpretation as before this study was published. The research demonstrates why lipoprotein(a) causes arterial harm by identifying 53 associated proteins enriched in immune-cell recruitment and inflammatory pathways. However, the study did not test a new treatment or establish a protein test for clinical use. If your blood work shows elevated lipoprotein(a), that finding indicates increased cardiovascular risk driven by particle atherogenicity and vascular inflammation. You should review your overall cardiovascular prevention strategy with your physician using current evidence-based clinical guidelines.