Bone and soft tissue tumors are heterogeneous, diagnostically challenging, and often defined by gene fusions.


To present our experience using a custom 34-gene targeted sequencing fusion panel.


Total nucleic acid extracted from formalin-fixed, paraffin-embedded (FFPE) tumor specimens was subjected to open-ended, nested anchored multiplex polymerase chain reaction and enrichment of 34 gene targets, thus enabling detection of known and novel fusion partners.


During a 12-month period, 145 patients were tested as part of routine clinical care. Tumor percentage ranged from 10% to 100% and turnaround time ranged from 3 to 15 (median, 7.9) days. The most common diagnostic groups were small round blue cell tumors, tumors of uncertain differentiation, fibroblastic/myofibroblastic tumors, and adipocytic tumors. In-frame fusion transcripts were identified in 64 of 140 cases sequenced (46%): in 62 cases, the detection of a disease-defining fusion confirmed the morphologic impression; in 2 cases, a germline TFG-GPR128 polymorphic fusion variant was detected. Several genes in the panel partnered with multiple fusion partners specific for different diagnoses, for example, EWSR1, NR4A3, FUS, NCOA2, and TFE3. Interesting examples are presented to highlight how fusion detection or lack thereof was instrumental in establishing accurate diagnoses. Novel fusion partners were detected for 2 cases of solid aneurysmal bone cysts (PTBP1-USP6, SLC38A2-USP6).


Multiplex detection of fusions in total nucleic acid purified from FFPE specimens facilitates diagnosis of bone and soft tissue tumors. This technology is particularly useful for morphologically challenging entities and in the absence of prior knowledge of fusion partners, and has the potential to discover novel fusion partners.

This content is only available as a PDF.

Author notes

The authors have no relevant financial interest in the products or companies described in this article.

An abstract based on this study was presented as a poster presentation at the 2019 Association for Molecular Pathology meeting; November 7–9, 2019; Baltimore, Maryland.