When researchers study how a peptide affects the kidneys, they usually look at a value called the glomerular filtration rate, or GFR. This number tells clinicians how well the kidneys are filtering waste from blood. The problem is that measuring true GFR requires a radioactive tracer and careful timing, so most clinical settings rely instead on estimated GFR, which is calculated from ordinary blood markers. A recent post-hoc analysis published in a nephrology journal asked a pointed question: does a glucagon-like peptide-1 receptor agonist change those blood markers on its own, independent of any real change in kidney function?
The peptide studied is a GLP-1 receptor agonist used in people with type 2 diabetes. Research over the past several years has suggested it may have kidney-protective properties alongside its glucose-lowering effects. But if the peptide also shifts the concentration of the very markers used to estimate kidney filtration, then clinicians and researchers could be reading a false signal. An apparent improvement or decline in kidney function might be a measurement artifact rather than a true biological change.
Study design
The analysis drew on data from a randomized, double-blind, placebo-controlled clinical trial. Participants were adults with type 2 diabetes who also had albuminuria, a sign of early kidney stress. Forty-eight individuals were assigned in equal numbers to receive either the GLP-1 peptide at 1 mg weekly or a matching placebo for 26 weeks. Both groups were also taking a background glucose-lowering medication throughout the study, so the researchers could focus specifically on what the GLP-1 peptide added.
True kidney filtration, called measured GFR or mGFR, was determined using a technetium-labeled tracer that the kidneys clear from plasma at a predictable rate. This is considered the gold standard for kidney function measurement. The researchers then compared mGFR against four different estimated GFR equations, each using a different combination of blood markers: creatinine alone, cystatin C alone, creatinine and cystatin C together, and a broader panel that also included beta-trace protein and beta-2 microglobulin.
What the filtration markers actually did
Among participants receiving the GLP-1 peptide, plasma creatinine and beta-trace protein both showed small but statistically significant increases compared with the placebo group. Cystatin C and beta-2 microglobulin, the other two markers tracked, did not change significantly.
Creatinine is the marker most commonly used in everyday clinical practice to estimate kidney filtration. A rise in creatinine normally signals a decline in kidney function. Here, though, the rise occurred without any corresponding drop in measured GFR. The median change in true filtration rate was zero milliliters per minute in the peptide group, compared with a small decline of about two milliliters per minute in the placebo group. Neither change was statistically significant, meaning the kidneys themselves were filtering at roughly the same rate either way.
This disconnect matters because it suggests the GLP-1 peptide may influence how much creatinine the body produces or retains, rather than how well the kidneys clear it. One plausible explanation noted in the literature is that GLP-1 receptor agonists are associated with reductions in lean muscle mass over time, since muscle is a primary source of creatinine production. A smaller muscle pool would ordinarily reduce creatinine, but other mechanisms, such as changes in creatinine secretion by kidney tubules, could push in the opposite direction. The analysis did not fully resolve which mechanism drove the observed rise.
Which estimation equation performed best
Because different markers behaved differently under the peptide, the accuracy of the four estimation equations varied. The researchers compared each equation against the directly measured GFR at both the start and end of the 26-week period.
Equations using creatinine alone or cystatin C alone were consistently less accurate than equations combining both markers. The combined creatinine-cystatin C equation yielded the most reliable estimates of true kidney filtration at both time points. Adding beta-trace protein and beta-2 microglobulin to form the full four-marker panel did not meaningfully improve accuracy beyond the two-marker combination in this cohort.
The practical implication, at least in a research context, is that relying solely on creatinine-based estimates when studying or monitoring people receiving a GLP-1 receptor agonist may introduce bias. The creatinine signal appears to drift upward independently of actual filtration changes, which could make a stable kidney look like a declining one.
Population context and limitations
The cohort was relatively small at 48 participants, with a median age of 70 years, and only about one in six participants was female. All participants had both type 2 diabetes and albuminuria, so the findings apply specifically to that population. The analysis was post-hoc, meaning it was designed after the original trial rather than as a prespecified hypothesis, which limits the strength of conclusions that can be drawn.
The 26-week duration is also relatively short for drawing conclusions about long-term kidney trajectories. Filtration marker concentrations could behave differently over months or years as body composition continues to shift. Additionally, both groups in the original trial were already taking another glucose-lowering agent, so the results reflect the GLP-1 peptide added on top of that background treatment rather than the peptide alone.
Broader significance for GLP-1 research
This analysis sits within a larger and growing body of research examining GLP-1 receptor agonists and kidney outcomes. Several large trials have reported reductions in markers of kidney damage with this class of peptides, and regulatory agencies have taken note. But those outcomes often rely on estimated rather than measured GFR. If the peptide systematically alters marker concentrations, the apparent kidney benefit could be at least partially a measurement effect rather than a true improvement in filtration capacity.
The literature suggests that cystatin C may be a more stable reference point than creatinine in people receiving this class of peptide, since cystatin C did not change significantly in this analysis while creatinine did. Researchers designing future trials may need to account for this when choosing primary endpoints and interpreting results. The analysis reinforces a broader methodological lesson: the tool used to measure an outcome can itself be affected by the intervention being studied.
Key takeaways from the data
The post-hoc analysis found that a GLP-1 receptor agonist raised plasma creatinine and beta-trace protein without producing a corresponding change in directly measured kidney filtration rate. Cystatin C and beta-2 microglobulin remained stable. Among the estimation equations tested, the combination of creatinine and cystatin C together produced the most accurate estimates of true filtration rate, outperforming either marker used alone.
Early data points toward the importance of using multi-marker GFR estimation when monitoring people receiving this class of peptide. A creatinine-only approach may overstate any apparent decline in kidney function, while cystatin C alone may underrepresent subtle shifts in filtration. The combined equation appears to partially offset the individual limitations of each marker, offering a more grounded read of what the kidneys are actually doing.
