Résumés des présentations de congrès & posters
TGFBR1 SNP’s are a genetic marker potentially useful for colorectal cancer screening.
ASCO Annual Meeting (May 31 - June 4, 2013)
Marc Pauly, Garry Mahon, Brigitte Metzger, Alain Menzel, Ben Weber, Mario-A Dicato; FRCS, Luxembourg, Luxembourg; Laboratoires Réunis Junglinster, Luxembourg, Luxembourg
Background
Colonoscopy (CS) used for colorectal cancer (CRC) screening is relatively expensive and intrusive. We have shown that CRC in Caucasian patients have a high frequency of certain TGFBR1 SNP's relative to normal controls. Here we extend the results to these SNP frequencies in normal tissue of the patients and the concordance of three different SNP's in the TGFBR1 gene across patients and subjects.
Methods
Tumor samples were obtained from 188 CRC patients and from 98 healthy control. After gDNA extraction, selected amplicons were amplified by PCR, followed by melting curve analysis. The statistical evaluation of the genotype frequency for the different SNPs comparing CRC patients with controls was carried out using logistic regression.
Results
Results previously reported were confirmed:- a highly significant association between TGFBR1 SNP genotype and colorectal cancer (P<0.000001). For example, with the TGFBR1 SNP rs334348, the frequency of the GG genotype was 43% for patients but 11% for controls, while the AA frequency was only 18% for patients and 43% for controls. All CRC patients and controls were typed for three SNP's in the TGFBR1 gene and the concordance was found to be perfect. That is, only 3 of the 27 possible genotypes were observed:- (i) rs334348, AA; rs334349, AA; rs1591, CC; (ii) rs334348, AG; rs334349, AC; rs1591, CT; and (iii) rs334348, GG; rs334349, CC; rs1591, TT. Blood and tumor samples from patients had the same identical SNP pattern.
Conclusions
The perfect concordance of the three SNP types in tumors and leukocytes shows that it reflects the underlying genotype of the patient rather than a somatic mutation in the tumour. Thus SNP type is a characteristic of subjects and a potential genetic marker for CRC susceptibility. We suggest that subjects aged 50 years or older be offered a test for a TGFBR1 variant. In case of a positive result, subjects should be informed that they are at risk of CRC, and strongly encouraged to have a colonoscopy. For this suggestion to become a definitive recommendation a cost-effectiveness analysis of the use of TGFBR1 SNP typing would be useful ( although such typing is inexpensive) and a prospective study of colonoscopy with TGFBR1 typing is mandatory.
European Journal of Cancer, Vol. 48, Suppl. 5 (July 2012), page S147, Abstract number 615
The PAI and TGFBR1 Genes Have a Significant Mutation Frequency at Single Nucleotide Polymorphism (SNP) Marker Sites in Caucasian Patients With Advanced Colorectal Cancer as Compared to Healthy Controls
M. Pauly1, B. Metzger1, M. Dicato1, A. Menzel2, G. Mahon1, B. Weber2.
1 Centre Hospitalier de Luxembourg, Recherche sur le cancer et maladies du sang, Luxembourg.
2 Laboratoires Réunis, Junglinster, Luxembourg.
Background
In order to evaluate various SNPs in different genes and chromosomal locations in caucasian patients as markers linked to the predisposition of the colorectal cancer (CRC) disease, we analyzed the frequency of different tumour-associated single-nucleotide polymorphisms in the following two genes: PAI (5G vs. 4G, rs1799899) and TGFbR1 (rs334348, A>G; rs334349, G>A; rs1591, A>C), chromosomal locations of caucasian CRC patients as compared to a healthy caucasian population.
Method
Tumour samples were obtained from 188 caucasian CRC patients at the adenocarcinoma stage and from 92 healthy caucasian individuals. After gDNA extraction, selected amplicons were amplified by PCR, followed by melting curve analysis. The statistical evaluation of the mutation frequency at the different SNPs while comparing CRC patients versus healthy individuals was calculated following the Chi-squared test.
Results
When analyzing 188 tumour samples and comparing with a healthy population to estimate the genotype distribution and mutation frequency in CRC cases, a significant frequency at the PAI (5G vs. 4G, rs1799899, P=0.025) locus. We had previously found a highly significant mutation frequency at three TGFbR1 SNPs (rs334348, rs334349, rs1591, P<0.005). Analysing the distribution of genotypes for TGFbR1 and PAI simultaneously did not lead to a significant reduction in chi-squared, suggesting that the effects of these variants were independent.
Conclusion
PAI (plasminogen activator inhibitor) was revealed as a significant marker and thus risk factor linked to the predisposition and/or occurrence of CRC in caucasian patients. The association seemed to be independent of that reported for SNPs rs334348, rs334349, rs1591 in the tumour growth factor beta receptor 1 (TGFbR1) gene.
Annals of Oncology, Vol. 23, Suppl. 4 (2012), page iv17, abstract number O-0017
M. Pauly1, B. Metzger1, G. Mahon1, M. Dicato1, A. Menzel2 and B. Weber2.
1 Centre Hospitalier de Luxembourg, Recherche sur le cancer et les maladies du sang, Luxembourg, Luxembourg.
2 Laboratoires Réunis, Junglinster, Luxembourg.
Background
In order to evaluate various SNP's in different genes and chromosomal locations in Caucasian patients as markers of the predisposition to the colorectal cancer (CRC) disease, we analyzed the frequency of different tumour-associated single-nucleotide polymorphisms in several genes including p53 and SMAD7 of Caucasian CRC patients as compared to a healthy Caucasian control population.
Method
Tumour samples were obtained from 188 CRC patients at the adenocarcinoma stage and from 98 healthy control individuals. After gDNA extraction, selected amplicons were amplified by PCR, followed by melting curve analysis. The statistical evaluation of the difference in genotype frequency at the different SNPs comparing CRC patients with healthy controls was carried out using the Chi-squared test.
Results
When analyzing 188 tumour samples and comparing with 98 healthy controls, a significant difference in the genotype distribution of the G429C SNP in the p53 gene was observed (P=0.046). Similarly, a significant difference was found for rs4939827 C>T in the SMAD7 gene (P=0.037), although no significant differences were found for rs4464148 T>C (P=0.585) or rs12953717 C>T (P=0.197) also in SMAD7. Differences in genotype distribution for CHR9 C>A rs719725 and CHR8 G>A rs7014346 were almost significant (P=0.050 and P=0.054 respectively). Significant results previously reported for two other genes were confirmed:- PAI, 5G vs. 4G, rs1799899 (P=0.047) and TGFBR1 A>G rs334348, G>A rs334349 and A>C rs1591 (P<0.0001). No significant differences were found for SNP's in 9 other genes.
Conclusion
The results for p53, SMAD7, PAI, CHR8 and CHR9 are suggestive rather than conclusive and invite confirmation through further study. Concerning the association between the TGFBR1 SNP's and colorectal cancer, on the other hand, the results are already clear. For example, with the TGFBR1 SNP rs334348, the frequency of the GG genotype was as much as 43% for the CRC patients but 11% for the controls, while the frequency of the AA genotype was only 18% for the patients and 43% for the controls.
