ALK-Positive Anaplastic Large Cell Lymphoma with the Variant EEF1G-, RNF213- and Atic-ALK Fusions Is Featured By Copy Number Gain of the Rearranged ALK Gene

Molecular pathogenesis of ALK-positive Anaplastic Large Cell Lymphoma (ALK+ ALCL) is not completely understood. Approximately 80% of ALK+ ALCL cases harbor the t(2;5)(p23;q35)-associated NPM1-ALK rearrangement, while several variant 2p23/ALK translocations involving at least nine partner genes have been identified in the remaining cases. The 5'ALK partners play a key role in the constitutive activation of the chimeric protein by mediating its oligomerization and its subcellular localization. Their role in ALK lymphomagenesis is debated, but experimental studies showed that the partners impact a range of biological activities of ALK chimeras, including proliferation, transformation and metastatic capacities. Comparative analysis of biological properties of ALK oncoproteins, however, is hampered by the relative low frequency of different variant ALK fusions.

To gain more insight into the molecular pathogenesis of ALK+ ALCL, we have characterized ALK fusions in five cases recently diagnosed in our institution. All tumors showed a cytoplasmic expression of ALK indicating the presence of a variant t(2p23/ALK). Using 5'RACE PCR, low coverage full genome sequencing and FISH, all five ALK rearrangements were characterized. The identified partner genes included EEF1G (Eukaryotic translation elongation factor 1 gamma), a novel ALK partner located at 11q12.3 (one case), and the already known partners, RNF213/ALO17 (17q25) (one case) and ATIC (2q35) (3 cases). Notably, all five cases displayed a similar LSI ALK break-apart FISH pattern indicative of a reciprocal t(2p23/ALK) associated with copy number gain of the rearranged ALK gene. The LSI 3'ALK/red signal was duplicated on der(11)t(2;11)(p23;q12.3) in the case with the novel EEF1G-ALK fusion and amplified in cases with the RNF213-ALK (5-7 extra red signals) and ATIC-ALK (2-4 extra red signals) rearrangements. The ATIC-ALK cases presumably harbor the inv(2)(p23q35) resulting in the reciprocal ATIC-ALK rearrangement, and 1 or 2 copies of the derivative i(2)(q10) carrying two additional ATIC-ALK loci each, as previously reported¹. Of note, the original case with RNF213-ALK was also featured by the ALK copy number gain². FISH findings were confirmed by array CGH in two available cases. These data together with our previous observations^1,2 provide strong evidence that ALCL driven by at least three variant ALKfusions (EEF1G-, RFN213- and ATIC-ALK), but not by the classic NPM1-ALK, requires an increased gene dosage of the rearranged ALK. To assess whether this need is caused by the weaker promoter of EEF1G, RFN213 and ATIC, compared to the NPM1 promoter, we determined at first the relative mRNA expression level of the four ALK partner genes. The study revealed a significantly lower expression of EEF1G, RNF213 and ATIC in nonmalignant lymph nodes when compared to NPM1. Therefore, we presume that lymphomas driven by the EEF1G-, RNF213- and ATIC-ALK fusions may compensate the relatively low expression of ALK by an increased gene dosage. In the next step, we will compare oncogenic potential of all four fusions in the murine hematopoietic IL-3 dependent Ba/F3 cell line. Cloning of the EEF1G-, RFN213-, ATIC- and NPM1-ALK expression vectors is ongoing. Interestingly, the three variant fusions predominantly occurred in adult patients (4/5 in age > 48 years). All patients were treated with chemotherapy, but three of them relapsed after 8 (EEF1G-ALK), 13 (RNF213-ALK) and 20 (ATIC-ALK) months. The child with EEF1G-ALK relapsed three times and died 73 months after diagnosis due to GVHD post stem cell transplantation.

In summary, we observed a recurrent gain of the rearranged ALK gene in ALK+ ALCL driven by three variant fusions: EEF1G-ALK (novel), RNF213-ALK and ATIC-ALK. Our findings suggest that expression pattern and/or transforming capacities of these fusions might be impacted by biological features of the partner genes. We plan to validate our hypothesis in functional studies. Preliminary clinical data suggest a poor clinical outcome of patients with the three variant fusions.

References

1. Wlodarska, I., et al. Blood.92, 2688-95 (1998)

2. Cools, J., et al. Genes Chromosomes Cancer. 34, 354-62 (2002)