Detection of early-stage lung cancer in sputum using automated flow cytometry and machine learning.

BACKGROUND: Low-dose spiral computed tomography (LDCT) may not lead to a clear treatment path when small to intermediate-sized lung nodules are identified. We have combined flow cytometry and machine learning to develop a sputum-based test (CyPath Lung) that can assist physicians in decision-making in such cases. METHODS: Single cell suspensions prepared from induced sputum samples collected over three consecutive days were labeled with a viability dye to exclude dead cells, antibodies to distinguish cell types, and a porphyrin to label cancer-associated cells. The labeled cell suspension was run on a flow cytometer and the data collected. An analysis pipeline combining automated flow cytometry data processing with machine learning was developed to distinguish cancer from non-cancer samples from 150 patients at high risk of whom 28 had lung cancer. Flow data and patient features were evaluated to identify predictors of lung cancer. Random training and test sets were chosen to evaluate predictive variables iteratively until a robust model was identified. The final model was tested on a second, independent group of 32 samples, including six samples from patients diagnosed with lung cancer. RESULTS: Automated analysis combined with machine learning resulted in a predictive model that achieved an area under the ROC curve (AUC) of 0.89 (95% CI 0.83-0.89). The sensitivity and specificity were 82% and 88%, respectively, and the negative and positive predictive values 96% and 61%, respectively. Importantly, the test was 92% sensitive and 87% specific in cases when nodules were < 20 mm (AUC of 0.94; 95% CI 0.89-0.99). Testing of the model on an independent second set of samples showed an AUC of 0.85 (95% CI 0.71-0.98) with an 83% sensitivity, 77% specificity, 95% negative predictive value and 45% positive predictive value. The model is robust to differences in sample processing and disease state. CONCLUSION: CyPath Lung correctly classifies samples as cancer or non-cancer with high accuracy, including from participants at different disease stages and with nodules < 20 mm in diameter. This test is intended for use after lung cancer screening to improve early-stage lung cancer diagnosis. Trial registration ClinicalTrials.gov ID: NCT03457415; March 7, 2018.

Sputum analysis by flow cytometry; an effective platform to analyze the lung environment.

Low dose computed tomography (LDCT) is the standard of care for lung cancer screening in the United States (US). LDCT has a sensitivity of 93.8% but its specificity of 73.4% leads to potentially harmful follow-up procedures in patients without lung cancer. Thus, there is a need for additional assays with high accuracy that can be used as an adjunct to LDCT to diagnose lung cancer. Sputum is a biological fluid that can be obtained non-invasively and can be dissociated to release its cellular contents, providing a snapshot of the lung environment. We obtained sputum from current and former smokers with a 30+ pack-year smoking history and who were either confirmed to have lung cancer or at high risk of developing the disease. Dissociated sputum cells were counted, viability determined, and labeled with a panel of markers to separate leukocytes from non-leukocytes. After excluding debris and dead cells, including squamous epithelial cells, we identified reproducible population signatures and confirmed the samples' lung origin. In addition to leukocyte and epithelial-specific fluorescent antibodies, we used the highly fluorescent meso-tetra(4-carboxyphenyl) porphyrin (TCPP), known to preferentially stain cancer (associated) cells. We looked for differences in cell characteristics, population size and fluorescence intensity that could be useful in distinguishing cancer samples from high-risk samples. We present our data demonstrating the feasibility of a flow cytometry platform to analyze sputum in a high-throughput and standardized matter for the diagnosis of lung cancer.

Early Detection of Lung Cancer with Meso Tetra (4-Carboxyphenyl) Porphyrin-Labeled Sputum.

INTRODUCTION: Early detection of lung cancer in high-risk individuals reduces mortality. Low-dose spiral computed tomography (LDCT) is the current standard but suffers from an exceedingly high false-positive rate (>96%) leading to unnecessary and potentially dangerous procedures. We, therefore, set out to develop a simple, noninvasive, and quantitative assay to detect lung cancer. METHODS: This proof-of-concept study evaluated the sensitivity/specificity of the CyPath Early Lung Cancer Detection Assay to correctly classify LDCT-confirmed cohorts of high-risk control (n = 102) and cancer (n = 26) subjects. Fluorescence intensity parameters of red fluorescent cells (RFCs) from tetra (4-carboxyphenyl) porphyrin (TCPP)-labeled lung sputum samples and subjects' baseline characteristics were assessed for their predictive power by multivariable logistic regression. A receiver operating characteristic curve was constructed to evaluate the sensitivity/specificity of the CyPath assay. RESULTS: RFCs were detectable in cancer subjects more often than in high-risk ones (p = 0.015), and their characteristics differed between cohorts. Two independent predictors of cancer were the mean of RFC average fluorescence intensity/area per subject (p < 0.001) and years smoked (p = 0.003). The CyPath-based classifier had an overall accuracy of 81% in the test population; false-positive rate of 40% and negative predictive value of 83%. CONCLUSIONS: The tetra (4-carboxyphenyl) porphyrin -based CyPath assay correctly classified study participants into cancer or high-risk cohorts with considerable accuracy. Optimizing sputum collection, sample reading, and refining the classifier should improve sensitivity and specificity. The CyPath assay thus has the potential to complement LDCT screening or serve as a stand-alone approach for early lung cancer detection.