Predicting Progression of Chronic Kidney Disease in Sickle Cell Anemia Using Machine Learning Models (PREMIER)

ABSTRACT
Sickle cell disease (SCD) is characterized by a vasculopathy affecting multiple end organs, with complications
including chronic kidney disease (CKD). Albuminuria, an early measure of glomerular injury, is common in
SCD and predicts progressive kidney disease. Kidney function decline is faster in SCD patients than in the
general African American population. The prevalence of rapid decline in SCD is 3-fold higher than in the
general population. Furthermore, high-risk APOL1 variants are associated with an increased risk of
albuminuria and progression of CKD in SCD. Kidney disease, regardless of severity, and rapid eGFR decline
are associated with increased mortality in SCD. As such, early identification of patients at risk for progression
of CKD is important to address potentially modifiable risk factors, slow eGFR decline and reduce mortality.
Despite the high prevalence of CKD and its contribution to increased morbidity and mortality, available
treatments for SCD-related kidney disease remain limited. Although angiotensin converting enzyme inhibitors
(ACE-I), angiotensin receptor blockers (ARBs), and hydroxyurea decrease albuminuria in short-term studies,
their benefits in preventing or slowing progressive loss of kidney function in SCD remain undefined.
We have recently reported that machine learning (ML) models can identify patients at high risk for rapid decline
in kidney function. Further, higher hemoglobin concentration is also an independent predictor of decreased
odds of rapid kidney function decline. With the contribution of intravascular hemolysis to the pathophysiology of
SCD-related glomerulopathy, voxelotor, a small molecule which modifies sickle hemoglobin oxygen affinity and
improves sickle RBC survival, may decrease glomerular injury and slow the progression of CKD in individuals
with SCD.
In this application, we propose the conduct of a prospective, multicenter study to build a ML-based predictive
model for progression of CKD in adults with SCD. Furthermore, in individuals predicted to be at risk for rapid
decline in kidney function, based on the presence of persistent albuminuria (urine ACR ≥ 100 mg/g), we will
evaluate the effect of voxelotor on albuminuria, rapid decline in kidney function and progression of CKD.
With advances in the understanding of the pathophysiology of SCD and its complications, combined with an
increasing number of approved drug therapies, early identification of patients at risk for progressive kidney
disease and subsequent increased risk of death is necessary to modify known risk factors, initiate targeted
therapies and possibly increase life expectancy. Further, with the known contribution of hemolytic anemia to
the pathogenesis of SCD-related glomerulopathy and progressive kidney disease, drugs that decrease
hemolysis are likely to be beneficial in preventing and/or slowing the progression of kidney disease in this
patient population.

Grant Number: 5R01HL159376-05
NIH Institute/Center: NIH
Principal Investigator: Kenneth Ataga