Neurotrophic tyrosine receptor kinase (NTRK) fusions are promising molecular targets that have been described in a broad range of malignant tumours. Fusions commonly lead to the expression of chimeric proteins with constitutive tyrosine kinase activation that drives tumorigenesis. Despite a low prevalence among most solid tumours (<1%), the first encouraging results with pan-NTRK tyrosine kinase inhibitors (TKIs) such as larotrectinib or entrectinib stimulated the search for eligible patients. Here, we report the first three cases of osteosarcoma harbouring NTRK fusions, among 113 patients sequenced. It is also the first report on NTRK fusions within a tumour type characterised by highly rearranged genomes and abundant passenger mutations. Whereas the presence of NTRK gene fusions in many tumours is considered to be one of the main driver events for tumour progression, the three chimeric transcripts described here appear non-functional and likely represent randomly occurring passenger alterations. Particularly in tumours with complex karyotypes, it may therefore be advisable to specifically investigate the fusion transcripts for functional impact before considering targeted treatment approaches using pan-NTRK TKIs.

Osteosarcoma is the most common primary malignant bone tumour in children and adolescents. More than a third of patients do not respond to standard therapy and urgently require alternative treatment options. Due to a high degree of inter- and intra-tumoural genomic heterogeneity and complexity, recurrent molecular alterations that could serve as prognostic predictors or therapeutic targets are still lacking in osteosarcoma. Copy number (CN) gains involving the IGF1R gene, however, have been suggested as a potential surrogate marker for treating a subset of patients with IGF1R inhibitors. In this study, we screened a large set of osteosarcomas and found specific CN gains of the IGF1R gene in 18 of 253 (7.1%) cases with corresponding IGF1R overexpression. Despite the discouraging results observed in clinical trials in other tumours so far, focusing only on selected patients with osteosarcoma that show evidence of IGF pathway activation might represent a promising new and innovative treatment approach.

Sporadic giant cell granulomas (GCGs) of the jaws and cherubism-associated giant cell lesions share histopathological features and microscopic diagnosis alone can be challenging. Additionally, GCG can morphologically closely resemble other giant cell-rich lesions, including non-ossifying fibroma (NOF), aneurysmal bone cyst (ABC), giant cell tumour of bone (GCTB), and chondroblastoma. The epigenetic basis of these giant cell-rich tumours is unclear and DNA methylation profiling has been shown to be clinically useful for the diagnosis of other tumour types. Therefore, we aimed to assess the DNA methylation profile of central and peripheral sporadic GCG and cherubism to test whether DNA methylation patterns can help to distinguish them. Additionally, we compared the DNA methylation profile of these lesions with those of other giant cell-rich mimics to investigate if the microscopic similarities extend to the epigenetic level. DNA methylation analysis was performed for central (n = 10) and peripheral (n = 10) GCG, cherubism (n = 6), NOF (n = 10), ABC (n = 16), GCTB (n = 9), and chondroblastoma (n = 10) using the Infinium Human Methylation EPIC Chip. Central and peripheral sporadic GCG and cherubism share a related DNA methylation pattern, with those of peripheral GCG and cherubism appearing slightly distinct, while central GCG shows overlap with both of the former. NOF, ABC, GCTB, and chondroblastoma, on the other hand, have distinct methylation patterns. The global and enhancer-associated CpG DNA methylation values showed a similar distribution pattern among central and peripheral GCG and cherubism, with cherubism showing the lowest and peripheral GCG having the highest median values. By contrast, promoter regions showed a different methylation distribution pattern, with cherubism showing the highest median values. In conclusion, DNA methylation profiling is currently not capable of clearly distinguishing sporadic and cherubism-associated giant cell lesions. Conversely, it could discriminate sporadic GCG of the jaws from their giant cell-rich mimics (NOF, ABC, GCTB, and chondroblastoma).