Novel <i>MAML2</i> fusions as oncogenic drivers in human malignancy.

Abstract

e15067 Background: Our group previously identified the oncogenic fusion CRTC1::MAML2 , arising from t(11;19) translocations in mucoepidermoid carcinoma, the most common type of salivary gland malignancy. Subsequent studies have reported additional fusions of MAML2 with CRTC3 , YAP1 , and NR1D1 . These fusions retain the C-terminal MAML2 transactivating domain (TAD), which is believed to drive aberrant gene transcription. While the oncogenic roles of these MAML2 fusions have been well-established in preclinical and clinical studies, we aimed to identify novel MAML2 fusions across a broader range of human malignancies. Methods: DNA and RNA sequencing were performed on tumor samples submitted to Caris Life Sciences. MAML2 fusions were identified from RNA transcripts and filtered to include only known pathogenic fusions or in-frame fusions containing a C-terminal MAML2 TAD that were observed in ≥3 samples. Fusion burden was defined as the number of unique fusion isoforms per sample. Statistical significance was evaluated using Mann-Whitney, Fisher’s Exact, or Chi-squared tests. Results: Among 180,124 tumor samples analyzed, 509 specimens were found to harbor MAML2 fusion transcripts. After filtering for non-recurrent fusions and those lacking a MAML2 TAD, 143 unique fusions remained. Over half of these specimens harbored known fusions involving CRTC1 (42/143), YAP1 (37/143), CRTC3 (7/143), and NR1D1 (2/143). In addition, novel fusions with MTMR2 (31/143), SESN3 (11/143), CCDC82 (6/143), FAM76B (4/143), and ATXN3 (3/143) were identified. Compared to known fusions, the novel fusions were identified in fewer sequencing reads (8 vs 13 median junction reads, p = 0.0064) and tended to have a higher median fusion burden (6 vs 2 median fusion isoforms per sample, p &lt; 0.0001), more frequent TP53 co-mutations (80% vs 11.5%, p &lt; 0.0001), and no clear association with tissue of origin. Novel fusions, excluding ATXN3::MAML2 , were located near MAML2 in the genome and appeared to result from duplication or deletion events. At least 75% of samples harboring each novel fusion also carried a concurrent pathogenic mutation, primarily involving TP53 . In contrast, ATXN3::MAML2 fusions, the product of which interacts with YAP1, arose via translocation, occurred in samples with a low fusion burden (2 fusion isoforms per sample), and were not associated with TP53 mutations. Conclusions: Our comprehensive analysis of tumor samples identified several novel MAML2 fusion partners. Most of these appear to represent passenger alterations arising from genomic instability, likely due to impaired p53 function. However, ATXN3::MAML2 fusions, previously reported in a precancerous pancreatic disease case, may represent a pathogenic alteration warranting further investigation.