More than 67% of deaths in breast cancer patients occur after the initial 5-year survival period while residual disease can be dormant for periods longer than 20 years. Patients are asymptomatic because circulating tumor cells (CTCs) remain dormant and are undetectable by current clinical tools. Dormant CTCs may retain their long-term tumor-initiating (LTI) potential by adhering to their original genome, unlike rapidly cycling cancer cells that are known to have increased genomic instability. We hypothesized that hyperactive mechanisms of DNA repair preserve the genomic make-up of dormant CTCs allowing them to retain their LTI potential, ultimately causing disease relapse.

We isolated and characterized EpCAM-negative breast cancer CTCs by mutiparametric flow cytometry and DEPArrayTM. Individually isolated breast cancer CTCs had a large proportion (>40%) of dormant (Ki67-/PCNA-) cells. Dormant CTCs had a lower incidence of double-strand DNA breaks (DSB) than proliferating cells as assessed by the phosphorylation status of Serine139 on gamma H2AX. This observation was further validated in a panel of eight genetically distinct breast cancer cell lines. Second, to understand whether dormant cells are inherently more resistant to DSB, we induced DSB in breast cancer cells by UV radiation and bleomycin treatment, and measured residual DSB at regular intervals. Results showed that besides being more resistant to DSB de novo, dormant breast cancer cells were also more efficient in repairing their DNA. There are two distinct phases of DSB repair - early [within 2 hours of DSB using Non-Homologous End Joining (NHEJ) methods] and late [evident after 24 hours using Homologous Recombination (HR)]. Unlike proliferating (S-G2M) cells, dormant (G0) cells lack the sister chromatid and repair their DNA exclusively by NHEJ methods. Therefore, and third, we investigated key players of the NHEJ pathway and examined their roles in maintaining genomic integrity. We found that the human telomere-associated protein RIF1, a mediator of alternative NHEJ, was significantly up-regulated in a dormant CTC subset. Dormant sub-populations of breast cancer cells confirmed RIF1 foci formation in areas of DNA damage. Fourth, mis-sense mutation of RIF1 in CAMA-1 cells (ΔRIF1 E1598K) attenuated resistance of the dormant subset to UV and bleomycin treatment.

Collectively, these findings suggest that RIF1 may play functional roles in maintaining the genomic integrity of dormant CTCs. Further investigations are being pursued to assess RIF1 contributions to retain CTC LTI potential leading to CTC-driven metastasis.

Citation Format: Debasish Boral, Haowen N. Liu, Wei Yin, Monika Vishnoi, Antonio Scamardo, Goldy C. George, David S. Hong, Dario Marchetti. Deciphering mechanisms of circulating tumor cells in breast cancer dormancy. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3143.