No abstract

Oncology, Newcastle upon Tyne, UK.Summary The epidermal growth factor receptor is homologous to the oncogene erb-,B and is the receptor for a class of tumour growth factors (TGF-ca). The clinical correlations with its expression were studied in 77 nonsmall cell lung cancers (NSCLC). They were stained for epidermal growth factor receptor (EGFr) by means of an indirect immunoperoxidase technique using a monoclonal antibody against the receptor. Normal lung tissue and normal bronchus were stained for comparison. Cancer tissue showed significantly increased staining compared to normal lung (P <0.05).Staining for EGFr in 40 squamous carcinomas was significantly stronger than in 37 specimens of other types of NSCLC (P<0.05), and staining in stage three NSCLC was stronger than in stage 1 and 2 (P<0.05).These results suggest that the presence of a high intensity of staining for EGF receptor is associated with spread of human non-small cell lung cancer and this receptor may be a suitable target for therapy.

Identification of prostate cancers at high risk of progression is difficult and a better understanding of how peptide growth factors influence cellular function might be useful. Fibroblast growth factors (FGFs) have been implicated in prostate cancer development. FGF8 was identified in the Shionogi mouse mammary carcinoma SC-3 cell line as an androgen-induced mitogen. We tested if FGF8 was over-expressed in human prostate cancer and if its expression correlated with clinical data and outcome. One hundred and six cases of prostate cancer and ten cases of BPH were examined. In situ hybridization was employed to detect FGF8 mRNA expression, which was identified within the malignant prostatic epithelium in 85/106 (80.2%) cases. Increased expression of FGF8 correlated significantly with higher Gleason scores (P=0.0004) and advanced tumour stage (P=0.0016). Using immunohistochemistry, we confirmed over-expression of the FGF8b isoform. Men with tumours which expressed high levels of FGF8 had worse survival (P=0.034), although FGF8 mRNA was not able to provide additional prognostic information in a multivariate analysis. Additionally, FGF8 expression was shown to persist in androgen independent prostate cancer. Using a range of normal adult tissues, FGF8 expression was restricted to neurones and the germinal epithelium in addition to the prostate. In vitro studies demonstrated that in the presence of neutralizing antibody to FGF8b there was significant inhibition of prostate cancer cell growth, confirming the biological significance of FGF8 in prostate carcinogenesis.

Overexpression of fibroblast growth factor 8 (FGF8) mRNA has been previously described in prostate cancer. Of its four isoforms, FGF8b is thought to be the most important in carcinogenesis. We hypothesised that immunodetection of FGF8b in archival prostate cancer specimens is of potential prognostic value. Using a selected cohort of prostate tumours from transurethral (n ¼ 30) and radical prostatectomies (n ¼ 59), an optimised protocol for FGF8b immunoreactivity was used to corroborate expression with clinical parameters. No expression was observed in benign prostates (n ¼ 10). In prostate cancer, immunoreactivity was localised to the malignant epithelium with weak signals in the adjacent stroma. Expression of FGF8b in stage T1 and T2 cancers were 40 and 67%, respectively. In contrast, FGF8b expression was present in 94% of T3 and 100% of T4 cancers. By histological grade, FGF8b was found in 41% of low-grade cancers (Gleason score 4 -6), 60% of intermediate-grade cancers (Gleason score 7 and 92% of high-grade cancers (Gleason score 8 -10). The intensity of expression was significantly associated with stage (P ¼ 0.0004) and grade (Po0.0001) of disease. We further hypothesised that FGF8b overexpression resulted from enhanced transcription and translation rather than from abnormalities involving the FGF8 gene locus. This was tested by means of fluorescent in situ hybridisation in 20 cancer specimens to map the FGF8 gene locus. FGF8 gene copy number in benign and malignant nuclei was found to be similar (2.3370.57 and 2.070.81, respectively P ¼ 0.51). Based on these findings, we propose a multicentre study on cohorts of patients to further evaluate FGF8b as a potential prognostic marker in prostate cancer.