Analysis of the dinucleotide repeat polymorphism in the epidermal growth factor receptor (EGFR) in tumor and normal tissue of head and neck cancer (HNC) patients

N. Magné(Centre Antoine Lacassagne), Sandrine Pereira(Centre Antoine Lacassagne), M C Etienne(Centre Antoine Lacassagne), J. L. Formento(Centre Antoine Lacassagne), M Francoual(Centre Antoine Lacassagne), F Démard(Centre Antoine Lacassagne), O. Dassonville(Centre Antoine Lacassagne), Gilles Poissonnet(Centre Antoine Lacassagne), Pierre Szepetowski(Centre Antoine Lacassagne), G. Milano(Centre Antoine Lacassagne)
Journal of Clinical Oncology
July 15, 2004
Cited by 1

Abstract

5518 Background: EGFR plays a major role in cell proliferation. EGFR overexpression is strongly associated with poor prognosis in HNC patients. First intron of EGFR gene contains a highly polymorphic microsatellite sequence (9 to 23 CA-repeats) and in vitro transcription declines with increasing number of repeats. We analyzed EGFR gene polymorphism and EGFR expression in tumor (T) and normal mucosa (N) of HNC patients. Methods: T biopsies (initial diagnosis) were taken in 113 patients (mean age 60, 101 men, 12 women; 13 stage II, 21 stage III, 79 stage IV; 53 cancer-related deaths) along with N biopsies for 100 patients. EGFR levels were measured by ligand-binding assay. The microsatellite marker was analyzed by semi-automated fluorescent genotyping. Prognostic values of EGFR genotype and expression on specific survival were analyzed by multivariate Cox regression including performance status and node involvement. Results: Number of CA repeats varied from 13 to 22. Allelic distribution in tumor was trimodal : 49 % 16 CA, 17.5 % 20 CA, 15.5 % 18 CA. A similar pattern was observed in N samples. Heterozygosity was 60 % and 65% in T and N, respectively. In 33% of patients genotype was discordant between T and N, strongly suggesting high EGFR microsatellite instability. There was no relationship between EGFR genotype and EGFR expression (by considering either the shorter allele, the longer, homozygous samples, samples having a common allele, or classifying genotypes as short/long/intermediary defined as 2 alleles<17 vs 2 alleles[tmsnew]63[/tmsnew]17 vs others). Cox analysis confirmed the prognostic significance of T EGFR expression (p = 0.030). T EGFR genotype did not influence survival. A poorer survival was observed in patients with short repeats in normal mucosa (short/long/intermediary, p = 0.040). Patients with EGFR microsatellite instability exhibited significantly shorter survival (p = 0.013). Furthermore, EGFR microsatellite instability (p = 0.003) and T EGFR expression (p = 0.009) were independent significant survival predictors. Conclusion: These promising new findings shed a new light on EGFR implication in HNC. No significant financial relationships to disclose.


Related Papers

No related papers found

Powered by citation graph analysis