Genetics, Genomics, and Precision Medicine in End-Stage Kidney Disease

Summary:

Recent advances in genetics of renal disease have deepened our understanding of progressive kidney disease. Here, we review genetic variants that are of particular importance to progressive glomerular disease that result in end-stage kidney disease (ESKD). Some of the most striking findings relate to APOL1 genetic variants, seen exclusively in individuals of sub-Saharan African descent, that create a predisposition to particular renal disorders, including focal segmental glomerulosclerosis and arterionephrosclerosis. We also review the genetics of cardiovascular disease in ESKD and note that little work has been published on the genetics of other ESKD complications, including anemia, bone disease, and infections. Deeper understanding of the genetics of ESKD and its complications may lead to new therapies that are tailored to an individual patient’s genetic profile or are discovered based on genetic approaches that identify novel pathways of renal cell injury and repair.

End-stage kidney disease (ESKD) is the end result of diverse renal diseases and is attended by much morbidity and mortality. Mortality rates among dialysis patients are very high, as tracked by the United States Renal Data System. Table 1 shows the mortality rates among the general Medicare population, for individuals 65 to 74 years of age, with various conditions; the ESKD population has by far the highest mortality rate. Mortality rates are similar between hemodialysis and peritoneal dialysis patients, with death rates being 220 per 1,000 patient-years and 210 per 1,000 patient-years, respectively. Age-adjusted mortality rates, given here for 2015 and for individuals 65 to 74 years of age and on dialysis, are higher among European Americans (253) compared with African Americans (189; 25% lower) and others (166; 34% lower) (Table 2).

Table 1.

Mortality Rates Among the Medicare Population: 2014 to 2015

Morbidity	Mortality, deaths/1,000 patient-years
	Males	Females
All Medicare	27	18
Dialysis	223	211
Transplant	66	60
Congestive heart failure	112	101
Acute myocardial infarction	87	94
Stroke	72	57
Cancer	73	64
Diabetes	40	31

Death rates are shown for the general Medicare population (age, 65–74 y) from 2014 to 2015, adjusted for race. Dialysis patients (indicated in boldface) included both hemodialysis and peritoneal dialysis patients; these patients have the highest mortality rates among any large disease population receiving Medicare.

Reprinted from US Renal Data Systems Report, 2017, Table 5.5.⁵¹

Table 2.

Racial/Ethnic Differences in Death Rates Among Dialysis Patients

Age group, y	White	Black	Black/White death rate ratio	Other
0–21	10	18	1.8	8
22–44	33	44	1.3	18
45–64	100	100	1	74
65–74	221	175	0.79	142
≥75	358	272	0.76	235

The all-cause mortality rates shown are from dialysis patients during 2015, expressed as deaths per 1,000 patient-years. Note that the black/white mortality rate ratio among dialysis patients was numerically higher among individuals <45 years of age, whereas the rate was numerically lower among those ≥65 years of age. The reasons for these disparities, and their shifts over time, are not well understood.

Reprinted from US Renal Data Systems Report 2017, Table 5.2.⁵¹

The lion’s share of ESKD mortality is owing to cardiovascular disease (CVD). Several recent reviews have summarized the current state of knowledge in this area, including articles by Bansal¹ and Regunathan-Shenk et al,² with a focus on disadvantaged minorities, by Tuegel and Bansal,³ Subbiah et al,⁴ and a review of inflammation and premature aging in chronic kidney disease (CKD) by Kooman et al.⁵

Many genetic loci and variants have been associated with glomerular disease, tubular disease, and kidney function. Several excellent recent reviews are available; these include a review of the genetics of diabetic complications by Dahlstrom and Sandholm.⁶ The focus of this article is to review what is known about the genetics of ESKD itself (as distinct from the genetics of nephrotic syndrome and systemic kidney diseases, all of which may progress to ESKD) and the genetics of three principal complications of ESKD (CVD, infection, and bone disease), and mortality among ESKD patients. Because much of the data were acquired via genomic studies (ie, genome-wide association studies), this review also draws from these studies. At this time, this body of work offers insights into possible mechanism of kidney disease and its complications, but as yet the field has not matured sufficiently that it can make recommendations that will affect the practice of medicine. This may change in the near future.

GENETIC STUDY DESIGNS

As a brief review of genetic methodology, there are three principal approaches to finding genetic variants that are associated with disease.⁷ First, family studies involve comparing the genomes of affected and unaffected family members. At present, this typically is performed by whole-exome sequencing, which will identify variants that affect the portion of the genome that codes for proteins. If the whole-exome sequencing approach does not provide an answer, then a whole-genome scan may be required to find variants in regulatory regions that might explain the disorder, although given the length and complexity of the human genome, this can be challenging. Second, for common diseases for which hundreds or ideally thousands of cases can be assembled, a genome-wide association study (GWAS) may identify common variants present on the array that are in linkage dysequilibrium with the as-yet-unknown disease variant, and these common variants suggest the genomic location of the disease variant.⁸ Third, a particularly powerful version of GWAS, termed mapping by admixture linkage disequilibrium (MALD), can identify disease-associated variants in an admixed population.⁷ MALD can be used only for diseases showing disparities in prevalence between the two ancestral populations contributing to the admixture (eg, ESKD in African Americans). The underlying hypothesis is that the causal variant will be more frequent on a chromosomal segment derived from the population with the higher disease prevalence. An admixed population carries ancestry from at least two distinct population groups that have been genetically isolated from each other for sufficient time for the processes of selection pressure or genetic drift to have led to genome-wide differences between the two study populations, in both coding and noncoding regions of the genome. The study design involves comparing cases and controls, all drawn from the same admixed population, genotyping for single-nucleotide polymorphisms (SNPs) differing in frequencies between the two population groups, and looking for substantial differences in local ancestry between the two study groups to localize a genomic region harboring the ancestry-specific causal variant.

GENETICS OF END-STAGE KIDNEY DISEASE

There is strong evidence for a genetic contribution to ESKD risk. A Swedish study identified 971 ESKD cases among adoptees, and studied the probands, their biologic parents, and their adoptive parents.⁹ For adoptees of biologic parents with ESKD, the odds ratio (OR) for ESKD was 6.1 (95% confidence interval [CI], 3–14), whereas the OR for ESKD was not significant for adoptees whose adoptive parents had ESKD. These findings suggest a strong genetic contribution to ESKD. Furthermore, African Americans, and indeed individuals with sub-Saharan African heritage and particularly those with West African heritage, living around the world, have a substantially higher risk for ESKD; the responsible locus was identified using the MALD approach and the mechanism for much of this risk involves APOL1 genetic variants, as will be discussed later.

It should be noted that genetic studies that recruit ESKD patients and compare them with non–kidney disease controls generally are unable to distinguish between genetic variants that increase the rate of progression to ESKD (thus, reducing the number of such subjects who die before reaching ESKD, all other things being equal) and those that increase the propensity to have any degree of progressive glomerular injury, resulting in ESKD. Although it may not matter practically from an epidemiologic perspective, the underlying biology may differ.

Diabetic Nephropathy

Many studies have sought genetic variants that are associated with diabetic nephropathy but fewer studies have examined the genetic risk of ESKD in diabetic patients. Although diabetic nephropathy unfortunately is common, and therefore recruitment for genetic studies is comparatively easy, most patients with a course that is typical for diabetic nephropathy do not undergo a kidney biopsy and therefore some subjects in a study may in fact have other kidney disease, such as focal segmental glomerulosclerosis, membranous nephropathy (which do not always manifest nephrotic-range proteinuria), and arterionephrosclerosis. These limitations would tend to weaken the observed genetic associations, particularly for nephropathy associated with type 2 diabetes. A recent article reviews the genetics of diabetic nephropathy.¹⁰

In 2014, Palmer et al¹¹ reported on a GWAS of African American diabetic kidney disease and found a peak on chromosome 22, with MYH9 having the strongest association, and other nearby peaks (APOL1, SFL1, and LIMK2) also being associated with an increased risk for diabetic kidney disease. After conditioning on the APOL1 renal disease–associated SNPs G1 and G2, the remaining significant peaks were on chromosome 3 (AGTR1), chromosome 7 (CHN2), and chromosome 18 (CNDP1). In the SUMMIT consortium, which lived up to its name, international collaborators studied individuals with type 1 diabetes and kidney disease (12,540 individuals in the GWAS, 997 subjects with whole-exome sequencing). Importantly, this 2016 report was unable to confirm single-allele associations with nephropathy in either study, suggesting that many prior studies suffered from type 1 error. Nevertheless, the study had important findings, including a role for smoking and obesity; pathway analysis, combining results for multiple genes, suggested a role for ascorbate and aldarate metabolism, as well as pentose and glucoronate interconversions.¹²

Nondiabetic Nephropathy

The quest for a genetic locus that would explain the increased risk among African Americans for CKD and ESKD lead two groups to perform MALD studies. These studies resulted in the simultaneous reports of a peak of excess African ancestry on chromosome 22 in a study of African Americans with focal segmental glomerulosclerosis and human immunodeficiency virus–associated nephropathy¹³ and in a study of nondiabetic kidney disease,¹⁴ both published in 2008. The gene directly under the peak was MYH9, encoding myosin heavy chain IIA. Subsequently, the main driver of this admixture peak was shown in 2010 to be APOL1, as discussed later.

Other loci also have been implicated in the pathogenesis of nondiabetic nephropathy. The protein reticulon 1, encoded by RTN1, contributes to the function of the endoplasmic reticulum and laboratory studies have suggested that it might contribute to CKD. Endoplasmic reticulum stress may protect cells from injury or alternatively may activate apoptotic pathways. Genetically induced dysfunction of the endoplasmic reticulum caused albuminuria in mice (reviewed by Cybulsky¹⁵). An isoform of RTN1A may contribute to CKD progression. Bonomo et al¹⁶ studied 219 SNPs in this gene and found that three intronic SNPs of RTN1 were associated with type 2 diabetic ESKD, with an OR of 0.67 to 0.77, and with the minor allele being protective. The effect was seen in African Americans and European Americans with type 2 diabetes, and African Americans with nondiabetic ESKD. The analyses suggested either a dominant model (one variant copy increased risk) and an additive model (one variant copy increased risk and two variant copies increased risk more), depending on the population.

In 2012, Bostrom et al¹⁷ performed a GWAS using samples from nondiabetic African Americans with CKD and identified APOL1 (discussed later) and CFH as candidate genes associated with CKD. These investigators identified six variants (5 exonic, 1 intronic) in complement factor H (encoded by CFH), variants of which promote complement activation within the glomerular mesangium via the alternative pathway. One SNP was associated with both diabetic and nondiabetic ESKD among African Americans, with an OR ranging from 0.8 to 2. Complement factor H is a plasma protein that regulates complement activation on cell surfaces, thereby limiting activation on normal cells while not interfering with complement activation on bacteria and viruses. Mutations in CFH have been associated with diverse conditions, including atypical hemolytic uremic syndrome and age-related macular degeneration; more limited evidence suggests a possible role in schizophrenia and ischemic stroke. In a follow-up study, Freedman et al¹⁸ and Bonomo et al¹⁹ examined 10 CFH SNPs, including those located in exons, an untranslated region, and within an intron. Of these SNPs, six were associated with nondiabetic ESKD and three were associated with type 2 diabetic ESKD. In an adjusted model, the strongest association with nondiabetic ESKD was for rs3753396 (a synonymous exonic SNP) with an OR of 1.62 (95% CI, 1.19–2.21) in a dominant model, whereas the strongest association with diabetic ESKD (but not with nondiabetic ESKD) was an exonic SNP, predicted to be benign, changing Asn to Tyr (rs35274867), with an OR of 2.0 (95% CI, 1.2–3.4).

Freedman et al¹⁸ also have identified variants in RREB1, encoding Ras-responsive element binding protein 1, as being associated with ESKD. The protein is a regulatory factor that represses transcription of the angiotensinogen gene, and thus modulates activity of the renin-angiotensinogen system. The variants, which are protective, are rs9379084 (a missense variant, D1771N) and rs41302867 (an intronic splice variant); these variants may be in linkage disequilibrium. Both variants are associated with protection against African American and European American type 2 diabetes ESKD, but not African American nondiabetic ESKD or hypertension-attributed ESKD.

Oxidative stress manifests as increased levels of reactive oxygen species, which are essential for normal cell function but when present in superabundance, can result in cellular dysfunction and progressive tissue injury. Two recent articles linked variants in genes related to cellular defenses against oxidative stress to ESKD. GSTM1, encoding glutathione-S-transferase mu 1, contributes to clearance of electrophilic toxins that bind glutathione; these toxins include carcinogens, therapeutic drugs, environmental toxins, and endogenous metabolites that arise as a consequence of oxidative stress. Without effective clearance, these electrophiles accumulate and damage DNA, proteins, and lipids. GSTM1 has a common null allele: approximately 50% of European Americans and approximately 25% of African Americans lack a functional allele. Tin et al²⁰ examined genotypes in the Atherosclerosis Risk in Communities (ARIC) study and found that individuals with zero or one copy of GSTM1 null variant had a higher risk for ESKD (OR, 1.7; 95% CI, 1.3–2.2) and heart failure (OR, 1.2; 95% CI, 1.04–1.3). The effect was similar in European Americans and African Americans. Second, investigators from Taiwan studied promoter variants in the HO-1 gene, encoding heme oxygenase 1, in hospitalized patients with coronary heart disease.²¹ Heme oxygenase 1 is the rate-limiting enzyme in heme degradation and plays a cytoprotective role, particularly against oxidant stress. HO-1 is inducible, whereas HO-2 is expressed constitutively. The HO-1 promoter has variable numbers of dinucleotide GT repeats, and the number of repeats correlates inversely with HO-1 messenger RNA level and enzyme activity. Previous studies had associated these variants with increased susceptibility to CVD events and mortality but a role for HO-1 variants in CKD had not been explored. Chen et al found that an increased number of GT repeats was associated with the composite renal outcome (doubling of creatinine or ESKD), with an OR of 2.0 (95% CI, 1.3–3.1), as well as cardiovascular events and mortality in this general population study. Plausibly, HO-1 variants may contribute to similar outcomes in the ESKD population.

Sickle cell trait, which confers protection against cerebral malaria, is associated strongly with CKD and ESKD. In the Reasons for Geographic and Racial Differences in Stroke study, 5.4% of sickle cell trait carriers compared with 2.6% of noncarriers progressed to ESKD. This relationship was independent of age, hypertension, diabetes, and APOL1 high-risk genotype.²² The mechanism for the association with both CKD and ESKD may be similar to glomerular injury in patients with sickle cell disease: the sickled hemoglobin causes hypoxia in the renal medulla and ischemic-reperfusion injury, leading to glomerulosclerosis.²³

GENETICS OF CARDIOVASCULAR DISEASE IN ESKD

Cardiovascular disease is a major source of morbidity and the primary cause of death among ESKD patients. As shown in Table 3, approximately 52% of dialysis patients will die a cardiovascular death. Manifestations of coronary heart disease include angina, myocardial infarction, and congestive heart failure. Manifestations of cerebrovascular disease include transient ischemic attacks and cerebrovascular accidents; the mechanisms include thrombosis of the cerebral arteries and rupture of endovascular plaque. Together, these are complex processes, driven by multiple factors. Atherosclerosis is accelerated in CKD in general and in ESKD in particular by hypertension and by uremic toxins, including homocysteine, indoxyl sulfate, asymmetric dimethyl arginine, advanced glycation end products, advanced oxidation protein products, as well as abnormalities in parathyroid hormone, vitamin D, calcium, and phosphate. Arrhythmias are more common among dialysis patients owing to multiple factors including altered and fluctuating plasma electrolyte levels.

Table 3.

Causes of Mortality Among Dialysis Patients in the Medicare Population in 2014

Cause	Fraction
Arrhythmia, cardiac arrest	40%
Acute myocardial infarction and atherosclerotic heart disease	6%
Congestive heart failure	3%
Other cardiac	3%
Stroke	3%
Septicemia	8%
Other infection	3%
Malignancy	4%
Withdrawal from dialysis	17%
Other causes	14%

The percentages shown are for causes of death among dialysis patients in 2015, with patients with missing causes of death excluded; these patients comprised 9% of the total number of deaths. Cardiovascular causes of death (the first 6 causes listed) accounted for 55% of deaths. The percentages sum to 101% owing to rounding.

Reprinted from US Renal Data Systems Report, 2017, Table 5.2.⁵¹

Genetic factors may contribute to cardiovascular disease in CKD, including ESKD, and this is the focus in the present article. In Spain, Rodrigo et al²⁴ devised an innovative approach to assessing cardiovascular risk in CKD patients that included both clinical and genetic variables. They devised a genetic risk score that involved variants in 9 genes that have been shown to be associated with coronary heart disease. They applied the score to 632 subjects with stage 4 or 5 CKD, or were on dialysis, or had a functioning allograft. The score was associated with CVD risk (hazard ratio, 1.34; 95% CI, 1.04–1.71). A prediction score that incorporated age, sex, and disease state (the status noted earlier) as variables had a C-statistic of 0.62, while incorporating the genetic risk score increased the C statistic to 0.70. Although this represents a modest improvement in prediction, it does suggest that as this field evolves, genetic risk scores might have clinical application, particularly with regard to the selection of interventions to reduce cardiovascular risk.

Sperati et al²⁵ hypothesized that cellular inflammation would increase cardiovascular risk so they examined known genetic variants in the Janus kinase/signal transducer and activator of transcription (STAT) signaling pathway. They studied tag SNPs, which mark a haplotype for a particular genetic region, in this case the coding region and nearby noncoding region for these genes. Thus, each tag SNP represents a genomic region that tends to be inherited as a block. They found that tag SNPs for Janus kinase 3 (but not STAT4 and STAT6) were associated with increased rates of cardiovascular events in dialysis patients; the ORs were approximately 2 for each association.

Sirtuins (SIRTs) are proteins with diverse functions. SIRT1 is a deacetylase that influences metabolism and inflammation, and participates in DNA repair. Increased expression of SIRT1 extends their lifespan. In endothelial cells, SIRT1 suppresses nuclear factor-ĸB activation, and in the endothelium in vivo, SIRT1 promotes vasodilation via stimulation of nitric oxide production.²⁶ Shimoyama et al²⁷ showed that three tagging SNPs (ie, those that identify haplotypes composed of multiple DNA variants) were associated with the coronary calcification score in Japanese hemodialysis patients.

There is increasing attention to the role of microRNAs (miRNAs) in post-transcriptional regulation of gene function. miRNAs are 22 base pairs in length and typically are complementary to the 3’ untranslated region, which promotes degradation of the transcripts. Chao et al²⁸ reported an inverse association between serum miRNA 125b levels and vascular calcification in 88 subjects receiving chronic hemodialysis and confirmed the association in another cohort of 135 subjects. The miRNA presumably acts on vascular cells but at present there is no identified molecular target.

APOL1 GENETIC VARIANTS AND KIDNEY DISEASE

The identification in 2010 of APOL1 genetic variants as a driver of kidney disease in individuals of African descent^29,30 opened a major new chapter in our understanding of the genetics of a diverse spectrum of kidney disease and its complications; this is treated as a separate topic here. APOL1 high-risk (HR) status is defined as the carriage of two APOL1 renal risk alleles, G1/G1, G2/G2, or G1/G2, and low-risk status by carriage of zero or one APOL1 risk variants. APOL1 HR status has been associated with focal segmental glomerulosclerosis,^31,32 human immunodeficiency virus–associated nephropathy in the United States^31,33 and South Africa,³⁴ hypertension-attributed CKD,³⁵ and collapsing glomerulopathy in the setting of lupus nephritis.³⁶

APOL1 HR status has been associated with the following aspects of CKD and ESKD.

More Rapid Progression of CKD

APOL1 HR status has been associated with more rapid progression of CKD in the following settings: focal segmental glomerulosclerosis,³¹ hypertension-attributed CKD,^35,37 lupus nephritis,³⁸ and diabetes mellitus.³⁷ In the ARIC population cohort, APOL1 variants were associated with faster estimated glomerular filtration rate declining rates.³⁹

Higher Rates of CKD and ESKD

In population-based cohorts, in whom a renal biopsy diagnosis generally is not available, studies have shown that APOL1 HR status is associated with a higher incidence or prevalence of CKD and/or ESKD. These include studies that have involved the following cohorts: the ARIC⁴⁰, the Mt Sinai Medical Center cohort,⁴¹ and in the Healthy Aging in Neighborhoods of Diversity Across the Lifespan study.⁴² APOL1 HR status was associated with ESKD outcomes in diabetic subjects (Table 4).⁴³

Table 4.

Genetic Loci Associated With ESKD and its Complications

Gene	Protein	Function	Association
RTN1	Reticulon	Endolysomal function	ESKD
CFH	Complement factor H	Adaptive immunity	ESKD
RREB1	Ras-responsive element binding protein 1	Angiotensin pathway	T2DM-ESKD
CD2AP	CD2-associated protein	Adapter protein	T2DM-ESKD
MMP2	Matrix metalloproteinase 2	Extracellular matrix turnover	T2DM-ESKD
MYH9	Myosin heavy chain 9	Cell motility	ESKD in Europeans and white Americans
APOL1	Apolipoprotein L1	HDL and podocyte function	ESKD
GSTM1	Glutathione S-transferase mu 1	Intracellular anti-oxidant	ESKD, heart failure
HO-1	Heme oxygenase 1	Heme degradation, intracellular anti-oxidant	CKD progression, ESKD
Hbs (sickle cell trait)	Hemoglobin β	Sickle cell trait	ESKD

Genes for which variants have been associated with altered risk for ESKD or mortality among ESKD subjects are shown. T2DM, type 2 diabetes mellitus.

Younger Age at Dialysis Initiation and Longer Survival on Dialysis

APOL1 HR status was associated with earlier age at dialysis initiation in the Accelerated Mortality in Renal Replacement study.⁴⁴ Similarly, in a study of nondiabetic African Americans and Hispanic Americans, subjects with APOL1 HR status started dialysis at a younger age.⁴⁵ African American and Hispanic individuals with APOL1 HR status had longer survival while on hemodialysis (hazard ratio, 0.57) among nondiabetic but not diabetic subjects.⁴⁶ The mechanisms for these relationships are poorly understood and is a fruitful area for future research, with implications for APOL1 HR individuals and with regard to the apparent survival benefits of APOL1 HR status, and possibly implications for the general dialysis population as well.

Because the relationship between APOL1 risk alleles and progression to ESKD is highly variable, further knowledge is needed about factors that influence this relationship. Tin et al, in studies of the ARIC cohort, found statistical interactions between APOL1 variants and factor VIIIc and protein C, with regard to progression to ESKD.⁴⁷ APOL1 is expressed in the microvasculature and this new finding suggests the interesting hypothesis that blood coagulation and/or vascular injury might contribute to CKD progression and cardiovascular disease in genetically at-risk individuals.

APOL1 AND CARDIOVASCULAR DISEASE

The literature addressing the relationship between APOL1 risk variants and cardiovascular risk has been complex and contradictory. Table 5 summarizes the results from population-based studies, modified from data summarized by Estrella and Parekh.⁴⁸ The only study examining CKD patients and involving children found an association with left ventricular hypertrophy.⁴⁹ Table 6 summarizes studies relating APOL1 genetic variants to all-cause mortality.

Table 5.

APOL1 Associations With Cardiovascular Outcomes

Study	N	Design	Outcome	Effect size
Jackson Heart Study	1,959	Cohort	Composite CVD events	Hazard ratio, 1.8*
Women’s Health Initiative Study	749	Cohort	Composite CVD events	Hazard ratio, 3.2*
Systolic Blood Pressure Intervention Trial	2,571	Cross-sectional	Self-reported history of CVD events	Hazard ratio, 1.0
Cardiovascular Health Study	798	Cohort	Incident myocardial infarction	Hazard ratio, 1.8*
ARIC	5,556	Cohort	Incident CVD	IRR, 0.88
Chronic Kidney Disease in Children	140	Cohort	Left ventricular hypertrophy	OR, 6.2

The results of six studies examining the role of APOL1 HR status on CVD are shown. Studies using the cardiac calcification score as an outcome were not included. The composite CVD events included incident myocardial infarction, stroke, or surgical or endovascular intervention. IRR, incident rate ratio.

^*Statistically significant. Adapted with permission from Estrella and Parekh,⁵² with an additional study included.⁵³

Table 6.

Association of APOL1 HR Status and Mortality

Study	Population	N	All-cause mortality	CVD mortality	Non-CVD mortality	Reference
Cardiovascular Health Study	General population	789	Decreased survival: P = .03; Increased mortality: HR, 1.3; 95% CI, 1.10–1.7; P =.05	NS	NS	54
African American Diabetes Heart study	Diabetes type 2	717	Decreased mortality: HR, 0.44 (two copies); P = .005, additive model; HR, 0.67 (one copy); dominant model, P = .078	ND	ND	18
Wake Forest/Emory	Diabetic ESKD	450	No effect	NS	ND	10
Wake Forest/Emory	Nondiabetic ESKD	275	Decreased mortality: P =.024; P =.038	ND	ND	10

Cohort studies that present data on the effect of APOL1 HR status on mortality are shown.

ND, not determined; NS, not significant.

GENETICS OF OTHER ESKD COMPLICATIONS

ESKD is associated with dysfunction in other organ systems, including bone physiology, host defense against infection, and hematopoietic function. To date, there is little evidence that genetic variants contribute to these functions in the setting of ESKD. Nonetheless, this is a potentially productive area for future research.

PRECISION MEDICINE: CONCLUSIONS

According to the National Institutes of Health Precision Medicine Initiative, “precision medicine is an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person.”⁵⁰ Here, we have summarized genetic variants that have been associated with ESKD and with ESKD complications, particularly those involving the cardiovascular system. It appears that for all genetic loci discussed, there are no trials as yet that examine whether genetic testing, followed by provision of these results to the clinician and subject, with or without an intervention, alters outcomes.

To exploit (for our patients’ benefit) the recent advances in the genetics of common kidney diseases and complications of CKD and ESKD, and particularly cardiovascular complications, we need a sequential approach. First, after securely identifying causal genetic variants, we need mechanistic studies, in cultured cells, experimental animal models, and human subjects to help us understand the impact each variant has on relevant pathways and to identify therapeutic approaches that will return these pathways to physiologic function, or at least to counter harmful effects. Second, with these approaches clearly defined, carefully designed clinical trials will be needed. Third, studies should examine whether a return of results, including information that an individual with CKD is at increased risk for progression to ESKD and/or for complications during progression of CKD and while in ESKD, would alter patient behavior (eg, adherence to diet, medications, and clinic visits) and physician behavior (selection of therapy when there are informed choices to be made, frequency of visits, and extent of patient counseling). Importantly, these studies should examine whether these approaches improve patient outcomes with regard to CKD progression and/or survival.