|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background Successful allogeneic bone marrow transplantation relies on global immunosuppression or elimination of T cells. In contrast, the induction of anergy can inactivate specific sets of alloreactive T cells in the donor marrow. Previous work has shown that anergy can be induced by blocking the interaction of the B7 molecule on the surface of antigen-presenting cells with the CD28 molecule on the surface of T cells, thus preventing key signaling events essential for the activation of T cells. To investigate the feasibility of this approach with respect to transplantation of histoincompatible bone marrow, we undertook a clinical trial of ex vivo induction of anergy in T cells present in donor marrow to recipient alloantigens.
Methods Outcomes in 12 transplant recipients were evaluated. The recipients' peripheral-blood lymphocytes were collected before myeloablation and served as alloantigen-presenting cells. To induce alloantigen-specific anergy, bone marrow from a donor mismatched with the recipient for one HLA haplotype was cocultured with irradiated cells from the recipient for 36 hours in the presence of CTLA-4–Ig, an agent that inhibits B7:CD28-mediated costimulation. After conventional myeloablation and immunoprophylaxis, the treated donor cells were transfused into the recipient.
Results After the induction of anergy, the frequency of T cells capable of recognizing alloantigens of the recipient in donor marrow was sharply reduced (P<0.001), whereas the responsiveness to alloantigens from persons unrelated to the recipient or the donor was unaffected (P=0.51). In the 11 patients who could be evaluated, the haploidentical bone marrow cells engrafted. Of these 11 patients, 3 had acute graft-versus-host disease (GVHD) confined to the gastrointestinal tract. No deaths were attributable to GVHD. Five of the 12 patients were alive and in remission 4.5 to 29 months after transplantation.
Conclusions Donor bone marrow treated ex vivo to induce anergy to alloantigens from the recipient can reconstitute hematopoiesis in vivo with a relatively low risk of GVHD.
A critical costimulatory signal is generated by the interaction between members of the B7 family of proteins on antigen-presenting cells and the CD28 molecule on the surface of T cells.17,18,19 Blockade of this interaction results in the induction of anergy.20 Studies in animals have shown that interference with the B7:CD28 interaction permits successful transplantation of histoincompatible and even xenogeneic allografts without the need for pharmacologic immunosuppression.21,22,23,24
In the case of bone marrow transplantation, T cells in the donor marrow are the cause of graft-versus-host disease (GVHD).25,26,27,28,29,30 We hypothesized that the induction of anergy in donor T cells that have the potential to react against the recipient's alloantigens might ameliorate GVHD while preserving the rest of the T-cell repertoire. One way to induce anergy in alloreactive T cells is by using a soluble preparation of CTLA-4, a counterreceptor for B7 that is found on activated T cells,31 to block the interaction between B7 proteins and CD28. As compared with CD28, CTLA-4 has a much higher affinity for B7, and thus it is a highly effective inhibitor of the B7:CD28 interaction.
We have previously shown that CTLA-4–Ig, a soluble fusion protein consisting of the extracellular domain of CTLA-4 linked to the constant region of IgG1, can reduce the alloreactivity of bone marrow T cells in vitro.32 We conducted a clinical trial of the feasibility of inducing anergy before the transplantation of histoincompatible bone marrow. We first incubated donor marrow cells with CTLA-4–Ig in the presence of mononuclear cells from the recipient. We then transferred the treated donor cells to the recipients and evaluated the outcomes in terms of the restoration of hematopoiesis and the incidence of GVHD.
Methods
Patients
Patients were eligible for this study if they were ineligible for autologous stem-cell transplantation at our institutions on the basis of disease status and if they had no phenotypically or genotypically HLA-matched or single-antigen–mismatched family member who was available to serve as a donor, nor was there a phenotypically matched unrelated donor. Those likely to have rapid progression of disease during a search for an unrelated donor and those for whom a search was unsuccessful after four or more months were also eligible. Initially, patients were eligible if they had acute myelogenous or lymphoblastic leukemia, non-Hodgkin's or Hodgkin's lymphoma, multiple myeloma, or chronic lymphocytic leukemia that met prescribed criteria for high risk (those with advanced disease, those in relapse, or those with other poor prognostic features). Those with myelodysplasia or bone marrow failure became eligible for the study subsequently. All the patients had to be free of clinically significant organ dysfunction or active infection and had to have a life expectancy of at least six weeks. The institutional review boards of Children's Hospital and Dana–Farber Partners Cancer Care approved the protocol. Each patient or a legal guardian or parent gave written informed consent.
Between March 1996 and October 1998, 12 patients entered this study and underwent transplantation with haploidentical bone marrow (i.e., marrow from a donor who shared only one of two major-histocompatibility-complex haplotypes with the recipient) (Table 1). Follow-up is reported through February 20, 1999. The median age of the patients was 10 years. Eleven patients had hematologic cancer, and one had congenital amegakaryocytic thrombocytopenia. Of the 11 patients with hematologic cancer, 5 had never had a remission despite undergoing multiple treatments, 1 had not initially entered remission but did so subsequently, and 5 had entered a first remission readily. Only four patients underwent bone marrow transplantation while they were in remission. Of the remainder, most had virtually complete replacement of their marrow by malignant cells. Autologous transplantation had failed 27 and 11 months previously in Patients 2 and 6, respectively. The patients with acute leukemia in second remission, lymphoma in remission, bone marrow failure, or myelodysplasia without excess blasts were considered "good-risk patients," whereas all the others were considered "poor-risk patients."
|
|
The patients underwent leukapheresis to collect 200 million to 600 million mononuclear cells per kilogram of body weight for use as the recipient's alloantigen-presenting cells. These cells were cryopreserved with the use of a standard method. Subsequently, the patient received 1400 cGy of total-body irradiation33 followed by cytarabine, cyclophosphamide, and methylprednisolone. Because of the high rate of toxic effects caused by this regimen in the first eight patients, cytarabine was not given to the last four patients. Donor marrow was harvested two days before transplantation. Prophylaxis against GVHD consisted of short-course methotrexate and cyclosporine34 starting on the day before transplantation, either by continuous infusion at 0.1 mg per kilogram per hour (in seven patients) or by a bolus of 1.5 mg per kilogram over a period of 2 to 3 hours every 12 hours (in five patients). The dose of cyclosporine was modified to alleviate toxic effects and was given orally as soon as gut function permitted. Methotrexate was withheld in cases of mucositis or renal insufficiency. Because of the high rate of regimen-related toxic effects in the first eight patients, the final dose of methotrexate was reduced and the administration of leucovorin was instituted for the last four patients (Table 3). GVHD was assessed with standard clinical criteria and graded according to the International Bone Marrow Transplant Registry (IBMTR) Severity Index.35
|
Ex Vivo Induction of Anergy
Cryopreserved mononuclear cells derived from the recipient's blood were thawed, washed, and irradiated at a midplane dose of 3300 cGy. The cells were resuspended at a concentration of 20 million cells per milliliter in RPMI 1640 medium with 5 percent human AB serum. CTLA-4–Ig (Repligen, Cambridge, Mass.) was added at a concentration of 20 µg per milliliter for a minimum of 30 minutes before the addition of the donor marrow cells. The erythrocyte-depleted, mononuclear-cell fraction of the marrow was resuspended at a concentration of 20 million cells per milliliter in RPMI 1640 medium with 5 percent human AB serum and added to the mixture of recipient cells and CTLA-4–Ig in a donor:recipient ratio of 1:1 to 2:1. Cells were cultured in Lifecell tissue-culture flasks (Baxter, Deerfield, Ill.) for 36 hours at 37°C in 5 percent carbon dioxide, washed, and then infused into the recipient.
Immunologic Assays
The frequency of precursor helper T cells was determined as previously reported.32 Briefly, we irradiated (with 2500 cGy) 1 million stimulator cells per milliliter in RPMI 1640 medium with 5 percent heat-inactivated human AB serum; 100 µl of the culture medium was then added to each well of a U-bottomed microtiter plate. Nonirradiated responder cells were added in aliquots of 100 µl to each of 72 replicate wells at dilutions ranging from 100,000 cells to 1 cell per well. After incubation for 64 hours at 37°C in 5 percent carbon dioxide, plates were irradiated (with 2500 cGy), CTLL-2 cells (interleukin-2–responsive T cells; 1000 cells in a volume of 50 µl) were added, and the culture was continued for another 24 hours. Proliferation was assessed by the incorporation of tritiated thymidine for the last 16 hours. The frequency of precursor cells was calculated on the basis of the proportion of wells showing no production of interleukin-2 by limiting-dilution analysis. Stimulation indexes were determined by mixed-leukocyte culture.32 In vitro production of antibodies with specificity against influenza A was determined.37 Influenza-specific antibodies were quantitated by enzyme-linked immunosorbent assay.
Donor–Recipient Hematopoietic Chimerism
Hematopoietic chimerism was evaluated by fluorescence in situ hybridization for sex chromosomes or by polymerase-chain-reaction (PCR) amplification of sequence-specific primers or oligonucleotide probes for HLA class I and class II donor and recipient antigens. Unfractionated blood or marrow samples were analyzed. In some cases, after Ficoll–Hypaque separation and red-cell lysis, the pellet was analyzed as the "granulocyte" fraction and the lymphomononuclear interface was analyzed as the "lymphocyte" fraction in order to differentiate between myeloid and lymphoid cells. PCR, as performed here with sequence-specific primers for class I and class II alleles, can detect 1 in 100 and 1 in 10,000 cells, respectively.38 Although expected to be substantially greater, the sensitivity of sequence-specific oligonucleotide probes in this setting has not been defined.
Detection of Leukemic Cells after Transplantation
In patients with a previously documented cytogenetic abnormality, specimens of bone marrow aspirate were sent to the clinical cytogenetic laboratory for analysis. To identify gene rearrangements associated with leukemia and rearrangements of the immunoglobulin gene in patients with acute lymphoblastic leukemia, genomic DNA was extracted from samples taken from the patients and amplified by PCR with a series of seven consensus primers for the framework-1 segment of the VH (variable heavy chain) region and a JH (joining heavy chain) consensus primer, as previously described.39 Clonal PCR products were excised and purified with Wizard PCR Preps (Promega, Madison, Wis.). Purified PCR fragments were sequenced directly by the Molecular Biology Core Facility at the Dana–Farber Cancer Institute with a DNA sequencer (model AB 373A, Perkin Elmer Applied Biosystems, Foster City, Calif.). The relevant VH family and JH consensus primers were used to obtain the sequence information from both strands, and sense and antisense sequences were aligned. VH, DH (diversity heavy chain), JH, and regions of the N-nucleotide addition were identified with the Basic Local Alignment Search Tool (National Center for Biotechnology Information, Bethesda, Md.).
Antisense allele-specific oligonucleotide primers for the CDRIII region were designed to have annealing temperatures of approximately 60°C and were synthesized by GIBCO BRL (Gaithersburg, Md.). Seminested PCR was performed as previously described.40 PCR products were then analyzed by electrophoresis on 2 percent agarose gels. Well-established precautions were taken to prevent carryover contamination of PCR reactions.41
The cellular immunophenotype was determined by dual-color fluorescence-activated cell sorting (Coulter Profile, Coulter, Miami) with the use of directly conjugated monoclonal antibodies for CD3+, CD4+, CD8+, CD56+, CD20+ (Coulter), and CD34+ (Becton Dickinson, San Jose, Calif.).
Statistical Analysis
Descriptive statistics are provided for determinations of the frequency of precursor cells. The Wilcoxon signed-rank test was used to assess changes in the logarithm (base 10) of the assay results before and after treatment with CTLA-4–Ig.
Results
Frequency of Alloreactive Precursor Helper T Cells after Induction of Anergy
The harvested donor bone marrow contained mature T cells, with medians of 16 million CD3+ cells per kilogram of the recipient's weight, including 8 million CD3+CD4+ cells per kilogram and 6 million CD3+CD8+ cells per kilogram. Medians of 28 million CD3+ cells per kilogram, including 14 million CD3+CD4+ cells per kilogram and 9 million CD3+CD8+ cells per kilogram, were infused after the ex vivo treatment (Table 4). Because not all of the irradiated recipient T cells were removed by density centrifugation after the coculture procedure, the number of T cells infused into the patient was generally greater than that in the harvested marrow.
|
|
A median of 3 million CD34+ donor cells per kilogram of the recipient's weight were treated in the coculture system, and a median of 2.2 million CD34+ cells per kilogram were infused into the recipient (Table 4). A second marrow harvest from the donor of Patient 4 yielded 900,000 CD34+ cells per kilogram of the recipient's weight, and 600,000 CD34+ cells per kilogram were infused.
Patient 2, who died on day 8, could not be evaluated for engraftment (Table 3). In Patient 4 the neutrophil count did not reach 500 per microliter despite cytogenetic evidence of donor cells in the recipient's bone marrow and blood. Therefore, cells from a second harvest were treated ex vivo and infused on day 35 without further conditioning of the patient or prophylaxis against GVHD. The neutrophil count was more than 500 per microliter 20 days later. Patient 1 achieved a neutrophil count of 100 per microliter on day 21, one day before her death. The remaining patients had prompt recovery of neutrophils (a median of 20 days until a neutrophil count 
500 per microliter was reached, Table 3). Patients 1, 2, 4, 5, 6, and 8 died before platelet or red-cell engraftment could be documented. Patients 3, 7, 10, 11, and 12 achieved a platelet count of more than 20,000 per microliter between 22 and 123 days after transplantation (median, 43). Patient 11 died before confirmed red-cell engraftment.
Patients 3, 7, 10, and 12 received their last red-cell transfusion 51 to 101 days after transplantation (median, 65). Patient 9 had persistent moderate pancytopenia, presumed to be secondary to recurrent cytomegalovirus infection. He received his last red-cell transfusion on day 108 and required intermittent platelet transfusions until day 197. The other surviving patients all have normal blood counts. Chimerism studies, performed in 8 of 10 patients in whom engraftment occurred, revealed no evidence of residual recipient hematopoietic cells at the first or subsequent evaluations (Table 5).
|
Despite the possibility of ex vivo sensitization of donor T cells against recipient alloantigens, no hyperacute GVHD was observed.42 Eleven of the 12 patients could be evaluated for acute GVHD, and 6 patients could be evaluated through day 100 (the entire risk period for acute GVHD). None of the patients had skin or liver findings consistent with acute GVHD (Table 3), but three (Patients 3, 5, and 11) had gastrointestinal symptoms and findings consistent with acute GVHD. Two of these patients had only diarrhea, and one had gastrointestinal hemorrhage. These nonspecific events might have been caused by concurrent influenza A (in Patient 3), persistent gastrointestinal toxic effects at day 20 from chemoradiotherapy (in Patient 5), and documented intolerance of infant formula (in Patient 11).
Nonetheless, on the basis of stool volume and biopsy specimens showing apoptosis or degeneration of crypt cells, Patient 3 was considered to have GVHD with an overall IBMTR index rating of B (skin and liver involvement, stage 0; gastrointestinal involvement, stage 2). This condition resolved within 10 days after treatment with methylprednisolone was begun. Patient 11, who had GVHD with an overall rating of B (skin and liver involvement, stage 0; gastrointestinal involvement, stage 2), also had a response to corticosteroids. Patient 5 had GVHD with an overall rating of C (skin and liver involvement, stage 0; gastrointestinal involvement, stage 3). Her stool output diminished temporarily with corticosteroid treatment, then diarrhea persisted until death on day 39. An autopsy disclosed ischemia and hemorrhage, without evidence of GVHD, suggesting that the GVHD had been successfully treated or that the patient had not had GVHD. Limited autopsy (Patient 4) or complete autopsy (Patients 2, 5, and 6) did not reveal pathological evidence of GVHD.
Only one of the patients given cytarabine for conditioning received all of the intended GVHD prophylaxis. Only half of the patients given cytarabine received the intended dose of methotrexate on day 3, only two of eight did so on day 6, and none of the seven patients who could be evaluated did so on day 11 (Table 3). It was anticipated that cyclosporine would be continued for 6 to 12 months after transplantation. As the study progressed, the target trough level of cyclosporine was decreased from 200 to 300 mg per milliliter to 100 to 150 mg per milliliter, and cyclosporine was discontinued much earlier than expected (on days 289, 88, 122, 101, and 120 for Patients 3, 7, 9, 10, and 12, respectively).
Clinical Outcome
Outcome was assessed according to disease classification (Table 3). Of the good-risk patients, one died early of regimen-related toxicity and three are alive, disease-free, and at home receiving no cyclosporine, corticosteroids, or other immunosuppressive treatment; they have no signs of chronic GVHD. Patient 10, who received a bone marrow transplant for congenital amegakaryocytopenia, has normal hematopoiesis and a platelet count of more than 250,000 per microliter. Neither Patient 3 nor Patient 7 has clinical evidence of leukemic relapse, and no evidence of leukemia was detected by PCR analysis of bone marrow samples (from day 754 for Patient 3 and day 391 for Patient 7) with the use of allele-specific oligonucleotide probes for rearrangements present in the patients' pretransplantation leukemic cells.
One of the eight poor-risk patients died on day 60 of a relapse of acute myelogenous leukemia. Five patients, including Patients 2 and 6, who had previously undergone autologous transplantation, died of infection and regimen-related toxicity before or shortly after day 100. Autopsies of Patients 2, 4, 5, and 6 disclosed no histologic evidence of cancer. Two poor-risk patients are alive and at home. In Patient 9, diarrhea and recurrent cytomegalovirus infection developed after discharge. Biopsy specimens taken from the small bowel and sigmoid colon on day 145 revealed GVHD, classified as chronic by virtue of its occurrence after day 100, but a biopsy of a rash showed no evidence of GVHD. Patient 9 was treated with cyclosporine and corticosteroids, with complete histologic resolution shown by a repeated biopsy of the gastrointestinal tract on day 197. The doses of corticosteroids and cyclosporine are being tapered. A bone marrow aspirate obtained from Patient 9 on day 197 was histologically normal and free of the prior cytogenetic abnormality (inversion of chromosome 12). As of this writing, Patient 12 is at day 132, two weeks after discontinuing cyclosporine, and has had no evidence of acute or chronic GVHD. Analysis of blood obtained from this patient on day 67 showed no evidence of leukemia on PCR amplification of allele-specific oligonucleotide primers for the T-cell–receptor rearrangement present in his pretransplant leukemic cells.
Infection and Immune Reconstitution
Infections developed in three patients after discharge. Influenza A developed in Patient 3 on day 56, after he was exposed to his infected sibling. It resolved without incident after treatment with amantadine. Dermatomal herpes zoster also developed on day 509 and was treated uneventfully with acyclovir. Patient 9 received his transplant immediately after an episode of cytomegalovirus viremia and acute hepatic necrosis. Recurrent cytomegalovirus infection developed on day 36 and was treated successfully with immune globulin and foscarnet. Another recurrence of cytomegalovirus infection soon after discharge was also treated successfully. Patient 10, who is currently in day care, was given a diagnosis of community-acquired respiratory syncytial virus on the basis of screening of nasal secretions nine months after transplantation. The nasal congestion resolved uneventfully without treatment.
As assessed by the expression of CD56, all the survivors except Patient 9 (Patients 3, 7, 10, and 12) had increased numbers of natural killer cells soon after transplantation. They also rapidly recovered and maintained normal numbers of CD20+ B cells. IgG levels were maintained in the age-adjusted normal range by 9 months in Patient 3, 3.5 months in Patient 7, 2.5 months in Patient 10, and 3 months in Patient 12. Patient 9 could not be evaluated because he received immune globulin for prophylaxis against cytomegalovirus infection.
By two months after transplantation, the CD3+CD4+:CD3+CD8+ ratio was 1 in all the patients evaluated (Patients 7, 9, 10, and 12). Patient 3, first evaluated four months after transplantation, also had a ratio of 1. All five surviving patients had a normal CD3+CD4+:CD3+CD8+ ratio (1.4) by seven months after transplantation. The absolute CD4+ T-cell counts rose to 400 per microliter or higher by six months in three of the four patients who survived until then; the sole exception was Patient 9, who had recurrent cytomegalovirus disease and chronic GVHD. Currently, Patient 12 has an absolute CD4+ T-cell count of 140 per microliter four months after transplantation.
Engrafted donor T cells from Patients 3 and 7 have been assessed for reactivity against recipient alloantigens. Peripheral-blood mononuclear cells collected from Patient 3 on day 379, 90 days after cyclosporine was discontinued, were cultured with either irradiated stimulator cells from three unrelated persons or with irradiated cryopreserved recipient peripheral-blood cells. The proliferative response to the cells from the unrelated persons was 5 to 11 times as high on day 5 of coculture as the response to the recipient cells. In an identical experiment with peripheral-blood mononuclear cells collected from Patient 7 on day 173, 85 days after the discontinuation of cyclosporine, the proliferative response to the cells from the unrelated persons was 6 to 12 times as high as the response to the recipient cells. In concurrent experiments, the frequencies of helper T-cell precursors responding to cryopreserved cells were 0.00001 in Patient 3 and 0.0000008 in Patient 7, as compared with frequencies of precursors responding to cells from unrelated persons of 0.0003 and 0.00003, respectively.
Community-acquired influenza A in Patient 3 provided an opportunity to measure the in vivo response to an infection that had not affected the recipient or donor before transplantation. Immediately before vaccination of both the recipient and the donor against influenza A on day 355, the recipient's engrafted cells responded to the influenza A virus in vitro, presumably because he had a naturally acquired infection two months after transplantation, whereas the cells from the uninfected donor showed no response. After vaccination, both the recipient and the donor had measurable responses in vitro.
Discussion
We found that ex vivo blockade of B7-mediated costimulation induces anergy in T cells residing in the bone marrow of a haploidentical donor that are ordinarily reactive to alloantigens of the recipient. Moreover, this treated marrow, after transplantation, supports the hematologic reconstitution of the recipient and replaces all of the recipient's hematopoietic cells. With conventional prophylaxis against GVHD, we would have expected to see severe GVHD in 60 to 90 percent of recipients of haploidentical marrow,43,44 yet only one case of acute GVHD with an IBMTR rating of C and no cases with a rating of D were observed, and no deaths were attributable to GVHD. Prophylaxis against GVHD could be discontinued uneventfully, and chronic GVHD has developed in only one patient. Despite an anticipated high rate of mortality in this population, 5 of the 12 patients are alive and disease-free.
It appears that donor marrow treated ex vivo to induce anergy can reconstitute hematopoiesis with full lymphoid and myeloid donor chimerism (Table 5). This study did not have adequate power for statistical certainty but, rather, was designed to investigate whether the expected high incidence of GVHD would be diminished after the ex vivo treatment. The occurrence of only two cases of acute GVHD with a rating of B and one case with a rating of C was particularly striking, since all the patients had multiple risk factors for GVHD, of which donor–recipient histoincompatibility was the most obvious.43,45,46,47 Furthermore, all the patients received donor bone marrow containing more than 10 million CD3+ T cells per kilogram of the recipient's weight, exceeding by one to two orders of magnitude the suggested threshold dose for minimizing GVHD.48,49 As a group, the 12 patients we treated had additional risk factors for GVHD, including mismatching of donor and recipient with respect to sex, donors who had been pregnant, incomplete methotrexate or cyclosporine prophylaxis, recurrent or persistent disease, irradiation-based conditioning (particularly with >1200 cGy), cytomegalovirus seropositivity, and older donor age.43,46,50,51,52,53,54,55 Moreover, the onset of GVHD is often early in cases of transplantation of haploidentical donor marrow with conventional immunoprophylaxis.56,57 All these elements support the contention that ex vivo induction of anergy in donor marrow T cells can prevent or minimize GVHD.
An alternative approach to the problems of graft failure and severe GVHD after haploidentical marrow transplantation involves intensive myeloablation with multiple agents, coupled with treatment of the recipient with antithymocyte globulin before transplantation, exhaustive depletion of T cells from the donor's marrow, and markedly increased doses of donor stem cells. Partial depletion of T cells and immunosuppression after transplantation, including that resulting from treatment with antithymocyte globulin, have also been tried.58,59,60,61 However, graft failure, substantial regimen-related toxicity, opportunistic infection, and delayed reconstitution of immune function are still problems.58,59,60,61,62 Because these approaches rely on nonspecific ablation of T cells, immunosuppression, and aggressive preparation of the recipient, these obstacles will probably be difficult to overcome.
As indicated by the frequency of precursor helper T cells with alloreactivity against haploidentical recipient alloantigens in the recipient–donor pairs that we studied, the frequency of donor helper T-cell precursors responsive to haploidentical recipient alloantigens is at most 0.1 percent (i.e., 0.001). This frequency suggests that more than 99 percent of the 500 million to 1.5 billion cells contained in the harvested donor marrow do not respond to recipient alloantigens and may be infused without the recipient's incurring a substantial risk of GVHD. This strategy of inducing anergy affects only a particular segment of the T-cell repertoire, leaving the rest to aid engraftment and restore immunocompetence. On the basis of recent assessments in this small cohort of patients, it appears that there is potential for rapid recovery of a normal ratio of CD4+ T cells to CD8+ T cells in peripheral blood and that the absolute number of CD4+ T cells can reach a normal level by approximately six months after transplantation.
More important, patients appear able to overcome acquired and reactivated viral infections. No lymphoproliferative disease associated with the Epstein–Barr virus has been seen clinically or on autopsy thus far, despite nearly uniform seropositivity in the patients and the donors. There were, however, four deaths related to infection. One patient died of bacterial sepsis (Patient 2), and two died of aspergillus infection (Patients 5 and 8). These infections occurred very soon after transplantation and only in patients with chronic neutropenia due to relapsed disease and severe mucositis who had received cytarabine as part of their conditioning regimen. Patient 4, another member of the group that received cytarabine, died of disseminated toxoplasmosis, discovered at autopsy. This patient had protracted leukopenia and neutropenia because of delayed engraftment. There were two deaths resulting from multiorgan failure in this cohort. No fungal disease, deaths from infection, or episodes of multiorgan failure have occurred since cytarabine has been eliminated from the conditioning regimen, suggesting that these events may have been due to the additional immunosuppression, mucositis, and toxic effects resulting from the aggressive conditioning regimen.60,61,63
In summary, we showed that inactivation of a selected segment of the alloreactive T-cell repertoire through blockade of B7-mediated T-cell costimulation is possible and appears to result in a significant reduction in the risk of GVHD. Further evidence that the portion of the T-cell repertoire that is not alloreactive is left unperturbed is essential in establishing whether this and similar methods are realistic alternatives to solving the difficult problem of transplanting haploidentical bone marrow.
Supported by grants from the National Institutes of Health (P01 AI41584 and P50 HL54785) and the Juvenile Diabetes Foundation International.
Dr. Nadler was a scientific advisor to Repligen, the manufacturer of the CTLA-4–Ig used in this study. He also received laboratory support from Repligen and held stock options in the company. These arrangements ended in September 1995, six months before this study began. Dr. Nadler also divested himself of all stock options he owned in the company.
We are indebted to Eileen A. Whyte, M.S.N., C.P.N.P., for her excellent management of outpatient care; to the other members of the medical and nursing staffs who cared for these patients and facilitated their treatment; to the clinical and laboratory personnel (particularly the dedicated staff of the Connell–O'Reilly Cell Manipulation Laboratory), whose efforts supported this program; to the referring physicians; and to the patients and families themselves who undertook this difficult and demanding study with us.
Source Information
From the Departments of Pediatric Oncology (E.C.G., L.L.B.), Adult Oncology (V.A.B., L.M.N., J.G.G.), and Biostatistics (D.N.), Dana–Farber Cancer Institute; the Division of Hematology and Oncology, Children's Hospital (E.C.G.); the Division of Medical Oncology, Brigham and Women's Hospital (V.A.B., N.H., L.M.N., J.G.G.); and the Departments of Pediatrics (E.C.G.) and Medicine (V.A.B., D.N., N.H., L.M.N., J.G.G.), Harvard Medical School — all in Boston.
Address reprint requests to Dr. Guinan at the Dana–Farber Cancer Institute, 44 Binney St., Boston, MA 02115, or at eva_guinan{at}dfci.harvard.edu.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Transplantation of Anergic Histoincompatible Bone Marrow Cells
de Carvalho Bittencourt M., Tiberghien P., Saas P., Gallardo D., Grañena A., García-López J.
Extract |
Full Text
N Engl J Med 1999;
341:1081-1082, Sep 30, 1999.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
