Germline predisposition in pediatric malignancies: Insights from a retrospective study at Dana-Farber Cancer Institute

Abstract Introduction Germline genetic testing is increasingly recommended at the time of a pediatric cancer diagnosis for children with solid tumors, but not routinely performed for patients with hematologic malignancies (HM). Studies regarding incidence of hereditary hematologic malignancies (HHM) are limited with variable rates reported in pediatric and adult cohorts (4-39%). In this retrospective study, we evaluated the prevalence of HHM in children and young adults with newly diagnosed or relapsed HM and the impact of these diagnoses on treatment decisions. Methods Patients with newly diagnosed or relapsed HM treated between 1/1/2020 and 12/31/2024 in the Pediatric Oncology Clinic at Dana-Farber Cancer Institute were eligible. We performed focused retrospective chart review on all patients in our study cohort with additional chart abstraction done for those patients identified with a HM who received genetic counseling (GC). All germline genetic testing was done using cell sources suitable for accurate detection of germline variants. Germline testing primarily consisted of multi-gene panels that screened 200 genes with known association to cancer predisposition and bone marrow failure. Results were obtained from a CLIA certified lab and variants were classified according to American College of Medical Genetics variant classification guidelines. Results 571 patients with a HM were included in our study cohort. 207/571 patients (36%) were referred for genetic counseling (GC). Of those seen for GC, 127 had a lymphoid neoplasm (acute lymphoblastic leukemia 48%, Hodgkin lymphoma 6%, non-Hodgkin lymphoma 4%, lymphoblastic lymphoma 3%) while 80 had a myeloid neoplasm (acute myeloid leukemia (AML), 32%, myelodysplastic syndrome 5%, myeloproliferative neoplasms 2%). The mean age of the GC-evaluated cohort at initial diagnosis of their HM was 9.1 years (0.2-28.4) with male predominance (62%). All 207 patients seen for GC were offered germline testing, and 141 patients (68%) consented. A pathogenic or likely pathogenic (P/LP) variant was identified in 67/141 (48%) patients. Specifically, 16 patients (16/141, 11%) of the tested cohort had a P/LP variant in a gene associated with a HHM and 11 patients (11/141, 8%) had a P/LP variant in a gene associated with solid tumor risks. One patient harbored two germline variants: one with HHM risk and the other with solid tumor risk. The remaining 41 patients (41/141, 29%) had a P/LP variant associated with carrier status of an autosomal recessive syndrome or risk for non-cancer phenotypes only. An additional 7 patients had a variant of uncertain significance (VUS) in a gene associated with HHM that was highly consistent with their phenotype and met at least one additional suggestive feature: functional evidence of pathogenicity, rarity in population databases, or familial segregation. In summary, 33 patients (33/141, 23%) had a P/LP variant or highly suspicious VUS in a cancer risk gene. Of that cohort, 54% (18/33) presented with myeloid disease while 45% (15/33) had a lymphoid malignancy. The P/LP variants associated with HHM included 3 patients with GATA2, two patients each with RUNX1, DDX41, POT1, and MPO, and one patient each with CBL, TP53, PAX5, CDKN2A and constitutional mismatch repair deficiency (PMS2 biallelic). For the 7 patients with highly suspicious VUS, the genes involved included ETV6, GATA2, LCP1, CEBPA, GATA1, TP53, and PIK3CD. AML treatment was upstaged to include hematopoietic stem cell transplantation in 2 patients due to their germline results. For all other patients, the germline results did not impact therapy but did lead to subsequent cancer surveillance and other practice changes, including cascade testing recommendations for family members, where applicable. ConclusionPrevalence of germline predisposition to pediatric HM remains unclear. Acknowledging the referral bias of this cohort, our findings show a significant rate (23%) of pediatric HM patients with a HHM, cancer predisposition syndrome, or VUS highly suggestive of such risk. Our data supports universal germline testing at time of HM diagnosis. In doing so, we will more accurately define the prevalence of HHM and determine if HHMs confer higher risk of relapse and adverse treatment effects in specific disease subsets.