Repurposing HLA genotype data of renal transplant patients to prevent severe drug hypersensitivity reactions.

Introduction: Specific alleles in human leukocyte antigens (HLAs) are associated with an increased risk of developing drug hypersensitivity reactions induced by abacavir, allopurinol, carbamazepine, oxcarbazepine, phenytoin, lamotrigine, or flucloxacillin. Transplant patients are genotyped for HLA as a routine practice to match a potential donor to a recipient. This study aims to investigate the feasibility and potential impact of repurposing these HLA genotype data from kidney transplant patients to prevent drug hypersensitivity reactions. Methods: A cohort of 1347 kidney transplant recipients has been genotyped in the Leiden University Medical Center (LUMC) using next-generation sequencing (NGS). The risk alleles HLA-A*31:01, HLA-B*15:02, HLA-B*15:11, HLA-B*57:01, and HLA-B*58:01 were retrieved from the NGS data. Medical history, medication use, and allergic reactions were obtained from the patient's medical records. Carrier frequencies found were compared to a LUMC blood donor population. Results: A total of 13.1% of transplant cohort patients carried at least one of the five HLA risk alleles and therefore had an increased risk of drug-induced hypersensitivity for specific drugs. HLA-A*31:01, HLA-B*15:02, HLA-B*57:01, and HLA-B*58:01 were found in carrier frequencies of 4.61%, 1.19%, 4.46%, and 3.35% respectively. No HLA-B*15:11 carrier was found. In total nine HLA-B*57:01 carriers received flucloxacillin and seven HLA-B*58:01 carriers within our cohort received allopurinol. Discussion: Our study shows that repurposing HLA genotype data from transplantation patients for the assignment of HLA risk alleles associated with drug hypersensitivity is feasible. The use of these data by physicians while prescribing drugs or by the pharmacist when dispensing drugs holds the potential to prevent drug hypersensitivity reactions. The utility of this method was highlighted by 13.1% of the transplant cohort patients carrying an actionable HLA allele.

Cell-free DNA measurement of three genomes after allogeneic MSC therapy in kidney transplant recipients indicates early cell death of infused MSC.

Introduction: Mesenchymal stromal cell (MSC) therapy is a promising treatment that allows for drug minimization in clinical kidney transplantation. While it is thought that MSCs rapidly go into apoptosis after infusion, clinical evidence for this is scarce since methods to detect cell death of infused cells in vivo are lacking. Cell-free DNA (cfDNA) has recently gained attention as a biomarker for cell death. Methods: In this study, we longitudinally measured cfDNA in plasma samples of the recipient, kidney donor, and allogeneic third-party MSC in the context of the Neptune study. cfDNA levels were measured at several time points before and after allogeneic MSC infusion in the 10 recipients who participated in the Neptune study. cfDNA ratios between the recipient, kidney graft, and MSC were determined. Results: We observed a peak in MSC-derived cfDNA 4 h after the first and second infusions, after which MSC-derived cfDNA became undetectable. Generally, kidney graft-derived cfDNA remained in the baseline-level range. Discussion: Our results support preclinical data that MSC are short-lived after infusion, also in a clinical in vivo setting, and are relevant for further research into the mechanism of action of MSC therapy.

Primary Trophoblast Cultures: Characterization of HLA Profiles and Immune Cell Interactions.

Introduction: Trophoblasts are essential in fetal-maternal interaction during pregnancy. The goal was to study HLA profiles of primary trophoblasts derived from placentas, and to investigate their usefulness in studying interaction with immune cells. Methods: After enzymatic digestion of first-trimester placental tissue from seven donors (6-9 weeks gestation) and trophoblast enrichment we cultured cytotrophoblasts (CTB) in stem cell medium. CTB were differentiated into EVT in a Matrigel-containing medium. A subset of CTB/EVT was profiled for microRNA levels. Expression of classical HLA molecules and of HLA-G was studied by flow cytometry, qPCR, and ELISA. Secondary trophoblast cell lines JAR and JEG-3 were studied as controls. Lymphocytes were investigated during co-culturing with EVT. Results: The trophoblasts could be easily maintained for several passages, upregulated classical trophoblast markers (GATA3, TFAP2C, chromosome-19 microRNAs), and upon differentiation to EVT they were selective in expressing HLA-C. EVT showed increasing expression of total HLA-G, an increasing proportion of HLA-G1 over G2- and G3 isoforms, and elevated excretion of soluble HLA-G. These features were distinct from those of the secondary trophoblast cell lines. TNF-α and IL-8 represented the most abundantly secreted cytokines by CTB, but their levels were minimal in EVT cultures. As proof of principle, we showed that EVT affect lymphocytes in three-day co-cultures (n=4) by decreasing activation marker HLA-DR. Conclusion: We verified the possibility culturing trophoblasts from first-term placentas, and their capability of differentiating to HLA-G expressing EVT. This culture model better represents the in-vivo situation than previously studied secondary trophoblast cell lines and enables mechanistic studies of fetal-maternal interactions.

A possible role for HLA-G in development of uteroplacental acute atherosis in preeclampsia.

HLA-G, a non-classical HLA molecule expressed by extravillous trophoblasts, plays a role in the maternal immune tolerance towards fetal cells. HLA-G expression is regulated by genetic polymorphisms in the 3' untranslated region (3'UTR). Low levels of HLA-G in the maternal circulation and placental tissue are linked to preeclampsia. Our objective was to investigate whether variants of the 3'UTR of the HLA-G gene in mother and fetus are associated with acute atherosis, a pregnancy specific arterial lesion of the decidua basalis that is prevalent in preeclampsia. Paired maternal and fetal DNA samples from 83 normotensive and 83 preeclamptic pregnancies were analyzed. We sequenced the part of the HLA-G 3'UTR containing a 14-bp insertion/deletion region and seven single nucleotide polymorphisms (SNPs). Associations with acute atherosis were tested by logistic regression. The frequency of heterozygosity for the 14-bp polymorphism (Ins/Del) and the +3142 SNP (C/G) variant in the fetus are associated with acute atherosis in preeclampsia (66.7 % vs. 39.6 %, p = 0.039, and 69.0 % vs. 43.4 %, p = 0.024). Furthermore, the fetal UTR-3 haplotype, which encompasses the 14-bp deletion and the +3142G variant, is associated with acute atherosis in preeclampsia (15 % vs. 3.8 %, p = 0.016). In conclusion, HLA-G polymorphisms in the fetus are associated with acute atherosis. We hypothesize that these polymorphisms lead to altered HLA-G expression in the decidua basalis, affecting local feto-maternal immune tolerance and development of acute atherosis.

HLA-G whole gene amplification reveals linkage disequilibrium between the HLA-G 3'UTR and coding sequence.

Polymorphic sites in the HLA-G gene may influence expression and function of the protein. Knowledge of the association between high-resolution HLA-G alleles and 3-prime untranslated (3'UTR) haplotypes is useful for studies on the role of HLA-G in transplantation, pregnancy, and cancer. We developed a next generation sequencing (NGS)-based typing assay enabling full phasing over the whole HLA-G gene sequence with inclusion of the 3'UTR region. DNA from 171 mother-child pairs (342 samples) was studied for: (a) HLA-G allele information by the NGSgo-AmpX HLA-G assay, (b) 3'UTR haplotype information by an in-house developed sequence-based typing method of a 699/713 base pair region in the 3'UTR, and (c) the full phase HLA-G gene sequence, by combining primers from both assays. The mother to child inheritance allowed internal verification of newly identified alleles and of association between coding and UTR regions. The NGSgo workflow compatible with Illumina platforms was employed. Data was interpreted using NGSengine software. In 99.4% of all alleles analyzed, the extended typing was consistent with the separate allele and 3'UTR typing methods. After repeated analysis of four samples that showed discrepancy, consistency reached 100%. A high-linkage disequilibrium between IPD-IMGT/HLA Database-defined HLA-G alleles and the extended 3'UTR region was identified (D' = 0.994, P < .0001). Strong associations were found particularly between HLA-G*01:04 and UTR-3, between HLA-G*01:01:03 and UTR-7, and between HLA-G*01:03:01 and UTR-5 (for all: r = 1). Six novel HLA-G alleles and three novel 3'UTR haplotype variants were identified, of which three and one, respectively, were verified in the offspring.

Soluble HLA-G levels in seminal plasma are associated with HLA-G 3'UTR genotypes and haplotypes.

Soluble HLA-G (sHLA-G) levels in human seminal plasma (SP) can be diverse and may affect the establishment of maternal-fetal tolerance and thereby the outcome of pregnancy. We investigated whether sHLA-G levels in SP are associated with polymorphisms in the 3'-untranslated region (UTR) and UTR haplotypes of the HLA-G gene. Furthermore, we compared the HLA-G genotype distribution and sHLA-G levels between men, whose partner experienced unexplained recurrent miscarriage (RM), and controls. Soluble HLA-G levels (n = 156) and HLA-G genotyping (n = 176) were determined in SP samples. The concentration of sHLA-G was significantly associated with several single-nucleotide polymorphisms (SNPs): the 14 base pair (bp) insertion/deletion (indel), +3010, +3142, +3187, +3196, and + 3509. High levels of sHLA-G were associated with UTR-1 and low levels with UTR-2, UTR-4, and UTR-7 (P < .0001). HLA-G genotype distribution and sHLA-G levels in SP were not significantly different between the RM group (n = 44) and controls (n = 31). In conclusion, seminal sHLA-G levels are associated with both singular SNPs and 3UTR haplotypes. HLA-G genotype and sHLA-G levels in SP are not different between men whose partner experienced RM and controls, indicating that miscarriages are not solely the result of low sHLA-G levels in SP. Instead, it is more likely that these miscarriages are the result of a multifactorial immunologic mechanism, whereby the HLA-G 3'UTR 14 bp ins/ins genotype plays a role in a proportion of the cases. Future studies should look into the functions of sHLA-G in SP and the consequences of low or high levels on the chance to conceive.

Increased HLA-G Expression in Term Placenta of Women with a History of Recurrent Miscarriage Despite Their Genetic Predisposition to Decreased HLA-G Levels.

Human leukocyte antigen (HLA)-G is an immune modulating molecule that is present on fetal extravillous trophoblasts at the fetal-maternal interface. Single nucleotide polymorphisms (SNPs) in the 3 prime untranslated region (3'UTR) of the HLA-G gene can affect the level of HLA-G expression, which may be altered in women with recurrent miscarriages (RM). This case-control study included 23 women with a medical history of three or more consecutive miscarriages who delivered a child after uncomplicated pregnancy, and 46 controls with uncomplicated pregnancy. Genomic DNA was isolated to sequence the 3'UTR of HLA-G. Tissue from term placentas was processed to quantify the HLA-G protein and mRNA levels. The women with a history of RM had a lower frequency of the HLA-G 3'UTR 14-bp del/del genotype as compared to controls (Odds ratio (OR) 0.28; p = 0.039), which has previously been related to higher soluble HLA-G levels. Yet, HLA-G protein (OR 6.67; p = 0.006) and mRNA (OR 6.33; p = 0.010) expression was increased in term placentas of women with a history of RM as compared to controls. In conclusion, during a successful pregnancy, HLA-G expression is elevated in term placentas from women with a history of RM as compared to controls, despite a genetic predisposition that is associated with decreased HLA-G levels. These findings suggest that HLA-G upregulation could be a compensatory mechanism in the occurrence of RM to achieve an ongoing pregnancy.

The combination of maternal KIR-B and fetal HLA-C2 is associated with decidua basalis acute atherosis in pregnancies with preeclampsia.

Acute atherosis is an arterial lesion most often occurring in pregnancies complicated by preeclampsia, a hypertensive pregnancy disorder. Acute atherosis predominates in the maternal spiral arteries in the decidua basalis layer of the pregnant uterus. This layer forms the fetal-maternal immunological interface, where fetal extravillous trophoblasts interact with maternal immune cells to promote decidual spiral artery remodeling and maternal immune tolerance towards the fetus. Of the classical polymorphic class I HLAs, extravillous trophoblasts express only HLA-C. HLA-C is a ligand for killer immunoglobulin-like receptors (KIR) on NK- and T-cells. Genetic combinations of fetal HLA-C and maternal KIRs affect pregnancy outcome. However, the role of HLA and KIR genes in acute atherosis is unknown. We hypothesized that specific genetic combinations of fetal HLA and maternal KIR are associated with the presence of acute atherosis lesions in the decidua basalis. We genotyped HLA class-I and II loci in paired fetal and maternal DNA samples from 166 pregnancies (83 preeclamptics, 83 controls). Acute atherosis was identified in 38 of these. Maternal KIR-loci were also genotyped. We found that the combination of maternal KIR-B haplotype and fetal HLA-C2 was significantly associated with acute atherosis in preeclampsia. In preeclamptic pregnancies with acute atherosis, 60% had this combination, compared to 24.5% in those without acute atherosis (p = 0.001). We suggest that interactions between fetal HLA-C2 and activating KIRs on maternal decidual NK-cells or T-cells may contribute to the formation of acute atherosis by promoting local decidual vascular inflammation.

HLA dosage effect in narcolepsy with cataplexy.

Narcolepsy with cataplexy is a sleep disorder caused by the loss of hypocretin-producing neurons in the hypothalamus. It is tightly associated with a specific human leukocyte antigen (HLA)-allele: HLA-DQB1*06:02. Based on this, an autoimmune process has been hypothesized. A functional HLA-DQ molecule consists of a DQα and a DQß chain. HLA-DQB1*06:02 (DQß) has a strong preference for binding to HLA-DQA1*01:02 (DQα), and together they form the functional DQ0602 dimer. A dosage effect would be expected if the HLA-DQ0602 dimer itself is directly involved in the aetiology. An increased expression of the HLA-DQ0602 dimer is expected in individuals homozygous for HLA-DQB1*06:02-DQA1*01:02, but is also hypothesized in individuals heterozygous for HLA-DQB1*06:02 and homozygous for HLA-DQA1*01:02. To study the impact of the expression of the HLA-DQ0602 dimer on narcolepsy susceptibility, 248 Dutch narcolepsy patients and 1272 Dutch control subjects, all of them positive for DQB1*06:02 (heterozygous and homozygous), were HLA-genotyped with attention not only to DQB1 but also to DQA1*01:02. DQB1*06:02-DQA1*01:02 homozygosity was significantly more often seen in patients compared to controls (O.R. 2.29) confirming previous observations. More importantly, a significantly higher prevalence of homozygosity for DQA1*01:02 was found in HLA-DQB1*06:02 heterozygous patients compared to controls (O.R. 2.37, p < 0.001). The latter finding clearly supports a direct role of the HLA-DQ molecule in the development of disease.

Naturally acquired microchimerism: implications for transplantation outcome and novel methodologies for detection.

Microchimerism represents a condition where one individual harbors genetically distinct cell populations, and the chimeric population constitutes <1% of the total number of cells. The most common natural source of microchimerism is pregnancy. The reciprocal cell exchange between a mother and her child often leads to the stable engraftment of hematopoietic and non-hematopoietic stem cells in both parties. Interaction between cells from the mother and those from the child may result in maternal immune cells becoming sensitized to inherited paternal alloantigens of the child, which are not expressed by the mother herself. Vice versa, immune cells of the child may become sensitized toward the non-inherited maternal alloantigens of the mother. The extent of microchimerism, its anatomical location, and the sensitivity of the techniques used for detecting its presence collectively determine whether microchimerism can be detected in an individual. In this review, we focus on the clinical consequences of microchimerism in solid organ and hematopoietic stem cell transplantation, and propose concepts derived from data of epidemiologic studies. Next, we elaborate on the latest molecular methodology, including digital PCR, for determining in a reliable and sensitive way the extent of microchimerism. For the first time, tools have become available to isolate viable chimeric cells from a host background, so that the challenges of establishing the biologic mechanisms and function of these cells may finally be tackled.

HLA-targeted flow cytometric sorting of blood cells allows separation of pure and viable microchimeric cell populations.

Microchimerism is defined by the presence of low levels of nonhost cells in a person. We developed a reliable method for separating viable microchimeric cells from the host environment. For flow cytometric cell sorting, HLA antigens were targeted with human monoclonal HLA antibodies (mAbs). Optimal separation of microchimeric cells (present at a proportion as low as 0.01% in artificial mixtures) was obtained with 2 different HLA mAbs, one targeting the chimeric cells and the other the background cells. To verify purity of separated cell populations, flow-sorted fractions of 1000 cells were processed for DNA analysis by HLA-allele-specific and Y-chromosome-directed real-time quantitative PCR assays. After sorting, PCR signals of chimeric DNA markers in the positive fractions were significantly enhanced compared with those in the presort samples, and they were similar to those in 100% chimeric control samples. Next, we demonstrate applicability of HLA-targeted FACS sorting after pregnancy by separating chimeric maternal cells from child umbilical cord mononuclear cells. Targeting allelic differences with anti-HLA mAbs with FACS sorting allows maximal enrichment of viable microchimeric cells from a background cell population. The current methodology enables reliable microchimeric cell detection and separation in clinical specimens.

HLA-DRBeta1, circulating Th1/Th2 cytokines and immunological homunculus in coronary atherosclerosis.

Coronary atherosclerotic disease is one of the most endangering health disorder worldwide. This study was designed to investigate the correlation between HLA-DR1 alleles and circulating Th1/Th2 type cytokines in coronary atherosclerosis. By Elisa, Th1/Th2 type cytokines were determined in serum samples of 31 subjects with unstable angina, 27 subjects with chronic stable angina and 24 individuals as normal control. By SSP-PCR, more than 100 alleles of HLA-DRBeta1 were typed in 24 subjects who had skewed serum levels of Th1/Th2 type cytokines. Lipid profiles were determined by the routine methods of clinical laboratory in all subjects. The mean serum concentration of IL-10 in normal control subjects was higher in comparison to the patient groups.0.33±0.59 pg/ml versus 0.064±0.3 pg/ml in unstable angina pectoris group (p<0.028) and 0.22±0.6 pg/ml in chronic stable subjects. There was no statistically significant difference among the groups in serum levels of other desired cytokines (IFN-Gamma, IL-4). 33.33% of normal control subjects were HLA-DR16 positive whereas none of the subjects with chronic stable angina or individuals with unstable angina pectoris was positive for this antigen. The mean concentration of serum LDL-cholesterol in normal control group was high 142.046±35.40 (pg/ml).This preliminary study shows that the atherogenic effect of the LDL- cholesterol may be dampened by HDL-cholesterol through anti inflammatory cytokine IL-10 and HLA-DR16, a phenomenon interpretable via immunological homunculus theory.

Non-HLA T-cell reactivity during the first year after HLA-identical living-related kidney transplantation.

BACKGROUND: It has been reported that donor-reactive T-cell responses may decrease during the first year after HLA-mismatched organ transplantation. We wondered whether donor-reactive T-cell responses directed to minor histocompatibility antigens (mHAgs) or other non-HLA antigens also decrease after HLA-identical living-related (LR) kidney transplantation. METHODS: We studied donor-reactive T-cell responses by IFN-gamma and granzyme B (GrB) Elispot assays in 15 HLA-identical LR kidney transplant recipients before, six months and one yr after transplantation. Third-party reactivity was used as control. Patient and donor peripheral blood mononuclear cells were typed for 11 known mHAgs. RESULTS: During the study period, 60% and 36% of the patients demonstrated donor-reactive IFN-gamma and GrB producing cells (pc), respectively. The number of donor-reactive IFN-gamma and GrB pc was significantly lower than the number of third-party reactive IFN-gamma and GrB pc. After transplantation, donor-reactivity and third-party reactivity were comparable to pre-transplant values. No relation was found in mHAg mismatches between donor and recipient and donor-reactive T-cell response. CONCLUSIONS: Donor-reactivity could be detected before and after HLA-identical LR kidney transplantation, but was not related with the number of mHAg mismatches, and did not decrease after transplantation.

Donor-reactive cytokine profiles after HLA-identical living-related kidney transplantation.

BACKGROUND: After HLA-identical living-related (LR) kidney transplantation, only non-HLA antigen mismatches between donor and recipient may exist. We questioned whether donor-reactive responses against non-HLA antigens could be found after HLA-identical LR kidney transplantation, and wondered whether donor reactivity in the HLA-identical setting was different from the HLA-mismatched setting during immunological quiescence. Healthy individuals served as controls. METHODS: Elispot assays were performed to determine the number of alloreactive IFN-gamma-producing cells (pc), IL-10 pc, granzyme B (GrB) pc and IL-13 pc from peripheral blood mononuclear cells (PBMC) of HLA-identical, HLA-mismatched LR kidney transplant recipients and healthy individuals. RESULTS: The frequency of alloreactive IFN-gamma pc, IL-13 pc and GrB pc was higher in healthy individuals compared to both transplant patient groups. In the HLA-identical group, significantly higher numbers of donor-reactive IL-10 pc were found compared to their autologous control. These frequencies were also higher compared to the HLA-mismatched and healthy control group. The number of donor-reactive GrB pc was higher in the HLA-mismatched group than in the HLA-identical group. Donor-reactive IFN-gamma pc and IL-13 pc were comparable in both transplant groups. CONCLUSIONS: In recipients of HLA-identical LR kidney transplant, high donor-reactive IL-10 pc, in combination with low donor-reactive IFN-gamma pc, IL-13 pc and GrB pc, suggests active downregulation of reactivity against non-HLA molecules.

Stable T-cell reactivity after successful tapering of azathioprine in HLA-identical living-related kidney transplant recipients despite minor histocompatibility antigen mismatches.

BACKGROUND: Human leukocyte antigen (HLA)-identical living-related (LR) kidney transplant recipients often receive the standard regimen of immunosuppression. We wondered whether these patients should be exposed to the side effects of these drugs any longer. Safe tapering of immunosuppression should not result in rejection and high donor-directed T-cell responses. In the present study, we investigated the effect of tapering azathioprine (AZA) on T-cell reactivity. METHODS: Fifteen HLA-identical LR kidney transplant recipients receiving a median of 150 mg/day AZA and 5-10 mg/day prednisone were tapered to a median of 50 mg/day AZA. Donor-, third-party and tetanus toxoid (TET)-reactivity were determined in interferon (IFN)-gamma and interleukin (IL)-13 Elispot assays, which reflect the T-helper (Th)1 and T-helper (Th)2 response. RESULTS: After the tapering of AZA, none of the patients developed acute rejection and the renal function remained stable, even at 1-year follow-up. The frequency of donor-specific IFN-gamma and IL-13 producing cells (pc) was low. Tapering of AZA did not influence the frequency of both IFN-gamma and IL-13 pc. Also, the reactivity against third-party cells and TET remained unchanged. CONCLUSIONS: The AZA-dose can be safely reduced in recipients of an HLA-identical LR kidney transplant without affecting kidney function and without increasing T-cell responses directed against donor or other antigens.

A uniform genomic minor histocompatibility antigen typing methodology and database designed to facilitate clinical applications.

BACKGROUND: Minor Histocompatibility (H) antigen mismatches significantly influence the outcome of HLA-matched allogeneic stem cell transplantation. The molecular identification of human H antigens is increasing rapidly. In parallel, clinical application of minor H antigen typing has gained interest. So far, relevant and simple tools to analyze the minor H antigens in a quick and reliable way are lacking. METHODOLOGY AND FINDINGS: We developed a uniform PCR with sequence-specific primers (PCR-SSP) for 10 different autosomal minor H antigens and H-Y. This genomic minor H antigen typing methodology allows easy incorporation in the routine HLA typing procedures. DNA from previously typed EBV-LCL was used to validate the methodology. To facilitate easy interpretation for clinical purposes, a minor H database named dbMinor (http://www.lumc.nl/dbminor) was developed. Input of the minor H antigen typing results subsequently provides all relevant information for a given patient/donor pair and additional information on the putative graft-versus-host, graft-versus-tumor and host-versus-graft reactivities. SIGNIFICANCE: A simple, uniform and rapid methodology was developed enabling determination of minor H antigen genotypes of all currently identified minor H antigens. A dbMinor database was developed to interpret the genomic typing for its potential clinical relevance. The combination of the minor H antigen genomic typing methodology with the online dbMinor database and applications facilitates the clinical application of minor H antigens anti-tumor targets after stem cell transplantation.

Transfusion-associated graft vs. host disease after donor-specific leukocyte transfusion before kidney transplantation.

Transfusion-associated graft vs. host disease (TA-GVHD) is a well-known but rare complication that follows infusion of histo-incompatible lymphoid cells, often seen in individuals with impaired cellular immunity. However, we present here a case report of fatal TA-GVHD in a 'presumed' immunocompetent patient after transfusion of a freshly isolated buffycoat from a relative as part of our protocol to prepare the patient for living-related kidney transplantation. To confirm the diagnosis of TA-GVHD, a polymerase chain reaction was used to detect donor cells in various affected tissues. Furthermore, the immune reactivity of the patient against donor and vice versa was tested on samples taken before transfusion using limiting dilution assays. Our patient received a transfusion with blood from a donor who was homozygous at the human leukocyte antigen (HLA) class I loci. Despite incompatibility for HLA class II, infused donor T lymphocytes were not rejected and became engrafted. The patient did not have cytotoxic T lymphocytes to reject the donor cells. DNA polymorphism studies on several organ biopsies confirmed the presence of infiltrating cells of donor origin. This report illustrates the possibility, in the general patient population, of developing TA-GVHD from whole blood transfusion. In the case of pre-transplant blood transfusion, the patient and donor have to be HLA-typed and special care should be taken in the situation of donor homozygosity for HLA class I, even in the presence of HLA class II incompatibility. Protocols in which donor-specific blood or bone marrow transfusions are given in an attempt to modulate the immune system should exclude these combinations.