Blood Grafts Research Articles

PCN291 ACUTE GRAFT VERSUS HOST DISEASE (GVHD), CHRONIC GVHD AND NON-RELAPSE MORTALITY (NRM) ASSOCIATED WITH MISMATCHED ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION (ALLO-HSCT): A SYSTEMATIC REVIEW

When HLA-matched stem cell grafts are unavailable, mismatched sources may be used, which are often associated with increased risk of serious complications. We reviewed published evidence to assess complications associated with mismatched allo-HSCT in acute myeloid leukemia (AML) patients by graft source. Embase®, MEDLINE®, Cochrane databases and conference abstracts (ASCO, ASH) were systematically reviewed in November 2018. Main inclusion criteria: studies from 2008-2018, AML patients, allo-HSCT, mismatched, interventional prospective studies which reported complications. Main exclusion criteria: non-interventional, retrospective studies. Two analysts reviewed each abstract and publication using PICOS (Patient, Intervention, Comparators, Outcomes, Study Design) criteria and all results were recorded using preferred reporting items for systematic reviews and meta-analyses (PRISMA) diagram. Rates of acute GVHD, chronic GVHD, and NRM were extracted and compared. A total of 1,335 records were reviewed and 8 studies satisfied inclusion criteria. The incidence of grade III-IV acute GVHD events was 10.0% among patients receiving UM171-expanded umbilical cord blood (UCB) grafts, 12.2% with NiCord®-expanded-UCB, 19.4% with StemEx®-expanded-UCB and ranged from 3.2-21.2% with other mismatched allo-HSCT (majority mixed bone marrow/peripheral blood). Incidence of moderate-severe chronic GVHD was found to be 0.0% with UM171-expanded-UCB at 24 months, 11.5% with NiCord®-expanded-UCB at 12 months, 19.0% in unmanipulated-UCB grafts at 12 months and ranged from 6.5-10.6% in mismatched allo-HSCT assessed at time points between 12-24 months. GVHD prophylaxis regimens varied across studies. NRM at 24 months was 4.7% with UM171-expanded-UCB, 20.3% with NiCord®-expanded-UCB, 19.0% with unmanipulated-UCB and 23.0-30.0% with alternate graft sources. Patients treated with UM171-expanded-UCB grafts experienced the lowest rates of NRM and moderate-severe chronic GHVD at 24 months relative to UCB and other mismatched graft sources identified in the SLR. Additionally, rates of grade III-IV acute GVHD were lower in UM171-expanded-UCB grafts compared to most, but not all, other sources identified.

A longitudinal analysis of TCRβ repertoire diversity in immune tolerance of pediatric patients post-UCBT: T cell clonal deletion over clonal anergy

Abstract The mechanisms of long term immune tolerance after successful hematopoietic stem cell transplantation are not fully elucidated. We set up serial experiments to test central and peripheral mechanisms sustaining tolerance after HLA-mismatched unrelated cord blood transplant (UCBT) in the GvH direction. TCRβ ImmunoSEQ® (Adaptive Biotechnology®) was applied for tracking and characterizing host-reactive T cell clones from graft or patient samples. Seven UCBT recipients with non-malignant diseases were studied. All patients enrolled on the same reduced intensity conditioning trial (NCT01852370). Patients achieved tolerance if discontinued immunosuppression, exhibited immunocompetence without GvHD or graft rejection. Purified T cells responses to host APCs were measured by mixed lymphocyte reaction, cytokine bioplex, and micro CTL assays. T cells clones expanding >0.1% frequency in response to host APC were tracked and characterized with TCRβ ImmunoSEQ®. ImmunoSEQ® revealed the disappearance of host-reactive clones that expanded from an aliquot of the infused cord blood graft by 7 days of stimulation with purified host APC. Most of these allo-reactive T cell clones were not identifiable even in the presence of IL2 supplementation to break anergy. Serial experiments post-UCBT revealed novel host-reactive clones that were not identifiable within the alloreactive pool of the graft pre-UCBT. In summary, the acquisition of immune tolerance in the GvH direction post-UCBT is characterized by hypo-reactivity in all assays employed towards host APC. The disappearances of most host-specific clones in the graft indicate that deletion is a major mechanism of long term tolerance with possibly a small degree of anergy in contribution.

New graft manipulation strategies improve the outcome of mismatched stem cell transplantation in children with primary immunodeficiencies

Mismatched stem cell transplantation is associated with a high risk of graft loss, graft-versus-host disease (GvHD), and transplant-related mortality. Alternative graft manipulation strategies have been used over the last 11years to reduce these risks. We investigated the outcome of using different graft manipulation strategies among children with primary immunodeficiencies. Between 2006 and 2017, 147 patients with primary immunodeficiencies received 155 mismatched grafts: 30 T-cell receptor (TCR) αβ/CD19-depleted grafts, 43 cord blood (CB) grafts (72% with no serotherapy), 17 CD34+ selection with T-cell add-back grafts, and 65 unmanipulated grafts. The estimated 8-year survival of the entire cohort was 79%, transplant-related mortality was 21.7%, and the graft failure rate was 6.7%. Posttransplantation viral reactivation, grade II to IV acute graft-versus-host disease (aGvHD), and chronic graft-versus-host disease (cGvHD) complicated 49.6%, 35%, and 15% of transplantations, respectively. Use of TCRαβ/CD19 depletion was associated with a significantly lower incidence of grade II to IV aGvHD (11.5%) and cGvHD (0%), although with a greater incidence of viral reactivation (70%) in comparison with other grafts. T-cell immune reconstitution was robust among CB transplants, although with a high incidence (56.7%) of grade II to IV aGvHD. Stable full donor engraftment was significantly greater at 80% among TCRαβ+/CD19+-depleted and CB transplants versus 40% to 60% among the other groups. Rapidly accessible CB and haploidentical grafts are suitable alternatives for patients with no HLA-matched donor. Cord transplantation without serotherapy and TCRαβ+/CD19+-depleted grafts produced comparable survival rates of around 80%, although with a high rate of aGvHD with the former and a high risk of viral reactivation with the latter that need to be addressed.

Association Between Adverse Clinical Outcomes After Coronary Artery Bypass Grafting and Perioperative Blood Transfusions.

Bleeding is a serious complication of coronary artery bypass grafting that often leads to blood transfusion. Approximately 50% of patients who have the surgery receive blood products, and blood transfusions play a role in adverse outcomes after the surgery. To examine the association between perioperative blood transfusion and postoperative adverse outcomes in patients undergoing coronary artery bypass grafting. A systematic review of the literature, via the matrix method of quality evaluation, was conducted. PubMed, CINAHL, and Science Direct databases for 2000 through 2016 were searched. Inclusion criteria were articles published in English and original research related to clinical outcomes of blood transfusion after coronary artery bypass grafting. Seventeen articles were included in the review. Mortality, both short- and long-term, was significantly higher in transfusion patients than in nontransfusion patients. Patients with transfusion of red blood cells had higher resource utilization and more complications, including infection, pneumonia, renal failure, graft occlusion, and atrial fibrillation, than did nontransfusion patients. An association exists between red blood cell transfusions and adverse clinical outcomes for patients undergoing coronary artery bypass grafting. Transfusion of red blood cells is sometimes unnecessary, may be injurious, and should be used cautiously. Even a single-unit increase in perioperative red blood cell transfusions can have a significant adverse impact on outcomes. Individual benefits and risks should be weighed before transfusion to avoid adverse outcomes. Transfusion guidelines should be reviewed to include the latest evidence to guarantee the most appropriate use of blood products.

Predictors of Donor/Host (D/H) Lineage Specific Chimerism Trends in Pediatric Patients Following Allogeneic Hematopoietic Cell Transplantation

The outcome of allogeneic HCT is dependent on several pre-transplant variables. We hypothesize that these pre-transplant factors may influence the donor/host chimerism post HCT, which in its turn may affect the outcome. We performed an analysis to study lineage specific D/H chimerism patterns of total blood leukocytes, myeloid cells/neutrophils, B-cells, NK-cells and T-cells. All consecutive patients who underwent an allogeneic HCT between 01/2010 to 06/2015 on the Pediatric BMT Service at MSKCC were reviewed. Chimerism analyses were performed by short tandem repeat (STR) polymorphism analysis at the American Red Cross Blood Services (Philadelphia, PA). Lineage specific donor chimerism post HCT was studied including donor chimerism trend, and factors with potential impact on chimerism including: age, disease, graft source, and T-cell depletion (TCD). Preliminary analyzes performed on this cohort included Wilcoxon Rank Test and cox proportional hazard analyses. 140 patients with a median age of 11.5 years were included. They included 97 patients with hematologic malignancies, 27 patients with non-malignant hematologic disorders and 16 patients with immunologic disorders. Cytoreduction included TBI in 48 patients or chemotherapy only in 98 patients. 105 patients received T-cell depleted grafts, 28 patients unmodified marrow or peripheral blood grafts, and 7 patients unrelated cord blood grafts. No patients received donor leukocyte infusions (DLI). Observational studies revealed that full donor chimerism of myeloid cells, B-cells and NK-cells, but not T-cells occurred early post-transplant. In Wilcoxon Rank Test, there was no difference in the percentage of total donor leukocytes at 3 months vs 12 months post HSCT (n=30), while the median of donor T-cell chimerism was 51% at 3 months and 91% at 12 months post HSCT (p The studies of the impact of different factors on chimerism were performed, including age, disease, and type of graft. Loss of donor total leukocytes chimerism between 3 and 12 months post-transplant was significantly more pronounced for patients This preliminary analysis of lineage specific chimerism post-transplant showed that donor T-cells may take one year to fully recover post-transplant, mostly following T-cell depleted grafts, without intervention. Cord blood grafts were associated with high donor chimerism throughout the post-transplant period. Lastly, factors associated with loss of donor chimerism post-transplant were younger age and non-malignant disorders. More in-depth analyses are being performed.

Comparison of Outcomes of Allogeneic Hematopoietic Cell Transplantation in Patients with Acute Myeloid Leukemia (AML) with Antecedent History of Philadelphia-Negative Myeloproliferative Neoplasm with De Novo AML and with AML Arising from Myelodysplastic Syndrome: A Study from the Center for International Blood and Marrow Transplant Research (CIBMTR)

Allogeneic hematopoietic cell transplantation (HCT) is the only treatment modality associated with long-term disease control in patients with acute myeloid leukemia (AML) with antecedent history of Myeloproliferative neoplasms (MPN), however outcomes data are scanty. We compared the outcomes of 177 patients with AML with prior MPN (post MPN AML) with 4749 patients with de no AML, and 1104 patients with AML with prior myelodysplastic syndrome (Post MDS AML) with an aim to identify patient, disease and transplant related factors associated with outcomes. The study cohort included patients aged ≥ 40 years undergoing first HCT between 2001 and 2015. Patients with Philadelphia or BCR-ABL positive MPN, undergoing transplants from syngeneic donors, cord blood transplants, or ex vivo T-cell depletion were excluded. Comparisons of outcomes of HCT were done in post MPN AML versus de novo AML and post MPN AML versus post MDS AML using multivariate regression analysis. A higher proportion of patients with post MPN AML had monosomal or non-monosomal poor risk karyotype (39%) in comparison to de novo AML (23%) and post MDS AML (34%). Active leukemia at HCT was seen in 52%, 26% and 42% in patients with post MPN AML, de novo AML and post MDS AML, respectively. In multivariate cox model, patients with post MPN AML in morphological leukemia free state (blasts Patients with post MPN AML are at high risk of relapse compared to de novo AML and post MDS AML. Peripheral blood grafts result in lower relapse risk in patients with post MPN and post MDS AML. Relapse prevention strategies are required to optimize the outcomes of HCT in post MPN AML.

Safety and Efficacy of Donor Leukocyte Infusion for Single Unit Umbilical Cord Blood Transplantation to Accelerate Immune Reconstitution Prior to Thymopoiesis

Success of reduced intensity umbilical cord blood transplantation (UCBT) in chemotherapy naive patients is limited by an increased risk of mixed chimerism, infection, and delayed immune reconstitution (IR). In a prospective clinical trial, we utilized risk-adapted alemtuzumab dosing, tested the safety of donor leukocyte infusion (cDLI) from a fraction of the thawed graft and evaluated it's efficacy in impacting IR, viral infection, and/or donor chimerism. Patients with inborn errors of metabolism, immunity, or hematopoiesis underwent a single unit UCBT following a reduced-intensity conditioning regimen of alemtuzumab, hydroxyurea, fludarabine, melphalan, and thiotepa (ClinicalTrials.gov: NCT00744692). Up to 5% of the graft was re-cryopreserved for potential cDLI and administered in cases of declining donor Tcell chimerism, viral infection, or delayed IR. Alemtuzumab levels were obtained on the day of UCBT. Chimerism and IR were measured at standard time points. 33 patients had cDLI stored and 18 patients received cDLI at a median of 65 days post-UCBT. Indications included mixed whole blood or CD3 chimerism, viral infection, or delayed IR. All patients engrafted with neutrophils at a median of 15 days post-UCBT, however, one patient experienced secondary graft failure at day +38, despite cDLI 28 days post-UCBT. Improvement in chimerism was seen in 5 of 9 patients, viral load and/or symptoms of infection in 5 of 9 patients and absolute CD3 count in 5 of 5 patients. 15 patients developed acute grade I-III GVHD at a median of 69 days post-UCBT. In those who received cDLI, 8 developed grade I-II skin GVHD at a median of 19 days post-cDLI. 2 patients with a history of GVHD received subsequent cDLI without GVHD flare. No patients who received cDLI developed grade III or IV acute or extensive chronic GVHD, while 4 patients who did not receive cDLI developed grade III GVHD affecting skin and/or the gastrointestinal tract. There was no difference in the cumulative incidence of acute GVHD between those who received cDLI and those who did not (44% vs. 33%; p=0.81). The 3-year overall and event-free survival of all patients was 94% and 91% respectively, with treatment-related mortality (TRM) of 5% 1 year post-UCBT. We report outcomes on the largest prospective trial utilizing cord blood grafts for nonmalignant disease. cDLI is safe in patients undergoing single-unit UCBT as all engrafted. Greater than half of the patients had improvement in the primary indication for its use, without increased GVHD risk. Greatest benefit was seen in patients with delayed IR and TRM at 1 year was exceptionally low. In sum, cDLI may be a simple and effective way to address mixed chimerism, infection, and accelerate IR during the vulnerable time prior to the onset of thymopoiesis. Further characterization of the merits of cDLI should be explored in a larger clinical trial.

Allogeneic Transplantation in Septuagenarians Using Post-Transplant Cyclophosphamide (PTCy): Similar Outcomes to Younger Patients Except for Higher Late Non-Relapse Mortality (NRM)

Advances in allogeneic blood and marrow transplantation (alloBMT), especially nonmyeloablative (NMA) conditioning and haploidentical donors, have made this therapy available to older people in whom it was historically contraindicated or not available. Since Jan 1, 2009, 108 consecutive patients older than 70 (Table 1) underwent alloBMT at Johns Hopkins (through March 31, 2018). The median age was 72 (range 70-78). Refined disease risk index (DRI) was low in 10 (9%), intermediate in 83 (77%), and high or very high in 14 (13%). All patients received NMA conditioning. Haplo donors were used in 94 (87%), including 80 children, 7 grandchildren, and 4 nieces or nephews; matched sibling donors were used in 7 patients and matched unrelated donors in 6. All patients received PTCy, MMF from day 5 to 35, and either tacro or siro from day 5 until day 60 - 180. With a median follow-up of 938 days, the 2 year overall survival is 49% (95% CI 40-61), and 2 year PFS was 41% (95% CI 32-53) (Figure 1). The 180 day cumulative incidence (CuI) of non-relapse mortality (NRM) is 14% (CI 8-21), but 2-year NRM was 29% (95% CI 20-39%), with many late deaths (Fig 1). The 2 year CuI of relapse was 30% (95% CI 20-39%). Graft failure was 6% (95% CI 1-10%) at 100 days and was not different between BM and PBSCT grafts (7.7% vs 6.1%). The CuI of grade III-IV acute GVHD was 7% (95% CI 2-12%), and cGVHD at 2 years was 8% (95% CI 3-14%) (Figure 2). In univariate analyses, there were no differences in NRM, OS, or PFS between bone marrow and peripheral blood grafts. HCT-CI (0, 1-2,or 3 or more) did not predict NRM. Of 78 patients who were alive and had a weight recorded on day 180, the median weight change was loss of 4.4 kg. NRM after day 180 was 10/39 (26%) in patients who lost more than 4.4kg and 3/39 (8%) in those who lost less than 4.4kg (p=0.035). In conclusion, alloBMT with PTCy is relatively safe and feasible in septuagenarians. Acute and chronic GVHD rates with PTCy are low and similar to those seen in younger patients. Relapse rates are also similar. Early (6 month) NRM is also low, but overall NRM is higher at least in part because of late deaths that appear to be in part age-related (e.g., secondary malignancy, dementia, fall, CVA). We are currently studying functional outcomes of older patients undergoing alloBMT in a prospective trial, as well as potential predictors of long-term outcomes.

Oral Brincidofovir Decreased HHV-6 Viremia in Hematopoietic Cell Transplant Recipients: Results from the Suppress Study

Introduction Human herpesvirus 6B (HHV-6) is detected in plasma in ∼40% of allogeneic hematopoietic cell transplant (HCT) recipients and ∼75% of cord blood recipients. HHV-6 viremia, particularly at levels >104 copies/mL, is associated with increased risk for HHV-6 encephalitis. Ganciclovir and foscarnet have antiviral activity against HHV-6, but studies of pre-emptive or prophylactic treatment have not been successful in preventing HHV-6 encephalitis. Prophylaxis for HHV-6 has never been studied in a randomized trial. Brincidofovir (BCV, CMX001) has potent in vitro activity against HHV-6 (EC50=0.007 µM), and unlike cidofovir, has good CNS penetration. Objectives In vivo activity of BCV against HHV-6 has not been demonstrated. We tested plasma from participants in the SUPPRESS clinical trial to explore the potential of BCV to prevent HHV-6 viremia. Methods SUPPRESS was a randomized, double-blind, placebo (PBO)-controlled trial of oral BCV for cytomegalovirus (CMV) prophylaxis after allo-HCT. 452 adult CMV-seropositive HCT recipients without CMV viremia at screening were randomized 2:1 to receive BCV or PBO twice-weekly until week 14 post-HCT. We selected subjects who were randomized within 2 weeks of allo-HCT, who did not have HHV-6 viremia at baseline, and who received at least 6 doses of BCV or PBO within the first 3 weeks of randomization for HHV-6 testing. Plasma samples from the first six weeks post-HCT were tested for HHV-6 with quantitative PCR (Viracor Eurofins, Lee's Summit, MO). We compared HHV-6 viremia between subjects who received BCV and those who received PBO using the Kaplan-Meier method, log rank test, and Cox proportional hazards model. Results 92 subjects in the BCV group and 61 in the PBO group met criteria for inclusion: 64% were male, 84% were white, and median age was 57 years (range: 18-76). 6% of subjects received cord blood grafts, 5% had haploidentical donors, and 8% had mismatched donors. 41% had ex vivo T-cell depletion or received serotherapy with antithymocyte globulin or alemtuzumab. Baseline characteristics were balanced between groups. 15% of BCV subjects vs. 31% of PBO subjects had detectable HHV-6 viremia within 6 weeks after HCT. The cumulative incidence of HHV-6 viremia was significantly lower in BCV subjects (Figure 1). Two subjects (2%) in the BCV group had HHV-6 viremia >103 copies/mL compared to 7 (11%) in the PBO group (Figure 2). In the study overall, no subjects in the BCV group and one in the PBO group developed HHV-6 encephalitis. Conclusion BCV reduced the incidence of detectable HHV-6 viremia in HCT recipients enrolled in a randomized, placebo-controlled clinical trial. BCV prophylaxis could prevent the morbidity and mortality associated with HHV-6 in HCT recipients.

Time to Transplantation (TTT) for Acute Myeloid Leukemia (AML) in First Complete Remission (CR1) Is Comparable Among Adolescent and Young Adults (AYAs) and Older Adults

Background AYA patients have a myriad of specific psychosocial and other challenges which may influence their ability to obtain appropriate treatment. Healthcare disparities regarding access to healthcare have been well described in this population. We hypothesized that hurdles faced by the AYA patient population also apply to patients with AML requiring allogeneic hematopoietic cell transplantation (HCT), resulting in AYAs having a longer TTT compared to older adults. Objectives In this study we aimed to compare TTT, defined as time from diagnosis to transplant for AML in CR1 between AYAs and older adults. Additionally, we aimed to determine if TTT has changed over time for this cohort and to identify factors associated with TTT. Methods BMT Program database was used to identify allogeneic HCT recipients with AML in CR1 from January 2007 to June 2018 for AYAs (age 18-39 years) and older adults (age 40-60 years). All the identified patients underwent allogeneic stem cell transplant in complete remission 1 (CR1). Patients receiving cord blood graft source were excluded. AYA and older adults were compared with Wilcoxon, Chi-square, or Fisher's exact test. Change in TTT over time was assessed with Spearman correlation (r). Factors associated with TTT were assessed with Wilcoxon or Jonckheere-Terpstra test. Results A total of 105 patients were identified: 24 AYAs and 81 older adults. Baseline characteristics were similar between AYAs and older adults aside from Karnofsky performance status; AYAs had better performance status than older adults (p=0.012) . Time to Transplant did not differ from between AYA and older adults (median 4.1 vs 4.0 months, p=0.61). There was no evidence of a change in TTT in more recent years in AYA (r=0.15, p=0.48) or older adults (r=-0.02, p=0.85). The only variable associated with TTT, as expected, was donor type (unrelated donors median 4.3 vs related donors 3.8 months, p=0.022) (Table 1). Conclusion Time to transplant for AYAs in AML is comparable to older adults and has not changed over time. Our analysis is limited to experience from a single center and a larger multicenter effort that takes into account other sociodemographic mediators of healthcare disparities is needed to better describe the association of age and time to allogeneic HCT in AML CR1. Our next step is to explore various psychosocial factors in this cohort and determine if they play a role in the TTT and also determine if TTT has an impact on outcomes.

Myeloablative Conditioning Using Timed Sequential Busulfan in Older Patients with Acute Myeloid Leukemia: Long Term Results of a Prospective Phase II Clinical Trial

Background Myeloablative conditioning (MA) reduces relapse as compared to reduced intensity (RIC), [Scott BL JCO 2017]. But it is avoided in older patients and those with comorbidities due to high non-relapse mortality (NRM), whose outcomes remain dismal. A CIBMTR study of AML patients ≥60 years (100% CR1) showed NRM of 32-34%, relapse rate of 33-37%, progression free survival (PFS) of 31-34% and overall survival (OS) of 34-36% at 2 years. [McClune. JCO 2010] A prospective trial in AML (≥60 years, 100% CR1) showed better 2-year PFS (42%) and OS (48%). [Devine. JCO 2015] Herein, we report the results of MA HCT using timed-sequential (TS) busulfan (Bu) in 71 AML patients up to age 73. Methods Patients received intravenous Bu in TS manner targeted to achieve an area under the curve of 16,000-20,000 μmol.min, with fludarabine (flu) 40 mg/m2 intravenously for four days. Results Median age was 64 years (range, 29-73). Donor was HLA matched unrelated (n=45, 63%) or related (n=26, 27%). Most received peripheral blood graft (n=46, 65%). Only 26% (n=18) had MRD-negative CR1/2 at HCT; others had MRD-positive CR1/2 (13%, n=9), primary induction failure (PIF; 45%, n=32) or relapsed disease (16%, n=11). Nearly half (47%) had the European LeukemiaNet (ELN) adverse risk disease. More than half (55%) had HCT-CI ≥ 3. Median follow-up was 40 months (range, 21-63). The cumulative incidences of grade II-IV and III-IV acute GVHD at day 100 were 39% (95% CI, 28-51%) and 10% (95% CI, 3-17%), respectively and extensive chronic GVHD at 1 year was 20% (95% CI, 10-29%). NRM was 6% (95% CI, 0-11%) at day 100 and 24% (95% CI, 14-34%) at 2 years. At 2 years, relapse was 28% (95% CI, 3-53%) in the CR/MRD-negative group vs 48% (95% CI, 34-62%) in others; progression-free survival (PFS) was 78% (95% CI, 61-99.6%) vs 25% (95% CI, 16-40%) and overall survival (OS) was 78% (95% CI, 61-99.6%) vs 33% (95% CI, 22-48%), respectively. [Fig 1a] Two year OS was was 67% (95% CI 51-87%) in the CR group, irrespective of MRD [Fig 1b]. In multivariate analysis (MVA) adjusted for covariates higher dose Bu vs lower dose [HR 0.39, 95% CI 0.18-0.85, p=0.02] was the only significant predictor of lower relapse risk. In MVA adjusted for age, Bu dose and ELN risk, not being in CR/MRD-negative state was the only predictor of significantly inferior PFS [HR 3.2, 95% CI, 1.3-8.2, p=0.01]. For OS, MVA adjusted for age, Bu dose and ELN risk, HCT-CI ≥3 [HR 1.98, 95% CI, 1-3.9, p=0.05] and not being in CR/MRD-negative state [HR, 2.7, 95% CI 0.98-7.2, p=0.05] were associated with higher mortality. Conclusion Our outcomes of MA HCT in an extremely high risk older population are promising especially in CR patients, whose outcomes approach that of younger patients. This was achieved by delivering Bu in TS manner which minimized toxicity and potentially increased efficacy. This approach warrants further investigation.

Impact of ABO Mismatch on Outcomes of Allogeneic Hematopoietic Stem Cell Recipients: 30 Years of Experience with 1016 Patients

Introduction Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is widely used to treat malignant and non-malignant hematological diseases.The impact of ABO mismatch on outcome following transplantation remains controversial.In this study, our aim is to define effects of ABO mismatch on engraftment, graft vs host disease, relapse free survival (RFS) and overall survival (OS) in patients who underwent allo-HSCT. Patients and Methods Between 1988 and 2016, we retrospectively identified 1016 patients who underwent allo-HSCT at Ankara University School of Medicine, Department of Hematology.Chi-square and FisherOs exact tests were used where appropriate in comparison.Cox regression model and Kaplan Meier curves were applied for survival analysis.P Results The median follow-up period was 34.7 months (range, 0.2-229).In our cohort, there were 420 (41.3%) ABO-mismatched transplants occurred including 167 (16.4%) major, 197 (19.4%) minor and 55 (5.4%) bidirectional mismatches.The pre-transplant characteristics of patients are summarized in Table.Allo-HSCTs from unrelated donors and peripheral blood grafts were detected higher in ABO mismatched patients vs ABO matched patients (28% vs 11%, P 1/128). Pure red cell aplasia was diagnosed in 5 (3%) major ABO mismatched patients.Major ABO mismatch (HR:1.46, 95% Cl:1.06-2.03;P=0.022) was found to be related with lower RFS and OS (HR:1.31, 95% Cl:1.06-1.62;P=0.013).3-year OS and 1-year RFS were lower with major ABO mismatch (38% vs 47%, P=0.02; 15% vs 24%; P=0.02) (Figure). Conclusion Engraftment failure was detected higher with ABO mismatch as well as major ABO mismatch was related with lower RFS and OS although the cohort is heterogeneous.Close monitoring and early treatment strategies for expectable complications would reduce the number fatal events by ABO mismatched allo-HSCT.

Phase I/II Study of Stem-Cell Transplantation Using a Single Cord Blood Unit Expanded Ex Vivo With Nicotinamide

PurposeIncreasing the number of hematopoietic stem and progenitor cells within an umbilical cord blood (UCB) graft shortens the time to hematopoietic recovery after UCB transplantation. In this study, we assessed the safety and efficacy of a UCB graft that was expanded ex vivo in the presence of nicotinamide and transplanted after myeloablative conditioning as a stand-alone hematopoietic stem-cell graft.MethodsThirty-six patients with hematologic malignancies underwent transplantation at 11 sites.ResultsThe cumulative incidence of neutrophil engraftment at day 42 was 94%. Two patients experienced secondary graft failure attributable to viral infections. Hematopoietic recovery was compared with that observed in recipients of standard UCB transplantation as reported to the Center for International Blood and Marrow Transplant Research (n = 146). The median time to neutrophil recovery was 11.5 days (95% CI, 9 to 14 days) for recipients of nicotinamide-expanded UCB and 21 days (95% CI, 20 to 23 days) for the comparator (P < .001). The median time to platelet recovery was 34 days (95% CI, 32 to 42 days) and 46 days (95% CI, 42 to 50 days) for the expanded and the comparator cohorts, respectively (P < .001). The cumulative incidence of grade 2 to 4 acute graft-versus-host disease (GVHD) at day 100 was 44%, and grade 3 and 4 acute GVHD at day 100 was 11%. The cumulative incidence at 2 years of all chronic GVHD was 40%, and moderate/severe chronic GVHD was 10%. The 2-year cumulative incidences of nonrelapse mortality and relapse were 24% and 33%, respectively. The 2-year probabilities of overall and disease-free survival were 51% and 43%, respectively.ConclusionUCB expanded ex vivo with nicotinamide shortens median neutrophil recovery by 9.5 days (95% CI, 7 to 12 days) and median platelet recovery by 12 days (95% CI, 3 to 16.5 days). This trial establishes feasibility, safety, and efficacy of an ex vivo expanded UCB unit as a stand-alone graft.

Efficacy Analysis of Unrelated Umbilical Cord Blood Transplantation for the Treatment of Wiskott-Aldrich Syndrome

To explore the clinical efficacy and safety of unrelated umbilical cord blood transplantation (UCBT) for the treatment of Wiskott-Aldrich syndrome(WAS). Five pediatric patients with WAS received single UCBT were retrospectively analyzed. The median age of these male patients was 268 days (range, 3 days -695 days). Among them, 2 patients were transplanted with a 6/6 matched cord blood graft，the other 3 patients received a 5/6 matched cord blood graft. Myeloablative conditioning regimen was applied, and all patients received a combination of cyclosporine and mycophenolate mofetil for the prophylaxis of graft versus host disease (GVHD). The recovery time of neutrophils and platelets as well as chimerism after transplantation were taken as the evidence of hematopoietic reconstruction. All the five pediatric patients had hematopoietic recovery. A median time of neutrophil cells after transplantation was at 15.8 days (range,11 days -25 days), and platelet recovery was at a median of 20.4 days(range,12 days-30 days). Chimerism data were available for 5 patients at 30 days after UCBT, 4 out of the 5 patients had full donor chimerism and only one patient had mixed chimerism. There were 2 cases with pre-engraftment syndrome, 3 cases with acute GVHD gradeⅠ-Ⅲ, 4 cases with pulmonary infection and cytomegalovirus infection, but chronic GVHD was not observed in 5 cases. Four patients were alive with a median follow-up of 12.3 months (range, 5 months-17 months), and one patient had died at 22 days after UCBT. Unrelated umbilical cord blood transplantation is a safe and effective treatment method for Wiskott-Aldrich syndrome.

Intrabone Delivery of CD34+ Cells Using an Optimized Delivery System Does Not Enhance Engraftment in a Rhesus Macaque Model of Hematopoietic Stem Cell Transplantation

Introduction: Umbilical cord blood (UCB) grafts are the only option for a significant minority of patients who require hematopoietic stem cell transplantation (HSCT) but lack a suitable related or unrelated donor. While UCB can serve as a suitable ‘off the shelf’ graft for many patients, units with the best HLA match often contain low and sometimes insufficient numbers of CD34+ cells for use in transplantation, particularly for adult patients. Furthermore, UCB grafts contain lower CD34+ cells number compared to BM or PBSC grafts, which leads to longer engraftment times and a higher risk of graft failure. BM and PBSC grafts are usually injected intravenously, homing to the bone marrow after several hours in the circulation. During this time, CD34+ cells are lost in the lungs, liver and spleen with typically < 20% making it to the bone marrow.1 Investigators have sought to overcome the limitation of low cell dose in UCB grafts and loss of CD34+ cells in the circulation by injecting CD34+ cells directly into the bone marrow space. However, we have recently shown that conventional intrabone delivery methods used in investigational trials to transplant UCB, result in low-level retention of hematopoietic progenitor cells in the intrabone space. Recently, we developed an optimized intrabone (OIB) transplant method using computer controlled low pressure and low volume injection (controlled infusion rate <0.2ml/min, total volume <5ml) under CT guidance. Utilizing porcine and rhesus macaque models, we have shown that OIB delivery of CD34+ cells improves intrabone retention, preventing circulation of CD34+ cells through the lungs, which is observed with conventional non-optimized intrabone methods.2 Here we conducted experiments using an autologous myeloablative transplant model in rhesus macaques comparing engraftment of gene-marked CD34+ cells transplanted intravenously, with cells transplanted using OIB delivery.Methods: Rhesus macaques received GCSF and plerixafor mobilization prior to apheresis. Products were CD34+ selected using Miltenyi beads. CD34+ cells were split equally for transduction with lentiviral vectors encoding the reporters GFP and YFP. After myeloablative conditioning with 10Gy total body irradiation, half the graft was injected directly intra-bone using the OIB method, with the other half of the autograft simultaneously being injected intravenously via slow iv push. Peripheral blood samples were measured daily by flow cytometry to assess the proportion of engrafting cells deriving from each source. To address whether OIB transplantation could allow engraftment of low CD34+ cell numbers, as found in UCB units, the cell doses transplanted in one rhesus macaque recipient were reduced to only 0.5 x 106CD34+ cells/kg.Results: CD34+ cells injected intrabone utilizing the OIB method engrafted in all 3 animals. However, flow cytometric analysis gating on GFP vs YFP positive neutrophils showed CD34+ cells injected utilizing OIB delivery did not engraft quicker than IV transplanted cells. Sequential monitoring of neutrophils over 30 days showed the contribution to hematopoiesis of OIB delivered cells was significantly lower than the cells injected IV (Table 1.)Conclusions: We developed a novel intrabone delivery system that optimizes the retention of CD34+ cells into the bone marrow space. Although autologous CD34+ cells injected using this OIB transplant method were capable of engrafting in rhesus macaques that had undergone myeloablative conditioning, they did not engraft faster and contributed less to hematopoiesis than CD34+ cells simultaneously transplanted using conventional IV infusion. These data raise questions over whether intrabone delivery, even when using techniques to optimize intrabone retention, has utility in improving CD34+ cell engraftment.

Risk Factors of Clostridium Difficile Infection in Hematopoietic Stem Cell Transplant Recipients: A Systemic Review and Meta-Analysis

Introduction:C. difficile infection (CDI) is a common nosocomial infection. Immune suppression is thought to be a major risk factor for CDI. Hematopoietic stem cell transplant (HSCT) recipients represent a major subgroup of immunologically vulnerable patients. It is unclear if these patients have novel risk factors for CDI development. We, therefore, performed a systematic review and meta-analysis to evaluate the predictors of CDI in HSCT recipients.Methods:A systematic search of Medline, Embase, and Web of Science was performed from January 2000 up to June 2018. All studies that assessed risk factors associated with CDI development in HSCT recipients were eligible for inclusion. Data on clinical characteristics and risk factors associated with CDI development were collected. Study quality was assessed using the Newcastle-Ottawa scale. Meta-analyses were performed using random effects models and pooled odds ratios with 95% confidence interval (CI) for risk factors reported in ≥ 2 studies were calculated.Results:Overall, 17 studies, including 6328 HSCT patients were included. Of those 874 patients developed CDI. The rate of CDI development in these patients was 13.81% with a follow up ranging from 1- 36 months. A total of 34 different risk factors were studied, of which 20 were identified in ≥ 2 studies. Our analysis showed that cephalosporin use (Odds ratio [OR] 2.21, 95% confidence interval [CI] 1.26-3.87), cord blood transplant (OR 1.36, 95% CI 1.01-1.83), graft versus host disease (GVHD) (OR 1.63, 95% CI 1.07-2.48) & prior hospitalization within 90 days of HSCT (OR 1.85, 95% CI 1.21-2.83) were associated with statistically significant increased risk of CDI in HSCT patients (Figure 1). On the other hand, many characteristics of HSCT patients including age, diabetes, lung disease, fluoroquinolones, proton pump inhibitors, rituximab, mucositis, myeloablative therapy, melphalan use before BMT, total body irradiation, growth factor use during admission, neutropenia, autologous vs allogenic transplant and use of matched related donor were not associated with increased risk of CDI.Conclusion:HSCT recipients are at an increased risk of CDI compared to general population. In addition to the common risk factors like cephalosporin use and prior hospitalization, these patients have novel characteristics including presence of GVHD and cord blood transplant that puts them at a higher risk of CDI. HSCT patients with these risk factors may warrant a closer surveillance for early detection and treatment of CDI. DisclosuresNo relevant conflicts of interest to declare.

Donor Lymphocyte Infusion from G-CSF-Primed, Unmanipulated Whole Blood Is Safe and Improves Chimerism in HLA-Matched and Haploidentical Transplantation

Introduction: Donor lymphocyte infusion (DLI) is a useful therapy for mixed chimerism (MC) or measurable residual disease (MRD) after allogeneic hematopoietic cell transplantation (HSCT) in patients with malignancies, while it is less effective for overt relapse and is limited by graft-versus-host disease (GVHD). Multiple DLI products can be obtained from a single leukoapheresis without granulocyte-colony-stimulating factor (G-CSF) priming, while cryopreserving surplus is an alternative for peripheral blood grafts. The ideal DLI CD3+ cell dose remains undefined ranging from 1-500x106/kg. However, leukoapheresis and cryopreservation are not risk-free, they are expensive and their costs prohibitive for many patients in developing countries. Furthermore, T-cell doses present in standard whole blood units can reach ≥1x106/kg in healthy donors which can be further increased after G-CSF exposure. For these reasons we have used G-CSF-primed whole blood units (WB-DLI) in HLA-matched and haploidentical (Haplo) transplant recipients and report our single-center experience.Methods: Patients ≥16 years of age who received WB-DLI at any time after HLA-matched related or Haplo peripheral blood HSCT due to relapse, MRD, MC or poor graft function were included from 2016-2018. Donors were stimulated with filgrastim 5 μg/kg every 24 hours for 3 days prior to donation. Standard pre-transfusion testing included major/minor crossmatch and flow cytometry analysis for lymphocyte subset quantification. The blood draw was performed on day 4 with the aim to collect 450 mL of whole blood in a standard collection bag. For patients with major and bidirectional ABO mismatch who had not undergone group switch or had a mixed population in solid-phase typing, a reactive major crossmatch was considered an absolute contraindication, while a reactive minor crossmatch was not. WB-DLI was infused fresh on the day of collection without manipulation in an outpatient basis. For relapsed patients, concurrent chemotherapy was administered on a case-by-case basis. Disease evaluation after relapse was performed on day +30 after treatment, while chimerism analysis was performed on day +60 through standard methods. GVHD prophylaxis was performed with oral cyclosporine A or tacrolimus and gradually tapered in a 60-day window.Results: Fourteen patients received a single dose of WB-DLI, median age was 32 years (16-67), 50% were female. Most common diagnoses were acute lymphoblastic leukemia, (n=5) and acute myeloid leukemia (n=5), followed by myelodysplastic syndrome (n=2), non-Hodgkin lymphoma (n=1) and chronic myeloid leukemia (n=1). Nine patients received Haplo grafts, while 5 were HLA-matched. Three patients had previous GVHD inactive at the time of WB-DLI. Indication for WB-DLI was relapse for 6 patients, 3 had MRD while 5 remained in remission; one patient received WB-DLI due to poor graft function. Median chimerism at the time of WB-DLI was 72% (range 38-100%). A single infusion was administered in all cases. Median mononuclear cell count obtained was 32x10x6/kg (range, 9-74), while median CD3+ cell count was 6.7x106/Kg (range 5.2-19). No immediate, severe adverse effects were observed. Febrile non-hemolytic transfusion reaction was documented in 4 cases after completing the procedure. No complications were observed in n=4 ABO mismatch cases including (n=2) major and bidirectional patients. Overall 50% of patients responded. Chimerism was improved in 50%, with a median increase of 28% (range 9-53%). Median post-DLI chimerism was 92.5% (range 20-100). Regarding patients treated for relapse or MRD, 3/9 patients responded (33%). Overall 8/14 patients eventually relapsed (57%) with a 12-month progression free survival of 27.5%, lower in relapsed/MRD positive patients (Figure I) (log rank test p=0.02 and 0.04, respectively). Overall survival at 12 months was 61.2%. Three patients developed acute GVHD, two of them grade III/IV in a median of 13 days (range 7-38). Four patients died due to relapse. Median follow-up after WB-DLI was 5 months (range 0.6-20.1 months). The cost of obtaining and administering WB-DLI was $350 USD per dose.Conclusions: DLI obtained from unmanipulated, GCSF-primed whole-blood is a safe and affordable adoptive cell therapy. It is effective and improves mixed chimerism after HSCT, while it was less effective for overt relapse. DisclosuresGomez-Almaguer:AbbVie: Consultancy; Novartis: Consultancy.

T-Replete Haploidentical Cell Transplantation Using Post-Transplant Cyclophosphamide for Acute Myeloid Leukemia, Acute Lymphoblastic Leukemia and Myelodysplastic Syndrome: Effect of Transplant Conditioning Regimen Intensity on Outcomes

▪The significance of conditioning regimen intensity on the outcomes of T-cell replete HLA-haploidentical transplants is not known. This study compared outcomes of commonly used myeloablative (MAC) to reduced intensity (RIC) conditioning regimens in 1325 such transplants (AML; n=818; ALL; n=286 and MDS; n=221) in the US between 2008 and 2016. The median age of the study population was 54 years (18 - 70). Most patients (80%) with AML and ALL were in first or subsequent remission; 83% of those with MDS had refractory anemia with excess blasts at transplantation. Fifty-one percent of patients with AML and ALL had intermediate disease risk index (DRI). In contrast, 50% of patients with MDS had high or very high DRI. Patients received MAC (n=526; 40%) or RIC (n=799; 60%) transplant conditioning regimens and a uniform graft-versus-host disease (GVHD) prophylaxis: post-transplant cyclophosphamide, calcineurin inhibitor and mycophenolate. Approximately 50% of patients reported a HCT-CI score of 0-2 in MAC and RIC groups. Sixty-six percent of MAC and 42% of RIC recipients received peripheral blood grafts. Total body irradiation (TBI) + fludarabine (TBI/Flu; 33%) and busulfan with cyclophosphamide with/without Flu (Bu/Cy ± Flu; 36%) were the predominant MAC regimens. Other MAC regimens included TBI/Cy or other agents (10%), Flu/Bu4 (13%), melphalan (140 mg/m2) + Flu ± thiotepa (Flu/Mel ± TT; 9%). TBI (200cGy)/Cy/Flu (84%) was the predominant RIC regimen. Other RIC regimens included TBI 200cGy + Bu or Mel + Flu (7%), Flu/Bu2 (1%) and Flu/Mel (100mg/m2) ± TT (5mg/kg) (8%). The primary endpoint was disease-free survival (relapse or death). Cox regression models were built to study the effect of conditioning regimens on transplant outcomes after adjusting for other factors significantly associated with outcomes. Differences in transplant-outcomes were observed between ages 18-54 years and 55-70 years. The effect of age was further tested within the 18-54 and 55-70 age groups and there were no differences in outcome. In patients aged 18-54 years (n=689), 55% received MAC and 54% received RIC regimens. In patients aged 55-70 years (n=636), 22% received MAC and 78% received RIC regimens. Table 1 shows the effect (hazard ratio; HR) of conditioning regimen intensity in the two age groups adjusted for HCT-CI, recipient CMV serostatus, disease, DRI and graft type and the 2-year probabilities for the outcomes of interest. In patients aged 18-54 years who were equally likely to receive MAC or RIC regimens, relapse risks were higher after RIC regimens that resulted in lower disease-free survival. There were no differences in non-relapse mortality (NRM) or overall survival by conditioning regimen intensity. In patients aged 55-70 years who were more likely to receive RIC regimen, NRM was lower after RIC but without an advantage for relapse, disease-free or overall survival. Figure 1A and 1B show the 2-year probability of disease-free survival by conditioning regimen intensity in patients aged 18-54 and 55-70 years, respectively. Consistent with the main analysis, a subset analysis limited to AML also confirmed higher relapse (HR 1.43, p=0.03) and lower disease-free survival (HR 1.38, p=0.02) after RIC regimens in patients aged 18-54 years but not in patients aged 55-70 years. Acute GVHD (HR 1.01, p=0.94) and chronic GVHD (HR 0.82, p=0.14) did not differ by conditioning regimen intensity. Table 2 compares the effect of TBI- and non-TBI containing MAC and RIC regimens adjusted for age, HCT-CI, recipient CMV serostatus, disease, DRI and graft type. NRM risks were higher after RIC non-TBI compared to RIC TBI regimens. The predominant RIC non-TBI regimen was Flu/Mel (100mg/m2) ± TT (5mg/kg). In conclusion, a MAC regimen offers higher disease-free survival for those aged 18-54 years and can tolerate MAC regimens. For patients who are unable to tolerate MAC regimens, regardless of their age, TBI200 cGy/Cy/Flu is preferred to Flu/Mel ± TT to minimize NRM risks. [Display omitted] DisclosuresShah:Juno Pharmaceuticals: Honoraria; Lentigen Technology: Research Funding; Exelexis: Equity Ownership; Geron: Equity Ownership; Miltenyi: Other: Travel funding, Research Funding; Oncosec: Equity Ownership. Brunstein:Gamidacell: Research Funding. Champlin:Otsuka: Research Funding; Sanofi: Research Funding. Hamadani:Celgene Corporation: Consultancy; Merck: Research Funding; Janssen: Consultancy; ADC Therapeutics: Research Funding; Sanofi Genzyme: Research Funding, Speakers Bureau; Cellerant: Consultancy; Takeda: Research Funding; Ostuka: Research Funding; MedImmune: Consultancy, Research Funding. McGuirk:Gamida Cell: Research Funding; Kite Pharma: Honoraria, Other: travel accommodations, expenses, speaker ; Fresenius Biotech: Research Funding; Pluristem Ltd: Research Funding; Bellicum Pharmaceuticals: Research Funding; Astellas Pharma: Research Funding; Novartis Pharmaceuticals Corporation: Honoraria, Other: speaker, Research Funding. Vasu:Boehringer Ingelheim Inc: Membership on an entity's Board of Directors or advisory committees. Waller:Pharmacyclics: Other: Travel Expenses, EHA, Research Funding; Cambium Medical Technologies: Consultancy, Equity Ownership; Celldex: Research Funding; Novartis Pharmaceuticals Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Kalytera: Consultancy.

Extracorporeal Photopheresis for First Line Treatment of Acute Graft Versus Host Disease

Background. Extracorporeal photopheresis (ECP) is increasingly used in the treatment of acute and chronic graft versus host disease (aGVHD). It is recognized as an immunomodulatory therapy that can decrease inflammation and allow for immune tolerance. Interesting results have obtained using ECP as salvage therapy in patients with steroid-refractory GVHD after failure of one or more lines of immunosuppressive therapy. With this background, we hypothesized that incorporation of ECP as part of first line treatment for aGVHD may improve response rate and avoid the development of steroid-refractory or steroid-dependent aGVHD, mainly in cases of aGVHD with predominant skin involvement.Patients and methods. Forty consecutive patients who received ECP as part of their first line treatment of aGVHD with skin involvement between 2013 and 2016 were included in this single center analysis. 36 patients (90%) received granulocyte-colony stimulating factor mobilized PBSC as grafts, 3 patients (7.5%) received a bone marrow graft and one patient (2.5%) received an umbilical cord blood graft. Donor was a matched related one for 10 patients, haploidentical for 8 patients, matched unrelated for 16 patients and mismatched unrelated for 6 patients. 4 patients received a reduced intensity conditioning regimen and 36 a myeloablative reduced toxicity conditioning regimen. All patients received a combination of cyclosporine A (CsA) and mycophenolate mofetil for GVHD prophylaxis, except for patients with a matched related donor who received cyclosporine A alone. Patients transplanted with a haploidentical donor, received post-transplant cyclophosphamide (PTCy; 50mg/kg/d at d+3/d+5). ECP was initiated as soon as possible after the diagnosis of aGVHD with predominant skin involvement. The primary endpoint was to determine the best response achieved through the use of ECP for first line treatment of aGVHD.Results. Median age was 57 (range, 22-66) years, with 28 male patients (49%). Diagnoses were myeloid (n=32) or lymphoid malignancies (n=8) and disease status at transplant was complete remission in 23 patients, partial remission in 8 patients, relapse/progression in 8 patients and untreated in 2 patients. Patients developed aGVHD at a median of 31 days after alloHCT (range, 11-129). 5 patients had late onset aGVHD and two patients developed aGVHD after DLI. None of them had an overlap syndrome. aGVHD grade was I in 13 patients, II in 18 patients and III-IV in 9 patients. All patients had skin involvement and 10 patients had gut (n=8) and/or liver (n=5) involvement. All patients, but one, were still receiving CsA at time of aGVHD diagnosis. Systemic corticosteroids were initiated in 29 patients, while 11 patients received only topical dermo-corticoids. ECP was initiated at a median of 9 days (range, 0-83) after the diagnosis of aGVHD. ECP was performed weekly for 26 patients and twice weekly for 14 patients. Overall response rate (ORR) was 80% after a median of 3 (range, 1-9) weeks of treatment, including 24 complete responses (CR, 60%), 6 very good partial responses (VGPR, 15%) and 2 partial responses (PR, 5%). According to aGVHD grade, in patients with grade I aGVHD, the ORR was 92%, including 11 CR and 1 VGPR. In patients with grade II aGVHD, the ORR was 83%, including 8 CR and 7 VGPR/PR. In patients with grade III-IV aGVHD, ORR was 55%, all of them being in CR. In responding patients, 1 patient presented a recurrence of aGVHD and 15 developed chronic GVHD, including 4 overlap syndromes. Median follow-up was 32 months (range, 22-58) among surviving patients. At 24 months after transplant, the OS and PFS rates were respectively 78% (95% CI, 61%-88%) and 56% (range, 39%-70%). The 2-years cumulative incidence of relapse was 33% (95% CI, 19%-49%) and the 2-years cumulative incidence of non-relapse mortality was 10% (95% CI, 3%-22%).Conclusion. ECP is an effective and well-tolerated option for treatment of acute GVHD with predominant skin involvement with a 2 years cumulative incidence of NRM of 10% and an ORR of 80%. Best responses were seen in patients with grade I-II aGVHD, suggesting that ECP should be started as early as possible after the diagnosis of aGVHD. Incorporation of ECP as part of first line treatment for aGVHD may improve response rate and avoid the development of steroid-refractory aGVHD. Prospective randomized trial are warranted to evaluate ECP as adjuvant treatment for first line treatment of skin aGVHD. DisclosuresMohty:Celgene: Consultancy, Honoraria; Takeda: Honoraria, Speakers Bureau; Janssen: Honoraria, Research Funding, Speakers Bureau; Jazz Pharmaceuticals: Honoraria, Research Funding, Speakers Bureau; MaaT Pharma: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bristol Myers: Consultancy, Research Funding; Sanofi: Consultancy, Honoraria, Research Funding, Speakers Bureau; Molmed: Consultancy; Servier: Consultancy; Amgen: Consultancy, Honoraria.

A Case Report of Refractory Immune Thrombocytopenia (ITP) Following Reduced Intensity Conditioning (RIC) Hematopoietic Cell Transplantation (HCT) for Myelodysplastic Syndrome (MDS) Successfully Treated with Off-Label Use of Daratumumab

Background:A source of treatment refractoriness in immune cytopenias appears to be residual CD138/38 positive lymphocyte populations (Audia S et al, Blood 118:4394-400,2011; Mahevas M et al, J Clin Invest 123:432-442, 2013). Persistence of recipient's plasma cells can lead to prolonged refractory thrombocytopenia following RIC-HCT. (Fasano RM et al, Br J Haematol 166(3):425-34, 2014). Daratumumab was effective in the treatment of a child with refractory autoimmune hemolytic anemia after HCT (Tolbert et al, Blood 128:4819, 2016).Case Report:The patient is a 60-year-old man with intermediate-high risk MDS who underwent RIC-HCT with total lymphoid irradiation and antithymocyte globulin with peripheral blood graft from a fully matched unrelated male donor. The patient had mild thrombocytopenia prior to HCT consistent with MDS and had not received platelet transfusions. He had not received any prior therapy for MDS. Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine and mycophenolate mofetil. Cytomegalovirus (CMV) serologic testing for exposure was negative for the recipient and positive for the donor. Both the patient and the donor had evidence for prior exposure to Epstein-Barr virus (EBV). He achieved engraftment on day +12. His peripheral blood chimerism on day + 30 showed full donor origin (WB 98%,CD3 96%,CD15 95%, CD19 98%, CD56 95%) and has been maintained to date. Acute skin GVHD responded to corticosteroids. While on corticosteroid therapy, he developed an abrupt decline in platelet count from 156,000/mcl on day +152 to 9, 000/mcl on Day + 166 without evidence for recurrent or active GVHD. While this was initially attributed to simultaneous EBV and CMV reactivations, severe thrombocytopenia persisted after viral clearance. An extensive work up for other etiologies of thrombocytopenia was negative. Repeated bone marrow biopsies were normal, including adequate megakaryocytosis and no MDS recurrence. Platelet associated antibody testing and platelet antigen genotyping were not conclusive for autoimmune versus alloimmune etiology. Testing for platelet HLA antibodies showed calculated Panel Reactive Antibody of 31% and unsatisfactory corrected count increment after transfusion of HLA compatible platelets units. The patient experienced prolonged severe thrombocytopenia for over 26 weeks with platelet count less than 5000/mcl for 22 weeks and only above 10,000 /mcl on 6 occasions. Potentially responsible medications were discontinued serially, but testing for drug inducted ITP was not conducted. Therapy included high dose corticosteroids, high dose immune globulin, rituximab, plasma exchange, splenectomy, romiplostim 10 mcg/kg/week, eltrombopag 100 mg to 150 mg daily for over 24 weeks, and low dose danazol. Fostamatinib was not available. Prednisone dose was tapered over many weeks to 5 mg daily. The patient experienced recurrent life-threatening and vision-threatening bleeding. Cumulative transfusions following Day + 166 were 133 single donor platelet units and 42 red cell units. All products were from CMV negative donors. Eltrombopag and danazol were deemed ineffective and tapered to discontinuation. CD38 positive cells were present in spleen and marrow by immunohistochemistry. Daratumumab 1300 mg was infused weekly x 4. Four weeks after the last dose of daratumumab, his platelet count increased to 91,000/mcl. Platelet count normalized to 150,000/mcl in week 5 or HCT Day + 383. Hypogammaglobulinemia has been the only detectable toxicity. Testing to determine autoimmune versus alloimmune origin is ongoing.Conclusion: Clinical trials of daratumumab for the treatment of severe refractory ITP are indicated. DisclosuresNo relevant conflicts of interest to declare.