Combined analysis of eIF4E and 4E-binding protein expression predicts breast cancer survival and estimates eIF4E activity

Abstract

Increased eukaryotic translation initiation factor 4E (eIF4E) expression occurs in many cancers, and makes fundamental contributions to carcinogenesis by stimulating the expression of cancer-related genes at post-transcriptional levels. This key role is highlighted by the facts that eIF4E levels can predict prognosis, and that eIF4E is an established therapeutic target. However, eIF4E activity is a complex function of expression levels and phosphorylation statuses of eIF4E and eIF4E-binding proteins (4E-BPs). Our hypothesis was that the combined analyses of these pathway components would allow insights into eIF4E activity and its influence on cancer. We have determined expression levels of eIF4E, 4E-BP1, 4E-BP2 and phosphorylated 4E-BP1 within 424 breast tumours, and have carried out analyses to combine these and relate the product to patient survival, in order to estimate eIF4E activity. We show that this analysis gives greater prognostic insights than that of eIF4E alone. We show that eIF4E and 4E-BP expression are positively associated, and that 4E-BP2 has a stronger influence on cancer behaviour than 4E-BP1. Finally, we examine eIF4E, estimated eIF4E activity, and phosphorylated 4E-BP1 as potential predictive biomarkers for eIF4E-targeted therapies, and show that each determines selection of different patient groups. We conclude that eIF4E's influence on cancer survival is modulated substantially by 4E-BPs, and that combined pathway analyses can estimate functional eIF4E.

Figure 1

The full range of expression intensities and proportions for eukaryotic translation initiation factor 4E (eIF4E), eIF4E-binding protein (4E-BP) 1, 4E-BP2 and p4E-BP1 occur within breast cancers. (A–H) Representative tumour tissue microarray cores showing immunoreactivity as labelled. These cores were scored A 4, B 7, C 3, D 7, E 3, F 7, G 3 and H 7. (I) Histograms showing distributions of immunohistochemistry scores within breast cancers. Scores (x-axis) and numbers of cores assigned to each score (y-axis) are shown. Percentages of the cohort are given above the bar for each score.

Figure 2

Expression of eukaryotic translation initiation factor 4E (eIF4E), but not eIF4E-binding protein (4E-BP) 1, 4E-BP2 or p4E-BP1, is associated with prognosis. (A–C) Kaplan–Meier survival analyses for overall (A), disease-free (B) and disease-specific survival (C) for patient groups with tumours with differing eIF4E scores. (D–F) Kaplan–Meier survival analyses for disease-free survival for patient groups with tumours with differing scores for 4E-BP1 (D), 4E-BP2 (E) or p4E-BP1 (F). As scores of 0, 2 and 3 were relatively rare, these have been grouped together. Censoring ticks have been omitted for clarity.

Figure 3

Additional prognostic information is gained by combining assessment of eukaryotic translation initiation factor 4E (eIF4E)-binding proteins (4E-BPs) with eIF4E analysis. Kaplan–Meier survival analyses of disease-free survival for patients with either high (6 or 7) eIF4E scores (A–C) or high (>5.4) Nottingham Prognostic Index scores (D–F). Patients were dichotomised using either eIF4E expression (A and D), the y function (which includes eIF4E and 4E-BP2 scores) (B and E), or the z function (which includes scores for all 4 markers) (C and F).

Figure 4

Use of biomarkers for dividing patients into potential treatment groups for eukaryotic translation initiation factor 4E (eIF4E)-directed therapy; different markers select very different groups. A Venn diagram demonstrating relationships between potential treatment groups selected on the basis of high eIF4E expression (1), high estimated eIF4E activity (‘z’) (2), and high p4E-BP1 expression (3).