Leveraging Real-World Patient Data to Support Recruitment in Diabetic Kidney Disease Studies

Practicing physicians frequently obtain laboratory assessments of kidney function in their routine management of patients with diabetes. Two tests that are commonly performed are the estimated glomerular filtration rate (eGFR) and the urinary albumin toDiabetic Kidney Disease Studies creatinine ratio (ACR). Results of these tests are often used to determine patient eligibility for clinical trials of drugs to treat patients with diabetic kidney disease (DKD).

One challenge that drug developers and clinical trialists face is in choosing eGFR and ACR criteria that support the aims of the clinical study—without hindering recruitment.

To address this issue, researchers at Covance and LabCorp queried a LabCorp database of 329,841 diabetic patients to analyze real-world data. They wanted to understand the distribution of eGFR and ACR values among diabetic patients in the United States and assess how these laboratory parameters predicted renal disease progression.

Defining the diabetic patient group

The team started by defining the data requirements. They first queried laboratory data for diabetic patients, 18-75 years old, without renal transplant, pregnancy or end-stage renal disease.

Then the team evaluated this group’s eGFR and ACR test results. Patient data were included in the analysis if there was at least one ACR and at least three eGFR results available between November 2011 and August 2015.


eGRF is the estimated glomerular filtration rate. This test utilizes a creatinine measurement in the blood, which is then included in a formula that factors in age, ethnicity and gender to create the estimated GFR (eGFR).  As kidney disease increases, the eGFR value decreases.

ACR is the albumin to creatinine ratio. It requires measurement of albumin and creatinine in a urine sample.  If a patient’s urine sample has the too much of a protein called albumin in it, it is called albuminuria and indicates that the kidneys are not properly filtering blood.


Looking at the population as a whole, this group of 329,841 diabetic patients had a similar number of men and women and an average age of 59 years.

Performing the analyses

To understand the relationship between the two tests, the team first compared the patients’ initial eGFR test results against their degrees of initial albuminuria. As expected, patients with higher levels of urinary albumin to creatinine ratio (ACR) had lower initial eGRF values.

Next, the team analyzed the rate of change of eGFR values as compared to ACR values. They also divided the data into subgroups (normal, mild, moderate or severely reduced eGFR) to evaluate initial eGFR values compared to the degree of albuminuria present.

Understanding the findings

When setting the eligibility criteria for a clinical study, it is important to understand if certain parameters can restrict recruitment and lower the possible pool of qualified patients. For this project, the team wanted to determine how to most effectively use both eGFR and ACR as eligibility criteria.

Leveraging the LabCorp patient database, the team found that the average rate of eGFR decline was low at 2.2 mL/min/year.  The team took a deeper look at the data and found that the ACR level was closely related to the eGFR decline rate.  Although the rate of decline was greatest in patients who have very high ACR results, the team observed higher rates of eGFR decline – even in patients whose ACR was only slightly increased.

 Implications for clinical trial design

These findings support the inclusion of patients with lower ACR levels in trials of drug treatment of DKD patients. This data can be useful for drug companies in choosing ACR and eGFR eligibility criteria to optimize the design of their clinical trials.

You may be interested in last week’s DKD blog:

Assessing Pharmacokinetics in Diabetic Patients with Impaired Renal Function