Lipoprotein(a) risk thresholds across the continuum of cardiovascular prevention

This editorial refers to ‘Clinical significance of elevated lipoprotein(a) in primary and secondary prevention: a multi-institutional study’, by D-.Y. Chen et al., https://doi.org/10.1093/eurjpc/zwaf649. Lipoprotein(a) [Lp(a)] is now firmly established as a mostly genetically determined, lifelong contributor to atherosclerotic cardiovascular disease (ASCVD) and calcific aortic valve stenosis.1 A substantial body of epidemiologic, genetic, and clinical evidence links elevated Lp(a) to myocardial infarction, ischaemic stroke, coronary revascularization, and cardiovascular mortality, independent of traditional risk factors and LDL cholesterol (LDL-C).1,2 As Lp(a)-lowering therapies approach clinical implementation, the central question has shifted from whether Lp(a) confers risk to how that risk should be interpreted across different clinical contexts. In this issue of the European Journal of Preventive Cardiology, Chen et al. report findings from a large multi-institutional cohort of 51 934 individuals from Taiwan, including 49 363 without established ASCVD and 2571 with established disease. Using an Lp(a) threshold of ≥30 mg/dL, the investigators demonstrate a consistent association between elevated Lp(a) and major adverse cardiovascular events across both primary [adjusted subdistribution hazard ratio (aSHR) 1.24] and secondary prevention (aSHR 1.36), populations over a mean follow-up of 6.6 years. Importantly, this threshold is below the level used by current guidelines to identify increased risk and is well below the inclusion criteria of ongoing clinical trials. The authors should be commended for this informative analysis of a large cohort; however, several limitations warrant consideration. The retrospective design precludes causal inference and allows for residual confounding, as clinically driven Lp(a) testing introduces potential selection bias towards individuals perceived to be at higher baseline risk. The exclusively Taiwanese cohort, which included mostly young males, further limits ethnic and geographic diversity, constraining generalizability given known population-level differences in Lp(a) distributions and associated risk.3,4 In addition, reliance on a single Lp(a) threshold oversimplifies what is likely a continuous and context-dependent risk relationship that varies by atherosclerotic burden and the presence of other risk factors. This is compounded by heterogeneity in Lp(a) assays and reporting in mg/dL rather than molar units, which—owing to variability in apolipoprotein(a) isoform size—do not reliably reflect particle number, complicating cross-study comparisons and biological interpretation. Although elevated Lp(a) is prognostically relevant across prevention categories, accumulating evidence suggests that its clinical implications differ according to baseline ASCVD status.1,4 This concept is well illustrated by the large contemporary US cohort reported by Berman et al.,4 in which higher Lp(a) levels among individuals with established ASCVD were associated with increased rates of myocardial infarction and coronary revascularization, with excess risk becoming most apparent above the ∼70th percentile. Notably, in this secondary prevention population, elevated Lp(a) was not significantly associated with ischaemic stroke or cardiovascular death, and relative risk appeared to plateau at higher levels. In contrast, among individuals without prior ASCVD, the relationship between Lp(a) and cardiovascular risk was more linear, with higher relative risk emerging only at very high Lp(a) levels (>90th percentile). These observations highlight that identical Lp(a) concentrations may carry different prognostic implications depending on clinical context, background risk, and competing mechanisms of disease. The traditional dichotomy of primary vs. secondary prevention is increasingly challenged by data demonstrating high event rates among ostensibly primary prevention patients with substantial subclinical atherosclerosis, such as that detected by coronary artery calcium scoring or coronary computed tomography angiography (CCTA).5,6 Rather than existing as discrete categories, cardiovascular risk appears to lie along a continuum defined by both prior clinical events and underlying atherosclerotic burden. Subclinical atherosclerosis occupies a critical intermediate position along this continuum. Patients without prior ASCVD but with imaging evidence of coronary plaque are typically classified as primary prevention, yet their biological risk profile and event rates may more closely resemble those of secondary prevention populations.6,7 In this setting, elevated Lp(a) may act as a potent accelerator of plaque development, progression, and transition to clinical events.5,8,9 Prior studies demonstrate that combining Lp(a) with imaging-based measures of atherosclerosis improves risk discrimination beyond either marker alone, suggesting that subclinical disease meaningfully modifies the prognostic significance of Lp(a).5,8 One potential explanation for this observation is that individuals with underlying atherosclerosis have higher vulnerability, while those without plaque have protective factors that make them more resilient.10 Importantly, most Lp(a) studies—including the present investigation—do not explicitly stratify patients by subclinical atherosclerotic burden, potentially obscuring clinically actionable risk within this intermediate population.11 While the use of a single Lp(a) cut-off enhances clinical simplicity and supports broader screening, it risks oversimplifying a biologically complex and context-dependent risk factor. Lipoprotein(a) behaves as a continuous exposure, and dichotomization may obscure clinically relevant gradients of risk, particularly across heterogeneous prevention populations. In secondary prevention—where absolute risk is high and background therapy is more intensive—the incremental prognostic value of very high Lp(a) levels may be attenuated.4 Conversely, among individuals without clinical ASCVD but with subclinical coronary plaque, even modest elevations in Lp(a) may identify patients at disproportionately high lifetime risk.2,4 These considerations argue against a universal ‘one-size-fits-all’ threshold and favour interpretation of Lp(a) in the context of baseline risk, atherosclerotic burden, and the presence of alternative mechanisms driving disease progression (Figure 1) Conceptual framework of lipoprotein(a) thresholds and associated cardiovascular risk across the prevention continuum. Lipoprotein(a)-related risk is context dependent, evolving from primary prevention through subclinical atherosclerosis to secondary prevention. Its clinical impact is modulated by patient vulnerability, traditional and genetic risk factors, and competing modifiers. Imaging with coronary artery calcium and coronary computed tomography angiography identifies subclinical atherosclerosis and integrates anatomic disease expression with lipoprotein(a)-associated risk. Lp(a), lipoprotein(a); CAC, coronary artery calcium; CCTA, coronary computed tomography angiography. As Lp(a)-lowering therapies advance towards clinical use, these findings have direct implications for patient selection and trial design. Most ongoing outcome trials have focused on secondary prevention populations with very high Lp(a) levels.11 However, growing evidence suggests that reliance on prior clinical events alone may fail to identify individuals most likely to benefit from intervention. The Phase 3 ACCLAIM Lp(a) trial (NCT06292013) and OCEAN(a)-PreEvent trial (NCT07136012) represent an important step forwards by incorporating imaging-defined atherosclerosis into eligibility criteria, thereby acknowledging subclinical disease as a clinically meaningful risk state. Future studies should continue to integrate imaging, baseline risk, and population-specific Lp(a) distributions to define context-appropriate thresholds across the full continuum of cardiovascular risk. The current study adds important evidence supporting Lp(a) as a determinant of cardiovascular risk across the prevention continuum. Moving forwards, effective integration of Lp(a) into clinical practice will require abandoning rigid thresholds in favour of a more individualized, biology-driven framework that accounts for baseline risk, atherosclerotic burden, and patient-specific vulnerability. Such an approach will help to ensure that Lp(a)-lowering therapies are deployed where they are most likely to deliver meaningful clinical benefit. Arthur Shiyovich (Writing—original draft [lead], Writing—review & editing [equal]), Rishi Rikhi (Writing—original draft [supporting], Writing—review & editing [equal]), and Ron Blankstein (Writing—original draft [supporting], Writing—review & editing [equal])