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.

Quality-Controlled Sputum Analysis by Flow Cytometry.

Sputum, widely used to study the cellular content and other microenvironmental features to understand the health of the lung, is traditionally analyzed using cytology-based methodologies. Its utility is limited because reading the slides is time-consuming and requires highly specialized personnel. Moreover, extensive debris and the presence of too many squamous epithelial cells (SECs), or cheek cells, often renders a sample inadequate for diagnosis. In contrast, flow cytometry allows for high-throughput phenotyping of cellular populations while simultaneously excluding debris and SECs. The protocol presented here describes an efficient method to dissociate sputum into a single cell suspension, antibody stain and fix cellular populations, and acquire samples on a flow cytometric platform. A gating strategy that describes the exclusion of debris, dead cells (including SECs) and cell doublets is presented here. Further, this work also explains how to analyze viable, single sputum cells based on a cluster of differentiation (CD)45 positive and negative populations to characterize hematopoietic and epithelial lineage subsets. A quality control measure is also provided by identifying lung-specific macrophages as evidence that a sample is derived from the lung and is not saliva. Finally, it has been demonstrated that this method can be applied to different cytometric platforms by providing sputum profiles from the same patient analyzed on three flow cytometers; Navios EX, LSR II, and Lyric. Furthermore, this protocol can be modified to include additional cellular markers of interest. A method to analyze an entire sputum sample on a flow cytometric platform is presented here that makes sputum amenable for developing high-throughput diagnostics of lung disease.

Molecular cytogenetics as a clinical test for prognostic and predictive biomarkers in newly diagnosed ovarian cancer.

BACKGROUND: There is a clinical need for routinely available genomic biomarker testing in newly diagnosed ovarian cancer. In the current study we performed molecular cytogenetics using a validated array based comparative genomic hybridization (array CGH) assay to screen for the presence of predictive and prognostic biomarkers in archival diagnostic tissue from ovarian cancer patients. We hypothesized that biomarkers of high-risk disease would be detectable in tumor samples from patients with treatment refractory, advanced disease, and would be detected less frequently in tumor samples from patients with more favorable outcomes. In addition, we predicted that the use of a genome-wide copy number analysis (CNA) testing platform would enable us to identify novel potentially targetable chromosomal alterations of therapeutic significance in a percentage of cases. METHODS: Formalin-fixed paraffin-embedded tissue (FFPE) tumor bank specimens were retrieved from the initial surgical resection for 18 ovarian cancer patients. Molecular cytogenetics was performed by array CGH for the detection of somatic chromosomal alterations associated with high-risk disease including amplifications of the CCNE1 and HER2 genes. Genomic risk stratification results were correlated with available clinical data. CGH data from each patient's tumor genome was also surveyed for the presence of potentially targetable aberrations. Relevant therapeutic agents and open studies for investigational drugs were reported for each patient. RESULTS: High-risk genomic alterations were identified in 12/18 (67%) of cases and all patients with high-risk markers had advanced, treatment refractory disease. Three tumors with minimal genomic changes had no high-risk markers and were from patients with Stage I/II disease that had been completely resected and under surveillance for recurrence. Eleven patients (61%) had at least one potentially targetable genomic alteration including CCNE1, HER2, KRAS gene amplifications, and somatic BRCA1 and/or BRCA2 gene deletions. Bi-allelic PTEN gene deletion was detected in one patient's tumor. CONCLUSIONS: Clinical genomic profiling of ovarian tumors by array CGH augments pathologic grade and stage to help stratify newly diagnosed ovarian cancer into high and low-risk disease. This personalized genomic information can also help guide treatment planning and disease monitoring by identifying novel potentially targetable genomic alterations that can be used by clinicians to choose rational directed therapies for patients with chemo-resistant disease.

Detection of recurrent copy number loss at Yp11.2 involving TSPY gene cluster in prostate cancer using array-based comparative genomic hybridization.

Prostate cancer is the second leading cause of cancer deaths among American men. The loss of Y chromosome has been frequently observed in primary prostate cancer as well as other types of cancer. Earlier, we showed that introduction of the human Y chromosome suppresses the in vivo tumorigenicity of the prostate cancer cell line PC-3. To further characterize the Y chromosome, we have developed a high-density bacterial artificial chromosome (BAC) microarray containing 178 BAC clones from the human Y chromosome. BAC microarray was used for array comparative genomic hybridization on prostate cancer samples and cell lines. The most prominent observation on prostate cancer specimens was a deletion at Yp11.2 containing the TSPY tandem gene array. Out of 36 primary prostate tumors analyzed, 16 (44.4%) samples exhibited loss of TSPY gene copies. Notably, we observed association between the number of TSPY copies in the blood and the incidence of prostate cancer. Moreover, PC-3 hybrids with an intact Yp11.2 did not grow tumors in nude mice, whereas PC-3 hybrids with a deletion at Yp11.2 grew tumors in nude mice.

Use of DNA sequencing analysis to confirm fungemia due to Trichosporon dermatis in a pediatric patient.

This is the first reported case of human disease caused by Tricosporon dermatis, an organism recently transferred to the genus Trichosporon from Cryptococcus and now confirmed to be a human pathogen.

Molecular characterization of a patient with central nervous system dysmyelination and cryptic unbalanced translocation between chromosomes 4q and 18q.

We report on a 12-year-old boy who presented with delayed development and CNS dysmyelination. Genetic studies showed a normal 46,XY karyotype by routine cytogenetic analysis, and 46,XY.ish del(18)(q23)(D18Z1+, MBP-) by FISH using a locus-specific probe for the MBP gene (18q23). Though the patient appeared to have normal chromosome 18s by repeated high resolution banding analysis, his clinical features were suggestive of a deletion of 18q. These included hearing loss secondary to stenosis of the external auditory canals, abnormal facial features, and foot deformities. FISH studies with genomic probes from 18q22.3 to 18qter confirmed a cryptic deletion which encompassed the MBP gene. In an attempt to further characterize the deletion, whole genome screening was conducted using array based comparative genomic hybridization (array CGH) analysis. The array CGH data not only confirmed a cryptic deletion in the 18q22.3 to 18qter region of approximately 7 Mb, it also showed a previously undetected 3.7 Mb gain of 4q material. FISH studies demonstrated that the gained 4q material was translocated distal to the 18qter deletion breakpoint. The 18q deletion contains, in addition to MBP, other known genes including CYB5, ZNF236, GALR1, and NFATC1, while the gained 4q material includes the genes FACL1 and 2, KLKB1, F11 and MTNR1A. The use of these combined methodologies has resulted in the first reported case in which array CGH has been used to characterize a congenital chromosomal abnormality, highlighting the need for innovative molecular cytogenetic techniques in the diagnosis of patients with idiopathic neurological abnormalities.

Molecular characterization of 18p deletions: evidence for a breakpoint cluster.

PURPOSE: To determine the size and parental origin of the deletion in individuals with 18p- syndrome. METHODS: Molecular and fluorescence in situ hybridization analyses of the pericentromeric region of chromosome 18 were performed on genomic DNA and chromosomes from study participants. RESULTS: The majority of the breakpoints were located between markers D18S852 on 18p and D18S1149 on 18q, a distance of approximately 4 Mb. The parental origin of these deletions appears to be equally distributed, half maternally derived and half paternally derived. CONCLUSION: The distributions of both the size and parental origin of the 18p deletions support the presence of a breakpoint cluster in the 18p- syndrome.