Bit Matrices

The microsatellite status is a marker of DNA mismatch repair integrity. DNA mismatch repair, one of the several mechanisms that cells have evolved to protect their genomes, is effected by a complex of repair enzymes which include MLH1, PMS2, MSH2, and MSH6. Failure of expression of any one of these enzymes as a result of mutations or epigenetic events leads to defective DNA mismatch repair (MMR). The consequence is failure to repair nucleotide mismatches especially in mono- and dinucleotide repeat sequences called microsatellites. These microsatellite repeats are distributed non-uniformly throughout the genome, and are found mostly in non-coding regions. Some are however found in coding regions and promoters of genes involved in cell growth regulation and DNA repair, eg, c-kit, TGFβRII, BAX, MSH2, p16/INK4a. Defective DNA mismatch repair causes MSI which is characterized by changes or instability in the length of the microsatellites that is not present in the normal cells of an individual. When these changes involve coding and promoter regions of cancer-related genes, they induce mutation in them and confer the risk of transformation on cells17, 20, 23, 26, 28, 66, 67. The presence of MSI is used to test for defective MMR. Since the instability is not uniform throughout the genome the chances of detecting instability is increased by testing more than one microsatellite locus. The combined use of more than one locus increases the specificity and sensitivity to near 100% ²⁵.

18

In addition to, MSI, immunohistochemistry is sometimes used to test for DNA MMR defects.

Tissue sections are stained with antibodies to MLH1, PMS2, MSH2, and MSH6 in order to detect the presence or absence of mismatch repair enzymes or to validate the results of MSI⁶⁸. The sensitivity of IHC in detecting MMR defect is 92% when all 4 antibody panel is used. The specificity is 95%. However, when only the MLH1 and MSH2 antibodies are used, the sensitivity falls to about 85%.⁶⁹ The concordance rate for MSI and IHC in detecting MMR is 92-99.4%.^{69, 70}

The National Cancer Institute, USA, recommended the use of at least 5 loci (the Bethesda panel) comprising 2 mononucleotide repeats loci BAT25 and BAT26 and dinucleotide repeats loci D5S346, D2S123, D17S250, and D4S531 in testing for MSI. However, the dinucleotide repeats have been found to be less specific (because of the high rate of polymorphisms in these loci in the population) and less sensitive.^70-73 Recent studies have tended to replace the dinucleotide markers with more sensitive and specific mononucleotide microsatellite loci, eg, NR21, NR 22 and NR 24.70, 73, 74 In contrast to the dinucleotide repeat loci, the mononucleotide repeats are generally quasi-monomorphic, ie, that exhibit little or no polymorphisms in the general population. For this reason, MSI can be tested using these quasi-monomorphic loci without the need for use of normal tissue from individual patients for negative control⁷¹. Several studies have, however, tended to use a limited number of loci, most commonly BAT25 and BAT26, because of their high sensitivity, either individually or together, in detecting MSI.

The BAT 25 locus maps to intron 16 of the c-kit oncogene and contains a 25 thymine repeat tract. The BAT 26 locus, on the other hand, is located in the 5^th intron of the MSH2 gene and contains a 26 adenine repeat tract. Both loci have been found to be sensitive in the detection of MSI which manifests as shortening in the mononucleotide repeat tracts in tumour DNA.⁶⁷ For

19

example, in the study by Schneider et al, among a panel of 8 markers, including the Bethesda dinucleotide markers above, BAT 25 showed a sensitivity of 85.7% in detecting MSI. BAT 26 showed a sensitivity of 92.9%, and combined together, sensitivity increased to 96.6%.⁷¹ Also Lee et al compared the sensitivity of BAT 25 and BAT 26 using IHC of MLH1 and MSH2 as the standard; and found that BAT 25 has a sensitivity of 87.1% and BAT 26 100%.⁷⁵ Zhou et al had earlier shown that BAT 26 has a sensitivity of 99.6% in the detection of MSI. ⁷⁶ Kim et al showed that BAT 26 has a sensitivity of 100% among a panel of 50 microsatellite markers.²⁴ Furthermore, Bianchi et al showed that in a panel of 5 Bethesda-recommended markers, BAT 25 showed a sensitivity of 84.6%, while BAT 26 showed a sensitivity of 92.3%; while Deschoolmeester et al found a sensitivity of 100% for both BAT25 and BAT 26 in detecting MSI. ^{71, 73}

Due to the high sensitivity rates which have been found with the use of these two microsatellite markers (84.6%-100%), and the limited resource at our disposal, BAT 25 and BAT 26 have been utilized in this study in testing for MSI in 47 gastric carcinoma cases diagnosed in UCH, Ibadan, between 2000 and 2011.

MSI are usually associated with mutations in other cancer-related genes especially tumour-suppressor genes such as TGFRβII, IGFRII, BAX, etc. ^{19, 20}

MSI was first described in HNPCC but is now known to be present in several familial and sporadic cancers. In nearly all cases of HNPCC and HNPCC-related extracolonic cancers, e.g., some gastric carcinomas, germline mutation in mismatch repair genes give rise to MSI. ^{20, 26} The rate of microsatellite instability in gastric carcinoma is highly variable both within one geographical region and from region to region. The quoted figures are between 8% and 67%.^17,

18, 21, 23- 26, 77 This variability may be due to a number of factors. One factor may be the difference

20

in methods used to determine MSI and the number of loci that are analyzed for instability. The other factor may be a real difference in biological behaviour that seems to exist among tumours from patients of different geographical background and ethnic origin. For example, MSI was detected more frequently among Japanese patients than in European Americans (39% versus 20%) in another study done by Theuer et al¹⁷.

However these ethnic differences were not duplicated in a work by Schneider et al which compared the rateof MSI in specimens from 142 Americans (29%), 20 Koreans (30%), and 6 Chileans (33%) undergoing a gastrectomy.⁷¹

Several studies have linked MSI-positive gastric carcinomas with distinct clinicopathological features. For example, gastric carcinomas with MSI tend to be located in the gastric antrum, occur in the elderly, be in the advanced stage, be of the intestinal histological subtype of the Lauren classification, and have lower rates of lymph node metastases and vascular permeation, and higher survival rates. With the exception of advanced tumour stage, the other features generally confer more favourable prognosis on the patients.17, 18, 21-26, 28, 29, 65, 71, 75, 77

However, Theuer et al found that MSI-positive tumours are associated more with proximally located tumours, while Lee et al found MSI-positive gastric carcinomas to be predominantly early cancers.^{17, 78} Schneider et al found that MSI status correlated with neither age nor sex of the patients, nor with depth of invasion, tumour grade, tumour differentiation, or histological type according to Laurén classification.⁷¹

Less consistent associations include patient’s gender, and a high intratumoural lymphocyte infiltrate.17, 29, 77

Microsatellite instability may also affect the response to therapy in gastric cancer. Mismatch repair–deficientcell lines are resistant to the antitumour agents cisplatin,doxorubicin, etoposide, and fluorouracil, and maybe sensitive to bleomycin.^{17, 27}

21

High Resolution Melt (HRM) Curve analysis has been utilized in this study to define the microsatellite status of gastric carcinomas in our patients. HRM is a sensitive and simple to interpret post PCR method for genotyping. It utilizes the difference in the melting characteristics of DNA sequences to separate them into different genotypes. The melting temperature of any DNA sequence is the temperature at which half of the DNA is double-stranded and half is single-stranded, and this melting temperature is dependent on DNA length, composition, GC content and complementarity. PCR products with different sequence characteristics will have different melt curve characteristics (Tm, melt curve shape and fluorescent intensity). HRM has been used for mutation screening (including MSI detection) and SNP genotyping.^{74, 79}