Multi-platform profiling of over 2000 sarcomas: Identification of biomarkers and novel therapeutic targets

Sujana Movva, Wenhsiang Wen, Wangjuh Chen, Sherri Z. Millis, Zoran Gatalica, Sandeep Reddy, Margaret von Mehren, Brian A. Van Tine

Research output: Contribution to journalArticlepeer-review

100 Scopus citations


Background: Drug development in sarcoma has been hampered by the rarity and heterogeneity of the disease and lack of predictive biomarkers to therapies. We assessed protein expression and gene alterations in a large number of bone and soft tissue sarcomas in order to categorize the molecular alterations, identify predictive biomarkers and discover new therapeutic targets. Methods: Data from sarcoma specimens profiled for protein expression, gene amplification/translocation and DNA sequencing was reviewed. Results: 2539 sarcoma specimens of 22 subtypes were included. TOPO2A was the most overexpressed protein at 52.8%. There was overexpression or loss of other sarcoma relevant proteins such as SPARC, PTEN and MGMT. Approximately 50% of the sarcomas expressed PD-L1 by IHC and presented with PD-1+ TILs, notably the LMS, chondrosarcomas, liposarcomas and UPS. Gene amplification/rearrangement of ALK, cMYC, HER2, PIK3CA, TOPO2A and cMET was relatively uncommon. EGFR gene amplification occurred at a rate of 16.9%. DNA sequencing of 47 genes identified mutations in 47% of the samples. The most commonly mutated genes were TP53 (26.3%) and BRCA2 (17.6%). Overexpression of TOPO2A was associated with TP53 mutation (P = 0.0001). Conclusion: This data provides the landscape of alterations in sarcoma. Future clinical trials are needed to validate these targets.

Original languageEnglish
Pages (from-to)12234-12247
Number of pages14
Issue number14
StatePublished - 2015


  • Biomarkers
  • DNA sequencing
  • Protein expression
  • Sarcoma
  • Targeted therapies


Dive into the research topics of 'Multi-platform profiling of over 2000 sarcomas: Identification of biomarkers and novel therapeutic targets'. Together they form a unique fingerprint.

Cite this