|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Adverse Prognostic Effect of Methylation in Colorectal Cancer Is Reversed by Microsatellite InstabilityFrom the Departments of Medical Oncology and Colorectal Surgery, St Vincent’s Hospital, Darlinghurst; and the Schools of Medicine and Medical Sciences, University of New South Wales, Sydney, Australia. Address reprint requests to Robyn Ward, MD, Department of Medical Oncology, St Vincent’s Hospital, Victoria St, Darlinghurst, NSW 2010, Australia; e-mail: r.ward{at}garvan.unsw.edu.au.
Purpose: DNA methylation is an important biologic event in colorectal cancer and in some cases is associated with the development of microsatellite instability (MSI). In this study, we sought to determine the prognostic significance of DNA methylation, both in univariate analysis and in concert with other clinicopathologic factors known to influence outcome. Patients and Methods: Fresh tissue (625 cancers) was obtained from 605 individuals (age range, 29 to 99 years) undergoing curative surgery for colorectal cancer at one institution during a period of 8 years. Clinicopathologic details were recorded for all tumors, including stage, grade, type, vascular space invasion, and clinical follow-up to 5 years. Microsatellite status was assessed using standard markers. Methylation of p16 and hMLH1 promoters was determined by methylation-specific polymerase chain reaction (PCR), whereas methylation at methylated-in-tumor loci (MINT)1, MINT2, MINT12, and MINT31 loci were assessed by bisulfite-PCR. Results: Patients with microsatellite unstable tumors (12%) had better disease-specific survival than those with microsatellite stable (MSS) tumors (univariate analysis: hazard ratio [HR], 0.53; 95% CI, 0.27 to 1.0). Overall survival of individuals with MSS tumors was influenced by three independently significant factors: tumor stage (HR, 7.3; 95% CI, 5.1 to 10.4), heavy tumor methylation (HR, 2.1; 95% CI, 1.1 to 4.0), and vascular space invasion (HR, 1.9; 95% CI, 1.3 to 2.9). In MSS tumors, methylation at any single site was not independently predictive of survival. Neither methylation nor microsatellite status predicted a favorable response to chemotherapy. Conclusion: DNA methylation is associated with a worse outcome in colorectal cancer, but this adverse prognostic influence is lost in those methylated tumors showing MSI. The mechanisms of these events warrant additional investigation.
THE CLINICOPATHOLOGIC staging of colorectal carcinoma currently is based on a combination of the extent of local tumor invasion, lymph node status, and the presence or absence of distant metastasis.1,2 This staging system provides an excellent indication of prognosis and identifies those likely to benefit from adjuvant therapy. Although stage remains the benchmark, there are a number of other pathologic factors known to provide additional prognostic information. Among these are the extent of residual tumor (R classification),3,4 the histologic grade of the tumor, and the presence of vascular space invasion.2 A large number of studies using multivariate analysis have unequivocally demonstrated that tumors of high grade, as well as those with vascular invasion, have worse outcomes independent of tumor stage (reviewed by Compton et al2). Despite this evidence, these pathologic factors are not universally reported, largely because of technical and interpretive problems. During the last two decades, many of the genetic and epigenetic changes that underpin the development of colorectal cancer have been identified. However, the hope that they would be integrated with the routine staging systems has not been realized. For instance, a recent meta-analysis showed that p53 status was of marginal prognostic value and that its use in clinical practice could not be justified.5 Similar conclusions have been drawn from studies of K-ras mutations,6 18q loss,7,8 and overexpression of HER-29 and c-myc10 (reviewed by Pasche et al11). In contrast to this trend, microsatellite instability (MSI) has shown promise as a prognostic marker, at least in select subgroups of patients, such as those with hereditary nonpolyposis colorectal cancer,12 in younger individuals,13 and in individuals with stage II or III disease.14,15 In sporadic colorectal cancers, microsatellite instability arises through methylation of the hMLH1 promoter,16–18 whereas in hereditary nonpolyposis colorectal cancer cancers it occurs through germline mutations of one of the mismatch repair genes.19 The survival advantage accruing in patients with microsatellite unstable tumors seems to occur regardless of the specific mismatch repair gene involved or the mechanism by which it is inactivated. It recently has been shown that the subset of sporadic cancers characterized by hMLH1 methylation and MSI overlaps with a broader group of tumors that display extensive methylation of CpG islands at methylated-in-tumor (MINT) loci and p16.20,21 This larger group of sporadic colorectal cancers, often referred to as CpG island methylator phenotype–positive tumors, shares many of the clinicopathologic features of microsatellite unstable cancers, including right-sided tumor location and mucinous phenotype.21–23 However, more than half of the tumors displaying widespread CpG island methylation are microsatellite stable (MSS).23 It is not known whether these MSS methylated tumors share the improved prognosis of their microsatellite unstable cousins. In this study, we evaluated the prognostic significance of methylation and microsatellite status in 605 consecutive colorectal cancer patients. We also undertook an extensive pathologic assessment of each tumor to establish the importance of these molecular changes in relation to standard clinicopathologic criteria.
Patients and Specimens After providing informed consent, 605 consecutive individuals (with 625 tumors) undergoing complete (RO or R1)1 surgical resection of colorectal cancer at St Vincent’s Hospital (Darlinghurst, Australia) were entered into this prospective study. This study was approved by the Hospital Ethics Committee. Enrollment was from January 1, 1994 to August 1, 2002, and patients were excluded if they had inflammatory bowel disease or a known hereditary colorectal cancer.
Treatment with chemotherapy and/or radiotherapy was recorded (Table 1
Collection and Histopathologic Analysis of Tumors Fresh representative tissue samples (500 µg) from all tumors and paired normal colonic mucosa were immediately frozen at -70°C. Of the 625 colorectal carcinomas collected, 121 (19.3%) were tumor-node-metastasis system stage I, 224 (35.7%) were stage II, 191 (31%) were stage III, and 89 (14%) were stage IV.1 There were 27 individuals with a total of 47 synchronous tumors. In 16 of these individuals, fresh tumor was collected from all lesions (36 tumors), whereas only the most advanced lesion was collected in the remaining 11 patients. Histopathologic assessment of all tumors was performed as previously described.23 Vascular space invasion was defined as the presence of viable tumor cells wholly within an endothelial-lined space or structure, as recognized by light microscopy.
Microsatellite Status
Detection of Methylation of p16, hMLH1, MINT1, MINT2, MINT12, and MINT31 For hMLH1, p16, and each of the MINT loci, methylation status was recorded as either unmethylated, methylated, or not assessable. Positive and negative controls23,30 were included for each reaction. The methylation status of each tumor was reported as the number of sites methylated from a total of five (four MINT loci and p16), and was not assessable in 57 (9%) tumors. A tumor was defined as heavily methylated when more than three of five loci were methylated.21–23
Statistical Analysis
Patient Characteristics The study population (Table 1  ) consisted of 282 females and 323 males, with a mean age of 68.3 ± 12.2 years (range, 29 to 99 years). At the time of resection, 13% of individuals were younger than 55 years, 54% were between 55 and 74 years, and 33% were older than 75 years. After surgical resection of the primary cancer, 53 individuals with rectal cancer were treated with either preoperative or postoperative radiotherapy, whereas 171 (42%) of the individuals with stage II or III cancer received adjuvant chemotherapy. Complete surgical resection of metastases was performed in three patients who had presented with a stage IV tumor, and in an additional eight individuals subsequent to hepatic recurrence. At the censor date of September 1, 2002, 434 patients were alive, 129 had died as a result of cancer, and 42 had died as a result of unrelated causes. The median follow-up period was 32 months (range, 1 to 60 months), with 144 individuals having more than 4 years of follow-up. Disease recurred subsequent to resection in 81 individuals, with a median time to recurrence of 16.6 months (range, 3.2 to 58.8 months).
CpG Island Methylation at Specific Loci
Prognostic Factors for Survival As expected, advanced tumor stage and the presence of vascular space invasion were associated with significantly decreased survival for all patients, as well as for those who had undergone curative resections for nonmetastatic disease (stages I to III; Table 3  ). When all tumor stages were considered, patients with microsatellite unstable tumors had better disease-specific survival than patients with MSS tumors (Table 3). Only nine of 73 (12%) individuals with microsatellite unstable cancers had died within the study period, compared with 120 of 532 individuals (23%) with MSS tumors. At first examination, it seemed that the degree of tumor methylation did not significantly affect disease outcome (Table 3). However, this analysis was confounded by the finding that MSI, which represented the majority of methylated cancers, correlated with both improved survival and with widespread CpG island methylation (P < .0001; Table 4). Examination of survival curves demonstrated that increasing levels of tumor methylation were not related to outcome for the microsatellite unstable tumors but were significantly associated with decreased survival within the MSS cancer group (Fig 1). This latter trend was most apparent with tumors showing methylation at more than three loci but was also noted in tumors with more than two methylated sites. Given the relationship between microsatellite status and methylation, we elected to divide tumors into stable or unstable groups before survival analysis (Fig 2).
For individuals with MSS tumors, Cox’s univariate analysis showed that tumor stage, vascular space invasion, poor differentiation, and tumor methylation were of prognostic significance overall (stages I to IV), as well as in nonmetastatic tumors (stages I to III). The final Cox regression model identified three of these as independent prognostic factors: tumor stage, tumor methylation, and vascular space invasion (Table 5 ).
Methylation of the MINT loci and p16 are closely correlated, and it was considered possible that the adverse effects of methylation of one member of the group could be incorrectly ascribed to tumors displaying a generalized methylator phenotype. To examine this possibility, the prognostic significance of each marker was independently examined. Methylation of MINT1, MINT2, and p16 were unrelated to disease outcome, with hazard ratios (HRs) of 1.1 (95% CI, 0.73 to 1.9), 1.5 (95% CI, 1.0 to 2.3), and 1.4 (95% CI, 0.93 to 2.2), respectively. In univariate analysis, it was evident that methylation of MINT12 and MINT31 was associated with poor prognosis (HR, 1.7; 95% CI, 1.2 to 2.3; P = .01 and HR, 2.0; 95% CI, 1.2 to 3.5; P = .02, respectively). However, methylation at either of these loci was not independently significant when analyzed with stage, vascular space invasion, and grade. With regard to individuals with microsatellite unstable cancers, tumor stage (HR, 6.2; 95% CI, 2.4 to 16.2; P < .0001), right-sided location (HR, 0.13; 95% CI, 0.03 to 0.48; P = .002), and age (HR, 1.0; 95% CI, 1.0 to 1.2; P = .03) were the only factors that correlated with disease outcome. In fact, five of the nine individuals who died as a result of a microsatellite unstable cancer had stage IV disease at presentation, and the mean age of the deceased group (80.5 ± 7.9 years) was significantly older than that of the remaining patients (mean, 70.5 ± 13.6 years; P = .006). Interestingly, vascular space invasion and mucinous phenotype showed no relationship to survival. Analysis of methylation at two regions of the hMLH1 promoter was performed in a subset of tumors (n = 383) in this study. Methylation of both regions of the promoter was strongly associated with MSI and with methylation of the five other loci examined (P < .0001). Eighty percent of the assessable microsatellite unstable tumors displayed methylation of region C, 83% were methylated at region A, and 70% of tumors showed methylation at both regions. There was no association between methylation of either region and survival. Methylation was found less frequently in assessable MSS tumors, with 11% methylated at region A, 23% at region C, and 2.4% at both regions. In individuals with MSS tumors, we noted a trend toward reduced survival in patients with methylation at either region, but this difference was not significant (HR, 1.6; 95% CI, 0.9 to 2.6; P = .09).
Factors Influencing Response to Chemotherapy
In this study, we have used a prospective collection of 625 colorectal cancers to examine the prognostic significance of methylation and MSI. There are a number of well-established pathologic criteria that readily identify prognostic subgroups of colorectal cancer. These criteria were determined in this study to establish a framework against which new molecular information could be assessed. A strength of this study is that it was broadly representative of standard clinical practice. For instance, more than 30% of patients were older than 75 years, a group that is often excluded when studies are performed using material collected in the course of a clinical drug trial. A key finding of this study was that individuals with heavily methylated but MSS tumors had a significantly worse outcome than those with nonmethylated MSS tumors. One explanation for this observation is that widespread methylation is simply a surrogate marker for some other well-established adverse pathologic characteristic. Although this possibility cannot be completely excluded in this study, it seems unlikely, given that the prognostic significance of methylation was shown to be independent of powerful predictors of outcome, such as tumor stage and vascular space invasion. A second possibility is that the presence of methylation obviates the positive effects of adjuvant chemotherapy. Although there are no published data concerning the interactions between CpG island methylation and sensitivity to fluorouracil, methylation of the DNA repair enzyme O(6)-methylguanine DNA methyltransferase has been shown to predict the response of gliomas to alkylating agents.32 Our study was not specifically designed to assess interactions between chemosensitivity and molecular changes. Nevertheless, we found no evidence that tumor methylation was related to disease response in those individuals receiving chemotherapy. It is also possible that methylation, by altering expression of a key gene or group of genes, has altered the biology of the tumor cells in ways that lead to a worse clinical outcome. It is well established that dense hypermethylation of promoter regions can reduce or silence gene expression. In genes such as hMLH1, this is strongly linked to the development of MSI colorectal cancers, which are known to have a less aggressive clinical course.17,33 Hypermethylation of other genes, such as BRCA1,34 E-cadherin,35 p16,36 MGMT,37 and p1438,39 have also been described in a variety of tumors, including colorectal cancer. Although these genes have well-defined cellular functions, their methylation does not necessarily alter gene expression,40,41 and thus influence tumor outcome. Our study included three genes with potentially important cellular functions, namely p16, hMLH1, and MINT31.42 However, we found no association between methylation at p16 or hMLH1 and changes in survival or disease-free recurrence. For MINT31, there was a weak association between methylation and survival in univariate analysis, but this was not an independent prognostic factor. It is thus unlikely that methylation of any one of the individual genes examined in this study was responsible for the poor outcome observed with methylated tumors. Clearly, we have not eliminated the possibility that methylation of one or more other critical genes has occurred and that this process is indirectly associated with methylation of our gene panel. Apart from speculating on its possible causes, the decreased survival seen in methylated MSS cancer group is of interest for at least one other reason. It has recently been suggested that although the extent of methylation is distributed widely among cancers, those tumors with the greatest degree of methylation also show preferential methylation of particular groups of CpG islands.43 Thus the assignment of some colorectal cancers into a hypermethylator group may reflect qualitative as well as quantitative changes in DNA methylation. Certainly, the finding of this study that heavily methylated cancers have a distinctive clinical course tends to support the biologic validity of this grouping. The second key finding of this study relates to prognosis in microsatellite unstable cancers. Approximately 12% of the tumors in this study demonstrated MSI, and in univariate analysis, these tumors were associated with a modest improvement in overall survival that did not reach statistical significance. In a study of 587 individuals younger than 50 years of age, MSI has previously been found to be an independent prognostic factor.13 Likewise, others have found MSI to be an important prognostic factor in specific stages of tumor.14 However, several studies have detected only a weak survival advantage44,45 or no significant survival advantage.46–48 As expected, many of the microsatellite unstable tumors displayed methylation of hMLH1, MINT loci, or p16,21,23 yet in striking contrast to MSS tumors, the extent or even presence of methylation in these tumors had no impact on outcome. It has been proposed that methylated microsatellite unstable and MSS tumors arise from similar precursor lesions that are susceptible to a process of CpG island methylation.49 Furthermore, there are striking similarities between tumors showing CpG island methylation, irrespective of their microsatellite status.23 Such tumors have a propensity to develop on the right side of the colon in older women and often display a mucinous phenotype. If this is the case, then it seems that the development of MSI may act as an antidote to the adverse prognostic effects of widespread methylation. Indeed, the apparently minor survival gains associated with the MSI phenotype become considerably more impressive when related to the outcome of methylated MSS cancers. The mechanisms underlying the improved survival seen in microsatellite unstable cancers remain unknown. It is possible they relate directly to instability, with MSI presumably making cells less fit to progress, or at least to metastasize. This is somewhat paradoxical,50 given that instability, either microsatellite or chromosomal in nature, is often invoked as a mechanism driving tumor progression. An alternate hypothesis for improved survival in microsatellite unstable tumors relates to immune mechanisms. Microsatellite unstable tumors typically display an intense intraepithelial infiltrate with CD8+CD103+ lymphocytes, and it is possible that these T cells play a role in modulating disease progression.51 In light of these data, it is interesting to note that the intraepithelial and peritumoral lymphocytic infiltrate typical of microsatellite unstable tumors is less common in MSS tumors with heavy methylation.23,52 The clinical implications of the findings of this study remain uncertain. It is premature to suggest that methylation or microsatellite DNA analysis should form part of the standard assessment of all colorectal cancers. These techniques cannot be justified in routine practice on the basis of either economics or clinical need. However, methylated tumors, both stable and unstable, have relatively distinct clinicopathologic features that can be recognized by pathologists and clinicians alike.53 Immunoperoxidase staining for the mismatch repair proteins may supplement standard histopathologic assessment of tumors to support a clinical suspicion of microsatellite instability. The important practical implication of this study is, however, that there is a close pathologic similarity between microsatellite unstable tumors and methylated MSS lesions. Given the significantly different prognosis for these two groups of cancers, it is important that clinicians and pathologists be aware of this distinction. In conclusion, this study shows that methylation status is an independent prognostic factor in colorectal carcinoma and highlights two important mechanistic questions: why do methylated MSS cancers behave so poorly, and how does the development of microsatellite instability so powerfully reverse this behavior? The answers to these questions may be of considerable relevance to the biology and therapeutics of colorectal cancer.
The authors indicated no potential conflicts of interest.
The authors acknowledge Rachael Williams for her assistance in the collection and verification of family history and survival data; Emma Quinn for microsatellite assays; Matthew Law, PhD, for advice on statistics; and Jenny Turner, MD, for review of histopathologic specimens.
1. Hermanek P, Sobin LH: TNM: American Joint Committee on Cancer—Manual for Staging of Cancer. Philadelphia, PA, Springer Verlag, 1998 2. Compton C, Fenoglio-Preiser CM, Pettigrew N, et al: American Joint Committee on Cancer Prognostic Factors Consensus Conference: Colorectal Working Group. Cancer 88:1739–1757, 2000[CrossRef][Medline] 3. Hermanek P, Wittekind C: Residual tumor (R) classification and prognosis. Semin Surg Oncol 10:12–20, 1994[Medline] 4. Newland RC, Dent OF, Chapuis PH, et al: Clinicopathologically diagnosed residual tumor after resection for colorectal cancer: A 20-year prospective study. Cancer 72:1536–1542, 1993[CrossRef][Medline] 5. Petersen S, Thames HD, Nieder C, et al: The results of colorectal cancer treatment by p53 status: Treatment-specific overview. Dis Colon Rectum 44:322–333, 2001[CrossRef][Medline] 6. Andreyev HJ, Norman AR, Cunningham D, et al: Kirsten ras mutations in patients with colorectal cancer: The "RASCAL II" study. Br J Cancer 85:692–696, 2001[CrossRef][Medline]
7. Jen J, Kim H, Piantadosi S, et al: Allelic loss of chromosome 18q and prognosis in colorectal cancer. N Engl J Med 331:213–221, 1994 8. Dix BR, Robbins P, Soong R, et al: The common molecular genetic alterations in Dukes’ B and C colorectal carcinomas are not short-term prognostic indicators of survival. Int J Cancer 59:747–751, 1994[Medline] 9. Sun XF, Carstensen JM, Stal O, et al: c-erbB-2 oncoprotein in relation to DNA ploidy and prognosis in colorectal adenocarcinoma. APMIS 103:309–315, 1995[Medline]
10. Augenlicht LH, Wadler S, Corner G, et al: Low-level c-myc amplification in human colonic carcinoma cell lines and tumors: A frequent, p53-independent mutation associated with improved outcome in a randomized multi-institutional trial. Cancer Res 57:1769–1775, 1997 11. Pasche B, Mulcahy M, Benson AB III: Molecular markers in prognosis of colorectal cancer and prediction of response to treatment. Best Pract Res Clin Gastroenterol 16:331–345, 2002[CrossRef][Medline] 12. Watson P, Lin KM, Rodriguez-Bigas MA, et al: Colorectal carcinoma survival among hereditary nonpolyposis colorectal carcinoma family members. Cancer 83:259–266, 1998[CrossRef][Medline]
13. Gryfe R, Kim H, Hsieh ET, et al: Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 342:69–77, 2000 14. Wright CM, Dent OF, Barker M, et al: Prognostic significance of extensive microsatellite instability in sporadic clinicopathological stage C colorectal cancer. Br J Surg 87:1197–1202, 2000[CrossRef][Medline]
15. Elsaleh H, Powell B, McCaul K, et al: P53 alteration and microsatellite instability have predictive value for survival benefit from chemotherapy in stage III colorectal carcinoma. Clin Cancer Res 7:1343–1349, 2001
16. Veigl ML, Kasturi L, Olechnowicz J, et al: Biallelic inactivation of hMLH1 by epigenetic gene silencing, a novel mechanism causing human MSI cancers. Proc Natl Acad Sci U S A 95:8698–8702, 1998
17. Herman JG, Umar A, Polyak K, et al: Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci U S A 95:6870–6875, 1998 18. Furukawa T, Konishi F, Masubuchi S, et al: Densely methylated MLH1 promoter correlates with decreased mRNA expression in sporadic colorectal cancers. Genes Chromosomes Cancer 35:1–10, 2002[CrossRef][Medline] 19. Menigatti M, Di Gregorio C, Borghi F, et al: Methylation pattern of different regions of the MLH1 promoter and silencing of gene expression in hereditary and sporadic colorectal cancer. Genes Chromosomes Cancer 31:357–361, 2001[CrossRef][Medline]
20. Ahuja N, Mohan AL, Li Q, et al: Association between CpG island methylation and microsatellite instability in colorectal cancer. Cancer Res 57:3370–3374, 1997
21. Toyota M, Ahuja N, Ohe-Toyota M, et al: CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A 96:8681–8686, 1999
22. Toyota M, Ohe-Toyota M, Ahuja N, et al: Distinct genetic profiles in colorectal tumors with or without the CpG island methylator phenotype. Proc Natl Acad Sci U S A 97:710–715, 2000 23. Hawkins N, Norrie M, Cheong K, et al: CpG island methylation in sporadic colorectal cancers and its relationship to microsatellite instability. Gastroenterology 122:1376–1387, 2002[CrossRef][Medline] 24. Moertel CG, Fleming TR, Macdonald JS, et al: Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 322:352–358, 1990[Abstract]
25. O’Connell MJ, Mailliard JA, Kahn MJ, et al: Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 15:246–250, 1997
26. Wolmark N, Rockette H, Fisher B, et al: The benefit of leucovorin-modulated fluorouracil as postoperative adjuvant therapy for primary colon cancer: Results from National Surgical Adjuvant Breast and Bowel Project protocol C-03. J Clin Oncol 11:1879–1887, 1993
27. Glimelius B: Radiotherapy in rectal cancer. Br Med Bull 64:141–157, 2002
28. Ward R, Meagher A, Tomlinson I, et al: Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut 48:821–829, 2001
29. Herman JG, Graff JR, Myohanen S, et al: Methylation-specific PCR: A novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci U S A 93:9821–9826, 1996 30. Suter C, Norrie M, Ku S, et al: CpG island methylation is a common finding in colorectal cancer cell lines. Br J Cancer 88:413–419, 2003[CrossRef][Medline]
31. Deng G, Chen A, Hong J, et al: Methylation of CpG in a small region of the hMLH1 promoter invariably correlates with the absence of gene expression. Cancer Res 59:2029–2033, 1999
32. Esteller M, Garcia-Foncillas J, Andion E, et al: Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 343:1350–1354, 2000
33. Kane MF, Loda M, Gaida GM, et al: Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res 57:808–811, 1997
34. Esteller M, Silva JM, Dominguez G, et al: Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst 92:564–569, 2000
35. Tamura G, Yin J, Wang S, et al: E-cadherin gene promoter hypermethylation in primary human gastric carcinomas. J Natl Cancer Inst 92:569–573, 2000
36. Esteller M, Gonzalez S, Risques RA, et al: K-ras and p16 aberrations confer poor prognosis in human colorectal cancer. J Clin Oncol 19:299–304, 2001
37. Esteller M, Toyota M, Sanchez-Cespedes M, et al: Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase by promoter hypermethylation is associated with G to A mutations in K-ras in colorectal tumorigenesis. Cancer Res 60:2368–2371, 2000
38. Zheng S, Chen P, McMillan A, et al: Correlations of partial and extensive methylation at the p14(ARF) locus with reduced mRNA expression in colorectal cancer cell lines and clinicopathological features in primary tumors. Carcinogenesis 21:2057–2064, 2000
39. Esteller M, Tortola S, Toyota M, et al: Hypermethylation-associated inactivation of p14(ARF) is independent of p16(INK4a) methylation and p53 mutational status. Cancer Res 60:129–133, 2000
40. Fearon ER: BRCA1 and E-cadherin promoter hypermethylation and gene inactivation in cancer-association or mechanism? J Natl Cancer Inst 92:515–517, 2000
41. Costello JF, Plass C: Methylation matters. J Med Genet 38:285–303, 2001
42. Toyota M, Ho C, Ohe-Toyota M, et al: Inactivation of CACNA1G, a T-type calcium channel gene, by aberrant methylation of its 5' CpG island in human tumors. Cancer Res 59:4535–4541, 1999 43. Costello JF, Fruhwald MC, Smiraglia DJ, et al: Aberrant CpG-island methylation has non-random and tumour-type-specific patterns. Nat Genet 24:132–138, 2000[CrossRef][Medline]
44. Halling KC, French AJ, McDonnell SK, et al: Microsatellite instability and 8p allelic imbalance in stage B2 and C colorectal cancers. J Natl Cancer Inst 91:1295–1303, 1999 45. Elsaleh H, Joseph D, Grieu F, et al: Association of tumour site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. Lancet 355:1745–1750, 2000[CrossRef][Medline] 46. Johannsdottir JT, Bergthorsson JT, Gretarsdottir S, et al: Replication error in colorectal carcinoma: Association with loss of heterozygosity at mismatch repair loci and clinicopathological variables. Anticancer Res 19:1821–1826, 1999[Medline] 47. Salahshor S, Kressner U, Fischer H, et al: Microsatellite instability in sporadic colorectal cancer is not an independent prognostic factor. Br J Cancer 81:190–193, 1999[CrossRef][Medline] 48. Messerini L, Ciantelli M, Baglioni S, et al: Prognostic significance of microsatellite instability in sporadic mucinous colorectal cancers. Hum Pathol 30:629–634, 1999[CrossRef][Medline]
49. Hawkins NJ, Ward RL: Sporadic colorectal cancers with microsatellite instability and their possible origin in hyperplastic polyps and serrated adenomas. J Natl Cancer Inst 93:1307–1313, 2001 50. Tomlinson I, Bodmer W: Selection, the mutation rate and cancer: Ensuring that the tail does not wag the dog. Nat Med 5:11–12, 1999[CrossRef][Medline] 51. Quinn E, Hawkins N, Yip L, et al: CD103+ intraepithelial lymphocytes: A unique population in microsatellite unstable sporadic colorectal cancer. Eur J Cancer 39:469–475, 2003[CrossRef][Medline]
52. Whitehall VL, Wynter CV, Walsh MD, et al: Morphological and molecular heterogeneity within nonmicrosatellite instability-high colorectal cancer. Cancer Res 62:6011–6014, 2002 53. Smyrk TC, Watson P, Kaul K, et al: Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma. Cancer 91:2417–2422, 2001[CrossRef][Medline] Submitted March 19, 2003; accepted July 30, 2003. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 test or the Fisher’s exact test. For individuals with synchronous tumors, the highest stage lesion alone was used in survival analysis. Survival was measured from the date of resection of colorectal cancer to the date of death, the completion of 5 years of follow-up, or the last clinical review before September 1, 2002. Only cancer-related deaths were analyzed as events. Time to recurrence was the interval from surgery to documented tumor recurrence. Differences between Kaplan-Meier survival curves were tested using the log-rank test. Univariate survival analysis was performed by Cox proportional hazards regression model, and multivariate analysis was used to determine independent prognostic factors. A P value of less than .05 was considered significant. All data were analyzed using SPSS statistical software, Version 11.0 (SPSS Inc, Chicago, IL). 
