Page 295 - Read Online

Lei et al. J Cancer Metastasis Treat 2019;5:38 I http://dx.doi.org/10.20517/2394-4722.2019.12 Page 11 of 16

Additional in-frame ESR1 translocations with diverse partner genes have now been identified in late-
stage, endocrine-refractory, ER+ metastatic cases. These include ESR1-e6>DAB2, ESR1-e6>GYG1, and
[77]
ESR1-e6>SOX9 . Like the ESR1-e6>YAP1 and ESR1-e6>PCDH11X fusions, the ESR1-e6>DAB2 and ESR1-
e6>GYG1 fusions produce stable ESR1 fusion proteins and all three were able to drive hormone-independent
[77]
activation of a ERE-luciferase reporter . Remarkably, these ESR1 fusions all follow a pattern preserving
the first six exons of ESR1, containing the N-terminal DNA binding domain fused in-frame to C-terminal
partner genes, thus excluding the LBD in ESR1 [Figure 1D]. Therefore, these additional ESR1 fusion proteins
likely drive pan-endocrine therapy resistance like our previously discovered ESR1-e6>YAP1 and ESR1-
[75]
e6>PCDH11X fusions . The functional and therapeutic significance of these additional ESR1 fusions are the
focus of ongoing investigation by our group and others.

In contrast to transcriptionally active ESR1 fusions, we also identified an in-frame ESR1-e6 fusion, ESR1-
e6>NOP2 in a treatment naïve primary breast tumor that was transcriptionally inactive despite producing
[75]
stable ESR1 fusion protein . ESR1-e6>NOP2 did not promote endocrine therapy resistant growth and was
found to bind relatively few sites in a genome-wide DNA binding assay, potentially explaining the weak
functional activity measured by our experimental systems. In addition, out-of-frame ESR1 fusions identified
in primary tumors preserving diverse exons of ESR1 gene, ESR1-e3, ESR1-e4, ESR1-e5, and ESR1-e6 did not
[75]
facilitate estrogen-independent proliferation . More studies are required to fully understand the contribution
of transcriptionally inactive in-frame and out-of-frame ESR1 fusions in breast cancer.

ESR1 fusion structural studies revealed that driver ESR1 fusions from metastatic patients follow the same
fusion pattern containing the first 6 exons of ESR1 (ESR1-e6) fused to C-termini of diverse gene partners
suggesting this pattern is strongly connected to endocrine therapy resistant, metastatic ER+ breast tumors.
The observation of a highly consistent and recurrent ESR1 breakpoint, together with the promiscuity of
ESR1 for a variety of fusion partners is certainly interesting. In prostate cancer, recurrent fusions involving
promoter regions of an androgen regulated gene, transmembrane protease serine 2 gene (TMPRSS2) fused
to coding sequences of erythroblastosis virus E26 gene (ETS) family members have been identified in
[78]
more than 50% of prostate cancer cases . Androgen receptor (AR) signaling has been shown to bring
the androgen regulated gene TMPRSS2 and the ERG gene in close proximity in prostate cancer cell line
[79]
models . Androgen signaling also generates DNA damage in the form of double strand breaks (DSBs)
at sites of TMPRSS2-ERG genomic breakpoints. These DSBs have been shown to be mediated by the class
[80]
II topoisomerase beta, TOP2B, which is recruited to AR, inducing DSBs . TMPRSS2-ERG gene fusions
can then arise from dysfunction of mechanisms to repair DSBs, such as homologous recombination
(HR) pathway and the error-prone non-homologous end-joining (NHEJ) pathway. AR-mediated DSBs in
prostate cancer may provide clues to the recurrent ESR1 breakpoints for ESR1 fusions seen in breast cancer.
Recruitment of TOP2B to ER and subsequent DSBs have been shown to occur at regulatory regions of ER
[81]
target genes as a consequence of ER-mediated transcriptional activation . Since regulatory regions of
[82]
ESR1 itself has also been shown to be bound by ER , transcription-induced DSBs by ER, coupled with
dysregulation of DSB repair mechanisms may contribute to the highly recurrent ESR1 breakpoints. Although
none of the fusion partners from endocrine-refractory, metastatic disease observed in our studies are known
ER targets, additional studies are needed to better understand the diversity of preferred ESR1 partner genes.

ESR1 fusions that contain the first six exons of ESR1 fused in-frame to partner genes are almost exclusively
observed in endocrine therapy resistant, metastatic ER+ breast cancer, with the exception of ESR1-
e6>NOP2, as described above, likely suggesting a role in driving disease pathogenesis. However, very few
functionally significant ESR1 fusions have been studied to date and therefore ESR1 fusion events remains
an understudied form of somatic mutation in breast cancer. The incidence of ESR1 fusions is also still not
well understood, especially in the metastatic setting, but the studies discussed here collectively suggest ESR1
[77]
fusions to be present in at least 1% of metastatic breast cancer cases , with the actual frequency likely to