Clinical validation of the tempus xT next-generation targeted oncology sequencing assay Clinical validation of the tempus xT next-generation targeted oncology sequencing assay
1 January 2019 in Journal of Pathology Informatics
Background: Tumor programmed death-ligand 1 (PD-L1) status is useful in determining which patients may benefit from programmed death-1 (PD-1)/PD-L1 inhibitors. However, little is known about the association between PD-L1 status and tumor histopathological patterns. Using deep learning, we predicted PD-L1 status from hematoxylin and eosin (H and E) whole-slide images (WSIs) of nonsmall cell lung cancer (NSCLC) tumor samples. Materials and Methods: One hundred and thirty NSCLC patients were randomly assigned to training (n = 48) or test (n = 82) cohorts. A pair of H and E and PD-L1-immunostained WSIs was obtained for each patient. A pathologist annotated PD-L1 positive and negative tumor regions on the training samples using immunostained WSIs for reference. From the H and E WSIs, over 145,000 training tiles were generated and used to train a multi-field-of-view deep learning model with a residual neural network backbone. Results: The trained model accurately predicted tumor PD-L1 status on the held-out test cohort of H and E WSIs, which was balanced for PD-L1 status (area under the receiver operating characteristic curve [AUC] =0.80, P << 0.01). The model remained effective over a range of PD-L1 cutoff thresholds (AUC = 0.67–0.81, P ≤ 0.01) and when different proportions of the labels were randomly shuffled to simulate interpathologist disagreement (AUC = 0.63–0.77, P ≤ 0.03). Conclusions: A robust deep learning model was developed to predict tumor PD-L1 status from H and E WSIs in NSCLC. These results suggest that PD-L1 expression is correlated with the morphological features of the tumor microenvironment.
The recent findings of the National Lung Screening Trial showed 24.2% of individuals at high risk for lung cancer having one or more indeterminate nodules detected by low-dose computed tomography-based screening, 96.4% of which were eventually confirmed as false positives. These positive scans necessitate additional diagnostic procedures to establish a definitive diagnosis that adds cost and risk to the paradigm. A plasma test able to assign benign versus malignant pathology in high-risk patients would be an invaluable tool to complement low-dose computed tomography-based screening and promote its rapid implementation. We evaluated 17 biomarkers, previously shown to have value in detecting lung cancer, against a discovery cohort, comprising benign (n = 67) cases and lung cancer (n = 69) cases. A Random Forest method based analysis was used to identify the optimal biomarker panel for assigning disease status, which was then validated against a cohort from the Mayo Clinic, comprising patients with benign (n = 61) or malignant (n = 20) indeterminate lung nodules. Our discovery efforts produced a seven-analyte plasma biomarker panel consisting of interleukin 6 (IL-6), IL-10, IL-1ra, sIL-2Rα, stromal cell-derived factor-1α+β, tumor necrosis factor α, and macrophage inflammatory protein 1 α. The sensitivity and specificity of our panel in our validation cohort is 95.0% and 23.3%, respectively. The validated negative predictive value of our panel was 93.8%. We developed a seven-analyte plasma biomarker panel able to identify benign nodules, otherwise deemed indeterminate, with a high degree of accuracy. This panel may have clinical utility in risk-stratifying screen-detected lung nodules, decrease unnecessary follow-up imaging or invasive procedures, and potentially avoid unnecessary morbidity, mortality, and health care costs.