Originally published April 27.
By Turna Ray
XDx last week published results of a study demonstrating that its AlloMap Molecular Expression Testing is non-inferior to standard methods of identifying which heart transplant recipients are likely to experience rejection.
The results of the study, published online in the New England Journal of Medicine, suggest that gene expression profiling of peripheral-blood specimens "may offer a reasonable alternative to routine biopsies" for certain heart transplant patients. However, future studies will be necessary before AlloMap would be considered as a substitute to biopsies, and those studies will need to be larger and have a narrower non-inferiority margin.
In the study, researchers from XDx, Stanford University, and members of the Invasive Monitoring Attenuation through Gene Expression (IMAGE) study group randomized 602 patients for monitoring of heart transplant rejection either with the use of gene-expression profiling or with the use of routine endomyocardial biopsies. The interval between transplantation and study randomization was six through 12 months for 15 percent of patients; 13 through 36 months for 68 percent; and 37 through 60 months for 17 percent.
According to Michael Pham, a cardiologist at the Stanford University Medical Center and a co-lead author of the study, the IMAGE trial is the first study in the field of cardiac transplantation to look at patient outcomes associated with the use of a biomarker.
AlloMap is an in vitro diagnostic multivariate index assay that analyzes the expression of 11 rejection biomarker genes and nine control genes at a single laboratory. The test was cleared by the US Food and Drug Administration in August 2008. Stanford University holds equity in XDx, and the company provided funding for the study.
The study authors reported that the two-year rate of the composite primary outcome — comprising rejection with hemodynamic compromise, graft dysfunction due to other causes, death, or retransplantation — was similar between the two arms, with patients in the gene expression profiling group experiencing a rate of 14.5 percent and biopsied patients experiencing a 15.3 percent rate. "Therefore, monitoring for rejection with gene-expression profiling was non-inferior to monitoring with routine biopsies with respect to the prevention of the primary outcome," the study authors wrote in the paper.
Additionally, a total of 409 biopsies were performed in the gene-profiling group, compared to 1,249 biopsies conducted for patients in the other arm. Approximately 88 percent of gene-profiling patients underwent two or fewer biopsies per patient-year, and 50 percent did not require a biopsy during the study period.
"Patient satisfaction was higher with the gene-expression profiling method of monitoring than with the biopsy method, reflecting the preference of many patients for avoiding an invasive procedure," the study authors reported.
XDx's AlloMap test, which analyzes patients' blood samples, yields a score that corresponds to patients' likelihood of transplant rejection. A score below 34 in the study conferred a low risk for rejection, and therefore, in 97 percent of study participants, biopsy was not needed. "Although the use of a higher threshold may further minimize the number of biopsies needed, the results of our trial suggest that a score below 34 represents a prudent threshold to use in clinical practice in the case of patients for whom the interval after transplantation is more than six months," the study authors noted.
Based on these results, "gene-expression profiling of peripheral-blood specimens may offer a reasonable alternative to routine biopsies for monitoring cardiac-transplant recipients for rejection if the interval since transplantation is at least 6 months and the patient is considered to be at low risk for rejection," the study authors concluded. "However, the study had limited power to allow for a firm conclusion to be reached regarding the use of gene-expression profiling as a substitute for the performance of biopsies."
For more definitive results, the researchers recommend a larger trial with a narrower non-inferiority margin and a longer follow-up period.
[ pagebreak ]
One factor in the study was the fact that many of the episodes of transplant rejection in the gene-profiling group were identified by observation of "overt symptoms of heart failure or echocardiographic evidence of graft dysfunction." Out of 34 transplant rejection cases in the gene-expression group, only six were detected through profiling with AlloMap.
"These observations raise the possibility that clinical observation may detect the majority of serious rejection episodes," the study authors point out. However, while some transplant centers in the US have stopped performing biopsies to detect rejection a year after the transplant has been performed, absent a comparative trial in this regard, many doctors have held off from broadly forgoing biopsy.
"Therefore, the majority of transplantation centers in the United States continue to perform biopsies beyond the first year post-transplantation, although there is considerable institutional variation in the frequency and duration of monitoring for rejection," the authors point out. "There has not been sufficient equipoise to justify a comparison of monitoring by means of clinical observation with monitoring by means of routine biopsies, but our findings may provide the basis for such comparisons in future studies."
The researchers also identified several limitations of the study. Since only those who had a cardiac transplant more than six months before the study started were enrolled, these study participants were inherently at a lower risk for transplant rejection than patients who had a transplant within the last six months. The decision to cautiously enroll patients at minimal risk for negative outcomes by the study researchers "reflects the characteristically conservative approach to the care of cardiac-transplant recipients and the reluctance of clinicians and patients to accept even a small possibility of causing harm."
Additionally, only 20 percent of potentially eligible patients were enrolled, with a preference toward those who had received a transplant less than three years prior to the start of the study, suggesting that patient selection was biased toward the inclusion of low-risk patients. This restricts the generalizability of the study results and should be "taken into account by clinicians when they consider the use of gene-expression profiling in the care of their patients," the study authors caution.
The study authors also noted that the low projected event rates and the small number of available patients necessitated a wide non-inferiority margin. "The trial's reduced power was reflected in a relatively wide confidence interval that does not exclude the possibility of a 33 percent decrease in primary event rates (or 1.8 fewer events per 100 patient-years) or of a 68 percent increase (3.7 excess events per 100 patient-years) among patients in the gene-profiling group," the researchers wrote in the paper.
The composite endpoint the trial pursued may have also reduced the trial's power, the authors note, adding that a "more robust test of non-inferiority would have necessitated a considerably larger sample than that which was feasible, given the limited number of cardiac transplantations performed worldwide."
Finally, researchers found that the gene expression and biopsy arms were similar in their overall intensity of immunosuppression, which may have been impacted by the lack of blinding in the study. Researchers did not observe any significant differences between the groups in mean levels of calcineurin inhibitors, serum creatinine levels, or in the incidence of neoplasms.
However, as reported in PGx Reporter sister publication The Sample, these study results were good enough for XDx to seek a change to AlloMap's labeling from the FDA to inform physicians that the gene-expression test is non-inferior to biopsy in patients who have received a transplant more than six months prior.
AlloMap, priced at around $3,000 per test, could offer a signification savings to transplant centers that currently perform around 12 biopsies for each patient in the first year after the transplant. Each biopsy costs between $4,000 and $5,000. On the other hand, considering that the NEJM study found that many of the transplant rejections were detectable by observation, refinement of an observation-based strategy would be far more cost-effective.
Currently, 65 transplant centers in the US offer gene expression profiling with AlloMap. According to the company's website, the test has been used in more than 21,000 cases to detect the likelihood of heart transplant rejection at the time of testing in more than 7,000 patients.