AUTHORS
Bhavisha A Patel, MD, Fieke W Hoff, MD, PhD2, Fernanda Gutierrez-Rodrigues, PhD3, Jennifer Lotter, MS4, Alice Fike5, Benjamin Turturice, MD6, Julie Onyechi7, Ivana Darden, RN8, Shanni Liu9, Neal Stuart Young, MD10, Peter Grayson, MD11, Emma Groarke, MD12
AFFILIATIONS
1 National Institutes of Health, Bethesda, Maryland, United States
2 National Institutes of Health, Bethesda, Maryland, United States
3 National Institutes of Health, Bethesda, Maryland, United States
4 National Institutes of Health, Bethesda, Maryland, United States
5 National Institutes of Health, Bethesda, Maryland, United States
6 National Institutes of Health, Bethesda, Maryland, United States
7 National Institutes of Health, Bethesda, Maryland, United States
8 National Institutes of Health, Bethesda, Maryland, United States
9 National Institutes of Health, Bethesda, Maryland, United States
10 National Institutes of Health, Bethesda, Maryland, United States
11 National Institutes of Health, Bethesda, Maryland, United States
12 National Institutes of Health, Bethesda, Maryland, United States
ABSTRACT
Introduction
Azacytidine (Aza) is an effective treatment for VEXAS, improving inflammatory and hematologic parameters and reducing size of the UBA1-mutated clone. Some patients achieve a complete molecular response (CMR), defined as UBA1 variant allele frequency (VAF) ≤ 2%, but the clonal dynamics of UBA1 and co-occurring somatic mutations during Aza therapy remains poorly characterized. We prospectively evaluated molecular responses (MR) to Aza, either CMR or partial MR (PMR), co-mutation dynamics, and the relationship with hematologic recovery.
Methods
Thirty men with VEXAS were followed from 2022 (median age=66 yrs, (range:41-84); p.M41L n=12; p.M41V n=7; p.M41T n=7; splice site n=3; p.G477A n=1). Clinical, laboratory and molecular responses were assessed after 6, 12, and 18 cycles of Aza. Somatic mutations were assessed in bone marrow (BM) or peripheral blood (PB; VAF ≥1%), and single-cell (sc) DNA sequencing was performed in select cases. PMR was defined as ≥25% UBA1 VAFs reduction from baseline. No MR (NMR) as minimal/no change/increase UBA1 clones.
Results
Twenty-nine patients (97%) were receiving steroids at Aza initiation (median dose 20mg/day [range 0-70mg/day]), and 11 were on additional immunosuppressive therapy (IL-6 inhibitor, n=8; JAK inhibitor, n=2; IL-4/IL-13 inhibitor, n=1). Ten patients (33%) were red blood cell transfusion dependent, 23/30 (77%) had baseline cytopenias (hemoglobin <10g/dL or platelets <100 K/µL), and 13/30 (43%) met diagnostic criteria for myelodysplastic syndrome. Pre-Aza, median UBA1 VAF was 60%. Co-mutations were present in 12 patients: DNMT3A (n=6; median VAF=40%), TET2 (n=3; median VAF=1.6%), and ZRSR2 (n=2; median VAF=42%).
Twenty-five patients had at least one post-Aza assessment (7 received standard 7-day regimen and 18 received reduced regimen, usually 5-day. Five patients were not evaluable due to proceeding with transplant (n=2), death (n=1), toxicities (n=1), lack of paired molecular testing (n=1). Post-6 cycles, 17/22 evaluable patients (77%) achieved MR (8 CMR, 9 PMR) and 5 had no MR. Among patients with PMR, mean UBA1 VAF decreased by 42% from baseline (range VAF at assessment 4–60%). CMR post-6 cycles was more frequent with standard Aza than with reduced dosing (4/7, 60% vs 4/15, 26%). Hemoglobin and platelet counts improved significantly after 6 cycles of Aza (median increases of 3.7 g/dL and 28 K/µL, respectively. These hematologic changes did not associate with MR. Median hemoglobin increase in patients with NMR (9.1 to 13.7g/dL), PMR (9.6 to 12.2g/dL), and CMR (9.4 to 12.9g/dL). Similar trend noted in platelet counts; 17/22 (77%) patients with baseline cytopenia responded per modified IWG criteria, regardless of MR. Five of 6 evaluable transfusion-dependent patients achieved transfusion independence within 3 cycles. Post-12 cycles (n=14), 4 achieved CMR (one was PMR post-6 cycles, reduced dose), 7 PMR (one was NMR post-6 cycles, reduced dose) and 3 NMR. Three out of 4 patients achieved CMR post-8 cycles (one was PR at 6 and 12 cycles, reduced dose).
scDNA sequencing in one patient with CMR showed that 0.6% BM cells harbored UBA1 mutations post-6 cycles. Residual mutant cells were mostly myeloid and NK cells, with only 0.04% of HSPC harboring UBA1 mutations. In PB, 1.2% of cells were UBA1-mutant and were exclusively mature myeloid cells.
Co-mutation dynamics were independent of Aza. DNMT3A mutations decreased in parallel to UBA1, consistent with acquisition of UBA1 within pre-existing DNMT3A-mutant clones. DNMT3A persisted at higher VAFs (12%–29%) in three patients despite UBA1 VAF<5%. In contrast, TET2 and spliceosome gene mutations remained largely stable, suggesting independent clones. Co-mutation status was not associated with MR or hematologic response. New somatic mutations, primarily involving DNMT3A or ASXL1, were uncommon during follow-up, and no patient developed acute myeloid leukemia.
Conclusions
Aza reduced UBA1 clonal burden in most patients, with higher CMR post-6 cycles in those treated with 7-day regimen. In contrast, responses were gradual with deeper MR occurring post 12 cycles with reduced Aza dosing. Hematologic improvement occurred early and was not associated with the depth of MR, indicating that restoration of hematopoiesis may occur by mechanisms not fully explained by direct reduction of the UBA1 clone. Co-mutation patterns were not associated with MR. UBA1 clonal reduction was not associated with expansion of alternative mutated clones.
DISCLOSURES
Bhavisha A Patel, MD: No relevant conflicts of interest to declare.