Evaluation of a novel blood-based lung cancer screening test in elevated-risk subjects.

10535 Background: Less than 20% of individuals recommended for lung cancer (LC) screening are screened each year by low dose computed tomography (LDCT), highlighting the need for a more acceptable screening modality. Tumor-associated extracellular vesicles and particles (EVPs) are highly abundant in blood and carry surface biomarkers from their tumor cell of origin, providing a readily measurable analyte class for blood-based screening tests. Test sensitivity may be improved by EVP abundance; specificity by simultaneous measurement of multiple co-localized cancer-related markers on a vesicle surface. We report performance of an EVP-based LC Screening Test in plasma samples from elevated risk participants in the National Lung Screening Trial (NLST). Methods: The LC Test (biomarkers and classifier) was locked following a training study, after which we validated consistent assay performance in an independent cohort. Here we evaluated the LC Screening Test in a blinded case-control study nested within the ACRIN Biomarker Arm of the NLST. The study included all available incident LC cases for which subject-matched blood samples were available from the year of diagnosis (T) and the year prior (T-1) (n = 48 cases; 96 samples). Controls were age-matched NLST subjects who had no LC detected before or within the trial follow up period (n = 60). LC Test results were determined blinded to case-control status, with unblinding by ECOG-ACRIN. LC Test sensitivity was calculated at a threshold pertaining to 90% specificity, superior to 87% reported specificity for LDCT in NLST (Lung RADS v1.1). LDCT positivity was based on Lung RADS v1.1 interpretation. Results: Across two screening encounters (T and T-1), 34.4% sensitivity, similar across stages, was observed for the blood-based LC Screening Test, compared to 39.5% sensitivity for LDCT, p = 0.53 McNemar test, indicating no significant difference in observed sensitivity between LDCT and the blood test. Complementarity in detection was observed between LDCT vs blood-test detected cases across all histologies. A mean 42% increase was observed in LC Screening Test scores from T-1 to T, indicating signal increase may provide additional sensitivity. Conclusions: The blood-based LC Screening Test exhibits similar performance to LDCT in detecting preclinical lung cancer in high-risk individuals and detects some LCs not found by LDCT. The data supports further evaluation of the LC Test for LC screening. Histology Dx Modality Total Detected Detected by One Test Detected by Both Detected by Neither Adenocarcinoma LDCT 37% (17/46) 24% (11/46) 13% (6/46) 44% (20/46) Adenocarcinoma Blood Test 33% (15/46) 20% (9/46) 13% (6/46) 44% (20/46) Squamous LDCT 42% (10/24) 29% (7/24) 13% (3/24) 50% (12/24) Squamous Blood Test 21% (5/24) 8% (2/24) 13% (3/24) 50% (12/24) Small Cell LDCT 43% (6/14) 14% (2/14) 29% (4/14) 36% (5/14) Small Cell Blood Test 50% (7/14) 21% (3/14) 29% (4/14) 36% (5/14) Other LDCT 42% (5/12) 50% (6/12) 17% (2/12) 25% (3/12) Other Blood Test 25% (3/12) 33% (4/12) 17% (2/12) 25% (3/12)