Cancer in Heterozygote Carriers of Fanconi Anemia Genes

Background: Fanconi anemia (FA) is an inherited bone marrow failure syndrome, characterized by a defect in DNA repair, increased frequency of birth defects, and high risks of malignancies. Relatives of patients with FA are concerned about the risk of cancer in themselves or other family members. Carrier grandparents with mutations in genes for FA were reported in 2007 to have no increase in overall cancer incidence, except for a higher rate of breast cancer among FANCC carrier grandmothers (Berwick et al, Cancer Research 67:9591, 2007); this study included 944 relatives among 312 families. Tischkowitz et al also did not find increased risks of cancer in 575 relatives among 36 families, except for prostate cancer (Tischkowitz et al, BMC Cancer 8:257, 2008). The relative risk of breast cancer associated with germline mutations in genes in the BRCA/FA DNA repair pathway in the general population not due to BRCA1 or BRCA2 is 5 to 10 fold (Couch et al, JAMA Oncology 3:1190, 2017), suggesting that heterozygotes for FA might be at increased risk. The FA cohort at the National Cancer Institute (NCI) is sufficiently large to again address the question of cancer in FA heterozygotes.

Hypothesis: Patients heterozygous for a mutated gene in the BRCA/FA DNA repair pathway have an increased risk of cancer. Identification of FA genes with increased cancer risk in heterozygotes, and specific cancer types in relatives of patients with FA, would guide screening recommendations for these categories of family members, and studies to determine whether FA heterozygotes with cancer should be managed with modifications of drug or radiation doses.

Methods: We evaluated cancer risks in relatives of probands in the NCI FA cohort. We collected self-report or proxy report data on the presence or absence of cancer, cancer type, age at cancer diagnosis for individuals with a history of cancer, and age at death from other causes or current age. We studied obligate heterozygotes (parents and offspring of probands with FA), relatives with a 50% probability of heterozygosity (grandparents), and relatives with a 2/3 probability of being carriers (siblings without FA). We determined the ratio of observed/expected cases (O/E) using data from SEER cancer registries, with adjustment for age, sex, and birth cohort. Confirmatory genotyping is being done by targeted panel next-generation sequencing for participants with DNA available.

Results and Discussion: Our study includes 94 families. Twenty-one cancers were reported in 188 parents, with 24 expected (O/E 0.86, 95% CI [confidence interval] 0.53-1.31). Three hundred and twenty-seven grandparents had 89 cases with cancer, expected 125 (O/E 0.71, CI 0.57-0.88). There were 122 siblings, with 1 cancer case and 4 expected (O/E 0.26, CI 0.01-1.46). Among 12 offspring there was 1 case of leukemia (O/E 1:0.12, CI 0.2-44.64). Specific cancers had increased O/E in parents (2 cases of salivary gland cancer, O/E 27.3, CI 3.31-98.66), grandparents (5 with liver cancer O/E 4.7, CI 1.53-10.96; 9 with leukemia O/E 2.42, CI 1.11-4.59), and a single case of acute myeloid leukemia in an offspring (O/E of 402.44, CI 10.19-2242.85). Four cases of prostate cancer in parents (O/E of 1.47, CI 0.4-3.76) and 8 in grandparents (O/E 0.47, CI 0.2-0.93) failed to support the suggestion by Tischkowitz et al that prostate cancer was significantly increased in male relatives (5 cases, O/E 3.1, CI 1.1-8.8). Missing data include birth dates, death dates, or cancer dates for 44 grandparents and 2 siblings.

Our results suggest that heterozygotes for mutations in FA genes do not appear to have an increased risk of cancer compared with the general population, in agreement with data from other cohorts. Further analyses will be done after completion of genotyping the relatives, with the caveat that DNA may not be available from some of the grandparents. The NCI FA study is limited by the numbers of participants, and the missing data, particularly in the grandparent generation. In addition, data are reported by the individuals or their proxies, and medical record validation may not be possible; the types and ages of cancers cannot be confirmed. Although the types of cancer and risks in FA heterozygotes are similar to those in the general population, the potential concern that management of patients who are carriers of a defect in DNA repair may need to be modified requires future investigation.