Skip to main content
Premium Trial:

Request an Annual Quote

26,000-Patient Study Suggests Triple-Biomarker Approach Could Improve CKD Diagnosis


By Adam Bonislawski

Results from a 26,643-subject study investigating chronic kidney disease indicate that a three-marker approach combining measurements of creatinine, albumin, and cystatin C could improve prediction of patients' progression to end-stage kidney disease and death.

The findings, which were published this month in the Journal of the American Medical Association, suggest that guidelines for CKD testing, which have come under increased scrutiny in recent years, could benefit from adding urine albumin and serum cystatin C levels to definitions of the disease. Currently, detection of CKD typically relies on serum creatinine measurements alone.

"Creatinine has been the standard-of-care biomarker for kidney function," Carmen Peralta, a researcher at the San Francisco VA Medical Center and leader of the study, told ProteoMonitor. "We take creatinine and put it into an equation [to estimate glomerular filtration rate], and that yields what percent of your kidneys are working to filter [the blood], and that gives you the percent function of the kidney."

The problem with using serum creatinine levels, though, is that creatinine is a function of muscle mass, meaning that patients' baselines can vary, depending on how much muscle they have.

"You can imagine that a very old person may have very low creatinine, which would make you believe that they're filtering it all and that their kidneys are normal when in fact it's just that they don't make any [creatinine]," Peralta said. "And then you could have a huge body builder with a lot of muscle that [has] elevated creatinine and it will signal that person has kidney disease when in fact they just make more creatinine."

This issue poses the risk of "false diagnoses in people who may not have [CKD] or missing people who do have it," she said. "So we thought that if we put together two different markers for kidney function, plus a marker for kidney injury, that combination would be more likely to yield accurate identification of people who have bad things going on in their kidneys."

For an indicator of kidney injury, the researchers chose urine albumin, which is an established biomarker for kidney damage in diabetics, but hasn't typically been used outside that community.

As the second kidney function marker they chose cystatin C, which has the advantages of not being affected by subject's muscle mass or race. The protein's link to kidney function has been known for more than 15 years, Peralta said, but has only in the last six years become recognized as a potential alternative to creatinine.

For the work, the researchers used subjects enrolled in the Reasons for Geographic and Racial Differences in Stroke, or REGARDS, study – a prospective cohort study following 26,643 US adults from January 2003 to June 2010. The participants were split into eight groups defined by measurement of the three markers.

Of the 26,643 participants, 2,904 (11 percent) were classified as having CKD based on creatinine alone and 3,863 (16 percent) were classified as having CKD based only on albumin or cystatin C levels.

Adding albumin and cystatin C allowed for improved determinations of which CKD sufferers were likeliest to progress to end-stage renal disease and death. The mortality rates of participants defined by creatinine and albumin or creatinine and cystatin C were 4.4 times higher than for those defined by creatinine alone, and those with CKD defined by all three biomarkers had 7.8 times higher mortality rates than those defined by creatinine alone.

Risk of developing end-stage renal disease was highest in the subset of participants with CKD defined by all three markers and second highest in participants with CKD defined by cystatin C and albumin but not creatinine.

Looking at the data, "two very different uses" for the three-marker approach emerge, Peralta said. "One would be to confirm high risk [of CKD in patients diagnosed by creatinine] and the other would be to pick up the people who were missed by creatinine."

The evidence from the study is likely sufficient to support the first use, she said, but added that more research was needed before cystatin C could be used to diagnose CKD in patients that test negative for creatinine levels.

Peralta said her team is planning additional trials to bolster evidence supporting the use of cystatin C for this purpose. In particular, they plan to look at the protein as a biomarker in minority populations most affected by CKD and in the elderly, who are one of the most likely cohorts to receive false negatives from creatinine testing.

One challenge facing cystatin C's adoption as a CKD biomarker is its relative novelty, Peralta said.

"A lot of clinicians aren't used to it or haven't heard of it," she said. "It's not part of a routine chemistry panel. So there's still a lot of research and education to be done."

"From a technical standpoint it would very simple to [measure] creatinine and cystatin C together because you have do them from the same blood draw," Peralta noted, adding that such cystatin C assays are internationally standardized; approved by the US Food & Drug Administration; commonly offered in hospital labs and by large clinical reference labs like Quest and LabCorp; and relatively inexpensive at around $10 to $15 per assay.

"Implementation is really a matter of understanding how best it fits," she said. "The triple-marker approach is definitely not clinical standard right now, but we hope that [it] will be considered clinical practice, particularly to confirm [CKD detected via creatinine]."

Peralta's team isn't alone in pursuing a multiple biomarker approach to diagnosing CKD. In October, researchers associated with the National Heart, Lung, and Blood Institute's Framingham Heart Study published a paper in the Journal of the American Society of Nephrology suggesting that serum homocysteine and aldosterone are both significantly associated with CKD (PM 10/22/2010). Caroline Fox, medical officer for the FHS, told ProteoMonitor at the time, however, that the additional clinical trials were needed and that the research "should be considered preliminary, without clinical implications at this point."

Last week, proteomics firm Intrinsic Bioprobes launched kidney disease assays for cystatin C as well as the biomarkers beta-2-microglobulin, and retinol binding protein. Based on the company's Mass Spectrometric Immunoassay platform, the tests are designed to detect levels of different variants of the target proteins, which it believes could provide additional diagnostic information not captured by measuring total protein levels alone.

The assays are being offered out of the CLIA-certified diagnostic laboratory of the Institute for Genomics Medicine at the University of Medicine and Dentistry of New Jersey, where, according to IGM executive director Peter Tolias, researchers "are interested in undertaking a clinical trial to evaluate structural heterogeneity within these three clinical markers in patients with renal dysfunction.”

Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.