Back to News

News at CRCH

Cancer Scientist Awarded Patent for a Method to Aid in the Identification of Cancer Genes
Computational Method May Predict Cancer Recurrence and Level of Aggressiveness
posted February 4, 2009

HONOLULU—Cancer Research Center of Hawai‘i (CRCH) bioinformatics researcher Gordon Okimoto, Ph.D., assistant professor in the Epidemiology Program, and the University of Hawai‘i was granted a patent for an innovative computational method for identifying genes involved in cancer. The method, called “Microarray ANalysis of InteNsities and RatIos” or “MANINI,” has been used in a number of studies at CRCH to compare the activation of every known gene in normal and cancerous tissue samples, and to identify the genes that are different in cancer. The MANINI algorithm must still undergo testing to validate its effectiveness in the patient setting but appears to have a promising future in the management of cancer patient care.

States Okimoto, “Based on studies using real human tumors, the MANINI algorithm has demonstrated tremendous potential in identifying a relatively small number of genes out of tens of thousands that are able to predict the recurrence and level of aggressiveness of specific cancers. This information will aid physicians plan more precise treatments tailored to individual patients, without subjecting them to unnecessary surgical trauma, radiation, or administration of excessive chemotherapeutic or biological agents.”

In one study, the activation patterns of 35 genes in breast cancer tumors were used to accurately determine, without invasive biopsy, whether the tumor had spread to a patient’s lymph nodes. The 35 genes were discovered based on a MANINI analysis of “genomic” data that measured the activation levels of over 12,000 genes in each of 60 breast cancer tumor samples. Different genes are activated in tumors that have spread to the lymph nodes versus tumors that have not, for reasons that are not clear at the moment. MANINI helps to identify these genes as a first step towards understanding this phenomenon, and most importantly, to predict which breast cancer tumors could or have already spread to the lymph nodes.

In another study, MANINI was able to identify 10 genes out of 32,000 that were predictive of early-onset cardiovascular disease after standard statistical methods failed to find any genes at all. Studies are planned or ongoing in which MANINI will be used to identify unique “genetic signatures” for different ethnic groups that are predictive of breast cancer recurrence. Such ethnic differences may help scientists better understand the genetic causes of breast cancer.

The MANINI method is unconventional as it focuses on genes with weak and erratic signals for activation that are often ignored by standard statistical methods. Okimoto has shown in a number of studies that such genes, when viewed as a group, are often strongly associated with cancer. On the other hand, genes with stronger and more consistent signals are often missing from the data, or are biologically uninteresting as a group. For many studies, Okimoto believes that MANINI will complement rather than replace standard methods for identifying genes that are useful for diagnosis and prediction of the likelihood of recurrence or of how well a patient’s cancer will respond to treatment. “Large genomic data sets contain many different kinds of signals that are important in cancer,” states Okimoto, “and it is likely that we will need more than one method to find all of them.”

Meanwhile, Okimoto’s expertise in bioinformatics, which broadly defined means the global analysis of biological systems using mathematical methods, is very relevant to the emerging science of cancer translational research. “With the flood of new genomic data, cancer research is fast becoming an information science,” he states, “where mathematics is the language that best describes the breakdown of communication that occurs between the molecular components of a cancerous cell.” Okimoto believes that cancer bioinformatics and systems biology will greatly accelerate the translation of basic results in cancer research into clinical applications, especially in the areas of early diagnosis, prediction of clinical outcomes, and how an individual may respond to different therapeutic strategies.

Okimoto’s specific areas of interest in cancer research include early diagnosis, and the prediction of cancer recurrence and response to therapy using gene activation patterns. He is also active in the field of cancer systems biology, which involves the computational modeling of cancer as a complex system based on genomic, clinical and environmental data. He has previously worked at the Cardiovascular Research Center (CVRC) in the John A. Burns School of Medicine (JABSOM), and obtained his Ph.D. under the guidance of Charles Boyd, Ph.D., former director of the CVRC, who is listed as a co-inventor on the MANINI patent along with Johann Urschitz, Ph.D., who is a biomedical researcher at the Institute of Biogenesis in JABSOM.

 

Back to News