Circulating Cell-Free DNA Reflects the Clonal Evolution of Breast Cancer Tumors

Abstract

Liquid biopsy of cell-free DNA (cfDNA) is proposed as a potential method for the early detection of breast cancer (BC) metastases and following the clonal evolution of BC. Though the use of liquid biopsy is a widely discussed topic in the field, only a few studies have demonstrated such usage so far. We sequenced the DNA of matched primary tumor and metastatic sites together with the matched cfDNA samples from 18 Eastern Finnish BC patients and investigated how well cfDNA reflected the clonal evolution of BC interpreted from tumor DNA. On average, liquid biopsy detected 56.2 ± 7.2% of the somatic variants that were present either in the matched primary tumor or metastatic sites. Despite the high discordance observed between matched samples, liquid biopsy was found to reflect the clonal evolution of BC and identify novel driver variants and therapeutic targets absent from the tumor DNA. Tumor-specific somatic variants were detected in cfDNA at the time of diagnosis and 8.4 ± 2.4 months prior to detection of locoregional recurrence or distant metastases. Our results demonstrate that the sequencing of cfDNA may be used for the early detection of locoregional and distant BC metastases. Observed discordance between tumor DNA sequencing and liquid biopsy supports the parallel sequencing of cfDNA and tumor DNA to yield the most comprehensive overview for the genetic landscape of BC.

Keywords: biomarker; intratumoral heterogeneity; liquid biopsy; metastasis; recurrence; sequencing; tumor evolution.

Figure 1

Functional and clinical relevance of detected somatic variants. (a) Distribution of somatic variants per gene is typical for BC and highlights the high mutation burden of genes TP53, AKT1, ARID1A, and NOTCH1. (b) Functional consequence of somatic variants as predicted by the CGI. Almost half of the variants were predicted to have gain-of-function or loss-of-function type consequences for the gene product. (c) CGI identified 56 known or predicted driver variants. In general, these variants were well prominent in the generated tumor evolution models, thus supporting the selective advantage provided by drivers. (d) Identified drug sensitivities as reported by the CGI. Most drugs that are reported to be sensitive for the detected somatic variants are still in pre-clinical or early clinical trials.

Figure 2

Evolutionary trees constructed from the multi-region sequencing results illustrate how locoregional metastases and distant metastases emerge from primary tumors and acquire new somatic variants over time. Each node in the trees represents a single subclone detected in the tumor samples, top node representing a hypothetical healthy cells where somatic variants are not detected. Abbreviations P, M, and DM with a running number refer to sequenced primary tumor, locoregional metastases, and distant metastases. The clonal structure of sequenced samples is visualized below the evolutionary trees as color-coded circles where the area of the circles is proportional to the estimated cancer cell fraction and each color represents a separate subclone. Predicted driver variants and variants with known pathogenicity are shown next to the evolutionary trees.

Figure 3

Concordance between tumor DNA and cfDNA sequencing results. (a) Comparison between matched tumor DNA and cfDNA samples visualized as a coMut plot. Each row represents one gene and each column represents one BC case. Bar plot at the top represent the observed concordance between matched samples while bar plot at the right represents the somatic variant count per gene. Circles within matrix cells represent somatic variants detected with the sequencing, blue and red circles corresponding to tumor-specific variants that originated either from the primary tumor or metastatic sites. Size of the circle corresponds to the VAF at the tumor. Circles with black line correspond to variants that were detected with the liquid biopsy as well while crosses represent variants that were detected only in the cfDNA. In general, liquid biopsy detects somatic variants that are represented with a relatively high VAF in primary tumor, locoregional recurrence (LR), or distant metastases (DM). (b) Venn diagram representation of detected somatic variant counts in different sample types illustrates how sequencing results of primary tumor, LRs and distant metastases, and cfDNA overlap with each other. (c) Statistically significant Pearson correlation (p = 0.003) was observed between matched tumor VAFs and cfDNA VAFs, thus supporting the idea that VAF in cfDNA reflects the clonal structure of primary tumor. All samples were included in the analysis.

Figure 4

Clonal evolution of KBCP cases and corresponding VAFs in the serial cfDNA samples. Lineplots represent the detected VAF in sequenced cfDNA samples at the time of diagnosis and at the latest follow-up prior to the detection of LR or distant metastases. Fishplots below represent the corresponding clonal evolution in matched tumor samples. Plots (a–f) represent the disease progression of separate BC cases and illustrate how detected cfDNA VAFs follow the corresponding tumor VAFs especially in the case of trunk variants which makes it possible to detect bottlenecks during the disease progression with liquid biopsy. An interesting example is case KBCP-1746 (f) where the sequencing of cfDNA is inconsistent with the tumor evolution model and suggests that tumor samples do not represent ITH reliably.