ALBERT

Description: Background: Despite adequate treatment and favorable initial outcome, the risk of recurrence of a previous invasive fungal infection (IFI) or acquisition of a new one is a major obstacle to the success of stem cell transplantation (SCT). The rate of IFI relapse is estimated to be 30 to 50%. Secondary prophylaxis could increase survival and reduce the burden of fungal infections to these patients. Methods: A prospective open multicenter trial was performed to evaluate the efficacy of voriconazole (VORI) (4mg/kg/12h IV or 200mg/12h oral) as secondary prophylaxis in allogeneic SCT patients with previous proven or probable IFI (2002 EORTC/MSG consensus criteria). Adult patients (Age 〉 18 y) were eligible if the prior IFI had occurred within the 12 months prior to SCT. Exclusion criteria were prior resistance to VORI, prior history of zygomycosis, or active IFI at time of transplant. VORI was started within 3 days before transplant and planned for 100 days. Prophylaxis could be prolonged up to 150 days in case of persistent graft-versus-host disease (GVHD) and ongoing immunosuppression on day 100. All patients were followed until 12 months post-transplant, or until death, whichever occurred first. The primary endpoint was the rate of proven and probable IFI between the start of VORI prophylaxis and the 12 month follow-up. Diagnosis of prior IFI, diagnosis of IFI occurring during the study, and cause of deaths, were assessed by a Data Review Committee including an independent radiologist. Results: 45 patients from 17 EBMT centers were included from Feb 2005 to March 2007. The mean age was 48 y (22–72). Forty-one had acute leukemia, and 24 were in first complete remission. Previous IFI included proven (n= 6) or probable (n=26) aspergillosis, proven candidiasis (n=5), and others (n=8). The mean time from diagnosis of the previous IFI to SCT was 151 days. Twenty four patients were transplanted from a family donor (HLA-id: 18; mismatched: 5; twin: 1), and 21 from an unrelated donor. The graft was bone marrow (n=6), peripheral blood stem cells (n=38) or cord blood (n=1). The conditioning regimen was myeloablative in 27 and non myeloablative in 18 patients. The median follow-up was 360 days (range 5–469). Twenty-six (58%) patients experienced at least one episode of GVHD. Three cases of IFI were recorded after transplant: one C. albicans candidemia, one S prolificans fungemia and one zygomycosis, at day 3, 16, and 66 after transplant, respectively (incidence: 7%). Two of these 3 IFI were relapses of a previous IFI. The median duration of the VORI prophylaxis was 94 days (5–180). Eleven patients (24%) died between 96 and 326 days post-transplant (median: 136 days) from scedosporiosis (n=1), leukemia relapse (n=4), respiratory failure or pneumopathy of unknown origin (n=3), GVHD (2), or sepsis (n=1). Conclusion: Voriconazole appears to be an efficient drug for secondary prophylaxis of proven and probable IFI after allogeneic SCT, with few adverse events. Considering an expected rate of IFI occurrence or recurrence after transplant of around 30%, the observed rate of 7%, and only one death from IFI in this high-risk adult population are promising.

Description: Background: Pneumococcal infections are causes of death after SCT. The IDWP01 trial compared early (E) vs late (L) pneumococcal immunization after allogeneic SCT, starting either at 3, or at 9 months post-transplant with 3 doses of PCV7 given at 1 month intervals,. This study has shown that the early immunization was not inferior in the proportion of responders one month after the last PCV7 dose. In addition, all patients received one dose of PPV23 after the 3 doses of PCV7 at 12 months and 18 months in the E and L groups, respectively. The goal of this supplementary analysis was to look at the immune response to the PPV23 according to the time of its administration. Methods: 158 patients were randomized in the IDWP01 trial, 75 in the E, and 83 in the L group, of whom 109 received the PPV23 (56 in the E, and 53 in the L group, respectively) and 86 (44 in the E, and 42 in the L group) were followed up to 24 months after SCT. The immune response to PPV23 was assessed by the antibody (Ab) titres, measured by ELISA to pn1 and pn5, which are both included in PPV23 but not in PCV7. Pn1 and pn5 Ab titres were assessed before, and one month after the PPV23 dose, and at 24 months post-SCT. Two cut-off levels for protection were considered for each antigen: □ .15 μg/ml and □ .5 μg/ml. The geometric mean concentrations (GMC) were also analyzed. Additionally, all Ab titres to the 7 PCV7 antigens were measured in parallel. Results: Ab titres to pn1 and pn5 before the PPV23 were not significantly different in the E (at 12 mo post-SCT) and in the L (at 18 mo post-SCT) group. Whatever the cut-off used to evaluating the response, the percentage of responders one month after the PPV23 was not significantly different in both groups: For pn1: □ .15μg/ml: E: 80% vs L: 87%; □ .5μg/ml: E: 42% vs L: 60%; For pn5: □ .15μg/ml: E: 84% vs L: 94%; □ .5μg/ml: E: 62% vs L: 74%). The GMC were not significantly different in the E vs. the L group for pn1. The GMC were significantly higher in the L vs. the E group for pn5 (P=.02). At 24 months post-SCT, there was no significant difference in the GMC of pn1 and pn5 Ab titres but there was a tendency for higher titres in the L group for both antigens. Additionally, among 36 non-responders to the PCV7 at the time of PPV23 administration, 21 remained non-responders, but 15 (42%) responded to all the seven PCV7 antigens 1 month later. Conclusion: Although there was a tendency for a better response rate and for higher GMC for pn1 and pn5 according to the timing of immunization with PPV23 (12 vs. 18 months) in favour of the L group, these differences were not significant. Additionally, PPV23 increase the response rate to the serotypes included in the PCV7. Therefore, we recommend starting PCV7 vaccination early after SCT (3 PCV7 from 3 months). A dose of PPV23 after 3 doses of PCV7 broadens the immune response to pneumococci as well as increases the response rate to the serotypes included in the PCV7 and should therefore be considered at 12 months after SCT. The authors are grateful to the Safety committee: D Engelhard, P Reusser, and P Reinert

Description: Abstract 468 AML is worldwide the most frequent indication for allo-SCT. This is most likely due to the curative potential of the graft versus leukemia (GVL) effect associated with the procedure. Unfortunately, GVL and GVHD are intimately linked. Thus, it is important to identify markers predictive of severe GVHD, to balance the risk of this complication and the risk of relapse in a particular AML patient. The innate immune system, as the initial regulator of the inflammatory response, mediates an important role in these reactions. After conditioning regimen, toll-like receptors initiate the innate inflammatory response by activating intracellular signaling cascades that converge on the activation of NF-κB and interferon regulatory factor-3 (IRF3). IRF3 activation in bone marrow derived dendritic cells (DCs) results in natural killer (NK) cell activation inducing an anti-tumor NK response. We hypothesized that genetic variability in IRF3 in either the donor or the recipient could impact on the degree of the inflammatory response and on GVHD and GVL effect allo-SCT. We analysed the effect of two frequent single nucleotide polymorphisms (SNPs) in IRF3 (rs7251 and rs2304205), which are inherited in a haplotype, on the GVHD and GVL in 249 patients diagnosed with AML and submitted to HLA-identical sibling allo-SCT. Those patients with a donor carrying dominant GG gene variant in rs7251 (45% of the donors) had, as compared to GC (44%) and CC (11%) variants, lower aGVHD III-IV incidence (4% vs 11% vs 27%; p=0.0078) (Figure 1A), higher relapse incidence (49% vs 35% vs 26%; p=0.018) (Figure 1B), and lower TRM (7% vs 24% vs 18%; p=0.0065). This clinical impact on severe aGVHD, relapse, and TRM was retained at multivariate analysis. Further, GG gene variant in rs7251 when present in the patient was also associated with lower aGVHD III-IV incidence (4% vs 13% vs 24%; p=0.009), higher relapse incidence (50% vs 34% vs 31%; p=0.014), and with a trend for a lower TRM (9% vs 19% vs 23%; p=0.064). Patients carrying AA dominant gene variant in rs2304205 in IRF3 presented a higher relapse incidence than the rest of genotypes (50% vs 39% vs 18%, p=0.0068). However, this impact was not retained at multivariate analysis. Functional studies in 180 healthy individuals showed that after stimulation of peripheral blood with cytomegalovirus (CMV) peptides, GG gene variant in rs7251 presented lower IFN-γ serum production than the rest of individuals, and GG gene variant was associated with lower number of IFN-γ producing mature NK cells, lower number of cytotoxic NK cells against K562 cell line and lower proliferation of T cells after antigen presentation by DCs. In conclusion, we show that a particular gene variant in IRF3 in the donors is associated with a low incidence of severe aGVHD and high incidence of relapse in AML patients submitted to allo-SCT. This finding could be explained by its effect in the inflammatory and adaptive immune response, with lower IFN-g production, lower lymphocyte proliferation after antigen presentation by DCs and lower mature NK cell response. Thus, these results suggest that, if possible, when transplanting an AML patient with a low risk of relapse it might be preferable to select a donor harbouring GG in rs7251 in IRF3, and when transplanting an AML patient with a high risk of relapse after the transplant it might be preferable to consider select a donor GC or CC in rs7251 in IRF3. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 465 A previous study from the IDWP of the EBMT showed that CMV seropositive patients undergoing unrelated donor HSCT had increased NRM and decreased survival if they were grafted from a CMV seronegative donor. This finding has been controversial and we therefore decided to revisit the question with a larger number of patients and including new factors such as conditioning intensity. Patients were selected from the EBMT database who had received an allogeneic HSCT from 1992 – 2008 and for whom both the donor and recipient CMV status were known. Patients receiving cord blood grafts were excluded. 54660 patients were identified and included in the study; 32320 seropositive and 22340 seronegative patients. The different donor categories (sibling, mis-matched family, unrelated) were analyzed separately. Cox multivariate models were fitted to estimate the effect of donor serological status (positive versus negative) on outcome both in CMV seropositive and CMV seronegative patients adjusted for year of HSCT, donor and patient gender, recipient age, stem cell source, diagnosis, use of alemtuzumab or ATG, country, and conditioning intensity (RIC vs. MAC). Seronegative patients receiving grafts from seropositive unrelated donors had an decreased overall survival (OS; HR 1.13; p

Description: Abstract 1302 Introduction: Interleukin 17 (IL-17) is a pro-inflammatory cytokine secreted by CD4+ helper T (TH) cells, TH17 cells. IL-17F is the most recently discovered member of the IL-17 cytokine family. IL-17F induces several cell types to express cytokines, chemokines, growth factors, and adhesion molecules, which are crucial molecules in leukocyte recruitment and activation. On this basis, IL-17F has been associated with the pathogenesis of multiple autoimmune diseases. Allele G for the polymorphism A7488G seems to play a protective role for the development of such diseases. However, no data is available about the possible impact of this polymorphism on the development of complications after allogeneic stem cell transplantation (allo-SCT). Objective: To analyze the influence of the donor (Dn) and recipient (Rc) genotype for the polymorphism A7488G in the IL-17F gene on the outcome of HLA-identical related allo-SCT. Material and Methods: The study comprised 143 allo-SCT patients and their donors (286 samples) included in the Spanish Group for Hematopoietic Stem cell Transplantation (GETH) DNA bank. Genomic DNA was purified from peripheral blood samples obtained, after written informed consent, from patients pre-SCT and donors. The A7488G polymorphism of the IL-17F gene was analyzed by polymerase chain reaction-restriction fragment length polymorphisms (PCR-RFLP) following the procedure developed by Tahara et al. (J Clin Immunol 2008, 28:44-49). Results were analyzed using the Pearson's Chi-square Test. Results: In accordance with previous reports, allele G was present in a reduced number of individuals, 30/286 (10.5%) corresponding to 19/143 Allo-SCT (13.3%). The genotypes of the Dn were 129/143 (90.2%) AA and 14/143 (9.8%) AG, while those of the Rc were 127/143 (88.8%) AA and 16/143 (11.2%) AG. No Dn or Rc with GG genotype was found. Dn and Rc genotype combinations were as follows: DnAA-RcAA, 124/143 (86.7%); DnAG-RcAG, 11/143 (7.7%), DnAA-RcAG, 5/143 (3.5%), DnAG-RcAA, 3/143 (2.1%). The genotype of the donor influenced transplant outcome in terms of relapse (35/129 (27.1%) when the donor was AA vs. 1/14 (7.1%) when the donor was AG; p=0.119) and survival (exitus 64/129 (49.6%) when the donor was AA vs. 3/14 (21.4%) when the donor was AG; p=0,045). No association between Dn or Rc genotype and graft versus host disease (acute or chronic) or other complications was observed. The differences observed in relapse and survival according to the genotype of the donor, were especially true for AG genotype patients (no significant differences were observed in AA genotype patients). In fact, relapse (3/5 (60%) when the Dn was AA vs. 0/11 (0%) when the Dn was AG; p=0.004) and survival (exitus 5/5 (100%) when the Dn was AA vs. 1/11 (9,1%) when the Dn was AG; p

Description: Abstract 890 Background: One of the most extensively used approaches to prevent GVHD in the RIC setting is the combination of CsAMMF. Other strategies have been described such as SiTac, mostly as single center experiences, with promising results. Nevertheless, data from multicenter studies are lacking using the latter approach and, furthermore, no studies have been performed comparing both approaches. Aim: in the current study we describe the results of the prospective multicenter trial 2007–006416-32 by GEL-TAMO/GETH using SiTac as GVHD prophylaxis and compare this approach to the combination of CsAMMF in a sequential analysis. Material and Methods: from May 2002, 90 patients were included. All of them received an URD transplantation after RIC based on fludarabine (150 mg/m2) plus busulphan (10 mg/kg) or melphalan (140 mg/m2). 45 transplanted between 2002 a 2007 received CsAMMF as GVHD prophylaxis while the remaining 45 patients undergoing transplantation from 2008 received SiTac. 41% of the patients were in CR, 30% were in PR and 29% had active disease at the time of transplantation. No differences were observed in terms of disease status. Patients in the SiTac trial had a higher median age (49 versus 43 years, p=0.02) while a higher percentage of patients in the CsAMMF were diagnosed with acute leukemia (10 versus 19 patients, p=0.05). Supportive care was similar in both subgroups except for the use of azols as antifungal prophylaxis which was not allowed in the SiTac. Results: 12% of patients receiving SiTac developed mycroangiopathy which required to modify immunesuppresive treatment although it did not increase the mortality of the procedure. No VOD was reported. No significant differences were observed neither in terms of hematopoietic recovery nor in the cumulative incidence of grades 2–4 or 3–4 aGVHD (49% versus 50% grades 2–4 and 15% versus 26% grades 3–4 for patients receiving SiTac versus CsAMMF, respectively). By contrast 18% of patients receiving SiTac versus 55% of those receiving CsAMMF developed gut aGVHD ≥ grade 2, p=0.007. Cumulative incidence of cGVHD was 55% versus 88% for SiTac versus CsAMMF, respectively, p=0.0002 while the incidence of extensive cGVHD was 27% versus 52%, respectively, p=0.03. Non relapse mortality was 10% versus 20% at 100 days and 19% versus 40% at 1 year, for patients receiving Si-Tac versus CsAMMF, respectively (p=0.028). Event free and overall survival at 2 years were 59% versus 35%, p=0.008 and 72 versus 48%, p=0.018, for patients receiving Si-Tac versus CsAMMF, respectively. Conclusions: the current study confirms in a multicenter trial the promising results of the combination Sirolimus plus Tacrolimus among patients undergoing RIC URD transplantation. The current is the first study comparing in a sequential study SiTac versus CsAMMF and confirms that the prior decreases the risk of chronic GVHD and the non-relapse mortality of the procedure which translates into a better event free and overall survival. Disclosures: No relevant conflicts of interest to declare.

Description: Introduction. Graft versus host disease (GvHD) is the main cause of morbimortality after allogeneic stem cell transplantation (allo-SCT). Several single-nucleotide polymorphisms (SNPs) in cytokine genes have shown to influence gene expression and are therefore associated with donor-recipient alloreactivity and, ultimately, with SCT outcome. Interleukin 1 (IL1) is a proinflammatory cytokine that induces the production of cytokines and chemokines and plays an important role in the inflammatory processes. IL1 is involved in various cellular functions, including cell proliferation, differentiation, and apoptosis. The IL1 family consists of two biologically actives forms (IL1A and IL1B). Several polymorphisms of these genes have been implicated in the pathogenesis of autoimmune diseases, however, their importance in SCT is not well-known. Objective To investigate the relationship between 4 SNPs in IL1A and IL1B genes and the susceptibility to the development of GvHD and other complications after HLA-identical allo-SCT. Patients and methods Genomic DNA obtained from peripheral blood samples belonging to 509 patients and their HLA-identical sibling donors (Table 1) included in the DNA Bank of the Spanish Group for Hematopoietic Stem Cell Transplantation (GETH). One SNP, rs1800587 (-889 C〉T), in the promoter region of the IL1A gene and three SNPs in the IL1B gene (two in the promoter region rs16944 (-511C〉T), rs1143627 (-31 T〉C) and one in exon 5 rs1143634 (+3954 C〉T) were genotyped by multiplex primer extension followed by mass spectrometry (MALDI-TOF; Sequenom MassArray). Univariate and multivariate regression analysis were performed using Cox regression in the presence of competing risks except for chronic GvHD for which we used logistic regression model. All variables with p≤0.10 according to univariate analysis were included in the multivariate analysis. For multivariate analyses, p values were two sided and the outcomes were considered to be significant for p

Description: Introduction Graft versus host disease (GVHD) is the main cause of morbi-mortality after allogeneic stem cell transplantation (allo-SCT). Despite considerable advances in our understanding of the pathophysiology, nowdays anticipation of GVHD is an unresolved matter. Several single-nucleotide polymorphisms (SNPs) in cytokine genes have shown to be associated with donor-recipient alloreactivity and, ultimately, with SCT outcome. In the present study, we propose a novel predictive model based on both clinical and genetic (SNP) variables applying an innovative estimation linear regression model, the least absolute shrinkage and selection operator (LASSO), in a large cohort of HLA-identical sibling donor allo-SCT. Patients and Methods The study evaluated 25 SNPs in 12 genes (Table 1) in genomic DNA obtained from PB samples from 273 patients with available acute GVHD (aGVHD) data and 213 patients with chronic GVHD (cGVHD) data included in the DNA Bank of the Spanish Group for Hematopoietic Stem Cell Transplantation (GETH) and their HLA-identical sibling donors. Each SNP was assessed for different models of transmission (recessive, dominant, co-dominant and additive), producing 25 SNPs x 4 models = 100 variables. Clinical variables known to influence the development of GVHD were also considered (Table 1). Univariate regression analysis was performed using Cox regression (data not shown). Multivariant analysis was made with LASSO, an innovative estimation method for linear regression models which is able to select a set of optimal predictors from a large set of potential predictor variables and was considered as a variables selection method under the estimation of a Logit regression model. In this model, the strength of the penalty term is controlled by a smoothing parameter (λ), which is chosen by maximizing the area under ROC curve (AUC) and the correct classification rate (CCR). The statistical model was fitted (goodness-of-fit assessment) by randomly selecting the 85% of the data (the so-called "training set"), and the predictive ability was computed with the remaining 15% (the so-called "testing set"). In order to evaluate the performance and the prediction ability of each model, training and testing samples were randomly selected a total of 100 times. The distribution of the CCR and the AUC over the 100 samplings, were shown by means of box plots and statistical summary in the results data. Finally, for prediction purposes, we considered a cut-off value according to the proportion of Y=1 in the sample (0.28 for grades II-IV aGVHD, 0.11 for grades III-IV aGVHD and 0.30 for extensive cGVHD). Results The best model to predict aGVHD II-IV included 11 genetic variables and no clinical variables with a CCR for patients who developed (CCR1) aGVHD II-IV of 63.6% (Figure 2). The best model to predict aGVHD III-IV included 20 genetic and 7 clinical variables with a CCR1 for aGVHD III-IV of 100%. The best model to predict extensive chronic GVHD included 10 genetic and 3 clinical variables with a CCR1 for extensive cGVHD of 80%. On the other hand, predictive models with only clinical variables showed a poorer CCR1 for patients who developed aGVHD II-IV, aGVHD III-IV and extensive cGVHD (55.6%, 50% and 66.7% respectively; Figure 1). Based on the results from LASSO multivariate analyses, a risk score was calculated for grades II-IV and III-IV aGVHD as well as for cGVHD and extensive cGVHD. Patients were categorized into two groups: low risk (below the cut-off value) and high risk (above the cut-off). Such risk model was able to stratify patients who develop grades II-IV aGVHD (p

Description: CTLA-4 is an inhibitory molecule that down-regulates T-cell activation. Although polymorphisms at CTLA-4 have been correlated with autoimmune diseases their association with clinical outcome after allogeneic hematopoietic stem cell transplantation (allo-HSCT) has yet to be explored. A total of 5 CTLA-4 single-nucleotide polymorphisms were genotyped on 536 HLA-identical sibling donors of allo-HSC transplants. Genotypes were tested for an association with patients' posttransplantation outcomes. The effect of the polymorphisms on cytotoxic T-lymphocyte antigen 4 (CTLA-4) mRNA and protein production were determined in 60 healthy control participants. We observed a reduction in the mRNA expression of the soluble CTLA-4 isoform in the presence of a G allele at CT60 and +49. Patients receiving stem cells from a donor with at least 1 G allele in position CT60 had worse overall survival (56.2% vs 69.8% at 5 years; P = .001; hazard ratio [HR], 3.80; 95% confidence interval [CI], 1.75-8.22), due to a higher risk of relapse (P = .049; HR, 1.71; 95% CI, 1.00-2.93). Acute graft-versus-host disease (aGVHD) was more frequent in patients receiving CT60 AA stem cells (P = .033; HR, 1.54; 95% CI, 1.03-2.29). This is the first study to report an association between polymorphisms at CTLA-4 and clinical outcome after allo-HSCT. The CT60 genotype influences relapse and aGVHD, probably due to its action on CTLA-4 alternative splicing.

Description: Background. Cytomegalovirus (CMV) is a widespread persistent β–herpesvirus, which can cause severe complications during primary infection or reactivation in immunocompromised patients, such as after allogeneic stem cell transplantation (alloSCT). Another major complication associated with alloSCT is graft-versus-host disease (GVHD). The pathogenesis of GVHD involves migration of the transplanted donor naïve T-cells into the secondary lymphoid organs in the recipient, which is mainly steered by CD62L and CCR7. As these homing molecules have been associated with both acute GVHD (aGVHD) and chronic GVHD (cGVHD), we studied whether the CMV serostatus of the donor affects the lymphoid composition of the graft product and whether this phenotype can predict CMV reactivation and GVHD. Methods. This single-center study included 77 donor-recipient pairs who underwent alloSCT. 64 pairs were HLA identical, 12 had 1 mismatch and 3 had 2 mismatch. 36 donors were related to their recipients. All recipients were followed at least for 100 (aGVHD) or 360 days (cGVHD) after transplantation. 43 donors were CMV-seropositive (CMVpos) and 34 were CMV-seronegative (CMVneg). 62 recipients were CMVpos, and 32 of them developed CMV reactivation, 25 aGVHD and 30 cGVHD. Samples from the graft product (donor) were phenotyped by flow cytometry (CD45, CD3, CD8, CD4, CD62L, CCR7) and both frequency (freq) and absolute number (abs) of each T-cell subpopulation were analyzed. Results. When the donors were divided based on their CMV serostatus, we observed that the grafts from CMVpos donors had a lower freq of naïve (CCR7+CD62L+) CD4+ T-cells (of lymphocytes p=0.06, of CD3 p=0.06, of CD4 p=0.07) and naïve CD8+ T-cells (of leukocytes p=0.03, of lymphocytes p=0.041, of CD3 p=0.011, of CD8 p=0.012) compared to CMVneg donors. Further, the abs of transplanted naïve CD8+ T-cells was significantly lower in the grafts from CMVpos donors (p=0.048). No differences were observed in T-cells (CD3+, CD4+, CD8+). We next studied if the CMV-serostatus and T-cell phenotype of the graft associates with GVHD. CMVpos donors whose recipients developed aGVHD had higher abs (p=0.05) and freq of naïve CD8+ T-cells (of lymphocytes p=0.08, of CD3 p=0.08, of CD8 p=0.11) compared to those without aGVHD. The same trend was observed with abs (p=0.11) and freq of CCR7+CD4+ T-cells (of leukocytes p=0.15). Similarly, those CMVpos donors whose recipients developed cGVHD had higher abs (p=0.05) and freq of CCR7+CD8+ T-cells (of leukocytes p=0.03, of lymphocytes p=0.06). Further, cGVHD patients who received the transplant from CMVpos donors were infused with a higher freq of CD3+ (of leukocytes p=0.03) and CD4+ T cells (of leukocytes p=0.04) than patients who received a graft from CMVpos donors but did not develop cGVHD. In contrast, CMVneg donors who developed aGVHD had a higher freq of CD3+ (p=0.018) and CD4+ T-cells (p=0.09), whereas no differences were seen in the T-cell subpopulations. Conversely, the abs (p=0.08) and freq of CCR7+CD4+ T-cells (of leukocytes p=0.11) were higher in those who later developed cGVHD. To study whether the graft lymphoid composition can be used to predict CMV reactivation, we analyzed the lymphoid composition in the graft product of those donors (both CMVpos and CMVneg) whose recipients were CMV seropositive but did not develop any form of GVHD (to avoid the influence of GVHD in the reactivation of CMV). Despite the low number of patients (CMV reactivation n=9, and no CMV reactivation n=13), we observed trends of higher portion of CD4+ T-cells (p=0.09 of lymphocytes, of CD3 p=0.20) and CCR7+CD4+ T-cells (of lymphocytes p=0.18, of CD4 p=0.16) in those grafts that were transplanted into CMV seropositive recipients who did not reactivate CMV. Conclusions. CMVpos donors whose recipient developed either aGVHD or cGVHD had a higher abs and freq of naïve CD8+ T-cells, which was not seen with CMVneg donors. This suggests that seropositivity sets the abs and freq of CD8 subpopulations near to a decisive cutoff for the development of GVHD. Conversely, other factors influences the development of GVHD in those patients whose donors were seronegative. In other words, seropositivity of the donor affects the graft composition and thus the risk of GVHD. Finally, our data indicate that a higher proportion of naïve or central memory CCR7+ CD4+ T cells in the donor graft could prevent CMV reactivation suggesting that graft composition affects also CMV reactivation. Disclosures No relevant conflicts of interest to declare.