Cell Biology & Immunology Laboratory
Head and Neck Cancer
Markers of Progression
We previously determined that loss of chromosome 18 is a consistent genetic change in squamous carcinoma. Loss of part or all of 18q was detected in 50-60% of SCCHN. Retrospective analysis of patients whose tumors had loss of 18q showed that these individuals were more likely to have died of their tumors than those with no loss. Consistent with this observation, a recent study by Frank et al. (1997) showed that loss of heterozygosity (LOH) on 18q is a genetic change that develops with tumor progression within individual patients. A prospective study also indicates that LOH on 18q is a bad prognostic sign (Pearlstein et al., 1998).
These studies strongly suggest that 18q contains the loci of one or more tumor suppressor genes (TSG). We postulate that inactivation of these TSGs then results in more aggressive tumor growth. We recently defined three minimal lost regions(s) (MLR) on 18q (Takebayashi et al. 2000) that correspond to the polymorphic markers D18S39 at 18q21.1, D18S61 at 18q22.2, and D18S70 at 18q23. The next stages of the search for important genes in this region are now underway including assessing the regions for homozygous deletions, physical mapping of the MLRs, and identification of candidate genes.
As candidates are identified these are tested for expression in normal squamous epithelium, evidence of mutation in tumors, loss of expression in tumors and cell lines, over-expression in tumors and cell lines and evidence of related genes having a role in cancers of other types. Together with Dr. Mark Prince, we are currently assessing whether LOH in the most commonly lost region is associated with clinical outcome.
Loss of Heterozygosity and Homozygous Deletion in the Search for Candidate Genes
Generally, genes involved in cancer development and progression have been categorized as either oncogenes or tumor suppressor genes based on the accelerator and brakes model popularized by Fearon and Vogelstein. In this model, oncogenes, which are frequently over-expressed, tend to stimulate growth and are also linked to gene amplification at the DNA level.
Tumor suppressors are genes that tend to be inactivated in one copy with loss of the wildtype allele in tumor cells. Thus, loss of heterozygosity has been used as a guidepost to the location of tumor suppressor genes. In some cases, the location of TSGs has been facilitated by examples of homozygous deletions that remove both copies of the gene. Alternatively, LOH without homozygous deletion might identify areas where loss of the wild type allele and mutation of the other copy activates a gene.
We are currently assessing 18q for evidence of homozygous deletion, but results thus far show that one copy of each allele is retained across the most distal MLR. Thus, gene activation is also a possible mechanism for this region.
Candidate genes:
As candidate genes are developed from genomic searches and mapping studies these are tested for functional activity as suppressors or stimulators of tumor growth and/or invasion in in vivo and in vitro assay systems. From such studies of genes on chromosome 18 plus data from other loci, we will catalog genes that are activated or inactivated with progression. This will be used to develop a biologic staging system.
Such a system can be used at the time of diagnosis to differentiate between tumors that are relatively non-aggressive versus those that are most likely to grow and spread aggressively. This type of information will allow the physician to make better treatment decisions. Furthermore, the identification of TSG(s) or activated genes that drive growth and invasion could allow for novel forms of therapy based on replacement of the missing function or blockade of the gained function.
We are currently investigating GALR1, a gene located at 18q23 within the most common region of loss. GALR1 is a G-protein coupled receptor (GPCR). GPCRs are a huge family of proteins that have seven transmembrane domains, are expressed on the cell surface and serve as receptors for a multitude of peptide hormones. GALR1 is one of three known receptors for galanin - a 31 amino acid neuropeptide.
We have observed abnormalities in GALR1 and GALR2 expression in HNSCC cell lines and we have demonstrated that galanin is expressed and secreted by most of the tumor cell lines of this type. Preliminary evidence implicated galanin and its receptors in growth regulation and cell survival and we have also found evidence of occasional mutation and methylation of the GALR1 gene in HNSCC tumor lines. This is consistent with the tumor suppressor gene hypothesis, however we also have evidence of overexpression of this gene in some tumors indicating that the role of galanin and its receptors in HNSCC may be complex. These studies are carried out in collaboration with Dr. Nisha D'Silva.
Predicting Response to Therapy
This study is carried out in collaboration with Dr. Carol Bradford and other members of the Head and Neck Oncology program. The goals are to test the hypothesis that chemotherapy response, organ preservation and survival are predicted by p53 expression and gene mutation and modulated by expression of the Bcl-2 family of anti-apoptotic proteins. We are testing p53 mutation status and expression of Bcl-2 family members in tumor populations of responding primary tumors, non-responding regional metastases as well as in pretreatment and salvage surgery specimens to determine if these genetic alterations can account for response and non-response patterns.
In addition we are using our large and well-characterized panel of cell lines from head and neck tumors to test our hypotheses in cells in vitro and in vivo. Our recent experiments have indicated that resistance to platinum based chemotherapy is associated with wild type p53 and overexpression of Bcl-xL. New agents designed to specifically inhibit Bcl-xL are effective in vitro at overcoming this resistance and show promise for clinical application.
This work is a collaboration with Dr. Shaomeng Wang and Dr. D'Silva. This project also has a collaboration with Dr. Reidar Grenman in Turku, Finland. Together with these collaborators we will test the hypothesis that p53 antibody is a serum marker for persistence, recurrence, and successful therapy.
