NEW YORK (GenomeWeb) – A new study by Veracyte and collaborators demonstrates that the company’s Envisia genomic test can improve the diagnosis of patients being evaluated for interstitial lung disease.
The study, published this week in Lancet Respiratory Medicine, includes both clinical validation and clinical utility data from a subset of patients who are part of an ongoing prospective 29-site study known as BRAVE (Bronchial Sample Collection for a Novel Genomic Test).
Veracyte funded the study and had a role in the design, data collection, analysis, and interpretation.Â
For the study, the company’s clinical collaborators recruited 237 individuals from the BRAVE trial who were being evaluated for interstitial lung disease. They used RNA sequencing data from transbronchial lung biopsy samples from 90 patients to train a machine learning algorithm to identify a pattern for usual interstitial pneumonia (UIP), which is required for diagnosing idiopathic pulmonary fibrosis, a deadly but common subtype of interstitial lung disease that is difficult to diagnose.
After training the algorithm, the team tested it on a different set of 49 patients, achieving 88 percent specificity and 70 percent sensitivity for identifying UIP. The positive predictive value was 84 percent, while the negative predictive value was 77 percent.
"IPF is often challenging to distinguish from other [interstitial lung disease], but timely and accurate diagnosis is critical so that patients with IPF can access therapies that may slow progression of the disease, while avoiding potentially harmful treatments," Ganesh Raghu, director of the Center for Interstitial Lung Diseases at the University of Washington and lead author of the publication, said in a statement. The study shows that physicians may be able to use Veracyte’s Envisia assay along with clinical and radiological information "as a diagnostic tool to make a more informed and confident diagnosis," he added.
Typically, idiopathic pulmonary fibrosis is diagnosed by identifying UIP on a high-resolution chest CT. However, according to retrospective analyses, only 28 percent to 58 percent of patients receive a confident diagnosis of IPF through a CT scan. A correct diagnosis is important because patients diagnosed with interstitial lung disease that is not IPF are often treated with immunosuppressants, which can be harmful if that patient does in fact have IPF. In cases where a CT scan is inconclusive, diagnosis is often achieved by a surgical lung biopsy, which can cause complications. Â The goal of Veracyte's Envisia is to help diagnose IPF in a less invasive manner, through a transbronchial lung biopsy, which does not require surgery.
Envisia is an RNA sequencing assay that uses expression patterns of 190 genes to differentiate between IPF and other lung diseases.
In 2017, Veracyte and collaborators published initial data on Envisia based on 84 patients from the BRAVE study, but since then, they have fine-tuned the algorithm used to classify samples. Â
In addition, aside from assessing the performance of Envisia, the Lancet Respiratory Medicine study also analyzed the clinical utility of the assay. To do this, the research team put together two multidisciplinary teams, each of which had a pathologist, radiologist, and pulmonologist. Medical files from study participants with and without a diagnostic pathology or with an unclassifiable fibrosis were randomly assigned to the two groups. The files included all the clinical information, as well as either the Envisia result or the diagnostic histopathology result, and were assigned so that each patient had both types of results independently reviewed.
Overall, for the 94 patients analyzed, the two teams agreed 86 percent of the time. For 46 cases in which histopathology was diagnostic, the teams reported being confident in their diagnosis for 29 cases when just using histopathology to diagnose and 22 cases when using Envisia, a statistically insignificant difference. In 48 patients for whom histopathology did not yield a diagnosis or a classifiable result, the teams reported making confident diagnoses in 30 cases using Envisia and 20 using histopathology results. "We found that confidence of an IPF diagnosis was substantially improved when classifier results were available than when histopathology results were available," the authors reported.
The researchers concluded that the study supports the use of Envisia "to aid clinicians and multidisciplinary teams in ascertaining the diagnosis of IPF, particularly for patients without a clear radiological diagnosis."
The publication follows a local coverage decision by Medicare administrative contractor Palmetto GBA MolDx last month, which said the test will be covered for patients for whom a CT scan was indeterminate or resulted in a probable UIP pattern, but not for patients without IPF or with a typical pattern of UIP.
"This new paper, combined with the Medicare coverage policy for the Envisia classifier issued recently, will fuel our efforts to make the classifier more widely available to the patients across the country who can benefit from it," Veracyte CEO Bonnie Anderson said in a statement.