BRCA1 and BRCA2 Nucleotide Variants in Young Women with Therapy Related Acute Myeloid Leukemia.

Therapy-related acute myeloid leukemia (t-AML) is a lethal iatrogenic complication of radiation therapy and chemotherapy. It is unclear whether t-AML represents a truly stochastic event, or if individuals have different degrees of susceptibility. It has been shown that women diagnosed with breast cancer younger than 50 are at the highest risk of developing subsequent AML compared to age-matched controls (RR 4.14). This may either reflect more aggressive therapy given to younger women or differences in their ability to tolerate chemotherapy, perhaps through germline polymorphisms in DNA repair or drug metabolism genes. BRCA1 and BRCA2 genes are involved in DNA repair and germline variants confer an increased risk of breast cancer at a young age (median age at diagnosis 40 and 41, respectively) as well as an increased risk of ovarian cancer. The age-dependent risk of t-AML in breast cancer survivors mirrored the risk of a subsequent diagnosis of ovarian cancer, suggesting a common genetic factor may be responsible. Therefore, we hypothesized that young women with breast cancer that develop t-AML would have an increased prevalence of BRCA1 or BRCA2 mutations compared to the predicted prevalence of mutations based on age, personal and family history.

We identified 13 women under the age of 50 that were treated at our institution for t-AML following a diagnosis of breast cancer (median age = 43.5). Each chart was reviewed for family and personal history of breast cancer and ovarian cancer. The predicted prevalence of mutations in BRCA1 and BRCA2 were calculated from tables developed by Myriad Genetic Laboratories and the number of expected cases was calculated with a weighted sum. Normal DNA obtained from a skin biopsy was sequenced for BRCA1 and BRCA2 in 11 women and bone marrow DNA was sequenced in 1 woman in whom a skin sample was unavailable. One woman had a known deleterious mutation in BRCA2 and only the exon containing the mutation was resequenced. Total exonic sequencing, including splice sites, was performed (excluding BRCA2 exon 12, and partial coverage of BRCA2 exon 20 that failed primer design). There are no described deleterious mutations in exon 12 in the On-line Breast Cancer Mutation Data Base (BIC) (http://research.nhgri.nih.gov/ projects/bic/). Statistical differences between the observed versus expected frequencies of nucleotide variants were compared using the Chi-square goodness-of-fit test. Functional significance of nucleotide variants were determined using BIC, SIFT (http://blocks.fhcrc.org/sift/SIFT.html), FastSNP (http://fastsnp.ibms.sinica.edu.tw/pages/-input_CandidateGeneSearch.jsp) and PolyPhen (http://genetics.bwh.harvard.edu/pph/) computational algorithms.

The expected number of patients harboring mutant BRCA1 or BRCA2 alleles was 1.46 in these 13 patients (11%). We identified 4 patients with sequence variants predicted to alter function in the 13 patients (31%), which represents a statistically significant increase over the expected frequency (p=0.026). Of the 4 mutations, two (BRCA2 - S1760X and BRCA2 – F1182X) are known to be deleterious in BIC, and one (BRCA1 - Q356R, found in 2 patients) is of uncertain significance in BIC, and is computationally predicted to disrupt protein function by SIFT, FastSNP, and PolyPhen algorithms. The functional consequence of each nucleotide change is being tested using DNA repair activity assays in isogenic cell lines expressing either the wild-type or mutant allele.

In this study, we observed an increased prevalence of mutations in BRCA1 and BRCA2 in young women with breast cancer treated with chemotherapy and/or radiotherapy that developed t-AML, suggesting that alterations in DNA repair genes may be important for the development of t-AML. A case-control study is being performed to validate this finding in a larger group of uniformly treated breast cancer patients.

No relevant conflicts of interest to declare.