Gene Expression Profiling as a Noninvasive Method to Monitor for Cardiac Allograft Rejection
Heart transplant patients are regularly monitored for allograft rejection. In addition to clinical surveillance, many patients undergo periodic endomyocardial biopsy, an invasive procedure, to evaluate whether there is asymptomatic rejection. The gene expression profiling test, AlloMap®, was developed as a noninvasive method to assist in the monitoring of heart transplant patients.
The objective of this Assessment is to determine whether AlloMap® testing improves health outcomes when used as an alternative to other methods of monitoring heart transplant patients for rejection.
A search of the MEDLINE® database (via PubMed) was completed for the period through September 2011. The search strategy used the terms “gene expression,” “heart transplant,” and “AlloMap” as textwords or subject terms. Articles were limited to those published in English language and enrolling human subjects. The MEDLINE® search was supplemented by an examination of article bibliographies and relevant review articles, which were searched for citations.
Studies that analyze the diagnostic test performance of AlloMap® testing in relevant patient populations were selected. A clinical trial in which a strategy of monitoring using AlloMap® testing was compared to a strategy of periodic endomyocardial biopsy was selected.
Question 1: Diagnostic Characteristics of AlloMap® Testing
Main Results. Studies evaluating the diagnostic characteristics of AlloMap® testing used various study populations and cutoff scores to denote a positive test. In validation studies, which only included patients with no rejection or class 3A rejection, the test was 76–84% sensitive and 38–41% specific at a cutoff score of 20. Post hoc analyses of subgroups of patients more distant from the time of transplant (>6 months and >12 months) and using higher cutoff scores showed sensitivities of 71.4–80% and specificities of 77.8–78.7%, but these numbers are based on very small sample sizes of fewer than 30 patients.
In other studies using study samples more representative of the spectrum of rejection grades present in a typical clinical population, publications reported positive predictive values that were generally below 7% and negative predictive values above 98%, depending on the particular cutoff score used to denote a positive test. However, the methods used in these studies are not clear. The data used to generate the values are not available, and the results are not consistent with results from the actual patient sample. One study in which the diagnostic characteristics are generated from an actual patient sample showed a positive predictive value of 7.8% and a negative predictive value of 100% using a cutoff score of 34; however, this is based on a sample of 5 cases of rejection in a sample of 243. The studies all lack sufficient description to determine whether there are biases such as verification bias. The emphasis in the studies on the negative predictive values is misleading because the values are uniformly high in study samples where the prevalence of rejection is low. The sensitivity of AlloMap® testing is not reported in several of the studies.
Author’s Comments and Conclusions. Although a higher score is associated with a greater likelihood of rejection class 3A or higher, the diagnostic characteristics of AlloMap® testing are uncertain. Study methods are unclear, study samples are incompletely described, numbers of cases of rejection are apparently small, and cutoff scores appear to have been determined post hoc. The sensitivity of the test for detecting rejection is uncertain.
Question 2: Patient Outcomes Using AlloMap® Testing
Main Results. A randomized, clinical trial compared patient outcomes in patients who were randomized to either a strategy of AlloMap® testing or periodic endomyocardial biopsy. In the AlloMap® group, only patients with high AlloMap® scores or clinical indications would undergo biopsy. It was hypothesized that the AlloMap® test would be successful in detecting rejection, allowing deferral of biopsy, and resulting in equivalent patient outcomes, thus improving health by avoidance of an invasive procedure and accompanying risk of complications.
The 2-year composite outcome of rejection with hemodynamic compromise, graft dysfunction, or death was similar in the two groups (AlloMap® 14.5%, biopsy 15.3%). There were fewer biopsies in the AlloMap® group. However, in examining the rates of treated asymptomatic rejection, there were 6 in the AlloMap® group detected by AlloMap® testing, and 22 detected by biopsy. This is best explained by low sensitivity of the test. As patients had equivalent outcomes in this trial, it is possible that the outcome of the study is not sensitive to the difference in treated rejection episodes, or that treatment of asymptomatic rejection is not necessary in these patients.
Author’s Comments and Conclusions. The appearance of equivalent outcomes in the two groups of the clinical trial does not provide strong evidence that AlloMap® testing is an effective strategy of rejection surveillance. The results raise questions about whether periodic biopsy is necessary in asymptomatic low-risk patients. An authoritative practice guideline is silent regarding whether biopsy is recommended in such patients after 1 year post-transplant, and states that the issue is the subject of debate. However, if one believes that asymptomatic rejection should be treated, then the clinical trial demonstrates that it is likely that AlloMap® testing misses a significant number of rejection episodes. Given the uncertain nature of the prior evidence regarding the diagnostic characteristics of AlloMap® testing, the potentially low sensitivity of AlloMap® testing in the clinical trial should not be regarded as an anomaly or chance occurrence.
Based on the available evidence, the Blue Cross and Blue Shield Association Medical Advisory Panel made the following judgments about whether AlloMap® gene expression testing for post-transplant monitoring of cardiac allograft rejection meets the Blue Cross and Blue Shield Association Technology Evaluation Center (TEC) criteria.
1. The technology must have final approval from the appropriate governmental regulatory bodies.
AlloMap® testing received 510(k) marketing clearance in 2008 for the purpose of aiding in the identification of heart transplant recipients with stable allograft function who have a low probability of moderate/severe acute cellular rejection at the time of testing in conjunction with standard clinical assessment.
2. The scientific evidence must permit conclusions concerning the effect of the technology on health outcomes.
The evidence is insufficient to permit conclusions concerning the effect of the technology on health outcomes. Existing studies on the diagnostic characteristics of AlloMap® testing to detect class 3A/2R grade of rejection in the clinical population of interest are small and the cutpoints for determination of a positive test have not been validated. The methodology of the studies is not well described; in many studies it cannot be determined how the test performance was calculated.
The clinical trial of AlloMap® testing showed no difference in the primary outcome, fewer biopsies, and fewer episodes of treated rejection in the group monitored with AlloMap® testing. The results are consistent with low sensitivity of AlloMap® testing. The study either had an insensitive outcome measure or it is not necessary to detect and treat asymptomatic rejection. With either explanation, the study does not permit conclusions regarding the effect of the technology on health outcomes.
3. The technology must improve the net health outcome; and
4. The technology must be as beneficial as any established alternatives.
The evidence is insufficient to make conclusions regarding whether AlloMap® testing either improves the net health outcome or is as beneficial as any established alternatives for the monitoring of heart transplant patients.
5. The improvement must be attainable outside the investigational settings.
It has not yet been demonstrated whether AlloMap® testing improves health outcomes in the investigational setting. Therefore, it cannot be demonstrated whether improvement is attainable outside the investigational settings.
For the above reasons, the use of gene expression profiling as a noninvasive method to monitor for cardiac allograft rejection does not meet the TEC criteria.
Full studies are in PDF format. You will need Adobe Acrobat Reader to view all studies.
Download Adobe Acrobat Reader here.
allograft; EMB; Grade 3A; HARDBALL; heart transplant; heterotopic; IMAGE; International Society of Heart and Lung Transplantation Guidelines; Invasive Monitoring Attenuation Through Gene Expression; ISHLT; endomyocardial biopsy; molecular expression testing; morbidity; nodular endocardial infiltrates; Quilty lesions; registry; rejection; surveillance; XDx