Risk of Breast Cancer in Women With a CHEK2 Mutation With and Without a Family History of Breast Cancer

Purpose

To estimate the risk of breast cancer in a woman who has a CHEK2 mutation depending on her family history of breast cancer.

Patients and Methods

Seven thousand four hundred ninety-four BRCA1 mutation–negative patients with breast cancer and 4,346 control women were genotyped for four founder mutations in CHEK2 (del5395, IVS2+1G>A, 1100delC, and I157T).

INTRODUCTION

The CHEK2 gene has been found to be a breast cancer–predisposing gene in several populations. In 2002, a single recurrent mutation in the CHEK2 gene (1100delC) was first reported as a cause of breast cancer, and numerous studies have confirmed this association. In Eastern and Northern Europe, the population frequency of the 1100delC allele ranges from 0.2% to 1.4%, but the allele is less frequent in North America. Three other founder variants of CHEK2 (IVS2+1G>A, del5395, I157T) have been associated with breast cancer in Eastern Europe. Two of these (IVS2+1G>A and del5395) are protein-truncating mutations, and one (I157T) is a missense variant. Both truncating mutations are associated with breast cancer in the Slavic populations of Poland, Belarus, Russia, and the Czech Republic. The I157T variant has a wider geographic distribution and has been associated with breast cancer in Poland, Finland, Germany, and Belarus.

It has been estimated that a mutation in CHEK2 increases the risk of breast cancer by between 1.5- and three-fold. A meta-analysis of all association studies estimated the risk of breast cancer among carriers of 1100delC to be increased by 2.7-fold. However, the risk of breast cancer for a woman with a CHEK2 mutation is not determined solely by the presence of the mutation; penetrance is also dependent on her family history of cancer. That is, the risk for a woman with a 1100delC mutation and a strong positive family history of breast cancer is greater than that of a carrier of the same mutation who has little or no family history of breast cancer. For this reason, CHEK2 has been proposed to be a risk modifier of other susceptibility genes. When counseling an unaffected patient with a CHEK2 mutation, it is important to take into consideration both the family history and the genotype. However, precise risk estimates for CHEK2 mutations in association with a given family history of breast cancer are not available. In particular, no risk estimates are available for the counseling of a woman with a CHEK2 mutation who has no affected relative. In this study, we investigate the risks of breast cancer associated with mutations of CHEK2 in the context of different family histories.

PATIENTS AND METHODS

Patients

Patient cases consisted of 7,931 prospectively ascertained unselected patients with invasive breast cancer who were diagnosed from 1996 to 2006 at 18 different hospitals in Poland. The study was initiated in Szczecin in 1996 and was extended to Poznan in 1997 and Olsztyn in 1998. Fifteen other centers began recruiting patients in 2001. All patients diagnosed with invasive breast cancer at participating centers were eligible. Patients with purely intraductal or intralobular cancer were excluded (ductal carcinoma in situ or lobular carcinoma in situ), but patients with ductal carcinoma in situ with microinvasion were included. In seven centers, patients with breast cancer diagnosed at all ages were recruited (n = 3,737; age range, 22 to 92 years), and at 11 centers, only patients diagnosed at age 50 years or younger were recruited (n = 4,194; age range, 21 to 50 years). Patients were unselected for family history. Altogether 10,545 women with breast cancer were invited to participate; of these, 7,931 women (75.2%) were enrolled. All 7,931 women were tested for the presence of three common founder alleles in BRCA1 (C61G, 4153delA, and 5382insC), which account for 90% of all BRCA1 mutations seen in the Polish population. 4153delA and 5382insC were detected using allele-specific oligonucleotide polymerase chain reaction (PCR), and C61G was detected using restriction fragment length polymorphism PCR. Four hundred thirty-five women carried one of three BRCA1 Polish founder mutations (C61G, n = 141; 4153delA, n = 27; 5382insC, n = 294) and were excluded. In total, there were 7,496 BRCA1 mutation–negative women with breast cancer (age range, 21 to 92 years; mean, 49.7 years); 5,152 patients were diagnosed at age 50 years or younger, and 2,344 patients were diagnosed at age greater than 50 years.

A family history was taken either by the construction of a family tree or the completion of a questionnaire. All first- and second-degree relatives diagnosed with invasive breast cancer and the age of diagnosis were recorded. The family history questionnaire was filled in by a patient or by a nurse. The detailed family tree was drawn by a physician. One thousand four hundred fifty-one women reported a first- or second-degree relative with breast cancer. In addition, information was recorded on stage, grade, lymph node status, estrogen receptor status, multicentricity, and bilaterality for more than 70% of patients. Women with a previous contralateral breast cancer or with a current diagnosis of bilateral cancer were considered to be bilateral. Patients with previous ipsilateral cancers were excluded. The vital status was available for 5,337 (71.2%) of 7,496 women. The year of death was obtained from the Polish Ministry of the Interior and Administration in April 2010. The study was approved by the Ethics Committee of Pomeranian Medical University in Szczecin, Poland.

Genotyping

DNA was isolated from 5 to 10 mL of peripheral blood. Four founder mutations in CHEK2 were genotyped as described previously. In brief, the CHEK2 del5395 mutation was detected by a multiplex PCR reaction. The IVS2+1G>A and I157T variants in CHEK2 were detected using restriction fragment length polymorphism PCR analysis, and the 1100delC mutation was analyzed using allele-specific oligonucleotide PCR.

Controls

The control group included 4,346 cancer-free women age 19 to 91 years (mean age, 52.2 years). These controls were derived from four sources. The first series consisted of 705 unselected young women (age range, 18 to 34 years; mean age, 22.0 years) from the Szczecin metropolitan region who submitted a blood sample for paternity testing between 1994 and 2001. The second series consisted of 1,079 unselected women (age range, 15 to 91 years; mean age, 58.3 years) selected at random from the computerized patient lists of five large family practices located in the region of Szczecin. These women were invited to participate by mail and participated in 2003 and 2004. The third subgroup consisted of 1,141 women from the region of Szczecin (age range, 24 to 84 years; mean age, 54.0 years) who were chosen for this study to be age and geographically matched with a series of patients with incident breast cancer diagnosed in Szczecin between 1996 and 2004. These women were part of a population-based study of the 1.5 million residents of West Pomerania designed to identify familial aggregations of cancer and were interviewed in 2007. Women with any cancer diagnosed in a first-degree relative were excluded from this control group. The fourth series included 1,421 Polish women (age range, 56 to 66 years; mean age, 59.7 years), who participated in population colonoscopy screening program for colorectal cancer between 2005 and 2010 and provided blood samples for DNA analysis (1,014 women were from Szczecin, 318 were from Białystok, and 89 were from Łódź). The allele frequencies for the CHEK2 variants were similar in controls derived from the four sources. The frequency of I157T in our controls (4.9%) is similar to that reported by Brennan et al in a nonoverlapping series of 790 controls from Poland (5.6%). The frequency of 1100delC in our controls and in controls genotyped by Brennan et al is almost identical (0.2%); Brennan et al did not genotype other truncating mutations of CHEK2 (IVS2+1G>A, del5395). 

Statistical Analysis

The prevalence of each of the four CHEK2 alleles in patient cases and population controls was compared. The odds ratios (ORs) were used to estimate relative risk. ORs were generated from two-by-two tables, and statistical significance was assessed using the Fisher's exact test. ORs were constructed for the patient cases, subdivided by the number of breast cancers reported in first- and second-degree relatives.

RESULTS

A truncating CHEK2 mutation (IVS2+1G>A, 1100delC, or del5395) was present in 227 (3.0%) of 7,496 women with breast cancer and in 37 (0.8%) of 4,346 controls. The OR for breast cancer, given a truncating mutation, was 3.6 (95% CI, 2.6 to 5.1). The OR was 3.5 (95% CI, 2.5 to 5.1) for women diagnosed at age ≤ 50 years and 3.8 (95% CI, 2.6 to 5.7) for women diagnosed at age greater than 50 years. A missense mutation I157T was detected in 535 patients with breast cancer (7.1%) and 215 (4.9%) of 4,346 controls. The OR for breast cancer, given the I157T mutation, was 1.5 (95% CI, 1.2 to 1.7). The OR was 1.4 (95% CI, 1.2 to 1.7) for women diagnosed at age ≤ 50 years and 1.6 (95% CI, 1.3 to 2.0) for women diagnosed at age greater than 50 years.

A truncating CHEK2 mutation was present in 60 (4.1%) of 1,451 patients with a family history of breast cancer (OR, 5.0; 95% CI, 3.3 to 7.6), in 167 (2.8%) of 6,045 patients with no family history of breast cancer (OR, 3.3; 95% CI, 2.3 to 4.7), and in 37 (0.8%) of 4,346 controls. The missense mutation was present in 115 (7.9%) of 1,451 patients with a family history of breast cancer (OR, 1.6; 95% CI, 1.3 to 2.1), in 420 (6.9%) of 6,045 patients with no family history of breast cancer (OR, 1.4; 95% CI, 1.2 to 1.7), and in 215 (4.9%) of 4,346 controls.

Among women with a first-degree relative affected by breast cancer, the OR for breast cancer, given a truncating mutation, was 5.7 (95% CI, 3.6 to 9.2). The OR was similar for women with an affected mother (OR, 6.1; 95% CI, 3.5 to 10.7) and women with an affected sister (OR, 5.9; 95% CI, 3.2 to 10.6). For carriers of the I157T missense mutation with a first-degree relative affected, the OR for breast cancer was 1.9 (95% CI, 1.4 to 2.5). The OR, given the I157T mutation, was 1.9 (95% CI, 1.3 to 2.8) for women with an affected mother and 1.8 (95% CI, 1.2 to 2.7) for women with an affected sister.

We also investigated how a history of breast cancer in a second-degree relative influences the risk  a. Among women with a second-degree relative affected, the OR for breast cancer, given a truncating mutation, was 4.7 (95% CI, 2.9 to 7.6). The OR was similar for women with a second-degree relative affected on the mother's side (maternal aunt or grandmother; OR, 5.0; 95% CI, 2.9 to 8.8) and women with a second-degree relative affected on the father's side (OR, 4.1; 95% CI, 2.2 to 7.9). For carriers of the I157T missense mutation with a second-degree relative affected, the OR for breast cancer was 1.5 (95% CI, 1.1 to 2.0). The OR was higher for women with a second-degree relative affected on the father's side (OR, 2.1; 95% CI, 1.4 to 3.0) than for women with a second-degree affected on the mother's side (OR, 1.0; 95% CI, 0.6 to 1.6). In this data set, there was no evidence for an increased risk if the I157T mutation was inherited through an unaffected mother, but the numbers in the subgroups were relatively small, and the interaction was not statistically significant (P = .07). The highest OR was observed for women with a first- and second-degree relative affected with breast cancer, both for the truncating mutations (OR, 7.3; 95% CI, 3.2 to 16.8) and for the missense mutation (OR, 2.0; 95% CI, 1.05 to 3.7).

The missense mutation I157T has less impact on breast cancer risk than the truncating mutations. This mutation has also been associated with breast cancer in Finland, Germany, and Belarus. The OR for breast cancer given the I157T mutation was 1.4 (95% CI, 1.1 to 1.9) in the Finish population, 3.6 (95% CI, 1.1 to 12.2) in the German population, and 4.5 (95% CI, 1.6 to 13.2) in the Byelorussian population. In our study, the OR for breast cancer associated with the I157T mutation was approximately 1.5 for unselected patients and 2.0 for familial patients (two first-degree relatives with breast cancer); this corresponds to a lifetime risk of breast cancer of 9% and 12%, respectively, in the Polish population. Therefore, the risk is increased (over the population risk of 6%), but there is no consensus of how women at this risk level should be managed.

Our data provide further support for the model that both family history and a CHEK2 mutation are risk factors for breast cancer that operate in concert. In the present study, we were unable to measure the effect of family history per se (ie, in the absence of a CHEK2 mutation) on the risk of breast cancer, but in a large meta-analysis, the OR for breast cancer given one first-degree relative affected was 1.8 (99% CI, 1.69 to 1.91). Therefore, the addition of the CHEK2 mutation to the family history augments the risk. This observation supports the hypothesis that the CHEK2 mutation modifies the risk of cancer imparted by other cancer genes. On the basis of the similar magnitude of the risks observed for sisters and mothers, it is expected that, in concert with truncating mutations, these other genes act in a dominant (or multifactorial) model. Interestingly, we saw a higher risk of cancer associated with the missense mutation if the cancer in a relative was on the father's side (OR, 2.1; P < .001) than on the mother's side (OR, 1.0; P = .9). Although the numbers are too small to be conclusive, it is interesting to speculate that there may be a role for imprinting of the I157T mutation or that there may be an X-linked modifier.

Date : 27/07/2018

Lo Staff / The Staff 

American Europen Medical Center 

(americaneuropeanmedicalcenter.blogspot.com) 

Direttore del sito : Daniel Viennese