Reduced intensity conditioning allogeneic hematopoietic stem cell transplantation in <scp>VEXAS</scp> syndrome: Data from a prospective series of patients

Abstract

Vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic (VEXAS) syndrome occurs secondary to somatic mutations in the X-linked UBA1 gene and presents with an adult-onset inflammatory syndrome with cytopenias and a predisposition to neoplasia (myeloid and plasma cell origin).1 The precise incidence of VEXAS is still unknown but estimated at approximately 9% of all cases with relapsing polychondritis,2 with cytopenias of varying severity occurring in almost all, and concurrent myelodysplastic syndrome (MDS) reported to occur in 25%–55% VEXAS patients.1, 3, 4 Therapeutic strategies have involved the use of inflammation-directed therapy (glucocorticoids, JAK inhibitors, and anti-IL6 therapy) and/or clone-directed therapy with hypomethylating agents; however, responses have been largely underwhelming.4 Allogeneic hematopoietic stem cell transplantation (alloHCT) remains a potentially curative strategy; however, published reports are largely retrospective.5, 6 In other words, UBA1 mutations were identified in stored blood samples collected from patients prior to undergoing alloHCT. In the few prospective cases where VEXAS diagnosis was established prior to alloHCT, there was lack of uniformity in the transplant approach.5-7 Additionally, unique VEXAS-specific clinical challenges such as transplant-specific glucorticoid dose tapering, appropriate indications, and complications remain undefined. We present one of the first prospective reports of VEXAS patients that underwent alloHCT using a uniform conditioning approach and graft versus host disease prophylaxis strategy. Within the auspices of institutional review board approval, we designed a consensus document for pursuing alloHCT in VEXAS by involving multidisciplinary experts from transplant, hematology, rheumatology, pharmacology, and genetics. Testing for UBA1 gene was carried out using Sanger sequencing (Supplementary methods). Indications for alloHCT were pre-determined as one of the following: (a) severe, glucocorticoid-refractory recurrent inflammatory symptoms (either relapsing polychondritis, vasculitis, Sweet syndrome or undifferentiated systemic symptoms in the setting of elevated inflammatory markers), (b) persistent (≥3 months) cytopenias, including need for packed red blood cell and/or platelet transfusions, (c) coexistent myeloid malignancy or clonal abnormalities predictive of myeloid transformation (higher risk clonal somatic myeloid mutations or MDS-defining cytogenetic abnormalities). Four patients underwent allogeneic HCT at Mayo Clinic (Rochester, MN), while one was treated at the University of Michigan (Ann Arbor, MI, in consultation with Mayo Clinic). Conditioning regimen strategy was pre-determined as reduced intensity conditioning (RIC) with fludarabine (Flu) at a dose of 125 mg/m2 over 5 days, and melphalan (Mel) 140 mg/m2 (appropriate dose reductions in the setting of renal insufficiency was allowed). RIC was chosen due to the advanced age at presentation and to limit transplant-related mortality (TRM). Donor choices included both matched sibling (MSD) and matched unrelated donors (MUD) with T-cell replete peripheral blood stem cell graft. Since VEXAS is a precursor clonal condition and not an advanced stage myeloid malignancy, graft versus leukemia (GVL) was deemed not necessary; and an intensive GVHD prophylaxis strategy (similar to haploidentical donor alloHCT) that involved post-transplant cyclophosphamide (PTCy, 50 mg/kg per dose, day+3 and + 4), mycophenolate mofetil (MMF) till day+35, and tacrolimus (to be started day 5 post alloHCT and tapered over time as per investigator's preference) was planned (appropriate substitutions in the setting of contra-indications to PTCy were allowed). Granulocyte colony stimulating factor (G-CSF) use post-transplant was avoided to limit exacerbating underlying systemic inflammation. Anti-inflammatory therapy (prednisone, tocilizumab, and ruxolitinib among others) prior to alloHCT was allowed (if patients were on prednisone at the time of alloHCT, it was recommended to switch to an equivalent dose of hydrocortisone in the peri-transplant period to limit interference with PTCy, followed by a slow taper left to investigator's discretion). Overall, five patients (all males) with VEXAS have undergone allogeneic HCT with a median follow up of 9.6 (range: 2.9–16.2) months. Among these, three patients were diagnosed with UBA1 p.Met41Thr variant (#1, #4, and #5), 1 with UBA1 p.Met41Val (#2), and 1 with UBA1 c.118-1G > C (splice variant, #3). The most common indication for alloHCT was refractory inflammatory symptoms (n = 3, 60%), bone marrow failure with red blood cell transfusion dependence along with relapsing polychondritis (n = 1), while one patient had both MDS (morphological diagnosis only without any karyotypic or molecular abnormalities), and relapsing polychondritis (n = 1). Pre-transplant glucorticoid-sparing therapies (Table 1) included methotrexate, MMF, decitabine, hydroxychloroquine, testosterone, cyclosporine, rituximab, adalimumab, abatacept, omalizumab, dapsone, leflunomide, etanercept, infliximab, ruxolitinib, and tocilizumab but these were not effective in reducing prednisone use or improving cytopenias. Types of donor included matched unrelated (matched at HLA-A, B, C, DR and DQ alleles, n = 3), and MSD (n = 2). Median time to neutrophil engraftment was 25 (range: 20–33) days, while median time to platelet engraftment was 33 (range: 24–51) days. All patients received Flu/Mel conditioning at the pre-specified dosing; however, PTCy/tacrolimus/MMF GVHD prophylaxis could be administered in only four patients [one of the patients (#3) received tacrolimus and methotrexate GVHD prophylaxis due to sub-normal ejection fraction (48%) and mild left ventricular hypokinesis at pre-transplant assessment]. Post-transplant complications included encephalopathy due to secondary adrenal insufficiency (treated with stress doses of glucorticoid, n = 1), mucositis (grade II, n = 2), bacteremia/sepsis (n = 1), drug-induced rash/mild dermal hypersensitivity reaction (n = 2), grade 1 acute skin GVHD (n = 1), gastrointestinal infection (n = 1), and metabolic acidosis (n = 1). In patients with at least 100 days of follow-up post alloHCT (#1, #2, #3, and #5, Table 1), normalization of bone marrow morphology, that is, eradication of vacuoles in myeloid and erythroid precursors was achieved at day+100 (post alloHCT), and they were weaned off glucocorticoid therapy. Similarly, inflammatory markers (CRP and ESR) normalized by day+100, and at least two patients (#1 and #2) have been successfully weaned off immunosuppressive therapy (Figure 1A,B). One patient (#3) developed peri-transplant (day+1) metabolic (lactic) acidosis in the absence of sepsis, hypoperfusion or organ dysfunction. Clinically, this was self-limiting (resolved by day+8 without intervention), and attributed to metformin use. Mixed chimerism in both myeloid and lymphoid fractions was seen in three patients (#1, #2, and #5) in the absence of elevated inflammatory markers (Figure 1B). Although these three patients met criteria for neutrophil and platelet engraftment, mild/moderate thrombocytopenia and leukopenia remained persistent at varying degrees but without VEXAS-associated clinical symptoms. In two patients (#1 and #2), we demonstrated disappearance of UBA1 mutant clone in peripheral blood through Sanger sequencing (although this is a not an MRD assay due to lower detection threshold at around 20% variant allele fraction). Further, in one patient (#1), donor chimerism improved to 95% in CD3 and 100% in CD33 fraction at long-term follow-up (day+444) with normalization of hematopoietic parameters (tacrolimus was weaned off completely at 109 days post alloHCT), while in another patient (#2), stable mixed chimerism (50% in CD3 and CD33 fractions at day 322 post alloHCT) continues to persist along with mild leukopenia and thrombocytopenia (tacrolimus was weaned off at 292 days post alloHCT). To date, none of the patients developed grade II-IV acute GVHD or chronic GVHD. To our knowledge, this is one of the first prospective reports of patients with VEXAS successfully treated with alloHCT following a consistent plan of reduced-intensity Flu/Mel conditioning. All patients are alive, and without any VEXAS-related inflammatory features, with normalization of inflammatory markers (ESR/CRP), and bone marrow morphology. Thus far, persistent cytopenias, infections, and mixed donor/recipient chimerism are among the observed post-transplant complications. Mixed chimerism (both myeloid and lymphoid) was observed in some patients; however, none of these patients experienced recurrent inflammatory symptoms or worsening cytopenias, suggesting that it should not warrant any immediate intervention in the absence of systemic inflammation, high-risk somatic myeloid mutations or myeloid neoplasia and progressive cytopenias. Long-term implications of mixed chimerism in VEXAS post alloHCT are yet to be determined. Self-limiting peri-transplant metabolic acidosis (lactic acidosis, in the absence of hypoperfusion or organ dysfunction) of uncertain etiology is a unique complication seen in one of our patients, also observed in other reports.8 In summary, we demonstrate feasibility and successful allogeneic hematopoietic stem cell transplantation approach for VEXAS syndrome. Patients with refractory inflammation, cytopenias, and concurrent MDS are all appropriate indications for alloHCT. With the increased awareness of VEXAS, larger prospective cohorts are necessary for identifying transplant strategies with best long-term outcomes. Abhishek A. Mangaonkar conceptualized and wrote the paper. Kimberly J. Langer and Hassan B. Alkhateeb played a major role in planning, data collection, and revising the paper. Terra L. Lasho and Christy Finke performed the UBA1 PCR assay. Mark R Litzow, William J Hogan, Mithun V. Shah, Ronald S. Go, Matthew Koster, Matthew Samec, Kenneth J. Warrington, Mary Riwes, and Mrinal M. Patnaik contributed patients. Gabriel Bartoo, Jade Kutzke, and Kristen B. McCullough provided critical input in drug use and monitoring. Kaaren K. Reichard and Horatiu Olteanu reviewed the pathology data. All authors played a role in writing the manuscript. Authors would like to acknowledge the patients and their families for their active cooperation, advocacy, and partnership. Graphical abstract was created using BioRender.com. None of the authors report any conflicts in relation to the manuscript. For additional data, please contact [email protected] Appendix S1: Supporting information 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.

Topics & Concepts