The genetic landscape of germline DDX41 variants predisposing to myeloid neoplasms

Abstract

Germline DDX41 variants are the most common mutations predisposing to acute myeloid leukemia (AML)/myelodysplastic syndrome (MDS) in adults, but the causal variant (CV) landscape and clinical spectrum of hematologic malignancies (HMs) remain unexplored. Here, we analyzed the genomic profiles of 176 patients with HM carrying 82 distinct presumably germline DDX41 variants among a group of 9821 unrelated patients. Using our proposed DDX41-specific variant classification, we identified features distinguishing 116 patients with HM with CV from 60 patients with HM with variant of uncertain significance (VUS): an older age (median 69 years), male predominance (74% in CV vs 60% in VUS, P = .03), frequent concurrent somatic DDX41 variants (79% in CV vs 5% in VUS, P < .0001), a lower somatic mutation burden (1.4 ± 0.1 in CV vs 2.9 ± 0.04 in VUS, P = .012), near exclusion of canonical recurrent genetic abnormalities including mutations in NPM1, CEBPA, and FLT3 in AML, and favorable overall survival (OS) in patients with AML/MDS. This superior OS was determined independent of blast count, abnormal karyotypes, and concurrent variants, including TP53 in patients with AML/MDS, regardless of patient's sex, age, or specific germline CV, suggesting that germline DDX41 variants define a distinct clinical entity. Furthermore, unrelated patients with myeloproliferative neoplasm and B-cell lymphoma were linked by DDX41 CV, thus expanding the known disease spectrum. This study outlines the CV landscape, expands the phenotypic spectrum in unrelated DDX41-mutated patients, and underscores the urgent need for gene-specific diagnostic and clinical management guidelines.

Graphical abstract

Figure 1.

Flowchart of this multi-institutional study and graphical representation of DDX41 variants found in this study. (A) In this study, 195 (2%) patients with HM with at least 1 DDX41 variant (MAF < 0.1%) are identified in 9821 unrelated and unselected adult patients from 6 medical centers and at ARUP Laboratories. Among these patients with HM, 3583 are diagnosed with AML, 2160 with MDS, 1030 with MPN, and 3048 with others including cytopenia and other myeloid and lymphoid neoplasms. These DDX41 variants are further classified into somatic variants alone (variants with a VAF < 40% in isolation) and presumed germline variants (VAFs of 40% or above, with or without concurrent somatic DDX41 variants). The germline variants are further classified into CV (PV/LPV, n = 116) and VUS (n = 60), according to the proposed gene-specific diagnostic criteria, modified from the ACMG guidelines. Among the 116 patients with germline DDX41 CV, 66 are diagnosed with AML, 28 with MDS, 4 with MPN, and 18 with cytopenia (others). Similarly, among the 60 patients with germline VUS, 18 are diagnosed with AML, 12 with MDS, 11 with MPN, and 19 with others. Among others, 4 are diagnosed with B-cell LPD, 1 with MM, and 14 with cytopenia. In addition, we select 4307 adult patients with HM (age of 18 years or above) with wild-type DDX41 (DDX41⁻), confirmed by NGS testing at ARUP laboratories during the same time period. Among these control patients (DDX41⁻), 1365 have a documented AML diagnosis, and the remaining cases include 1109 MDS, 479 MPN, and 1354 others, most of which are cytopenia, similar to those in the cohort of 9821 patients described above. Patients’ age, sex, and cytogenetic and molecular profiles are summarized and sorted by each distinct MN entity and correlated with their DDX41 genotypes (short double-headed arrows indicate the epidemiologic and molecular profile comparisons in between DDX41⁺ CV, VUS, and DDX41⁻ cohorts). Furthermore, we summarize the OS in patients with AML and MDS who were treated at Huntsman Cancer Institute and other medical centers in comparison with the age-matched cohorts (long double-headed arrows indicate the OS comparisons in between DDX41⁺ CV, VUS, and DDX41⁻ cohorts). ^#Of note, 24 patients with AML with DDX41 CV have been documented in a previous study. (B) Graphic distribution of variants identified in this study, positioned on the protein sequence (NM_016222.4) with major functional domains (red, DEAD domain; green, helicase domain; orange, Znf, zinc finger domain; teal, NLS, nuclear localization signaling domain) is separated by germline (above-protein sequences) or somatic (below) variants. Each symbol in germline variants represents 1 patient. The underline indicates novel variants reported in this study. Red, DDX41 CV; blue, DDX41 VUS; orange, p.R164W, likely CV in lymphoma. *With specific exceptions (eg, p.M155I and p.P510S).

Figure 2.

Summary of DDX41 variants and ethnic difference in germline CV identified in this study and literature. (A) Summary of DDX41 germline (above the protein sequence) and somatic (below the protein sequence) variants. The colors in the boxes above and the horizontal bars below the protein sequence are designated corresponding to the protein functional domains. Numbers in parentheses alone or before a slash indicate the total times of a certain variant was reported in literature including those reported in this study, whereas numbers after a slash represent variants seen in the current study. Red, CV; blue, VUS; orange, likely CV for lymphoma. (B-E) Ethnic difference in DDX41 CV as data combined in this study and collected and reanalyzed in literature. (B) Germline variants of p.M1I (98%, 39 White and 1 Asian patients) and p.D140fs (95%, 23 White and 1 African American patients) are the leading CVs in White patients. (C) Missense germline variants, although uncommon in Whites (15%), are seen in 49% of Asian patients with HM (P < .0001). (D) p.Y592C (92%, 11 Asians and 1 non-Asian) and p.A500fs (100%, 10 all in Asian) appear the most common germline CV in Asian patients. (E) Somatic DDX41 variants alone, in the absence of associated germline variants, appear more frequently in Asian than White patients (36% in Asian vs 15% in White, P = .0007). ***, P < .001; ****, P < .0001.

Figure 3.

Integrated genetic profiles of the 195 HM patients with epidemiologic characteristics grouped by different HM diagnoses. A total of 176 patients with presumed germline (gl DDX41) 116 CV and 60 VUS are grouped (CV in red and VUS in blue, respectively), along with the associated somatic DDX41 (s DDX41), concomitant somatic variants, and cytogenetics. In addition, 19 patients with HM with somatic DDX41 variants in the absence of germline variants are appended to the right of the variant table, 13 CV in red and 6 VUS in blue. Each column represents 1 patient. The concomitant variants are grouped into 6 categories based on gene function: epigenetic, epigenetic regulators, genes involving DNA methylation or histone acetylation, and deacetylation (light green); SFs, RNA splicing factors (purple); TFs, transcription factors (orange); signaling, molecules in tyrosine kinase pathway or RAS/MAPK pathways (pink); C, cohesins (light blue); and others (dark blue), genes with function beyond the above categories. Each bar represents 1 variant, and split bars indicate 2 or more variants in the same gene. A&B, AML and breast cancer; CLL & B, chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) and breast cancer; complex, complex karyotype; DLBCL, diffuse large B-cell lymphoma; gl DDX41, germline DDX41 variants; LPL, lymphoplasmacytic lymphoma (MYD88 negative); LR, low risk; MBL, monoclonal B-cell lymphocytosis; MM, multiple myeloma; NI, no information; NL, normal; NM, no mutation; RCA, recurrent cytogenetic abnormalities in AML; s DDX41, somatic DDX41 variants.

Figure 4.

Genetic characteristics of patients with AML/MDS with germline variants in DDX41. (A) The occurrence of somatic DDX41 variants in patients with AML is closely linked to the presence of germline DDX41 CV (79%; 52 of 66), in comparison with patients with VUS (5.6%, 1 of 18, P < .0001) or patients not carrying germline DDX41 variants (DDX41⁻, 0.2%; 3 of 1365, P < .0001). A similar trend is seen in patients with MDS (79% in CV, 0% in VUS, and 0.1% in DDX41⁻, P < .0001). (B) A lower somatic mutation burden, calculated by the number of total concomitant somatic variants (excluding somatic DDX41 variants) per case, is seen in patients with AML with CV (mean ± standard error of the mean: 1.6 ± 0.2) compared with patients with AML with wild-type DDX41 (DDX41⁻, 3.7 ± 0.07, P < .0001) and those with VUS (2.6 ± 0.4, P = .03). Similarly, in MDS, a lower somatic mutation burden is seen in patients with CV (0.7 ± 0.1) in contrast to those with wild-type DDX41 (DDX41⁻, 2.7 ± 0.06, P < .0001) or VUS (3.2 ± 0.5, P < .0001). (C-D) Circos plot diagrams illustrate the pairwise co-occurrence of somatic variants and cytogenetic abnormalities in 94 patients with AML/MDS with germline CV (C) and 30 with VUS (D). Genetic variants and cytogenetic events listed in Figure 3 appear in descending order clockwise, starting at 12 o’clock. Each link (ribbon) indicates pairwise co-occurrence of mutational events, and the width of the ribbons indicates the frequency of the co-occurrent events. The occurrence of germline and somatic DDX41 variants is indicated in red and yellow ribbons, respectively. Variants in signaling and RAS/MAPK pathways are labeled in pink; NPM1 and TP53 variants are labeled in black and green, respectively; the remaining variants are labeled in gray. s (yellow) in D, somatic DDX41; SF (purple), RNA splicing factors; TF (orange), transcription factors; S (pink) in panel C, signaling; O (gray), others; # (blue), cohesin; ! (black), NPM1; @ (green), TP53; nl CG, normal cytogenetics; a CG, abnormal cytogenetics. (E) Frequencies of somatic DDX41 and other concurrent variants in AML/MDS patients. For each gene or genetic category, the percentage of mutations is displayed, associated with either germline CV (red bars, gl DDX41⁺ CV), VUS (blue bars, gl DDX41⁺ VUS), or wild-type DDX41 (green bars, gl DDX41⁻). s DDX41, somatic DDX41 variants; epigenetics, genes involving DNA methylation or histone acetylation and deacetylation; signaling transduction, molecules in tyrosine kinase pathway or RAS/MAPK pathways; RCA, recurrent cytogenetic abnormalities in AML. *P < .05; ****P < .0001.

Figure 5.

Common clinical, hematologic, pathologic, and genetic features and superior OS in AML/MDS patients with germline DDX41 CV. (A) Both patients with AML and MDS with germline DDX41 CV present a similarly indolent and chronic course of cytopenia years before the diagnosis of an overt myeloid neoplasm. Furthermore, the bone marrow examination shows predominantly normal to hypocellular marrow in AML (86%) and MDS (75%, P > .05), and a borderline increase in blasts is seen in patients with AML (31% in AML vs 8% in MDS, P < .0001). Most patients with AML (79%) and MDS (85%, P > .05) carry normal karyotypes with similar germline DDX41 variant subtypes and somatic mutation profiles. (B) The median OS of 57 patients with AML with CV (red line, CV AML, not reached) is significantly longer than that of 13 patients with AML with VUS (blue line, VUS AML, 613 days, P = .02) or 158 patients with DDX41 wild-type AML (dark green line, DDX41⁻ or WT AML, 630 days, P < .0001) in the current study and 1040 patients documented in cBioPortal (lavender line, DDX41⁻ or WT AML R 433 days, P < .0001). Similarly, the median OS of 24 patients with MDS with CV (orange line, CV MDS, not reached) is significantly longer than that of 7 patients with MDS with DDX41 VUS (green line, VUS MDS, 468 days, P = .003) or 87 patients with DDX41 WT MDS (purple line, DDX41⁻ or WT MDS 1425 days, P = .03) in this study and 3128 patients reported recently (navy blue line, DDX41⁻ or WT MDS R, 1116 days, P = .003). (C-D) Statistical characteristics of the median OS in each genotype and disease group (C) and P values in pairwise comparisons (D) are listed in the tables. (E) The results of univariate analysis for different factors predicting OS in patients with AML/MDS with DDX41 CV show that the superior OS is not impacted by patient’s age, duration or severity of cytopenia, blast count, presence of abnormal cytogenetics, somatic DDX41 or other concomitant variants, somatic mutation burden, or different types of germline DDX41 CV. Each circle represents the mean HR calculated by Cox proportional hazards regression, and the horizontal lines represent the 95% confidence interval (CI) for the subgroup’s HR. Right of the dashed vertical line (HR = 1), unfavorable OS; left of the dashed line, favorable OS. WBC, white blood cells; Hgb, hemoglobin; PLT, platelet count; BM, bone marrow; gl DDX41, germline DDX41 CV, s DDX41, somatic DDX41 variants. *P < .05; **P < .01; ***P < .001; ****P < .0001; NS, not significant, P > .05.

Figure 6.

MPN and lymphoma predisposed by germline DDX41 CV. (A) The occurrence of somatic DDX41 variants in MPN patients is more frequent in patients with germline CV (25%) compared with patients with VUS (9%; 1/11, P = .4) or patients not carrying germline DDX41 variants (DDX41⁻, 0%, P < .0001). (B) There appears to be a lower concomitant somatic mutation burden in patients with CV (mean ± standard error of the mean: 0.5 ± 0.5), compared with those with WT DDX41 (DDX41⁻, 2.6 ± 0.1, P = .05) and VUS (1.5 ± 0.4, P > .05). (C) No mutations in JAK2 or CALR were seen in MPN with CV, whereas these canonical variants are seen in 73% (8 of 11) of MPN patients with VUS and 82% in the WT cohort (P < .0001). (D-E) Circos plot diagrams illustrate the pairwise co-occurrence of variants and cytogenetic events in MPN patients with germline CV (D) and VUS (E). Genetic variants and cytogenetic events listed in Figure 2 appear in descending order clockwise starting at 12 o’clock. Each link (ribbon) indicates the pairwise co-occurrence of mutational events, and the width of the ribbons indicates the frequency of the co-occurrent events. TF (orange), transcription factors; nl CG (green), normal cytogenetics; s (yellow), somatic DDX41; E (light green), epigenetic modulators; SF (purple), RNA splicing factors; C (pink), CALR. (F) HM predisposed by germline DDX41 CV. AML (55%) and MDS (24%) are the most common entities predisposed by DDX41 CV, followed by cytopenia (16%), MPN (3%), and lymphoma (3%). ****P < .0001.