Implementing a germline testing program using fingernail specimens for patients with hematologic malignancies and initial clinical findings.

Abstract

10632 Background: About 5-20% of hematologic malignancies (HM) occur in the setting of a genetic cancer predisposition syndrome. Additionally, overt constitutional syndromes involving multiple systems may also predispose to HM. The spectrum of cancer predisposition in patients with HM has been limited partially due to the difficulty of germline sample ascertainment from patients with liquid tumors as skin biopsies and fibroblast cultures are often required for testing. Methods: Hematologic malignancy patients enrolled in Memorial Sloan Kettering Cancer Center’s institutional 12-245 study who were having somatic testing performed on their tumor were given the option to consent to germline testing using fingernail specimens. The validated germline assay using nails as a sample type was approved as a clinical test by the New York State Department of Health. Samples were analyzed using a custom panel consisting of 82 cancer predisposition genes. Single nucleotides, insertions deletions, and copy number variants were detected using our analytic pipeline and analyzed according to ACMG criteria for pathogenicity. Only pathogenic and likely pathogenic germline variants (gPV) were returned. Results: Among 86 HM patients (acute myeloid leukemia, n = 19; myelodysplastic syndrome, n = 16; acute B lymphoblastic leukemia, n = 9; acute T-cell leukemia, n = 4; mixed phenotypic acute leukemia, n = 1; chronic lymphocytic leukemia, n = 7; hairy cell leukemia, n = 1; chronic myeloid leukemia, n = 4; chronic myelomonocytic leukemia, n = 1, non-Hodgkin lymphoma, n = 12; Hodgkin lymphoma, n = 8; and multiple myeloma, n = 4), gPV were detected in 16 % (14/86). Heterozygous gPV identified in high/moderate penetrant genes included 2 BRCA1, 2 ATM, 1 PALB2, and 3 CHEK2. Monoallelic carriers for homozygous disease included 2 MUTYH and 1 MSH3. Biallelic gPV were identified in 3 patients (1 FANCA, 1 ATM, 1 MSH6); the patients with Fanconi anemia (age 25) manifesting with MDS and constitutional mismatch repair deficiency (age 13) presenting with T-cell lymphoblastic lymphoma had not been previously diagnosed with these conditions and did not have obvious clinical features outside of their cancer diagnosis. In all 3 patients, identification of an autosomal recessive syndrome impacted therapeutic planning. Conclusions: Our study demonstrates the feasibility of utilizing nails, a non-contaminated germline sample source, for HM predisposition with a significant proportion of pediatric and adult HM patients identified to harbor gPV in a cancer predisposition gene. Autosomal recessive diseases should be considered not only in pediatrics but also adult patients and requires confirmation of allele segregation in parents. Our data also demonstrate the importance of gPV in HM for therapeutic planning.