NEW YORK (GenomeWeb) – An international team of researchers has developed a qPCR-based assay to detect kidney transplant patients who are at high risk of acute rejection.
The team, led by Minnie Sarwal at the University of California, San Francisco, examined gene expression data from more than 430 renal transplant patients to develop a 17-gene set that could distinguish patients at high risk of rejection. They also developed an algorithm to classify patients into high- or low-risk categories using a numerical risk score.
This assay, dubbed kSORT for Kidney Solid Organ Response Test, "has the potential to become a simple, robust, and clinically applicable blood test," Sarwal and her colleagues said in their PLOS Medicine paper.
In the US, there are some 17,000 kidney transplants a year, according to the National Institute of Diabetes and Digestive and Kidney Diseases, and the researchers noted that acute rejection occurs in some 15 percent to 20 percent of patients even with immunosuppressive therapy. Currently, clinicians monitor creatinine levels as a marker for rejection, followed by a biopsy to confirm rejection. Sarwal and her team argued that a less invasive approach, and one that can identify rejection events early on, is needed.
To develop such a test, they collected 558 blood samples from 436 adult and pediatric renal transplant patients in the US, Mexico, and Spain. Using a set of 143 adult samples, the researchers found 31 genes that were differentially expressed between adult acute rejection (AR) and non-AR patients, as determined through kidney biopsies. A subset of 15 genes was able to classify nearly 92 percent of the AR and non-AR adult samples, but not in pediatric cases.
As Sarwal and her team wanted to develop a test that would work no matter the patient's age, they turned to a panel of 10 genes they previously developed in pediatric cases, and folded in a further seven genes to optimize the panel for both an adult and pediatric population.
These 17 genes could predict 39 of 47 AR samples correctly as AR and 87 of 96 non-AR samples as non-AR, a sensitivity of nearly 83 percent and a specificity of just more than 90 percent.
More than half of the genes in the set, the researchers noted, are either directly or indirectly involved in molecular pathways like apoptosis, immune phenotype, and cell surface. Eleven of the 17 genes have a role in cell death or survival networks, they added.
Sarwal and her colleagues tested the gene panel in a validation set of 124 samples, including 59 adults and 65 children. In this cohort, the gene panel correctly classified 21 of 23 AR samples as AR and 100 of 101 non-AR samples as non-AR, for a sensitivity of 91.3 percent and specificity of 99 percent.
The researchers also split the samples by collection site and found that the 17-gene set had similar performances at Emory University, the University of Pittsburgh Medical Center, University of California, Los Angeles, and California Pacific Medical Center transplant sites.
In a set of 191 serial samples, kSORT could predict rejection up to three months before biopsy, the researchers added.
Sarwal and her team also developed an algorithm they called kSAS to classify the patient's blood samples by comparing them to a reference AR or non-AR expression profile using a number of gene models.
The researchers then validated the use of these 13 12-gene models to classify AR risk in a dataset of 100 AR and non-AR samples. Most of the gene models agreed on whether a sample should be classified as AR or non-AR, though some were called as indeterminate. Among patients predicted to be at low risk of AR, 93.5 percent were classified correctly, and among patients predicted to be at high risk, 92.3 percent were classified correctly as AR, yielding a sensitivity of 92.3 percent and a specificity of 93.5 percent.
Sarwal and her colleagues added that kSORT needs to be evaluated in a prospective clinical trial so that its clinical efficacy may be gauged, and said that kSORT is being used to monitor patients in the Steroid Avoidance and Low-Dose CNI by ATG-Induction in Renal Transplantation, or SAILOR, study.
"The results of this trial will further [demonstrate] whether kSORT can be used as an objective and quantitative measure for the risk of AR, and whether kSORT can be used serially post-transplant to complement current clinical practice guidelines for stratifying patient immune risk, medication load, and requirement for biopsy," the researchers said.