Scientific Reports (Mar 2024)

Identification of an early survival prognostic gene signature for localized osteosarcoma patients

  • Tajhal D. Patel,
  • Sandra L. Grimm,
  • Rupa S. Kanchi,
  • Tanmay Gandhi,
  • Amrit Koirala,
  • Jason T. Yustein,
  • Cristian Coarfa

DOI
https://doi.org/10.1038/s41598-024-57527-8
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Osteosarcoma is the most prevalent bone tumor in pediatric patients. Neoadjuvant chemotherapy has improved osteosarcoma patient survival, however the 5-year survival rate for localized osteosarcoma is 75% with a 30–50% recurrence rate. We, therefore, sought to identify a prognostic gene signature which could predict poor prognosis in localized osteosarcoma patients. Using the TARGET osteosarcoma transcriptomic dataset, we identified a 13-hub gene signature associated with overall survival and time to death of localized osteosarcoma patients, with the high-risk group showing a 22% and the low-risk group showing 100% overall survival. Furthermore, network analysis identified five modules of co-expressed genes that significantly correlated with survival, and identified 65 pathways enriched across 3 modules, including Hedgehog signaling, which includes 2 of the 13 genes, IHH and GLI1. Subsequently, we demonstrated that GLI antagonists inhibited growth of a recurrent localized PDX-derived cell line with elevated IHH and GLI1 expression, but not a non-relapsed cell line with low pathway activation. Finally, we show that our signature outperforms previously reported signatures in predicting poor prognosis and death within 3 years in patients with localized osteosarcoma.