Development of databases, summarising the genetic events associated with or
al squamous cell carcinoma (SCC), should increase our understanding of the
molecular basis of these lesions. Additionally, databases will help establi
sh whether different cancer subtypes show different growth characteristics,
because the multistage carcinogenic process is different in the various tu
mour subtypes. This new knowledge may also provide new prognostic informati
on, as these aberrations represent fundamental biological characteristics o
f each tumour. To assess the value of incorporating the results from loss o
f heterozygosity (LOH) analysis into patient-specific mutation databases, w
e have carried out microsatellite analysis with 52 polymorphic markers at 1
3 key chromosomal regions implicated in the pathogenesis of head and neck c
ancers. Altered expression of the Rb, p53 and DCC tumour suppressor genes h
as also been studied by immunohistology. Our results shed light on the diff
erent pathways that lead to cancer and reveal that a variety of different p
atterns of allelic imbalance (AI) were detected at all TNM stages, reflecti
ng the different clinical behaviour that tumours classified as being of the
same TNM stage may exhibit. Summarising the level of genetic damage as a f
ractional allelic loss (FAL) score and the presence of AI at 3p22-26, 3p14.
3-12.1 and 9p21 was found to be a better predictor of outcome than the TNM
system. This finding suggests that molecular data can be incorporated into
conventional staging systems to provide more accurate prognostic informatio
n for this group of patients. Int. Cancer (Pred. Oncol) 84:284-292, 1999. (
C) 1999 Wiley-Liss, Inc.