APOL1 Gene Variants and Risk for Cardiovascular Disease

Studies have demonstrated that polymorphisms in the gene for APOL1 confer an increased risk for kidney disease progression in people of African ancestry [1]. These APOL1 risk variants denominated G1 and G2 are associated with focal segmental glomerulosclerosis, HIV-associated nephropathy, advanced lupus nephritis, and hypertension-attributed chronic kidney disease (CKD) [2, 3]. However, the association between APOL1 risk alleles and cardiovascular disease (CVD) is unclear. Several recent studies suggest the possibility of APOL1 risk variants also contribute to an increase in the cardiovascular risk observed in people of African ancestry. In a report from the Jackson Heart Study, Ito et al. [4] reported a higher burden of CVD in people with APOL1 risk variants; Mukamal et al. [5] found APOL1 risk variants associated with increased risk of mortality and myocardial infarction. In contrast, several studies with patients with diverse CKD grades have found no associations between APOL1 risk variants and higher CVD risk.

The Systolic Blood Pressure Intervention Trial (SPRINT) failed to detect an association between APOL1 G1/G2 risk alleles and prevalent CVD [6]. Grams et al. [7] reported in a large meta-analysis that APOL1 high-risk (HR) variants were not associated with incident CVD or death independent of kidney measures. Colombian population is highly diverse with substantial admixture between African, Native American, and European ancestral populations [8]. It is currently estimated that 10.4% of the Colombian population is Afro-descendant. We aim to assess the relationship between the presence of APOL1 risk variants and the CVD risk in Afro-descendant patients with end-stage renal disease (ESRD) treated at Fundación Valle del Lili and a university hospital in Cali, Colombia.

We performed an observational, cross-sectional study of Afro-descendant adult patients with ESRD who were on the waitlist for a kidney transplant. Patients with acute kidney failure, cognitive impairment, or psychiatric disease that would affect their ability to voluntarily consent to participate in the study were excluded. The race was assessed by self-recognition of the race and physician-assessed classification of skin color.

Data from cardiovascular events (myocardial infarction, stroke, coronary heart disease, and congestive heart failure) and cardiovascular risk factors were collected by interviews and review of medical records from patients attended our institution, at the transplant program, from January to November 2017. Cardiovascular risk factors were defined as: hypertension as office SBP values ≥140 mm Hg and/or diastolic BP values ≥90 mm Hg [9], dyslipidemia like an imbalance of lipids such as cholesterol, low-density lipoprotein cholesterol, triglycerides, and high-density lipoprotein, and diabetes mellitus with the clinical practice recommendations of American Diabetes Association [10]. The Institutional Review Board approved the study protocol. Blood samples for genomic DNA extraction were obtained from participants after signing the informed consent.

Echocardiography was performed in a standard manner with the same equipment by cardiologists with expertise in echocardiographic recording and interpretation. An echocardiography examination was undertaken using two-dimensional measurements. LVEF was determined by modified biplane Simpson’s rule. Left ventricular (LV) diameters, volumes, diastolic function, and mass were measured as recommended.

The primary clinical endpoint was the cardiovascular risk and for estimating the probability of CVD, the risk assessment tool was the Framingham score. This score has been validated in whites and blacks in the USA and is transportable with calibration to culturally diverse populations. In our study, we adjusted the risk by a correction factor of 0.75. This adjustment is based on a study that showed that the Framingham scale overestimates cardiovascular risk in Colombia by 30% [11].

Genomic DNA extraction was performed using E.Z.N.A^® Tissue DNA Kit (Omega Bio-Tek) according to the manufacturer’s instructions. Quantity and purity of gDNA extracted were assessed using a NanoDrop 2000^® UV-Vis spectrophotometer (Thermo Fisher Scientific). Amplification of gDNA was performed by PCR. The set of primers (amplifying a 421 bp in the exon 7 of the APOL1 gene), amplification reaction setup, and thermocycling conditions have been described in a previous study [12]. PCR products were run on an agarose gel to verify the size of the amplified fragment: 5 μL of each PCR reaction was analyzed in a 1% agarose gel containing 0.5% ethidium bromide and was visualized by UV illumination. The PCR products were purified using an E.Z.N.A. Cycle-Pure Kit (Omega Bio-Tek) and sequenced using BigDye chemistry in an ABI 3500 automated system (Applied Biosystems) according to the manufacturer’s instructions.

Baseline demographic, socioeconomic, and clinical characteristics were compared by APOL1 risk status (HR versus low risk [LR]). The categorical variables were summarized in proportions, and the continuous variables were expressed as the mean and standard deviation.

APOL1 HR status was defined as the presence of 2 risk alleles (G1/G1, G2/G2, or G1/G2) versus the LR status, defined as having 1 or 0 risk variants (G1/G0, G2/G0, or G0/G0). Clinically significant variables or those with p < 0.20 in the univariate analysis were included in a multivariate Cox regression analysis, which yielded odds ratio and confidence intervals (CI). The model between APOL1 risk variants and cardiovascular risk was the primary prespecified analysis. The proportional hazards model adjusted for baseline covariates of age, sex, diabetes mellitus, and dyslipidemia. Values of p < 0.05 were considered statistically significant. All statistical analyses were performed using the statistical software package STATA 12.

We enrolled a total of 102 patients, of which 56% (n = 57) were female. The mean age was 48 years (standard deviation 13 years). There were 44 patients on the transplant waitlist and 58 in the posttransplant period. The mean time of dialysis before a kidney transplant was 4 years. The mean age at the start of dialysis in the LR group was 41 years and in the HR group was 38 years without significant differences according to the groups.

In the LR group, 25% (n = 16) had diabetes compared to the HR group where 7.9% (n = 3) had it. In both groups, 92% of the patients had hypertension (n = 59 in the LR and n = 35 in the HR group). 23.4% (n = 15) of patients in the LR group had peripheral artery disease compared to 13.2% (n = 5) in the HR group. The LR group had 9.4% (n = 6) of obese patients, while the HR group had 15.7% (n = 6). Regarding dyslipidemia, 26.5% (n = 17) of the patients from the LR group had it, compared to 15.7% (n = 6) in the HR group. The demographic characteristics of the patients by APOL1 genotype status are described in Table 1. The etiological cause of ESRD was as follows: unknown in 82%, diabetic nephropathy in 5%, lupus nephritis in 5%, polycystic disease in 4%, and others in 4%. The comorbid condition more frequently found was arterial hypertension in 88% of patients. Ongoing drug therapy at the time of the score assessment was aspirin (28.4%, n = 29), ACEI/ARBs (42.2%, n = 43), beta blockers (17.6%, n = 18), diuretics (28.4%, n = 29), calcium channel blockers (43%, n = 44), alpha-2-adrenergic agonists (12.7%, n = 13), and statins (24.5%, n = 25).

37% had APOL1 HR status in two alleles in homozygous (G1/G1 = 21 and G2/G2 = 3) or compound heterozygote (G1/G2 = 14) form and 63% had APOL1 LR status. Figure 1 shows allele and genotype frequencies.

Lipid profile and comorbidities did not show differences between HR versus LR groups. There was no significant association between APOL1 genotypes and the adjusted Colombia Framingham Risk Score. APOL1 HR versus LR status was not independently associated with LV hypertrophy or systolic dysfunction. Three no-cardiovascular deaths occurred during the follow-up.

The cardiovascular risk-associated factors were evaluated in all patients and multivariate analyses were performed. Table 2 shows multivariable analysis of cardiovascular risk by clinical characteristics and APOL1 risk genotype. Diabetes was a statistically significant risk factor associated with a high cardiovascular risk definite for a Framingham score of more than 10 points.

This study shows that 37% of the Afro-descendant patients with end-stage CKD had an APOL1 HR allele. We did not find an association between APOL1 risk alleles and major cardiovascular risk.

Previous studies have shown that the G1 and G2 APOL1 risk alleles do not represent a strong independent risk factor for CVD. In the African American participants from the SPRINT, APOL1 association was tested with baseline estimated glomerular filtration rate, urine albumin:creatinine ratio, and prevalent CVD in 2,571 Afro-descendant patients. APOL1 was positively associated with CKD (odds ratio 1.37, 95% CI 1.08–1.73), but the risk variants were not significantly associated with prevalent CVD (odds ratio 1.02, 95% CI 0.82–1.27) [6]. The Atherosclerosis Risk in Communities (ARIC) Study found a strong association between the HR APOL1 genotype and ESRD but failed to demonstrate any association with incident CVD in 3,676 African American participants [13].

However, these studies primarily consisted of individuals with normal kidney function or non-dialysis-dependent CKD. Chen et al. evaluated the associations of APOL1 risk variants with subclinical CVD and mortality in a cohort of black incident hemodialysis patients. This study included a population more like ours. 27% had 2 risk alleles and 73% had 0 or 1 risk allele. APOL1 HR status was associated with better baseline measures of subclinical CVD, namely, lower likelihood of LVH and CAC and lower LV mass, compared with those with LR status [12].

Our study did not assess subclinical CVD but rather cardiovascular risk as measured by the Framingham score. That is a simplified and common tool for the assessment of the risk level of CAD over 10 years, considering six coronary risk factors, including age, gender, total cholesterol, high-density lipoprotein cholesterol, smoking, and systolic blood pressure. We used this score because it is the most applicable method for predicting the person’s chance of developing CVD in long term [14].

In the risk factor analysis, diabetes was identified as an important risk factor for high cardiovascular risk. Like diabetes mellitus, CKD may be a powerful enough risk factor for ASCVD that obscures any independent effect of APOL1 genotype. In previous reports, the association of APOL1 genotype with incident composite CVD differed by diabetes mellitus status; for Gutierrez et al. [15] in the REGARDS study, patients without diabetes mellitus, APOL1 HR genotypes associated with greater risk of incident composite CVD (hazard ratio, 1.67; 95% CI, 1.12–2.47), but there was no statistically significant association of APOL1 genotypes with incident CVD in subjects with diabetes mellitus.

CVD is a major issue for patients with ESRD, especially those on hemodialysis. Risk alleles of the APOL1 gene have been associated with higher risk-adjusted odds of earlier initiation of chronic hemodialysis [16, 17]; however, in our study, we did not find statistically significant differences in the age of onset of dialysis. The risk of cardiovascular events and mortality in dialysis patients is related to the prevalence of traditional and no traditional cardiovascular risk factors and the time in dialysis. In our study, 82% of the patients had been on dialysis therapy with a mean of 4 years of dialysis therapy; however, the effect of the time in dialysis on the mortality risk of hemodialysis patients does not follow a straight line and is influenced by comorbidities and treatment factors not evaluated in this study.

The etiological cause of ESRD was unknown in 82%. Our hospital is a kidney transplant referral center, and many of the patients arrive in stage 5 of the disease for evaluation of possible kidney transplant. There is a very high proportion of patients with CKD of unknown cause, most likely due to the late diagnosis of CKD and, the result of an insufficient number of renal biopsies.

Our study is not without limitations, and the results should be interpreted in the context of the study design. First, the race was not assessed by mitochondrial haplogroup or autosomal genetic ancestry. Second, the size of the cohort was determined based on the available budget for the molecular analysis, which limits the applicability of our results. Third, we do not have longitudinal blood pressure, lipid panel, or glycated hemoglobin, and data may have changed differentially over time between genotype groups. Fourth, common CV risks were not different among the two groups, but whether there are additional factors determining differences in CV events between the two groups is unclear and longitudinal studies are required.

Despite our limitations, we believe that our analysis advances the knowledge of the APOL1 mutations and their impact on outcomes among the Afro-descendant population. However, a furthermore extensive multicenter study should be performed to better understand the frequency and role of these mutations.

In Afro-descendent patients with ESRD, the APOL1 HR status is not associated with the increase in cardiovascular risk profile and metabolic disturbances. More follow-up and the number of patients are required to determine future cardiovascular event associations.

This study meets ethical standards in accordance with the World Medical Association Declaration of Helsinki. Blood samples were obtained from participants after signing the informed consent. Data were collected through interview and review of medical records of patients who attended our institution. The Institutional Review Board of Fundacion Valle del Lili approved the study, protocol number 1030. Written informed consent was obtained from participants prior to the study.

The authors declare no conflict of interest.

This research received no external funding.

All those designated as authors meet all four criteria for authorship. Carlos Duran: conception, design, and analysis of the work. Mayra Estacio and Daniela Espinosa: analysis and interpretation of data for the work and writing assistance. Juan G Posada, Liliana Mesa, and Johanna Schweineberg: conception and design of the work. Eliana Manzi: statistic analysis.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.