EPIC-Norfolk Study Supports Universal Screening for Three Cardiovascular Biomarkers

A single measurement of LDL cholesterol, high-sensitivity C-reactive protein (hsCRP), and lipoprotein(a) [Lp(a)] at study entry predicted 20-year cardiovascular risk in 17,087 initially healthy European men and women, with the highest risk observed in participants who had all 3 biomarkers elevated, according to findings from the EPIC-Norfolk study published in the European Heart Journal.¹

Validating a Novel Screening Strategy

The research directly addresses a critical gap in preventive cardiology: validating whether universal one-time screening for these 3 biomarkers, each representing distinct modifiable pathways to atherosclerosis, can reliably identify at-risk individuals in primary prevention. The study, led by Paul Ridker, director of the Center for Cardiovascular Disease Prevention and Eugene Braunwald Professor of Medicine at Harvard Medical School, in Boston, MA, replicates recent findings from the US Women's Health Study, which followed 27,939 American women for 30 years,² and extends those results to a European population that includes both sexes.

"While risk algorithms and imaging tests can inform preventive cardiologists about whom to treat, modifiable blood-based biomarkers can inform physicians about what to treat with," the authors noted.¹ “In this respect, we believe the ‘one size fits all’ approach to pharmacologic primary prevention is outdated” as novel therapeutics beyond statins enter clinical practice, creating opportunities for more personalized prevention strategies.¹

The EPIC-Norfolk Trial

The prospective EPIC-Norfolk cohort recruited participants aged 40–79 through general practices in Norfolk, UK, between 1993 and 1997. Researchers measured plasma levels of LDL cholesterol, hsCRP, and Lp(a) at baseline and followed participants for major adverse cardiovascular events (MACEs), defined as fatal or nonfatal coronary artery disease and ischemic stroke, through March 2016.

The cohort had a mean age of 59.0 years and included 9,745 women (57%). At baseline, mean LDL cholesterol was 4.0 ± 1.0 mmol/L, median hsCRP was 1.5 (0.7–3.2) mg/L, and median Lp(a) was 11 (6–27) mg/dL. Notably, only 1.1% were using lipid-lowering therapy at study entry, and traditional risk factors remained relatively low, with 11.3% current smokers, 2.9% with diabetes, and 16.3% with hypertension.

Independent and Additive Risk Prediction

During 20.5 years of median follow-up, 3,249 MACEs occurred. Ridker and colleagues reported that increasing quintiles of all 3 biomarkers independently predicted cardiovascular risk. Comparing the highest to lowest quintile, multivariable-adjusted hazard ratios (HRs) were 1.78 (95% CI: 1.57–2.00) for LDL cholesterol, 1.55 (95% CI: 1.37–1.74) for hsCRP, and 1.19 (95% CI: 1.07–1.33) for Lp(a).

The biomarkers showed minimal correlation with each other. Investigators said that Spearman coefficients ranged from 0.04 to 0.19, supporting their assessment of distinct pathogenic pathways. Each standard deviation increase in log-transformed levels yielded multivariable-adjusted HRs of 1.14 for LDL cholesterol, 1.07 for hsCRP, and 1.07 for Lp(a).

The combined analysis revealed striking additive effects. Compared with participants having no biomarkers in the top quintile, those with a single elevated biomarker were at a 33% increased risk (HR 1.33, 95% CI: 1.23–1.43) and those with 2 elevated biomarkers were at more than double that risk (68%) (HR 1.68, 95% CI: 1.51–1.87). Among participants with increased levels of all 3 biomarkers, the risk for MACE was approximately 2.5 times greater (HR 2.41, 95% CI: 1.90–3.07).

Sex-Specific Patterns

Sex-stratified analyses showed generally lower risk associations in women than men, though both sexes demonstrated significant trends. For women with all 3 biomarkers elevated, the multivariable-adjusted risk was 2.5 times greater (HR 2.61, 95% CI: 1.93–3.51); for men, the HR was 1.85 (95% CI: 1.22–2.81), though only 79 men (1.1%) had all 3 biomarkers in the highest quintile, limiting statistical power.

Clinical Implications

The consistency between UK and US findings proves particularly compelling, the autthors wrote. Among women, the UK data showed an HR per quintile of LDL cholesterol of 1.09 (95% CI: 1.05–1.14) vs 1.09 (95% CI: 1.07–1.12) in American women, with similarly aligned results for hsCRP.

"The fact that a single random blood measure of hsCRP provided long-term risk estimates comparable with that of LDL cholesterol and greater than that of Lp(a) should dispense concerns occasionally voiced by some clinicians that variability in hsCRP limits its prognostic value," the authors emphasized.

Cohort Limits Generalizability of Findings

Among the study’s limitations, Ridker et al acknowledged that diabetes, hypertension, and obesity prevalence was lower during 1993–1997 recruitment than today, potentially making findings more relevant to current populations. Data on statin initiation during follow-up were unavailable, though prior research suggests minimal impact when accounting for treatment initiation. The cohort's predominantly white European ancestry limits generalizability to other ethnicities.

“One-time simultaneous evaluation of LDL cholesterol, hsCRP, and Lp(a) is also likely to assist clinicians as they select different agents to address the unique aspects of risk presented by individual patients," Ridker and team wrote. "In this respect, we believe the ‘one size fits all’ approach to pharmacologic primary prevention is outdated as several novel therapeutics beyond statins are entering clinical practice."

The researchers concluded that “physicians will not treat what they do not measure.” With commercial assays for hsCRP and Lp(a) now standardized, inexpensive, and widely available, they believe “the time has come for universal screening of these three biomarkers in primary as well as secondary prevention,” they said.

At-A-Glance

Clinical Implications for Primary Prevention

Extending Risk Assessment Horizons

• Current data evaluate 20-year cardiovascular risk, far exceeding traditional 10-year risk estimates from guideline algorithms
• Longer assessment periods better reflect cardiovascular disease as a lifetime disorder where early prevention achieves greatest gains
• Findings reinforce the value and clinical utility of early life risk detection

Validating hsCRP as a Robust Risk Marker

• A single random blood measure of hsCRP provided long-term risk estimates comparable to LDL cholesterol and greater than Lp(a)
• These findings should dispense concerns that hsCRP variability limits its prognostic value
• Since variability biases toward the null, the results actually underscore the underappreciated clinical and pathophysiologic importance of inflammation in early vascular disease development

Guiding Lifestyle and Pharmacologic Interventions

• European Society guidelines consistently indicate that early intervention through dietary modification, smoking cessation, and daily exercise provides greatest benefit
• While Lp(a) levels remain unaffected by lifestyle changes, these interventions reduce both risk and levels of LDL cholesterol and hsCRP
• Early identification enables targeting of modifiable risk factors when intervention yields maximum benefit

Personalizing Treatment Selection

• Simultaneous evaluation of all three biomarkers assists clinicians in selecting agents to address unique aspects of individual patient risk
• The "one size fits all" approach to pharmacologic primary prevention is outdated as novel therapeutics beyond statins enter clinical practice
• Emerging treatment options target specific pathways:
• SGLT2 inhibitors and GLP-1 receptor agonists lower hsCRP and cardiovascular risk
• Low-dose colchicine approved by FDA for high-risk primary prevention
• Interleukin-6 inhibition under evaluation in large-scale global trials
• Targeted therapeutics for elevated Lp(a) in development, as traditional lipid-lowering strategies do not affect Lp(a) levels

Supporting Pathway-Specific Prevention

• Combined effects analyses provide strong evidence that multiple pathways contribute to atherosclerosis
• Targeted prevention strategies should differ between individuals based on their specific risk profile
• Three-biomarker screening enables identification of which pathways require intervention in each patient

References

1. Kraaijenhof JM, Nurmohamed NS, Nordestgaard AT, et al. Low-density lipoprotein cholesterol, C-reactive protein, and lipoprotein(a) universal one-time screening in primary prevention: the EPIC-Norfolk study. Eur Heart J. 2025;46:3875–3884. doi:10.1093/eurheartj/ehaf209
2. Ridker PM, Moorthy MV, Cook NR, Rifai N, Lee I-M, Buring JE. Inflammation, cholesterol, lipoprotein(a), and 30-year cardiovascular outcomes in women. N Engl J Med 2024;391:2087–97. doi:10.1056/NEJMoa2405182