No abstract

Nasopharyngeal carcinoma (NPC) in Tunisia is characterized by its bimodal age distribution involving juvenile patients of 10-24 years and adult patients of 40-60 years. Three serological techniques were compared for primary diagnosis (N = 117) and post-treatment monitoring (N = 21) of NPC patients separated in two age groups. Immunofluorescence assay (IFA) was used as the "gold standard" for detection of IgG and IgA antibodies reactive with Epstein-Barr virus (EBV) early (EA) and viral capsid (VCA) antigens. Results were compared with ELISA measuring IgG and IgA antibody reactivity to defined EBNA1, EA, and VCA antigens. Immunoblot was used to reveal the molecular diversity underlying the anti-EBV IgG and IgA antibody responses. The results indicate that young NPC patients have significantly more restricted anti-EBV IgG and IgA antibody responses with aberrant IgG VCA/EA levels in 78% compared to 91.7% in elder patients. IgA VCA/EA was detected in 50% of young patients versus 89.4% for the elder group (P < 0.001). Immunoblot revealed a reduced overall diversity of EBV antigen recognition for both IgG and IgA in young patients. A good concordance was observed between ELISA and IFA for primary NPC diagnosis with 81-91% overall agreement. Even better agreement (95-100%) was found for antibody changes during follow-up monitoring, showing declining reactivity in patients in remission and increasing reactivity in patients with persistent disease or relapse. ELISA for IgA anti-VCA-p18 and immunoblot proved most sensitive for predicting tumor relapse. VCA-p18 IgA ELISA seems suitable for routine diagnosis and early detection of NPC complication.

Epigenetic modification is one of the mechanisms leading to gene silencing in neoplastic cells. By methylation-specific PCR, we analyzed the promoter methylation of three cancer-related genes: Ras Association domain Family 1A (RASSF1A), Death Associated Protein kinase (DAP-kinase) and Retinoic Acid Receptor β2 (RARβ2) in two NPC xenografts (C15 and C17), 68 primary NPC tumors, and 9 normal nasopharyngeal epithelia. We showed that C15 and C17 displayed a complete promoter methylation of RASSF1A, RARβ2 and DAP-kinase genes. In primary NPC tumors, the incidence of promoter methylation was very high for all three tested genes: 91% for RASSF1A, 88% for both RARβ2 and DAP-kinase whereas all normal nasopharyngeal epithelia were unmethylated. Interestingly, our study revealed that aberrant promoter methylation of the three genes were statistically associated with the lymph node involvement (p < 0.0001). In addition, hypermethylation of RASSF1A was correlated with age at diagnosis (p = 0.047) and T stage (p = 0.037) while the RARβ2 hypermethylation was associated with histological type (p = 0.011). Taken together, our results demonstrate that silencing of RASSF1A and RARβ2 expression by promoter hypermethylation is associated with highly differentiated tumors, advanced tumor stage and the presence of lymph node metastasis.To assess the functional significance of the epigenetic silencing of RARβ2 and DAP-kinase in NPC, we analysed the expression of two downstream target genes COX-2 and p53 by reverse PCR (RT-PCR) and immunohistochemistry (IHC). We revealed a significant association between expression of COX-2 and loss of RARβ2 through aberrant methylation (p = 0.003) in NPC biopsies.We concluded that the inactivation of RASSF1A, RARβ2 and DAP-Kinase by hypermethylation is a key step in NPC tumorigenesis and progression.

Our preliminary findings suggested that aberrant methylation of RASSF1A and RARbeta2 occurs frequently in Tunisian breast cancer patients compared with others. Furthermore, RASSF1A hypermethylation could be used as a potential marker of poor prognosis.

The TP53 gene, frequently mutated in human cancers, carries several polymorphisms. The one most informative and studied concerns codon 72; a single base changes the CGC (arginine) to CCC (proline). The arginine form was considered to be a significant risk factor in the development of cancer. However, various reports on this polymorphism are controversial. We carried out the same investigation in two groups of patients, a group with bladder cancer and another with breast cancer, and in healthy controls in two regions of our country, using an improved PCR-RFLP method. The number of Arg/Arg, Arg/Pro, and Pro/Pro genotypes was as follows: 21, 23, 3 and 13, 19, 2 for patients (total 47) and controls (34), respectively, in the first group; 18, 9, 3 and 19, 26, 4 for patients (30) and controls (49), respectively, in the second group. Statistical analysis of the genotype and allele frequencies did not reveal any difference between patients and controls in both groups except for a weak difference between the homozygotes to heterozygotes in the second group with a chi square of 4.1 (P = 0.045); the number of breast cancer patients is actually low (30) and should be increased in order to assess such a conclusion. Our overall results are therefore not consistent with a high risk associated with TP53 codon 72 polymorphism in breast and in bladder cancers.