Predicting Kidney Disease Progression in APOL1 Variants Using Protein Biomarkers
Hidden Risks of APOL1-Associated Kidney Disease
While APOL1(G1/G2) variants are known to be a major cause of focal segmental glomerulosclerosis (FSGS) in African Americans, accurately predicting when the disease will progress has been challenging. Existing genetic testing alone cannot distinguish the onset of proteinuria or the rate of glomerular fibrosis, creating a significant gap in personalized patient management. In this context, capturing protein expression patterns that reflect the microenvironment of kidney tissue has emerged as a new challenge. However, with thousands of proteins fluctuating simultaneously, identifying meaningful signals is difficult without high-sensitivity mass spectrometry and large-scale cohorts. The research team aimed to overcome these technical and statistical barriers and interpret the pathological mechanisms associated with APOL1 variants at the molecular level.
Building a Protein-Based Predictive Model and Identifying Key Signaling Pathways
The researchers performed LC-MS/MS-based quantitative profiling of serum proteins from 1,200 APOL1 carriers and refined data from over 3,500 peptides. They then used LASSO and Elastic Net algorithms to select 27 candidate proteins most strongly correlated with kidney damage, among which APOL1-associated signals, such as NF-κB(p65) activation and complement C3 precursor, were prominent. In particular, a protein network linked to mitochondrial stress marker PGC-1α, observed in APOL1 variant cell lines, improved prediction accuracy by 15%. A score model combining the selected biomarkers achieved an ROC AUC of 0.92 and detected kidney function decline within 3 years more than twice as early as existing genotype-based risk scores. In this process, the research team discovered a new link through protein-protein interaction maps and KEGG pathway analysis, revealing that APOL1 variants cross-activate the JAK-STAT and MAPK pathways.
New Biomarkers and Linking to Cell-Level Pathology
When the model was applied to an independent cohort of 500 validation subjects, the prediction score increased most significantly in patients with high levels of CCL2 and CXCL10, which are associated with the activation of CD68+ macrophages in the kidney. These chemokines support the mechanism in which APOL1 variant podocytes promote TRPC6 channel overexpression, leading to cytoskeletal rearrangement and eventual rupture of the glomerular filtration barrier. Furthermore, it was additionally revealed that an increase in the ratio of TIMP-1 and MMP-9, signaling proteins detected in serum, accelerates ECM (extracellular matrix) remodeling and progresses glomerular sclerosis. The risk score based on these molecular signatures showed 30% higher sensitivity than existing eGFR-based estimates, and early therapeutic intervention significantly reduced glomerular hypertension. As a result, the research team presented a complex mechanism in which the protein network associated with APOL1 variants simultaneously regulates immune, metabolic, and cytoskeletal signals, providing important clues for the discovery of next-generation therapeutic targets.
Prospects for Clinical Application and Future Treatment Strategies
This proteomics-based score is now expected to enable precise stratification of kidney risk in APOL1 variant patients in clinical settings with a single blood test. In particular, ongoing Phase II clinical trials suggest that personalized application of a combination therapy of ACE inhibitors and SGLT2 inhibitors based on the score may reduce the proportion of patients requiring dialysis within 5 years by more than 40%. At the same time, a new drug development pipeline based on the biomarker panel is underway in collaboration with pharmaceutical companies, with the goal of submitting the first IND application by 2028. Market research suggests that the market size for APOL1-related chronic kidney disease therapeutics is expected to exceed $800 million by 2025, and early diagnosis technology will be a key driver of this growth. Therefore, future multi-institutional large-scale cohort studies and integrated multi-omics studies combining genomic, transcriptomic, and metabolomic data are expected to play a crucial role in further refining this model and realizing personalized kidney protection strategies.
Nature Genetics, Published online: 12 June 2026; doi:10.1038/s41588-026-02650-7Proteomic prediction of APOL1-associated kidney disease
Approximately 13 million African Americans with APOL1 variants are currently at risk of rapid kidney function decline, but existing tests cannot accurately determine when treatment should begin. In the past, the use of blood creatinine levels or genetic testing alone failed to detect early-stage micro-damage, and clinical practice often involved treating already progressed nephrotic syndrome. This study precisely analyzed 3,500 peptides and created a predictive model by combining 27 key proteins using machine learning, which enabled the identification of high-risk individuals more than twice as quickly as existing methods. As a result, early identification of high-risk patients allows for personalized administration of existing drugs such as ACE inhibitors and SGLT2 inhibitors, potentially reducing the need for dialysis within 5 years by more than 30%, with an expected annual cost savings of billions of dollars. In the future, applying this proteomics score to multi-institutional clinical trials and collaborating with pharmaceutical companies to develop new kidney-protective agents could reduce the number of APOL1-related kidney failure patients by more than 50% by 2030.