טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
Ph.D Thesis
Ph.D StudentKislitsin Dmitry
SubjectUtilization of Ki-ras Mutations for Early Detection of
Colorectal Cancer
DepartmentDepartment of Biology
Supervisors Assistant Professor Ze'ev Lev
Professor Aaron Lerner


Abstract

We exploited mutations in the oncogene Ki-ras as a model for detection marker of cancerous cells in feces samples of colorectal cancer patients. We tested 371 tumor samples for the presence of Ki-ras mutations in codons 12 and 13, and 40 for the presence mutations in codon 61. 40% of tumor samples were found to bear Ki-ras mutations either in codons 12 or 13. 78% of all mutations were in codon 12 and the remnants in codon 13.  No mutations were found in codon 61. We exploited Ki-ras mutations as a marker for cancerous cell infiltration beyond visible tumor edges. We found very significant infiltration in the close proximity to tumor (up to 5 cm) in some cases, but distant surgical safety zones (15 cm approximately) were clean in all cases tested. In order to develop a procedure for detecting Ki-ras mutations in stool samples, we developed a method for storage of feces samples under conditions preventing DNA degradation for a half a year at least. We developed DNA extraction method allowing successful PCR amplification in all cases with fresh stool samples, and over 80% with dried stool samples from FOBT card. A method of detecting   mutant Ki-ras alleles allowing reliable detection of 1% mutant Ki-ras alleles was developed. In order to test the abundance of mutant Ki-ras alleles specific for primary tumor we tested seven feces samples received from colorectal cancer patients and matching tumor tissue samples, which were positive for Ki-ras mutations. We found unexpectedly high proportion of mutant tumor specific Ki-ras alleles in feces of colorectal patients. The proportion of mutant Ki-ras alleles reached 15-50% (30-100% of cancerous cells) in feces of colorectal cancer patients. We tested about 300 dry feces samples for the presence of mutant Ki-ras alleles. Unfortunately, the extremely high background in part of cases made this screening uninformative. We found that an extremely small amount of human DNA in the significant part of samples caused high background in these samples. The background in these cases may be explained by errors accumulation during PCR reaction. Several samples of fresh feces samples received from healthy donors were tested for the presence of human DNA.  It was found that only diminutive human DNA amount, as small as 0.5-2ng per gram of fresh stool, could be extracted from stool of healthy persons.  In contrast, it was found, that the feces samples of colorectal cancer patients contain 2-3 orders of magnitude more human DNA than feces of healthy persons. Therefore, the test for human DNA in feces samples could be a useful test for colorectal cancer detection.