It is generally assumed that tumor suppressor and mutator pathways cooperate in the progression of tumors. Recent mathematic modeling has  suggested that mutator mutations that create genetic instability are an important process in carcinogenesis and are of greatest importance when they occur as an initial step. In a significant fraction of gastric adenocarcinomas, microsatellite instability (MSI) is present as a defect in the repair of DNA replication errors owing to mutations in DNA mismatch repair genes. By decreasing genomic stability, mutator mutations accelerate the accumulation of random mutations, including those in oncogenes and tumor suppressor genes. Inactivation of tumor suppressor genes such as APC, nm23, Rb and p53 frequently occur in gastric adenocarcinomas. MSI from DNA mismatch repair deficiency and loss of heterozygosity (LOH) from inactivation of tumor suppressor genes can be detected in microsatellite alleles by quantitative multiplex fluorescent PCR.
In the current study by Gazvoda et al. , the authors have investigated how tumor status and patient survival is correlated with multiple markers of MSI and LOH, and how these markers can be used prognostically. Importantly, a major aim of the study was to challenge the concept of the use of single or few prognostic markers in favor of multiple markers. Although not a large study, the group correlated tumor suppressor and MSI microsatellite markers with tumor types and grades, and in particular, first reported no observed LOH for Rb in any tumors originating in the upper third of the esophagus. Different MSI and LOH groupings were associated with particular types of histological differentiation, anatomical origin, postsurgical survival and prognosis; the findings suggest different roles and mechanisms for these individual markers.
MSI can be classified as high-level MSI (MSI-H), low-level MSI (MSI-L), and microsatellite stable (MSS) type according to the frequency of mutations. One of the main conclusions from the study is that MSI-L combined with mutations in several chromosomal alterations gives the poorest prognosis. This study confirms previous reports that MSI status is important in gastric cancer [3–5]. It has been suggested that, as MSI-L status is similar to MSS tumors (tumors with no observable microsatellite variation), few MSI markers might need to be compared with the MSI status in staging and treatment. This might be an oversimplification. For example, it is pointed out that MSI-L occurs infrequently in breast cancer, but occurs with relatively high frequency in colorectal, ovarian and endometrial cancers. This study found poorer differentiation in tumors with low MSI status than MSI-high status. LOH and MSI appear to be inversely correlated where less than 10% of the MSI-H samples had LOH status which was lower than MSI-stable samples. LOH-H and MSI-H were not associated and thus might be independent.
A conclusion reached by this group is that the use of less than five different screening markers for gastric adenocarcinoma might not provide an accurate predictor of risk, supporting the concept that overlapping effects from errors in both tumor suppression and repair produce statistically more risk than either alone. Blanes and Diaz-Cano  arrived at a similar conclusion regarding MSI and also pointed out the need to select and accurately amplify the correct locus to arrive at meaningful conclusions based on MSI. Therefore, employing a wider panel of screening markers (as is suggested in this report) should increase the precision of this technique. Additionally, whenever possible, sampling from multiple intratumoral sites would minimize problems from MSI heterogeneity. Some additional factors which might warrant further consideration would include the contributions of gut flora (which correlate with MSI-L status, ), diploid/nondiploid karyotype, activation of oncogenes and epigenetic (mutation independent) inactivation of tumor suppressor genes such as the methylation of the hMLH1 (mutL homolog1) promoter causing decreased expression of hMLH1, an important factor in mismatch repair [7–10]. Interestingly, mismatch repair appears to provide significant protection against tumors induced by methylation agents, suggesting that mismatch repair is self-preserving . MSI-H status, although correlated with MLH1 inactivation, was paradoxically with the longest postsurgical survival . Thus, apparently, the accumulation of large numbers of mutations does not necessarily correlate with the worst-case scenario clinically.
Studies from several investigators have now revealed a prominent contribution of aberrant CpG promoter hypermethylation in gastric carcinogenesis. Although it is true that silencing tumor genes (TSGs) via promoter methylation contributes to the aggressiveness of various cancers, it is also clear that methylation of specific TSGs can decrease the responsiveness of some tumors to chemotherapy. For example, methylation of the DNA repair enzyme O6-methylguanine-DNA methyltransferase (MGMT) in gliomas is a useful predictor of the responsiveness of the tumor to alkylating agents . With respect to prolonged overall and disease-free survival, MGMT methylation was shown to be a stronger prognostic factor than age, stage, tumor grade or performance status. MGMT is a key factor in resistance to alkylating agents, because the transfer of alkyl groups to MGMT prevents the formation of lethal cross-links to DNA. The LOH-L subpopulation described by Gazvoda and colleagues  might possess a greater extent of epigenetic defects as epigenetic silencing and DNA mutation of a common target is often mutually exclusive. Second, MGMT inactivation by promoter hypermethylation has been linked with MSI-L more than MSI-H and stable colorectal carcinomas . Considering this, it will be interesting to determine the relationship between MSI-L and MGMT hypermethylation in gastric cancer. Thus, the findings of Gazvoda et al.  might hint of an epigenetic basis for the finding that the poorest prognosis was associated with patients with both MSI-L and LOH-L status. Unlike the MSI-L and LOH-L patients, perhaps MSI-H patients are more responsive to adjuvant therapy administered after surgery.
In this study, MSI-L was correlated with poorest prognosis, lowest differentiation and greatest tumor mass. When combined with LOH-L, MSI-L patients had the poorest outcome. Interestingly, MSS survival was actually lower than that in MSI-H patients. These findings suggest that MSI-L with chromosome loss is the worst condition. Therefore, this study suggests that MSI and LOH pathways converge in complex and unanticipated ways. Thus while confirming that LOH is essential to the progression of many tumors, MSI status, particularly low MSI status might be significant and have relevant clinical implications. Finally these results suggest that in future studies, the use of multiple markers, for example hypermethylated genes (CDH1, p16/INK4A, p14/ARF and MLH1), will test the experimental design proposed by this group and improve the accuracy of diagnosis and prognosis.
Conflict of interest: none declared.
1. Gazvoda B, Juvan R, Zupanič-Pajnič I, Repše S, Ferlan-Marolt K, Balazic J, Komel R. Genetic changes in Slovenian patients with gastric adenocarcinoma evaluated in terms of microsatellite DNA. Eur J Gastroenterol Hepatol 2007; 19:1082–1089.
2. Beckman RA, Loeb LA. Efficiency of carcinogenesis with and without a mutator mutation. Proc Natl Acad Sci U S A 2006; 103:14140–14145.
3. Leung WK, Kim JJ, Kim JG, Graham DY, Sepulveda AR. Microsatellite instability in gastric intestinal metaplasia in patients with and without gastric cancer
. Am J Pathol 2000; 156:537–543.
4. Kelsen DP. Gastric cancer
. Semin Oncol 1996; 23:279–407.
5. Tamura G. Genetic and epigenetic alterations of tumor suppressor and tumor-related genes in gastric cancer
. Histol Histopathol 2002; 17:323–329.
6. Blanes A, Diaz-Cano SJ. Complementary analysis of microsatellite tumor profile and mismatch repair
defects in colorectal carcinomas. World J Gastroenterol 2006; 12:5932–5940.
7. Cerezo L, Cardenes H, Michael H. Molecular alterations in the pathogenesis of endometrial adenocarcinomas. Therapeutic implications. Clin Transl Oncol 2006; 8:231–241.
8. Hamilton JP, Meltzer SJ. A review of the genomics of gastric cancer
. Clin Gastroenterol Hepatol 2006; 4:416–425.
9. Hudler P, Vouk K, Liovic M, Repse S, Juvan R, Komel R. Mutations in the hMLH1 gene in Slovenian patients with gastric carcinoma. Clin Genet, 2004; 65:405–411. Erratum in: Clin Genet 2005; 66
10. Wheeler JM. Epigenetics, mismatch repair
genes and colorectal cancer
. Ann R Coll Surg Engl 2005; 87:15–20.
11. Colussi C, Fiumicino S, Giuliani A, Rosini S, Musiani P, Macrí C, Potten CS, Crescenzi M, Bignami M.1,2-Dimethylhydrazine-induced colon carcinoma and lymphoma in msh2(–/–) mice. J Natl Cancer
Inst 2001; 93:1534–1540.
12. Li JH, Shi XZ, Liu M, Wang Y, Yu ZH, Xu GW, Lu S. Methylation of hMLH1 gene promoter in gastric carcinoma with microsatellite instability. Ai Zheng 2005; 24:273–277.
13. Esteller M, Garcia-Foncillas J, Andion E, Goodman SN, Hidalgo OF, Vanaclocha V, et al. Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 2000; 343:1350–1354.
14. Whitehall VL, Walsh MD, Young J, Leggett BA, Jass JR. Methylation of O6-methylguanine DNA methyltransferase characterizes a subset of colorectal cancer
with low-level DNA microsatellite instability. Cancer
Res 2001; 61:827–830.