Abstract Background: Serial cell free DNA (cfDNA) analysis in metastatic breast cancer (MBC) is a noninvasive method for tracking tumor evolution and the emergence of new somatic alterations through the course of the disease. The detection of new actionable alterations may impact treatment selection; thus, serial cfDNA collection is increasingly being performed in the clinic. However, it is not well-understood how often serial cfDNA testing identifies new actionable alterations. We evaluated changes in the genomic spectrum of actionable alterations in serial cfDNA analysis of HER2 negative (HER2-) MBC. Method: Patients with HER2- MBC who underwent plasma-based cfDNA testing (Guardant360, 74-gene assay) between February 2015 and February 2021 at an academic institution were included. A retrospective review of records was conducted to identify subtype, demographics, lines of therapy, and cfDNA results. At baseline cfDNA testing, all cfDNA alterations were considered new. For patients with serial draws, new cfDNA alterations in each draw, compared to previous draws, were quantified and characterized. The pathogenicity of new alterations was determined using the OncoKB precision oncology database. New pathogenic alterations were further classified as actionable alterations (AA) using the ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT). Alterations that met the ESCAT I (alteration-drug match was associated with improved outcomes in clinical trials) or ESCAT II (alteration-drug match was associated with antitumor activity) criteria were considered as AA. Results: 344 patients with hormone receptor positive (HR+) MBC and 95 patients with triple negative (TN) MBC had a baseline cfDNA draw. Among the HR+ cohort, 139, 79 and 48 patients had 2nd, 3rd, and 4th serial cfDNA draws, respectively. Among the TN cohort, 33 and 18 patients had 2nd and 3rd serial cfDNA draws, respectively. Table 1 depicts the prevalence of AA found in each of these draws, as well as the median number of prior therapies at each point of collection and the mean time between serial draws. The median number of new genomic alterations was lower in subsequent draws compared to the baseline, regardless of subtype (HR+: 4 vs. 2, TN: 4 vs. 1.5-3, at baseline vs. subsequent draws, respectively). In the HR+ cohort the proportion of patients with new AA (ESCAT I/II) decreased from 63% at baseline to 27-33% in the 2nd-4th draws. While some of the new AA in subsequent draws were new actionable variants in the same genes that were known to be altered in previous draws (in 19%, 68% and 57% of patients with new AA in the 2nd, 3rd and 4th draws, respectively), the remainder were new AA in previously unaltered genes. In the TN cohort 25% of patients had new AA at baseline and this proportion of TN patients with new AA continued to decrease in subsequent draws (9% and 0% in the 2nd and 3rd draws, respectively). PIK3CA and ESR1 were the most frequent genes with new AA in the ER+ cohort, regardless of draw number. In the TN cohort, PIK3CA was the most frequent gene with new AA in the first draw, and in the second draw, new AA in PIK3CA, BRCA2, and ERBB2 were present at an equal frequency. Conclusion: While the proportion of patients with HER2- MBC identified to have new AA in serial cfDNA decreased with time for both HR+ and TNBC, new alterations continue to emerge, particularly for patients with HR+ MBC. Further research is needed to determine the impact of serial cfDNA testing on the selection of genotype-matched therapy and outcomes. Table 1. Citation Format: Yael Bar, Jennifer C. Keenan, Lianne Ryan, Dejan Juric, Jennifer Shin, Seth A. Wander, Laura M. Spring, Beverly Moy, Leif Ellisen, Steven J. Isakoff, Aditya Bardia, Neelima Vidula. Changes in the Genomic Spectrum of Actionable Alterations in HER2 Negative Metastatic Breast Cancer in Serial Cell Free DNA (cfDNA) Analysis [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P4-01-14.
Read full abstract