Innovation and Partnerships Office

Genetic Profile Determines Prognosis and Response to Cancer Therapy WIB-2709


  • Diagnostic testing
  • Drug design, laboratory testing, and clinical trials
  • Clinical management of cancer and other diseases


  • Enables more efficient clinical trials
  • Identifies drug targets
  • Applicable to multiple tumor types
  • Identifies normal tissues that may be adversely effected by drugs


Joe Gray and colleagues at Berkeley Lab have identified a network of functionally related genes that controls mitosis in cancer. The researchers used this network to develop a measurable mitotic network activity index (MNAI) that can identify poor outcome tumors and predict tumor response to specific drugs. The MNAI can also be applied to biopsies to diagnose tumors and determine how far advanced they are; identify appropriate patients for clinical trials; and choose targets for new drug development.

The researchers initially identified 272 genes that, in breast cancer cell lines, are strongly co-expressed with the mitosis-related genes PLK1, CENPE,and AURB—targets of recently developed anti-cancer drugs. They narrowed this group of 272 to the 54 genes most closely linked to the target genes to constitute the network and the MNAI. When applied to multiple cell lines, the MNAI predicted in vitro drug sensitivity more effectively than did the traditional classification between luminal and basal cell cancers. The scientists then applied the MNAI to published datasets that included survival information and tumor gene profiles from more than 500 patients with breast cancer. The index greatly simplified the analysis by focusing on the 54 network genes among the thousands included in each tumor profile and correlated significantly with survival. Additional analysis identified 22 of the 54 genes whose inhibition resulted in a marked suppression of growth. Several of these genes appear modifiable with drugs making them appealing therapeutic targets.

The network genes were also highly expressed in other types of cancers, including several solid and hematopoietic tumors, indicating that the MNAI may be useful in these diseases. Even more broadly, the process of determining the network could be applied to any disease in which the genes for an abnormal physiologic pathway have been identified, leading to more effective drug design and clinical decisions.

In cancer research, identifying drug targets and selecting patients for trials remain significant challenges. In the clinical setting, the challenge is to choose the drug likely to have the greatest efficacy and least toxicity in a particular patient with a particular type of tumor. The Berkeley Lab technology has the potential to greatly improve these processes.

STATUS: Published patent application US2010/034274 available at Available for licensing or collaborative research.

DEVELOPMENT STAGE: Proof of principle.


ANXA9: A Therapeutic Target and Predictive Marker for Early Detection of Aggressive Breast Cancer, JIB-2371

Identifying Biomarkers of Physiological Responses of Cancer Cells, IB-2546

High Selectivity Peptides for Cancer Therapy, Screening for Cancer Therapies and Therapy Sensitization, IB-2489