Veracyte said the accuracy of its investigational nasal swab genomic classifying test rivals the clinical standard for lung nodule assessment and has potential to spare unnecessary follow-up in suspected cancer cases, based on preliminary data presented on October 22 at the Chest annual meeting of the American College of Chest Physicians in New Orleans.
The South San Francisco, CA-based company is developing the nasal product as a laboratory-developed test for its own CLIA-certified facility and plans to introduce it commercially in early 2021, following additional analytical work and studies, according to Giulia Kennedy, PhD, chief scientific and medical officer. Development is ongoing as part of a broader partnership with Johnson & Johnson related to the early detection of lung cancer.
The study detailed in a poster presentation at the Chest meeting evaluated nasal epithelium swab samples collected prospectively as part of the Airway Epithelial Gene Expression in the Diagnosis of Lung Cancer (AEGIS) I and AEGIS II studies, which enrolled current and former smokers with lung lesions seen on chest computed tomography (CT) scans. The nasal test is based on the field-of-injury technology that Veracyte used to create its Percepta Genomic Sequencing Classifier (GSC). Field of injury refers to gene expression changes associated with malignancies.
The researchers shared experience with the product following development in a training set of 411 nasal samples and testing in an independent set of 261 samples. The samples were evaluated using whole-transcriptome RNA sequencing technology and informed by machine-learning algorithms. Results were compared to a clinical model for evaluating pulmonary nodules established by Kaiser Permanente pulmonologist and health services researcher Dr. Michael Gould and colleagues in practice guidelines (Chest, May 2013, Vol. 143:5, pp. e93S-e120S).
For classifying low-risk lesions at a fixed sensitivity of 96.6% in a population with 25% cancer prevalence, the nasal test outperformed the Gould model by almost 20 absolute percentage points -- that is, 45.6% versus 26.3%, according to Veracyte. For high-risk lesions with specificity fixed at 94.7%, the nasal test demonstrated sensitivity of 50%, compared with 42.2% for the Gould model.
|Nasal swab test results at 25% cancer prevalence|
|Low risk||High risk|
|Sensitivity||Specificity||NPV (CI)||Sensitivity||Specificity||PPV (CI)|
|Nasal classifier||96.6%||45.6%||97.6% (94.8%-98.9%)||50%||94.7%||76% (51.1%-90.6%)|
|Gould model||96.6%||26.3%||95.8% (90.8%-98.2%)||42.2%||94.7%||72.8% (46.7%-89%)|
Presented by Lamb et al, Chest 2019.
In search of a less-invasive test
The field-of-injury technology in the noninvasive nasal swab test was validated in the marketed Percepta GSC test, which measures genomic changes in cells caused by smoking, using samples taken from the main airway of the lung during bronchoscopy procedures.
The company noted that every year in the U.S., approximately 2 million lung nodules are detected on CT scans, but tools for assessing whether they are malignant are lacking, which results in unnecessary follow-up procedures. Methods for guiding imaging using blood tests to focus follow-up on those most in need had a high profile at the World Lung Conference on Lung Cancer in Barcelona, Spain, this fall.
"The preliminary data being presented at Chest on October 22 suggest that Veracyte's noninvasive nasal swab classifier can accurately classify lung cancer risk in patients with lung nodules -- less invasively and more easily than lung biopsies and even liquid biopsies," Kennedy said in an emailed response to questions from LabPulse.com.
The data show that performance is consistent, regardless of lung nodule size or location, as well as cancer subtype or stage, and the company is exploring opportunities to deploy the technology across the lung cancer care continuum, she said.
In addition to the poster presentation on the new nasal swab test, the company also reported supportive data for its marketed tests of bronchoscopy samples at the Chest meeting -- Percepta GSC for lung cancer detection and the Envisa Genomic Classifier, which is used for distinguishing idiopathic pulmonary fibrosis from other interstitial lung diseases.