5077

Introduction: Eukaryotic initiation factor 4E (eIF4E) is overexpressed in breast cancer cells, but not in benign breast tissue. eIF4E is a 25 kDa protein that plays a pivotal role in translational regulation of genes containing 5-untranslated regions (5’-UTRs). This is accomplished by eIF4E binding to the 5’-UTR, facilitating ribosomal binding and therefore protein synthesis. Some gene products upregulated by eIF4E overexpression are known oncogenes. The overexpression of eIF4E has been shown to predict cancer recurrence and death in patients with breast cancer. It has been assumed that the elevation of eIF4E in cancer specimen was due to its overexpression in tumor cells. An alternate hypothesis is that highly proliferative cancers may have higher tumor cell density. Therefore a higher elevation of eIF4E, as detected by Western blots, is a marker of tumor cell density. Thus, in this study, we examined whether eIF4E elevation in breast cancer specimens is correlated with tumor cell density.

Methods: 101 patients were accrued from our prospective breast cancer database. eIF4E levels were quantified using a standardized 10µg of protein extracts from cancer specimens by western blot analysis, and expressed as x-fold over normal breast tissue from non-cancer patients. Tumor specimens were marked and confirmed by the study pathologist, and the mean tumor cell density was measured manually by counting tumor cells per 40X microscopic high power field at differing locations on three separate representative H & E slides. Statistical tests used include Spearman rank, Chi-square, Kaplan-Meier survival analysis, and log-rank test.

Results: All 101 breast cancer specimens had eIF4E overexpression, with a mean elevation of 12.1 ± 6.8-fold (mean ± standard deviation). The mean tumor cell density of cancer specimens was 429 ± 192 cells per high power field. Tumor cell density was not statistically correlated with the degree of eIF4E overexpression (rho=0.12, p=0.26, Spearman Rank). Tumor cell density and eIF4E overexpression did not correlate with tumor size, tumor grade, nodal status, hormone receptor, or HER2 status (chi-square analysis). Consistent with previous studies, high eIF4E overexpression correlated with cancer recurrence and cancer-related death (p=0.01 and p=0.05, log-rank test, respectively). In contrast, tumor cell density did not correlate with either cancer recurrence or cancer-related death (p=0.65 and p=0.59, respectively).

Conclusion: Tumor cell density and eIF4E overexpression varies widely amongst cancer specimens. The degree of eIF4E overexpression did not correlate with tumor cell density. High eIF4E elevation predicted a worse disease-free and overall survival; tumor cell density did not. Therefore, the degree of eIF4E overexpression in breast cancer appears to be a biological consequence of malignant transformation and not of tumor cell density.

98th AACR Annual Meeting-- Apr 14-18, 2007; Los Angeles, CA