Corticosteroid-refractory graft-versus-host disease (SR-GVHD) remains a significant source of morbidity after allogeneic hematopoietic cell transplantation. No standard therapy exists in this setting; however, recent studies have demonstrated a very promising role for ruxolitinib, an oral Janus kinase 1/2 inhibitor. With increasing evidence of efficacy for SR-GVHD, limited data exist describing complications of ruxolitinib use, specifically infectious complications during use in SR-GVHD. In this study we report outcomes and infectious complications at our institution with ruxolitinib use. Overall, 43 patients were treated with ruxolitinib for SR-GVHD, 19 for acute SR-GVHD and 24 for chronic SR-GVHD. With respect to acute SR-GVHD, 15 patients had grade III acute GVHD and 4 patients had grade IV acute GVHD. At 28 days, a response rate of 84% was detected. With respect to chronic SR-GVHD, 16 patients had moderate refractory disease and 8 had severe refractory disease. At around 28 days, a 63% response rate was detected. Overall, 42% of patients (n = 18) treated with ruxolitinib had a documented infectious event. Infectious events were significantly more common among patients treated for acute SR-GVHD (P < .005). Among patients treated for acute SR-GVHD, both viral (n = 11) and bacterial (n = 10) events were frequently encountered. Cytomegalovirus reactivation was detected in 4 patients without organ involvement in any patient. Bacteremia was the most common bacterial event (n = 8), and 2 patients died after development of bacteremia. Only 5 of 24 patients treated with ruxolitinib for chronic SR-GVHD developed infectious complications after initiation of therapy. Nearly an even number of viral (n = 3) and bacterial (n = 4) were detected. This study supports the use of ruxolitinib in SR-GVHD, with impressive responses observed in both acute and chronic SR-GVHD. Infectious complications were particularly frequent among patients treated for acute SR-GVHD, and nearly all these patients were concurrently on high-dose steroids while on ruxolitinib. This study suggests careful monitoring for viral reactivation is required for patients initiated on ruxolitinib, supports the role of continuing prophylactic antimicrobial measures in ruxolitinib-treated GVHD patients, and raises the question of whether bacterial prophylaxis should be considered among patients initiated on ruxolitinib for acute SR-GVHD, particularly while on high-dose steroids.

HLA matching by allele-level genotyping is largely based on genetic similarity between a few exons that encode the antigen recognition domain (ARD) of the HLA protein. Next-generation sequencing (NGS) can identify HLA genetic polymorphisms in non-ARD-encoding exons, introns, and untranslated regions, but the impact of these polymorphisms on hematopoietic cell transplantation (HCT) outcome is unclear. We performed NGS-based sequencing of 11 HLA loci on a well-characterized retrospective cohort of 166 unrelated donor-recipient HCT pairs. Genetic differences between HCT pairs were identified and visualized using a novel bioinformatics approach that directly compares phased full-length HLA sequences. Our approach was able to correctly classify HCT pairs without known HLA allele-level mismatches and also to identify a subset of HLA allele-matched HCT pairs with very few to no genetic differences in the sequenced HLA regions. This highly HLA genetically matched unrelated HCT group shows improved overall survival and reduced acute graft-versus-host disease compared with HCT pairs with HLA allele-level mismatches. These results suggest that direct genetic matching of HLA loci may offer an additional means of HCT donor selection beyond traditional HLA allele comparisons and suggests that genetic similarity as defined by HLA sequencing may have a novel role in unrelated HCT donor selection. Finally, our approach can enable larger cohort studies with adequate power to detect differences in other HCT outcomes based on genetic similarity within the HLA loci.

T cell-replete post-transplant cyclophosphamide (PTCy)-based protocols have led to increasing use of haploidentical allogeneic hematopoietic cell transplantation (haploHCT). With this approach, bidirectional alloreactivity causing nonengraftment or severe graft-versus-host disease (GVHD) is no a longer major barrier to haploHCT. PTCy eliminates alloreactive lymphocytes but spares CD34 stem cells and regulatory T lymphocytes, resulting in reliable hematopoietic recovery with relatively low incidence of GVHD. The immediate post-haploHCT course, usually before PTCy administration, is often complicated by cytokine release syndrome (CRS). The predictors of CRS and its effect on outcomes post-transplant have not been fully ascertained. We analyzed the outcomes of 66 patients who received haploHCT at our institution. Using published CRS criteria we identified 48 patients who developed CRS. In multivariate analysis peripheral blood grafts were significantly associated with grade ≥ 2 CRS, compared with bone marrow. Grade ≥ 2 CRS (compared with grade < 2) was not associated with differences in overall survival or nonrelapse mortality. Severe CRS was associated with a statistically nonsignificant trend toward higher incidences of grades III to IV acute GVHD, especially in the context of peripheral blood grafts. CRS is a common complication after T cell-replete peripheral blood haploHCT, but post-transplant survival outcomes may not be affected in those with severe CRS.

No abstract

Allogeneic hematopoietic cell transplantation (allo-HCT) is the only curative treatment modality for primary myelofibrosis (MF) and related myeloproliferative neoplasms. Older age at diagnosis and age-related comorbidities make most patients ineligible for allo-HCT, given concerns for nonrelapse mortality (NRM). Here we report the outcomes of 37 consecutive recipients of allo-HCT for MF performed at a single center between 2009 and 2018 with a standardized institutional protocol. Most patients received ruxolitinib before HCT (n = 32), and those with splenomegaly >22 cm received pretransplantation splenic irradiation. The median age at HCT was 60 years (range, 40 to 74 years), and 68% of the cohort carried a JAK2 driver mutation. All patients received fludarabine/busulfan-based conditioning; 22 patients (59%) received a reduced-intensity conditioning regimen. All patients received peripheral blood grafts, from a matched sibling donor in 16 patients (43%), an unrelated donor in 20 patients, and a haploidentical-related donor in 1 patient. Sixty-one percent had a Hematopoietic Cell Transplantation Comorbidity Index 3, 40% had a Karnofsky Performance Status score <90, and 24% had a high-risk DIPSS Plus score. With a median follow-up of 40.2 months (range, 16.9 to 115 months), the 3-year overall survival and relapse-free survival were 81.1% (95% confidence interval [CI], 64.4% to 90.5%) and 78.4% (95% CI, 61.4% to 88.5%), respectively. Only 2 patients relapsed/progressed after transplant. NRM at 2 years was 16.2% (95% CI, 6.5% to 29.9%). All patients engrafted. Sixteen patients were treated with ruxolitinib post-transplantation for graft-versus-host disease, graft rejection/relapse, or persistent MF. These results suggest that pretransplantation ruxolitinib, fludarabine/busulfan-based conditioning, and splenic management are keys to improved transplantation outcomes in patients undergoing allo-HCT for MF.