Abstract 4218: Validation of a new blood-based biomarker strategy for the early detection of lung cancer

Abstract

Abstract Rationale: The management of indeterminate pulmonary nodules (IPNs) remains a challenging problem. Our new assay methodology, the Free Solution Assay (FSA), measured by the Compensated Interferometric Reader (CIR), consisting of a diode laser, capillary and CCD, provides 40-fold lower limits of quantitation (LOQ) than ELISA for known candidate serum protein biomarkers, while speeding assay development, accuracy, and sensitivity. We hypothesized that lowering the LOQ of previously investigated biomarkers can increase the biomarker discriminatory power, enabling patients in an intermediate risk group (15-80%) to be reclassified into a low (<15) or high (>80) risk group. Methods: In this retrospective case-control study, FSA-CIR was used to measure the serum concentration of CYFRA 21-1 in two patient cohorts, a training cohort (N=274) of patients with IPNs collected at Vanderbilt University Medical Center and an external validation cohort (N=103) collected at the University of Pittsburgh Medical Center. Patient malignancy was determined by tissue biopsy or 2-year follow-up CT scan showing no signs of nodule growth. Baseline risk for cancer was calculated using nodule size. The added value of CYFRA 21-1 was assessed by comparing the difference in risk for lung cancer after incorporating CYFRA 21-1 into the model. Results: The limit of detection (LOD) of 6pg/mL and an average LOQ for standards was determined to be 60pg/mL. Patient samples in the training cohort were found to have CYFRA 21-1 concentrations ranging 100 pg/mL to 10 ng/mL, with a median of 0.79 (0.28-1.22, interquartile range) ng/mL in the control population and 1.90 (1.33-3.35) ng/mL in the case population, providing a ROC-AUC of 0.86 across three histological subtypes (adenocarcinoma, squamous cell carcinoma, and small cell lung cancer). The CYFRA 21-1 + nodule size risk model correctly reclassified 28 (25%) [DB1] of intermediate-risk benign nodules into the low-risk group and 28(24%) of intermediate-risk malignant nodules into the high-risk group. The independent validation cohort’s controls had a median of 0.35 (0.20-0.56) ng/mL, while cases had 0.97(0.66-1.22) ng/mL, providing a ROC-AUC of 0.84 and correct reclassification of 12(34%) of intermediate-risk benign nodules and 6(15%) of intermediate-risk malignant nodules. The CYFRA 21-1 + nodule size risk model correctly classified 90.6% in the low-risk group and 87.0% in the high-risk group. Conclusions: FSA-CIR measurements requiring only a few microliters of serum allowed for reclassification of patients in the intermediate risk group in both the training and validation cohorts. The results suggest that CYFRA 21-1 measured by FSA-CIR represents a strong candidate biomarker for risk stratification of patients with IPNs. Citation Format: Michael N. Kammer, Amanda K. Kussrow, Sanja L. Antic, Rina Nguyen, Heidi Chen, Darryl J. Bornhop, Pierre P. Massion. Validation of a new blood-based biomarker strategy for the early detection of lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4218.