Poster Board I-948
Eukaryotic initiation factor 4E binding protein 1 (4E-BP1) regulates the function of eukaryotic initiation factor 4E (eIF-4E) which mediates nucleo-cytoplasmic transport and cap-dependent translation of specific transcripts. eIF-4E expression itself is elevated in several carcinomas and hematopoietic malignancies. Hypophosphorylated 4E-BP1 binds eIF-4E and inhibits cap-dependent translation. Phosphorylated 4E-BP1 dissociates from eIF-4E, thereby allowing translation of several cap-dependent transcripts responsible for cell cycling, survival and angiogenesis. Hyper-phosphorylation of 4E-BP1 may thus be associated with oncogenesis and the malignant phenotype; its influence on the response to chemotherapy and prognosis of diffuse large B cell lymphoma (DLBCL) is unknown. We examined diagnostic tissues from 85 patients with DLBCL for 4E-BP1 expression and phosphorylation, and correlated these with clinical outcomes. Samples were stained for total and phosphorylated 4E-BPI using immunohistochemistry (IHC). Slides were scored for percent cells positive for the two markers (0% to 100%) and the intensity of expression (0 to 3+) by hematopathologists (A.R. and H.M.) blinded to the clinical data. An Expression Score was calculated from their product ([% cells positive × intensity]/10), thus ranging from 0-30. In parallel, expression of bcl-2 was determined by IHC on the same samples. Patients had a median age of 75 years (range, 25-91). The majority received CHOP-based chemotherapy with/without rituximab. Total 4E-BP1 was expressed uniformly (by 97% of the cases). However, there was marked heterogeneity in the level of phosphorylation of 4E-BP1 (none [Expression Score = 0] in 35%, uniformly high [Expression Score = 30] in 34%, and variable [Expression Score = 1-29] in the remainder [31%]). We therefore examined if the level of phosphorylation of 4E-BP1 varied with clinical features at diagnosis, and whether it correlated with response to chemotherapy and survival. The level of expression or phosphorylation of 4E-BP1 did not correlate with the clinical stage, IPI score or bcl-2 expression at diagnosis, and did not influence the likelihood of achieving complete remission with CHOP-based chemotherapy. However, in patients with low IPI scores (IPI 0-2), the level of phosphorylation of 4E-BP1 correlated significantly with overall survival (OS). In these patients, the OS was significantly different (P = 0.031) in patients with low levels of phosphorylation of 4E-BP1 (Expression Score = 0-15) compared to those with higher levels of 4E-BP1 phosphorylation (Expression Score = 16-30). The median survival was >146.4 months vs 20.6 months, and the 5-year survival was 73% vs 31%, in patients with low and high levels of 4E-BP1 phosphorylation, respectively. The difference in survival between the two groups was not a consequence of differences in age or IPI score at diagnosis or inclusion of rituximab in the treatment regimen. The level of phosphorylation of 4E-BP1 did not have any further impact on the survival of patients with high IPI scores (IPI 4-5). We conclude that while the large majority of DLBCL express the translation regulatory protein 4E-BP1, the level of phosphorylation (indicative of biological activity) of this protein varies widely. Importantly, the level of phosphorylation of 4E-BP1 further stratifies DLBCL patients who would be expected to have a favorable prognosis based on low IPI scores, and identifies a subset of these patients who do poorly following standard CHOP-based chemotherapy. Since hyper-phosphorylation of 4E-BP1 increases the activity of eIF-4E, this study provides the rationale for trials incorporating novel targeted agents that inhibit eIF-4E into the therapeutic regimen, specifically in DLBCL patients with low IPI scores and high levels of 4E-BP1 phosphorylation.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.