Cystatin C: why clinical laboratories should be measuring it

Abstract

Cystatin C is a 13-kDa cysteine protease inhibitor that satisfies many of the criteria required of a marker of glomerular filtration rate. It can be readily measured in laboratories using automated, standardised immunoassays. Hitherto there has been reluctance to adopt cystatin C measurement in the assessment of kidney function, despite demonstrated superiority compared to the current standard of practice, serum creatinine. This may be due to increased direct reagent costs. Recent strong evidence has confirmed the superior ability of cystatin C compared to creatinine to predict risk of a range of poor outcomes (including kidney failure, cardiovascular events and mortality) among patients with chronic kidney disease. On the back of this, national and international guidelines including those of the National Institute for Health and Clinical Excellence have recommended the use of cystatin C to risk stratify individuals with mild to moderately reduced glomerular filtration rate and no albuminuria. Cystatin C presents an opportunity for laboratories to engage with their users in developing clinical pathways which will usefully focus chronic kidney disease management on those most likely to benefit while reassuring patients at low risk.

Glomerular filtration rate (GFR) is widely regarded as the best overall measure of kidney function. In recent years, the importance of this measure has increased, as it has been incorporated as the primary criterion within an international kidney disease classification system. ¹ Cystatin C was first described in 1982. ² Over the next two decades, many studies reported a close association between serum cystatin C concentration and GFR. By the beginning of the present century, the literature was mature enough for this to be tested in meta-analyses, which confirmed the superiority of cystatin C over creatinine as a marker of GFR. ³ Despite this, at present, very few laboratories offer cystatin C as part of their standard assessment of kidney function. Why is this, and is this situation likely to change?

Cystatin C is a 13-kDa cysteine protease inhibitor that satisfies many of the criteria required of a marker of GFR. It is produced at a constant rate by all nucleated cells. It is relatively freely filtered at the glomerulus and reabsorbed and catabolised within the proximal tubule, with no known extra-renal elimination or tubular secretion. It has comparatively low biological variability and appears to accurately reflect GFR throughout life, including during childhood and older age. It is unaffected by acute dietary meat intake and the spectral interferences that devalue the use of creatinine in this setting, although it does appear to be influenced by thyroid status, and possibly by obesity, independently of GFR. It appears to be clinically useful in a variety of clinical situations including malignancy, pregnancy and transplantation monitoring. Cystatin C is stable ex vivo and is readily measured by specific and precise automated immunoassays that are available on standard clinical laboratory platforms and which have recently been calibrated against a certified reference material (ERM-DA471/IFCC).

GFR is measured using reference procedures that follow the clearance of an infused exogenous substance (e.g. inulin). However, these methods are cumbersome and impractical for widespread use. Clinical laboratories report estimates of GFR using equations based on measurement of serum creatinine concentration, taking into account other variables including age, gender and ethnicity. Various equations are used by laboratories to estimate GFR. The Modification of Diet in Renal Disease (MDRD) Study equation and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine-based GFR-estimating equation have been widely used since 2000 to estimate GFR. Equations incorporating cystatin C, either alone or in combination with creatinine, have also been proposed for estimating GFR, for example the CKD-EPI-cystatin C equation, the Caucasian, Asian, paediatric and adult (CAPA) equation and the Berlin Initiative Study 2 equations. At best, such equations offer a modest improvement in accuracy over creatinine based equations. Direct reagent costs of cystatin C measurement are approximately 10-fold higher than those of creatinine. It seems likely that most health economies will not bear this increased cost on the basis of slightly improved GFR estimation alone.

Routine availability of GFR estimates have improved recognition of chronic kidney disease (CKD), but knowledge of GFR alone is insufficient to predict risk of adverse outcomes in CKD patients. There is concern that not all individuals so identified, especially those without albuminuria, are at increased risk of significant CKD-related outcomes. ⁴ Outcomes of interest in CKD include progressive loss of kidney function (including kidney failure), acute kidney injury, cardiovascular events, complications related to progression of CKD and mortality. Earlier studies focussed on the relative accuracy of GFR estimators compared to ‘true’ GFR. However, absolute accuracy in GFR estimation may not be the prime criterion and some more recent literature has focussed on the relative ability of equations to predict disease progression and poor outcomes in CKD. Such an ability would usefully focus management on those most likely to benefit, while reassuring people at low risk. This risk prediction may be achieved by use of laboratory-generated GFR estimates, in addition to albuminuria.

An increasing body of evidence has confirmed superior ability of cystatin C-based estimated GFR to predict poor outcomes. The Reasons for Geographic and Racial Differences in Stroke (REGARDS) study was a cohort study of >26,000 middle-aged north Americans. ⁵ All individuals had GFR estimated using separate cystatin and creatinine equations at baseline, in addition to urinary albumin to creatinine ratio measurements. Over a 4-year follow-up period, there were 1940 deaths. Individuals whose only abnormality was creatinine-estimated GFR <60 mL/min/1.73 m² had identical risk of death to those with creatinine-estimated GFR ≥60 mL/min/1.73 m². However, if either albuminuria was present or cystatin C-estimated GFR was also <60 mL/min/1.73 m², then the hazard ratio for death more than doubled, and if both were present, there was an additive effect. A meta-analysis of data from 16 studies including >90,000 participants found that using cystatin C in addition to creatinine predicted risk classification for kidney failure and death more accurately. These observations have been confirmed in other large-scale general and disease population studies (see Lamb et al. ⁶ ). Increasing cystatin C concentration predicts increased risk independently of GFR. This phenomenon (i.e. improved predictive ability over and above any improvement in GFR estimation) has been attributed to, as yet undefined, non-GFR determinants of serum cystatin C concentration that predict outcome.

Not all people with reduced GFR are at increased risk. Recent guidelines recommend the use of cystatin C-based GFR estimation as a confirmatory test of CKD-associated risk among the largest identified group of those with CKD (i.e. those with GFR between 45 and 59 mL/min/1.73 m² and no albuminuria: Stage 3 a, A1 CKD).^1,7 Cystatin C should be measured once at initial identification of such individuals to enable risk stratification. A cystatin C-estimated GFR <60 mL/min/1.73 m² confirms the presence of CKD, while a value ≥60 mL/min/1.73 m² refutes the diagnosis. Partly due to increased costs, cystatin C measurement outside of this population and repeated measurements within this group are not currently recommended. However, this new guidance, coming on the back of sound evidence around risk and concerns of over-diagnosis, should be seen as an opportunity for clinical laboratories to engage with their users and offer added value.