|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on June 7, 2002; DOI 10.1182/blood-2002-01-0176.
TRANSPLANTATION
From the Johns Hopkins University School of Medicine,
Oncology Center and the Kanazawa University Graduate School of Medical
Science, Cellular Transplantation Biology.
Administration of the immunosuppressive drug cyclosporine A (CsA)
following autologous stem cell transplantation paradoxically elicits a
systemic autoimmune syndrome resembling graft-versus-host disease
(GVHD). This syndrome, termed autologous GVHD, is associated with
autoreactive CD8+ T cells that recognize major
histocompatibility complex (MHC) class II determinants in association
with a peptide from the invariant chain. To investigate the potential
role of cytokines and chemokines in autologous GVHD, interleukin 2 (IL-2), IL-4, IL-10, interferon High-dose chemoradiotherapy combined with
autologous stem cell transplantation (SCT) can be used successfully in
the treatment of patients with malignant lymphoma. A number of clinical
studies demonstrate that among the therapeutic options available for
relapsing lymphoma, autologous SCT is the most effective approach to
achieve long-term survival.1-3 However, clinical trials
have failed to demonstrate any significant advantage of autologous SCT
over conventional chemotherapy in the treatment of patients with either
chemotherapy-resistant non-Hodgkin lymphoma (NHL), with NHL in
remission with poor prognostic factors,4,5 or in patients
with breast cancer with extensive lymph node
involvement.6,7 Raising the dose intensity of chemotherapy
facilitated by the use of autologous SCT does not necessarily improve
the outcome. For patients who do not achieve sustained remission after
conventional chemotherapy, novel approaches including immunotherapy are
needed. Based on the findings that the relapse rate after allogeneic
SCT is remarkably lower in patients with graft-versus-host disease
(GVHD) compared with patients who do not develop this
syndrome,8-10 one potential approach is the induction of
autologous GVHD after autologous SCT.
Autologous GVHD can be induced in recipients of autologous bone marrow
by administration of cyclosporine A (CsA) for a short period following
transplantation.11-16 This autoaggression syndrome shares
similar dermal pathology with acute GVHD after allogeneic SCT. Several
initial clinical trials in patients with acute myeloid leukemia and in
patients with NHL suggest that the induction of autologous GVHD can
reduce the rate of relapse.17-20 Analysis of the effector
mechanisms involved in autologous GVHD reveal promiscuous recognition
of major histocompatibility complex (MHC) class II determinants by
CD8+ T cells.21-24 Clonal expansion of
autoreactive T cells are observed in both humans and in the rat
model.24,25 Interestingly, the CD8+
autoreactive T cells lyse myeloma, lymphoma, and breast cancer cell
lines.21,22 Moreover, the antitumor effect can be enhanced by administration of interferon Autologous and allogeneic GVHD are multistep processes. During the
"induction phase," T cells react to Ag disparities and clonally
expand ("expansion phase"). The activated T cells also release
cytokines and chemokines, resulting in the recruitment of other cells
such as macrophages and natural killer (NK) cells in the "recruitment
phase." Finally, the concert of T lymphocytes and other cell types
mediate the pathology associated with GVHD (the "effector phase").
Cytokines drive the immune response and play a pivotal role in all
phases of GVHD. However, cytokines play a complex and dual role in
GVHD, and can have either protective or deleterious effects. In
particular, cytokines such as interleukin 10 (IL-10) and IFN- To investigate the role of cytokines and chemokines in autologous GVHD,
the current study examined IL-2, IL-4, IL-10, IFN- Patients
Preparative regimens and induction of autologous GVHD
Cell-mediated lympholysis assay Lytic activity was assessed sequentially after autologous SCT using PBMCs isolated by Ficoll-Hypaque density centrifugation. Target cells used in these studies included autologous PHA-blasts cryopreserved before transplantation, the MHC class II-positive breast cancer cell line T47D,42,43 and the NK target cell line K562. The pretransplantation lymphocytes were thawed and stimulated with PHA for 72 hours (RPMI 1640; 20% normal human serum) before use as targets. The K562 cell line was grown in suspension culture in RPMI 1640 tissue culture medium supplemented with 10% fetal calf serum. The cells were washed 3 times before the cell-mediated lympholysis (CML) assay. T47D is an adherent breast cancer line grown in Dulbecco modified Eagle medium (DMEM) tissue culture medium supplemented with 5% fetal calf serum, glutamine, and sodium pyruvate.42,43 Before assay, the cells were mildly trypsinzed and grown in Nalgene Teflon flasks (Thomas Scientific, Swedesboro, NJ) for 24 hours to allow for the recovery of cell surface antigens. The target cells were labeled with 250 µCi (9.25 MBq) of 51Cr for 1 hour at 37°C. The effector lymphocytes were cocultured with 2.5 × 103 labeled target cells in triplicate in round-bottom microtiter wells. After 4 hours incubation, 51Cr release was assessed and the percent specific lysis was determined from triplicate cultures as previously described.22,25RNA and genomic DNA extraction Heparinized peripheral blood was collected from the patients after informed consent was obtained, and PBMCs were separated using density-gradient centrifugation. Monocytes were isolated by plastic adherence and gentle scraping (> 85% CD14+ by flow cytometry). Lymphocytes were also separated into distinct T-cell subsets by immunomagnetic bead separation chromatography using monoclonal antibodies (MoAbs) to the CD4+ and CD8+ cell surface determinants (Dynal Biotech, Oslo, Norway), as previously described.22,44 The purity of CD4+ or CD8+ cells was typically more than 97%. Skin biopsies (4-mm punch biopsies) were obtained with informed patient consent before transplantation and upon initial development of erythematous rash. After initial fracturing of the tissue in the presence of liquid nitrogen with pestle,44,45 total RNA was purified with Trizol reagent (Life Technologies, Gaithersburg, MD). Cell lysate was prepared from 5 × 106 PBMCs in 1 mL Trizol reagent with adequate mixing. After adding 200 µL chloroform, the solution was well mixed and centrifuged. The supernatant was collected and extracted once with chloroform. RNA was precipitated with 2-propanol and rinsed with 70% ethanol. Purified RNA was dissolved in 30 µL diethyl-pyrocarbonate-treated distilled water. Genomic DNA was prepared from PBMCs with DNAzol reagent (Life Technologies), according to the manufacturer's protocol.Quantification of cytokine and chemokine mRNA levels by real-time PCR Reverse transcription (RT) was conducted as follows: 32 µL water containing 1 µg total RNA was added to 0.4 µg random primers (Life Technologies) and incubated at 65°C for 10 minutes. Samples were chilled on ice and cDNA was prepared with Ready-To-Go You-Prime First-Strand kit (Amersham Pharmacia Biotech, Piscataway, NJ), according to the protocol provided by the manufacturer.Real-time polymerase chain reactions (PCR) were performed using TaqMan
assay (Applied Biosystems of PerkinElmer [ABI-PE], Foster City, CA)
and PCR amplifications in ABI-PE prism 7700 sequence detection
system.46,47 Briefly, a solution of TaqMan Universal PCR
Master Mix (25 µL; ABI-PE) containing sense, antisense primer (300 nM each) and dual-labeled fluorogenic probes (100 nM) was prepared and
aliquoted into individual MicroAmp Optical Plate (ABI-PE) and 5 µL
cDNA was added to give a final volume of 50 µL. Conditions for PCR
included 50°C for 2 minutes, 95°C for 10 minutes, and 40 cycles of
95°C for 15 seconds (denaturation) and 60°C for 1 minute
(annealing/extension). Data were analyzed with Sequencer Detector
version 1.6 software (ABI-PE). Threshold cycle (CT) during the
exponential phase of amplification was determined by real-time
monitoring of fluorescent emission after cleavage of sequence-specific
probes by nuclease activity of Taq polymerase. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control gene. Primers and fluorogenic probes for GAPDH, IL-2,
IL-4, IL-10, IFN- Relationship between cytokine production and mRNA expression CD4+, CD8+, and monocyte subsets (2.5 × 106 cells) were diluted with 1 mL RPMI-1640 culture medium supplemented with 10% autologous serum and incubated for 3, 12, or 24 hours in 24-well cell culture plates at 37°C in 5% CO2. Concanavalin A (ConA; Sigma, St Louis, MO) was added at a final concentration of 0.4 µg/mL. The supernatants were collected by centrifugation (1500g, 10 minutes) and stored at 70°C until assay. The concentration of IL-10 or IFN-
(mean ± SE) was determined by enzyme-linked immunosorbent assay
(ELISA) using a polyclonal assay (PharMingen, San Diego, CA), as
previously described.48
IL-10 polymorphisms by allele-specific polymerase chain reaction An allele-specific polymerase chain reaction (ASPCR) was used to detect the G A transition polymorphism at position 1082 of IL-10
gene as previously described.49,50 There were 3 primers used for ASPCR: the 3' primer (5'-AGCAACACTCCTCGTCGCAAC-3') was combined with either the 5' primer (1082G: 5'-CCTATCCCTACTTCCCCC-3'), complementary to the IL-10 1082 G allele, or the 5' primer
(1082A: 5'-CCTATCCCTACTTCCCCT-3'), which is complementary to the
IL-101082 A allele. Primers 1082G and 1082A differ only
in their 3' terminal nucleotide. Similarly, ASPCR was used to detect
the CA transition polymorphism at position 592 of IL-10
gene.51 The 3' primer (5'-TGAGAAATAATTGGGTCCCC-3') was
combined with either the 5' primer (592C: 5'-ATCCTGTGACCCCGCCTGTC-3'),
complementary to the IL-10592 C allele, or the 5' primer
(592A: 5'-ATCCTGTGACCCCGCCTGTA-3'), which is complementary to the
IL-10592 A allele. For microsatellite typing at position
1064, the following primers were used: the sense primer was labeled
with 6-FAM fluorescent dye (IDT): sense,
5'-(6FAM)-GTCCTTCCCCAGGTAGAGCAACACTCC-3' and antisense,
5'-CTCCCAAAGCCTTAGTAGTGTTG-3'. DNA (1 µg) was PCR-amplified through
30 cycles (95°C for 30 seconds, 60°C for 1 minute, 72°C for 1 minute) with sense primers and antisense primers. PCR products (1 µL)
plus modified Genescan molecular weight makers (ABI-PE) were
sized using an ABI 377 automatic sequencer equipped with the computer
program Genotyper 2.0 software (ABI-PE).
Statistical analysis Data were analyzed by the Welch t test, Fisher exact test, or ANOVA using Statview software (SAS, Cary, NC), with P values less than .05 considered statistically significant.
Cytotoxic activities and autologous GVHD Table 1 summarizes the data from patients with biopsy-confirmed autologous GVHD, patients without clinical manifestations of autologous GVHD, and control autologous SCT patients (non-CsA treated) evaluating maximal lytic activity during the course of treatment (day 12 through day 33). Cytolytic activity against autologous PHA-blasts was significantly higher in the patients who developed GVHD compared with control patients and to patients who did not develop autologous GVHD (P < .01). These data are consistent with previous data.22,41 The cytolytic activity against the T47D cell line by posttransplantation lymphocytes of patients with cutaneous GVHD was also significantly enhanced when compared with patients without GVHD (P = .047). Significant lysis of PHA-blasts, T47D cells, and K562 cells was observed in the autologous GVHD-induced patients, evaluated at the onset of GVHD versus patients without GVHD (P < .01, P < .01, P = .034, respectively).
Gene expression of cytokines in PBMCs Autologous GVHD can be inducible in recipients of autologous stem cell transplantation by administration of CsA during the first 4 weeks after transplantation. Onset of disease can vary but can be correlated with the development of autocytolytic T cells detected in the peripheral blood, findings in accordance with a clonal expansion up to day 36. Therefore, PBMCs from the patients on the autologous GVHD induction protocol were serially monitored for cytokine/chemokine gene expression by real-time PCR during the course of treatment (day 12 through day 33) at 4 different time points (days 12, 19, 25, and 33 after transplantation) and compared with control autologous SCT patients (non-CsA treated), patients before transplantation, healthy individuals, and allogeneic SCT patients. Figure 1 summarizes the data from patients evaluated at the onset of GVHD, patients without clinical manifestations of autologous GVHD (highest levels detected), and healthy individuals evaluating cytokine mRNA levels normalized against the housekeeping gene, GAPDH. IL-10 mRNA levels in patients at the onset of autologous GVHD were 29.1-fold higher than in healthy individuals (P < .01) and 4.5-fold higher than in those who did not develop clinical manifestations of autologous GVHD. IFN-
or IL-2 mRNA levels were 3.8-fold and 2.8-fold higher compared with
healthy individuals (P < .01 or P = .089,
respectively). In contrast, IL-4 mRNA levels for all groups were
comparable. No significant differences were observed in cytokine
expression comparing 8 healthy individuals, 6 control autologous SCT
patients (non-CsA treated), and 4 patients who were evaluated
before transplantation (data not shown). Figure 2A-D summarizes the temporal analysis of
IL-10, IFN-, IL-2, and IL-4 mRNA levels for patients who developed
autologous GVHD and patients who did not develop this experimental
autoaggression syndrome. IL-10 and IFN- mRNA levels in patients with
autologous GVHD were significantly higher than the levels in patients
without GVHD at day 26 (P = .011 or
P = .039). The difference was particularly pronounced, reflecting the development of autologous GVHD and cytolytic
activity against autologous lymphocytes. Comparatively, mRNA levels for
all cytokines were remarkably elevated in 4 patients following
allogeneic SCT. The IL-2, IL-4, IL-10, and IFN- mRNA transcripts
were detected in levels 160-fold, 91-fold, 730-fold, or 260-fold higher
compared with healthy individuals, respectively (data not presented).
The temporal changes in cytolytic activity and cytokine mRNA expression
were directly assessed in PBMCs from patients with autologous GVHD
following transplantation. A representative experiment evaluating
lymphocytotoxicity and cytokine expression is shown in Figure
3. The results reveal that the ability of
PBMCs to lyse autologous PHA-blasts and the T47D tumor cells parallels the temporal relationship with IL-10 and/or IFN- gene
expression.
Gene expression of chemokines in PBMCs Chemokines have been implicated in the recruitment of inflammatory cells and modulate the function of both T cells and NK cells. Analysis of chemokine mRNA transcripts (MIP-1 and RANTES, both CCR5 ligands and IP-10, a CXCR3 ligand) in PBMCs is summarized in
Figure 4. Compared with healthy controls,
PBMCs obtained from patients at the onset of autologous GVHD exhibited
significantly higher transcript levels for MIP-1 (44-fold), and
IP-10 (53-fold) mRNA transcript levels for these chemokines persisted
during CsA treatment (P < .01 and P = .024,
respectively). Interestingly, the increase in MIP-1 and IP-10 levels
was not confined to the patients developing autologous GVHD but was
also observed in patients who were treated with CsA and IFN- but who
did not develop clinical evidence of this autoaggression syndrome and
in autologous SCT control recipients (not treated with CsA and
IFN-). RANTES mRNA levels were comparable for all groups.
Comparatively, mRNA levels for MIP-1, RANTES and IP-10 in patients
with grade IV allogeneic GVHD were markedly elevated (> 100-fold; data not presented).
Cytokine and chemokine gene expression in skin lesions of GVHD In order to confirm the pathogenic involvement of cytokines and chemokines in GVHD, gene expression was evaluated in the skin lesions of a patient with autologous GVHD. Comparison was also made to skin biopsies from patients with allogeneic GVHD by real-time PCR. Figure 5 summarizes the data evaluating cytokine and chemokine mRNA levels normalized against GAPDH. IL-10, IFN-,
MIP-1, and IP-10 mRNA transcripts were detected in the skin lesions
of both patients with autologous GVHD and allogeneic GVHD. Levels of
IL-10 and IFN- mRNA in the skin from a patient with autologous GVHD were found approximately 4-fold higher compared with the levels found
in PBMCs from healthy individuals. On the other hand, levels of
MIP-1 mRNA transcripts were pronouncedly elevated in the autologous GVHD skin lesion with levels comparable to the levels detected in skin
lesions from patients with allogeneic GVHD.
Cellular origin of cytokine transcription To identify the cellular origin of the IL-10, IFN-, IL-2, and
MIP-1 mRNA transcripts, CD4+ and CD8+ T
cells and monocytes were separated from PBMCs of 7 patients who
received CsA following autologous SCT. Gene expression was determined
in each population. The results in Table
2 reveal that IL-10 mRNA levels in the
CD4+ subset were 2.3-fold higher compared with the levels
detected in the CD8+ subset. Higher IL-10 mRNA levels in
the CD4+ subset were observed in patients who developed
autologous GVHD (patient no. 6 and no. 7). On the other hand, IFN-
mRNA levels were 42-fold higher in the CD8+ subset compared
with the levels detected in the CD4+ subset. IL-2 and
MIP-1 mRNA levels were also found to be increased in the
CD8+ subset. Higher levels of MIP-1 mRNA transcripts
were detected in both CD8+ and monocytes subsets, whereas
RANTES mRNA levels were detected principally in the
CD8+ subset.
Relationship between cytokine production and mRNA expression To quantify and correlate cytokine mRNA expression with cytokine protein production, PBMCs were isolated from a patient with autologous GVHD and from 3 control individuals. Isolated CD4+ T cells, CD8+ T cells, and monocytes were cultured separately for 3, 12, or 24 hours in the absence (nonstimulated) or presence (stimulated) of ConA. Cytokine concentrations in the culture supernatant and mRNA expression in the culture cells were measured using ELISAs or real-time PCR assays, respectively. Interestingly, IL-10 mRNA expression by PBMCs from the patient with autologous GVHD increased in the absence of any stimulus and correlated with IL-10 production assessed by ELISA. Significant IL-10 production was not detected in 3 control individuals, findings correlated with decreased IL-10 mRNA levels (Figure 6A). On the other hand, IL-10 production by CD4+ cells and monocytes from control individuals in response to ConA stimulation correlated with mRNA expression of IL-10. Similarly, IFN- production was also detected in both
CD4+ and CD8+ cells, and significantly
correlated with the levels of mRNA expression for this cytokine (Figure
6B). Analysis of cytokine protein production directly related to levels
of mRNA expression (Figure 6C-D).
Development of autologous GVHD and IL-10 polymorphisms Because of significantly elevated levels of IL-10 mRNA transcripts, polymorphisms in the IL-10 promoter were examined at positions 592, 1064, and 1082, which correlate with high IL-10 production. The association of allelic composition with the susceptibility to the induction of autologous GVHD is summarized in Table 3. Interestingly, the presence of an A nucleotide at either position 592 or 1082 correlates with low IL-10 production and a reduced incidence of autologous GVHD. Conversely, the presence of G/G at position 1082 correlates with high IL-10 production and an increased frequency of patients who develop autologous GVHD. In accordance are the results comparing IL-10 mRNA levels and the development of autologous GVHD with promoter polymorphisms (Figure 7). As revealed by analysis of variance, IL-101082 G/G homozygotes had
significantly higher levels of IL-10 mRNA transcripts in patients who
develop autologous GVHD (P < .01). Patients who expressed
the IL-10592 A/C and A/A genotypes had reduced levels of
IL-10 mRNA transcripts and a reduced incidence of developing autologous
GVHD. On the other hand, IL-101064 polymorphism did not
correlate with the development of autologous GVHD. Attempts to
correlate single nucleotide polymorphisms in other cytokine promoter
regions (tumor necrosis factor alpha [TNF-], IL-6, and IFN-)
with the induction of autologous GVHD did not reveal any significant
associations (data not shown).
The present analysis of cytokine and chemokine profiles in both
PBMCs and skin lesions reveal several new findings regarding the
posttransplantation immune response in the pathophysiology of human
autologous GVHD. The most pronounced and perhaps unexpected finding is
that IL-10 mRNA levels in PBMCs from patients with autologous GVHD were
29-fold higher compared with healthy individuals. IFN- Given the fact that IL-10 has powerful immunoregulatory potential, the
most surprising finding of the current studies is that mRNA transcripts
for this cytokine were found to be markedly elevated during the onset
of autologous GVHD. One simple explanation for the findings of the
current studies is that there is a regulatory response that ensues with
the onset of autoaggression. The increased IL-10 mRNA levels may
reflect a compensatory response correlating with the induction of
autologous GVHD. However, the onset of autocytolytic activity exhibited
an identical temporal relationship with elevated expression of IL-10
(and IFN- Westendrop et al48 demonstrated that genetic factors could
account for 74% of differences in IL-10 production. High IL-10 production may be a genetic risk factor for disease
susceptibility.57,58 Single nucleotide polymorphisms at
IL-10 is thought to be a powerful regulatory T-helper type 2 (TH-2) cell cytokine produced by lymphocytes and monocytes that limits alloreactivity including GVHD in vivo, ostensibly inhibiting macrophage/monocyte and T-cell replication and secretion of inflammatory cytokines.62,63 The results from the present studies suggest that IL-10 may have immunostimulatory properties, particularly important in the stem cell transplantation setting. Several recent studies suggest that CsA inhibits thymic-dependent clonal deletion leading to the release of autoreactive T cells.64-66 Furthermore, CsA reduces MHC class II antigen expression in the thymus, resulting in a failure of mature T cells to recognize this Ag as self.11,67 In this setting, IL-10 may act synergistically, inducing MHC class II Ag down-regulation on thymic epithelial cells facilitating a failure to clonally delete autoreactive T cells. In support of this hypothesis is the recent finding indicating that IL-10 enhances the proliferative response of thymocytes.68 Thus, IL-10 may enhance the proliferative capacity of autoreactive T cells prematurely released from the thymus. Alternatively, IL-10 can augment the proliferative response of IL-2- and/or IL-4-activated T cells.31 In accord are the results indicating that IL-10 increases the IL-2-stimulated cytolytic activity of CD8+ T cells.69 Interestingly, this immunoregulatory cytokine can prevent priming of autoreactive T cells, but IL-10 cannot suppress ongoing immune responses.30,70 This cytokine can even promote IL-2-independent growth of activated CD8+ T cells in the absence of costimulatory signals mediated by APCs.30 Both CD4+ and CD8+ T cells are required for the development of autologous GVHD.15 A vigorous cellular immune reaction, dominated by activated proliferating CD8+ cytotoxic T lymphocytes, is mounted during both autologous GVHD15,22 and allogeneic GVHD.71,72 It is not yet clear from functional studies whether most of these activated T cells are Ag-specific expanded during GVHD or result from additional cytokine-induced activation. If activation of CD8+ effector cells by IL-10 is pivotal in the transplantation setting, such activation is most likely to account for the failure to suppress allogeneic GVHD by blocking costimulation during development of GVHD.73 Blazar and colleagues32 provide substantial evidence that
IL-10 can exacerbate GVHD presumably mediated by enhanced IFN- Elucidating the interactive cytokine cascade in autologous GVHD
should provide novel insights into the immunobiology of this unique
autoaggression syndrome. Based on the results of the current studies it
appears that secretion of IL-10 by CD4+ cells and monocytes
leads to the activation and proliferation of IL-2-driven
CD8+ autoreactive T cells into IFN-
Expression of chemokines such as MIP-1 Clearly, the regulation and expression of IL-10 appears to play an unexpected and critical role in the susceptibility to and intensity of autologous GVHD. Further understanding of these cytokine/chemokine pathways should promote more effective immunotherapeutic strategies for enhancing the antitumor efficacy of autologous GVHD. Moreover, polymorphic analysis may allow prediction of individual patient susceptibility to the induction of autologous GVHD and provide a possible solution (administration of IL-10) for these patients with resistant alleles.
Submitted January 28, 2002; accepted May 27, 2002.
Prepublished online as Blood First Edition Paper, June 7, 2002; DOI 10.1182/blood-2002-01-0176.
Supported by grants CA 15396, CA 82583, and AI 24319 from the National Institutes of Health, grant USAMRMC DAMD 17-99-1-9238 from the Department of Defense, and a grant from the Avon Foundation.
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: Allan D. Hess, Johns Hopkins University School of Medicine, Oncology Center, Bunting-Blaustein Cancer Research Building, 1650 Orleans St, Room 484, Baltimore, MD 21231; e-mail:adhess{at}jhmi.edu.
1.
Armitage JO.
Bone marrow transplantation.
N Engl J Med.
1994;330:827-838 2. Philip T, Guglielmi C, Hagenbeek A, et al. Autologous bone marrow transplantation as compared with salvage chemotherapy in relapses of chemotherapy-sensitive non-Hodgkin's lymphoma. N Engl J Med. 1995;333:1540-1545 | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
may be critical
mediators for the development of autologous GVHD
Top
Abstract
Introduction
-
