Khandan Keyomarsi, Ph.D.

Hubert L. and Olive Stringer Professor in Medical Oncology
Department of Experimental Radiation Oncology
University of Texas MD Anderson Cancer Center
1515 Holcombe Blvd, Unit 66, Y3.5314a
Houston, Texas 77030
713-794-5369 (FAX)

My laboratory is currently involved in 5 areas of research, for identifying novel therapeutic strategies and prognostic markers for breast cancer based on the alteration in the G1/S checkpoint in tumor cells. Below I have summarized 2 of the translational projects in more detail.

  1. Investigation of the low-molecular-weight (LMW) forms of cyclin E as prognostic markers in breast cancer for routine use in the clinic.
  2. Delineation of how the alteration of cyclin E, a G1 cyclin, could lead to the tumorigenic phenotype and determination of the oncogenic potential of the altered forms of cyclin E in breast cancer. 
  3. Inhibition of LMW forms of cyclin E as a therapeutic target.
  4. Examination of the in vivo and subsequent clinical feasibility of a novel treatment strategy designed to protect normal proliferating cells against the toxic effects of chemotherapeutic agents.
  5. Determination of whether the proteasome is the target of G1 arrest/growth inhibition mediated by farnasyl-transferase inhibitors and structurally related compounds.

Investigation of the LMW forms of cyclin E as potential prognostic markers for routine use in the clinic. My laboratory has made the novel observation that cyclin E is present and overexpressed as LMW isoforms in breast cancer cells and tumor tissues. In a retrospective analysis of tumor specimens from 395 breast cancer patients, we established the overexpression of the LMW forms of cyclin E to be a strong predictor of poor survival (Keyomarsi, et al. New England Journal of Medicine, 2002).  The hazard ratio for breast cancer death for patients with overexpression of cyclin E was 10-fold higher than the hazard ratio we obtained using patients’ lymph node status. Based on preliminary findings in the laboratory, we are also evaluating the use of LMW cyclin E as a predictor of poor response to hormonal therapy and chemotherapy. Our long-term goal is that once patients with deregulated cyclin E expression are identified, they may then be singled out for tumor directed therapy with cyclin E as the target.  In addition, we will work toward the routine use of cyclin E as a biomarker for breast cancer patients by development of assays, which can be readily performed in tumor pathology laboratories, nationwide. 

Inhibition of LMW forms of cyclin E as a therapeutic target. We have 2 different approaches to target the LMW forms of cyclin E: a pharmacologic one and a molecular one. 

The pharmacologic approach involves the identification of kinase inhibitors specific to LMW cyclin E/cyclin-dependent kinase 2 (CDK2) complexes by screening a library of 12,000 compounds using an in vitro kinase assay developed by the laboratory of Dr Laurent Meijer in Roscoff, France. After this screening, we will identify the candidate compounds that selectively inhibit the proliferation of LMW cyclin E–over expressing cell lines.  These steps will be followed by preclinical testing of the candidate CDK2 inhibitors using xenografts in athymic mice. Lastly, we will initiate clinical trials with the most promising of the  cyclin E/CDK2 inhibitor in combination with chemotherapy in metastatic breast cancer.

The molecular approach targets the mechanism responsible for generating the LMW forms of cyclin E. This approach involves inhibiting the generation of the LMW forms of cyclin E through upregulation of elafin. We have shown that elafin can eradicate xenograft tumor growth in vivo.  Our long-term goal is to use the LMW cyclin E/CDK2 specific inhibitors and/or elafin to inhibit control progression of the cell cycle in cancer cells, thus limiting the ability of these cells to populate distant metastatic sites. The information gained through these studies may have tremendous clinical relevance for women with early stage and advanced breast cancer.  We already know that cyclin E overexpression correlates with poor outcome and that no alternative treatment strategy can currently be offered to these patients with altered expression of cyclin E.  We believe there is a need to develop novel treatment strategies that could target the altered forms of cyclin E. Consequently, our hypothesis is that the tumor-specific, hyperactive, LMW forms of cyclin E can be targeted differentially from the full-length cyclin E by novel CDK2 inhibitors and elafin. These inhibitors would then be used as targeted therapy for those breast cancer patients whose tumors overexpress the LMW forms of cyclin E to potentially improve long term patient outcome in this poor prognostic patient cohort.