Comparing the Performance of Serum Creatinine and Cystatin C-based GFR Estimations in Predicting Directly Measured GFR in Patients With or Without Nephrotic Syndrome

Nephrotic syndrome (proteinuria >3.5 g/24-hour with or without hypoalbuminemia) can be a clinical manifestation of different primary and secondary glomerulopathy (including membranous nephropathy, minimal change disease, focal and segmental glomerulosclerosis, C3 glomerulopathy, diabetic glomerulosclerosis, amyloidosis, and many others). It is a marker of disease severity and a major risk factor for renal disease progression, worsening kidney function, and cardiovascular and thromboembolic complications. Accurate measurement of the glomerular filtration rate (GFR) is paramount to monitoring disease progression and response to treatment in patients with Nephrotic syndrome (NS).

GFR can be reliably measured by plasma clearance of exogenous markers such as inulin clearance, 51Cr-EDTA, 99mTc-DTPA, iohexol, and iothalamate. However, because measuring GFR requires specialized facilities to perform clearance measurements, serum concentrations of endogenous filtration markers are used in routine clinical practice to estimate GFR (eGFR) through different equations. Over the past 50 years, several eGFR equations focusing primarily on serum creatinine (Cr) were developed and implemented to varying degrees into clinical practice. These formulas incorporate demographic variables to explain variations in concentrations of the markers unrelated to GFR and provide rapid results. The most used endogenous filtration marker in clinical practice is Cr value or its reciprocal, alone or in conjunction with 24-hour urine collections for Cr clearance. However, Cr is produced by the normal breakdown of muscle tissue, and its concentration is inaccurate for predicting actual GFR, particularly in patients with NS. Indeed, heavy proteinuria and hypoalbuminemia with secondary fluid overload can affect Cr levels. In hypo-albuminemic patients with NS, tubular secretion of Cr may increase, leading to a misleading reduction in serum Cr level and increased creatinine urinary excretion that affects the reliability of GFR estimates based on Cr levels and even directly measured. In a pilot study with a limited number of patients, it was observed that Cr clearance overestimated GFR (measured by inulin clearance) in patients with hypoalbuminemia. Therefore, the inaccuracies associated with serum Cr measurements in NS patients highlight the need for alternative biomarkers to provide a more reliable estimate of GFR in NS patients. Cystatin C (Cys-C), a low-molecular-weight protein produced by all nucleated cells, is accumulating interest. Cys-C is a cysteine proteinase inhibitor with a molecular weight of 13.3 kD and its serum concentration is constant regardless of age, gender, and weight. After filtration into the glomeruli, most Cys-C is reabsorbed by the proximal tubule and degraded making it a possible further useful biomarker for estimating GFR. Since Cys-C levels are not influenced by non-kidney factors which instead can affect Cr levels, Cys-C could provide a more accurate estimation of GFR in healthy and disease. Moreover, the validity and accessibility of Cys-C testing have been greatly improved by the release of certified reference material for calibrating laboratory assays and by the development of new, less expensive methods for automated analyzers. The introduction of certified reference material has markedly improved the accuracy of GFR estimates with Cys-C as an endogenous biomarker, especially in patients with normal or near-normal kidney function. Furthermore, recent research has shown that combining Cys-C and Cr is more accurate by estimates as GFR and improves the classification of chronic kidney disease. However, there is no formal demonstration of superior accuracy of Cys-C-based GFR estimations compared to serum creatinine-based GFR estimations. Over 100 studies have compared the performance of Cys-C and Cr as filtration markers. However, the lack of direct comparisons of serum creatinine or Cys-C-based GFR estimations with directly measured GFR (mGFR) values did not allow us to determine which between serum Cr or Cys-C was the primary source of bias in cases of wide eGFR discordances. This is an issue of major clinical relevance in patients with nephrotic syndrome because it is already established that serum-creatinine-based GFR estimations are not accurate in this context. Thus, finding that Cys-C-based GFR estimations could overcome the limitations of serum Cr-based GFR estimations in patients with NS (with or without hypalbuminemia) would have major implications in clinics and research.

To address this issue, we will take advantage of the availability of serum samples collected by patients with membranous nephropathy associated with NS (with or without hypoalbuminemia) that are stored in our certified Biobank at the Clinical Research Center for Rare Disease - Istituto di Ricerche Farmacologiche Mario Negri IRCCS. Twelve months after specific interventions [PEPTIDE study (NCT04095156), MONET study (NCT04893096), and ORION study (NCT05050214)], some patients achieved complete (defined by a urinary protein excretion < 0.3 g/day (urinary P/C < 300 mg/g), along with normal serum albumin concentration and stable renal function) or partial (defined by a urinary protein excretion < 3.5 g/day (urinary P/C < 3500 mg/g) and a 50% or greater reduction from baseline values, accompanied by improvement or normalization of serum albumin concentration and stable renal function) remission or NS or remained persistently nephrotic. All patients had their GFR directly measured with the iohexol plasma clearance technique at baseline (before treatment, when all of them were nephrotic) and at 12-month follow-up when some of them had achieved complete or partial remission of the NS or remained persistently nephrotic.

The measurement of serum Cr and Cys-C levels in all these samples will allow us to compare the accuracy of serum Cr and Cys-C-based GFR estimation in predicting true mGFR in different clinical conditions such as humoral NS with hypoalbuminemia, NS without hypoalbuminemia, sub-nephrotic proteinuria and normal-proteinuria. In addition, to verify the actual superiority of Cys-C-based estimates vs serum creatinine-based estimates in these different clinical conditions, this analysis will allow us to confirm whether the superiority of Cys-C vs serum creat-based equations in predicting mGFR increases in parallel with the worsening of proteinuria and the NS.