Correlation of serum apolipoprotein B with the Framingham Risk Score among a group of Iraqi subjects: a cross-sectional study

Introduction

Apolipoprotein B (apoB) is a key structural component of all atherogenic lipoproteins, including low-density lipoprotein (LDL), playing a pivotal role in lipoprotein assembly, receptor binding, and metabolic regulation. As the primary structural protein of LDL, apoB is fundamental to hepatic lipoprotein synthesis and is directly implicated in atherogenesis (Olofsson, Wiklund & Borén, 2007).

Cardiovascular disease (CVD) remains a leading cause of morbidity and mortality (Gaidai, Cao & Loginov, 2023), with diabetes mellitus, hypertension, and dyslipidemia being the primary risk factors for coronary artery disease (CAD) (Duggan et al., 2022). Addressing these risk factors is central to clinical practice and preventive cardiology (Wong et al., 2022). Several algorithms have been developed to estimate CAD risk, including the Reynolds Risk Score (Ridker et al., 2007), Pooled Cohort Equations (Lloyd-Jones et al., 2019), and the Framingham Risk Score (FRS). The FRS, derived from the landmark Framingham Heart Study, is a widely adopted tool for estimating the 10-year risk of CAD in middle-aged, asymptomatic individuals (Andersson et al., 2019; Dehghan et al., 2024). It incorporates key risk variables such as age, sex, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), systolic blood pressure, antihypertensive therapy, and smoking status (D’Agostino Sr et al., 2008). Despite the widespread use of the FRS, its reliance on conventional lipid parameters raises concerns. Previous studies have demonstrated a significant proportion of CAD patients exhibit normal lipid profiles (Silverman et al., 2016; Hoog et al., 2022), requesting the predictive accuracy of standard risk equations. Notably, serum apoB has demonstrated a stronger association with CAD development than traditional lipid measures (Lawler et al., 2017). Unlike low density lipoprotein cholesterol (LDL-C), which estimates cholesterol content, apoB directly quantifies the number of atherogenic lipoprotein particles, offering a more precise assessment of atherogenic burden (Sniderman, Pedersen & Kjekshus, 1997). From a pathophysiological standpoint, cholesterol within non-HDL particles is only harmful when transported by apoB-containing lipoproteins (Skålén et al., 2002). Studies found that each lipoprotein particle that penetrates the endothelium of the coronary artery can initiate atherosclerosis regardless of its cholesterol content (Tabas, Williams & Borén, 2007), and as the number of atherosclerotic lipoprotein particles increases there will be an increase in the retention of lipoprotein particles in the arterial wall where the latter can induce inflammation and leads to an increase in the coronary artery permeability and makes the endothelium layer loss its barrier function (Khalil, Wagner & Goldberg, 2004). The best example for this explanation is the stronger correlation of atherogenicity with the small dense LDL particles compared to normal-size LDL particles although small-dense LDL particles carry less amount of cholesterol in comparison to normal-size LDL particles (Froyen, 2021). Additionally, apoB measurement is cost-effective, does not require fasting, and remains reliable even in the presence of hypertriglyceridemia or reduced LDL-C levels due to lipid lowering therapy (Glavinovic et al., 2022). Approximately 20% of individuals with metabolic disorders, including hypertriglyceridemia, type 2 diabetes, and obesity, exhibit discordance between apoB and LDL-C levels, underscoring the limitations of LDL-C-based risk assessment (Glavinovic et al., 2022). Unlike LDL-C and non-HDL-C, which are derived through mathematical estimation, apoB is directly measured in laboratories, offering greater precision.

In recent years, increasing attention has been given to the pathophysiological role of serum apoB in CAD progression, particularly in explaining the presence of CAD in individuals with an accepted range of lipid profiles (Mach et al., 2020). However, despite its well-established role in atherogenesis, most previous studies have examined apoB in the context of diagnosed CAD, rather than as a predictor of future risk. To our knowledge, only one study has directly correlated apoB with the FRS (Ryoo et al., 2011). Given these insights, our study explores the correlation between serum apoB levels and FRS in an Iraqi cohort, aiming to clarify its utility in CAD risk prediction and primary prevention. This approach offers an added advantage by assessing apoB’s role in risk stratification before the clinical onset of CAD, especially in a population that remains underrepresented in cardiovascular research.

Methods

Sample size determination and participant selection

The required minimum sample size was estimated using a modified two-mean formula with a power of 80% and a Confidence Interval of 95%. The mean and the standard deviation of serum apoB in the three FRS groups were derived from a previous study (Ryoo et al., 2011). The equation determined that at least 120 subjects were necessary to achieve adequate statistical power. A total of 287 participants were initially screened. After applying the eligibility criteria, 86 individuals were excluded, comprising 15 with diabetes, two pregnant women, 13 with a history of CVD, and three who declined participation. An additional 53 participants were excluded due to incomplete data, resulting in a final study population of 201 eligible participants. These exclusions were due to technical issues such as hemolyzed, clotted, or insufficient blood samples, which rendered biochemical analysis incomplete. Since the exclusions were based solely on laboratory sample quality and not participant characteristics, they are unlikely to have introduced selection bias. Figure 1 illustrates the participant selection process, detailing the exclusions due to eligibility criteria, refusal, and incomplete data.

Results

Predictive performance of apoB for high CAD risk

The ROC curve analysis demonstrated that a serum apoB level of 97.75 mg/dL serves as a potential cutoff value for predicting high CAD risk as estimated by FRS as shown in Fig. 2. The AUC was 92% (p = 0.0001), indicating strong discriminatory power for CAD risk stratification. Where FRS was classified into two groups FRS <  20% and FRS > 20% where the second group was considered as CAD risk equivalent. apoB value of 97.75 mg/dL was the cutoff point with the highest sensitivity (94%) and specificity (79%) for predicting high FRS (FRS > 20%).

Discussion

The present study provides compelling evidence that serum apoB levels vary significantly across FRS-defined risk groups, with the highest levels observed in the high-FRS group, followed by the moderate and low-risk groups. This trend, coupled with the strong correlation between apoB and FRS, as well as ROC analysis, suggests a potential role for apoB in refining CAD risk stratification as estimated by FRS. Although the FRS includes total cholesterol and HDL-C, it might not account for the full spectrum of atherogenic lipoproteins. ApoB, by quantifying the number of atherogenic particles directly, offers additional predictive value beyond traditional lipid parameters included in FRS.

Despite the fact that the apoB/apoA1 ratio has also been proposed as a meaningful marker, prior evidence suggests that apoB alone is a superior predictor of CAD mortality, with the predictive power of the ratio largely driven by apoB itself. Therefore, the current study focused on apoB as a stand-alone marker, which is both pathophysiologically grounded (Galimberti, Casula & Olmastroni, 2023; Sierra-Johnson et al., 2009).

The current findings align with a large observational study of 13,523 Korean males, which demonstrated a significant association between serum apoB and CAD risk in individuals without diabetes or hypertension (Ryoo et al., 2011). However, beyond this study, there remains a paucity of research directly correlating apoB with FRS. Instead, most prior investigations have focused on apoB levels in established CAD cases. For instance, a case-control study examining patients with unstable angina diagnosed via electrocardiography found a significant association between elevated serum apoB and unstable angina, compared to controls (Al-Tu’ma, Mohammedad & Al-Sarraf, 2017). Similarly, another study involving 160 participants (80 with acute coronary syndrome and 80 controls) confirmed a strong correlation between apoB, apoA, and apoB/A ratio with acute coronary syndrome incidence (Jassim & Ali, 2016). While these studies highlight the role of apoB in diagnosed CAD cases, our study extends this evidence by demonstrating its association with CAD risk prediction via FRS. Further supporting our findings, a prospective study of 9,231 asymptomatic adults in Denmark, followed over eight years, reported that apoB outperformed LDL-C in predicting incident CVD (Benn et al., 2007). Likewise, the Health Professionals Follow-up Study, which followed 18,225 men for six years, found that apoB was superior to non-HDL-C in predicting nonfatal myocardial infarction (Pischon et al., 2005). These longitudinal studies reinforce the clinical relevance of apoB in cardiovascular risk assessment, underscoring its potential role as a valuable biomarker in primary prevention. However, apoB is still not recommended by the current guidelines for clinical practice in the primary prevention of CAD.

A key contribution of this study is the identification of a cutoff value of 97.75 mg/dL for predicting high CAD risk using FRS. However, the cutoff value was previously investigated in other populations but the results were controversial. The variation in the cutoff value may be attributed to differences in the design of each study, where the present study has used FRS as a determinant factor of the predictive role of apoB while the previous studies have used the presence or the prospective occurrence of CAD instead. In addition to the ethnicity of participants and the methodology employed. For instance, a case-control study conducted among 437 participants aged 40–70 years old admitted to the hospital for angiography indicated a higher cutoff value of 120 mg/dl for serum apoB for diagnosis of CAD (Levinson, 2007). Another case-control study of 271 participants conducted in Iran reported a lower cutoff value of serum apoB of 89.5 mg/dl for angiography-diagnosed CAD patients (Mashayekhi et al., 2014). Notably, the present cutoff value falls within the range recommended by the American Diabetes Association (ADA) and the American College of Cardiology Foundation (ACCF) for high-risk patients (<90 mg/dL) (Brunzell et al., 2008) and the Adult Treatment Panel III (ATP III) recommendations for moderate-risk individuals (<110 mg/dL) (Grundy, 2002). This positioning suggests that apoB may serve as a clinically meaningful biomarker for risk stratification, bridging the gap between high-risk thresholds established by ADA/ACC and moderate-risk thresholds outlined in ATP III, as both of them have two or more risk factors with no previous history of diabetes or CAD. Despite these similarities, current clinical guidelines from the American Heart Association (AHA) and American College of Cardiology (ACC) do not yet incorporate apoB as a routine parameter for cardiovascular risk assessment. Instead, LDL-C and non-HDL-C remain the primary targets for lipid-lowering therapy. However, the Canadian Cardiovascular Society and the European Society of Cardiology/European Atherosclerosis Society (ESC/EAS) do recommend apoB as a co-primary or secondary target, particularly in patients with elevated triglycerides or those at high risk (Arnett et al., 2019; Pearson et al., 2021; Pirillo, Casula & Catapano, 2023).

Given the strong correlation between apoB and FRS observed in this study, further prospective research is needed to validate apoB’s predictive utility and refine its cutoff values for different risk groups, particularly in Middle Eastern populations.

Sex-based differences in serum apoB levels were evident in this study, with males exhibiting a significantly higher apoB concentration than females. This discrepancy may be attributed to the higher catabolic rate of apoB-containing lipoproteins in females (Matthan et al., 2008), a phenomenon previously observed in a large Swedish cohort of 147,576 participants (Jungner et al., 1998). Furthermore, a strong positive correlation between apoB levels and age was noted, likely due to the age-related decline in LDL receptor activity and reduced catabolism of apoB-containing lipoproteins (Ericsson et al., 1991). This trend aligns with prior research conducted in 100 males and 100 females (Avogaro et al., 1979), reinforcing the notion that aging contributes to increased apoB concentrations. Given that age is a key determinant in FRS calculations, this age-related rise in apoB may partially explain its correlation with FRS in the current study. A more detailed age-stratified analysis among females revealed significantly higher apoB levels in women aged ≥50 years compared to those <50 years. This could be attributed to postmenopausal hormonal changes, particularly the decline in estrogen levels, which has been shown to correlate with higher apoB concentrations (Ej, 1994). Supporting this, previous study had demonstrated a decrease in serum apoB levels following estrogen replacement therapy (Granfone et al., 1992), highlighting the hormonal regulation of apoB metabolism in females.

Conclusion

The findings of this study underscore the potential role of serum apoB as a promising predictive biomarker for CAD risk stratification in the Iraqi population, demonstrating a strong positive correlation with FRS. apoB may offer advantages over LDL-C and non-HDL-C in assessing cardiovascular risk, particularly in populations where conventional lipid markers may be less predictive. Unlike previous studies that primarily associated apoB with established CAD, this study contributes population-specific insights by identifying a cutoff value of 97.75 mg/dL for high CAD risk. However, the study’s reliance on FRS as a surrogate measure presents a limitation, as FRS was developed for Western populations and may not fully capture Iraqi-specific risk factors. Further prospective studies are needed to validate these findings and assess apoB’s role in clinical guidelines for CAD risk prediction, dyslipidemia diagnosis, and preventive strategies.

Supplemental Information