G388R mutation of the FGFR4 gene is not relevant to breast cancer prognosis

Many components of mitogenic signalling pathways have been identified over the past 15 years, including the large family of protein kinases. Dysregulation of these biological processes may play an important role in the growth and survival of cancer cells. The 20 members of the fibroblast growth factor (FGF) family transduce a variety of biological signals via distinct transmembrane tyrosine kinase receptors (FGFR1 – FGFR4) encoded by four closely related genes (Powers et al, 2000). Mature FGFR proteins, which act as membrane-spanning tyrosine kinase receptors, have an extracellular ligand-binding domain, a transmembrane domain and a split intracellular tyrosine kinase domain (Green et al, 1996). Fibroblast growth factor -mediated signals have mitogenic, angiogenic and hormone regulatory effects and play key roles in a wide variety of crucial biological activities requiring cell growth, differentiation, migration and chemotaxis. Molecular anomalies of FGFR genes (inappropriate expression, single nucleotide polymorphism (SNP), splice variations, genomic alterations) have been described in several types of human cancer (bladder, cervical, colorectal carcinomas and multiple myeloma) and in skeletal deformities (achondroplasia, Crouzon syndrome and thanatophoric dysplasia type II) (Muenke and Schell, 1995; Avet-Loiseau et al, 1998; Cappellen et al, 1999; Vajo et al, 2000. Bange et al, (2002) recently studied the role of an SNP responsible for a missense mutation (G388R) in the transmembrane domain of FGFR4, in the progression and prognosis of nodepositive breast cancer. This G → A transition changes the sense of codon 388 from Gly (GGG) to Arg (AGG). The authors concluded that the G388R mutation in heterozygous or homozygous state was significantly over-represented in node-positive breast cancer patients with early relapse, but has no role in tumour formation, making this SNP a prognostic marker. On a worldwide basis, they found that allele distribution in normal controls and breast cancer patients did not differ significantly, showing an estimated 45.4% for the Gly/Gly allele, 45.7% for the Gly/Arg allele, and 8.9% for the Arg/Arg allele (Bange et al, 2002). No relationship has been found between this SNP and the prognosis for node-negative breast cancer. Our aim was to confirm the findings of Bange et al on a larger cohort, using an improved PCR – RFLP analysis and a reinforced and more adapted statistical analysis for investigating the possible association between G388R mutation and early cancer progression. Screening was conducted in a cohort of node-positive breast cancer patients who received different adjuvant therapies (endocrine (n=67), or chemotherapeutic (n=72)) and in a cohort of node-negative breast cancer patients (n=95). These cohorts were followed up, respectively, for median periods of 89 and 87 months. PCR – RFLP analysis was modified to make molecular screening more convenient and less time-consuming. Curves for disease-free survival (DFS) were plotted according to the Kaplan – Meier method, and the log-rank test was used for comparisons between groups, as in the study of Bange et al, (Kaplan and Meier, 1958; Cox and Oakes, 1984). Statistical analysis was reinforced by using three other nonparametric linear rank tests (Breslow, Peto – Prentice and Tarone – Ware) (Breslow, 1970; Peto and Peto, 1972; Tarone and Ware, 1977; Prentice, 1978).