BCOR-CCNB3 Ewing-like tumors form a histologically heterogeneous family belonging to the group of small round cell tumors, just as EWS-ETS positive "classical" Ewing.

Description of this new clinical entity in the literature is recent (Pierron et al., 2012). Clinical data of this small cohort of patients is under current study and only few pieces can be displayed for now.

BCOR-CCNB3 RT-PCR screening of a series of transcript-negative sarcomas yielded a total of 24 positive cases, out of 594 (4%), meaning this entity is rare among the rare transcript-negative sarcomas.

Clinical description is highly similar to that of Ewing sarcoma, with a median age of occurrence of 13 years, a sex ratio desequilibrated towards males, and tumors affecting preferentially long bones, the spine and pelvis. 20% of patients presented with a soft tissue tumor.

Pathological reports were consistent with an undifferentiated, small round cell sarcoma, suggestive of the Ewing sarcoma family of tumors. However, in contrast to Ewing sarcoma, approximately half of the cases lacked strong membrane positivity for CD99. BCOR-CCNB3-positive cases that were tested exhibited strong nuclear CCNB3 staining. This finding highlights the potential usefulness of a simple CCNB3 immunohistochemical assay as a diagnostic test for this subgroup of sarcoma.

Due to the recent description of this entity, retrospective work on received treatment is on-going. Still, the treatment of these tumors is generally based on combined therapy with chemotherapy associated to best local treatment.

Retrospective work on outcome of patients harboring BCOR-CCNB3 positive tumors in on-going. As in all sarcomas, prognosis might be influenced by the presence of metastases at the time of diagnosis.

BCOR-CCNB3 Ewing-like tumors show a X-chromosome paracentric inversion; the inversion results in fusion of the complete coding sequence of the BCOR gene to the last 8 exons of CCNB3, leading to a hybrid transcript and an oncogenic chimeric protein.

SNP6.0 arrays detected no alteration in 11 of the 18 BCOR-CCNB3 Ewing-like cases studied. In the other seven cases, the only recurrent abnormalities were deletions at chromosomes 17p and 10q, each observed in two cases. Neither deletion nor amplification of the fusion partners was observed. Of note, the most frequent copy-number abnormalities of Ewing sarcoma, particularly the gain at chromosome 8 that is observed in ~50% of Ewing sarcoma cases, and the gain of 1q or chromosome 12 or the loss of 16q, which are each observed in ~25% of Ewing sarcoma cases, were not recurrent characteristics of BCOR-CCNB3-positive cases. These data indicate that classic Ewing sarcomas and BCOR-CCNB3-positive tumors do not share common copy-number changes.

BCOR (BCL6 corepressor)

Xp11.4

interacting co-repressor of BCL6.

CCNB3 (cyclin B3)

Xp11.22

Cyclin B3, early meiotic cyclin in spermatogenesis.

The complete coding sequence of the BCOR gene is fused to the last 8 exons of CCNB3, leading to a hybrid transcript and an oncogenic chimeric protein. The stop codon of BCOR (TGA) is included within a putative GGTGAG donor splice-site sequence. Thus, the most likely hypothesis accounting for the fusion mRNA is that this cryptic donor site is spliced with the acceptor site of exon 5 of CCNB3. Splicing was found to be incomplete, leading to both BCOR-CCNB3 and wild-type BCOR expression.

Sequence should yield a unique and tumor specific PCR product of 290pb.

The full fusion protein is expected to have a molecular weight of 337 kDa and to include the three known ankyrin domains of BCOR as well as the two known cyclin domains of CCNB3.

5-bromodeoxyuridine (BrdU) staining of mouse NIH3T3 fibroblasts overexpressing BCOR-CCNB3 or ΔCCNB3, a truncated version of CCNB3 corresponding to the portion involved in the fusion gene, was performed. Compared to the empty vector, both expression vectors led to a twofold increase in the proportion of cells in S phase. Although cell cycle analyses clearly show that the CCNB3 portion of the fusion is potentially sufficient to mediate cell cycle effects, it is expected that oncogenesis also relies on additional phenotypic effects from the full-length fusion protein.