A simple nasal swab may help in the diagnosis of lung cancer in smokers who have undergone CT screening and had lung nodules detected on the scan.
Only about 5% of the nearly 1.6 million lung nodules identified as incidental findings on low-dose CT screening tests will turn out to be malignant. The new test helps to distinguish between benign and malignant nodules, say researchers reporting a validation study.
The results show that the test identified those at low risk for cancer with a sensitivity of 96.3% and specificity of 41.7%, as well as identifying those as high risk, with a specificity of 90.4% and sensitivity of 58.2%.
The Percepta nasal swab is a first-of-its-kind genomic test, says the manufacturer Veracyte.
It is based on “field of injury” technology, which examines genomic changes in the lining of the respiratory tract for evidence of active cancer cells, coupled with a machine learning model that includes factors such as age, gender, and smoking history.
Veracyte hopes to begin to make the test available to a select number of sites in the second half of 2021. “The test is intended to be performed in the physician’s office on patients referred with suspicious lung nodules found on CT scans,” said Giulia C. Kennedy, PhD, chief scientific officer and chief medical officer at Veracyte. “This could include patients with nodules found through screening programs, as well as incidentally.”
“It will be made available as a laboratory developed test in the U.S. through Veracyte’s centralized CLIA laboratory,” she said in an interview. “In global markets, we will offer the test as an IVD product that can be performed on the nCounter instrument by laboratories locally. Outside of the United States, the test will require a CE mark, which we are equipped to support.”
Results with the test were presented during the American Society of Clinical Oncology (ASCO) 2021 Annual Meeting, which was held virtually this year.
It was first tested in a training set, which consisted of more than 1,100 patients. All were current or former smokers who had a lung nodule detected on chest CT scanning and were followed for up to 1 year or until a final diagnosis of lung cancer or benign disease.
Brushings of the nasal epithelium were prospectively collected in patients with lung nodules from multiple cohorts.
A total of 502 genes were used in the classifier, and performance was evaluated in an independent clinical validation set consisting of 249 patients.
The test identified true benign patients as low risk with 41.7% specificity and 96.3% sensitivity, resulting in a negative predictive value (NPV) of 97.1% in a population with a cancer prevalence of 25%. The risk of malignancy for patients in this low-risk group was less than 3% (1-NPV), and for this group, clinical guidelines recommend surveillance.
Patients with true malignancies were identified as high risk, with 58.2% sensitivity and 90.4% specificity, resulting in a positive predictive value of 67.0% in a population with 25% cancer prevalence. The risk of malignancy for patients deemed to be high risk by the classifier was 67.0%, which exceeds the current guideline threshold for consideration of surgical resection or other ablative therapy if a staging evaluation confirms early stage disease, the authors point out.
The remaining patients, who did not meet the stringent cut-offs for low or high risk, were identified as intermediate risk. In this population, the prevalence of malignancy for patients identified as intermediate risk was 20.7%, which is consistent with guidelines that provide a range for intermediate-risk patients as between 5% and 65% for whom diagnostic biopsy is recommended.