Identification of translocations in single disseminated cancer cells

Metastasis is the major cause of cancer-related deaths. Occult cancer cells when found in the bone marrow of patients with carcinomas have been termed disseminated cancer cells (DCCs) and when found in the blood circulating tumor cells (CTCs). CTCs and DCCs may comprise metastasis-initiating cells and surrogate markers for minimal residual disease (MRD) after curative surgery of the primary tumor. Numerous studies have shown that the presence of DCCs in bone marrow or CTCs in blood detected at time of curative surgery or during the follow-up of the patient put the patients at risk for disease progression and death. Therefore, there is a medical need to better characterize these cells and identify early oncogenic changes that drive metastatic progression. Fusion genes have been shown to be among the very early genomic changes necessary for cancer progression, as demonstrated for the TMPRSS2-ERG rearrangement in prostate cancer, but the identification of these structural variations has been challenging until now, especially in carcinomas, due to the lack of methods able to provide high resolution. Therefore, this work aimed to develop a method for the enrichment of breakpoint sequences that can be applied to DCCs and CTCs. Since these cells are extremely rare the method should enable fusion gene identification in single cells. The technique was established as downstream application of a deterministic method to amplify the genome of single cells. The method was based on the reduction of genome complexity and subtractive hybridization. Reduction of complexity was achieved by portioning the genome into ten fractions based on deterministic fragment selection. Several approaches of subtractive hybridization were subsequently compared and applied. In order to establish the technique, single cells derived from the K562 cell line (Tester), harboring the BCR-ABL gene fusion, and a pool of unrelated lymphocytes (Driver) were used. The enrichment of the breakpoint fusion was monitored by quantitative PCR which demonstrated an enrichment of 1664-fold, sufficient to detect fusion sequences by low coverage sequencing. Thus combination of genome fractioning with subtractive hybridization is a valid approach for the enrichment of breakpoint sequences and could be applied for the genetic characterization of CTCs and DCCs. Future applications of the method might include monitoring cancer specific sequences during the course of therapy of solid cancers, as it is routinely performed for the BCR-ABL fusion gene to monitor therapy response and disease remission, or the identification of novel therapy targets, such as neo-antigens