Kerstin Schaefer-Eckart

Therapeutic platelet transfusion versus routine prophylactic transfusion in patients with haematological malignancies: an open-label, multicentre, randomised study

BACKGROUND Routine prophylactic platelet transfusion is the standard of care for patients with severe thrombocytopenia. We assessed the effect of a new strategy of therapeutic platelet transfusion on the number of transfusions and safety in patients with hypoproliferative thrombocytopenia. METHODS We did a multicentre, open-label, randomised parallel-group trial at eight haematology centres in Germany. Patients aged 16-80 years, who were undergoing intensive chemotherapy for acute myeloid leukaemia or autologous haemopoietic stem-cell transplantation for haematological cancers, were randomly assigned via a computer-generated randomisation sequence to receive either platelet transfusion when bleeding occurred (therapeutic strategy) or when morning platelet counts were 10×10(9) per L or lower (prophylactic strategy). Investigators undertaking interventions were not masked to group assignment. The primary endpoint was the number of platelet transfusions. Analysis was by intention to treat. This trial is registered, NCT00521664. FINDINGS 197 patients were assigned the prophylactic strategy and 199 the therapeutic strategy. Of 391 patients analysed, the therapeutic strategy reduced the mean number of platelet transfusions by 33·5% (95% CI 22·2-43·1; p<0·0001) in all patients (2·44 [2·22-2·67] in prophylactic group vs 1·63 [1·42-1·83] in therapeutic group), 31·6% (18·6-42·6; p<0·0001) in those with acute myeloid leukaemia (2·68 [2·35-3·01] vs 1·83 [1·58-2·10]), and 34·2% (6·6-53·7; p=0·0193) in those who had had autologous transplantation (1·80 [1·45-2·15] vs 1·18 [0·82-1·55]. We noted no increased risk of major haemorrhage in patients who had undergone autologous transplantation. In those with acute myeloid leukaemia, risk of non-fatal grade 4 (mostly CNS) bleeding was increased. We recorded 15 cases of non-fatal haemorrhage: four retinal in each transfusion group, and one vaginal and six cerebral in the therapeutic group. 12 patients died in the study: two from fatal cerebral haemorrhages in the therapeutic group, and ten (five in each treatment group) unrelated to major bleeding. INTERPRETATION The therapeutic strategy could become a new standard of care after autologous stem-cell transplantation; however, prophylactic platelet transfusion should remain the standard for patients with acute myeloid leukaemia. The new strategy should be used by some haematology centres only if the staff are well educated and experienced in the new approach and can react in a timely way to first signs of CNS bleeding. FUNDING Deutsche Krebshilfe eV (German Cancer Aid).

Reduced intensity conditioning for allogeneic hematopoietic stem cell transplantation in patients with acute myeloid leukemia: disease status by marrow blasts is the strongest prognostic factor.

Summary:We analyzed predictive factors for the outcome of 113 acute myeloid leukemia patients receiving reduced-intensity conditioning prior to allogeneic hematopoietic stem cell transplantation (HSCT). Patients were ineligible for conventional-intensity HSCT. Conditioning consisted of fludarabine and 50% of the conventional dose of busulfan (n=93) or total body irradiation (n=20). The source of stem cells was blood in 102 patients, marrow in 10, and both in one. In total, 50 (44.2%) donors were HLA-matched siblings, 50 (44.2%) unrelated fully matched and 13 (11.5%) partially mismatched family (n=1) or unrelated (n=12) donors. In all, 107 (94.6%) patients showed neutrophil and platelet engraftment after a median time of 13.5 and 13 days. The probabilities of event-free survival (EFS) (median follow-up: 12 months) were 49% for patients with less than 5% blasts in the marrow, 24% for patients with 5–20% blasts (P=0.002) and 14% with >20% blasts (P⩽0.001). Death occurred because of relapse in 29 patients (25.6%), infection in 12 patients (10.6%), acute graft-versus-host disease in eight patients (7.0%) and organ toxicity in nine patients (7.9%). In multivariate analysis, higher number of blasts in the marrow, alternative donors and low Karnofsky performance score were independent adverse prognostic factors for EFS.

A therapeutic platelet transfusion strategy is safe and feasible in patients after autologous peripheral blood stem cell transplantation

Prophylactic platelet transfusions are considered as standard in most hematology centers, but there is a long-standing controversy as to whether standard prophylactic platelet transfusions are necessary or whether this strategy could be replaced by a therapeutic transfusion strategy. In 106 consecutive cases of patients receiving 140 autologous peripheral blood stem cell transplantations, we used a therapeutic platelet transfusion protocol when patients were in a clinically stable condition. Platelet transfusions were only used when relevant bleeding occurred (more than petechial). Median duration of thrombocytopenia <20 × 109/l and <10 × 109/l was 6 and 3 days, which resulted in a total of 989 and 508 days, respectively. In only 26 out of 140 transplants (19%), we observed clinically relevant bleeding of minor or moderate severity. No severe or life-threatening bleeding was registered. The median and mean number of single donor platelet transfusions was one per transplant (range 0–18). One-third of all transplants, and 47% after high-dose melphalan could be performed without any platelet transfusion. Compared with a historical control group, we could reduce the number of platelet transfusions by one half. This therapeutic platelet transfusion strategy can be performed safely resulting in a considerable reduction in prophylactic platelet transfusions.

Allogeneic hematopoietic SCT in patients with AML following treosulfan/fludarabine conditioning

An alternative reduced-toxicity conditioning regimen for allogeneic transplantation, based on treosulfan and fludarabine, has recently been identified. The safety and efficacy of this new conditioning regimen has been investigated prospectively in patients with AML. A total number of 75 patients with AML in CR were treated with 3 × 14 g/m2 treosulfan and 5 × 30 mg/m2 fludarabine, followed by matched sibling or unrelated SCT. Patients were evaluated for engraftment, adverse events, GVHD, and for non-relapse mortality, relapse incidence, overall and disease-free survival (DFS). All patients showed primary engraftment of neutrophils after a median of 20 days. Non-hematological adverse events grade III–IV in severity included mainly infections (59%) and gastrointestinal symptoms (7%). Acute GVHD grade II–IV occurred in 21% and extensive chronic GVHD occurred in 16% of the patients. After a median follow-up of 715 days, the 2-year overall and DFS estimates were 61% and 55%, respectively. The 2-year incidences of relapse and non-relapse mortality reached 34% and 11%, respectively. In summary, our data confirm promising safety and efficacy of the treosulfan-based conditioning therapy in AML patients, ClinicalTrials.gov Identifier: NCT01063660.

Matching for the MICA-129 polymorphism is beneficial in unrelated hematopoietic stem cell transplantation

Major histocompatibility complex class I polypeptide-related sequence A (MICA) is a highly polymorphic ligand of the activating NKG2D receptor on natural killer (NK) cells, γδ-T cells, and NKT cells. MICA incompatibilities have been associated with an increased graft-versus-host disease (GVHD) incidence, and the MICA-129 (met/val) dimorphism has been shown to influence NKG2D signaling in unrelated hematopoietic stem cell transplantation (uHSCT). We investigated the effect of MICA matching on survival after uHSCT. We sequenced 2172 patients and their respective donors for MICA. All patients and donors were high-resolution HLA-typed and matched for 10/10 (n = 1379), 9/10 (n = 636), or 8/10 (n = 157) HLA alleles. Within each HLA match group, cases matched and mismatched for MICA and MICA-129 were analyzed for the end points overall survival (OS), disease-free survival (DFS), nonrelapse mortality (NRM), relapse-incidence (RI), and GVHD. Mismatches at the MICA locus as well as MICA-129 increased with the number of HLA mismatches (MICA mismatched 10/10, 9.2% [n = 127]; 9/10, 22.3% [n = 142]; 8/10, 38.2% [n = 60]; MICA-129 mismatched 10/10, 3.9% [n = 54]; 9/10, 10.2% [n = 65]; 8/10, 17.2% [n = 27]). Adverse OS was observed in the 10/10 match group if MICA-129 was mismatched (10/10, hazard ratio [HR], 1.77; confidence interval [CI], 1.22-2.57; P = .003). MICA-129 mismatches correlated with a significantly worse outcome for DFS in the 10/10 HLA match group (HR, 1.77; CI, 1.26-2.50; P = .001). Higher rates of aGVHD were seen in MICA-129 mismatched cases. Our results indicate that MICA-129 matching is relevant in uHSCT. Prospective typing of patients and donors in unrelated donor search may identify mismatches for MICA-129, and compatible donor selection may improve outcome for this small but high-risk subgroup.

Clinical-scale single-step CD4(+) and CD8(+) cell depletion for donor innate lymphocyte infusion (DILI).

The ability to selectively deplete or enrich cells of specific phenotype by immunomagnetic selection to reduce the risk of GVHD holds significant promise for application in adoptive immunotherapy. Current clinical-scale approaches for T-cell depletion (e.g., CD34+ selection, CD3+ depletion), usually deplete γδ T cells, which may be advantageous in mediating graft-versus-tumor (GVT) effects and augmenting the innate immune response against infections. Here, we present a new method for depletion of T cells with potential GVHD reactivity by using a single-step immunomagnetic protocol, which efficiently depletes CD4+ and CD8+ αβ T cells under good manufacturing practice (GMP) conditions. Depletion from unstimulated leukapheresis products (n=6) containing up to 2.0 × 1010 cells showed high efficiency (mean log depletion of CD4+ cells: 4.12, CD8+ cells: 3.77). In addition, immunomagnetic CD4/CD8 depletion resulted in passive enrichment of innate lymphocytes (mean recovery of natural killer (NK) cells: 38%, γδ T cells: 50%). We demonstrated that γδ/NK cells preserved their proliferative and cytotoxic capacity and conclude that simultaneous large-scale depletion of CD4+/CD8+ T cells is feasible and can be performed under GMP conditions with high-depletion efficacy for αβ T cells and recovery of functionally intact innate effector lymphocytes for potential use in adoptive immunotherapy studies.

Sex and Body Mass Index but Not CXCL12 801 G/A Polymorphism Determine the Efficacy of Hematopoietic Cell Mobilization: A Study in Healthy Volunteer Donors

Analyses of healthy donors of granulocyte colony-stimulating factor (G-CSF) mobilized hematopoietic stem and progenitor cells (HSPCs) and of patients undergoing autologous stem cell transplantation have suggested that individuals harboring the CXCL12-A allele mobilize a higher number of CD34 + HSPCs after G-CSF administration. We typed 463 healthy unrelated donors (376 men and 87 women) who had received daily subcutaneous injections at a mean dose of 7.36 ± 1.71 μg/kg G-CSF for 5 days for CXCL12 801 G/A using a real-time PCR assay. Interestingly, the median concentration of mobilized CD34 + cells on day 5 was almost identical in donors with the A-allele (79/μL; range, 11 to 249/μL) and the G/G-group (82/μL; range, 15 to 268/μL). In addition, the allelic distribution was not different in donors (n = 11) who mobilized less than 20/μL CD34 + cells. No difference in the overall yield of CD34 + cells in the apheresis product and in the number of CD34 + cells/kg recipient could be detected between both groups. In a multivariate regression model for the endpoint CD34 + cells/μL at day 5, only male sex (regression coefficient, 11.5; 95% confidence interval, 1.7 to 21.2, P = .021) and body mass index as continuous variables (regression coefficient, 3.5; 95% confidence interval, 2.5 to 4.5, P = .0001) but not age, smoking status, or CXCL12 allelic status represented independent variables. Our data derived from a large well-controlled cohort contradict previous analyses suggesting an association between CXCL12 allelic status and the yield of CD34 + HSPC after G-CSF mobilization. Concentration of CD34 + cells in the peripheral blood, the most objective parameter, could not be predicted by CXCL12 genotype.