Abstract A037: Investigating the role of the somatic DDX41 variant in the context of DDX41 hereditary myeloid malignancy syndromes (HMMS)
Ludovica Marando,Talha Badar,Yael Kusne,Audrey Morris,Terra Lasho,Abhishek Mangaonkar,Christy Finke,Carles Crawley,James M. Foran,Aref Al-Kali,Andrew King,Hassan Alkhateeb,Naseema Gangat,Chi Wong,Rong He,David Viswanatha,Faisal Basheer,Mark Litzow,Alejandro Ferrer,George Vassiliou,Mrinal Patnaik
DOI: https://doi.org/10.1158/1538-7445.canevol23-a037
IF: 11.2
2024-02-02
Cancer Research
Abstract:Heterozygous germline variants in DDX41 are identified in ~5% of patients with Myelodysplastic Syndrome (MDS) or Acute Myeloid Leukemia (AML). Recognition of the hereditary nature of such cancers and a better understanding of the factors that affect progression to malignancy has deep implications for health surveillance of patients and relatives that share a pathogenic DNA variant. DDX41 has a role in ribosome biogenesis and in opposing R-loop accumulation at promoters, but the mechanisms through which DDX41 loss drives cancer development are poorly understood. In a collaboration between Cambridge, UK and Mayo Clinic, USA we identified 85 patients with germline DDX41 variants [73/85 with germline pathogenic variants (GPV) and 12 with variants of uncertain significance (VUS)]. The majority were identified through targeted myeloid NGS upon presentation with AML (30/85) or MDS (44/85). Average age at diagnosis was 65 (35-93) with male predominance (57/85). The most common GPV was the NM_016222.4(DDX41):c.415_418dup; p.D140Gfs*2 (15/85). The majority (53.4% 39/73), also acquired a somatic variants in the other DDX41 allele (referred to as GPV+Som). The most common somatic variant (82% 32/39) was the R525H, a variant predicted to reduce the ATPase function of the helicase domain. We hypothesise that the markedly reduced DDX41 activity in cells with biallelic variants, is a key cause of progression to overt malignancy. In our cohort, we observed that somatic variants occurred at a similar frequency in association with start loss, truncating, splice, in-frame deletion, and missense GPV. However, whereas MDS was the predominant diagnosis within GPV (67.6% 23/34, AML 26.4% 9/34), AML was the most frequent diagnosis within GPV+Som (AML 51%.2 20/39, MDS 38.4% 15/39). Overall, there was a significantly higher likelihood of AML in the presence of a somatic variant (p<0.05), suggesting that there might be a dosage effect on the phenotype. This pattern was observed across all the GPV, with the exception of missense, suggesting that many might be benign polymorphysms. In addition, although somatic variants are typically subclonal (overall mean variant allele fraction VAF=9.87%, 3-27.8%), we observed a higher somatic VAF in AML over MDS cases (12.4% vs 7.7% respectively p<0.05). A lower bone marrow cellularity was also characteristic of GPV+Som vs GPV alone (mean observed/expected cellularity for age 0.9 vs 1.32 respectively p<0.05). In two cases, serial bone marrow morphology and NGS data were available, and we observed marked reduction in bone marrow cellularity upon acquisition of the R525H clone. The mechanisms of AML evolution appear different in DDX41 HMMS compared to sporadic cases. Interestingly, over half of our patients (21/34 GPV and 23/39 GPV+Som) did not exhibit additional somatic drivers. Our work suggests that acquisition of a somatic variant is a critical event in disease progression. Furthermore, the near universally subclonal nature of somatic variants suggest that they may affect disease phenotype through non-cell autonomous mechanisms. Citation Format: Ludovica Marando, Talha Badar, Yael Kusne, Audrey Morris, Terra Lasho, Abhishek Mangaonkar, Christy Finke, Carles Crawley, James M. Foran, Aref Al-Kali, Andrew King, Hassan Alkhateeb, Naseema Gangat, Chi Wong, Rong He, David Viswanatha, Faisal Basheer, Mark Litzow, Alejandro Ferrer, George Vassiliou, Mrinal Patnaik. Investigating the role of the somatic DDX41 variant in the context of DDX41 hereditary myeloid malignancy syndromes (HMMS) [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2) nr A037.
oncology