- Email: [email protected]
CHEK2 variants predispose to benign, borderline and low-grade invasive ovarian tumors
Gynecologic Oncology 102 (2006) 429 – 431 www.elsevier.com/locate/ygyno
Rapid Communication
CHEK2 variants predispose to benign, borderline and low-grade invasive ovarian tumors J. Szymanska-Pasternak a,1 , A. Szymanska a,1 , K. Medrek a , E.N. Imyanitov b , C. Cybulski a , B. Gorski a , P. Magnowski c , I. Dziuba a , K. Gugala d , B. Debniak e , S. Gozdz f , A.P. Sokolenko b , N.Y. Krylova b , O.S. Lobeiko b , S.A. Narod g,⁎, J. Lubinski a a
International Hereditary Cancer Center, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland b N.N. Petrov Institute of Oncology, St. Petersburg, Russia c Clinic of Gynecology Oncology, Poznan Medical University, Poznan, Poland d Regional Oncology Hospital, Olsztyn, Poland e Regional Hospital, Gorzow Wielkopolski, Poland f Regional Oncology Center, Kielce, Poland g Centre for Research on Women's Health, 790 Bay Street, Toronto, Ontario, Canada M5G 1N8 Received 11 April 2006 Available online 10 July 2006
Abstract Objective. Three founder alleles of the CHEK2 gene have been associated with predisposition to a range of cancer types in Poland. Two founder alleles (1100delC and IVS2 + 1G N A) result in a truncated CHEK2 protein and the other is a missense substitution, leading to the replacement of a threonine with an isoleucine (I157T). Methods. To establish if these variants play a role in the etiology of ovarian tumors, we genotyped 1108 Polish women with various types of ovarian tumors and 4000 controls for the three CHEK2 variants. We included 539 Polish women with benign ovarian cystadenomas, 122 women with borderline ovarian malignancies and 447 women with invasive ovarian cancer. Results. Positive associations were seen with the CHEK2 I157T missense variant and ovarian cystadenomas (OR = 1.7; P = 0.005), with borderline ovarian cancers (OR = 2.6; P = 0.002) and with low-grade invasive cancers (OR = 2.1; P = 0.04). There was no association with ovarian cancer of high grade (OR = 1.0). The association between the I157T missense variant was then confirmed in a second sample of Russian patients with borderline ovarian cancers (OR = 2.7; P = 0.06). Conclusion. These data indicate that CHEK2 variants may predispose to a range of ovarian tumor types of low malignant potential, but not to aggressive cancers. © 2006 Elsevier Inc. All rights reserved. Keywords: Cystadenoma; Ovarian cancer; CHEK2
Introduction The CHEK2 gene encodes the human analogue of the yeast checkpoint kinases Cds1 and Rad53 [1]. Activation of CHEK2 in response to DNA damage prevents the cell from entering into mitosis. In Poland, there are three polymorphic variants of
⁎ Corresponding author. Fax: +1 416 351 3767. E-mail address: [email protected] (S.A. Narod). 1 Both authors contributed equally to this work. 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.05.040
CHEK2, which, in aggregate, are present in 5.5% of the population; two of these (IVS2 + 1G N A and 1100delC) are rare and result in premature protein truncation [2]. The IVS2 + 1G N A appears to be characteristic founder mutation in the Polish population. The frequency of the 1100delC variant ranges from 0.2% to 1.4% in various European populations [3,4] but this allele is infrequent in North America [5]. The 1100delC allele has been found to confer an elevated risk of breast cancer [3,4] and prostate cancer [2,6,7]. The third variant is a common missense variant (I157T) that results in the substitution of an isoleucine for a threonine. We have
430
J. Szymanska-Pasternak et al. / Gynecologic Oncology 102 (2006) 429–431
Table 1 Association between CHEK2 I157T variants and types of ovarian tumors in Poland Tumor type
Number of Positive subjects tested (%)
Ovarian cystadenoma 539 Borderline malignancy 122 Ovarian cancers 447 Controls 4000
42 (7.8%) 14 (11.5%) 26 (5.8%) 193 (4.8%)
Odds P 95%CI ratio va1ue 1.7 2.6 1.2 –
0.005 1.2–2.3 0.002 1.4–4.5 0.4 0.8–1.8
recently reported that this variant is associated with increased risks of several cancer types in Poland [8] but these sites did not include ovarian cancer. Ovarian cystadenoma is a benign tumor of the ovarian epithelium with histologic features similar to those of malignant cystadenocarcinoma. Both cystadenomas and carcinomas may be of serous or mucinous histology. We found that ovarian cystadenomas may be feature of the familial breast/ ovarian cancer syndrome in Poland, after excluding families with BRCA1 mutations [9]. This observation prompted us to evaluate other candidate genes that may underlie the association between benign ovarian tumors and malignant breast tumors. Because CHEK2 has been associated with a wide range of cancer types in Poland, including breast cancer, it was selected for study. Material and methods Study subjects Cases were collected from 32 hospitals in 5 provinces of Poland (Szczecin, Olsztyn, Gorzow, Kielce, Poznan) between 1997 and 2005. Patients were identified through the pathology departments of the participating hospitals. Study subjects were unselected for age and family history. Eligible patients were invited to participate in the study by attending an outpatient clinic where they signed informed consent and provided a blood sample. We included 539 Polish women with benign ovarian cystadenomas, 122 women with borderline ovarian malignancies and 447 women with invasive ovarian cancer. Patient participation rates exceeded 80%. The mean age of the women with benign tumors was 45.8 years (range 14 to 86), with tumors of low malignant potential was 44.0 years (range 17 to 77 years) and of malignant tumors was 54.5 years (range 18 to 80 years). Among women with invasive cancers, the mean age of the women with grade I tumors was 52.7 years (range 19 to 79 years), the mean age of the women with grade II tumors was 55.3 years (range 18 to 80 years) and the mean age of the women with grade III tumors was 54.7 years (26 to 78 years). Three control groups from Poland were combined. The first group consisted of 2000 newborn children from ten hospitals throughout Poland (Szczecin, Bialystok, Gorzow, Katowice, Wroclaw, Poznan, Opole, Lodz and Rzeszow) in 2003 and 2004. Samples of cord blood from unselected infants were forwarded to the study center in Szczecin. The second control group was taken from adult patient rolls of three family doctors practicing in the Szczecin region. 1000 controls were selected at random from the patient lists of these family doctors. The third control group consisted of adults from Szczecin who submitted blood for paternity testing. A sample of DNA was forwarded to the reference laboratory without identifying information. To ensure comparability of the control groups, the allele frequencies of the three alleles were computed separately for the adult and neonatal control groups and compared. In addition, we studied 77 Russian women with borderline serous ovarian malignancies from the N.N. Petrov Institute of Oncology in St. Petersburg and the Regional Oncology Hospital (St. Petersburg). Russian controls consisted of 150 females, who were either volunteer blood donors (ages 18–91) from St. Petersburg (N = 50) or cancer free hospital controls (ages 56–91) collected in the general hospitals of St. Petersburg (N = 100).
Laboratory analysis We assessed the frequency of three CHEK2 variants, IVS2 + 1G N A, 1100del C and I157T in the Polish cases and controls. The IVS2 + 1G N A mutation was identified by RFLP-PCR using Hpy188 III (New England Biolabs) and primers Ch2/3F (5′-ATT TAT GAG CAA TTT TTA AAC G) and Ch2/3R (5′-TCC AGT AAC CATAAG ATA ATA ATATTA C). The PCR product was digested in samples where a mutation was present. The 430T N C variant (IleI57Thr) was analyzed by restriction fragment length polymorphism polymerase chain reaction, using Ch157F (5′- A CCC ATG TAT CTA GGA GAG CTG) and Ch157R (5′- CCA CTG TGA TCT TCT ATG TCT GCA) primers. The reverse primer introduced artificial restriction site for PstI enzyme. The PCR products were digested in mutation positive cases. The 1100delC was analyzed using an allele specific polymerase chain reaction assay using primers Chk2ex10F (5′-TTA ATT TAA GCA AAA TTA AAT GTC), Chk2ex10R (5′-GGC ATG GTG GTG TGC A TC) and Chk2de1C (5′-TGG AGT GCC CAA AAT CAT A). All reactions were carried out in ThermalCycler 9600 (Perkin Elmer). PCR products were separated in 2–3% agarose gels and visualized in UV light. In cases positive in RFLP- and ASA-PCR analyses, a separate DNA sample was sequenced to confirm the presence of the mutation. The Russian cases and controls were tested for the I157T mutation only. Statistical analysis included the comparison of the proportions of the prevalence of the allele in cases and controls. Odds ratios were generated from two-bytwo tables and statistical significance was assessed using the Fisher exact test and Chi-square test. The Polish and Russian populations were analyzed separately and the data were then combined using the Mantel–Haenszel estimate for combining odds ratios across strata.
Results The frequencies of the CHEK2 missense variant I157T in Polish cases and controls are presented in Tables 1–3. There was no overall association between this CHEK2 variant and invasive ovarian cancer in the data set (OR = 1.2). However, we observed positive associations with the missense I157T variant and benign cystadenomas (OR = 1.7; P = 0.005) and with borderline lesions (OR = 2.6; P = 0.002). Because of the observed relationship between degree of invasiveness, we looked at the subgroup of patients with low-grade invasive ovarian cancers. For this group, an association with I157T was present, but was of borderline significance (OR = 2.1; P = 0.04). The association was stronger, however, for women with early-onset disease; there were 19 women diagnosed with invasive low-grade serous cancers before the age of 50, of whom four carried the I157T variant (OR = 4.4; P = 0.006). Among women with ovarian cancers of intermediate or high grade, the frequency of the I157T CHEK2 variant was not higher than expected (OR = 1.0). Table 2 Association between CHEK 1157T variant and invasive ovarian cancers in Poland Tumor type
Number of subjects tested
Positive (%)
Odds ratio
P value
95%CI
All b50 years N50 years Grade 1 Grade 2 Grade 3 Serous Mucinous Others Controls
447 158 289 88 178 181 257 49 141 4000
26 (5.8%) 7 (4.5%) 19 (6.5%) 9 (10.2%) 11 (6.1%) 6 (3.3%) 15 (5.8%) 1 (2.0%) 10 (7.1 %) 193 (4.8%)
1.2 0.9 1.4 2.1 1.3 0.7 1.1 0.4 1.5
0.4 0.97 0.2 0.04 0.52 0.4 0.6 0.57 0.3
0.8–1.8 0.4–1.9 0.8–2.2 1.1–4.5 0.7–2.4 0.3–1.5 0.6–1.9 0.06–3.0 0.8–2.9
J. Szymanska-Pasternak et al. / Gynecologic Oncology 102 (2006) 429–431 Table 3 Association between CHEK2 I157T variant and borderline ovarian cancers in Poland Tumor type
Number of subjects tested
Positive (%)
Odds ratio
P value
95%CI
All b50 years N50 years Serous Mucinous Others Controls
122 85 37 87 32 3 4000
14 (11.5%) 10 (11.7%) 4 (10.8%) 12 (13.8%) 2 (6.3%) 0 (0%) 193 (4.8%)
2.5 2.6 2.4 3.2 1.3 0
0.0019 0.008 0.2 0.0004 0.7 –
1.4–4.5 1.3–5.1 0.8–6.8 1.7–6.0 0.3–5.5 –
Data were available for 77 Russian women with serous cystadenomas of borderline malignancy and 150 controls and the missense variant I157T. A mutation was present in 9 of 77 cases and 7 of 150 controls (OR = 2.7; P = 0.06). The magnitude of the odds ratio was almost identical to that of the Polish cases. The odds ratios for the Polish women and the Russian women with borderline ovarian malignancies were then combined, using the Mantel–Haenszel method of adjustment. The overall odds ratio was 2.7 (95% CI 1.6 to 4.4; P = 0.00009). Truncating CHEK2 variants were studied in the Polish cases only. These rare variants were observed in 1.3% of women with benign tumors, in none of the 120 borderline tumors, in none of 209 of cancers and in 0.7% of 4000 controls. The odds ratio for a truncating variant and a benign ovarian tumor was 1.8, but this association was not significant (P = 0.2).
431
(data not shown). We were able to confirm this association in a second population of Russian women, but due to small numbers, the association in Russia was not statistically significant. It will be of interest to see if this relationship holds in other ethnic groups. We have previously found the missense variant I157T to be associated with an increased risk of breast cancer (OR 1.4; P = 0.02), colon cancer (OR 2.0; P = 0.001), kidney cancer (OR 2.1; P = 0.0006), prostate cancer (OR 1.7; P = 0.002) and thyroid cancer (OR 1.9; P = 0.04) [8]. The magnitude of the risk increase is similar for all of these sites (about two-fold). Thus, CHEK2 appears to be a multi-site cancer susceptibility gene. This study further expands the range of tumor types associated with the CHEK2 I157T variant. Acknowledgments We would like to acknowledge the contributions of Menkiszak J, Rzepka-Gorska I, Kowalska E, Lener M, Matyjasik J, Stawicka M, Spaczynski M, Foszczynska-Kloda M, Czeszynska B, Czajka R, Fabian W, Raczynski A, Chosia M, Domagała W, Starczewski A, Niedzielski A, Lokociejewski J, Kaluzynski W, Abucewicz A, Sygut J, Starzewski J, Rudzinski R, Palys M, Witek A, Graniczka M, Mizgiert W, Pawlak L, Brych M, Klusewicz H, Rams L, Gromacki S, Gizewski A, Stefanowicz M, Frackowiak L, Malarkiewicz J, Wisniewski R, Janowski J, Wichlinski W, Maczuga Z, Jaworska M, Swiderski R, Klukowski M, Sztencel K, Grobelny R in collection of patients and controls.
Discussion
References
This report suggests an association between a founder missense mutation in the CHEK2 gene (I157T) and benign, borderline and low-grade malignant ovarian tumors. To our knowledge, this is the first identified example of a susceptibility gene for borderline ovarian tumors. The association was first seen among Polish women, but was confirmed in a second group of Russian patients. In contrast, the BRCA1 and BRCA2 genes predispose to invasive ovarian malignancies, but not to borderline malignancies [10–12]. Ovarian cancers associated with BRCA1 mutations are typically of high-grade [10,11], whereas the relationship with CHEK2 was restricted to low-grade tumors. To our knowledge, benign ovarian tumors have not been reported to be found in excess in families with BRCA1 mutations, nor have they been reported to be a feature of CHEK2 families studied to date. There are several strengths of our study. We had a relatively large number of cases and controls to study. Poland is a genetically homogeneous country and is well-suited for association studies. Most of our cases were recruited from the Szczecin region, which is populated by ethnic Poles who immigrated to the region from throughout Poland after the second world war, as German residents were relocated elsewhere. Our control group was drawn both from the adult population of Szczecin and from newborns in ten cities throughout Poland. However, the frequency of the CHEK2 alleles was similar in the newborn and adult controls. Also, there was no statistical difference in the CHEK2 allele frequencies in the newborns recruited from the Szczecin metropolitan region compared to other Polish cities
[1] Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science 1998;282:1893–7. [2] Cybulski C, Huzarski T, Gorski B, et al. A novel founder CHEK2 mutation is associated with increased prostate cancer risk. Cancer Res 2004;64:2677–9. [3] Meijers-Heijboer H, van den Ouweland A, Klijn J, et al. Low-penetrance susceptibility to breast cancer due to CHEK2* 1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 2002;3:55–9. [4] CHEK2 Breast Cancer Case-Control Consortium. CHEK2*1100delC and susceptibility to breast cancer: a collaborative analysis involving 10,860 breast cancer cases and 9,065 controls from 10 studies. Am J Hum Genet 2004;74:1175–82. [5] Offit K, Pierce H, Kirchhoff T, et al. Frequency of CHEK2* 1100delC in New York breast cancer cases and controls. BMC Med Genet 2003;4:1. [6] Dong X, Wang L, Taniguchi K, et al. Mutations in CHEK2 associated with prostate cancer risk. Am J Hum Genet 2003;72:270–80. [7] Seppala EH, Ikonen T, Mononen N, et al. CHEK2 variants associate with hereditary prostate cancer. Br J Cancer 2003;89:1966–70. [8] Cybulski C, Gorski B, Huzarski T, et al. CHEK2 is a multiorgan cancer susceptibility gene. Am J Hum Genet 2004;75:1131–5. [9] Menkiszak J, Brzosko M, Gorski B, et al. Ovarian cystadenoma as a characteristic feature of families with hereditary ovarian cancers unassociated with BRCA1 and BRCA2 mutations. J Appl Genet 2004;45:255–63. [10] Moslehi R, Chu W, Karlan B, et al. BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer. Am J Hum Genet 2000;66:1259–72. [11] Risch HA, McLaughlin JR, Cole DE, et al. Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. Am J Hum Genet 2001;68:700–10. [12] Gotlieb WH, Chetrit A, Menczer J, et al. Demographic and genetic characteristics of patients with borderline ovarian tumors as compared to early stage invasive ovarian cancer. Gynecol Oncol 2005;97:780–3.
Rapid Communication
CHEK2 variants predispose to benign, borderline and low-grade invasive ovarian tumors J. Szymanska-Pasternak a,1 , A. Szymanska a,1 , K. Medrek a , E.N. Imyanitov b , C. Cybulski a , B. Gorski a , P. Magnowski c , I. Dziuba a , K. Gugala d , B. Debniak e , S. Gozdz f , A.P. Sokolenko b , N.Y. Krylova b , O.S. Lobeiko b , S.A. Narod g,⁎, J. Lubinski a a
International Hereditary Cancer Center, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland b N.N. Petrov Institute of Oncology, St. Petersburg, Russia c Clinic of Gynecology Oncology, Poznan Medical University, Poznan, Poland d Regional Oncology Hospital, Olsztyn, Poland e Regional Hospital, Gorzow Wielkopolski, Poland f Regional Oncology Center, Kielce, Poland g Centre for Research on Women's Health, 790 Bay Street, Toronto, Ontario, Canada M5G 1N8 Received 11 April 2006 Available online 10 July 2006
Abstract Objective. Three founder alleles of the CHEK2 gene have been associated with predisposition to a range of cancer types in Poland. Two founder alleles (1100delC and IVS2 + 1G N A) result in a truncated CHEK2 protein and the other is a missense substitution, leading to the replacement of a threonine with an isoleucine (I157T). Methods. To establish if these variants play a role in the etiology of ovarian tumors, we genotyped 1108 Polish women with various types of ovarian tumors and 4000 controls for the three CHEK2 variants. We included 539 Polish women with benign ovarian cystadenomas, 122 women with borderline ovarian malignancies and 447 women with invasive ovarian cancer. Results. Positive associations were seen with the CHEK2 I157T missense variant and ovarian cystadenomas (OR = 1.7; P = 0.005), with borderline ovarian cancers (OR = 2.6; P = 0.002) and with low-grade invasive cancers (OR = 2.1; P = 0.04). There was no association with ovarian cancer of high grade (OR = 1.0). The association between the I157T missense variant was then confirmed in a second sample of Russian patients with borderline ovarian cancers (OR = 2.7; P = 0.06). Conclusion. These data indicate that CHEK2 variants may predispose to a range of ovarian tumor types of low malignant potential, but not to aggressive cancers. © 2006 Elsevier Inc. All rights reserved. Keywords: Cystadenoma; Ovarian cancer; CHEK2
Introduction The CHEK2 gene encodes the human analogue of the yeast checkpoint kinases Cds1 and Rad53 [1]. Activation of CHEK2 in response to DNA damage prevents the cell from entering into mitosis. In Poland, there are three polymorphic variants of
⁎ Corresponding author. Fax: +1 416 351 3767. E-mail address: [email protected] (S.A. Narod). 1 Both authors contributed equally to this work. 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.05.040
CHEK2, which, in aggregate, are present in 5.5% of the population; two of these (IVS2 + 1G N A and 1100delC) are rare and result in premature protein truncation [2]. The IVS2 + 1G N A appears to be characteristic founder mutation in the Polish population. The frequency of the 1100delC variant ranges from 0.2% to 1.4% in various European populations [3,4] but this allele is infrequent in North America [5]. The 1100delC allele has been found to confer an elevated risk of breast cancer [3,4] and prostate cancer [2,6,7]. The third variant is a common missense variant (I157T) that results in the substitution of an isoleucine for a threonine. We have
430
J. Szymanska-Pasternak et al. / Gynecologic Oncology 102 (2006) 429–431
Table 1 Association between CHEK2 I157T variants and types of ovarian tumors in Poland Tumor type
Number of Positive subjects tested (%)
Ovarian cystadenoma 539 Borderline malignancy 122 Ovarian cancers 447 Controls 4000
42 (7.8%) 14 (11.5%) 26 (5.8%) 193 (4.8%)
Odds P 95%CI ratio va1ue 1.7 2.6 1.2 –
0.005 1.2–2.3 0.002 1.4–4.5 0.4 0.8–1.8
recently reported that this variant is associated with increased risks of several cancer types in Poland [8] but these sites did not include ovarian cancer. Ovarian cystadenoma is a benign tumor of the ovarian epithelium with histologic features similar to those of malignant cystadenocarcinoma. Both cystadenomas and carcinomas may be of serous or mucinous histology. We found that ovarian cystadenomas may be feature of the familial breast/ ovarian cancer syndrome in Poland, after excluding families with BRCA1 mutations [9]. This observation prompted us to evaluate other candidate genes that may underlie the association between benign ovarian tumors and malignant breast tumors. Because CHEK2 has been associated with a wide range of cancer types in Poland, including breast cancer, it was selected for study. Material and methods Study subjects Cases were collected from 32 hospitals in 5 provinces of Poland (Szczecin, Olsztyn, Gorzow, Kielce, Poznan) between 1997 and 2005. Patients were identified through the pathology departments of the participating hospitals. Study subjects were unselected for age and family history. Eligible patients were invited to participate in the study by attending an outpatient clinic where they signed informed consent and provided a blood sample. We included 539 Polish women with benign ovarian cystadenomas, 122 women with borderline ovarian malignancies and 447 women with invasive ovarian cancer. Patient participation rates exceeded 80%. The mean age of the women with benign tumors was 45.8 years (range 14 to 86), with tumors of low malignant potential was 44.0 years (range 17 to 77 years) and of malignant tumors was 54.5 years (range 18 to 80 years). Among women with invasive cancers, the mean age of the women with grade I tumors was 52.7 years (range 19 to 79 years), the mean age of the women with grade II tumors was 55.3 years (range 18 to 80 years) and the mean age of the women with grade III tumors was 54.7 years (26 to 78 years). Three control groups from Poland were combined. The first group consisted of 2000 newborn children from ten hospitals throughout Poland (Szczecin, Bialystok, Gorzow, Katowice, Wroclaw, Poznan, Opole, Lodz and Rzeszow) in 2003 and 2004. Samples of cord blood from unselected infants were forwarded to the study center in Szczecin. The second control group was taken from adult patient rolls of three family doctors practicing in the Szczecin region. 1000 controls were selected at random from the patient lists of these family doctors. The third control group consisted of adults from Szczecin who submitted blood for paternity testing. A sample of DNA was forwarded to the reference laboratory without identifying information. To ensure comparability of the control groups, the allele frequencies of the three alleles were computed separately for the adult and neonatal control groups and compared. In addition, we studied 77 Russian women with borderline serous ovarian malignancies from the N.N. Petrov Institute of Oncology in St. Petersburg and the Regional Oncology Hospital (St. Petersburg). Russian controls consisted of 150 females, who were either volunteer blood donors (ages 18–91) from St. Petersburg (N = 50) or cancer free hospital controls (ages 56–91) collected in the general hospitals of St. Petersburg (N = 100).
Laboratory analysis We assessed the frequency of three CHEK2 variants, IVS2 + 1G N A, 1100del C and I157T in the Polish cases and controls. The IVS2 + 1G N A mutation was identified by RFLP-PCR using Hpy188 III (New England Biolabs) and primers Ch2/3F (5′-ATT TAT GAG CAA TTT TTA AAC G) and Ch2/3R (5′-TCC AGT AAC CATAAG ATA ATA ATATTA C). The PCR product was digested in samples where a mutation was present. The 430T N C variant (IleI57Thr) was analyzed by restriction fragment length polymorphism polymerase chain reaction, using Ch157F (5′- A CCC ATG TAT CTA GGA GAG CTG) and Ch157R (5′- CCA CTG TGA TCT TCT ATG TCT GCA) primers. The reverse primer introduced artificial restriction site for PstI enzyme. The PCR products were digested in mutation positive cases. The 1100delC was analyzed using an allele specific polymerase chain reaction assay using primers Chk2ex10F (5′-TTA ATT TAA GCA AAA TTA AAT GTC), Chk2ex10R (5′-GGC ATG GTG GTG TGC A TC) and Chk2de1C (5′-TGG AGT GCC CAA AAT CAT A). All reactions were carried out in ThermalCycler 9600 (Perkin Elmer). PCR products were separated in 2–3% agarose gels and visualized in UV light. In cases positive in RFLP- and ASA-PCR analyses, a separate DNA sample was sequenced to confirm the presence of the mutation. The Russian cases and controls were tested for the I157T mutation only. Statistical analysis included the comparison of the proportions of the prevalence of the allele in cases and controls. Odds ratios were generated from two-bytwo tables and statistical significance was assessed using the Fisher exact test and Chi-square test. The Polish and Russian populations were analyzed separately and the data were then combined using the Mantel–Haenszel estimate for combining odds ratios across strata.
Results The frequencies of the CHEK2 missense variant I157T in Polish cases and controls are presented in Tables 1–3. There was no overall association between this CHEK2 variant and invasive ovarian cancer in the data set (OR = 1.2). However, we observed positive associations with the missense I157T variant and benign cystadenomas (OR = 1.7; P = 0.005) and with borderline lesions (OR = 2.6; P = 0.002). Because of the observed relationship between degree of invasiveness, we looked at the subgroup of patients with low-grade invasive ovarian cancers. For this group, an association with I157T was present, but was of borderline significance (OR = 2.1; P = 0.04). The association was stronger, however, for women with early-onset disease; there were 19 women diagnosed with invasive low-grade serous cancers before the age of 50, of whom four carried the I157T variant (OR = 4.4; P = 0.006). Among women with ovarian cancers of intermediate or high grade, the frequency of the I157T CHEK2 variant was not higher than expected (OR = 1.0). Table 2 Association between CHEK 1157T variant and invasive ovarian cancers in Poland Tumor type
Number of subjects tested
Positive (%)
Odds ratio
P value
95%CI
All b50 years N50 years Grade 1 Grade 2 Grade 3 Serous Mucinous Others Controls
447 158 289 88 178 181 257 49 141 4000
26 (5.8%) 7 (4.5%) 19 (6.5%) 9 (10.2%) 11 (6.1%) 6 (3.3%) 15 (5.8%) 1 (2.0%) 10 (7.1 %) 193 (4.8%)
1.2 0.9 1.4 2.1 1.3 0.7 1.1 0.4 1.5
0.4 0.97 0.2 0.04 0.52 0.4 0.6 0.57 0.3
0.8–1.8 0.4–1.9 0.8–2.2 1.1–4.5 0.7–2.4 0.3–1.5 0.6–1.9 0.06–3.0 0.8–2.9
J. Szymanska-Pasternak et al. / Gynecologic Oncology 102 (2006) 429–431 Table 3 Association between CHEK2 I157T variant and borderline ovarian cancers in Poland Tumor type
Number of subjects tested
Positive (%)
Odds ratio
P value
95%CI
All b50 years N50 years Serous Mucinous Others Controls
122 85 37 87 32 3 4000
14 (11.5%) 10 (11.7%) 4 (10.8%) 12 (13.8%) 2 (6.3%) 0 (0%) 193 (4.8%)
2.5 2.6 2.4 3.2 1.3 0
0.0019 0.008 0.2 0.0004 0.7 –
1.4–4.5 1.3–5.1 0.8–6.8 1.7–6.0 0.3–5.5 –
Data were available for 77 Russian women with serous cystadenomas of borderline malignancy and 150 controls and the missense variant I157T. A mutation was present in 9 of 77 cases and 7 of 150 controls (OR = 2.7; P = 0.06). The magnitude of the odds ratio was almost identical to that of the Polish cases. The odds ratios for the Polish women and the Russian women with borderline ovarian malignancies were then combined, using the Mantel–Haenszel method of adjustment. The overall odds ratio was 2.7 (95% CI 1.6 to 4.4; P = 0.00009). Truncating CHEK2 variants were studied in the Polish cases only. These rare variants were observed in 1.3% of women with benign tumors, in none of the 120 borderline tumors, in none of 209 of cancers and in 0.7% of 4000 controls. The odds ratio for a truncating variant and a benign ovarian tumor was 1.8, but this association was not significant (P = 0.2).
431
(data not shown). We were able to confirm this association in a second population of Russian women, but due to small numbers, the association in Russia was not statistically significant. It will be of interest to see if this relationship holds in other ethnic groups. We have previously found the missense variant I157T to be associated with an increased risk of breast cancer (OR 1.4; P = 0.02), colon cancer (OR 2.0; P = 0.001), kidney cancer (OR 2.1; P = 0.0006), prostate cancer (OR 1.7; P = 0.002) and thyroid cancer (OR 1.9; P = 0.04) [8]. The magnitude of the risk increase is similar for all of these sites (about two-fold). Thus, CHEK2 appears to be a multi-site cancer susceptibility gene. This study further expands the range of tumor types associated with the CHEK2 I157T variant. Acknowledgments We would like to acknowledge the contributions of Menkiszak J, Rzepka-Gorska I, Kowalska E, Lener M, Matyjasik J, Stawicka M, Spaczynski M, Foszczynska-Kloda M, Czeszynska B, Czajka R, Fabian W, Raczynski A, Chosia M, Domagała W, Starczewski A, Niedzielski A, Lokociejewski J, Kaluzynski W, Abucewicz A, Sygut J, Starzewski J, Rudzinski R, Palys M, Witek A, Graniczka M, Mizgiert W, Pawlak L, Brych M, Klusewicz H, Rams L, Gromacki S, Gizewski A, Stefanowicz M, Frackowiak L, Malarkiewicz J, Wisniewski R, Janowski J, Wichlinski W, Maczuga Z, Jaworska M, Swiderski R, Klukowski M, Sztencel K, Grobelny R in collection of patients and controls.
Discussion
References
This report suggests an association between a founder missense mutation in the CHEK2 gene (I157T) and benign, borderline and low-grade malignant ovarian tumors. To our knowledge, this is the first identified example of a susceptibility gene for borderline ovarian tumors. The association was first seen among Polish women, but was confirmed in a second group of Russian patients. In contrast, the BRCA1 and BRCA2 genes predispose to invasive ovarian malignancies, but not to borderline malignancies [10–12]. Ovarian cancers associated with BRCA1 mutations are typically of high-grade [10,11], whereas the relationship with CHEK2 was restricted to low-grade tumors. To our knowledge, benign ovarian tumors have not been reported to be found in excess in families with BRCA1 mutations, nor have they been reported to be a feature of CHEK2 families studied to date. There are several strengths of our study. We had a relatively large number of cases and controls to study. Poland is a genetically homogeneous country and is well-suited for association studies. Most of our cases were recruited from the Szczecin region, which is populated by ethnic Poles who immigrated to the region from throughout Poland after the second world war, as German residents were relocated elsewhere. Our control group was drawn both from the adult population of Szczecin and from newborns in ten cities throughout Poland. However, the frequency of the CHEK2 alleles was similar in the newborn and adult controls. Also, there was no statistical difference in the CHEK2 allele frequencies in the newborns recruited from the Szczecin metropolitan region compared to other Polish cities
[1] Matsuoka S, Huang M, Elledge SJ. Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science 1998;282:1893–7. [2] Cybulski C, Huzarski T, Gorski B, et al. A novel founder CHEK2 mutation is associated with increased prostate cancer risk. Cancer Res 2004;64:2677–9. [3] Meijers-Heijboer H, van den Ouweland A, Klijn J, et al. Low-penetrance susceptibility to breast cancer due to CHEK2* 1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Genet 2002;3:55–9. [4] CHEK2 Breast Cancer Case-Control Consortium. CHEK2*1100delC and susceptibility to breast cancer: a collaborative analysis involving 10,860 breast cancer cases and 9,065 controls from 10 studies. Am J Hum Genet 2004;74:1175–82. [5] Offit K, Pierce H, Kirchhoff T, et al. Frequency of CHEK2* 1100delC in New York breast cancer cases and controls. BMC Med Genet 2003;4:1. [6] Dong X, Wang L, Taniguchi K, et al. Mutations in CHEK2 associated with prostate cancer risk. Am J Hum Genet 2003;72:270–80. [7] Seppala EH, Ikonen T, Mononen N, et al. CHEK2 variants associate with hereditary prostate cancer. Br J Cancer 2003;89:1966–70. [8] Cybulski C, Gorski B, Huzarski T, et al. CHEK2 is a multiorgan cancer susceptibility gene. Am J Hum Genet 2004;75:1131–5. [9] Menkiszak J, Brzosko M, Gorski B, et al. Ovarian cystadenoma as a characteristic feature of families with hereditary ovarian cancers unassociated with BRCA1 and BRCA2 mutations. J Appl Genet 2004;45:255–63. [10] Moslehi R, Chu W, Karlan B, et al. BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer. Am J Hum Genet 2000;66:1259–72. [11] Risch HA, McLaughlin JR, Cole DE, et al. Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer. Am J Hum Genet 2001;68:700–10. [12] Gotlieb WH, Chetrit A, Menczer J, et al. Demographic and genetic characteristics of patients with borderline ovarian tumors as compared to early stage invasive ovarian cancer. Gynecol Oncol 2005;97:780–3.