New tissue test for idiopathic pulmonary fibrosis changes the diagnostic game
Medicare announced it will now pay for a new molecular tissue test for idiopathic pulmonary fibrosis, called the Envisia Genomic Classifier (Veracyte). The availability of the new assay reshuffles the algorithm for the diagnosis of IPF, forcing consideration of transbronchial biopsy in some patients. However, professional societies do not yet endorse its general use.
The Envisia genetic test uses RNA sequencing of 190 genes in tissue obtained by transbronchial biopsies or cryobiopsies during bronchoscopy. The data is fed to a machine learning algorithm which provides a probabilistic estimate of the likelihood of IPF. In its development, the algorithm was ‘trained’ on patients with a definite diagnosis of IPF.
Envisia is intended to be used as a complement to high-resolution chest CT (HRCT) to differentiate IPF from other interstitial lung diseases. When IPF has a classic appearance, HRCT is considered diagnostic. However, a substantial proportion of cases eventually diagnosed as IPF are not initially obvious, and may be misdiagnosed or the diagnosis delayed. In these cases, tissue biopsy and genomic testing can increase or decrease the likelihood of IPF.
In the BRAVE validation study testing the Envisia algorithm on undiagnosed tissue samples, its sensitivity was 70% and its specificity 88% for diagnosing IPF, with histopathology as the gold standard comparator.
Challenging interstitial lung disease diagnoses are sometimes made by multidisciplinary teams (pulmonologists, radiologists, and pathologists), who are encouraged to talk each other into a consensus diagnosis. In the CATALYST study, when such teams were provided the Envisia classifier data, they achieved 86% agreement in IPF vs. non-IPF diagnoses, when also using HRCT and histopathology results. Teams using only histopathology were only 56% confident in their conclusions.
Professional society guidelines do not yet endorse routine tissue biopsies in the diagnosis of IPF. A 2018 interim guideline update advised:
Blood testing (e.g. ANA) to identify possible connective tissue disease as a cause or contributor to ILD.
Biopsy is not advised for patients with a usual interstitial pneumonitis pattern (UIP) on high resolution chest CT.
For patients with indeterminate HRCT findings, bronchoalveolar lavage or surgical lung biopsy were advised for additional diagnostic testing.
For transbronchial lung biopsies or lung cryobiopsy, no recommendation was made (for or against) due to lack of evidence.
The authors specifically addressed the use of the new genomic testing, saying,
Machine learning using molecular signatures is being developed to make a molecular diagnosis of UIP in transbronchial biopsy specimens but is not yet available in routine clinical practice. The recent studies about the utility of molecular diagnostic tools that involve machine learning using TBBx samples are promising; further studies to validate this are pending. This recommendation will be revisited in future iterations of this guideline as related evidence accumulates.”