Identifying individuals with DDX41 germline predisposition syndrome

Abstract Background: Deleterious DDX41 germ line variants cause a predisposition syndrome (DDX41-GPS) that increases the risk of myeloid neoplasms (MNs), particularly myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), with a penetrance of roughly 50% by age 90 years. First identified in 2015, there remains much to learn about the pathogenesis, presentation, and survival of individuals with DDX41-GPS. Methods: Records from patients identified during workup of a MN or through targeted family variant testing with DDX41-GPS at Huntsman Cancer Institute (HCI) between January 1, 2021, and August 1, 2025, were reviewed. We examined the clinical presentation, laboratory findings, genetic profile, and survival. DDX41 variants were classified per American College of Medical Genetics guidelines as likely pathogenic (LP), pathogenic (P), or variants of uncertain significance (VUS). When possible, germ line status was confirmed by sequencing cultured fibroblasts. When germ line testing was not an option, variants with an allele frequency >40% were assumed to be of germ line origin. Results: We identified 48 individuals with DDX41-GPS from 35 unrelated families. 66.7% (32/48) of individuals were alive as of the data cutoff. 62.5% (30/48) were male and 37.5% (18/48) female. DDX41-GPS identification occurred during the work-up of MN diagnosis in 60.4% (30/48) (median age at diagnosis 68 years, range 47-89 years). Seventeen patients (10 males, 7 females) were diagnosed with AML and 12 with MDS (11 males, 1 female). Fifteen patients were identified through cascade testing. Eighteen unique germ line DDX41 variants were identified, with 11 classified as P/LP and 7 classified as VUS. The most common germ line variants were DDX41 c.3G>A (27.1%, n=13) and DDX41 c.415_418dup (20.8, n=10%). Of those with MNs, 51.7% (15/29) had a somatic mutation identified in the alternate allele. 60% (9/15) of those had the canonical somatic DDX41 c.1574G>A variant. Of individuals without MN (18/48), the median age was 58 years, range 38-96 years. At the time of DDX41-GPS identification, hemoglobin was a median of 13.75 g/dL (12.5-15.6), white blood cell count was a median of 6.45 k/uL (2.69-9.39), absolute neutrophil count was a median of 3.77 k/uL (1.31-6.82), and platelet count was a median of 238.5 K (123-388). 11.1% (2/18) of individuals had pancytopenia at the time of DDX41-GPS identification but ultimately were not diagnosed with MDS/AML. For 6/18 individuals, CBCs were unavailable. None of the unaffected individuals have developed MNs since starting annual follow-up at HCI. The median survival of patients with MN was 5.0 years (95% CI 4.0-NS), consistent with previously published works. Median OS if a patient developed a second somatic alteration in DDX41 was 6.0 years (95% CI 5.0-NS) versus 2.0 years (95% CI 1.0-NS) in those for which a somatic mutation was not identified (p=0.19). Of patients with DDX41-AML (n=17), cytopenia was noted a median of one month (range 0-31 months) before AML diagnosis. Of patients with DDX41-MDS (n=12), cytopenia was identified a median of 8.5 months before MDS diagnosis (range 1-74 months). Conclusion: Most individuals with DDX41-GPS were identified at the same time as MN diagnosis. Lifelong cytopenias, such as those observed in germ line ETV6 or ANKRD26 predispositions, were uncommon in our DDX41-GPS cohort. A longer latency between cytopenia development and MN diagnosis was observed in those with DDX41-MDS versus DDX41-AML. More research is needed to determine whether these cytopenias are associated with DDX41-GPS or whether they represent early clonal changes in MDS. Cascade testing of affected individuals' family members will help pave the way for prospective studies aimed at characterizing the natural history of DDX41-GPS before individuals develop MNs.