Next-generation sequencing of circulating tumor DNA (ctDNA) is integral to optimal cancer treatment, as exploring ctDNA can provide a complete picture of a tumor’s genetic landscape and can thus help guide treatment for the patient. The IMAGE II study (NCT02965755) investigated the clinical utility of serial ctDNA assessment in patients with metastatic breast cancer (mBC).
The study enrolled 198 patients with mBC of any subtype who were planning to initiate a new line of therapy. Plasma samples were collected at baseline, 1 to 2 weeks after line-of-therapy initiation, first restaging, and disease progression. Comprehensive genomic profiling (CGP) was performed on these samples by Foundation Medicine, which included a real-time review of plasma CGP and sequencing results by the Johns Hopkins Genetic Alteration in Tumors With Actionable Yields (GAITWAY) Tumor Board, calculation of ctDNA tumor fraction by FoundationOne® Liquid CDx, and calculation of maximum variant allele frequency (maxVAF). Tumor fraction and maxVAF dynamics were then compared with clinical response at first restaging. Alterations were also explored, and classified as emerging, disappearing, or persistent.
Samples collected at weeks 1 to 2 showed that a decrease in ctDNA tumor fraction after starting a new therapy was associated with a clinical response at the time of first restaging. Specifically, 44% of patients with a clinical response showed a ≥90% decrease in tumor fraction from baseline compared with only 9% of patients with clinical improvement. In contrast, changes in maxVAF estimated for 176 samples at baseline and 106 samples at 1 to 2 weeks posttreatment did not show a statistically significant association with clinical response at first restaging (23% vs 12%; P=.27). These findings suggest that early changes in ctDNA levels could serve as an indicator of treatment efficacy, and that tumor fraction may be a more sensitive marker for early treatment response assessment compared with maxVAF.
Beyond predicting treatment response, the study also highlighted the dynamic nature of tumor genomics over time. Of 958 alterations not classified as variants of unknown significance, 43% were emerging, 17% were disappearing, 28% were persistent, and 12% were intermittently detected.
When restricting analysis to clinically actional genes in breast cancer, 27% had emerging mutations, most frequently BRCA2 and ESR1, and 29% had disappearing mutations, most frequently ESR1 and PIK3CA. Emerging mutations were often detected at the time of disease progression, though disappearing mutations were frequently observed at the 1- to 2-week time point, potentially reflecting the impact of treatment. The identification of emerging mutations, particularly in actionable genes, could have implications for treatment strategies and the selection of targeted therapies.
The IMAGE II study demonstrates the value of serial ctDNA analysis in mBC. Early changes in ctDNA tumor fraction can predict treatment response at restaging, providing valuable information for clinical decision-making. Furthermore, serial liquid biopsies offer insights into the dynamic genomic evolution of tumors, revealing emerging and disappearing alterations that may inform treatment strategies and personalize cancer care.
Linville L, et al. Serial circulating tumor DNA (ctDNA) assessment to predict treatment response and identify genomic evolution in patients with metastatic breast cancer (mBC). Presented at: San Antonio Breast Cancer Symposium. December 10-13, 2024; San Antonio, TX. Abstract SESS-2246.