9-Protein Plasma Score Predicts Kidney Failure Risk in APOL1 High-Risk Individuals Before Visible CKD

A plasma proteomic risk score built from 9 circulating proteins can identify individuals with high-risk APOL1 genotypes who are most likely to progress to kidney failure, years before standard clinical markers such as estimated glomerular filtration rate (eGFR) or albuminuria become abnormal.

The findings, derived from the Penn Medicine BioBank (PMBB) and validated in 2 independent cohorts, offer a potential precision medicine framework to address predicting which APOL1 carriers will actually develop kidney disease, one of the most persistent gaps in nephrology.

"APOL1 variants are common — present in about one third of African Americans — and while they confer substantial risk, only a minority of individuals develop kidney disease. So the challenge is not identifying risk, but distinguishing who will get sick from who will remain healthy," said Katalin Susztak, MD, PhD, Professor of Medicine in the Renal-Electrolyte and Hypertension Division at the University of Pennsylvania Perelman School of Medicine.

What the APOL1 Proteomic Risk Score Study Found

Researchers profiled plasma proteomes using the SomaScan platform, quantifying approximately 7549 proteoforms, in 851 PMBB participants of African ancestry who carried high-risk APOL1 genotypes (G1/G1, G2/G2, or G1/G2) and had preserved kidney function, defined as ≥ eGFR of 60 ml/min/1.73m2. Using elastic net Cox regression adjusted for age, sex, eGFR, and urine albumin-creatinine ratio (UACR), the team derived a 9-protein APOL1 Proteomic Risk Score (APRS) comprising SPON1, SUMO2, EPHA10, REG3A, WFDC2, LYZ, MMP7, NPPB, and CILP2.

The composite primary outcome of a ≥ 40% eGFR decline, kidney failure, or death occurred in 18.0% of participants, with preserved eGFR in 10 years of follow-up, underscoring the event burden in this population even before CKD is clinically apparent. APRS achieved a time-dependent area under the receiver operating characteristic curve (tAUC) of 86.5% for the composite outcome, substantially outperforming the Kidney Failure Risk Equation (KFRE; tAUC 66.1%) and polygenic risk scores (tAUC 58.5%) in this group. Risk stratification by quintile showed 10-year cumulative event rates of 62.5% in the highest APRS quintile versus 3.3% in the lowest (P = 4.73 × 10−23). At 5 years, APRS demonstrated a sensitivity of 76.3% and specificity of 86.7%.1

How APRS Performs Compared to Existing Kidney Failure Prediction Tools

The KFRE, the most widely validated clinical equation for kidney outcomes, is known to perform poorly when eGFR is preserved, precisely the population where early intervention would have the greatest benefit. Polygenic risk scores, which aggregate inherited susceptibility variants across the genome, similarly showed limited discrimination in the APOL1 high-risk group, consistent with prior observations that static genetic measures provide modest predictive value in diverse populations.2

The APRS was also benchmarked against the Chronic Renal Insufficiency Cohort (CRIC) proteomic score, a 65-protein model developed in patients with established CKD. Although the CRIC score was derived from a more advanced disease population, APRS outperformed it in the preserved-eGFR group (tAUC 86.5% versus 79.0%) while performing comparably in participants with eGFR below 60 ml/min/1.73 m2.

Decision curve analysis confirmed greater net clinical benefit from APRS than either a treat-all or treat-none strategy across a range of decision thresholds. Assuming a treatment that reduces composite endpoint risk by 27%, the reduction reported for the investigational APOL1 inhibitor inaxaplin in prior trials, APRS at or above the 95th percentile, yielded a number needed to treat of 4.9, compared with 8.4 for KFRE and 23.9 for a CKD polygenic risk score.1, 3

External Validation of the 9-Protein Score Across Independent Cohorts

To assess generalizability, APRS was validated in 2 external population-based cohorts. In the Atherosclerosis Risk in Communities (ARIC) study, 314 APOL1 high-risk African American participants with preserved kidney function at baseline (mean eGFR 96.2 ml/min/1.73 m2) were evaluated; APRS achieved a tAUC of 82.2% over 10 years. In the UK Biobank, 204 APOL1 high-risk participants of African ancestry (mean eGFR 87.1 ml per minute per 1.73 m2) showed similarly strong discrimination, with a tAUC of 84.7%. Despite differences in recruitment era, geographic setting, and healthcare systems, performance remained consistent across both cohorts, supporting reproducibility beyond the academic medical center setting where the score was developed. 1

Among the 9 component proteins, 4 were significantly associated with interstitial fibrosis in independent human kidney tissue samples at both the mRNA and protein levels, suggesting biological plausibility. Susztak noted that some proteins are linked to fibrosis and tissue injury, others to inflammatory pathways, and some to endothelial dysfunction. MMP7 has been reported as a marker of tubular injury and fibrosis; WFDC2 correlates with interstitial fibrosis and rapid eGFR decline; and LYZ is associated with fibroblast proliferation and tubular cell senescence.

The investigators acknowledge that these associations are correlative and that whether these proteins actively drive disease progression or reflect altered renal clearance remains to be established through experimental validation.

What Is Needed Before Clinical Use

"APRS is currently a research tool," Susztak said. "While we have validated it in independent cohorts, including studies in the United Kingdom and NIH-funded populations, clinical implementation requires further work. The challenge is demonstrating that acting on this information improves outcomes, particularly in populations that have historically experienced a disproportionate burden of disease."

The investigators outline several translational steps, including independent prospective validation, assay standardization across platforms, and incorporation of APRS into clinical trials of APOL1-targeted therapies to determine whether score-guided enrollment and treatment decisions improve patient outcomes. As APOL1-directed agents such as inaxaplin advance through development, APRS may also serve as a pharmacodynamic readout of treatment effect and a subject enrichment tool to improve trial efficiency.3, 4

Editor’s Note: Susztak reports relevant disclosures with Ilead, Novo Nordisk, Novartis, and others.

References

1. Susztak K, et al. Plasma proteomic risk score predicts kidney failure in APOL1 high-risk individuals. Nature Medicine. 2025.

2. Inker LA, Tangri N, Greene T, et al. Kidney Failure Risk Equation performance across eGFR strata. J Am Soc Nephrol. 2019;30(9):1731-1742.

3. Olabisi OA, Zhang JY, VerPlank L, et al. APOL1 nephropathy risk variants induce cytotoxicity by disrupting cellular electrolyte homeostasis. Proc Natl Acad Sci USA. 2016;113(4):830-837.

4. Gharavi AG, Ahmad T, Roberts JD, et al. Genome-wide association analysis improves polygenic risk prediction for CKD. J Am Soc Nephrol. 2023. [Note: Verify full citation details; PRS reference sourced from study methods.]