HLA-identical sibling bone marrow transplants vs chemotherapy for acute myelogenous leukemia in first remission.

There is controversy whether adults with acute myelogenous leukemia (AML) in first remission are best treated with chemotherapy or an HLA-identical sibling bone marrow transplant. We studied 1097 adults, 16-50 years old, with AML in first remission. Results of transplants from HLA-identical siblings reported to the International Bone Marrow Transplant Registry (IBMTR; n = 901) were compared with results of chemotherapy in comparable persons treated by the German AML Cooperative Group (GAMLCG; n = 196). Preliminary analyses identified subject- and disease-related variables differing between the cohorts and associated with treatment outcome within each cohort. We adjusted for these variables and differences in time-to-treatment in subsequent comparisons of treatment-related mortality, relapse, survival and leukemia-free survival (LFS). Five-year probability of treatment-related mortality was greater for transplants than chemotherapy (43% (95% confidence interval, 37-49%) vs 7% (3-11%); P< 0.0001). Five-year relapse probability was less for transplants than chemotherapy (24% (20-28%) vs 63% (55-71%); P< 0.0001). Five-year probability of survival was similar with transplants and chemotherapy (48% (43-53%) vs 42% (33-51%); P = 0.24). Five-year LFS probability was higher for transplants than chemotherapy (46% (42-50%) vs 35% (28-41%); P= 0.01). These data indicate that bone marrow transplants from HLA-identical siblings result in comparable survival but greater LFS than chemotherapy in adults with AML in first remission.

Experimental immunologically mediated aplastic anemia (AA) in H-2k identical, Mls (M) locus different mice.

Immunologically mediated aplastic anemia (AA) was experimentally induced in mice by injecting 10(7) lymph node cells (LNC) from donor mice of one inbred strain to another H-2k identical but Mls mismatched strain previously given 600 rad total body gamma irradiation (TBI). AA developed after 2 weeks to 6 months in selected strain combinations used and usually 60 to 90% of the mice died. Clinical signs of graft-versus-host disease did not occur and splenic atrophy rather than splenomegaly was the rule. Histologically these mice had a lesion of the hematopoietic microenvironment characterized by sinusoidal injury and stromal necrosis. Others have demonstrated injury to hematopoietic stem cells. C3H/He LNC induced AA whereas C3H/HeJ LNC failed to induce AA. The C3H/HeJ strain carries a macrophage defect and these results suggest that a macrophage-like cell may be a mediator of immunological injury in this experimental model. Although all strain combinations evaluated were H-2k identical and Mls mismatched, certain Mls combinations resulted in AA and identical Mls mismatched but different strains did not. Both strong (Mlsd) and weak (Mlsc) stimulating LNC induce AA but simple Mls differences do not explain the AA as similar Mls combinations but different strain combinations fail to induce AA.

Colony growth in cultures from bone marrow and peripheral blood after curative treatment for leukemia and severe aplastic anemia.

The recovery of colony-forming cell numbers after curative treatment for leukemia and severe aplastic anemia (SAA) was studied. We examined 191 patients (85 acute myeloid leukemia [AML], 48 acute lymphocytic leukemia [ALL], 32 chronic myeloid leukemia [CML], 17 SAA, and nine myelodysplastic syndrome [MDS]) who were in hematologic remission 6 months to 13 years after either curative chemotherapy (n = 69) or allogeneic bone marrow transplantation (BMT) (n = 122) by culturing their precursor cells from bone marrow (BM) (n = 548) and peripheral blood (PB) (n = 529) in methylcellulose. Thirty-six BM donors and 25 PB donors served as controls. BM colony-forming cell numbers were abnormally low in all patients (p < 0.002) irrespective of underlying disorder and type of treatment (chemotherapy or irradiation). These numbers did not normalize with time--colony-forming cells were still strongly reduced up to 10 years after therapy, whether or not the patient had received an allogeneic bone marrow graft (p < 0.002). We also compared patients who remained in stable hematologic remission with those who later relapsed (6 months to 2 years after treatment). BM colony-forming cell numbers were significantly lower in patients who subsequently relapsed (p = 0.004). In contrast to BM cultures, we found normal colony-forming capacity by PB precursors in all patients. We conclude that (1) after chemotherapy or BMT, colony-forming cell numbers of BM in culture are permanently reduced; (2) this defect is probably due to a dysfunction of the BM environment rather than to a numerical reduction of the precursor cell pool; and (3) very low colony-forming capacity may be related to relapse.

High T cell colony numbers do not reflect immune competence.

The number of T cell colonies growing from ficoll separated PHA-stimulated peripheral blood cells was compared to 3H-thymidine uptake in 133 normal donors of all ages. Unexpectedly, the two parameters did not correlate: The tendency of growing cells to aggregate during the incubation period resulted in few, but large spherical colonies in persons who by young age and high counts for thymidine uptake were judged to have good T cell function. This clustering effect was less pronounced in old persons and those with a low PHA response; it was virtually absent in 17 of 18 patients with undoubted cellular immune deficiency early after allogeneic bone marrow transplantation. This resulted in a growth pattern consisting of a high number of small scattered colonies. The capacity of growing cells to aggregate was sensitive to sublethal irradiation. We conclude that the number of T cell colonies growing from a given cell suspension not only depends on the number of lymphocytes responding to the mitogen. It is negatively determined by the drive of growing cells to cluster. This phenomenon deserves further study, since it appears to be a parameter of cellular immune competence which is not readily recognized with 3H thymidine incorporation.

High-dose recombinant human erythropoietin for treatment of anemia in myelodysplastic syndromes and paroxysmal nocturnal hemoglobinuria: a pilot study.

In a dose escalation study we tested the feasibility and tolerance of high-dose recombinant human erythropoietin (r-HuEPO) therapy in four patients with ineffective erythropoiesis due to myelodysplastic syndromes (MDS) or paroxysmal nocturnal hemoglobinuria (PNH). Recombinant human EPO was administered i.v. with an initial dose of 50 U/kg body weight (BW) three times per week. The dose was increased by steps of 25 or 50 U/kg bW with intervals of 1-4 weeks up to a maximum dose of 500 U/kg BW three times per week. All patients were treated as outpatients. Pre-study treatment with cyclosporin A and/or Danazol was continued in three patients. In one patient r-HuEPO was discontinued after 20 weeks because of relapse of severe aplastic anemia. No major side effects were observed even at the maximum dose. One patient with PNH showed an increase of hemoglobin from 89 to 139 g/liter that permitted monthly phlebotomies to reduce his iron overload. In one patient with MDS the reticulocyte count increased from 2.5 to 50 x 10(9)/liter, and the transfusion requirement decreased to 2 U every 3-4 weeks instead of every 2 weeks. Two patients did not complete the whole treatment period and showed no rise in reticulocyte count. We conclude that high dose r-HuEPO therapy is feasible in patients with anemia due to MDS or PNH. High-dose r-HuEPO appears to have some effect on anemia due to ineffective erythropoiesis in a subgroup of patients. Further studies are needed to identify potential responders and to define the optimal administration of r-HuEPO.

Treatment of severe aplastic anemia.

A total of 100 patients with severe aplastic anemia were treated and evaluated in a prospective study at our hospital between January 1976 and October 1983: 28 patients had an HLA-identical sibling donor and were treated with bone marrow transplantation, and 72 patients without an HLA-identical sibling donor were given antilymphocyte globulin followed by oral low-dose androgen therapy. At 1 1/2-9 years after treatment, 13 patients (46%) survive in the transplant group and 45 patients (75%) survive in the second group. All except one in the second group have self-sustaining hematopoiesis without need for transfusions. There is one major difference between the two therapies. Marrow transplantation restores bone marrow function completely and no late hematologic complications have been seen in this group. The majority of patients treated with antilymphocyte globulin in contrast have residual abnormalities of hematopoiesis: macrocytosis, mild granulocytopenia, and mild thrombopenia. Relapse (11 of 72 patients) and clonal hematologic disorders, such as paroxysmal nocturnal hemoglobinuria (four patients) and leukemia (one patient) can occur years after complete bone marrow reconstitution with antilymphocyte globulin. These late disorders are of concern. Despite this, we conclude that antilymphocyte globulin treatment is an effective therapy with low early mortality and morbidity and a high chance for a long, sustained remission. Results are better or at least equivalent to bone marrow transplantation and patients with donors should be given the option of transplant or antilymphocyte globulin.

Autologous marrow recovery following allogeneic marrow transplantation in a patient with severe aplastic anemia.

A 45-year-old woman with severe idiopathic marrow failure was prepared for marrow transplantation by administration of cyclophosphamide (cy) 50 mg/kg on each of four successive days. She then received an intravenous infusion of 20 X 10(9) nucleated marrow cells from an HL-A matched and mixed lymphocyte culture (MLC) non-reactive sister. There was evidence for minimal marrow recovery in 1-2 months and a second marrow infusion was carried out 69 days after the first without additional immunosuppression. There was a continued slow recovery of peripheral blood counts with complete reconstitution of erythropoiesis, return of the white blood cell count to between 3 and 4000/mm3, with 50% granulocytes, and platelets to 60--70,000/mm3, 11 months after the initial grafting attempt. Red cell antigens and gamma globulin allotypes were of recipient type. The MLC and the indirect cell mediated lympholysis (CML) test became positive possibly indicating cellular sensitization to non HL-A antigens. This report of a patient with severe marrow failure documents autologous recovery of marrow function after receiving a large dose of cy and allogeneic marrow. The implications of this are discussed.

Treatment of Aplastic Anemia.

There is general agreement that in children and adolescents with an HLA-identical or syngeneic sibling, bone-marrow transplantation (BMT) should be performed without delay. More controversial are young-to-middle-aged adults with an HLA-identical sibling. Because of comparable survival rates, some centers advocate BMT; others advocate immunosuppression as primary treatment. BMT cures severe aplastic anemia (SAA), but has a higher early mortality. Immunosuppression usually induces hemopoietic improvement and has hardly any early mortality. However, there are late problems. After immunosuppression, there may be relapse and clonal hematopoietic disease. After BMT, there may be late graft failure, severe or even fatal chronic graft-versus-host-disease (GVHD). Both procedures are followed by a slight increase in second malignancies. Retrospective studies show no significant differences in long-term survival between the two treatment modalities. In patients over the age of 40, immunosuppression is favored by most centers because in this age group there is a very high transplant-related mortality with BMT. Androgens and hematopoietic growth factors have hardly any role in therapy of SAA. Splenectomy helps some patients with severe thrombocytopenia and problems with supportive care, but does not improve the overall prognosis. BMT with unrelated donors has usually been performed in late stages of the disease with disappointing results.

B/T cell ratio of rabbit peripheral blood lymphocytes. Influence of separation technique on results.

Spontaneous in vitro T cell rosette formation at room temperature leading to enrichment of B cells has been reported. We tested 10 individual New Zealand White rabbits sequentially, separating the lymphocytes either at room temperature or at 37 degrees C. T cells are lost constantly at room temperature but to a lesser extent at 37 degrees C. The determination of the yield of lymphocytes after Ficoll separation gives the best control for the accuracy of the results. If lymphocytes are separated at 37 degrees C and if the yield of lymphocytes is greater than 45%, the variation in T cells is small and their number is constant between 62 and 74%. These data show that the reported wide range of T cells in rabbit peripheral blood is due to methodological errors and not inherent in the rabbit.

Histoincompatible skin and marrow grafts in rabbits on cyclosporin A.

Cyclosporin A (Cy A; 15 mg/kg s.c.) allows allogeneic histoincompatible skin graft survival in 10 of 10 rabbits as long as the drug is given. The same dose of Cy A does not affect acute graft-versus-host disease (GVHD) and increase survival of rabbits after allogeneic bone marrow transplantation between the same two strains, when two groups of 15 animals are tested. Median survival of 14 engrafted animals without Cy A was 23 days, of 12 engrafted animals with Cy A 22 days (not significant). Cryopreservation of bone marrow delays the onset of GVHD and increases survival of engrafted animals. Median survival of 5 engrafted animals without Cy A was 33 days, of 14 engrafted animals with Cy A 35 days. Our hypothesis is that one part of early GVHD as well as of early graft rejection is mediated by a subclass of cells which is resistant to Cy A. This hypothesis is supported by the finding that most transplanted skins show a self-limited period of infiltration and induration. Skin grafts survive this period of infiltration. If the same potentially self-limited process occurs in the liver or the intestine during acute GVHD, animals die. this hypothesis could explain why Cy A allows skin graft survival but does not affect acute GVHD.

Characteristics and clinical relevance of autolymphocytotoxins in patients with aplastic anemia.

Serum of 68 patients with aplastic anemia was tested for the presence of autolymphocytotoxins (auto-LTs). Prior to specific disease treatment, 16 patients (24%) displayed antibodies cytotoxic to their own lymphocytes. These antibodies had the characteristics of cold-reactive lymphocytotoxins. Their detection in patients' sera was found unrelated to a viral or toxic cause of the disease or the patients' HLA genotype. Broadly reactive anti-HLA antibodies were less frequent in pretreatment sera containing auto-LTs, suggesting that these autoantibodies could modulate alloantibody production. However, after specific disease treatment, the alloantibody frequency was comparable in patients with or without auto-LTs. We found no significant difference in response to antilymphocyte serum or bone marrow graft outcome in the patients in relation to the presence or absence of pretreatment auto-LTs. This observation suggests that the detection of these autoantibodies in aplastic anemia has no clinical relevance.

Cyclosporine in human bone marrow transplantation. Serum concentration, graft-versus-host disease, and nephrotoxicity.

Serum concentrations of cyclosporine were studied in 42 patients given cyclosporine for the prevention of graft-versus-host-disease (GVHD) following allogeneic bone marrow transplantation (BMT). Serum trough levels for cyclosporine were determined in each patient at least once weekly during the first 3 months and were compared with the occurrence of GVHD and with nephrotoxicity. Cyclosporine was given as 20 mg/kg i.m. or as a 24-hr infusion for the first 5-7 days. This was followed by a single daily oral dose of 12.5 mg/kg for 6 months. Cyclosporine was then gradually reduced and stopped after one year. After a median observation period of 2 years 25 of the 42 patients (59%) are alive. Twenty six patients (62%) developed GVHD, which was stage II or more in 11 (26%) and fatal in 2 patients (5%). Five patients developed GVHD 6-8 weeks after withdrawal of cyclosporine one year after BMT. All patients improved after restarting cyclosporine. No correlation between cyclosporine serum concentration and GVHD was observed, but patients with GVHD had greater fluctuations of their serum trough levels. Serum creatinine increased in all patients soon after BMT and was correlated with cyclosporine serum concentration during the first month. Serum creatinine, however, rose further despite lower cyclosporine concentrations in the second month. These results show that cyclosporine effectively reduces the severity, but not the incidence, of GVHD suggesting that there is a subset of cells resistant to cyclosporine. The therapeutic range, however, between high doses (which are often associated with nephrotoxicity) and the minimal effective dose of cyclosporine, has yet to be defined.

Immunologic reactivity in marrow graft recipients receiving cyclosporine to prevent graft-versus-host disease.

In vitro and in vivo immunologic parameters were determined in 26 patients treated continuously with cyclosporine to prevent graft-versus-host-disease (GVHD) after allogeneic bone marrow transplantation (BMT) for acute and chronic leukemia and for aplastic anemia. A group of 18 patients was tested 6 months after BMT and another group of 10 patients was tested after one year. At 6 months after BMT, 94% of the patients had normal serum IgG and IgM levels, whereas at one year 29% of them had low IgA levels. The proportion of patients with normal lymphocyte responses in vitro at 6 months after BMT was 69% and 76% for the responses to concanavalin A and to soluble antigens; 75% and 53% for the responses to allogeneic cells and pokeweed mitogen, respectively; and 89% for the response to phytohemagglutinin. All but one were able to generate suppressor cells upon con A stimulation. At one year after the graft, only one patient had demonstrable multiple abnormalities in in vitro tests. Skin test reactivity at one year was comparable to pre-graft reactivity. After BMT a lymphopenia persisted for 6 months. The rate of infectious complications was high during the first 3 months after BMT, and it diminished progressively as immune functions returned to normal. Infection was the cause of death in two cases (one disseminated cytomegalovirus infection and one septicemia). GHVD occurred in 12 patients; in nine of them the disease was transient and mild, only 1 patient developed severe chronic GVHD. Acute GVHD did not influence the tempo of immunologic reconstitution. In comparison to other studies, it seems that cyclosporine does not delay immune restoration, or increase morbidity from infection, while preventing GVHD and its complications efficiently.

Neopterin as a new biochemical marker in the clinical monitoring of bone marrow transplant recipients.

In previous reports we demonstrated that increased amounts of the pyrazinopyrimidine compound neopterin are released in the context of T lymphocyte activation. The aim of this investigation was twofold: (1) to define the contribution of hemopoetic cells to neopterin excretion, and (2) to search for the clinical utility of this biochemical marker in the monitoring of such patients. Thirteen patients were grafted with allogeneic, 1 with syngeneic, and 2 with autologous marrow. Urinary neopterin excretion was measured daily by means of high-performance liquid chromatography from the time before transplantation until the patients' discharge from the isolation unit. In all patients bone marrow aplasia was associated with depressed, and engraftment with increased, neopterin values. Rising neopterin levels invariably preceded the cytological definition of "take," on the average by seven days. After hematological reconstitution, neopterin excretion continuously declined in all 5 patients lacking infectious complications and/o-graft-versus-host disease (GVHD). A transitory increase of urinary neopterin followed by normalization was observed in 5 further patients. At the time of increased neopterin excretion, 4 experienced either herpetic infection or GVHD, both of which resolved promptly under the appropriate treatment. Neopterin values remained elevated after engraftment in 6 patients who suffered from persistent GVHD. Results of this pilot study suggest that (1) bone marrow derived cells are crucially involved in production of neopterin in vivo and (2) evaluation of neopterin excretion patterns after hemopoietic reconstitution enables one to discriminate between patients with and without an increased risk of developing GVHD or viral disease.

Chronic cyclosporine-associated nephrotoxicity in bone marrow transplant patients.

This study describes the prevalence and degree of chronic cyclosporine-associated nephropathy and its risk factors. For this purpose we reviewed all available renal histology specimens in 169 bone marrow transplant recipients treated during an eight year period with cyclosporine for prevention of graft-versus-host-disease, and determined their pattern and degree of histomorphological changes. A total of 51 specimens obtained from 49 patients by biopsy (n = 12) or autopsy (n = 39) was evaluated. The pattern of histomorphological changes was compared with diagnosis, age, sex, and potential risk factors--such as cyclosporine dose, levels, duration of therapy, changes in serum creatinine and onset of hypertension. Morphological lesions of chronic cyclosporine-associated nephropathy were found in 67% of the specimens. They were more frequent and more severe with increasing duration of cyclosporine therapy, in patients with a higher increase in serum creatinine during the first 3 months and in patients given total-body irradiation for conditioning. These latter findings suggest that additional damage sensitizes the kidney to irreversible toxic effects of cyclosporine.

Kinetics of cytomegalovirus IgG antibody following infusion of a hyperimmune globulin preparation in allogeneic marrow transplant recipients.

We investigated prospectively cytomegalovirus (CMV) IgG antibody kinetics following infusion of a CMV hyperimmune globulin preparation in 19 consecutive patients undergoing allogeneic bone marrow transplantation (BMT). CMV IgG against late antigen was measured by enzyme-linked immunosorbent assay. On a prophylactic hyperimmune globulin regimen (100 mg/kg body weight every 20 days), the mean half-life (t1/2) +/- SD of CMV IgG in seronegative patients with seronegative marrow donors was 19.7 +/- 7.9 days after the first infusion on day -7 before BMT, 24.9 +/- 12.5 days after the second infusion on day +13 after BMT, and 19.4 +/- 6.9 days after the third infusion on day +33. The t1/2 in individual patients showed a wide variation ranging from 10.9 to 47.6 days. A therapeutic high-dose hyperimmune globulin regimen (200-400 mg/kg body weight at 4-day intervals) given to five patients resulted in high serum levels of CMV IgG, which were maintained throughout therapy, even in two patients with severe intestinal graft-versus-host disease. The variation of CMV IgG kinetics between individual BMT recipients may contribute to the lack of protection from active CMV infection observed in some patients.

Allogeneic peripheral blood precursor cell transplants in rabbits.

This report summarizes a series of experiments undertaken to evaluate the role of mobilized peripheral blood precursor cells (PBPC) for transplantation across a major histocompatibility barrier. Adult outbred red Burgundy rabbits were used as donors, New Zealand white rabbits of the opposite sex as recipients. Conditioning consisted of single dose total body irradiation (TBI) of 10 Gy supported by a short course of cyclosporine to enhance engraftment. Human recombinant G-CSF at a dose of 10 micrograms/kg was used for mobilization of precursor cells. Three methods of PBPC transplants were tested initially in 5 animals each. PBPC were collected and infused at once on day 0; collected initially, cryopreserved for one month, infused on day 0 and followed by 3 additional fresh donations or collected and infused on 6 occasions between days 0 and + 11. 13 animals engrafted, 2 became complete, longterm chimeras. Survival was best in the group given repetitive infusions (39 days median, 12 days to > 180 days, range). 10 additional animals were transplanted as in the last group and the number of transplanted nucleated cells (10.5 x 10(8)/kg median, 7.3 - 15.7 x 10(8)/kg range) and colony forming units CFU-GM (42 x 10(4)/kg median, 12.3 - 176.8 x 10(4)/kg range), were compared with outcome. Median survival of the 10 animals was 29 days (12 - 55 days range; 1 autologous reconstitution). Survival did not correlate with total nucleated cells per kg (r = 0.10; p = 0.79), but there was a trend to prolong survival with higher numbers of CFU-GM per kg (r = 0.47; p = 0.19). These data show that allogeneic PBPCT can engraft across a major histocompatibility barrier, that the high number of CFU-GM per kg might be advantageous, but also that additional methods are warranted to reduce acute GvHD.

Irradiated donor buffy coat following T cell-depleted bone marrow transplants.

Twenty patients aged 27-47 years (median, 35 years) with hematological malignancies, treated with T cell-depleted bone marrow transplantation received in a pilot study five donations of 15 Gy irradiated donor buffy coat cells at days +1, +3, +5, +7, +14 in order to prevent rejection and leukemic relapse. Patients were conditioned with etoposide, cyclophosphamide and 12 Gy fractionated total body irradiation and given cyclosporine postgrafting. Donor bone marrow was T cell-depleted (median 3% remaining T cells) by counterflow elutriation. All patients engrafted. Fourteen (70%) are alive. Two are living with relapse, 12 (60%) are alive and well without any signs of disease, 2-27 months (median, 9 months) post-transplant. Three patients died of interstitial pneumonitis and/or graft-versus-host-disease; three died of relapse. This pilot study supports previous animal data. Repeated infusions of 15 Gy irradiated donor buffy coat are feasible and do not appear to increase transplant related mortality. Whether this approach ultimately will reduce the rate of rejection and relapse following T cell depletion needs to be confirmed in a larger, prospective study.

Transplantation of G-CSF mobilized allogeneic peripheral blood stem cells in rabbits.

Mobilized peripheral blood precursor cells (PBPC) are used with increasing frequency to restore autologous hematopoiesis following high-dose radio-chemotherapy. The success of this method has aroused interest in the use of mobilized PBPC for allogeneic transplants. This approach would eliminate the need for marrow aspiration and general anesthesia. In this project we tested the feasibility of allogeneic histoincompatible PBPC transplants in rabbits. Adult outbred Red Burgundy rabbits were used as donors, histoincompatible New Zealand White rabbits of the opposite sex as recipients. One individual donor was used for one individual recipient. Conditioning consisted of 10 Gy total body irradiation (TBI). Donor animals were pre-treated with recombinant human granulocyte colony-stimulating factor (rh G-CSF) given s.c. at 10 micrograms/kg daily. Three schedules of PBPC collection and reinfusion were tested in 3 groups of animals, each consisting of 5 donor recipient pairs: (A) PBPC were collected either on days -2, -1 and 0, and infused at once after TBI on day 0; (B) collected and infused on days 0, +2, +4, +7, +9, and +11; (C) collected on 3 consecutive days, cryopreserved for 1 month and infused on day 0 followed by 3 fresh donations on days +4, +8 and +11. The median amount of blood processed from donor animals was 470 ml (312-602) containing about 10 x 10(8) (5-71 x 10(8)) nucleated cells. Recipient animals received a median of 2.7 x 10(8) cells/kg equivalent to 9.6 x 10(4) colony-forming units granulocyte-macrophages (CFU-GM)/kg (data derived from Group C of the animals).(ABSTRACT TRUNCATED AT 250 WORDS)

Celiac disease transmitted by allogeneic non-T cell-depleted bone marrow transplantation.

We observed the occurrence of celiac disease following allogeneic bone marrow transplantation in a patient transplanted for acute leukemia. The marrow donor was his HLA-identical sister, who had suffered from celiac disease since birth. The post-transplant period was characterized by recurrent episodes of diarrhea. Detailed workup showed atrophic intestinal mucosa on histology and anti-gliadin and anti-endomysium antibodies in the serum, features that were not present before transplantation. GVHD was absent at that time. The patient remains free of symptoms on gluten-free diet and slight immunosuppression. This case suggests transmission of celiac disease by bone marrow transplantation and supports the T cell concept in celiac disease.