AbstractBackgroundApproximately 1% of clinically treatable tyrosine kinase fusions, including anaplastic lymphoma kinase, neurotrophic tyrosine receptor kinase, RET proto-oncogene, and ROS proto-oncogene 1, have been identified in soft tissue sarcomas via comprehensive genome profiling based on DNA sequencing. Histologic tumor-specific fusion genes have been reported in approximately 20% of soft tissue sarcomas; however, unlike tyrosine kinase fusion genes, these fusions cannot be directly targeted in therapy. Approximately 80% of tumor-specific fusion-negative sarcomas, including myxofibrosarcoma and leiomyosarcoma, that are defined in complex karyotype sarcomas remain genetically uncharacterized; this mutually exclusive pattern of mutations suggests that other mutually exclusive driver oncogenes are yet to be discovered. Tumor-specific, fusion-negative sarcomas may be associated with unique translocations, and oncogenic fusion genes, including tyrosine kinase fusions, may have been overlooked in these sarcomas.Questions/purposes(1) Can DNA- or RNA-based analysis reveal any characteristic gene alterations in bone and soft tissue sarcomas? (2) Can useful and potential tyrosine kinase fusions in tumors from tumor-specific, fusion-negative sarcomas be detected using an RNA-based screening system? (3) Do the identified potential fusion tumors, especially in neurotrophic tyrosine receptor kinase gene fusions in bone sarcoma, transform cells and respond to targeted drug treatment in in vitro assays? (4) Can the identified tyrosine kinase fusion genes in sarcomas be useful therapeutic targets?MethodsBetween 2017 and 2020, we treated 100 patients for bone and soft tissue sarcomas at five institutions. Any biopsy or surgery from which a specimen could be obtained was included as potentially eligible. Ninety percent (90 patients) of patients were eligible; a further 8% (8 patients) were excluded because they were either lost to follow-up or their diagnosis was changed, leaving 82% (82 patients) for analysis here. To answer our first and second questions regarding gene alterations and potential tyrosine kinase fusions in eight bone and 74 soft tissue sarcomas, we used the TruSight Tumor 170 assay to detect mutations, copy number variations, and gene fusions in the samples. To answer our third question, we performed functional analyses involving in vitro assays to determine whether the identified tyrosine kinase fusions were associated with oncogenic abilities and drug responses. Finally, to determine usefulness as therapeutic targets, two pediatric patients harboring an NTRK fusion and an ALK fusion were treated with tyrosine kinase inhibitors in clinical trials.ResultsDNA/RNA-based analysis demonstrated characteristic alterations in bone and soft tissue sarcomas; DNA-based analyses detected TP53 and copy number alterations of MDM2 and CDK4. These single-nucleotide variants and copy number variations were enriched in specific fusion-negative sarcomas. RNA-based screening detected fusion genes in 24% (20 of 82) of patients. Useful potential fusions were detected in 19% (11 of 58) of tumor-specific fusion-negative sarcomas, with nine of these patients harboring tyrosine kinase fusion genes; five of these patients had in-frame tyrosine kinase fusion genes (STRN3-NTRK3, VWC2-EGFR, ICK-KDR, FOXP2-MET, and CEP290-MET) with unknown pathologic significance. The functional analysis revealed that STRN3-NTRK3 rearrangement that was identified in bone had a strong transforming potential in 3T3 cells, and that STRN3-NTRK3–positive cells were sensitive to larotrectinib in vitro. To confirm the usefulness of identified tyrosine kinase fusion genes as therapeutic targets, patients with well-characterized LMNA-NTRK1 and CLTC-ALK fusions were treated with tyrosine kinase inhibitors in clinical trials, and a complete response was achieved.ConclusionWe identified useful potential therapeutic targets for tyrosine kinase fusions in bone and soft tissue sarcomas using RNA-based analysis. We successfully identified STRN3-NTRK3 fusion in a patient with leiomyosarcoma of bone and determined the malignant potential of this fusion gene via functional analyses and drug effects. In light of these discoveries, comprehensive genome profiling should be considered even if the sarcoma is a bone sarcoma. There seem to be some limitations regarding current DNA-based comprehensive genome profiling tests, and it is important to use RNA testing for proper diagnosis and accurate identification of fusion genes. Studies on more patients, validation of results, and further functional analysis of unknown tyrosine kinase fusion genes are required to establish future treatments.Clinical RelevanceDNA- and RNA-based screening systems may be useful for detecting tyrosine kinase fusion genes in specific fusion-negative sarcomas and identifying key therapeutic targets, leading to possible breakthroughs in the treatment of bone and soft tissue sarcomas. Given that current DNA sequencing misses fusion genes, RNA-based screening systems should be widely considered as a worldwide test for sarcoma. If standard treatments such as chemotherapy are not effective, or even if the sarcoma is of bone, RNA sequencing should be considered to identify as many therapeutic targets as possible.
