Association of EWS-FLI1 type 1 fusion with lower proliferative rate in Ewing's sarcoma

Abstract

The Ewing's sarcoma (ES) family of tumors, including peripheral neuroectodermal tumor (PNET), is defined genetically by specific chromosomal translocations resulting in fusion of the EWS gene with a member of the ETS family of transcription factors, either FLI1 (90-95%) or ERG (5-10%). A second level of molecular genetic heterogeneity stems from the variation in the location of the translocation breakpoints, resulting in the inclusion of different combinations of exons from EWS and FLI1 (or ERG) in the fusion products. The most common type of EWS-FLI1 fusion transcript, type 1, is associated with a favorable prognosis and appears to encode a functionally weaker transactivator, compared to other fusion types. We sought to determine whether the observed covariation of structure, function, and clinical course correlates with tumor cell kinetic parameters such as proliferative rate and apoptosis, and with expression of the receptor for insulin-like growth factor I (IGF-1R). In a group of 86 ES/PNET with defined EWS-ETS fusions (45 EWS-FLI1 type 1, 27 EWS-FLI1 non-type 1, 14 EWS-ERG), we assessed proliferation rate by immunostaining for Ki-67 using MIB1 antibody (n = 85), apoptosis by TUNEL assay (n = 66), and IGF-1R expression by immunostaining with antibody 1H7 (n = 78). Ki-67 proliferative index was lower in tumors with EWS-FLI1 type 1 than those with non-type 1 EWS-FLI1, whether analyzed as a continuous (P = 0.049) or categorical (P = 0.047) variable. Logistic regression analysis suggests that this association was secondary to the association of type 1 EWS-FLI1 and lower IGF-1R expression (P = 0.04). Comparing EWS-FLI1 to EWS-ERG cases, Ki-67 proliferative index was higher in the latter (P = 0.01, Mann-Whitney test; P = 0.02, Fisher's exact test), but there was no significant difference in IGF-1R. TUNEL results showed no significant differences between groups. Our results suggest that clinical and functional differences between alternative forms of EWS-FLI1 are paralleled by differences in proliferative rate, possibly mediated by differential regulation of the IGF-1R pathway.

Figure 1.

Ki-67 proliferation rate assesed by MIB1 immunohistochemistry in ES/PNET. Left panel shows ES/PNET case with non-type 1 EWS-FLI1 and high Ki-67. Right panel shows ES/PNET case with type 1 EWS-FLI1 and low Ki-67.

Figure 2.

IGF-1R immunohistochemistry in ES/PNET. ES/PNET case with high IGF-1R expression is shown.

Figure 3.

Scattergram showing correlation between Ki-67 (MIB1) and IGF-1R immunoreactivities. Significant Ki-67 immunoreactivity (>10%) was only present when over 50% of cells show expression of IGF-1R. Tumors having non-type 1 EWS-FLI1 fusions (green triangles) had significantly higher values of IGF-1R and Ki-67 (see Tables 1 and 2 ▶ ▶ ).