Utilizing cell-free DNA to predict risk of developing brain metastases in patients with metastatic breast cancer

Neelima Vidula; Andrzej Niemierko; Katherine Hesler; Lianne Ryan; Beverly Moy; Steven Isakoff; Leif Ellisen; Dejan Juric; Aditya Bardia

doi:10.1038/s41523-023-00528-z

Utilizing cell-free DNA to predict risk of developing brain metastases in patients with metastatic breast cancer

NPJ Breast Cancer. 2023 Apr 19;9(1):29. doi: 10.1038/s41523-023-00528-z.

Authors

Neelima Vidula¹, Andrzej Niemierko², Katherine Hesler², Lianne Ryan², Beverly Moy², Steven Isakoff², Leif Ellisen², Dejan Juric², Aditya Bardia²

Affiliations

¹ Massachusetts General Hospital Cancer Center, Boston, MA, 02114, USA. nvidula@mgh.harvard.edu.
² Massachusetts General Hospital Cancer Center, Boston, MA, 02114, USA.

Abstract

We compared cell-free DNA (cfDNA) results at MBC diagnosis in patients who developed brain metastases (BM) vs those without (non-BM) to understand genomic predictors of BM. Patients with cfDNA testing at MBC diagnosis (Guardant360®, 73 gene next generation sequencing) were identified. Clinical and genomic features of BM and non-BM were compared (Pearson's/Wilcoxon rank sum tests). Eighteen of 86 patients (21%) with cfDNA at MBC diagnosis developed BM. Comparing BM vs non-BM, a higher prevalence of BRCA2 (22% vs 4.4%, p = 0.01), APC (11% vs 0%, p = 0.005), CDKN2A (11% vs 1.5%, p = 0.05), and SMAD4 (11% vs 1.5%, p = 0.05) was observed. Seven of 18 BM had ≥1 of the following 4 mutations in baseline cfDNA: APC, BRCA2, CDKN2A or SMAD4 vs 5/68 non-BM (p = 0.001). Absence of this genomic pattern had a high negative predictive value (85%) and specificity (93%) in excluding BM development. Baseline genomic profile varies in MBC that develops BM.