IDENTIFICATION OF LIKELY MALIGNANT INDETERMINATE PULMONARY NODULES BY ANALYSIS OF AUTOANTIBODIES AGAINST LUNG CANCER-ASSOCIATED ANTIGENS

Abstract

SESSION TITLE: Lung Cancer Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: October 18-21, 2020 PURPOSE: Cancer risk prediction based on clinical guidelines alone of indeterminate pulmonary nodules (IPNs) is imprecise, potentially leading to unnecessary procedures on benign IPNs and delays in the diagnosis of malignant IPNs. ˜75% IPNs fall into the moderate risk category (probability of cancer (pCA) 5-65%) by the ACCP guidelines, where clinical follow up is less clear. A blood test evaluating levels of autoantibodies to lung cancer associated antigens has been shown to identify “likely malignant” IPNs with high specificity may accelerate diagnosis. Here we report the retrospective analysis of the PANOPTIC clinical trial to further validate the clinical utility of the test in patients with incidentally identified lung nodules. METHODS: Plasma samples collected in PANOPTIC were retrospectively evaluated with the Nodify CDT test (CDT) and correlated with clinical outcomes (average of 2 year follow up). A total of 317 patients were eligible for analysis based on sample availability, of which 263 patients met the clinical use (CU) criteria for the test (8-30mm diameter, age ≥40 years and no previous history of cancer). Lung cancer prevalence was 54% in the CU population. RESULTS: The number of CDT positive (CDT-P) results was 34 (13%) with 25/34 (74%) being malignant resulting in the following overall test performance: specificity 93% (95% confidence interval (CI) 86-97%); sensitivity 18% (CI 12-25%); positive predictive value (PPV) 74% (CI 57-85%). Performance of the High-Level (HL) cutoff indicated: specificity 94% (88-98%); sensitivity 11% (CI 7-18%) and PPV 70% (49-84%). Comparison to previously reported test performance in Healey et al. (2017) was similar: specificity 93% (HL: 98%), sensitivity 41% (HL: 28%) and PPV 59% (HL: 78%). Following testing, the risk of malignancy was adjusted using the positive likelihood ratios for CDT. Of the patients with CDT-P results, 18 / 24 (75%) patients were reclassified from the low-moderate risk category to the high-risk category (>65% risk) per ACCP guidelines. In patients reclassified as high risk by the test, 12/18 (66%) were confirmed to be malignant. Comparison of the care utilization patterns between CDT-P patients reclassified to high risk and standard of care (SOC) indicated 3 patients with cancer, representing 12% of the CDT-P patients with cancer, were routed to CT surveillance by SOC. CONCLUSIONS: The results of this retrospective analysis confirm the previously reported (Healey et al 2017) performance of the Nodify CDT test including high specificity and PPV necessary for accurately identifying patients with likely malignant lung nodules that will benefit from a more rapid evaluation and extended follow up if initially benign. CLINICAL IMPLICATIONS: The Nodify CDT blood test can identify likely malignant lung nodules that may benefit from a more rapid evaluation. DISCLOSURES: Consultant relationship with Biodesix Please note: $1001 - $5000 Added 06/02/2020 by Kevin Doubleday, source=Web Response, value=Consulting fee Consultant relationship with Biodesix Please note: $20001 - $100000 Added 05/31/2020 by James Jett, source=Web Response, value=Consulting fee Employee relationship with Biodesix, Inc. Please note: >$100000 Added 06/02/2020 by Laura Peek, source=Web Response, value=Salary Employee relationship with Biodesix Please note: >$100000 Added 06/01/2020 by Trevor Pitcher, source=Web Response, value=Salary Removed 06/01/2020 by Trevor Pitcher, source=Web Response Employee relationship with Biodesix Please note: >$100000 Added 06/01/2020 by Trevor Pitcher, source=Web Response, value=Salary