Circulating epithelial tumor cell analysis in CSF in patients with leptomeningeal metastases.

OBJECTIVE: The primary objective was to determine the sensitivity and specificity of epithelial cell adhesion molecule (EpCAM) immunoflow cytometry circulating tumor cells (CTC) analysis in CSF in patients with suspected leptomeningeal metastases (LM). The secondary objective was to explore the distribution of driver mutations in the primary tumor, plasma, cell free CSF (cfCSF), and isolated CTC from CSF in non-small cell lung cancer (NSCLC). METHODS: We tested the performance of the CTC assay vs CSF cytology in a prospective study in 81 patients with a clinical suspicion of LM but a nonconfirmatory MRI. In an NSCLC subcohort, we analyzed circulating tumor (ct)DNA of the selected driver mutations by digital droplet PCR (ddPCR). RESULTS: The sensitivity of the CTC assay was 94% (95% confidence interval [CI] 80-99) and the specificity was 100% (95% CI 91-100) at the optimal cutoff of 0.9 CTC/mL. The sensitivity of cytology was 76% (95% CI 58-89). Twelve of the 23 patients with NSCLC had mutated epidermal growth factor receptor (EGFR). All 5 tested patients with LM demonstrated the primary EGFR driver mutation in cfCSF. The driver mutation could also be detected in CTC isolated from CSF. CONCLUSION: CTC in CSF are detected with a high sensitivity for the diagnosis of LM. ddPCR can determine EGFR mutations in both cfCSF and isolated CTC from CSF of patients with EGFR-mutated NSCLC and LM. CLASSIFICATION OF EVIDENCE: This study provides Class III evidence that EpCAM-based immunoflow cytometry analysis of CSF accurately identifies patients with LM.

Preanalytical blood sample workup for cell-free DNA analysis using Droplet Digital PCR for future molecular cancer diagnostics.

In current molecular cancer diagnostics, using blood samples of cancer patients for the detection of genetic alterations in plasma (cell-free) circulating tumor DNA (ctDNA) is an emerging practice. Since ctDNA levels in blood are low, highly sensitive Droplet Digital PCR (ddPCR) can be used for detecting rare mutational targets. In order to perform ddPCR on blood samples, a standardized procedure for processing and analyzing blood samples is necessary to facilitate implementation into clinical practice. Therefore, we assessed the technical sample workup procedure for ddPCR on blood plasma samples. Blood samples from healthy individuals, as well as lung cancer patients were analyzed. We compared different methods and protocols for sample collection, storage, centrifugation, isolation, and quantification. Cell-free DNA (cfDNA) concentrations of several wild-type targets and BRAF and EGFR-mutant ctDNA concentrations quantified by ddPCR were primary outcome measurements. Highest cfDNA concentrations were measured in blood collected in serum tubes. No significant differences in cfDNA concentrations were detected between various time points of up to 24 h until centrifugation. Highest cfDNA concentrations were detected after DNA isolation with the Quick cfDNA Serum & Plasma Kit, while plasma isolation using the QIAamp Circulating Nucleic Acid Kit yielded the most consistent results. DdPCR results on cfDNA are highly dependent on multiple factors during preanalytical sample workup, which need to be addressed during the development of this diagnostic tool for cancer diagnostics in the future.