Molecular Staging of Human Squamous Cell Skin Cancer

Gale, James

Abstract

) Lifetime risk for developing skin cancer is currently l in 5 for the USA. While the non-melanoma skin cancers (basal and squamous cell) are generally not life threatening (2-5 percent of squamous cell carcinoma will become metastatic), they account for substantial morbidity and health-care expenditures. Our understanding of the molecular pathology and critical gene-networks involved with the development and progression of non-melanoma skin cancers is incomplete, resulting in poor markers for progression, prognosis and largely ineffective treatments for invasive stages. The broad objective of this proposal is to characterize the changes in gene expression in different stages of squamous cell carcinoma (SCC) to identify the different gene pathways involved. To characterize the most complete gene-expression profile currently possible, a set of reusable nylon microarrays containing 30,000 different human cDNA's (all of the approximately 7,000 """"""""named"""""""" genes and EST's of unknown function) will be used. SCC is well suited for an initial human cancer study because (a) there is a better consensus about the clinical stages of the disease, (b) samples of SCC are relatively easy to obtain from the clinic, and (c) SCC is most associated with sun-exposure (most frequently found on sun-exposed skin and over 90 percent of SCC tumors have p53 mutations, mostly of the type caused by UV). Of the new technologies available, cDNA microarrays have ability to screen a huge portion of all human genes (30,000 of approximate 100,000 total human genes) with the expected potential to identify """"""""all"""""""" the genes whose expression has been changed at different stages of cancer.
The specific aim of the R33 component of this proposal is to profile the gene-expression levels in the three most distinct stages of the disease (actinic keratoses; a pre-malignant lesion, SCC and metastatic SCC) from 24 subjects (12 men and 12 women, 8 subjects in each lesion group). The necessary controls of normal-uninvolved skin will be obtained by punch biopsy from both the forearm (sun-exposed skin) and the buttocks (sun-protected skin). Having two normal skin controls will add power to the data analysis and help identify genes that are naturally responsive to sunlight (may be involved in early stages of skin cancer or show natural high variability in skin which would confuse the comparison with SCC lesions). Moreover, the potential power of reusable microarrays (multiple stripping and probing) is the ability to compare results between arrays. Once the variability-profile of comparing many microarrays is calculated (hybridization of 5 microarrays with the same probe, a milestone of this proposal), we will calculate the average and standard deviation for each of the 30,000 cDNA's for the sun-exposed skin controls and sun-protected skin controls form both the 12 men and 12 women in the study.