Improving FOBT based colorectal cancer screening by faeces DNA testing: Copy number and promoter methylation profiling of faeces derived DNA for secondary prevention of colorectal cancer
Samenvatting van de aanvraag
Colorectal cancer (CRC) is the second leading cause of cancer death in the western world with 4400 deaths per year in The Netherlands. Secondary prevention is the most realistic approach for reducing this high number of colorectal cancer deaths. A recent expert meeting in The Netherlands initiated by the Dutch Cancer Society and ZonMw (Netherlands Organisation for Health Research and Development) yielded the consensus that a national population screening for colorectal cancer should be implemented and carried out in the Netherlands, based on faecal occult blood testing (FOBT) (Visser et al, 2005). In addition, a research agenda for continued improvement of the FOBT-based screening was drafted. A main topic here was to explore the performance of faecal tests additional to FOBT, such as tests for DNA hypermethylation or other DNA alterations that go undetected by existing (commercial) multiple DNA point mutation tests. New tests should be compared with results of screening with FOBT only. Also in respect of possible future replacement of FOBT as a screening tool in 10 to 15 years, multi-biological-variables testing on faeces, including DNA methylation and additional DNA alterations, was labeled as a research priority. The present grant application fits exactly in these priorities. Existing commercial fecal DNA tests have yielded promising results but need further improvement. A major drawback is that only a limited number of genes are tested, and that a substantial part of the test relies on the actual detection of point mutations. This is technically demanding, requires high tech faeces sampling and consequently is expensive. There obviously is room for improvement e.g. by including additional genes, which are associated with adenoma to carcinoma progression, and alternative, more versatile test methods. The present grant application explicitly focuses on these issues. In addition to the genetic alterations described in the well known (but oversimplified) Vogelstein model of colorectal carcinogenesis (e.g. APC, kRAS, p53) we have identified a number of other genetic and epigenetic alterations specifically associated with high adenoma to carcinoma progression risk (Hermsen et al., Gastroenterology 2002). In addition we have performed pilot studies demonstrating the feasibility of detecting these targets in DNA isolated from stool by measuring DNA copy number changes and detecting promoter methylation. The aim of the present grant application is to investigate whether promoter methylation- and DNA copy number profiling of free, tumor derived DNA isolated from faeces, can be used for improving FOBT based screening for CRC. Plan of investigation: Phase 1: Selecting high-risk markers for designing dedicated MLPA and MSP probe sets: Dedicated MLPA and MSP probe sets will be designed by selecting high-risk markers. These markers will be selected based on high resolution microarray CGH and expression microarray data of colorectal adenomas and carcinomas combined with cutting edge bioinformatics for selecting the most likely candidate methylated tumor suppressor genes. Phase 2: Determine the prevalence of MLPA and MSP marker alterations in normal colorectal mucosa (n=50), adenoma (n=50), progressed adenoma (n=50) and carcinoma (n=50) tissue samples Phase 3: Determine the sensitivity and specificity of MLPA/MSP/MSI based faeces DNA testing in a case control study: Faeces of CRC cases (n=100), colorectal adenoma cases (n = 100) as well as colonoscopic negative controls (n=300) will be collected. DNA will be isolated from these stool samples and promoter methylation and MLPA profiles obtained from free DNA will be compared to the profiles obtained from DNA isolated from the matching tumor tissue. A pilot study had already demonstrated the logistic and technical feasibility of this approach. MSI is included because it is a genetic hallmark of a subgroup of 15% of CRCs. Phase 4: First line investigation of the performance of MLPA/MSP/MSI based faeces DNA testing in a screening setting: Faeces will be obtained in the context of CRC screening trials in the Amsterdam and Maastricht area which will yield at least ca. 2000 screenees. Faeces and tumor samples of the anticipated 100 individuals tested positive as well as faeces samples of a random selection of 300 individuals tested negative will be evaluated. This will allow to estimate the positive and negative predictive value of the test in a colorectal cancer screening setting, and to compare these test characteristics to those of the currently used colonoscopy and FOBT tests. Relevance: This research will bring us a major step forward towards new low-cost/high performance DNA based diagnostic tests for CRC which will be highly suitable for improving FOBT based population screening for CRC in the Netherlands.