BIOM-04. CORRELATIVE PERFORMANCE OF CEREBROSPINAL FLUID (CSF) CIRCULATING TUMOR CELLS (CTC) AND CELL-FREE TUMOR-DERIVED DNA (CTDNA) IN LEPTOMENINGEAL METASTASES (LM) AND PARENCHYMAL BRAIN METASTASES (BM). (14th November 2022)

• Record Type: Journal Article
• Authors: Diaz, Maria; Reiner, Anne; Huereca, Claudia; Young, Robert; Arcila, Maria; Rosenblum, Marc; Ramanathan, Lakshmi; Bale, Tejus; Pentsova, Elena
• Is Part Of: Neuro-oncology. Volume 24(2022)Supplement 7
• Page Start: vii4
• Page End: vii4
• Publication Date: 2022-11-14
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noac209.014
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: BACKGROUND: CSF-CTC analysis through the CellSearch® platform has been studied as a diagnostic and prognostic tool for LM from solid tumors. CSF-ctDNA can be detected in LM and/or BM, with unclear significance. We aimed to evaluate CSF-CTCs and CSF-ctDNA as predictors of LM diagnosis and survival in LM and BM. METHODS: We retrospectively reviewed charts of patients with solid tumors who underwent CSF-CTC and ctDNA evaluation from the same lumbar puncture between 2016-2021 at Memorial Sloan Kettering Cancer Center. LM was diagnosed by MRI and/or CSF cytology; CSF-ctDNA was considered "positive" if next-generation sequencing (NGS) detected ≥ 1 mutations using clinically validated thresholds. Associations with newly diagnosed LM were performed using logistic regression, and with development of LM using cause-specific competing risks models. Survival analyses were performed using Cox proportional hazards modeling from date of first CSF collection, and adjusted by LM as a time-dependent variable. RESULTS: Out of 151 total patients, 99 had LM at the time of CSF analysis (58/99 with newly diagnosed LM, within 30 days of first CSF collection), and 52 had BM only, of which 8/52 (15%) developed LM later (median time from CSF to LM=235 days, range 57-364). CSF-CTCs and CSF-ctDNA were predictive of newly diagnosed LM (OR=1.006, 95% CI:1.002-1.01, p=0.002 for continuous CSF-CTCs; OR=3.60, 95% CI:1.65-7.87, p=0.001 for positive CSF-ctDNA). In a small cohort of patients without LM, CSF-ctDNA was almost statistically significant in predicting later development of LM (HR=4.14, 95% CI: 0.98-17.57, p=0.054) while continuous CSF-CTCs were not (HR=1.02, 95% CI: 0.99-1.04, p=0.21). In all patients, both CSF-CTCs (OR=1.004, 95% CI: 1.001-1.006, p=0.002) and CSF-ctDNA (OR=2.49, 95% CI: 1.51-4.15, p=0.0004) predicted survival. CONCLUSION: CSF-ctDNA could predict development of LM in patients with BM. Presence of CTCs and ctDNA in CSF were associated with poor survival in patients with central nervous system metastases.