Favorable outcomes of DDX41-mutated myelodysplastic syndrome and low blast count acute myeloid leukemia treated with azacitidine lenalidomide

Myelodysplastic neoplasms (MDS) are clonal myeloid neoplasms with an increased risk of progression to acute myeloid leukemia (AML). For patients unfit for hematopoietic cell transplantation (HCT), hypomethylating agents, such as azacitidine (AZA), represent the main therapeutic option for patients with higher-risk MDS, with an overall response rate (ORR) of 51% and median overall survival 25 months [1, 2]. Lenalidomide is an immunomodulatory drug that is effective in lower-risk MDS with isolated del(5q) [3], but may also have activity in non-del(5q) MDS [4]. Germline DDX41 mutations are found in approximately 5% of patients with MDS and AML, representing the most common myeloid neoplasms associated with germline predisposition [5-7]. To date, it is unclear if DDX41-mutated MDS should be treated differently than the standard treatment, such as AZA [8, 9]. While DDX41 (5q35.3) is outside the minimal deleted region (q31–q33) of most MDS cases with del(5q), it is variably deleted in 25% of patients [10]. Thus, there is considerable interest if DDX41-mutated MDS might also respond favorably to lenalidomide [10-13]. The Australasian Leukaemia and Lymphoma Group (ALLG) MDS4 phase II trial (ACTRN12610000271000) randomized 160 patients with higher-risk MDS, chronic myelomonocytic leukemia (CMML), and low blast AML to either AZA or combination AZA with lenalidomide (LEN, from cycle 3 onward) [14]. While the combination of LEN and AZA was tolerable, overall there was no improvement in clinical benefit, response rates, or overall survival in patients compared to treatment with AZA alone [14]. We performed a post-hoc genomic analysis to correlate genomic lesions in this trial cohort with clinical outcomes. This study was approved by the local ethics committee and conducted in accordance with the Declaration of Helsinki. A unique molecular index-based QIAseq targeted DNA panel (QIAGEN) was performed with a sensitivity of 0.5% variant allele frequency (VAF) [15]. HumanCytoSNP-12 BeadChip array (SNP-A) was previously reported [14]. The primary endpoint of the study was an ongoing clinical benefit (alive and progression-free) at 12 months. Kaplan−Meier overall survival was measured from the date of treatment commencement. R statistical software version 4.2 was used for analysis. A total of 66 patients had baseline DNA available for testing: 36 on AZA alone and 30 on AZA+LEN. Baseline characteristics are summarized in Figure 1. The median age was 71.6 years (range 52–87). IPSS-R risk groups were very low or low (n = 20), intermediate (n = 18), high or very high (n = 25), and not available (n = 3 due to missing/failed karyotype). Disease subtypes by WHO 2008 were MDS (n = 52), CMML (n = 9), and AML (n = 5). SNP-A was abnormal in 47% of patients, including 18% with complex (≥3) copy number changes. The most common mutations were ASXL1 (48.5%), TET2 (47%), and RUNX1 (30%). Median follow-up among survivors was 49 months and median overall survival was 33.2 months (95% CI, 21.7 to NE). DDX41 variants were found in six patients with all patients harboring ≥2 DDX41 variants (including the Arg525His variant in three). The germline versus somatic origin of these variants could not be proven; however, all six patients had one DDX41 variant at approximately 50% VAF with a second lower VAF DDX41 detected (Table S1). All patients with DDX41 variants had a normal karyotype and SNP-A (5/5, 100%) in comparison to 48% (26/55) DDX41 wildtype patients among evaluable patients. Additionally, five patients had monosomy 5 by G-banded karyotyping (expected to delete the DDX41 locus) but all were associated with complex karyotype and these five patients died after median 4.9 months (range 0.6−21.6). The most common co-mutation in DDX41 mutant group was ASXL1 (4/6 patients) (Table S2). One patient with a DDX41 mutation (low blast AML) died before receiving therapy due to disease progression. The remaining five patients, aged 64–76 years, received AZA (n = 4) or AZA+LEN (n = 1) and achieved complete remission (CR, n = 1), marrow CR (n = 1), hematologic improvement (HI, n = 1), and stable disease (n = 2), and remarkably all remained alive at last follow-up of median 53 months (range 47–58; Figure 2), despite having intermediate (n = 3) and high (n = 2) R-IPSS. No patients underwent HCT. Mutations in TP53, U2AF1, and EZH2, and abnormal SNP-A profile, including complex and abnormal 5q/7q/17p, were associated with inferior outcomes consistent with previous studies (Figures S1A and S2). Additional molecular subgroup analyses did not identify any significant interaction with the treatment arms (Figure S1B). Mutations were serially assessed on therapy in 45 patients after two (n = 38) to four cycles (n = 7) of therapy. The majority (67%) of the 166 mutations observed at baseline remained stable (< 10% VAF change). With the caveat of short interval between testing, no patient achieved complete molecular clearance. Fourteen (31%) patients had newly acquired mutations (n = 18 variants) or > 10% VAF increase (n = 2 variants). Twenty-six (58%) patients had mutation clearance of ≥1 variant (n = 20/166 [12%] variants) or > 10% VAF reduction (n = 33 [20%] variants). Of note, eight patients had mixed changes in VAF. Patients with mutation clearance/reduction had a trend toward better overall survival (Figure S3). Two patients with DDX41-mutated disease were assessed after two cycles of AZA with no significant change in the somatic mutations. One patient with DDX41-mutated MDS with increased blasts-1 achieved marrow CR after four cycles of AZA with approximately a 50% decrease in somatic DDX41, NRAS, and ASXL1 mutations (Figure S4). Our cohort demonstrated excellent outcomes among patients with DDX41-mutated myeloid neoplasms treated with AZA. Earlier retrospective studies suggested that patients with DDX41-mutated myeloid neoplasms might respond favorably to lenalidomide [10, 12, 13]. Sebert et al. studied 11 patients with DDX41-mutated MDS/AML who received AZA with 73% ORR and prolonged response duration (median 2.5 years); note seven patients underwent HCT [8]. A pooled analysis of additional patients (total 33 patients) showed similar ORR of 70% to AZA [9]. The small number (n = 5) in our trial cohort prevents any conclusion to be drawn regarding the impact of the addition of lenalidomide (n = 1) on DDX41-mutated MDS in comparison to AZA alone (n = 4). However, until further data become available, the management of DDX41-mutated MDS should follow the current standard of care. In summary, our correlative molecular analysis of the ALLG MDS4 trial of AZA versus AZA+LEN in MDS/AML has identified that patients with DDX41 mutations are associated with favorable outcomes with either AZA or AZA+LEN supporting the use of this therapy in these patients. This observation and subgroup of DDX41-mutated patients warrants further study in randomized trials. PB and WSS designed the study. IST, MW, YZY, and PB performed the molecular analyses. JFS and MK contributed essential clinical data. IST and PB wrote the manuscript. All authors reviewed and approved the submitted version. The authors gratefully acknowledge funding from the Wilson Centre for Blood Cancer Genomics and the Snowdome Foundation. The authors would also like to acknowledge all the participating patients in the Australasian Leukaemia and Lymphoma Group MDS4 study from the following sites (Principal Investigators): Austin Hospital (Daniela Zantomio); Barwon Health (Philip Campbell); Border Medical Oncology (Richard Eek); Cabrini Hospital (Melita Kenealy); Calvary Mater Newcastle (Sandra Deveridge); Canberra Hospital (James D'Rozario); Coffs Harbour Hospital (Martin Browne); Concord Hospital (Ilona Cunningham); Flinders Medical Centre (David Ross); Fremantle Hospital (Michael Leahy); Gosford Hospital (Campbell Tiley); Greenslopes Hospital (Anthony Mills); ICON (Kerry Taylor); Liverpool Hospital (Anne-marie Watson); Monash Medical Centre (Stephen Samuel Opat); Nepean Hospital (John Taper); Peter MacCallum Cancer Centre (John Seymour); Port Macquarie Hospital (Richard Stark); Princess Alexandra Hospital (Anthony Mills); Royal Adelaide Hospital (Devendra Hiwase); Royal Hobart Hospital (Rosemary Harrup); Royal Melbourne Hospital (Ashish Bajel); Royal North Shore Hospital (William Stevenson); Sir Charles Gardiner Hospital (Gavin Cull); St. George Hospital (Jing Hu) St. Vincent's Hospital Melbourne (Robin Filshie); St. Vincent's Hospital Sydney (Keith Fay); Tweed Hospital (Ehtesham Abdi); Western Health Hospital (Duncan Carradice); Westmead Hospital (Warwick Benson); Wollongong and Shoalhaven Hospital (Pauline Warburton). The authors declare no conflict of interest. Wilson Centre for Blood Cancer Genomics; Snowdome Foundation. This study was approved by the Northern Sydney Local Health District Human Research Ethics Committee (2019/ETH08385). All patients provided written informed consent to the ALLG MDS4 study and the storage and use of blood and tissue samples for research. anzctr.org.au (ACTRN12610000271000). The data that support the findings of this study are available from the corresponding author upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.