Requiem for Urate Lowering in CKD?

The kidney plays a key role in the genesis of hyperuricemia, in that underexcretion of urate through renal and intestinal transport pathways is the major mechanism for human hyperuricemia. In renal epithelia, primarily within the proximal tubule, separate sets of apical and basolateral urate transporters mediate opposing reabsorption and secretion of urate, with net excretion of urate in urine determined by the relative activity of these two pathways.¹ The primacy of urate transport in urate homeostasis is underscored by the observation that variation in genes encoding urate transporters and associated regulatory proteins are the major contributors to serum urate level, despite the involvement of almost 200 genes in determining this trait. The kidney has also been labeled an unwilling accomplice in urate and salt retention associated with metabolic syndrome because of preserved renal insulin sensitivity in the face of systemic insulin resistance.² Hyperinsulinemia thus causes hyperuricemia by activating net renal urate reabsorption, and the basolateral reabsorptive urate transporter GLUT9a appears to be the major target for insulin in the proximal tubule.² In contrast, insulin-like growth factor 1, whose receptor is genetically linked to variation in the serum urate level, antagonizes insulin effects on urate transport and activates secretory urate transporters,³ resulting in a urate-reducing effect.

Notably, the fledgling literature on regulation of urate transport in hyperuricemia is dwarfed by an impressive body of animal and epidemiological data that have implicated hyperuricemia in the pathophysiology of hypertension, CKD, and cardiovascular disease.⁴ However, an evolving theme has been that despite compelling pathophysiology, interventional studies in human hyperuricemia have been disappointing. So for example, in a rat model, hyperuricemia induced hypertension and a renal arteriopathy that were responsive to urate-lowering therapy and therapy with enalapril or losartan, but not hydrochlorothiazide, consistent with a role for an activated renin-angiotensin-aldosterone system (RAAS).⁴ In humans, increasing the serum urate level progressively blunts the renal plasma flow response to infused angiotensin-II, a phenotype that is indirectly linked to an activated intrarenal RAAS.⁵ However, in overweight or obese human adults, urate-lowering therapy with allopurinol or probenecid had no effect on BP or either intrarenal or systemic RAAS activation.⁴ To some degree, the discrepancies between these human studies and animal models of hyperuricemia are reflective of the major differences in the genetics and physiology of urate homeostasis in humans versus rodents.¹ For example, in addition to the loss of a functional uricase gene in humans, xanthine oxidase transcription in humans is tightly repressed when compared with the mouse gene,¹ presumably to limit urate production in the absence of a functional human uricase enzyme. Other differences include a complete lack of GLUT9 expression in mouse proximal tubular cells, versus the abundant expression of this key urate transporter in human proximal tubules. Therefore, although considerable insight into selected aspects of uric acid homeostasis can be gained from rodent models, when it comes to uric acid pathophysiology, it seems that the human model does not always replicate the rodent disease.

With respect to CKD and diabetic kidney disease, provocative evidence in humans with type 1 diabetes indicated that hyperuricemia was associated with early reduction in GFR before development of proteinuria, suggesting that urate lowering might help prevent progression of diabetic kidney disease.⁶ These and other data led to the Preventing Early Renal Loss in Diabetes trial, published in 2020⁷, which tested the effect of allopurinol on the progression of diabetic kidney disease. Unfortunately, despite sustained and significant lowering of the serum urate level over a 3-year period, there was no appreciable difference in GFR slope between the experimental and control arms.⁷ In the CKD-FIX trial, also published in 2020⁷, 369 patients with stage 3 or 4 CKD determined to be at high risk of progression were randomized to urate lowering with allopurinol; again, there was no evidence of slower eGFR decline in the allopurinol group over the 2-year follow-up period.

Whatever the criticisms might be, the negative Preventing Early Renal Loss in Diabetes and CKD-FIX trials have inarguably lessened the prior enthusiasm for urate-lowering therapy to prevent progression of human CKD. However, some glowing embers of enthusiasm re-emerged in early 2021 with the publication of the 60-participant CITRINE trial, on the effects of a verinurad and febuxostat combination on albuminuria in patients with type 2 diabetes; this combination therapy reduced albuminuria by approximately 39%.⁸ The uricosuric verinurad, developed by Ardea Biosciences and licensed by AstraZeneca, is a high-affinity inhibitor of the apical urate-anion exchanger URAT1. Like its predecessor lesinurad, verinurad monotherapy can be associated with AKI, potentially analogous to the exercise-associated AKI seen in patients with renal hypouricemia due to loss-of-function mutations in URAT1.¹ To mitigate this nephrotoxicity, trials of verinurad have been performed in combination therapy with a xanthine oxidase inhibitor.

To follow up on the promising results of CITRINE, AstraZeneca and collaborators designed the Study of verinurAd and alloPurinol in Patients with CKD and hyperuricemia trial, the results of which are published in this issue of JASN.⁸ Owing to the results of the CARES trial, which suggested an excess of cardiovascular mortality for febuxostat compared with allopurinol,⁷ the investigators substituted allopurinol as the xanthine oxidase inhibitor for the Study of verinurAd and alloPurinol in Patients with CKD and hyperuricemia trial. The trial enrolled a total of 861 adults with CKD and eGFR ≥25 ml/min per 1.73 m², urinary albumin-to-creatinine ratio between 30 and 5000 mg/g, and serum urate ≥6.0 mg/dl. All participants were required to be receiving a stable dose of an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker for at least 4 weeks before screening. Participants were randomized to one of five treatment arms: placebo, placebo+allopurinol 300 mg/d, verinurad 3 mg+allopurinol 300 mg/d, verinurad 7.5 mg+allopurinol 300 mg/d, or verinurad 12 mg+allopurinol 300 mg/d in a 1:1:1:1:1 ratio. Ultimately, despite significant and sustained reduction in the serum urate level in the treatment arms, urinary albumin-to-creatinine ratio and eGFR change from baseline did not differ among treatment groups after 60 weeks. So again, to be succinct, a dramatically negative result for urate-lowering therapy in CKD, from a large, well-designed trial.

Does this trial really sound the death knell for urate-lowering therapy in hyperuricemic CKD? Perhaps it does for all comers with hyperuricemia and CKD. However, several lines of evidence suggest that the role of hyperuricemia in kidney disease progression deserves more attention in patients with gout and CKD, versus hyperuricemia and CKD. First, it is worth emphasizing in this context the role of monosodium urate crystals in activating the inflammasome in gout.¹ Tophi, in turn, develop out of a chronic, foreign-body granulomatous inflammatory response to monosodium urate crystals. The whole-body burden of tophi and monosodium urate deposition in gout can be estimated using dual-energy computed tomography scans. Notably, although some 25% or so of asymptomatic patients with hyperuricemia have detectable monosodium urate deposits by dual-energy computed tomography, the burden is considerably less than in patients with clinical gout.⁷ Point being, the same may apply to patients with hyperuricemia and CKD without gout that they lack extensive, whole-body monosodium urate deposition. However, a recent report emphasized the presence of medullary echogenicity on kidney ultrasounds in patients with severe gout,⁹ raising once again the specter of a bona fide gouty nephropathy due to medullary crystal deposition and medullary tophi. In addition, a recent provocative study in a mouse model of hyperuricemia—with all the caveats discussed above regarding non-human hyperuricemic models—underscored the role of medullary tophi deposition in hyperuricemic kidney dysfunction.¹⁰ Notably, it took acidic urine to generate monosodium urate crystals and medullary tophi in these mice. That observation suggests a role for urinary alkalinization in preventing kidney disease progression in gouty nephropathy, particularly given the observations of excessively acidic urine in insulin-resistant patients with a gouty diathesis and uric acid stones.¹¹ Regardless, channeling James Carville, one wonders whether, when it comes to hyperuricemia and CKD, it is the gout?

Finally, it is worth emphasizing in this context that CKD is a major comorbidity in gout. Analysis of the 2007–2008 NHANES data thus estimated that approximately 71% of patients diagnosed with gout had a diagnosis of CKD stage 2 or worse and that approximately 20% had CKD stage 4.⁷ CKD tends to compromise the treatment of gout, increasing the risks of medication toxicity and/or undertreatment because of outdated misconceptions in gout management in CKD.⁷ I would like to take the opportunity to emphasize just how much nephrologists could affect the morbidity of our patients with gout, were we to take a more proactive, involved role in their gout management.

Supplementary Material

jasn-35-525-s001.pdf ^{(1.3MB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/E609.

Funding

D.B. Mount: NIAMS (5P50AR060772).

Author Contributions

Conceptualization: David B. Mount.