gms | German Medical Science

49. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds)
19. Jahrestagung der Schweizerischen Gesellschaft für Medizinische Informatik (SGMI)
Jahrestagung 2004 des Arbeitskreises Medizinische Informatik (ÖAKMI)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie
Schweizerische Gesellschaft für Medizinische Informatik (SGMI)

26. bis 30.09.2004, Innsbruck/Tirol

A strategy for the analysis of complex genotype-phenotype relationships: MDR1 genotypes are associated with susceptibility for colorectal cancer

Eine Strategie für komplexe Genotyp-Phänotyp-Beziehungen: MDR1-Genotypen sind mit der Suszeptibilität für das kolorektale Karzinom assoziiert

Meeting Abstract (gmds2004)

  • corresponding author presenting/speaker Karla Köpke - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland
  • Elena Gaikovitch - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie und Voronezh Burdenko State Medical Academy, Berlin und Voronezh, Deutschland
  • Andreas Johne - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland
  • Uwe Malzahn - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland
  • Farhad Arjomand - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland
  • Christian Meisel - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland
  • Przemyslaw Michael Mrozikiewicz - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie und Research Institute of Medical Plants, Berlin und Poznan, Deutschland
  • Jury Chernov - Voronezh Burdenko State Medical Academy, Voronezh, Russland
  • Ivar Roots - Charité, Universitätsmedizin Berlin, Institut für Klinische Pharmakologie, Berlin, Deutschland

Kooperative Versorgung - Vernetzte Forschung - Ubiquitäre Information. 49. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 19. Jahrestagung der Schweizerischen Gesellschaft für Medizinische Informatik (SGMI) und Jahrestagung 2004 des Arbeitskreises Medizinische Informatik (ÖAKMI) der Österreichischen Computer Gesellschaft (OCG) und der Österreichischen Gesellschaft für Biomedizinische Technik (ÖGBMT). Innsbruck, 26.-30.09.2004. Düsseldorf, Köln: German Medical Science; 2004. Doc04gmds134

Die elektronische Version dieses Artikels ist vollständig und ist verfügbar unter: http://www.egms.de/de/meetings/gmds2004/04gmds134.shtml

Veröffentlicht: 14. September 2004

© 2004 Köpke et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielf&aauml;ltigt, verbreitet und &oauml;ffentlich zug&aauml;nglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.


Gliederung

Text

Introduction

Colorectal cancer (CRC) is among the three most common cancers in most industrial nations. Polymorphisms in different genes were found associated with CRCs. Recently a possible role of the multidrug-resistance gene (MDR1) was reported in colon tumorigenesis [1], [2]. The MDR1 gene encods P-glycoprotein (P-gp), a transmembrane transporter, that acts as an efflux pump of xenobiotics. Furthermore, P-gp might play a fundamental role in regulating cell death and differentiation, proliferation, and in immunology. A large body of evidence indicates that, apart from genetic factors, several dietary and lifestyle factors such as high-fat diet, red meat intake, alcohol consumption or smoking are likely to influence the risk of colorectal cancer [3], [4], [5], [6].

This population-based case-control study was performed to elucidate the potential role of MDR1 genotypes combining the three polymorphisms -129 T →C (exon-1b), G2677T/A (exon 21) and C3435T (exon 26) in tumorgenesis of colorectal carcinomas; additionally, lifestyle risk factors were taken into account.

Patients and Methods

Patients. A group of 285 colorectal cancer patients and 275 individuals without malignancies was recruited in five hospitals in the Voronezh region (Russia). The study was approved by the responsible Central Moscow Ethics Committee.

Lifestyle parameters. Heaviness of smoking was expressed in pack-years. Alcohol consumption was divided into 3 categories,meat quantity into 4, and meat quality into 2.

DNA extraction and genotyping of MDR1. The samples were analysed in the Institute of Clinical Pharmacology, Charité, in Berlin using standard procedures.

Statistical analysis.To uncover the potentially complex relationships between categorical variables, a multiway crosstabulation loglinear model from the SPSS Hierarchical Loglinear Analysis procedure was used.The unsaturated model with the smallest number of variables which still retains statistical significance is the best. Multiple logistic regression was used to reveal the relationship between status variable on the one hand and lifestyle parameters andgenotypes on the other hand.Associations between categorical variables were tested with contingency table analysis using chi-square likelihood ratio statistics and the exact test, if necessary. All procedures were performed with SPSS™ (version 11.0.1, SPSS Inc., Chicago, USA).

Haplotype analysis. Haplotypes were constructed as combinations of the SNPs -129 T →C, 2677 G→T/A, and 3435 C→T. Haplotype pairs were predicted with the help of our algorithm [7].

Results

MDR1 gene sequence variability. The SNP -129 T →C was rare with a frequency of approximately 3%. For the variant at position 2677 all three different nucleotides were found. The rare nucleotide 2677A had a frequency <3%. Statistically significant deviations from Hardy-Weinberg equilibrium in the patient group were found for 2677 G→T (P=0.041) and 3435 C→T (P=0.032).

MDR1 genotypes and haplotypes. Eighteen different genotypes combining the variants -129 T→C, 2677 G→T/A, and 3435 C→T were found. Haplotype pairs were in silico predicted on the basis of the entire genotype sample. Eleven genotypes have an unambiguous haplotype pair. For seven genotypes more than one haplotype pair was predicted. Eleven different haplotypes were inferred. All haplotypes including nucleotide A at position 2677 were rare with a frequency <1.5%. The same holds true for haplotypes which are mutated at position -129. Rare haplotypes with a frequency <1.5% were excluded from further analysis.

Complex relationships between all categorical variables. Saturated and unsaturated loglinear models which contain all variables were tested. The paramount influence of smoking on cancer status was confirmed by the results of a logistic regression between cancer status (response variable) and lifestyle parameters and genotypes (independent variables). Consequently, the lifestyle parameters explained the greater part of the variance for the disease status and obscured influences of genetic factors if they existed at all.

Influence of smoking behavior on cancer status. The expected association between smoking grade and cancer status was highly statistically significant (P=0.00016). A sample stratification was inevitable.

Association between MDR1 genotypes and cancer. Analysis of all never smoking cases and controls pointed to an association between colorectal cancer risk and MDR1 genotypes (P=0.051). In contrast, for all smoking cancer patients no differences in genotype frequencies between cases and controls could be found. Bearing in mind the potential gender differences a statistically significant association between genotype and colorectal cancer was found (P=0.007) ) in the group of never smoking women, considering the late age of onset. In the group of never smoking men no significant association could be found presumably due to the small sample size (n=17; P=0.107). In the next step we tested colon and rectal cancer patients separately. This analysis revealed significant genotype differences between cases and controls for women with colon cancer (P=0.043) and rectal cancer (P=0.014) as well as for men for rectal cancer (P=0.047).

Analysis ofMDR1 haplotypes and cancer. No significant association was found.

Discussion

A strategy of examination for complex genotype-phenotype relationships should be used which considers its characteristics. Our analysis allowed for detailed insights to the interaction of genetic and lifestyle risk factors for colorectal cancer. Further analyses are necessary, which should take into account more variants of the MDR1 gene and gene variability of further genes, playing a possible role in colorectal cancer susceptibility.

Acknowledgements

This study was supported by grants of the German Federal Ministry of Education and Research ("Berlin Center for Genome Based Bioinformatics" grant no. 031U209B).


References

1.
Potocnik U, Ravnik-Glavac M, Glavac D. Functional MDR1 polymorphism (G2677T and C3435T) and TCF4 mutations in colorectal tumors with high microsatellite instability. Cell Mol. Biol. Lett. 2002;7: 92-5.
2.
Potocnik U, Glavac MR, Golouh R, Glavac D. The role of P-glycoprotein (MDR1) polymorphisms and mutations in colorectal cancer. Pflügers Arch. 2001;442: R182-3.
3.
Giovannucci E. Modifiable risk factors for colon cancer. Gastroenterol. Clin. North Am. 2002;31: 925-43.
4.
Kim JI, Park YJ, Kim HH, Kim JI, B.J. Song BJ, Lee MS, Kim CN, Chang SH. hOGG1 Ser326Cys polymorphism modifies the significance of the environmental risk factor for colon cancer. World J. Gastroenterol. 2003; 9: 956-60.
5.
Slattery ML, Samowitz W, Ballard L, Schaffer D, Leppert M, Potter JD. A molecular variant of the APC gene at codon 1822: its association with diet, lifestyle, and risk of colon cancer. Cancer Res. 2001;61: 1000-4.
6.
Giovannucci E, Martinez ME. Tobacco, colorectal cancer, and adenomas: a review of the evidence. J. Natl. Cancer Inst. 1996;88: 1717-30.
7.
Terhalle W, Schmidt W, Köpke K. New method for predicting haplotype pairs for genotypes. In: Spang R, Beziat, P, Vingron, M Hrsg. Currents in Computational Molecular Biology 2003. Berlin: MPI for Molecular Genetics and BCB;2003: 95-6.