Apolipoprotein L1 high-risk (APOL1 HR) genotypes—defined by the G1/G2 variants at the APOL1 locus1—are increasingly being tested in clinical practice and as novel targets. However, only some individuals with APOL1 HR have CKD or kidney failure. This indicates that along with social and clinical risk determinants, genetic background contributes to the APOL1 HR kidney disease relationship. However, few genome-wide significant genetic variants have been identified that affect the APOL1 HR kidney disease relationship.

A polygenic risk score (PRS) aggregates the cumulative effects of millions of common genetic variants.2 A recent study has shown that a linear combination of PRS and APOL1 genotype enhances CKD prediction across ancestries.3 However, the study did not investigate further the performance of their score among APOL1 HR.

In this article, we evaluated the performance of a PRS for CKD stage 3 or higher in individuals with APOL1 HR. We tested the PRSs using African American (AA) individuals from BioMe, an electronic health record–linked diverse clinical cohort, and then validated the result in the All of Us Research Program data.

We first generated two new PRSs (PRS1 and PRS2) using the same method by which the most recent PRS for CKD (PRS3) was developed (Khan et al., 2022).3 The first score, PRS1, differs from PRS3 in using individuals of African ancestry in BioMe as the optimization cohort instead of European ancestry (Figure 1). Our second score, PRS2, differs from PRS1 in using effect size from different summary statistics. Namely, we computed PRS2 on the basis of summary statistics from a genome-wide association study (GWAS) for the eGFR in an African American cohort of 16,474 participants.5 We then tested for the association between the three scores with CKD stage 3 and higher, defined by a validated electronic algorithm in BioMe.6 In All of Us, we used preidentified cases from the database on the basis of diagnostic codes. We calculated variance explained for the PRS component, using the Nagelkerke's pseudo R2 and adjusted odds ratios (aOR) from logistic regression models adjusted for age, sex, type 2 diabetes, and four principal ancestry components.

Figure 1:

The performance of the PRSs for CKD stage 3 and above in APOL1 HR. (A) The performance of the PRSs for CKD stage 3 and above in optimization cohort (AAs in BioMe). (B) The performance of the PRSs for CKD stage 3 and above in testing cohort (AAs with APOL1 HR in BioMe) and in validation cohort (AAs with APOL1 HR in All of Us). (C) The proportion of CKD stage 3 and above by PRS1 quintiles in testing (left) and validation (right) cohort. We calculated variance explained for the PRS component using the Nagelkerke's pseudo R2. We calculated aOR from logistic regression models adjusted for age, sex, type 2 diabetes, and four principal ancestry components. All PRSs were normalized before association testing. *PRS3 is a linear combination of a PRS optimized for the UK Biobank individuals of European ancestry and the APOL1 HR genotype. AA, African American; aOR, adjusted odds ratio; APOL1 HR, apolipoprotein L1 high-risk; CI, confidence interval; GWAS, genome-wide association study; PRS, polygenic risk score.

In the optimization cohort of 10,497 BioMe AAs, PRS1 obtained with tuning parameters r2=0.2 and P = 0.3 (see the pruning and P value thresholding method in the study of Khan et al. 20223) and PRS2 obtained with r2=0.2 and P = 0.1 were best associated with CKD. PRS2 performed better than PRS1 (aOR=1.51, P = 5.05×10−18, and R2=1.2% versus aOR=1.36, P = 1.08×10−4, and R2=0.93%). However, in the testing cohort of 1452 APOL1 HR, PRS1 performed better than PRS2 (R2, 0.97% versus 0.59%). When we compared the performance of PRS1 and PRS2 with the most recently developed PRS (PRS3) for CKD across ancestry, we observed better performance of both PRS1 and PRS2 in the testing cohort (Figure 1, A and B).

We replicated the analysis in an independent validation cohort of 6118 AAs with APOL1 HR from All of Us and observed a significant association of PRS1 and PRS2 with CKD (Bonferroni-corrected P value = 0.017). Consistently, the most significant association was observed for PRS1 (aOR=1.44; 95% confidence interval [CI], 1.16 to 1.80; P = 8.80×10−4). The association was weaker with PRS2 (aOR=1.19; 95% CI, 1.05 to 1.35; P = 0.005) and PRS3 (aOR=1.14; 95% CI, 1.02 to 1.29; P = 0.03) (Figure 1B).

We further examined the changes in CKD prevalence along the PRS1 distributions. We found that the prevalence of CKD increases from 34% in the first quintile to 42% in the last quintile in BioMe and from 6% to 8% in All of Us (Figure 1C).

We showed that polygenic risk weakly associates with CKD in individuals with APOL1 HR. Although no significant interactions between APOL1 genotypes and PRS were detected (data not shown), the associations of PRS differed in APOL1 HR versus in AAs in general. We also showed that in individuals with APOL1 HR, polygenic scores computed from summary statistics for eGFR in AAs outperformed the most recently developed PRS3 (R2, 0.5% versus 0.2%), suggesting the polygenicity of kidney disease among APOL1 HR is better captured by these summary statistics.

Understanding polygenic and APOL1 HR interplay has clinical implications. APOL1 genetic testing is now frequent, especially in transplant evaluation and direct-to-consumer testing. Thus, it is important to understand which individuals will develop the disease and polygenicity is a component of disease risk. Although the elevation of CKD risk by PRS is mild (the top quintile has 1.3 times higher risk of developing CKD than the bottom quintile), there may be value to including polygenic risk as a genetic marker, along with other nongenetic risk factors, for more optimal kidney allocation and improved pretransplant living donor counseling. However, further validation and implementation testing is needed before practical application.

Disclosures

S.G. Coca reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. S.G. Coca reports consultancy for 3ive, Axon, Bayer, Boehringer-Ingelheim, Nuwellis, Renalytix, Reprieve Cardiovascular, Takeda, and Vifor; ownership interest in pulseData and Renalytix; research funding from ProKidney, Renal Research Institute, Renalytix, and XORTX; patents or royalties from Renalytix; advisory or leadership role as an Associate Editor for Kidney360; and role on the Editorial Boards of CJASN, JASN, and Kidney International. R. Cooper reports research funding from NIH. R. Do reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. R. Do reports consultancy for Pensieve Health (pending) and ownership interest in Pensieve Health (pending). O.M. Gutierrez reports consultancy for QED; research funding from Amgen; honoraria from Akebia, Amgen, Ardelyx, AstraZeneca, and QED Therapeutics; and an advisory or leadership role as an Associate Editor of CJASN. C.R. Horowitz reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. R.J.F. Loos reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. R.J.F. Loos reports consultancy for Eli Lilly and Regeneron and an advisory or leadership role as a Board member of the European Association of the Study of Diabetes. G.N. Nadkarni reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. G.N. Nadkarni reports consultancy for Daiichi Sankyo, GLG Consulting, Qiming Capital, Reata, Renalytix, Siemens Healthineers, and Variant Bio; ownership interest in Data2Wisdom LLC, Doximity, Nexus iConnect, Pensieve Health, Renalytix, and Verici; research funding from Renalytix; honoraria from Daiichi Sankyo; patents or royalties from Renalytix; advisory or leadership role for Renalytix; and speakers bureau for Daiichi Sankyo. A. Sakhuja reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. A. Sakhuja reports an advisory or leadership role for Carolinas/Virginia's chapter of SCCM—unpaid, current funding from NIH/NIDDK 5K08DK131286 (PI: A. Sakhuja) and past funding from NIH/NIGMS 5U54GM104942 (PI: Sally Hodder). A. Sawant reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix. A. Sawant reports ownership interest in Tesla, Inc. and provisional patent application Serial No. 63/500,262; “System, Method, And Apparatus For Facilitating Spatial Resolved Temporal Networks”; Filed: May 4, 2023. H.M.T. Vy reports employment with Icahn School of Medicine at Mount Sinai; Mount Sinai owns part of Renalytix.

Funding

G.N. Nadkarni: National Institute of Diabetes and Digestive and Kidney Diseases (R01DK127139).

Acknowledgments

The All of Us Research Program is supported by the National Institutes of Health, Office of the Director: Regional Medical Centers: 1 OT2 OD026549; 1 OT2 OD026554; 1 OT2 OD026557; 1 OT2 OD026556; 1 OT2 OD026550; 1 OT2 OD 026552; 1 OT2 OD026553; 1 OT2 OD026548; 1 OT2 OD026551; 1 OT2 OD026555; IAA: AOD 16037; Federally Qualified Health Centers: HHSN 263201600085U; Data and Research Center: 5 U2C OD023196; Biobank: 1 U24 OD023121; the Participant Center: U24 OD023176; Participant Technology Systems Center: 1 U24 OD023163; Communications and Engagement: 3 OT2 OD023205; 3 OT2 OD023206; and Community Partners: 1 OT2 OD025277; 3 OT2 OD025315; 1 OT2 OD025337; 1 OT2 OD025276. In addition, the All of Us Research Program would not be possible without the partnership of its participants. Because Dr. Orlando M. Gutiérrez is an Associate Editor of CJASN, he was not involved in the peer-review process for this manuscript. Another editor oversaw the peer-review and decision-making process for this manuscript.

Author Contributions

Conceptualization: Girish N. Nadkarni, Ha My T. Vy.

Data curation: Ha My T. Vy.

Formal analysis: Ha My T. Vy.

Funding acquisition: Girish N. Nadkarni.

Methodology: Ron Do.

Resources: Richard Cooper, Orlando M. Gutierrez, Carol R. Horowitz, Ruth J.F. Loos, Ankit Sakhuja, Ashwin Sawant.

Supervision: Steven G. Coca, Girish N. Nadkarni.

Writing – original draft: Ha My T. Vy.

Writing – review & editing: Steven G. Coca, Girish N. Nadkarni.

Data Sharing Statement

Previously published data were used for this study. Wuttke M, et al. A catalog of genetic loci associated with kidney function from analyses of a million individuals. Nat Genet. 51, 957–972 (2019). Pattaro C, Teumer A, Gorski M, et al. Genetic associations at 53 loci highlight cell types and biological pathways relevant for kidney function. Nat Commun. 7, 10023 (2016).

References 1. Pollak MR, Friedman DJ. APOL1 and APOL1-associated kidney disease: a common disease, an unusual disease gene - proceedings of the Henry Shavelle professorship. Glomerular Dis. 2023;3(1):75–87. doi:10.1159/000529227 2. Fahed AC, Wang M, Homburger JR, et al. Polygenic background modifies penetrance of monogenic variants for tier 1 genomic conditions. Nat Commun. 2020;11(1):3635–3639. doi:10.1038/s41467-020-17374-3 3. Khan A, Turchin MC, Patki A, et al. Genome-wide polygenic score to predict chronic kidney disease across ancestries. Nat Med. 2022;28(7):1412–1420. doi:10.1038/s41591-022-01869-1 4. Wuttke M, Li Y, Li M, et al. A catalog of genetic loci associated with kidney function from analyses of a million individuals. Nat Genet. 2019;51(6):957–972. doi:10.1038/s41588-019-0407-x 5. Pattaro C, Teumer A, Gorski M, et al. Genetic associations at 53 loci highlight cell types and biological pathways relevant for kidney function. Nat Commun. 2016;7:10023. doi:10.1038/ncomms10023 6. Nadkarni GN, Gottesman O, Linneman JG, et al. Development and validation of an electronic phenotyping algorithm for chronic kidney disease. AMIA Annu Symp Proc. 2014;2014:907–916.