Increased activity of nm23-H1 gene in squamous cell carcinoma of the head and neck is associated with advanced disease and poor prognosis.

The present study was undertaken to determine whether the nm23-H1 gene is expressed in squamous cell carcinoma of the head and neck (SCCHN) and whether the level of nm23-H1 protein or mRNA in cells vary as they progress to a more malignant phenotype. Of the 120 SCCHN studied 54 (45%) stained positively for nm23-H1 protein. Protein expression was significantly higher in more advanced stages of disease. Expression of nm23-H1 was significantly higher in cancer tissues than in normal, adjacent tissue, dysplasia, or carcinoma in situ. The nm23-H1 rate increased with progression of synchronous lesions from dysplasia to carcinoma in situ and finally to carcinoma (P<0.05). Northern blot analyses of tissues with various clinicopathological characteristics also revealed differences in nm23-H1 mRNA expression. When levels of nm23-H1 mRNA were compared to tumor stage, intensity of expression was found to be higher in stages 3 and 4 than stages 1 and 2 (P<0.01). Malignant tumors had a higher level of mRNA nm23-H1 expression than normal or premalignant tissues. The nm23-H1 negative patients survived significantly longer than nm23-H1 positive ones (P<0.05). To study the possible relationship between nm23-H1 gene expression and cell growth rate in tumor cells, the mRNA level in each tumor was compared to proliferative activity. The nm23-H1 gene expression levels were directly related to the [3H]thymidine labeling index in tumor cells (R=0.6681). Our results strongly indicate that the nm23-H1 gene is involved in progression of SCCHN. Together with results obtained on lung cancer, our observations suggest that increased expression of nm23-H1 in cancers of the upper aerodigestive tract may have different implications than elsewhere in the body.

Molecular pathology of hemangiopericytomas accompanied by severe hypoglycemia: oncogenes, tumor-suppressor genes and the insulin-like growth factor family.

Relatively little is known about molecular genetic events that participate in the genesis and progression of hemangiopericytoma. In this study, we describe two cases of hemangiopericytoma accompanied by severe hypoglycemia. Tumor cells from patient 1 exhibited insulin-growth factor I (IGF I) and insulin-like growth factor I receptor (IGF IR) mRNA transcripts. Tumor cells from patient 2 exhibited IGF II, IGF IR and IGF binding proteins 1-3 mRNA. Serum from patient 2 contained IGF II, mostly in a large molecular form ("big" IGF II); the IGF II level did not change after the tumor removal. The presence of IGF IR in tumor cells was confirmed by immunoprecipitation with antibodies that recognize human IGF IR subunit (visualized as a 460-kDa band). The hemangiopericytoma cells derived from patient 1 expressed 210000 IGF I receptors/cell. Specific binding of IGF I to the tumor cell membrane fraction was higher in tissue from patient 1, while the tissue of patient 2 showed relatively low IGF I binding. In contrast, IGF II binding was much higher in tissue from patient 2. Both tumor tissues showed positive immunostaining for c-Jun; one tumor showed strong immunostaining for c-Myc, H-Ras and p53, while the other exhibited strong reaction with H-Ras antibodies only. No loss of the heterozygosity at the genes APC, NFI and nm23-H1 loci in tumor tissue obtained from patient 1 was found. In effect, our results suggest multiple molecular genetic changes in hemangiopericytoma -- activation of some oncogenes and the IGF growth factor family. IGF ligands together with IGF IR could be responsible for hypoglycemia and perhaps the transformed phenotype.

Prostaglandin H synthase isoenzyme distribution in the gingival tissue of patients with periodontitis: pronounced expression adjacent to gram-positive bacteria.

Prostaglandin (PGE(2)) is an inflammatory mediator that plays a critical role in the pathogenesis of periodontal disease. Prostaglandin H synthase (PGHS) a rate-limiting enzyme in PGE(2) biosynthesis exists as two separate isoforms (PGHS-1 and PGHS-2). We have previously demonstrated that both isoforms are generally present in the gingival tissue of periodontitis patients. This study explores in greater detail the variable distribution of each isoenzyme in both inflamed and non-inflamed gingival tissues of patients with periodontitis, and the relationship to adjacent bacteria. Although the positive staining for PGHS-1 was never as intense as for PGHS-2 in the same tissue specimen, either in inflamed or non-inflamed tissues, there was strong staining for both isoenzymes in the epithelium. The keratin layer did not stain. Non-keratinizing crevicular and junctional epithelium contained both isoenzymes through their full thickness in both inflamed and non-inflamed tissues. Pronounced staining of PGHS-2 was evident in the epithelia adjacent to Gram-positively stained organisms. In non-inflamed tissue, PGHS-1 and PGHS-2 were particularly evident in the spinous cell layer; however, fewer of the fibroblasts, endothelial cells, and resident mononuclear inflammatory cells stained positively for PGHS-1 as compared to PGHS-2, but this was less apparent in the inflamed tissues. The immunohistochemical staining patterns indicate that both crevicular and gingival epithelium are important sources of prostaglandin production in the gingival tissue of patients with periodontitis and that bacteria entrapped near to these sites may be important in promoting expression of inducible PGHS-2.

Stage of disease confounds apparent relationship between levels of N-ras and duration of survival in head and neck tumours.

The objective of this study was to determine whether elevated levels of N-ras correlated with clinicopathological data. Complete clinical data were available on 133 of 481 patients surgically treated for squamous cell carcinoma of the head and neck (SCCHN) who had immunohistochemical data for N-ras. Advanced stages of disease were strongly related to the staining for N-ras in tumour cells (P = 0.0031). The stage of disease was inversely related to duration of survival (P = 0.0017). Initial statistical evaluation revealed an apparent correlation between survival and N-ras staining. However, duration was found to be independent of the level of N-ras. The illusory relationship initially was a result of the confounding effect of the stage of disease.

Altered p53 in microdissected, metachronous, premalignant and malignant oral lesions from the same patients.

Aims-To determine whether mutant p53 alleles harboured by malignant tumours of the oral cavity were also present in previous premalignant lesions at the same site.Methods-Paraffin embedded tumour specimens along with their premalignant counterparts were analysed for p53 alterations using immunohistochemistry, microdissection, polymerase chain reaction amplification, and DNA sequencing.Results-Malignant lesions from five of eight patients showed overexpression of p53 protein by immunohistochemistry. Upon DNA sequencing, two of these five specimens had p53 mutations. Of the five patients whose cancers showed p53 overexpression by immunohistochemistry, three had previous premalignant lesions that also had immunohistochemically detectable p53 protein. However, DNA sequencing showed that none of these three had mutations in the p53 gene. The remaining five premalignant lesions had no immunohistochemically detectable p53 protein.Conclusions-Some premalignant lesions have increased p53 protein which can be detected by staining with antibody to p53. This staining is not caused by mutations in p53 that are found in subsequent tumours at the same site.

Immunohistochemical detection of the H-ras, K-ras, and N-ras oncogenes in squamous cell carcinoma of the head and neck.

The expression of H-ras, K-ras and N-ras oncogenes was analyzed on frozen sections of squamous cell carcinoma of the head and neck (SCCHN) by immunohistochemistry using anti-ras monoclonal antibodies. Of 22 primary SCCHN, 15 (68%) stained positive for H-ras, 10 (45%) for K-ras and seven (32%) for N-ras. Thirteen specimens (59%) stained positive for at least two anti-ras monoclonal antibodies. The presence of immunohistochemically detectable H-ras, K-ras and N-ras proteins was most frequently associated with an increase in tumor size and later stages of disease (T3 and T4), with no apparent correlation with lymph node involvement, site of occurrence, degree of differentiation, age, sex, or race. Thus, overexpression of members of the ras gene family occurs as a relatively common even in SCCHN and may be an important event in the later stages of tumorigenesis.