Highlights
- •Compared phenotypes of founder and non-founder CHEK2 mutation carriers.
- •No significant phenotypic differences were observed between founders and non-founders.
- •CHEK2 founder mutation cancer risks may be generalizable to all CHEK2 mutations.
CHEK2 mutations are associated with increased cancer risks, including breast; however, published risk estimates are limited to those conferred by CHEK2 founder mutations, presenting uncertainty in risk assessment for carriers of other CHEK2 mutations. This study aimed to assess phenotypes and molecular characteristics of CHEK2 mutation carriers (CHEK2 + s) from a multi-gene panel testing (MGPT) cohort, focusing on comparing phenotypes of founder and non-founder CHEK2 + s. Clinical histories and molecular results were reviewed from 45,879 patients who underwent MGPT including CHEK2 at a commercial laboratory. Of individuals tested, 2.4% (n = 1085) were CHEK2 + s. Sixteen individuals harbored biallelic CHEK2 mutations, bringing the total number of CHEK2 mutations detected in this cohort to 1101. Personal/family cancer histories were compared between founder (n = 576; included c.1100delC, p.S428F, c.444 + 1G > A, and EX8_9del) and non-founder (n = 259) CHEK2 + s using Fisher's exact test and multivariate logistic regression analysis. Individuals carrying the p.I157T moderate risk founder mutation (n = 231), additional mutations in non-CHEK2 genes (n = 83), or biallelic mutations (n = 16) were excluded from phenotype analysis, as were cases with no clinical information provided. No significant phenotypic differences were observed between founder and non-founder CHEK2 + s. These data suggest that cancer risks reported for founder mutations may be generalizable to all CHEK2 + s, particularly for breast cancer.
Keywords
Introduction
Mutations in the CHEK2 gene are associated with increased risks of developing breast and other cancers. While the role of CHEK2 in cancer susceptibility has been generally well studied, published cancer risk estimates are limited to those conferred by the more common CHEK2 founder mutations (Supplemental Table S1). The c.1100delC Eastern European founder mutation is the most comprehensively studied CHEK2 mutation. Results from large case-control studies have shown a 2-fold increased risk for breast and colorectal cancers (
1
, 2
) with cancer risks further increased among individuals with family histories of these cancers. After initial reports of c.1100delC's association with cancer, several other European founder mutations have been reported including c.444 + 1G > A, p.I157T, and EX8_9del, a ~5.5 kb deletion spanning coding exons 8 and 9 (exons 9 and 10). The c.444 + 1G > A and EX8_9del mutations have been associated with increased breast, prostate, gastric and thyroid cancer risks (3
, 4
, 5
, 6
, 7
, 8
). The p.I157T mutation has been associated with increased risks of breast, colon, kidney, prostate, and thyroid cancers; however, the female breast cancer risk of approximately 1.5-fold is attenuated compared to other CHEK2 mutations (3
, 9
, 10
). An additional founder mutation, p.S428F, has been identified in the Ashkenazi Jewish population and is reported to confer an approximate 2-fold increase in breast cancer risk among female carriers (11
).In a number of CHEK2 cancer risk studies, CHEK2 genotyping was limited to one or more founder mutations for at least part of the study cohort, and limited to the founder ethnic group. As such, little is known about the full spectrum of CHEK2 mutations in cancer cohorts across diverse ethnic groups, and cancer risk estimates for CHEK2 mutation carriers (CHEK2 + s) of non-founder mutations remain largely undefined. With the clinical availability of multi-gene panel testing (MGPT) for hereditary cancer, an increased number of CHEK2 + s is being reported. In multiple breast/ovarian cancer MGPT cohort studies, CHEK2 mutations have accounted for 15–33% of non-BRCA1/2 mutations identified (
12
, 13
, 14
, 15
).In a recent update to the National Comprehensive Cancer Network (NCCN) Genetic/Familial High-Risk Assessment Guideline, screening breast MRIs are recommended for CHEK2 + s as their lifetime breast cancer risk exceeds 20% (
16
). It is currently unclear whether these recommendations are appropriate for all CHEK2 + s. In this study, we aimed to assess the phenotypes and molecular characteristics across a full spectrum of CHEK2 mutations from a multi-gene cancer panel cohort, with a focus on comparing phenotypes of founder and non-founder mutation carriers.Methods
Study population
Clinical histories and molecular test results were reviewed for all patients who underwent MGPT including CHEK2 between March 2012 and June 2015, regardless of personal cancer history (n = 45,879). The following information was extracted from test requisition forms and clinic notes submitted by ordering providers: gender, age at testing, ethnicity, and personal/family cancer history.
Laboratory methods
MGPT was performed as previously described (
17
). In summary, genomic deoxyribonucleic acid (gDNA) was isolated from whole blood or saliva samples using QIAsymphony DNA kits (Qiagen, Valencia, CA) and then quantified using a spectrophotometer (Nanodrop; Thermo Scientific, Pittsburgh, PA, or Infinite F200; Tecan, San Jose, CA). Sequence enrichment was performed (RainDance Technologies, Billerica, MA), followed by next-generation sequence (NGS) analysis (Illumina, San Diego, CA) of all coding domains plus at least five bases into the 5′ and 3′ ends of all introns and untranslated regions (5'UTR and 3′UTR) of 14–49 cancer susceptibility genes, depending on the panel ordered. Sanger sequencing was performed for any region with insufficient depth of coverage for reliable heterozygous variant detection (<10×) and for verification of all variant calls other than known nonpathogenic alterations. A targeted chromosomal microarray designed with increased probe density in regions of interest was used for the detection of gross deletions and duplications for each sample (Agilent, Santa Clara, CA). Initial data processing and base calling, including extraction of cluster intensities, were done using RTA 1.12.4 (HiSeq Control Software 1.4.5; Illumina). Sequence quality filtering was executed with the CASAVA software (version 1.8.2; Illumina, Hayward, CA). Sequence fragments were aligned to the reference human genome (GRCh37), and variant calls were generated using CASAVA. A minimum quality threshold of Q20 was applied, translating to an accuracy of >99.9% for called bases. Variants were annotated with the Ambry Variant Analyzer, a proprietary alignment and variant annotation software (Ambry Genetics). All variants, with the exception of previously characterized benign alterations, underwent thorough assessment and review of available evidence (e.g., population frequency information, published case reports, case/control and functional studies, internal co-occurrence and co-segregation data, evolutionary conservation, and in silico predictions). Variants were further classified per Ambry's five-tier variant classification protocol (pathogenic mutation; variant, likely pathogenic (VLP); variant of unknown significance (VUS); variant, likely benign (VLB); and benign), which is based on published recommendations/guidelines by the American College of Medical Genetics and Genomics and the International Agency for Research on Cancer (18
, 19
).Data analysis
CHEK2 pathogenic mutation and VLP carriers were further classified as “founder mutation carriers” if they harbored the c.1100delC, p.I157T, p.S428F, EX8_9del, or c.444 + 1G > A pathogenic mutations or as “non-founder mutation carriers” if they harbored any other CHEK2 alterations classified as pathogenic or VLP (
17
). The overall frequencies of founder, non-founder and all CHEK2 mutations were calculated and compared between Caucasians and non-Caucasians. Phenotypes of founder and non-founder mutation carriers were compared based on personal history of any cancer, multiple primary cancers, and history of the following cancer types: breast (any, female, male and multiple breast primaries), colorectal, ovarian, endometrial, thyroid, kidney, prostate, pancreatic, leukemia, lymphoma, brain, and gastric. History of these cancer types among first, second, or third-degree relatives was also compared between the two groups. The reduced penetrance p.I157T mutation was excluded from phenotype comparisons. CHEK2 + s lacking clinical history information, and individuals harboring biallelic CHEK2 mutations or mutations in additional cancer susceptibility genes were excluded from the analyses.Additional personal and family history comparisons were performed for the CHEK2 + s cohort (excluding p.I157T) vs. MGPT negative controls (MGPT-) as a means of validating the ability to detect significant differences within a highly selected cancer cohort. Subsequent comparisons of p.I157T vs. other CHEK2 + s and MGPT- were performed. Statistical analyses were performed using the Fisher's exact test and multivariate logistic regression analysis adjusted for multiple comparisons via false discovery rate (FDR cutoff of 5% applied) estimation, controlling for age at testing, MGPT ordered, ethnicity, and gender. The final cohort for phenotype comparisons consisted of 31,080 individuals.
Results
CHEK2 positive cohort
The overall frequency of CHEK2 + s in this MGPT cohort was 1085/45,879 (2.4%), including 83 individuals carrying additional mutations in non-CHEK2 genes and 16 individuals harboring biallelic CHEK2 mutations. The majority of CHEK2 + s were female (n = 1001, 92.3%) and Caucasian (n = 824, 75.9%). Other ethnicities included African American (n = 8, 0.7%),Ashkenazi Jewish (n = 120, 11.1%), Asian (n = 10, 0.9%), Hispanic (n = 14, 1.3%), Middle Eastern (n = 6, 0.6%), Mixed Ethnicity (n = 31, 2.9%), “Other” (n = 6, 0.6%), and Native American (n = 1, 0.1%). Ethnicity was not provided for 65 participants (6.0%).
Of the 1101 CHEK2 mutations identified, 841 (76.4%) were founder mutations, and 260 (23.6%) were non-founder mutations (Table 1). The most commonly detected founder mutations were c.1100delC (n = 416) and p.I157T (n = 258), which represented 80.1% of the total number of founder mutations observed. Founder mutations accounted for 96.7% of Ashkenazi Jewish, 76.9% of Caucasian, 57.1% of Hispanic, and 100% of Native American CHEK2 + s. Additionally, founder mutations accounted for 64.5% of mixed ethnicity, 66.2% of unknown ethnicity, and 66.7% of “other” ethnicity CHEK2 + s. Conversely, founder mutations accounted for 37.5% of CHEK2 mutations in African Americans and were not detected among Middle Eastern or Asian subjects.
Table 1CHEK2 mutation spectrum
| Total | Ashkenazi Jewish | Caucasian | Middle Eastern | Hispanic | African American | Asian | Native American | Other | Mixed ethnicity | Unknown | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | n | n | n | n | n | n | n | n | n | n | n | n | n | n | |
| Total tested | – | 45,879 | – | 3,039 | – | 31,196 | – | 262 | 2,129 | 2,335 | 1,630 | 54 | 162 | 2,016 | 3,056 |
| Any CHEK2 mutation | 1,101 | 1,085 | 122 | 120 | 837 | 824 | 7 | 6 | 14 | 8 | 10 | 1 | 6 | 31 | 65 |
| CHEK2 founders | 841 | 831 | 118 | 116 | 644 | 636 | 0 | 0 | 8 | 3 | 0 | 1 | 4 | 20 | 43 |
| c.1100delC | 416 | 412 | 21 | 21 | 352 | 348 | 0 | 0 | 6 | 3 | 0 | 1 | 2 | 14 | 17 |
| p.S428F | 108 | 108 | 82 | 82 | 20 | 20 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 4 |
| EX8_9del | 30 | 30 | 1 | 1 | 25 | 25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 |
| c.444 + 1G > A | 29 | 29 | 0 | 0 | 27 | 27 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.I157T | 258 | 257 | 14 | 14 | 220 | 219 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 5 | 17 |
| CHEK2 non-founders | 260 | 259 | 4 | 4 | 193 | 193 | 7 | 6 | 6 | 5 | 10 | 0 | 2 | 11 | 22 |
| Recurrent | 221 | 220 | 3 | 3 | 165 | 165 | 7 | 6 | 4 | 4 | 6 | 0 | 2 | 11 | 19 |
| p.T476M | 67 | 66 | 3 | 3 | 46 | 46 | 6 | 5 | 0 | 0 | 0 | 0 | 2 | 3 | 7 |
| p.R117G | 40 | 40 | 0 | 0 | 35 | 35 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 3 |
| c.1263delT | 12 | 12 | 0 | 0 | 11 | 11 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.R145W | 9 | 9 | 0 | 0 | 7 | 7 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.G167R | 8 | 8 | 0 | 0 | 6 | 6 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0 |
| c.591delA | 8 | 8 | 0 | 0 | 7 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| p.G306A | 7 | 7 | 0 | 0 | 6 | 6 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.R95* | 6 | 6 | 0 | 0 | 4 | 4 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| c.1368dupA | 6 | 6 | 0 | 0 | 5 | 5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.H371Y | 5 | 5 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0 | 0 | 1 | 0 |
| p.R519* | 5 | 5 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 |
| EX2_3del | 5 | 5 | 0 | 0 | 4 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| c.1567delC | 4 | 4 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| c.277delT | 4 | 4 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| p.R137* | 3 | 3 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 |
| p.Y390S | 3 | 3 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| p.W93* | 3 | 3 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| c.908 + 1G > A | 3 | 3 | 0 | 0 | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| c.683 + 1G > T | 3 | 3 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| 5'UTR_EX1del | 3 | 3 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 |
| c.793-1G > A | 3 | 3 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0 | 0 | 0 | 0 | 0 | 0 |
| c.1434delA | 2 | 2 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| c.247delC | 2 | 2 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 |
| EX3_4del | 2 | 2 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| c.1462-2A > G | 2 | 2 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| c.319 + 2T > A | 2 | 2 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| 5′UTRdel | 2 | 2 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| 5′UTR_EX14del | 2 | 2 | 0 | 0 | 2 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Non-recurrent | 39 | 39 | 1 | 1 | 28 | 28 | 0 | 0 | 2 | 1 | 4 | 0 | 0 | 0 | 3 |
a Coding exon nomenclature.
b Number of CHEK2 mutations detected (when number of mutations and number of individuals is equal, this designation is used).
c Number of CHEK2 mutation-positive individuals (if different from number of mutations detected).
Overall, CHEK2 mutations were significantly more frequent among Ashkenazi Jews and Caucasians than non-Caucasians/non-Ashkenazi Jews (OR = 4.63, p = 1.48e-32, 95%CI [3.36, 6.56]), with the frequency of founder mutations greater between these groups as well (OR = 11.97, p = 1.29e-40, 95%CI [6.81, 23.29]). The c.1100delC founder mutation was more frequent among Caucasians than non-Caucasians/non-Ashkenazi Jews (OR = 14.75, p = 5.83e-14, 95%CI [4.98, 72.11]). The p.S428F founder mutation was more frequent among Ashkenazi Jews than non-Ashkenazi Jews (OR = 52.16, p = 2.97e-73, 95%CI [31.60, 89.97]). One recurrent non-founder mutation, p.H371Y, was less frequent among Caucasians than non-Caucasians (OR = 0.00, p = 0.003, 95%CI [0.00, 0.46]). This may represent an Asian founder mutation since 4 of 5 carriers reported Asian ancestry, although background haplotype analysis was not performed.
Phenotype comparisons
When compared to MGPT-, CHEK2 + s (excluding p.I157T) were significantly more likely to have a personal history of any cancer, multiple primaries, female and multiple primary breast cancers, and thyroid cancer, and were significantly less likely to have a personal history of ovarian cancer (Table 2). CHEK2 + s were also significantly more likely to have a family history of female breast cancer, leukemia, thyroid and prostate cancers (Table 2). No significant differences were observed between personal and family cancer histories of founder and non-founder CHEK2 + s (Table 2).
Table 2Phenotype comparisons across individuals tested for CHEK2 (excluding p.I157T carriers)
| Positive | Negative | Founder | Non-founder | Positive vs. negative | Founder vs. non-founder | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Personal history | % | p-value | OR | 95% CI | Multivariate p-value | p-value | OR | 95% CI | Multivariate p-value | |||
| Any cancer | 85.0% | 77.9% | 85.5% | 83.9% | 1.97e-6 | 1.61 | [1.31, 2.00] | 8.60e-06 | 0.58 | 1.13 | [0.71, 1.77] | 0.99 |
| Multiple primary cancers | 24.0% | 17.8% | 25.1% | 21.4% | 0.00 | 1.32 | [1.10, 1.58] | 3.18e-05 | 0.38 | 1.21 | [0.81, 1.83] | 0.84 |
| Breast | 70.3% | 56.9% | 69.9% | 71.0% | 2.20e-13 | 1.79 | [1.52, 2.11] | 5.99e-14 | 0.79 | 0.95 | [0.66, 1.36] | 0.99 |
| Multiple primary breast | 12.6% | 8.3% | 13.6% | 10.3% | 0.00 | 1.59 | [1.26, 1.99] | 8.32e-05 | 0.23 | 1.37 | [0.82, 2.38] | 0.98 |
| Female breast | 74.9% | 61.0% | 74.8% | 75.1% | 4.16e-14 | 1.91 | [1.60, 2.29] | 1.80e-13 | 1.00 | 0.99 | [0.66 1.46] | 0.99 |
| Male breast | 12.7% | 7.5% | 12.5% | 13.3% | 0.19 | 1.80 | [0.68, 4.08] | 0.07 | 1.00 | 0.93 | [0.13, 10.91] | 0.99 |
| Colorectal | 6.7% | 8.6% | 7.5% | 4.9% | 0.07 | 0.76 | [0.56, 1.02] | 0.10 | 0.26 | 1.56 | [0.76, 3.46] | 0.84 |
| Ovarian | 6.0% | 9.1% | 6.0% | 6.2% | 0.01 | 0.64 | [0.46, 0.88] | 0.01 | 0.86 | 0.96 | [0.47, 2.06] | 0.98 |
| Endometrial | 2.5% | 3.4% | 2.6% | 2.4% | 0.24 | 0.73 | [0.42, 1.18] | 0.28 | 1.00 | 1.07 | [0.35, 3.94] | 0.99 |
| Thyroid | 3.7% | 2.1% | 3.3% | 4.5% | 0.01 | 1.77 | [1.15, 2.62] | 0.01 | 0.52 | 0.74 | [0.31, 1.84] | 0.84 |
| Kidney | 1.8% | 1.0% | 1.8% | 1.8% | 0.06 | 1.76 | [0.92, 3.07] | 0.19 | 1.00 | 0.99 | [0.27, 4.46] | 0.99 |
| Prostate | 10.9% | 8.0% | 5.0% | 26.7% | 0.45 | 1.41 | [0.49, 3.35] | 0.73 | 0.04 | 0.15 | [0.01, 1.21] | 0.74 |
| Pancreas | 1.1% | 0.7% | 1.0% | 1.3% | 0.27 | 1.49 | [0.63, 3.01] | 0.49 | 0.71 | 0.73 | [0.14, 4.76] | 0.99 |
| Leukemia | 0.5% | 0.3% | 0.6% | 0.4% | 0.30 | 1.76 | [0.47, 4.69] | 0.38 | 1.00 | 1.32 | [0.11, 69.83] | 0.99 |
| Lymphoma | 0.7% | 0.8% | 1.0% | 0.0% | 0.84 | 0.81 | [0.26, 1.93] | 0.73 | 0.33 | Inf | [0.40, Inf] | 0.99 |
| Brain | 0.7% | 0.4% | 1.0% | 0.0% | 0.26 | 1.58 | [0.50, 3.82] | 0.48 | 0.33 | Inf | [0.40, Inf] | 0.99 |
| Gastric | 0.5% | 0.5% | 0.6% | 0.4% | 0.78 | 1.12 | [0.30, 2.95] | 0.79 | 1.00 | 1.32 | [0.11, 69.83] | 0.99 |
| Family history | % | p-value | OR | 95% CI | Multivariate p-value | p-value | OR | 95% CI | Multivariate p-value | |||
| Any cancer | 97.8% | 97.6% | 98.2% | 96.8% | 0.81 | 1.10 | [0.67, 1.95] | 0.76 | 0.27 | 1.82 | [0.57, 5.58] | 0.84 |
| Breast | 77.1% | 72.3% | 78.3% | 74.3% | 0.00 | 1.29 | [1.08, 1.55] | 0.01 | 0.25 | 1.25 | [0.84, 1.84] | 0.98 |
| Multiple primary breast | 11.2% | 9.1% | 11.3% | 11.0% | 0.06 | 1.26 | [0.98, 1.60] | 0.12 | 1.00 | 1.03 | [0.61, 1.79] | 0.99 |
| Female breast | 76.1% | 70.6% | 77.5% | 72.9% | 0.00 | 1.33 | [1.12, 1.59] | 0.00 | 0.18 | 1.28 | [0.87, 1.87] | 0.84 |
| Male breast | 3.1% | 2.4% | 3.6% | 1.8% | 0.26 | 1.29 | [0.80, 1.99] | 0.33 | 0.25 | 1.98 | [0.64, 8.16] | 0.84 |
| Colorectal | 37.9% | 34.7% | 38.4% | 36.7% | 0.08 | 1.15 | [0.98, 1.34] | 0.13 | 0.74 | 1.07 | [0.76, 1.52] | 0.99 |
| Ovarian | 18.9% | 22.2% | 18.3% | 20.2% | 0.04 | 0.82 | [0.67, 0.99] | 0.07 | 0.60 | 0.88 | [0.58, 1.36] | 0.84 |
| Endometrial | 9.0% | 10.5% | 10.5% | 5.5% | 0.24 | 0.85 | [0.64, 1.10] | 0.30 | 0.03 | 2.02 | [1.04, 4.24] | 0.74 |
| Thyroid | 8.2% | 5.4% | 8.9% | 6.4% | 0.00 | 1.58 | [1.18, 2.07] | 0.00 | 0.30 | 1.43 | [0.75, 2.89] | 0.84 |
| Kidney | 7.5% | 6.0% | 7.4% | 7.8% | 0.10 | 1.27 | [0.94, 1.69] | 0.17 | 0.88 | 0.94 | [0.50, 1.82] | 0.99 |
| Prostate | 26.9% | 21.8% | 27.6% | 25.2% | 0.00 | 1.32 | [1.11, 1.57] | 0.00 | 0.52 | 1.13 | [0.78, 1.66] | 0.99 |
| Pancreas | 13.3% | 13.5% | 14.3% | 11.0% | 0.96 | 0.99 | [0.78, 1.23] | 0.91 | 0.28 | 1.35 | [0.81, 2.31] | 0.84 |
| Leukemia | 13.0% | 8.6% | 13.5% | 11.9% | 0.00 | 1.59 | [1.26, 1.99] | 0.00 | 0.63 | 1.15 | [0.70, 1.95] | 0.99 |
| Lymphoma | 8.7% | 7.3% | 9.3% | 7.3% | 0.15 | 1.22 | [0.93, 1.59] | 0.21 | 0.47 | 1.30 | [0.70, 2.52] | 0.84 |
| Brain | 12.1% | 9.8% | 12.3% | 11.5% | 0.04 | 1.27 | [1.00, 1.60] | 0.09 | 0.80 | 1.08 | [0.65, 1.86] | 0.99 |
| Gastric | 10.4% | 11.0% | 9.3% | 12.8% | 0.67 | 0.94 | [0.73, 1.20] | 0.76 | 0.18 | 0.70 | [0.42, 1.20] | 0.74 |
a The p-value of the Fisher's exact test.
b The p-value of the multivariate logistic regression analysis, adjusted for multiple comparisons via false discovery rate (cutoff of 5%) estimation, controlling for age at testing, MGPT ordered, ethnicity, and gender.
No significant phenotypic differences were observed between p.I157T mutation carriers (p.I157T + s) and all other CHEK2 + s (Supplemental Table S2). When compared to MGPT-, the only observed difference was that p.I157T + s were significantly more likely to have leukemia (OR = 5.85, p = 5.94e-3, 95%CI [1.55, 15.72]; multivariate logistic regression analysis p = 0.03).
Discussion
To our knowledge, we have examined the largest CHEK2 + cohort ascertained through comprehensive CHEK2 analysis to date. While the majority of CHEK2 + s (76.4%) harbored founder mutations with published cancer risk estimates, 23.6% of patients harbored non-founder mutations for which cancer risk information is undefined. No significant differences were observed between personal and family histories of CHEK2 founder and non-founder mutation carriers, suggesting that cancer risks reported in CHEK2 founder populations may be generalizable to all CHEK2 + s. Compared to MGPT-, CHEK2 + s were more likely to have a personal history of breast or thyroid cancer; however, no significant differences were observed for other reported CHEK2-associated cancers (colorectal, prostate, and kidney). This observation can be partially attributed to an ascertainment bias of the MGPT cohort as the most common indication for testing was breast cancer, with 70.3% of CHEK2 + s and 56.9% of MGPT- reporting a personal history of breast cancer. All other cancer types assessed were reported at relatively lower frequencies (i.e. ≤10%) among CHEK2 + s and MGPT- patients. Future studies incorporating larger numbers of mutation carriers with cancers other than breast are needed to further evaluate whether risks for other cancers are similar among founder and non-founder mutation carriers.
Several other limitations should be noted. While all patients underwent comprehensive analysis of CHEK2, additional testing varied. It is possible that some MGPT- harbor mutations in genes not tested or that CHEK2 + s carry an additional non-CHEK2 mutation, although the spectrum of MGPT ordered was controlled for in multivariate logistic regression analysis. Also, grouping the founder mutations together results in over-representation of the most common mutation in this cohort (c.1100delC) and under-representation of the least common founder mutation (c.444 + 1G > A). This is reasonable for estimating breast cancer risk, as the non-p.I157T founder mutations have been associated with an increased breast cancer risk of similar magnitude, but could confound the analysis for other cancer types since the various founder mutations have somewhat differing cancer profiles. Finally, phenotype data were collected primarily from clinician-completed requisition forms, introducing potential for incomplete reporting of a patient's cancer history.
Due to the previously reported lower breast cancer risk with the p.I157T mutation relative to the other four founder mutations, p.I157T + s were analyzed separately. The observation that cancer risks among p.I157T + s were moderate, falling between the higher risks in other CHEK2 + s and lower risks in MGPT- suggests a gradient of risk, with p.I157T representing a lower penetrance allele, as expected.
Continued CHEK2 testing via MGPT will further our knowledge of the full CHEK2 mutation spectrum and allow for additional founder mutations to emerge among ethnic minorities with sample sizes historically inadequate to identify obvious trends. The data presented here supports the generalizability of cancer risks from founder populations to carriers of other CHEK2 mutations, with the exception of the p.I157T known reduced penetrance allele. Additional studies are warranted to confirm precise non-breast cancer risks, age-related penetrance and appropriate screening protocols. However, in the setting of any (non-p.I157T) clinically identified CHEK2 mutation, patients should be targeted to updated NCCN screening protocols. This provides an effective framework for supporting NCCN recommendations for intensive breast cancer surveillance among CHEK2 + s.
Appendix. Supplementary data
The following is the supplementary data to this article:
- Table S1
CHEK2 founder mutations and associated cancer risks
- Table S2
Phenotype comparisons of CHEK2 p.I157T carriers
- Table S3
Type of panel ordered
References
- CHEK2*1100delC and susceptibility to breast cancer: a collaborative analysis involving 10,860 breast cancer cases and 9065 controls from 10 studies.Am J Hum Genet. 2004; 74: 1175-1182
- Meta-analysis of CHEK2 1100delC variant and colorectal cancer susceptibility.Eur J Cancer. 2011; 47: 2546-2551
- CHEK2 is a multiorgan cancer susceptibility gene.Am J Hum Genet. 2004; 75: 1131-1135
- A novel founder CHEK2 mutation is associated with increased prostate cancer risk.Cancer Res. 2004; 64: 2677-2679
- A large germline deletion in the Chek2 kinase gene is associated with an increased risk of prostate cancer.J Med Genet. 2006; 43: 863-866
- Risk of breast cancer in women with a CHEK2 mutation with and without a family history of breast cancer.J Clin Oncol. 2011; 29: 3747-3752
- Synergistic interaction of variants in CHEK2 and BRCA2 on breast cancer risk.Breast Cancer Res Treat. 2009; 117: 161-165
- The risk of gastric cancer in carriers of CHEK2 mutations.Fam Cancer. 2013; 12: 473-478
- Association of two mutations in the CHEK2 gene with breast cancer.Int J Cancer. 2005; 116: 263-266
- CHEK2 variant I157T may be associated with increased breast cancer risk.Int J Cancer. 2004; 111: 543-547
- Functional and genomic approaches reveal an ancient CHEK2 allele associated with breast cancer in the Ashkenazi Jewish population.Hum Mol Genet. 2005; 14: 555-563
- Multigene panel testing detects equal rates of pathogenic BRCA1/2 mutations and has a higher diagnostic yield compared to limited BRCA1/2 analysis alone in patients at risk for hereditary breast cancer.Ann Surg Oncol. 2015; 22: 3282-3288
- Hereditary predisposition to ovarian cancer, looking beyond BRCA1/BRCA2.Gynecol Oncol. 2015; 137: 86-92
- Frequency of mutations in individuals with breast cancer referred for BRCA1 and BRCA2 testing using next-generation sequencing with a 25-gene panel.Cancer. 2015; 121: 25-33
- Pathogenic and likely pathogenic variant prevalence among the first 10,000 patients referred for next-generation cancer panel testing.Gen Med. 2016; 18: 823-832
- The NCCN clinical practice guidelines in oncologyTM genetic/familial high-risk assessment: breast and ovarian V2.(2015; Available at:) (Accessed 14 March, 2016)
- Utilization of multigene panels in hereditary cancer predisposition testing: analysis of more than 2000 patients.Gen Med. 2014; 16: 830-837
- Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.Genet Med. 2015; 17: 405-424
- Sequence variant classification and reporting: recommendations for improving the interpretation of cancer susceptibility genetic test results.Hum Mutat. 2008; 29: 1282-1291
Article Info
Publication History
Published online: August 15, 2016
Accepted:
August 10,
2016
Received in revised form:
July 22,
2016
Received:
March 17,
2016
Identification
Copyright
© 2016 The Authors. Published by Elsevier Inc.
User License
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
