P506: clinical validation of the nordic guidelines for germline testing in myeloid neoplasms: results from a multi-center prospective cohort study

Background: Most available data on germline predisposition for myeloid neoplasms are based on retrospective studies. Moreover, due to great variability in resources and institutional policies, there is currently no consensus on how to investigate germline predisposition in the clinical setting. Aims: To prospectively investigate the clinical validity of the Nordic guidelines for germline testing in patients with suspected germline predisposition to myeloid neoplasms. Methods: Between October 2019 and January 2023, 84 sequential patients fulfilling the Nordic criteria for germline investigation were included in the study [acute myeloid leukemia (AML): n=38, myelodysplastic syndrome (MDS): n=33, thrombocytopenia: n=11, other: n=2]. In brief, the inclusion criteria were: (i) personal history indicative of a germline condition (criterion A1); (ii) “family history” defined as two or more first-degree (FDR) or second-degree relatives (SDR) with AML/MDS/thrombocytopenia or two FDR/SDR with other cancers, with at least one individual diagnosed before the age of 50 (criterion A2 and A3 respectively) or 3 FDR/SDR with AML/MDS/thrombocytopenia independently of age (A4); (iii) a potential germline variant detected during the somatic diagnostic work-up (Twist myeloid-panel, criterion B). Patients fulfilling A1-A4 were analyzed with whole exome sequencing (WES) including copy number variant analysis and analysis of non-coding regions in genes related to germline predisposition. Patients fulfilling criterion B were analyzed with targeted analysis. In patients with malignant disease the analysis was performed on genomic DNA from cultured fibroblasts (n=33), T-lymphocytes (n=3), or peripheral blood in remission (n=3). Genetic counselling was provided before and after the analysis. Results: As of today, the analysis is completed in 79/84 (94%) patients fulfilling one of the above criteria (A1: n=18; A2: n=6; A3: n=10; A4: n=15; B: n=30) from 13 Swedish institutions. A pathogenic or likely pathogenic (P/LP) germline variant according to the ACMG-criteria was identified in 29/79 (37%) patients. The detection rate of P/LP variants varied depending on the applied criterion: A1: 22%, A2-A3: 6%, A4: 47%, B: 57% (Figure 1). DDX41 variants were the most common aberration accounting for 50% of all positive cases (14/29) followed by ANKRD26: n=3, RUNX1: n=3, CEBPA: n=2, and ACTN1, DNAJC21, ETV6, PARN, TERT, SBDS: n=1 each). DDX41 aberrations were highly enriched (13/14) in patients fulfilling criterion B (13/14). Ten patients (13%) analyzed with WES carried a variant of unknown significance (VUS, TERT: n=2, RTEL1: n=2, SAMD9: n=1, SAMD9L: n=1, GATA2: n=1, RUNX1: n=1, PARN: n=1, ETV6: n=1) which warranted further investigation. Four patients carried a P/LP variant in genes not commonly associated with myeloid neoplasms (BRCA1, PALB2, PMS2, KCNQ1), a finding that prompted reassessment of the clinical management. Summary/Conclusion: The application of the Nordic guidelines for germline predisposition for myeloid neoplasms leads to the identification of a P/LP variant in more than one third of the patients with a detection rate depending on the applied criteria (range 6-57%). The spectrum of germline variants is criterion-depended, with DDX41 variants dominating among cases investigated due to the findings of the somatic profiling. In a significant proportion of patients with myeloid neoplasms and suspected germline predisposition the underlying genetic cause remains unknown despite the application of high-throughput sequencing.Keywords: Acute leukemia, Hereditary malignancies, Myelodysplastic syndrome