Zoltan Szallasi, MD

2009-2010 BCRF Project:
(made possible by BlackBerry®)

Dr. Szallasi's group is developing computational methods to analyze the various genome scale molecular profiles of breast cancer samples such as gene expression microarrays, simultaneously measuring all gene expression levels in a cancer sample, or CGH arrays, simultaneously detecting all chromosomal aberrations in the same cancer sample. While solving the theoretical problem of how to identify robust consistent information in multiple high genome scale measurements, the researchers identified reproducible predictors of clinical chemotherapy response in breast tumors that are negative for the estrogen, progesterone and HER2 receptors.

Genome scale analysis of cancer helps researchers to expand their research focus from a few genes to a more complete view of the entire cancerous genetic network. The expression levels or DNA copy number of virtually all genes in a cell can be quantified simultaneously. This more complete analysis of human cancer cells will likely hold the key to solving the difficulties associated with the fact that breast cancer cells comprise complex, robust and evolving systems.

The Szallasi group is focusing on several aspects of genome scale analysis of human breast cancer. First, they are developing bioinformatics methods that increase the accuracy of high throughput measurements. Second, from the data sets of increased accuracy they aim to extract quantitative measures of key biological processes in cancer. They are particularly interested in the various subtypes of genomic instability, their relative level in a given breast tumor and whether this could guide more effective therapeutic decisions. Third, they are combining genome scale molecular profiling of chemotherapy resistant breast cancer cell lines with bioinformatics analysis in order to determine whether clinical response to chemotherapy in breast cancer can be predicted by gene expression signatures derived from cell lines. The most immediate outcome of this research is to develop tools that would predict with high accuracy which breast cancer patient will respond to a given chemotherapeutic agent.

Mid-Year Progress Report:
Dr. Szallasi's group has produced several lines of evidence that genome scale molecular profiling of breast cancer is an effective method to identify the specific genomic instability subtype of a given breast tumor and this information is one of the key determinants whether a patient will respond to a given chemotherapeutic agent. His current aim is to develop genome scale molecular profiling based clinical tests that predict response to therapy with high accuracy.

Bio:
Dr Szallasi received his Doctor of Medicine (MD) degree from the University of Medicine in Debrecen, Hungary in 1988. He did his postdoctoral research in the field of molecular pharmacology of cancer at the National Cancer Institute. As a faculty member, first at the Uniformed Services University of Health Sciences and currently at the Children's Hospital, Boston at Harvard Medical School, he has become active in the high throughput analysis of breast cancer. He has published over 60 peer reviewed articles, mainly on the molecular pharmacology and high throughput analysis of cancer.

Genome scale molecular analysis, such as microarray based gene expression profiling, offers a more complete view of the biochemical changes associated with cancer. However, more data means more noise, more uncertainty and an explosion of the hypothesis space, all impeding association based learning often applied both in basic and clinical cancer research. Dr. Szallasi's group is interested in the meaningful and responsible application of high throughput measurements for cancer research. They implemented several methods that increased the reliability of microarray measurements. They are also interested in approaches that combine high throughput measurements in a manner that describe essential biology in a robust fashion, such as developing a gene expression signature of chromosomal instability.

His earlier projects have led Dr. Szallasi to the current main focus of his research: How is the robust phenotype of a given cancer type coded in gene expression networks? This problem could (and perhaps should) be approached both from a computational and an experimental direction. The success of genome scale analysis of breast cancer may in fact depend on the effective combination of developing experimental models yielding robust information about human tumors and their statistically sound exploitation. Dr. Szallasi's group is working on such "dual" approaches to answer whether one can identify which patient will respond to a given chemotherapeutic agent or whether there exist different subtypes of genomic instability in breast cancer with prognostic and therapeutic relevance.