Prognostic impact of P53 status in Ewing sarcoma

Cancer. 2000 Aug 15;89(4):783-92.


Background: Disease stage at the time of diagnosis and response to therapy are the main prognostic factors for patients with Ewing sarcoma or peripheral neuroectodermal tumor (ES/PNET). The primary genetic alteration in ES/PNET, the fusion of the EWS gene with FLI1 or ERG, is diagnostically highly specific for these tumors, and molecular variation in the structure of the EWS-FLI1 fusion gene also is of prognostic significance. In contrast, secondary genetic alterations, such as P53 alterations, are relatively uncommon in ES/PNET, and their prognostic impact has not been extensively studied.

Methods: Prechemotherapy, paraffin embedded, nondecalcified, primary tumor material in a well-characterized series of 55 patients with ES/PNET with defined EWS-FLI1 fusion transcripts (32 patients with type 1 and 23 patients with other types) was studied retrospectively by immunohistochemical techniques for cell cycle regulators and proliferative markers, such as P53, P21(WAF1), and Ki-67, as well as by the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) technique for apoptosis. Nuclear P53 expression in > 20% of tumor cells was scored as aberrant overexpression. Histologic response to neoadjuvant chemotherapy was assessed.

Results: Aberrant P53 expression (in > 20% of tumor cells) was present in 6 patients (11%) but showed no statistically significant correlation with disease stage, tumor size, proliferation rate (Ki-67), apoptotic rate (TUNEL), or EWS-FLI1 fusion type. By univariate analysis, the P53 > 20% group showed a significantly poorer overall survival among patients with localized disease (n = 43 patients) (P = 0.001) and in the entire study group (P = 0.01). In multivariate Cox analyses of overall survival, P53 > 20% was the strongest negative factor among prognostic factors available at the time of diagnosis (P = 0.001; relative risk [RR] = 9) and when chemotherapy response was included in the analysis (P53 > 20%: P = 0.01; RR = 10).

Conclusions: P53 alteration appears to define a small clinical subset of patients with ES/PNET with a markedly poor outcome. The current observations warrant a systematic prospective study with comprehensive P53 mutation analysis. [See related article on pages 793-9, this issue.]

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Antineoplastic Agents / therapeutic use
  • Biomarkers, Tumor / metabolism*
  • Bone Neoplasms / diagnosis
  • Bone Neoplasms / drug therapy
  • Bone Neoplasms / metabolism*
  • Bone Neoplasms / mortality
  • Child
  • Child, Preschool
  • Female
  • Humans
  • Immunohistochemistry
  • Male
  • Neoplasm Staging
  • Oncogene Proteins, Fusion / genetics
  • Oncogene Proteins, Fusion / metabolism
  • Point Mutation
  • Prognosis
  • Prospective Studies
  • Proto-Oncogene Protein c-fli-1
  • RNA-Binding Protein EWS
  • Sarcoma, Ewing / diagnosis
  • Sarcoma, Ewing / drug therapy
  • Sarcoma, Ewing / metabolism*
  • Sarcoma, Ewing / mortality
  • Transcription Factors / genetics
  • Transcription Factors / metabolism
  • Treatment Outcome
  • Tumor Suppressor Protein p53 / genetics
  • Tumor Suppressor Protein p53 / metabolism*


  • Antineoplastic Agents
  • Biomarkers, Tumor
  • EWS-FLI fusion protein
  • Oncogene Proteins, Fusion
  • Proto-Oncogene Protein c-fli-1
  • RNA-Binding Protein EWS
  • Transcription Factors
  • Tumor Suppressor Protein p53