Tumor Marker E2F4 Linked to Breast Cancer Survival

Dr. Chao Cheng PhD Department of Genetics Geisel School of Medicine at Dartmouth Hanover 03755, NH MedicalResearch.com Interview with:
Dr. Chao Cheng PhD
Department of Genetics
Geisel School of Medicine at Dartmouth
Hanover 03755, NH


MedicalResearch: What is the background for this study? What are the main findings?

Dr. Chao Cheng: Cancer survival prognosis—“How long do I have, Dr.?” is a topic of great importance to cancer patients and their families. While clinical and pathological variables, such as cancer type, stage, grade, and patient demographics, have long been used to predict survival outcomes, only recently have molecular signatures become incorporated into survival prediction. A molecular approach holds great promise for improving prediction accuracy and additionally elucidating mechanisms of disease, however it is fraught with difficulty due to assay “noise” and “big data” statistical issues, such as the multiple comparisons problem

In this study, we began by analyzing transcription factor binding profiles across available cell lines. By restricting our analysis to transcription factors, DNA expression regulators known to be involved in tumor genesis, we reasoned that we could avoid many of the “big data” issues and achieve results that would make mechanistic and biological sense. We first employed a statistical method we described previously to calculate which genes were the major downstream targets of our transcription factors. With these targets identified, we then analyzed gene expression data using a bioinformatics method to infer the relative activity of each transcription factor based upon the overall expression levels of their gene targets. From here, we incorporated cancer survival data and examined how each transcription factor’s regulatory activity did, or did not, correlate with survival.

The most prognostic transcription factor was E2F4, a member of the E2F family and a known regulator of the cell cycle. We therefore restricted our analysis to E2F4 and examined how its activity level impacted survival in breast cancer patients.

We found that tumors with high E2F4 regulatory activity as compared to low E2F4 regulatory activity had much worse survival outcomes. These results were stable even after controlling for tumor stage, grade, patient age, and treatment, and were based on data from over 1900 patients across eight independent datasets. These results demonstrate that E2F4 is an independent and enhancing predictor of survival above the currently examined variables.

MedicalResearch: What should clinicians and patients take away from your report?

Dr. Chao Cheng: E2F4 regulatory activity is a powerful prognostic factor in breast cancer. Our next step will be a genomic assay developed to quantify the expression of E2F4 signature genes in patient samples collected at the time of diagnosis or surgery. Based on the resulting E2F4 scores from the assay, patients will be stratified into different groups to guide physicians in choosing the most appropriate therapeutic strategy.

Oncotype DX assay has been widely used to predict the risk of recurrence in estrogen receptor positive lymph node negative breast cancer. Based on the assay, patients can be stratified into high-, intermediate- and low-risk groups. Until now, there has been no standard of care for those with intermediate risk. Our results suggest that 20-30% of those intermediate patients are actually in high-risk for recurrence, indicating they should receive aggressive follow-up treatment.

More broadly, the study reveals the power of using the activity of regulators, like transcription factors, from the expression of their downstream targets.

MedicalResearch: What recommendations do you have for future research as a result of this study?

Dr. Chao Cheng: E2F4’s regulatory activity should be examined across more cancer types and patients to see if it is further generalizable. Refinement of the core genes in the regulatory activity signature should also be pursued.


E2F4 regulatory program predicts patient survival prognosis in breast cancer
Khaleel SS, Andrews EH, Ung M, Direnzo J, Cheng C.

Breast Cancer Res. 2014 Dec 2;16(6):486.