(718c) Investigation of Circulating Tumor Cells in DCIS Patients Enabled By an Unbiased Isolation and Characterization Workflow

Raised awareness of the importance of early diagnostics in breast cancer is leading to an increased incidence of ductal carcinoma in situ (DCIS) being detected through mammographic screening. DCIS is regarded as a ‘Stage 0’ carcinoma by clinicians and widely defined by neoplastic cells still confined to the basement membrane. Most DCIS patients will receive widespread overtreatment for their carcinoma; however, the 20-year breast cancer mortality rate following a DCIS diagnosis, with or without treatment, remains at 3.3%. To better understand breast cancer carcinogenesis, recent studies have drawn attention to the clonal diversity within the DCIS tumor microenvironment and the presence of disseminated tumor cells in DCIS patients. However, there still exist gaps in understanding of characteristics at the time of diagnosis that make certain DCIS patients more likely to progress or recur. Ultimately, the goal of the project is to better inform DCIS patient stratification via the high throughput isolation and analysis of circulating biomarkers.

In this study, we establish a workflow that targets the gaps in traditional tissue-based diagnostics by pivoting towards liquid biopsies. Our study not only builds on emerging literature on the early dissemination of breast cancer cells during carcinogenesis, but also investigates the value of circulating tumor cells (CTCs) as a biomarker in DCIS. To achieve a more holistic understanding of a patient’s carcinoma and potential risk of progression to invasive disease and/or recurrence, we supplement our study with the Mouse INtraDuctal (MIND) models - an animal model that recapitulates the tumor microenvironment of DCIS in the mammary ducts of mice.

We employed a high throughput, label-free microfluidic platform to enrich CTCs from 22 DCIS patient samples. We used immunofluorescent staining to characterize our CTCs and their inherent heterogeneity. We also conducted a pilot single cell RNA sequencing study of the captured DCIS CTCs to find expression of known migratory and stem-like cells genes. Finally, to correlate findings in patient samples to an animal model wherein the timeline of progression to invasive disease can be shortened to 9 to 12 months (vs. 10+ years), we analyzed blood samples from three different MIND models.

Applying our workflow to 22 patient samples and 10 healthy controls, we identified CTCs in 95% of the patient samples. The average concentration of CTCs per mL of blood across all DCIS patients was significantly greater than that in a cohort of 10 healthy controls (7.8 vs 1.2, p = 0.0003, Mann-Whitney U Test). Single cell transcriptomic analysis of CTCs revealed a diverse tumor cell population expressing genes characteristic of epithelial to mesenchymal phenotypes. In addition, CTCs were also found in MIND models prior to the detection of invasion, providing preliminary support for the early dissemination of cancer cells in DCIS.

Overall, our study employs a workflow that can identify characteristics correlated with greater risk of progression. By studying CTCs in patient samples and their derived MIND models in parallel, we aim to continue to build an understanding of the drivers behind early dissemination. Ultimately, we will continue to explore the value of CTCs as an early biomarker of aggressive breast cancer that can help improve risk stratification of DCIS patients.