Search

Intro: High-Grade Gliomas (HGGs) are the most common and aggressive primary brain tumors in adults and are almost always fatal. Liquid biopsy provides a noninvasive window into the cancer genome and the underlying biology of the tumor. Circulating-tumor DNA (ctDNA) is a versatile analyte for tumor diagnosis, monitoring treatment response, detecting resistance, and tracking tumor evolution. The central hypothesis of our work is that detection of cerebrospinal fluid (CSF) ctDNA correlates with clinically significant events and can be used as a surrogate for tissue biopsy to guide treatment decisions in the clinic. Methods: Our study includes CSF ctDNA samples from 140 recurrent glioma patients at Memorial Sloan Kettering Cancer Center who underwent collection of CSF as part of their clinical evaluation for neurological signs and symptoms. For each patient, glioma subtype and grade were confirmed by a neuropathologist. Overall survival was calculated as the time interval between the date of diagnosis and the date of death. All samples were sequenced using the MSK-IMPACT targeted sequencing assay (468 genes). Allele-specific copy number calls were assessed using the FACETS algorithm. Mutations were classified as drivers based on OncoKB. To evaluate the relationship between CSF ctDNA detection and clinico-pathologic correlates, CSF ctDNA status was determined positive by the presence of at least one somatic mutation and CSF ctDNA status was correlated with clinico-pathologic features. Results: Within this cohort, we found 68 CSF ctDNA positive and 72 CSF negative samples. The most frequently mutated genes were: TERT (58.2%), TP53 (47.8%), IDH1 (20.9%) and EGFR (26.9%). We observed a concordance between contemporaneously sampled tumor and CSF. In a multivariate analysis accounting for established prognostic factors including: % extent of resection at diagnosis; tumor burden at the time of lumbar puncture; and IDH status, we found that CSF ctDNA positivity was negatively correlated with overall survival (HR: 2.52, p Conclusion: ctDNA from CSF depicts a powerful analyte with the potential to alter the standard of care. We have established a robust liquid biopsy program across the neuro-oncology department at MSK, and MSK-IMPACT is now certified by NYS DOH for use on CSF ctDNA enabling routine integration into clinical care. In summary, we are now able to monitor the changing genome along the disease course and have the potential to detect disease occurrence at an earlier time point, however further validation of CSF ctDNA for disease monitoring is needed. Additionally, our data suggests that CSF ctDNA may be used as a prognostic biomarker for survival, but confirmation requires further validation in a prospective study. Citation Format: Subhiksha Nandakumar, Christoph Kreitzer, Charli Ann Hertz, Johnathan Rafailov, Timothy Song, Nicholas D. Socci, A. Rose Brannon, Maria E. Arcila, David B. Solit, Michael F. Berger, Nikolaus Schultz, Ingo K. Mellinghoff, Alexandra M. Miller. Circulating tumor DNA from cerebrospinal fluid (CSF) allows for characterization and monitoring of glioma patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1053.

On November 2017, the US Food and Drug Administration (FDA) authorized Memorial Sloan Kettering Cancer Center's (MSKCC) IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets) tumor profiling assay and bioinformatics pipeline for clinical decisions in treating all solid tumors. The IMPACT test uses next-generation sequencing (NGS) to identify the presence of mutations in 468 unique genes, fusions at known sites, copy-number alterations, and microsatellite instability (MSI). A version freeze of the bioinformatics pipeline was made public in May 2017 in Zehir et al. "Mutational Landscape of Metastatic Cancer Revealed from Prospective Clinical Sequencing of 10,000 Patients", Nat Med. 23(6): 703-713. Active development of the pipeline has continued in its CLIA-compliant computing environment, and the resulting improvements have been incorporated into an open-source portable research-grade bioinformatics pipeline. Here, we present results from Roslin, our research bioinformatics pipeline built to be functionally equivalent to the IMPACT clinical reporting workflow that includes manual filtering/annotation and signout by molecular oncologists. Roslin can be installed and executed on laptops, workstations, local compute clusters, or cloud compute servers. As a research-grade pipeline it is flexible enough to handle new assays including exomes/genomes, and extensible enough to add or replace components like reference genomes, sequence aligners, variant callers, false-positive filters, and functional/clinical annotation. Tracking all components and parameters used in each pipeline run allows researchers to easily generate method sections for manuscripts that are sufficiently detailed for peer-review and reproducibility of published results. Roslin has been deployed and tested on multiple high-performance clusters and cloud computing resources. It offers end users GUI driven workflow logging, run reporting, and real-time tracking (https://github.com/mskcc/roslin). Evaluated on NGS data from 128 tumors with signed out IMPACT clinical reports, Roslin achieves a recall rate of 98% for both somatic substitutions and indels, and 60% for fusions. Precision was 96% for substitutions, and 86% for indels. We present our work in maximizing precision and recall rates while retaining the portability, modularity, and reproducibility of Roslin. Citation Format: Zuojian Tang, Nikhil Kumar, Allan Bolipata, Timothy Song, Cyriac Kandoth, Feng Liu, David Mcmanamon, Oliver Hampton, Nicholas Socci, David Solit. A portable bioinformatics pipeline for the FDA authorized IMPACT DNAseq assay [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 2482.