Biomarker Discovery in Atherosclerotic Diseases Using Quantitative Nuclear Magnetic Resonance Metabolomics

Despite great progress in the management of atherosclerosis (AS), its subsequent cardiovascular disease (CVD) remains the leading cause of morbidity and mortality. This is probably due to insufficient risk detection using routine lipid testing; thus, there is a need for more effective approaches relying on new biomarkers. Quantitative nuclear magnetic resonance (qNMR) metabolomics is able to phenotype holistic metabolic changes, with a unique advantage in regard to quantifying lipid-protein complexes. The rapidly increasing literature has indicated that qNMR-based lipoprotein particle number, particle size, lipid components, and some molecular metabolites can provide deeper insight into atherogenic diseases and could serve as novel promising determinants. Therefore, this article aims to offer an updated review of the qNMR biomarkers of AS and CVD found in epidemiological studies, with a special emphasis on lipoprotein-related parameters. As more researches are performed, we can envision more qNMR metabolite biomarkers being successfully translated into daily clinical practice to enhance the prevention, detection and intervention of atherosclerotic diseases.

Nuclear Magnetic Resonance Derived Biomarkers for Evaluating Cardiometabolic Risk in Youth and Young Adults Across the Spectrum of Glucose Tolerance

Youth with obesity have an increased risk for cardiometabolic disease, but identifying those at highest risk remains a challenge. Four biomarkers that might serve this purpose are “by products” of clinical NMR LipoProfile® lipid testing: LPIR (Lipoprotein Insulin Resistance Index), GlycA (inflammation marker), BCAA (total branched-chain amino acids), and glycine. All are strongly related to insulin resistance and type 2 diabetes (T2DM) in adults (glycine inversely) and are independent of biological and methodological variations in insulin assays. However, their clinical utility in youth is unclear. We compared fasting levels of these biomarkers in 186 youth (42 lean normal glucose tolerant (NGT), 88 obese NGT, 23 with prediabetes (PreDM), and 33 with T2DM. All four biomarkers were associated with obesity and glycemia in youth. LPIR and GlycA were highest in youth with PreDM and T2DM, whereas glycine was lowest in youth with T2DM. While all four were correlated with HOMA-IR (Homeostatic Model Assessment for Insulin Resistance), LPIR had the strongest correlation (LPIR: r = 0.6; GlycA: r = 0.4, glycine: r = −0.4, BCAA: r = 0.2, all P < 0.01). All four markers correlated with HbA1c (LPIR, GlycA, BCAA: r ≥ 0.3 and glycine: r = −0.3, all P < 0.001). In multi-variable regression models, LPIR, GlycA, and glycine were independently associated with HOMA-IR (Adjusted R2 = 0.473, P < 0.001) and LPIR, glycine, and BCAA were independently associated with HbA1c (Adjusted R2 = 0.33, P < 0.001). An LPIR index of >44 was associated with elevated blood pressure, BMI, and dyslipidemia. Plasma NMR-derived markers were related to adverse markers of cardiometabolic risk in youth. LPIR, either alone or in combination with GlycA, should be explored as a non-insulin dependent predictive tool for development of insulin resistance and diabetes in youth.Clinical Trial RegistrationClinicaltrials.gov, identifier NCT:02960659

Sitosterolemia: A Rare Case with Broad Implications

Sitosterolemia is an ultra-rare autosomal recessive dyslipidemia characterized by mutations in genes encoding the ATP-binding cassette (ABC) G5/8 transporters. We describe the case of a 20-month-old female presenting with xanthomas and serum low density lipoprotein cholesterol of 657 mg/dL. Diagnostic workup revealed a previously undescribed sitosterolemia-causing mutation. After elimination of dietary sterols and initiation of ezetimibe therapy, the patient’s xanthomas resolved, and serum low density lipoprotein cholesterol was reduced to 104 mg/dL. Importantly, pathologically elevated serum phytosterols were found in each of the proband’s heterozygous parents. Elevated phytosterols, an established cause of atherosclerosis, are typically unrevealed by standard lipid testing. As heterozygous mutations for ABCG5/8 are relatively common, this has implications for a broader population than the ultra-rare sitosterolemia cohort. Thus, insights gleaned from this case highlight underappreciated matters in the prevention of atherosclerotic disease in both heterozygous and homozygous carriers alike.

Lipid Testing, Lipid-Modifying Therapy, and PCSK9 (Proprotein Convertase Subtilisin-Kexin Type 9) Inhibitor Eligibility in 27 979 Patients With Incident Acute Coronary Syndrome

Background: While registry-based studies have shown that as many as 1 in 2 patients with stable atherosclerotic cardiovascular disease would be eligible for PCSK9i (proprotein convertase subtilisin-kexin type 9 inhibitor) therapy, this has not been studied in a large population-based postacute coronary syndrome (ACS) cohort. Methods: We examined lipid testing performed in hospital or within 90 days of discharge and lipid-lowering therapies dispensed within 90 days of discharge in patients surviving for at least 1 year after their first ACS between 2012 and 2018 in the province of Alberta, Canada. We estimated the proportion of patients eligible for PCSK9i and the expected benefits of treatment. Results: Of the 27 979 patients (median age 64.0 years, 29.3% female, 28.0% diabetic), 3750 (13.4%) did not have lipid testing in-hospital or within 90 days postdischarge. Untested patients were more likely to be older, female, from rural areas, to have more comorbidities, to already be on cardioprotective therapies, to present with unstable angina, and were less likely to have invasive interventions (all P <0.0001). Of the 24 229 tested, 18 767 (77.5%) had at least one lipid value above guideline-recommended threshold (LDL [low-density lipoprotein] ≥1.8 mmol/L [70 mg/dL] and non-HDL [high-density lipoprotein] ≥2.6 mmol/L [100 mg/dL]), of which 7284 (38.8%) did not have repeat testing within the year after discharge. Lipid testing in hospital was associated with higher rates of initiation or escalation of statin therapy within 90 days of their ACS (adjusted odds ratio, 2.13 [95% CI, 1.97–2.30). In total, 9592 patients (39.6% of the tested cohort) would be eligible for PCSK9i use, which could result in 184 fewer cardiovascular events over 3.4 years, including cardiovascular death, nonfatal ACS (myocardial infarction or unstable angina requiring hospitalization), and ischemic stroke. Conclusions: Within 90 days of incident ACS, ≈80% of patients did not meet guideline-recommended lipid thresholds and more than one-third would potentially be eligible for PCSK9i.

927. Antiretroviral Laboratory Monitoring and Implications for HIV Clinical Care in the Era of COVID-19 and Beyond

Background In the era of COVID-19, providers are delaying laboratory testing in people with HIV (PWH) to avoid unnecessary exposures despite antiretroviral guidelines recommending periodic testing. The purpose of this study was to examine the clinical significance of periodic renal, liver, and lipid testing. Methods We reviewed the charts of 265 people with HIV (PWH) who initiated outpatient care at HIV clinic between 1/1/16 and 12/21/18 and had at least two clinic visits. Analysis included frequency distributions, descriptive statistics, one-sided binomial exact tests, and Poisson models with 95% confidence intervals (CI). Results Eighty-five percent (221) of PWH had no laboratory abnormalities while on antiretroviral therapy (ART). The most common abnormality was a glomerular filtration rate (GFR) &lt; 60 ml/min found in 10% of PWH. Multivariate analysis revealed that diabetes mellitus (DM) was associated with an increased risk of GFR &lt; 60 ml/min (estimated rate ratio 2.68, 95% CI 1.35-5.33) and age &lt; 60 years (estimated rate ratio .122, 95% CI .05-.32) was associated with a decreased risk (estimated rate ratio .24, CI .14 –.43). When a GFR was &lt; 60 ml/min or an AST or ALT was &gt;2X upper limit of normal (ULN), no action was taken in 52% of the cases. When an action was taken, the most common action was to repeat testing (18%). After a lipid panel result, the most common actions were to calculate a 10-year cardiovascular risk score (32%) and add a statin (18%). Taking action after lipid panel results was strongly associated with age ≥ 40 (estimated rate ratio 9.1, 95% CI 3.3-25). ART was changed in seven PWH based on GFR, AST/ALT, or lipid panel results. There were four individuals with poor outcomes including cerebrovascular accident, acute renal failure, end stage renal disease, congestive heart failure, myocardial infarction, and death. Contributing factors were hypertension, DM, and hypercholesterolemia. Conclusion Individuals &lt; 40 years without ithout comorbidities had a low risk of having clinically significant renal and liver function abnormalities and rarely had actions taken after renal, liver, or lipid results. In the era of COVID-19 and beyond, it may be prudent for in certain groups to delay or eliminate liver, renal, and lipid testing to eliminate exposure, reduce cost, and avoid patient anxiety. Disclosures All Authors: No reported disclosures

Impact of Lipoprotein(a)-Cholesterol on Accurate Classification of Familial Hypercholesterolemia

Lipoprotein(a) [Lp(a)] is a pro-atherogenic and pro-thrombotic LDL-like particle recognized as an independent risk factor for cardiovascular disease (CVD) that is resistant to typical lipid-lowering treatments. The cholesterol within Lp(a) (Lp(a)-C) contributes to the reported LDL-cholesterol (LDL-C) concentration by nearly all available methods including beta-quantification, direct homogenous assays, and all estimating equations. Accurate LDL-C measurements are critical for identification of genetic hyperlipidemia conditions such as familial hypercholesterolemia (FH). FH risk estimators such as the Dutch Lipid Clinic Network (DLCN) criteria utilize LDL-C concentration cut-offs and other clinical inputs to assess the likelihood of FH. Therefore, failure to adjust for Lp(a)-C can impact accurate FH classification, appropriate follow-up testing and treatments, and interpretation of cholesterol-lowering treatment efficacy. Lp(a)-C can be estimated from Lp(a) mass as measured by immunoassay using an average cholesterol content per particle. However, Lp(a)-C size and composition varies significantly within individuals resulting in inaccurate Lp(a)-C estimates. In this study, we use direct Lp(a)-C measurements to assess the potential misclassification of FH risk due to the contribution of Lp(a)-C to LDL-C in patient samples submitted for advanced lipoprotein profiling. A total of 28,200 samples submitted for lipoprotein profiling were included. The profiling included lipid testing in a CDC-certified laboratory on Roche cobas 501 (cholesterol and triglycerides by enzymatic method, high-density lipoprotein cholesterol by MgCl2/dextran sulfate precipitation). LDL-C was measured by beta-quantification, and Lp(a)-C by quantitative lipoprotein electrophoresis (SPIFE Vis Cholesterol, Helena Laboratories). The DLCN LDL-C cut-offs (155, 190, 250, and 330mg/dL) were applied to LDL-C results before and after accounting for Lp(a)-C contribution. Lp(a)-C was detected in 3,728 (13.2%) samples. The median (range) concentrations of Lp(a)-C and LDL-C were 11mg/dL (5-108mg/dL) and 121mg/dL (27-678mg/dL), respectively. Overall, subtracting Lp(a)-C would reclassify 6.5% of all samples into a lower LDL-C category within the DLCN algorithm. Within the LDL-C scoring categories, 7.0% (n=222) of subjects with LDL-C 155-189mg/dL, 5.6% (n=66) of subjects with LDL-C 190-249mg/dL, 5.2% (n=10) of subjects with LDL-C 250-329mg/dL, and 3.4% (n=4) of subjects with LDL-C &gt;330mg/dL would be down-classified after adjusting for Lp(a)-C. Limiting to subjects with measurable Lp(a)-C, reclassification to a lower diagnostic threshold occurred in 47.4% of subjects with LDL-C 155-189mg/dL, 37.5% with LDL-C 190-249mg/dL, 41.6% with LDL-C 250-329mg/dL, and 33.3% with LDL-C &gt;330mg/dL after adjustment. Current guidelines recommend screening for elevated Lp(a) in patients with family history of CVD. Our data show that a high percentage of samples evaluated for advanced lipid testing contain measurable Lp(a)-C that could cause mis-classification in FH prediction algorithms. If labeled high probability of FH, these mis-classifications could trigger inappropriate work-up for suspected FH. As clinical follow-up and therapeutic strategies differ between FH and elevated Lp(a), proper distinction between LDL-C and Lp(a)-C is needed to guide appropriate patient management.

Abstract MP72: Using Non-HDL-C/Triglyceride Ratio To Screen For Direct LDL-C Testing: Improving LDL-C Estimates At A Cost

Introduction: Despite more accurate low-density lipoprotein cholesterol (LDL-C) estimates by the Martin (mLDL-C) method, most laboratories still use the Friedewald (fLDL-C) equation. Low non-high density lipoprotein (non-HDL-C) and high triglycerides (TG) drive inaccuracy in LDL-C estimation. We compared a strategy of identifying errors in fLDL-C using non-HDL-C/TG ratios with subsequent reflex direct LDL-C testing to a strategy of using mLDL-C. Methods: We included 4,939,542 individuals (2/3 derivation, 1/3 validation dataset) with TG <400 mg/dL with lipid profiles directly measured via Vertical Auto Profile from the Very Large Database of Lipids. We compared directly measured LDL-C with estimated fLDL-C and mLDL-C. The direct LDL-C assay has an allowable error of 12% which was used as the threshold for accuracy assessment. LDL-C estimates above and below non-HDL-C/TG cutpoints (range 0-2.0) were evaluated for accuracy from the derivation dataset and the 4 best performing ratios were tested in the validation set. Individuals with non-HDL-C/TG ratios below the cutpoints were assumed to require direct LDL-C measurement. Medicare costs ($17 lipid panel; $12 direct LDL-C) were used to estimate added costs of direct LDL-C measurement. Results: Nearly 8% of fLDL-C results deviated >12% from direct LDL-C compared with only 2.5% of mLDL-C results in the entire population. Non-HDL-C/TG ratios of 0.6-0.9 performed best in the derivation dataset. In the validation dataset, a non-HDL-C/TG ratio of 0.7 had the highest area under the curve for identifying >12% error in fLDL-C estimates (Table). At this ratio, 14.5% of fLDL-C samples would need direct LDL-C measurement at an increase in costs of lipid testing by 10%. Conclusions: Use of non-HDL-C/TG ratio <0.7 as screening for direct LDL-C measurement can improve the accuracy of LDL-C estimates in labs using the Friedewald equation. However, such an approach is significantly costlier than using mLDL-C.