- Email: [email protected]
RAF mutation and defective DNA mismatch repair in endometrial cancers
American Journal of Obstetrics and Gynecology (2004) 190, 935e42
www.elsevier.com/locate/ajog
RAS/RAF mutation and defective DNA mismatch repair in endometrial cancers David G. Mutch, MD,a Matthew A. Powell, MD,a Mary Ann Mallon, BS,b Paul J. Goodfellow, PhDb,c Division of Gynecologic Oncology,a Departments of Surgeryb and Genetics and Obstetrics and Gynecology,c Washington University School of Medicine, St. Louis, Mo
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS BRAF mutation KRAS2 mutation Colon Endometrial cancer Microsatellite instability
Objective: Defective DNA mismatch repair is a common genetic abnormality in both colon cancers and endometrial cancers. Cancers with defective DNA mismatch repair have the so-called mutator phenotype and accumulate genetic errors at an increased rate. An early mutational target in cells with defect DNA mismatch repair may be the RAS/RAF pathway. Colon cancers often have KRAS2 mutations and, if not KRAS2 mutations, may have BRAF mutations. This study investigated the spectrum and frequency of mutations in BRAF and KRAS2 in endometrial carcinomas on the basis of mismatch repair status. Study design: Four hundred forty-one patients with endometrial cancer were staged properly and graded and evaluated for mismatch repair status. These patients were then stratified to groups by the degree of microsatellite instability that was observed in their tumors. One hundred forty-six of the selected tumors were then evaluated for KRAS2 and BRAF mutations on the basis of their microsatellite instability. Results: One hundred forty-six endometrioid endometrial cancers were evaluated for KRAS2 and BRAF mutations. Thirty-five cancers (24%) had activating KRAS2 mutations, but only a single BRAF mutation was identified in an microsatellite instabilityepositive cancer. Twenty-four of 81 microsatellite instability high cancers (29.6%) in which the MLH1 repair gene was methylated had KRAS2 mutations. When compared with the other groups, this finding approached statistical significance (P = .06). KRAS2 mutation status was associated with increasing age at diagnosis (P = .02). Conclusion: Despite many similarities between colon and endometrial cancers, the mechanism of the development of endometrial cancers appears to be different from colon cancers in that BRAF is not affected by a mismatch repair problem, because only KRAS2 mutations were seen. In addition, increasing age appears to lead to an increased likelihood that such a mutation will occur. Ó 2004 Elsevier Inc. All rights reserved.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Presented at the Twenty-Second Annual Meeting of the American Gynecological and Obstetrical Society, Napa, California, September 18-20, 2003. Supported by grant CA71754 from the National Institutes of Health. 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.01.017
Reprint requests: David G. Mutch, MD, Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Washington University School of Medicine, 4911 Barnes Hospital Plaza, St. Louis, MO 63110. E-mail: [email protected]
936
Figure 1 Selection of cases for KRAS2 and BRAF mutation analysis. The tumors that were investigated came from our patient series, which was characterized previously for tumor MSI and MLH1 methylation.9 One hundred sixteen of 127 MSI-H cases in the series (91%) were studied for KRAS2 and BRAF mutation along with 10 MSI-L and 30 MSS tumors. The asterisk denotes unsuccessful mutation analyses for 11 MSI-H M tumors (repeat failures with R1 of the 5 PCR reactions for KRAS2 and BRAF mutation analysis).
Cancers arise because of the accumulation of alterations in genes that control DNA repair, cell growth, proliferation, differentiation, and death. The molecular basis of the stepwise progression from a normal to malignant cell is perhaps best described in cancers of the colon.1 Endometrial cancers are associated with colon cancer and are the second most common cancer to be associated with hereditary nonpolyposis colon cancer (HNPCC).2-4 Furthermore, endometrial cancers share many of the molecular characteristics of colon cancers, which is evident from the finding that the 2 types of cancers frequently exhibit a DNA mismatch repair abnormality.4-6 This can occur through a germline mutation or somatic mutation or through hypermethylation of promoter regions that effectively silence the gene.6-10 Thus, both types of cancers, colon and endometrial, are linked by mismatch repair, and both tumor types often demonstrate the so-called mutator phenotype.2-4,10-13 Similarly, both colon and endometrial cancers share the trait that mutation in the KRAS2 gene is a common and early event in the genesis of their uncontrolled growth.14-16 It has been shown that KRAS2 mutations occur predominantly during the transformation of a small-to-intermediate sized adenoma in colon cancers and in the transition of hyperplasia to endometrial carcinoma.14,16,17 It is unclear how closely these mutations are linked to mismatch repair. However, the RAS/RAF family does seem to be an early target for this type of abnormality.16-18 Genes of the RAF family encode kinases that are important regulators of cellular response to growth signals. Protein products of these genes participate in the RASRAF-MEK-ERK-MAP kinase pathway.19-21 There are 3 RAF genes, each of which code for a threonine kinase. The RAF function is regulated by RAS.19,21 Colorectal cancers often have mutations in the RAS family; when this occurs, it is uncommon forO1 gene that is involved with the expression pathway to be disrupted.18,22 That is
Mutch et al to say that, when RAS is mutated, BRAF is not, and similarly, when BRAF is mutated, RAS is not. This makes sense because there is no need forO1 activating mutation to allow the cancer cell to progress to the next step toward malignancy. Given the similarity of mismatch repair and RAS mutations in colon and endometrial cancers, we investigated the occurrence and spectrum of KRAS2 and BRAF mutations in endometrial cancers of known DNA mismatch repair status.
Material and methods Patient and specimen selection This study was approved by the Human Studies Committees; an informed written consent was obtained from all participants. Tissue specimens from 441 women with primary endometrial cancers were collected at the time of treatment by hysterectomy, and FIGO surgical stage was assigned. Mixed Mu¨llerian tumors and other sarcomas were excluded from this analysis. The cases that were analyzed for KRAS2 and BRAF mutations were selected from 441 tumors on the basis of the results of microsatellite instability (MSI) analysis (Figure 1). All cases with MSI were evaluated, and a representative subset of 30 microsatellite stable (MSS) tumors was analyzed. MSI and MLH1 promoter methylation analyses were preformed as previously described.9 A tumor was classified as MSI-high (MSI-H) if R2 markers showed instability, MSS if no instability was noted, and MSIlow (MSI-L) if a single marker revealed novel bands. MSI-H with hMLH1 promoter methylation was designated as MSI-H M; those without promoter methylation were designated MSI-H U.
KRAS2 and BRAF mutation analysis Codons 12 and 13 of KRAS2 were evaluated for mutation with the use of previously described polymerase chain reaction (PCR) amplification/restriction digestion assays.23 Codon 61 was evaluated by single-strand conformational variant analysis, as described elsewhere.15 The codon 12, 13, and 61 mutations were determined by the direct sequencing of the amplified PCR products (codons 12 and 13 in exon 1 and codon 61 in exon 2), as reported previously by our group.15,16 BRAF exons 11 and 15 were PCR amplified and subjected to single-strand conformational variant analysis on MDE (FMC Bioproducts, Rockland, Maine) gels with and without 5% glycerol. In addition to the singlestrand conformational variant analyses, PCR products from 23 tumors were sequenced directly. The primers that were used for amplification and sequencing were as follows: BRAF exon 11: Forward 5# TTTCTTAAGGGGATCTCTTCCTG 3#, Reverse 5# CCGACTGCTGTGAACAGTTTT 3#; BRAF exon 15: Forward 5#
937
Mutch et al
Figure 2 Frequency of KRAS2 mutation according to molecular classification. The MSI-H M tumor group had the highest rate of KRAS2 mutation, but the rate of mutation was not significantly different than for the other 3 classes.
Table KRAS2 mutation by grade and stage in uterine endometrioid adenocarcinomas Endometrioid Adenocarcinoma KRAS2 Mutation Status Grade 1 2 3 Stagey I II III IV
Positive (No.)
Negative* (No.)
14 (21.5%) 18 (29%) 3 (16%)
51 (78.5%) 44 (71%) 16* (84.2%)
26 2 6 0
(24.5%) (18.2%) (27.3%) (0%)
80 9 16 4
(75.5%) (81.8%) (72.7%) (100%)
* 10 cases in series were clear cell or papillary carcinomas and not included in this analysis (3 Stage II, 6 Stage III, and 1 Stage IV). y Incomplete staging for 1 case with KRAS2 mutation and 2 cases without mutation.
TGCTTGCTCTGATAGGAAAATG 3# and Reverse 5# TC AGGGCCAAAAATTTAATCA 3#.
Statistical analysis Fisher’s exact test, the Student t test, and the MannWhitney test were performed with InStat version for Macintosh (GraphPad Software Inc, San Diego, Calif). A probability value of !.05 was considered statistically significant.
Results Of 156 endometrial cancers that were evaluated for KRAS2 and BRAF mutation, 146 cancers were of endometrioid histologic condition and 10 were of clear-cell or papillary serous histologic condition. The 10 clear cell
Figure 3 BRAF intron 11 mutation that was identified in MSI-H U tumor 1184. The arrow indicates the insertion of a T at position C74 from the 3# end of exon 11.
and papillary serous were eliminated from this evaluation. Of the remaining 146 specimens, 35 specimens (24%) had activating KRAS2 mutations, although a single case had a BRAF mutation of uncertain functional significance. Most of the KRAS2 mutations (29/35) were in codon 12. Five mutations were in codon 13, and 1 mutation was in codon 61. Thirty-four of the 35 tumors with KRAS2 mutation were of the endometrioid histologic subtype, and 1 case (1411) was a mixed endometrioid/ serous carcinoma. For case 1411, the mutation analysis was performed for the endometrioid component that made up most of the tumor. The rate of KRAS2 mutation varied with tumor molecular class. Twenty-four of 81 tumors (29.6%) in the MSI-H M group had KRAS2 mutation, whereas 4 of 35 MSI-H U tumors (11.4%), 1 of 10 MSI-L tumors (10%), and 6 of 30 MSS tumors (20%) had mutation (Figure 2). The difference in the rate of mutation between the MSI-H M and the MSI-H U groups approached statistical significance (P = .06; Fisher’s exact test). It is noteworthy that all 4 MSI-H U tumors with KRAS2 mutations arose in women with inherited mutations in DNA mismatch repair. Tumor 1150, which has a codon 12 Asp mutation, is from a case with an MSH2 mutation.10 Tumor 1319 (codon 12 Ala mutation), tumor 1524 (codon 12 Asp mutation), and tumor 1401 (codon 13 Asp) all arose in MSH6 mutation carriers.9 KRAS2 mutation was not associated with either tumor grade or stage (Table). Mutation status, however, was associated with increased age at diagnosis. The mean and median age at diagnosis for the 35 patients with tumors-harboring mutations were 68.3 and 70.1 years, respectively; the mean and median ages for the 121 mutation-negative cases were 63.2 and 62.2 years, respectively. The 5.1-year increase in mean ages and 7.9-year increase in median ages of the patients with tumor KRAS2 mutations were statistically significant (P = .02; 2-sided Student t test; and P = .02; MannWhitney test).
938 A single example of BRAF mutation was identified among 146 endometrioid cases that were evaluated for sequence alterations in exons 11 and 15. An insertion mutation (insert T) in intron 11 was seen in the MSIpositive cancer case 1184 (Figure 3). Direct sequencing of amplicons that corresponded to exons 13 and 17 from 19 tumors revealed intronic polymorphisms, which indicated that the direct sequencing of tumor-derived PCR products reliably could reveal sequence alterations (data not shown).
Comment The results of this study demonstrate that the path to malignant transformation in endometrial cancers appears to be different than that of colon cancers, despite other similarities. There were no significant BRAF mutations in tumors with defective DNA mismatch repair. On the basis of the higher KRAS2 mutation rate in MSI-positive cancers, it appears that the inability to repair mismatch errors may contribute to KRAS2 mutation Also, increasing age seems to increase the likelihood that a mutation will occur. The rate of KRAS2 mutation that was observed in this study parallels what has been reported previously. Lax et al24 made the observation that KRAS2 mutation is a molecular feature that distinguishes endometrioid endometrial cancers from serous carcinomas. The rate of KRAS2 mutation in our series (35/146 mutations [24%]) is similar to the 26% rate for codon 12 mutation that was reported by Lax et al24 for endometrioid carcinomas. In our series, we saw a significant increase in the age at diagnosis of women with tumors with KRAS2 mutations compared with mutation-negative cases. Because our series was enriched for tumors with MSI and because our previous demonstration that the MSI-H M tumor phenotype is associated with increased age at diagnosis,9 we compared the age at diagnosis of mutation-positive and mutation-negative cases among the MSI-H M group. The higher age at diagnosis was also observed in this subset of patients. The mean and median ages of mutationpositive cases were 70.4 and 72.0 years, respectively; the mean and median ages of mutation-negative cases were 66.4 and 66.4 years, respectively. The increase in mean and median age approached statistical significance (P = .06; 1-sided Student t test; and P = .06; MannWhitney test). Further studies will be required to determine whether KRAS2 mutation is a feature of late age of onset in endometrial carcinoma. The absence of BRAF mutations in MSI-positive endometrial cancers was unexpected. Previous studies of colorectal neoplasms with MSI suggested that tumors with defective DNA mismatch repair frequently have BRAF mutations.18 In a series of 330 colorectal tumors that were investigated for KRAS2 and BRAF mutation,
Mutch et al 32 tumors (10%) had BRAF defects, and 169 tumors (51%) had KRAS2 mutations. BRAF and KRAS2 mutations were not seen in the same tumor, which is consistent with the defects that exert equivalent effects in tumor formation.22 The rate of BRAF mutation was much higher in tumors with MSI; 15 of 49 MSI-positive tumors had BRAF mutations. Because both endometrial and colorectal cancer show frequent MSI and KRAS2 mutation, we reasoned that a similar relationship might exist in MSI-positive endometrial cancers. Our studies have demonstrated that endometrial cancers differ from colorectal tumors in that BRAF mutations are not seen in MSI-positive cases. One possible explanation for the absence of BRAF mutation in MSI-positive endometrial cancers relates to the relative roles that defective DNA mismatch repair and the RAS/RAF pathway play in tumor formation in the 2 different tissues. In our series, 127 of 441 cases (29%) were MSI-positive (Figure 1); only 49 of 330 of the colorectal tumors (15%) that were studied by Rajagopalan et al18 were MSI-positive. Based on the higher rate of MSI in endometrial cancers, we speculate that the loss of DNA mismatch repair plays a more important role in endometrial tumorigenesis than in colorectal tumorigenesis. On the other hand, the rate of KRAS2 mutation in our series of endometrial cancers (35/146 cancers [24%]) is less than in the colorectal tumors that were studied for KRAS2 and BRAF mutations. Overall, 51% of colorectal tumors (169/330) had KRAS2 mutations. Because we investigated a selected subset of endometrial cancers, which were enriched for cases with defective DNA mismatch repair, the most direct comparisons of KRAS2 and BRAF mutation rates can be made between MSI-positive endometrial and colorectal tumors. Twenty-one of 49 of MSI-positive colorectal tumors (43%) had KRAS2 mutation18; only 28 of 116 of MSI-positive endometrial cancers (24%) had KRAS2 mutation. Our studies and the studies of Rajagopalan et al18 suggest that KRAS mutations are nearly twice as frequent in colorectal tumors than in endometrial cancers with MSI. The relative role that BRAF plays in these 2 groups of tumors, however, is very different. No examples of activating BRAF mutation were found among 116 MSI-positive endometrial cancers, whereas 21 of 49 MSI-positive colorectal cancers had BRAF mutations. Although colorectal and endometrial cancers share the common features of MSI and KRAS2 mutation, they differ with respect to the involvement of the RAS/RAF pathway, which is evidenced by the lower rate or KRAS2 mutation and the absence of activating BRAF mutations in endometrial cancers.
Acknowledgments We thank Ms Erin Ball for her assistance with the preparation of this manuscript.
939
Mutch et al
References 1. Feron ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1995;61:197-205. 2. Watson P, Lynch HT. Extracolonic cancer in the hereditary nonpolyposis colorectal cancer. Cancer 1993;71:679-85. 3. Watson P, Vasen HFA, Mecklin JP, Jarvinen H, Lynch HT. The risk of endometrial cancer in hereditary nonpolyposis colorectal cancer. Am J Med 1994;96:516-20. 4. Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshelman JR, Burt RW, et al. A national cancer institute workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998;58:5248-57. 5. Kowalski LD, Mutch DG, Herzog TJ, Rader JS, Goodfellow PJ. Mutational analysis of MLH1 and MSH2 in 25 prospectively acquired RERC endometrial carcinomas. Genes Chromosomes Cancer 1997;18:219-27. 6. Katabuchi H, van Rees B, Lambers AR, Ronnett BM, Blazes MS, Leach FS, et al. Mutations in DNA mismatch repair genes are not responsible for microsatellite instability in most sporadic endometrial carcinomas. Cancer Res 1995;55:5556-60. 7. Estellar M, Levine R, Baylin SB, Ellenson LH, Herman JG. MLH1 promoter hypermethylation is associated with the microsatellite instability phenotype in sporadic endometrial carcinomas. Oncogene 1999;16:2413-7. 8. Simpkins SB, Swisher EM, Mutch DG, Gersell DJ, Kovatich AJ, Palazzo JP, et al. MLH1 promoter methylation and gene slicing is the primary cause of microsatellite instability in sporadic endometrial cancers. Hum Mol Genet 1999;8:661-6. 9. Goodfellow PJ, Buttin BM, Herzog TJ, Rader JS, Gibb RK, Swisher E, et al. Prevalence of defective DNA mismatch repair and MSH6 mutation in an unselected series of endometrial cancers. Proc Natl Acad Sci USA 2003;100:5908-13. 10. Whelan A, Babb S, Mutch DG, Rader J, Herzog TJ, Todd C, et al. MSI in endometrial carcinoma: absence of MLH1 promoter methylation is associated with increased familial risk for cancers. Int J Cancer 2002;99:697-704. 11. Wijnin J, De Leeuw W, Vasen H, van der Klift H, Moller P, Stormorken A, et al. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Gen 1999;23:142-4. 12. Vasen HFA, Wijnen JT, Menko FH, Kleibeuker JH, Tall BG, Giffioen G, et al. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. Gastroenterology 1996;110:1020-7. 13. Millar AL, Pal T, Madlensky L, Sherman C, Temple L, Mitri A, et al. Mismatch repair gene defects contribute to the genetic basis of double primary cancers of the colorectum and endometrium. Hum Mol Genet 1999;9:823-9. 14. Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319:525-32. 15. Swisher EM, Peiffer-Schneider S, Mutch DG, Herzog TJ, Rader JS, Eibenaary A, et al. Differences in patterns of TP53 and KRAS2 mutations in a large series of endometrial carcinomas with or without microsatellite instability. Cancer 1999;85:119-26. 16. Cohn DE, Mutch DG, Herzog TJ, Rader JS, Dintzis SM, Gersell DJ, et al. Genotypic and phenotypic progression in endometrial tumorigenesis: determining when defects in DNA mismatch repair and KRAS2 occur. Genes Chromosomes Cancer 2001;32: 295-301. 17. Yuen TS, Davies H, Chan TL, Ho JW, Bignell GR, Cox C, et al. Similarity of the phenotypic patterns associated with BRAF and KRAs mutations in colorectal neoplasia. Cancer Res 2002;62: 6451-5.
18. Rajagopalan H, Bardelli A, Lengauer C, Kinzler KW, Vogelstein B, Velculescu VE. RAF/RAS oncogenes and mismatch-repair status. Nature 2002;418:934-7. 19. Peyssonnaux C, Eychene A. The Raf/MEK/ERK pathway: new concepts of activation. Biol Cell 2001;93:53-62. 20. Avruch JA. Ras activation on the Raf kinase:tyrosine kinase recruitment of the MAP kinase cascade. Recent Prog Horm Res 2001;56:127-55. 21. Kolch W. Meaningful relationships: the regulation of the Ras/Raf/ MEK/ERK pathway by protein interactions. Biochem J 2000;351: 289-305. 22. Davies H, Bignell GR, Cox C, Stephans P, Edkinss S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417: 949-54. 23. Lin SY, Chen PH, Wang CK, Liu JD, Slauw CP, Chen YJ, et al. Mutation analysis of K-ras oncogenes in gastroenterologic cancers by the amplified created restriction sites method. Am J Clin Pathol 1993;100:686-9. 24. Lax SF, Kendall B, Tashiro H, Slebos RJ, Hedrick L. The frequency of p53, K-ras mutations, and microsatellite instability differs in uterine endometrioid and serous carcinoma: evidence of distinct molecular genetic pathways. Cancer 2000;88:814-24.
Discussion DR DAVID M. GERSHENSON, Houston Tex. In 1913, Warthin1 first reported a clustering of colorectal cancers in ‘‘Family G.’’ Eighty years later, HNPCC, or the Lynch syndrome, crystallized.2 HNPCC is an autosomal dominantly inherited cancer susceptibility syndrome that is characterized by the development of early-onset colorectal cancer and extracolonic cancers (endometrium, gastrointestinal tract, genitourinary tract, ovary, and brain). Colorectal cancer is the most common malignancy in the syndrome, which accounts for approximately 5% of all colorectal cancers. The most common extracolonic cancer is endometrial cancer, which is the subject of this article. The diagnosis of HNPCC is based solely on family history, with the use of the internationally accepted ‘‘Amsterdam criteria.’’3,4 Included in the criteria are the requirements for at least 3 family members with a histologically verified HNPCC-associated cancer. In addition, affected individuals must span at least 2 generations; 1 individual must be a first-degree relative of the other 2 individuals, and 1 case must be diagnosed before age 50 years. A major clue to heighten one’s suspicion of the HNPCC syndrome is multiple tumors, either synchronous or metachronous, in the same individual. The molecular basis of HNPCC is a germline defect in mismatch repair.5,6 The primary function of the mismatch repair system is to eliminate single-base mismatches and insertion-deletion loops that may occur during DNA replication.7 To date, germline mutations in 1 of 4 major mismatch repair genes (MLH1, MSH2, MSH6, and PMS2) have been detected in up to 80% of HNPCC families.8 Two other putative mismatch repair gene mutations (MLH3 and EXO1) have been
www.elsevier.com/locate/ajog
RAS/RAF mutation and defective DNA mismatch repair in endometrial cancers David G. Mutch, MD,a Matthew A. Powell, MD,a Mary Ann Mallon, BS,b Paul J. Goodfellow, PhDb,c Division of Gynecologic Oncology,a Departments of Surgeryb and Genetics and Obstetrics and Gynecology,c Washington University School of Medicine, St. Louis, Mo
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS BRAF mutation KRAS2 mutation Colon Endometrial cancer Microsatellite instability
Objective: Defective DNA mismatch repair is a common genetic abnormality in both colon cancers and endometrial cancers. Cancers with defective DNA mismatch repair have the so-called mutator phenotype and accumulate genetic errors at an increased rate. An early mutational target in cells with defect DNA mismatch repair may be the RAS/RAF pathway. Colon cancers often have KRAS2 mutations and, if not KRAS2 mutations, may have BRAF mutations. This study investigated the spectrum and frequency of mutations in BRAF and KRAS2 in endometrial carcinomas on the basis of mismatch repair status. Study design: Four hundred forty-one patients with endometrial cancer were staged properly and graded and evaluated for mismatch repair status. These patients were then stratified to groups by the degree of microsatellite instability that was observed in their tumors. One hundred forty-six of the selected tumors were then evaluated for KRAS2 and BRAF mutations on the basis of their microsatellite instability. Results: One hundred forty-six endometrioid endometrial cancers were evaluated for KRAS2 and BRAF mutations. Thirty-five cancers (24%) had activating KRAS2 mutations, but only a single BRAF mutation was identified in an microsatellite instabilityepositive cancer. Twenty-four of 81 microsatellite instability high cancers (29.6%) in which the MLH1 repair gene was methylated had KRAS2 mutations. When compared with the other groups, this finding approached statistical significance (P = .06). KRAS2 mutation status was associated with increasing age at diagnosis (P = .02). Conclusion: Despite many similarities between colon and endometrial cancers, the mechanism of the development of endometrial cancers appears to be different from colon cancers in that BRAF is not affected by a mismatch repair problem, because only KRAS2 mutations were seen. In addition, increasing age appears to lead to an increased likelihood that such a mutation will occur. Ó 2004 Elsevier Inc. All rights reserved.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
Presented at the Twenty-Second Annual Meeting of the American Gynecological and Obstetrical Society, Napa, California, September 18-20, 2003. Supported by grant CA71754 from the National Institutes of Health. 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.01.017
Reprint requests: David G. Mutch, MD, Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Washington University School of Medicine, 4911 Barnes Hospital Plaza, St. Louis, MO 63110. E-mail: [email protected]
936
Figure 1 Selection of cases for KRAS2 and BRAF mutation analysis. The tumors that were investigated came from our patient series, which was characterized previously for tumor MSI and MLH1 methylation.9 One hundred sixteen of 127 MSI-H cases in the series (91%) were studied for KRAS2 and BRAF mutation along with 10 MSI-L and 30 MSS tumors. The asterisk denotes unsuccessful mutation analyses for 11 MSI-H M tumors (repeat failures with R1 of the 5 PCR reactions for KRAS2 and BRAF mutation analysis).
Cancers arise because of the accumulation of alterations in genes that control DNA repair, cell growth, proliferation, differentiation, and death. The molecular basis of the stepwise progression from a normal to malignant cell is perhaps best described in cancers of the colon.1 Endometrial cancers are associated with colon cancer and are the second most common cancer to be associated with hereditary nonpolyposis colon cancer (HNPCC).2-4 Furthermore, endometrial cancers share many of the molecular characteristics of colon cancers, which is evident from the finding that the 2 types of cancers frequently exhibit a DNA mismatch repair abnormality.4-6 This can occur through a germline mutation or somatic mutation or through hypermethylation of promoter regions that effectively silence the gene.6-10 Thus, both types of cancers, colon and endometrial, are linked by mismatch repair, and both tumor types often demonstrate the so-called mutator phenotype.2-4,10-13 Similarly, both colon and endometrial cancers share the trait that mutation in the KRAS2 gene is a common and early event in the genesis of their uncontrolled growth.14-16 It has been shown that KRAS2 mutations occur predominantly during the transformation of a small-to-intermediate sized adenoma in colon cancers and in the transition of hyperplasia to endometrial carcinoma.14,16,17 It is unclear how closely these mutations are linked to mismatch repair. However, the RAS/RAF family does seem to be an early target for this type of abnormality.16-18 Genes of the RAF family encode kinases that are important regulators of cellular response to growth signals. Protein products of these genes participate in the RASRAF-MEK-ERK-MAP kinase pathway.19-21 There are 3 RAF genes, each of which code for a threonine kinase. The RAF function is regulated by RAS.19,21 Colorectal cancers often have mutations in the RAS family; when this occurs, it is uncommon forO1 gene that is involved with the expression pathway to be disrupted.18,22 That is
Mutch et al to say that, when RAS is mutated, BRAF is not, and similarly, when BRAF is mutated, RAS is not. This makes sense because there is no need forO1 activating mutation to allow the cancer cell to progress to the next step toward malignancy. Given the similarity of mismatch repair and RAS mutations in colon and endometrial cancers, we investigated the occurrence and spectrum of KRAS2 and BRAF mutations in endometrial cancers of known DNA mismatch repair status.
Material and methods Patient and specimen selection This study was approved by the Human Studies Committees; an informed written consent was obtained from all participants. Tissue specimens from 441 women with primary endometrial cancers were collected at the time of treatment by hysterectomy, and FIGO surgical stage was assigned. Mixed Mu¨llerian tumors and other sarcomas were excluded from this analysis. The cases that were analyzed for KRAS2 and BRAF mutations were selected from 441 tumors on the basis of the results of microsatellite instability (MSI) analysis (Figure 1). All cases with MSI were evaluated, and a representative subset of 30 microsatellite stable (MSS) tumors was analyzed. MSI and MLH1 promoter methylation analyses were preformed as previously described.9 A tumor was classified as MSI-high (MSI-H) if R2 markers showed instability, MSS if no instability was noted, and MSIlow (MSI-L) if a single marker revealed novel bands. MSI-H with hMLH1 promoter methylation was designated as MSI-H M; those without promoter methylation were designated MSI-H U.
KRAS2 and BRAF mutation analysis Codons 12 and 13 of KRAS2 were evaluated for mutation with the use of previously described polymerase chain reaction (PCR) amplification/restriction digestion assays.23 Codon 61 was evaluated by single-strand conformational variant analysis, as described elsewhere.15 The codon 12, 13, and 61 mutations were determined by the direct sequencing of the amplified PCR products (codons 12 and 13 in exon 1 and codon 61 in exon 2), as reported previously by our group.15,16 BRAF exons 11 and 15 were PCR amplified and subjected to single-strand conformational variant analysis on MDE (FMC Bioproducts, Rockland, Maine) gels with and without 5% glycerol. In addition to the singlestrand conformational variant analyses, PCR products from 23 tumors were sequenced directly. The primers that were used for amplification and sequencing were as follows: BRAF exon 11: Forward 5# TTTCTTAAGGGGATCTCTTCCTG 3#, Reverse 5# CCGACTGCTGTGAACAGTTTT 3#; BRAF exon 15: Forward 5#
937
Mutch et al
Figure 2 Frequency of KRAS2 mutation according to molecular classification. The MSI-H M tumor group had the highest rate of KRAS2 mutation, but the rate of mutation was not significantly different than for the other 3 classes.
Table KRAS2 mutation by grade and stage in uterine endometrioid adenocarcinomas Endometrioid Adenocarcinoma KRAS2 Mutation Status Grade 1 2 3 Stagey I II III IV
Positive (No.)
Negative* (No.)
14 (21.5%) 18 (29%) 3 (16%)
51 (78.5%) 44 (71%) 16* (84.2%)
26 2 6 0
(24.5%) (18.2%) (27.3%) (0%)
80 9 16 4
(75.5%) (81.8%) (72.7%) (100%)
* 10 cases in series were clear cell or papillary carcinomas and not included in this analysis (3 Stage II, 6 Stage III, and 1 Stage IV). y Incomplete staging for 1 case with KRAS2 mutation and 2 cases without mutation.
TGCTTGCTCTGATAGGAAAATG 3# and Reverse 5# TC AGGGCCAAAAATTTAATCA 3#.
Statistical analysis Fisher’s exact test, the Student t test, and the MannWhitney test were performed with InStat version for Macintosh (GraphPad Software Inc, San Diego, Calif). A probability value of !.05 was considered statistically significant.
Results Of 156 endometrial cancers that were evaluated for KRAS2 and BRAF mutation, 146 cancers were of endometrioid histologic condition and 10 were of clear-cell or papillary serous histologic condition. The 10 clear cell
Figure 3 BRAF intron 11 mutation that was identified in MSI-H U tumor 1184. The arrow indicates the insertion of a T at position C74 from the 3# end of exon 11.
and papillary serous were eliminated from this evaluation. Of the remaining 146 specimens, 35 specimens (24%) had activating KRAS2 mutations, although a single case had a BRAF mutation of uncertain functional significance. Most of the KRAS2 mutations (29/35) were in codon 12. Five mutations were in codon 13, and 1 mutation was in codon 61. Thirty-four of the 35 tumors with KRAS2 mutation were of the endometrioid histologic subtype, and 1 case (1411) was a mixed endometrioid/ serous carcinoma. For case 1411, the mutation analysis was performed for the endometrioid component that made up most of the tumor. The rate of KRAS2 mutation varied with tumor molecular class. Twenty-four of 81 tumors (29.6%) in the MSI-H M group had KRAS2 mutation, whereas 4 of 35 MSI-H U tumors (11.4%), 1 of 10 MSI-L tumors (10%), and 6 of 30 MSS tumors (20%) had mutation (Figure 2). The difference in the rate of mutation between the MSI-H M and the MSI-H U groups approached statistical significance (P = .06; Fisher’s exact test). It is noteworthy that all 4 MSI-H U tumors with KRAS2 mutations arose in women with inherited mutations in DNA mismatch repair. Tumor 1150, which has a codon 12 Asp mutation, is from a case with an MSH2 mutation.10 Tumor 1319 (codon 12 Ala mutation), tumor 1524 (codon 12 Asp mutation), and tumor 1401 (codon 13 Asp) all arose in MSH6 mutation carriers.9 KRAS2 mutation was not associated with either tumor grade or stage (Table). Mutation status, however, was associated with increased age at diagnosis. The mean and median age at diagnosis for the 35 patients with tumors-harboring mutations were 68.3 and 70.1 years, respectively; the mean and median ages for the 121 mutation-negative cases were 63.2 and 62.2 years, respectively. The 5.1-year increase in mean ages and 7.9-year increase in median ages of the patients with tumor KRAS2 mutations were statistically significant (P = .02; 2-sided Student t test; and P = .02; MannWhitney test).
938 A single example of BRAF mutation was identified among 146 endometrioid cases that were evaluated for sequence alterations in exons 11 and 15. An insertion mutation (insert T) in intron 11 was seen in the MSIpositive cancer case 1184 (Figure 3). Direct sequencing of amplicons that corresponded to exons 13 and 17 from 19 tumors revealed intronic polymorphisms, which indicated that the direct sequencing of tumor-derived PCR products reliably could reveal sequence alterations (data not shown).
Comment The results of this study demonstrate that the path to malignant transformation in endometrial cancers appears to be different than that of colon cancers, despite other similarities. There were no significant BRAF mutations in tumors with defective DNA mismatch repair. On the basis of the higher KRAS2 mutation rate in MSI-positive cancers, it appears that the inability to repair mismatch errors may contribute to KRAS2 mutation Also, increasing age seems to increase the likelihood that a mutation will occur. The rate of KRAS2 mutation that was observed in this study parallels what has been reported previously. Lax et al24 made the observation that KRAS2 mutation is a molecular feature that distinguishes endometrioid endometrial cancers from serous carcinomas. The rate of KRAS2 mutation in our series (35/146 mutations [24%]) is similar to the 26% rate for codon 12 mutation that was reported by Lax et al24 for endometrioid carcinomas. In our series, we saw a significant increase in the age at diagnosis of women with tumors with KRAS2 mutations compared with mutation-negative cases. Because our series was enriched for tumors with MSI and because our previous demonstration that the MSI-H M tumor phenotype is associated with increased age at diagnosis,9 we compared the age at diagnosis of mutation-positive and mutation-negative cases among the MSI-H M group. The higher age at diagnosis was also observed in this subset of patients. The mean and median ages of mutationpositive cases were 70.4 and 72.0 years, respectively; the mean and median ages of mutation-negative cases were 66.4 and 66.4 years, respectively. The increase in mean and median age approached statistical significance (P = .06; 1-sided Student t test; and P = .06; MannWhitney test). Further studies will be required to determine whether KRAS2 mutation is a feature of late age of onset in endometrial carcinoma. The absence of BRAF mutations in MSI-positive endometrial cancers was unexpected. Previous studies of colorectal neoplasms with MSI suggested that tumors with defective DNA mismatch repair frequently have BRAF mutations.18 In a series of 330 colorectal tumors that were investigated for KRAS2 and BRAF mutation,
Mutch et al 32 tumors (10%) had BRAF defects, and 169 tumors (51%) had KRAS2 mutations. BRAF and KRAS2 mutations were not seen in the same tumor, which is consistent with the defects that exert equivalent effects in tumor formation.22 The rate of BRAF mutation was much higher in tumors with MSI; 15 of 49 MSI-positive tumors had BRAF mutations. Because both endometrial and colorectal cancer show frequent MSI and KRAS2 mutation, we reasoned that a similar relationship might exist in MSI-positive endometrial cancers. Our studies have demonstrated that endometrial cancers differ from colorectal tumors in that BRAF mutations are not seen in MSI-positive cases. One possible explanation for the absence of BRAF mutation in MSI-positive endometrial cancers relates to the relative roles that defective DNA mismatch repair and the RAS/RAF pathway play in tumor formation in the 2 different tissues. In our series, 127 of 441 cases (29%) were MSI-positive (Figure 1); only 49 of 330 of the colorectal tumors (15%) that were studied by Rajagopalan et al18 were MSI-positive. Based on the higher rate of MSI in endometrial cancers, we speculate that the loss of DNA mismatch repair plays a more important role in endometrial tumorigenesis than in colorectal tumorigenesis. On the other hand, the rate of KRAS2 mutation in our series of endometrial cancers (35/146 cancers [24%]) is less than in the colorectal tumors that were studied for KRAS2 and BRAF mutations. Overall, 51% of colorectal tumors (169/330) had KRAS2 mutations. Because we investigated a selected subset of endometrial cancers, which were enriched for cases with defective DNA mismatch repair, the most direct comparisons of KRAS2 and BRAF mutation rates can be made between MSI-positive endometrial and colorectal tumors. Twenty-one of 49 of MSI-positive colorectal tumors (43%) had KRAS2 mutation18; only 28 of 116 of MSI-positive endometrial cancers (24%) had KRAS2 mutation. Our studies and the studies of Rajagopalan et al18 suggest that KRAS mutations are nearly twice as frequent in colorectal tumors than in endometrial cancers with MSI. The relative role that BRAF plays in these 2 groups of tumors, however, is very different. No examples of activating BRAF mutation were found among 116 MSI-positive endometrial cancers, whereas 21 of 49 MSI-positive colorectal cancers had BRAF mutations. Although colorectal and endometrial cancers share the common features of MSI and KRAS2 mutation, they differ with respect to the involvement of the RAS/RAF pathway, which is evidenced by the lower rate or KRAS2 mutation and the absence of activating BRAF mutations in endometrial cancers.
Acknowledgments We thank Ms Erin Ball for her assistance with the preparation of this manuscript.
939
Mutch et al
References 1. Feron ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1995;61:197-205. 2. Watson P, Lynch HT. Extracolonic cancer in the hereditary nonpolyposis colorectal cancer. Cancer 1993;71:679-85. 3. Watson P, Vasen HFA, Mecklin JP, Jarvinen H, Lynch HT. The risk of endometrial cancer in hereditary nonpolyposis colorectal cancer. Am J Med 1994;96:516-20. 4. Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshelman JR, Burt RW, et al. A national cancer institute workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998;58:5248-57. 5. Kowalski LD, Mutch DG, Herzog TJ, Rader JS, Goodfellow PJ. Mutational analysis of MLH1 and MSH2 in 25 prospectively acquired RERC endometrial carcinomas. Genes Chromosomes Cancer 1997;18:219-27. 6. Katabuchi H, van Rees B, Lambers AR, Ronnett BM, Blazes MS, Leach FS, et al. Mutations in DNA mismatch repair genes are not responsible for microsatellite instability in most sporadic endometrial carcinomas. Cancer Res 1995;55:5556-60. 7. Estellar M, Levine R, Baylin SB, Ellenson LH, Herman JG. MLH1 promoter hypermethylation is associated with the microsatellite instability phenotype in sporadic endometrial carcinomas. Oncogene 1999;16:2413-7. 8. Simpkins SB, Swisher EM, Mutch DG, Gersell DJ, Kovatich AJ, Palazzo JP, et al. MLH1 promoter methylation and gene slicing is the primary cause of microsatellite instability in sporadic endometrial cancers. Hum Mol Genet 1999;8:661-6. 9. Goodfellow PJ, Buttin BM, Herzog TJ, Rader JS, Gibb RK, Swisher E, et al. Prevalence of defective DNA mismatch repair and MSH6 mutation in an unselected series of endometrial cancers. Proc Natl Acad Sci USA 2003;100:5908-13. 10. Whelan A, Babb S, Mutch DG, Rader J, Herzog TJ, Todd C, et al. MSI in endometrial carcinoma: absence of MLH1 promoter methylation is associated with increased familial risk for cancers. Int J Cancer 2002;99:697-704. 11. Wijnin J, De Leeuw W, Vasen H, van der Klift H, Moller P, Stormorken A, et al. Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Gen 1999;23:142-4. 12. Vasen HFA, Wijnen JT, Menko FH, Kleibeuker JH, Tall BG, Giffioen G, et al. Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. Gastroenterology 1996;110:1020-7. 13. Millar AL, Pal T, Madlensky L, Sherman C, Temple L, Mitri A, et al. Mismatch repair gene defects contribute to the genetic basis of double primary cancers of the colorectum and endometrium. Hum Mol Genet 1999;9:823-9. 14. Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. N Engl J Med 1988;319:525-32. 15. Swisher EM, Peiffer-Schneider S, Mutch DG, Herzog TJ, Rader JS, Eibenaary A, et al. Differences in patterns of TP53 and KRAS2 mutations in a large series of endometrial carcinomas with or without microsatellite instability. Cancer 1999;85:119-26. 16. Cohn DE, Mutch DG, Herzog TJ, Rader JS, Dintzis SM, Gersell DJ, et al. Genotypic and phenotypic progression in endometrial tumorigenesis: determining when defects in DNA mismatch repair and KRAS2 occur. Genes Chromosomes Cancer 2001;32: 295-301. 17. Yuen TS, Davies H, Chan TL, Ho JW, Bignell GR, Cox C, et al. Similarity of the phenotypic patterns associated with BRAF and KRAs mutations in colorectal neoplasia. Cancer Res 2002;62: 6451-5.
18. Rajagopalan H, Bardelli A, Lengauer C, Kinzler KW, Vogelstein B, Velculescu VE. RAF/RAS oncogenes and mismatch-repair status. Nature 2002;418:934-7. 19. Peyssonnaux C, Eychene A. The Raf/MEK/ERK pathway: new concepts of activation. Biol Cell 2001;93:53-62. 20. Avruch JA. Ras activation on the Raf kinase:tyrosine kinase recruitment of the MAP kinase cascade. Recent Prog Horm Res 2001;56:127-55. 21. Kolch W. Meaningful relationships: the regulation of the Ras/Raf/ MEK/ERK pathway by protein interactions. Biochem J 2000;351: 289-305. 22. Davies H, Bignell GR, Cox C, Stephans P, Edkinss S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417: 949-54. 23. Lin SY, Chen PH, Wang CK, Liu JD, Slauw CP, Chen YJ, et al. Mutation analysis of K-ras oncogenes in gastroenterologic cancers by the amplified created restriction sites method. Am J Clin Pathol 1993;100:686-9. 24. Lax SF, Kendall B, Tashiro H, Slebos RJ, Hedrick L. The frequency of p53, K-ras mutations, and microsatellite instability differs in uterine endometrioid and serous carcinoma: evidence of distinct molecular genetic pathways. Cancer 2000;88:814-24.
Discussion DR DAVID M. GERSHENSON, Houston Tex. In 1913, Warthin1 first reported a clustering of colorectal cancers in ‘‘Family G.’’ Eighty years later, HNPCC, or the Lynch syndrome, crystallized.2 HNPCC is an autosomal dominantly inherited cancer susceptibility syndrome that is characterized by the development of early-onset colorectal cancer and extracolonic cancers (endometrium, gastrointestinal tract, genitourinary tract, ovary, and brain). Colorectal cancer is the most common malignancy in the syndrome, which accounts for approximately 5% of all colorectal cancers. The most common extracolonic cancer is endometrial cancer, which is the subject of this article. The diagnosis of HNPCC is based solely on family history, with the use of the internationally accepted ‘‘Amsterdam criteria.’’3,4 Included in the criteria are the requirements for at least 3 family members with a histologically verified HNPCC-associated cancer. In addition, affected individuals must span at least 2 generations; 1 individual must be a first-degree relative of the other 2 individuals, and 1 case must be diagnosed before age 50 years. A major clue to heighten one’s suspicion of the HNPCC syndrome is multiple tumors, either synchronous or metachronous, in the same individual. The molecular basis of HNPCC is a germline defect in mismatch repair.5,6 The primary function of the mismatch repair system is to eliminate single-base mismatches and insertion-deletion loops that may occur during DNA replication.7 To date, germline mutations in 1 of 4 major mismatch repair genes (MLH1, MSH2, MSH6, and PMS2) have been detected in up to 80% of HNPCC families.8 Two other putative mismatch repair gene mutations (MLH3 and EXO1) have been