(P 014) A PLETHORA OF DISEASE-ASSOCIATED ALTERATIONS IN OSTEOSARCOMA

Objective: Osteosarcoma is a primary bone malignancy with a wide histomorphological spectrum which, in rare cases, can make the diagnosis challenging. Furthermore, the extreme number of numerical and structural chromosomal alterations makes it difficult to identify genetic alterations of biological relevance. A number of cancer-associated genes and regions have been reported to be altered in osteosarcoma, such as amplification of FGFR1, IGF1R, YAP1, regions on chromosome arms 6p, 8q (including MYC), 12q (including CDK4 and MDM2), and 17p, as well as homozygous loss, structural variations and single nucleotide variants (SNVs) affecting TP53, CDKN2A, RB1, ATRX, and NF2. However, none of these alterations are unique to osteosarcoma and therefore of little use in clinical diagnostics. For a large proportion of conventional osteosarcoma, particularly those affecting adults, specific mutations that would explain tumor development remain unidentified. In the present study, we aimed to further characterize the genetic alterations that underlie our cohort and several unexpected specific mutations emerged.

Methods: Genetic analyses were conducted on fresh frozen tumor biopsies from 150 conventional osteosarcomas. The patient age ranged from 3-81 years (median and mean age 16 and 21 years, respectively). DNA and RNA were extracted according to standard procedures and analyzed by: (i) whole genome mate pair sequencing to detect structural variants; (ii) whole genome paired-end sequencing to detect single nucleotide variants and insertions/deletions; (iii) SNP-array analysis to detect genome-wide DNA copy numbers; and (iv) RNA sequencing to detect single nucleotide variants, gene fusions and gene expression levels. As control, RNA sequencing data on 13 osteoblastomas were included in the global gene expression analysis.

Results: The genetic complexity of cases in our cohort varied from an extreme amount of copy number alterations to single or a few whole chromosome gains or losses. Three main groups were detected based on their mutational profile. The first group, representing approximately 40% of the cases, harbored genomic copy number alterations, structural variants and/or SNVs affecting TP53. Five of the cases with a TP53 mutation harbored a NF2 homozygous loss. The second group included four cases that harbored gene fusions and single nucleotide variants that are characteristic markers for other tumor types e.g., EWSR1::NFATC2, FUS::TCFP2, H3-3A p.G34W, IDH1 p.R132S and NRAS p.G61L. The third group included cases with gene fusions of potential biological importance that involved at least one partner gene previously reported in bone and soft tissue tumor development e.g., THRAP3::NFATC2, FOSL1::RELA, and CTNNB1::DICER. To evaluate pathways activated in osteosarcoma, gene set enrichment analysis was performed on genes that were differentially expressed between osteosarcoma and osteoblastoma. We found that gene ontology terms upregulated in osteosarcoma were related to DNA replication, sister chromatid segregation, mitotic nuclear division and cell division, and the ones downregulated in osteosarcoma compared to osteoblastoma were related to bone resorption, ossification, bone development and osteoblast differentiation.

Conclusion: Several cases in our cohort harbored mutations indicative of other bone and soft tissue tumors, which led us to question their diagnosis. These findings highlight the potential additional value of genetic analysis, particularly in diagnostically challenging cases. Our findings also show that further studies, using unbiased mutation screening, are needed to find new biologically relevant drivers in this genetically heterogenous disease.