Detection of Novel Occult Germline Multi-Exon Deletions in Patients with DDX41 Familial Predisposition to Myeloid Malignancy

Background and aims Myeloid malignancy in patients (pts) with a hereditary predisposition to myeloid malignancy as a result of deleterious germline variants in DDX41 is typically accompanied by acquisition of a somatic variant (most commonly Arg525His) in the unaffected allele of DDX41. Approximately 10% of patients with MDS/AML with somatic variants in DDX41 harbor no detectable germline DDX41 variant. It is currently unclear whether these pts have true somatically-mutated only DDX41 disease or whether they harbor occult deleterious germline variants given that the majority of studies to date have only assessed exonic regions of DDX41 and focused on single nucleotide variants (SNVs) or small insertions/deletions. We aimed to further genomically characterize pts categorized as having 'somatic-only’ DDX41 variants in order to identify potential underlying causative germline DDX41 variants.

Methods 4868 samples were received for clinical testing for myeloid malignancy. Samples underwent NGS molecular testing (sensitivity of approximately 2% variant allele frequency [VAF]) with targeted panels including all coding exons of DDX41 (NM_016222.4). Single nucleotide variants and small insertions/deletions were called with Pisces or CLC Genomics workbench. Structural variants (SVs)/large insertion-deletion variants were called with GRIDSS.

Results Of 4868 pts referred for clinical sequencing we identified 252 pts (5.2%) with sequence variants in DDX41. 77% (195/252) had either confirmed or assumed (based on clinicogenomic features including VAF, family history and other testing) germline DDX41 variants classified as likely pathogenic or pathogenic according to ACMG criteria. 12% (30/252) of pts were categorized as having 'somatic-only’ DDX41 variants. 'Somatic-only’ DDX41 mutated pts were defined as having no detectable DDX41 variant with a VAF greater than 35% (in the absence of detected copy number change at the DDX41 locus) or having only somatic variants detectable after sequencing of a true germline specimen.

Of the 30 pts with 'somatic-only’ DDX41 variants, 18 pts had one DDX41 variant and 12 pts had two (median VAF 7.2%, range 2-46%). The spectrum and frequency of co-mutated genes (ASXL1, DNMT3A, SRSF2) was similar to those with confirmed/assumed germline variants in DDX41. In order to further characterize DDX41 in 'somatic-only’ pts we performed SV/large insertion-deletion detection whereby we detected three large multi-exon deletions in four pts (two unrelated pts had an identical deletion) of 2.1kb, 1.5kb and 0.6kb in size. The three deletions encompassed exons 6 (partial) - 9 [Chr5(GRCh37):g.176,941,255_176,942,766del], exons 15 - 17 (and extending through the intergenic region into DOK3) [Chr5(GRCh37):g.176,937,300_176,939,456del] and exons 14 (partial) - 17 (partial) [Chr5(GRCh37):g.176,938,890_176,939,513del]. Deletions were present at approximately 50% VAF and were detected in germline (hair follicle) and remission samples from pts with the exons 14 (partial) - 17 (partial) deletion. Five somatic DDX41 variants were present in these 4 pts: p.(Arg525His) (n=2), p.(Gly530Asp) (n=2) and p.(Met378Ile) (n=1) and were detected at VAFs ranging from 2.4% to 19%. We assessed SVs in DDX41 in the 195 pts with assumed/confirmed germline variants but did not detect any similar variants. In addition, we performed variant calling on all intronic regions covered by targeting sequencing and did not detect any causal germline variants in these non-coding regions.

Conclusion We have described multiple novel and recurrent germline multi-exon deletions in DDX41 in pts presumed to have 'somatic-only’ DDX41 variants based on routine clinical sequencing. These clinically relevant data highlight the importance of using dedicated bioinformatics processes or alternative assays that may detect such abnormalities (e.g. MLPA) when assessing DDX41 in the clinic. Future studies should further assess non-coding/regulatory regions in order to fully understand the spectrum of variants causative of germline DDX41 predisposition to myeloid malignancy.

Maierhofer:MLL Munich Leukemia Laboratory: Current Employment. Hutter:MLL Munich Leukemia Laboratory: Current Employment. Baer:MLL Munich Leukemia Laboratory: Current Employment. Tiong:Servier: Consultancy, Speakers Bureau; Pfizer: Consultancy, Speakers Bureau; Amgen: Speakers Bureau. Nadarajah:MLL Munich Leukemia Laboratory: Current Employment. Meggendorfer:MLL Munich Leukemia Laboratory: Current Employment. Kern:MLL Munich Leukemia Laboratory: Current Employment, Other: Ownership. Haferlach:MLL Munich Leukemia Laboratory: Current Employment, Other: Ownership. Haferlach:Munich Leukemia Laboratory: Current Employment, Other: Part ownership. Blombery:Adaptive Biotechnologies: Consultancy, Honoraria; AstraZeneca: Consultancy, Honoraria; Servier: Honoraria.