Tree nut allergies: Allergen homology, cross-reactivity, and implications for therapy.

Tree nut allergy is a potentially life-threatening disease that is increasing in prevalence, now affecting 1% of the general population in the United States. While other food allergies often resolve spontaneously, tree nut allergies are outgrown in less than 10% of cases. Due to the likelihood of cross-sensitization to multiple tree nut allergens, the current treatment guideline is strict avoidance of all nuts once one tree nut allergy has been diagnosed. For example, walnut and pecan are highly cross-reactive, along with cashew and pistachio, but the extent of clinical, IgE-mediated cross-reactivity among other tree nuts remains unclear, therefore making avoidance of all tree nuts a safe approach. There have been recent advances in immunotherapy for food allergies. For instance, there are investigational immunotherapies for milk, egg and peanut allergies, specifically oral immunotherapy, sublingual immunotherapy and epicutaneous immunotherapy. However, there are no large randomized controlled clinical trials for tree nut allergies. Even though there has been less research into tree nut allergy immunotherapies, the evidence of T-cell cross-reactivity among tree nuts exists in animal models and in T cells from allergic patients indicates that immunotherapeutic interventions may be possible. Here, we review the literature regarding epidemiology, allergen homology and cross-reactivity among tree nuts, and explore how current findings can be employed for effective therapy.

Utility of component analyses in subjects undergoing sublingual immunotherapy for peanut allergy.

BACKGROUND: Sublingual immunotherapy (SLIT) with peanut changes clinical and immune responses in most peanut-allergic individuals, but the response is highly variable. OBJECTIVE: We sought to examine the component-specific effects of peanut SLIT and determine whether peanut component testing could predict the outcome of a double-blind, placebo-controlled food challenge (DBPCFC) after 12 months of peanut SLIT. METHODS: We included 33 subjects who underwent peanut SLIT with a DBPCFC of 2500 mg of peanut protein performed after 12 months of therapy. Plasma samples from baseline and after 12 months of peanut SLIT were assayed using ImmunoCAP for IgE and IgG4 against whole peanut, Ara h 1, Ara h 2, Ara h 3, Ara h 8, and Ara h 9. RESULTS: Following 12 months of SLIT, 10 subjects (30%) passed the DBPCFC without symptoms and were considered desensitized. Subjects that failed the DBPCFC tolerated a median of 460 mg peanut protein (range: 10-1710 mg). The desensitized group had significantly lower baseline levels of IgE against peanut (median 40.8 vs. 231 kUA /L, P = 0.0082), Ara h 2 (median 17 vs. 113 kUA /L, P = 0.0082), and Ara h 3 (median 0.3 vs. 8.5 kUA /L, P = 0.0396). ROC curves indicated that baseline IgE against peanut and Ara h 2 were equally effective at discriminating between the two groups (AUC = 0.7957, P = 0.007752 for both). CONCLUSION AND CLINICAL RELEVANCE: In this cohort of subjects undergoing SLIT for peanut allergy, lower baseline levels of IgE against Ara h 2, Ara h 3, and peanut were associated with successful desensitization.

Component-resolved analysis of IgA, IgE, and IgG4 during egg OIT identifies markers associated with sustained unresponsiveness.

BACKGROUND: In a previously reported CoFAR study, 55 subjects with egg allergy underwent randomized, placebo-controlled egg oral immunotherapy (eOIT). Active treatment induced desensitization in most and sustained unresponsiveness (SU) in a smaller subset. We hypothesized that component-resolved analysis of IgE, IgG4, IgA, IgA1, and IgA2 may identify potential biomarkers of SU in OIT subjects. METHODS: Longitudinal samples for 51 egg-allergic subjects (37 active and 14 placebo) were available. Egg white (EW)-, ovalbumin (OVA)-, and ovomucoid (OVM)-specific levels of IgA, IgA1, and IgA2 were quantified by ELISA. IgE and IgG4 to these antigens were quantified using ImmunoCAP® . Clinical responders achieved SU to egg; all others were considered nonresponders. Between-group comparisons were made among active and placebo, as well as responders and nonresponders. RESULTS: No placebo subjects achieved responder status. Through month 48, among the 37 active subjects, baseline IgE-OVM was lower in responders (median 3.97 kU/l, n = 19) than in nonresponders (10.9 kU/l, n = 18, P = 0.010). Logistic regression analysis revealed that lower baseline IgE-EW (P = 0.038), IgE-OVM (P = 0.032), and a higher IgG4/IgE-OVM ratio (P = 0.013) were associated with clinical response. Relative increases in IgG4-EW, IgA-EW, and IgA2-EW were observed in responders (P = 0.024, 0.024, and 0.029, respectively). IgG4/IgE, IgA/IgE, and IgA2/IgE ratios for EW and IgA/IgE ratio for OVA were found to be significantly elevated among responders (P = 0.004, 0.009, 0.028, and 0.008, respectively). CONCLUSIONS: Increased IgG4-EW, IgA-EW, and IgA2-EW during eOIT are associated with clinical response to eOIT. Lower pretreatment IgE-EW and IgE-OVM are also associated with SU. Future studies are needed to evaluate and validate these potential biomarkers.

Effects of a pre-existing food allergy on the oral introduction of food proteins: findings from a murine model.

Cashew-allergic mice develop elevated walnut-specific IgE upon oral feeding of walnut proteins. Ingestion of tree nuts in the presence of a known nut allergy could lead to additional sensitizations and anaphylaxis following subsequent exposure.

The 2S albumin allergens of Arachis hypogaea, Ara h 2 and Ara h 6, are the major elicitors of anaphylaxis and can effectively desensitize peanut-allergic mice.

BACKGROUND: Ara h 2 and Ara h 6, co-purified together in a 13-25 kD fraction (Ara h 2/6; 20 kD fraction) on gel filtration chromatography, account for the majority of effector activity in a crude peanut extract (CPE) when assayed with RBL SX-38 cells sensitized with IgE from human peanut allergic sera. OBJECTIVES: To determine if Ara h 2/6 are the primary peanut allergens responsible for allergic reactions in vivo and to determine if Ara h 2/6 would be sufficient to prevent allergic reactions to a complete CPE. METHODS: An oral sensitization mouse model of peanut allergy was used to assess the activity of Ara h 2/6 (20 kD) and CPE without the 20 kD fraction (CPE w/o 20 kD) for allergic provocation challenge and immunotherapy. The activity of these preparations was also tested in an assay of histamine release from human basophils in whole blood. RESULTS: Compared with mice challenged with control CPE, mice challenged with CPE w/o 20 kD experienced reduced symptoms (P < 0.05) and a smaller decrease in body temperature (P < 0.01). Results with the basophil histamine release assay corroborated these findings (P < 0.01). The mouse model was also used to administer Ara h 2/6 (20 kD) in an immunotherapy protocol, in which peanut-allergic mice treated with the 20 kD fraction experienced significantly reduced symptoms, changes in body temperature, and mast cell protease (MMCP-1) release compared with placebo (P < 0.01 for all parameters). Importantly, immunotherapy with the 20 kD fraction was just as effective as treatment with CPE, whereas CPE w/o 20 kD was significantly less effective for higher dose peanut challenges. CONCLUSIONS AND CLINICAL RELEVANCE: Ara h 2/6 are the most potent peanut allergens in vivo and can be used to desensitize peanut-allergic mice. These results have potential implications for clinical research in the areas of diagnosis and immunotherapy for peanut allergy.

Evidence of pathway-specific basophil anergy induced by peanut oral immunotherapy in peanut-allergic children.

BACKGROUND: In Westernized countries, over 1% of the population is allergic to peanuts or tree nuts, which carries a risk of severe allergic reactions. Several studies support the efficacy of peanut oral immunotherapy (OIT) for reducing the clinical sensitivity of affected individuals; however, the mechanisms of this effect are still being characterized. One mechanism that may contribute is the suppression of effector cells, such as basophils. Basophil anergy has been characterized in vitro as a pathway-specific hyporesponsiveness; however, this has not been demonstrated to occur in vivo. OBJECTIVE: To evaluate the hypothesis that basophil anergy occurs in vivo due to chronic allergen exposure in the setting of a clinical oral immunotherapy trial. METHODS: Samples of peripheral blood were obtained from subjects during a placebo-controlled clinical trial of peanut OIT. Basophil reactivity to in vitro stimulation with peanut allergen and controls was assessed by the upregulation of activation markers, CD63 and CD203c, measured by flow cytometry. RESULTS: The upregulation of CD63 following stimulation of the IgE receptor, either specifically with peanut allergen or non-specifically with anti-IgE antibody, was strongly suppressed by active OIT. However, OIT did not significantly suppress this response in basophils stimulated by the distinct fMLP receptor pathway. In the subset of subjects with egg sensitization, active peanut OIT also suppressed CD63 upregulation in response to stimulation with egg allergen. Allergen OIT also suppressed the upregulation of CD203c including in response to stimulation with IL-3 alone. CONCLUSION: Peanut OIT induces a hyporesponsive state in basophils that is consistent with pathway-specific anergy previously described in vitro. This suggests the hypothesis that effector cell anergy could contribute to clinical desensitization.

Prediction of nonrelapse mortality for patients surviving 100 days after allogeneic hematopoietic stem cell transplantation.

BACKGROUND: Patients who survive 100 days after allogeneic hematopoietic stem cell transplantation (alloHSCT) are at risk for chronic graft-versus-host disease and other potentially fatal complications. As the symptoms overlap and the differential diagnosis is difficult, the goal of this study was to verify whether basic laboratory evaluation performed on day +100 may allow identification of patients who are at high risk for nonrelapse mortality (NRM), independent of the underlying complications. PATIENTS AND METHODS: We analyzed 255 patients, mean age 29 years (range, 10-56 years), who remained alive and disease-free on day +100 after myeloablative alloHSCT from an HLA-identical sibling (n=177) or a matched unrelated volunteer (n=78), performed in a single institution between 1992 and 2003. RESULTS: Upon univariate analysis, the following laboratory parameters were associated with increased incidence of NRM: peripheral blood neutrophils<1.5x10(9)/L, platelets<100x10(9)/L, hemoglobin<11 g/dL, total protein<60 g/L, elevated plasma aspartate aminotransferase, elevated alkaline phosphatase, and elevated bilirubin. Upon multivariate analysis, only decreased protein (hazard ratio [HR]=6.97 [3.3-14.7], P<.0001) and elevated bilirubin (HR=3.52 [1.91-6.48], P<.0001) independently influenced the risk for NRM. The cumulative incidence of NRM equaled 6% if none of the above factors was present; 10% for hyperbilirubinemia alone; 22% for hypoproteinemia alone; and 70% for hyperbilirubinemia and hypoproteinemia, both present. CONCLUSIONS: A simple laboratory evaluation is highly predictive of the risk for NRM in patients surviving 100 days after alloHSCT. The prognosis is particularly poor for patients with hypoproteinemia and hyperbilirubinemia. These abnormalities may reflect impaired liver and intestine functions due to various posttransplantation complications.

Impact of activating killer immunoglobulin-like receptor genotype on outcome of unrelated donor-hematopoietic cell transplantation.

BACKGROUND: In a previous study we demonstrated that incompatibility regarding ligands for inhibitory killer immunoglobulin-like receptors (KIRs) is associated with a survival advantage following unrelated donor-hematopoietic cell transplantation (URD-HCT). The goal of the present analysis was to evaluate whether genotype of activating KIRs of the donor may have an impact on the outcome of URD-HCT. PATIENTS AND METHODS: Twenty-five URD-HCT recipients with hematological malignancies, mean age 27 years (range, 14-43 years), were included in the analysis. The conditioning regimen was myeloablative and based on chemotherapy alone (n = 20) or total body irradiation (n = 5). Graft-versus-host disease (GVHD) prophylaxis consisted of cyclosporine, methotrexate, and pretransplant antithymocyte globulin. Patients were grouped according to their donors' activating KIR genotype including two loci: KIR2DS1 and KIR2DS2. RESULTS: The presence of KIR2DS1 in the donor (n = 16/25) was not demonstrated to influence outcome. In contrast, the presence of KIR2DS2 (n = 13/25 donors) was associated with decreased probability of overall survival (0% vs 92%, P = .04) and disease-free survival (0% vs 92%, P = .046). The reason for failures in the KIR2DS2-positive group was chronic GVHD (n = 4), acute GVHD (n = 2), and relapse (n = 1). The cumulative incidence of nonrelapse mortality equaled 90% for the KIR2DS2-positive group and 8% for the KIR2DS2-negative group (P = .09). CONCLUSION: The presence of KIR2DS2 gene in the donor is associated with a high risk of mortality following URD-HCT, resulting mainly from the incidence of severe GVHD. Whether this effect is associated with the activity of natural killer cells or KIR-bearing T lymphocytes requires further investigation.

Addition of cladribine to induction/consolidation regimen does not impair peripheral blood stem cell mobilization and bone marrow harvest for autotransplantation in acute myeloid leukemia patients.

BACKGROUND: The previous study by the Polish Adult Leukemia Group has demonstrated that addition of cladribine to standard DNR+AraC induction potentiates the antileukemic activity. The goal of this study was to compare the efficacy of bone marrow or peripheral blood hematopoietic cell collection in patients who obtained remission after daunorubicine plus cytarabine induction with cladribine (DAC-7) or without addition of cladribine (DA-7) in preparation for autotransplantation. PATIENTS AND METHODS: Sixty-six patients aged 41 years (range, 17-58 years) were included in this study: 33 cases in the DAC-7 and 33 in the DA-7 arm. Hematopoietic cells were collected from the bone marrow (ABMT, n = 29) or from the peripheral blood (ABCT, n = 37) using cytopheresis after administration of AraC (2 x 2 g/m2) on days 1, 3, 5 and subsequent G-CSF (10 microg/kg) from day 7 as mobilization therapy. RESULTS: The numbers of harvested CD34+ cells were similar in the DAC-7 and DA-7 pretreated patients both after harvesting from peripheral blood (2.55 x 10(6)/kg vs 2.5 x 10(6)/kg) and from bone marrow (1.62 x 10(6)/kg vs 1.55 x 10(6)/kg), respectively. The proportion of patients with sufficient material for autologous bone marrow transplantation was higher in the DAC-7 compared with the DA-7 arm. All patients engrafted; hematopoietic recovery was similar in both subgroups. CONCLUSION: Addition of cladribine to a standard DA induction does not impair the harvesting of hematopoietic cells and their engraftment after autotransplantation.

A B-cell epigenetic signature defines three biologic subgroups of chronic lymphocytic leukemia with clinical impact.

Prospective identification of patients with chronic lymphocytic leukemia (CLL) destined to progress would greatly facilitate their clinical management. Recently, whole-genome DNA methylation analyses identified three clinicobiologic CLL subgroups with an epigenetic signature related to different normal B-cell counterparts. Here, we developed a clinically applicable method to identify these subgroups and to study their clinical relevance. Using a support vector machine approach, we built a prediction model using five epigenetic biomarkers that was able to classify CLL patients accurately into the three subgroups, namely naive B-cell-like, intermediate and memory B-cell-like CLL. DNA methylation was quantified by highly reproducible bisulfite pyrosequencing assays in two independent CLL series. In the initial series (n=211), the three subgroups showed differential levels of IGHV (immunoglobulin heavy-chain locus) mutation (P<0.001) and VH usage (P<0.03), as well as different clinical features and outcome in terms of time to first treatment (TTT) and overall survival (P<0.001). A multivariate Cox model showed that epigenetic classification was the strongest predictor of TTT (P<0.001) along with Binet stage (P<0.001). These findings were corroborated in a validation series (n=97). In this study, we developed a simple and robust method using epigenetic biomarkers to categorize CLLs into three subgroups with different clinicobiologic features and outcome.

Allogeneic and autologous bone marrow transplantation in single centre experience.

Among 290 BMT procedures: 74 AML, 78 ALL, 34 CML, 6 SAA, 3 MDS, 42 HD, 35 NHL, 11 MM, and 7 solid tumours (breast or testis cancer) Allogeneic BMT was performed in 76 patients and ABMT/APBCT in 214 patients. Survival, DFS and relapse curves were calculated using the Kaplan-Meier product limit method. Variables potentially affecting survival and DFS were assessed in a multivariate analysis by the Cox proportional hazard model in a stepwise regression. The promising results were obtained in high risk adult ALL in the first CR. DFS in CR1 patients transplanted after full dose induction and high dose consolidation was significantly longer if compared to those who received dose/time reduced or postponed treatment. For CR> or =2 patients and with CNS involvement at diagnosis ABMT offers a salvage therapy that needs further improvement. In relapsed and refractory HD better results are obtained in patients relapsing > 1 year after first CR and in patients with entirely nodal localisation of this relapse. In NHL bone marrow and spleen infiltration at diagnosis appear to be an unfavourable prognostic factor.

2-chlorodeoxyadenosine (2-CdA) in 2-hour versus 24-hour intravenous infusion in the treatment of patients with hairy cell leukemia.

Forty one patients with hairy cell leukemia (HCL) were treated with 2-chloro-deoxyadenosine (2-CdA) administered in various schedules. Complete remission (CR) was achieved in 31 (76%) patients and partial remission (PR) in 9 (22%). The mean duration of remission (CR + PR) was 25.2 months (range 9-45 months). One patient did not respond to therapy. Twelve out of 16 patients (75%) achieved CR after 5-day intravenous infusions of 2-CdA and 19 out of 25 patients (76%) after 7-day courses. In 19 out of 23 patients (82.6%) CR was achieved after intermittent 2-hour infusions and in 12 out of 18 (66.7%) after continuous 24-hour infusion. The differences were not statistically significant. Side effects of 2-CdA were similar in both groups except for infections, which were less frequently observed in the group treated for 5 days. The results of our study suggest that 2-CdA can be effectively administered to patients with HCL using 5-day courses and a 2-hour daily infusion.

Application of a combined rosette-fluorescent test in studies on leukemia cells.

In 13 healthy blood donors and 42 patients with leukemia and lymphosarcoma, the combined rosette-fluorescent test allowed simultaneous evaluation of peripheral blood lymphocytes and leukemia cells for presence of membrane immunoglobulins and sheep erythrocyte receptors. In healthy subjects, the mean value for T cells was 68 +/- 9%, for B cells 16 +/- 5%, for nonreceptor cells 15 +/- 8%, and for cells possessing both markers 0.6 +/-0.8%. In cases of acute leukemia, nonreceptor cells predominated (59 +/-24%), and values for T cells were usually low (19 +/- 13%). In chronic lymphocytic leukemias and lymphosarcoma, B and "null" cells usually predominated more or less. In all patients values for double-receptor cells were low. The combined test proved to be a convenient method for simultaneous evaluation of both markers. It is also useful for recognizing blastic leukemia cells devoid of both receptors and for determing values for T and B lymphocytes.

Parallel examination of receptors of malignant lymphoma cells using combined tests with mouse erythrocytes and monoclonal VIB-C5 antibody or polyvalent serum against human immunoglobulins.

This report presents the results of studies on two membrane markers of lymphocytes isolated from patients with non-Hodgkin's lymphoma. The combined methods were applied allowing for a simultaneous testing of: a) presence of the membrane receptor for mouse erythrocytes (Em) and membrane immunoglobulins (SmIg) or b) presence of the receptor for mouse erythrocytes and antigen binding monoclonal VIB-C5 antibody (VIB). In 12 patients with the low grade malignant lymphoma, up to 50% of lymphocytes possessed both receptor for mouse erythrocytes as well as membrane immunoglobulins and higher frequency of occurrence of Em+ cells were associated mainly with a high percentage of SmIg-- cells. In 9 patients with low grade malignant lymphoma, 24% of cells were observed to possess Em receptor and antigen detectable with VIB antibody, 34-70% of cells were Em-VIB-lymphocytes and 13-54% were Em-VIB+ cells. The results obtained have revealed that the application of monoclonal antibody renders possible the diagnosis of lymphoma and may be of particular value in the cases with weak expression of receptor for mouse erythrocytes.

Studies on receptor for mouse erythrocytes in lymphocytes of patients with malignant lymphoma.

In a group of 30 healthy subjects and 27 patients with the diagnosis of non-Hodgkins, lymphoma, the rosette test was performed with erythrocytes of mice of the C57 BL strain according to the method of Stathopoulos and Elliott. Lymphocytes were isolated from peripheral blood, and in four cases also from lymph nodes, on an Uropolinum-Polfa gradient. In healthy subjects, 12.6 +/- 6.9% of lymphocytes possessed receptors for mouse erythrocytes. The values in the group of patients were significantly higher (31.8 +/- 26.8%). Moreover, in the group of chronic lymphocytic leukemia, the values (68.0 +/- 20%) were significantly higher than in the group with lymphosarcoma (9.7 +/- 7.7%). Results of the rosette test with lymphocytes isolated from lymph nodes were the same as with blood lymphocytes of these patients: a high level of 68% was observed in a patient with chronic lymphocytic leukemia, and low levels (between 2% and 9%) in three patients with lymphosarcoma. It was concluded that the rosette test with erythrocytes from mice of the C57 BL strain detects low differentiated forms of B lymphocytes and could be useful in the diagnosis of chronic lymphocytic leukemia and differentiation from lymphosarcoma.

The monitoring of lymphocyte subpopulations in non Hodgkin lymphoma patients during chemotherapy.

Numerous examinations of the changes in B, T, null subpopulations were systematically carried out in 20 non Hodgkin lymphoma patients 8, 24, 48 hours after the therapy onset and subsequently at intervals of a few days during and after each chemotherapy cycle (COP, Knospe). T cells were evaluated by examining the receptor for neuraminidase pretreated sheep erythrocytes (nE), B lymphocytes by detecting surface immunoglobulins (SmIg) with normal and Fc deprived anti Fab2 globulin, moreover receptors for mouse erythrocytes (Em) and for C3 (EAC) were examined. A characteristic rise of the dominating subpopulation was observed during the first phase of the drug application which, most probably, was due to an expulsion of the pathological cells from involved lymphatic organs to the blood. On the basis of our clinical observations a conclusion has been drown, that this type of monitoring possibly renders a maximal effective treatment, enables the control of the accompanying immunosuppression and offers an additional criterion of the remission.

Application of series of monoclonal antibodies against human white cell differentiation antigens and the common ALL antigen in the characterization of leukemia cells.

The specificity and the clinical usefulness of the hybridoma derived monoclonal antibodies raised against the differentiation antigens of granulocytes (VIM-D5, VIM-C6), monocytes (VIM-D2), B lymphocytes (VIB-C5, VIC-Y1), erythrocytes (VIE-G4) and CALLA (VIL-A1) was studied in leukemic cells isolated from peripheral blood and bone marrow of 41 adults with acute leukemia by using indirect fluorescence method. The VIL-A1 positivity was observed in 4/9 of ALL and in none of myeloid leukemia. It was accompanied in 3/4 of cases by VIB-C5 positivity and in one case by VIE-G4 positivity. It is important that 2 out 3 unclassifiable cases (PAS-) could be diagnosed as common ALL due to their VIL-A1 positivity. VIM-D5 like VIM-C6 reacted specifially with granulocytic cells only and gave positive results in 20/30 of acute myeloid leukemias. When classified according to the FAB scheme, the proportion of VIM-D5 + patients rose from M1 toward more mature subtypes, including M4/5. It allowed to identify as AML 2/6 of unclassifiable-Mo leukemias, VIC-Y1 appeared to be helpful in characterization of B cell malignancies and of myelomonocytic leukemias. It is concluded that the monoclonal antibodies of VI series are specific and allow a more precise definition of leukemia, thus helping in optimalization of treatment.

Sonography in the diagnosis and therapy monitoring of lymphomas.

Multiple ultrasonographic (USG) evaluations of abdomen were performed in 97 patients: 30 with Hodgkin's disease (HD), 60 with non-Hodgkin lymphomas (NHL) and 7 with primary gastric lymphoma (PGL) before or during chemo- or X-ray therapy. In 33% of HD patients USG was normal, while in 63% splenomegaly, in 40% hepatomegaly and in 20% lymph node enlargement were observed. After therapy, in 57% USG was improved and in only 3% of patients worsening was observed. In NHL patients splenomegaly was observed in 70%, hepatomegaly in 60% and lymph node enlargement in 35%. During follow-up, in 49% of patients improvement and in 3% worsening was observed. In 5 patients with PGL no changes were observed, in further 2 patients in Stage IV stomach wall and infiltration of nearest lymph nodes was observed. USG evaluation of abdomen may be useful in staging and therapy monitoring of malignant lymphomas.

Allogeneic transplantation of selected peripheral CD34+ cells with controlled CD3+ cells add-back in high-risk patients.

We evaluated the feasibility of allogeneic transplantation of CliniMACS-selected peripheral CD34+ cells from siblings (four patients: AML-M4, M2, CLL, MDS); nonoptimal related donors (two patients: AML-M4, CML); and unrelated donors (two patients: CML, ALL, both without engraftment after preceding URDBMT). All patients had high-risk of aGVHD and/or graft failure due to multiple transplantation risk factors. Conditioning treatment was myeloablative (n=7) or nonmyeloablative (n=1). Immunosuppression consisted of CsA (n=8), Mtx (n=5), ATG (n=4). Selected CD34+ cells were transplanted (average 3.91 x 10(6)/kg, range 1.29 to 7.27 x 10(6)/kg) together with 0.01 to 0.5 x 10(7) CD3+ cells/kg to assure proper engraftment. The remaining CD34-negative fraction was cryopreserved for further CD3+ cell add-back. Average recovery and purity of CD34+ cells following CliniMACS selection were 74% and 97%. No severe complications were observed in the first 100 days. Regeneration times were satisfactory in seven of eight patients (87.5%) with ANO > 0.5 g/L and Plt > 50 g/L reached on average on days +26 and +32 (range 15 to 29 and 15 to 67), respectively. In three patients (37.5%) T-lymphocytes were added-back one to three times (due to low numbers of initially transfused CD3+ cells in two patients, in one patient with PRCA caused by ABO incompatibility). One to four additional transplantations of nonselected peripheral cells were performed on days +28 to +270 in consequence of infections (CMV-two patients; parvovirus-one patient), poor regeneration and residual disease (one patient) and prolonged transfusion dependency (one patient). Severe aGVHD grade III or IV developed in three patients (37.5%) following the nonselected cells transplantation. Finally, five patients (62.5%) are alive and in remission (median follow-up 815 days). We conclude that allogeneic transplantation of selected peripheral CD34+ cells (CliniMACS) with controlled add-back of CD3+ cells is an effective, well, tolerated procedure in high-risk patients.