Biology of Blood and Marrow Transplantation 9:292-303 (2003) 䊚 2003 American Society for Blood and Marrow Transplantation 1083-8791/03/0905-0002$30.00/0 doi:10.1016-S1083-8791(03)0087-9
Role of Tumor Necrosis Factor-␣ in Graft-versus-Host Disease and Graft-versus-Leukemia Responses Robert Korngold1, Joseph C. Marini,1 Monica E. de Baca,2 George F. Murphy,2 Jill Giles-Komar3 1
Kimmel Cancer Institute and 2the Department of Pathology, Jefferson Medical College, Philadelphia, Pennsylvania; 3Research and Development, Centocor, Inc, Malvern, Pennsylvania Correspondence and reprint requests: Robert Korngold, PhD, Kimmel Cancer Institute, Jefferson Medical College, 233 South 10th Street, Philadelphia, PA 19107 (e-mail:
[email protected]). Received January 22, 2003; accepted March 3, 2003
ABSTRACT Tumor necrosis factor–alpha (TNF-␣) antagonist therapy has proven effective in inflammatory conditions such as rheumatoid arthritis and Crohn’s disease. There is substantial evidence that TNF-␣ also plays a role in the development of graft-versus-host disease (GVHD) after allogeneic hematopoietic cell transplantation, which along with leukemia relapse remains one of the 2 major impediments to success of the approach. Using a recently developed potent rat/mouse chimeric monoclonal antibody directed against murine TNF-␣ (CNTO2213), the authors investigated the effect of TNF-␣ blockade on GVHD mediated by either CD4ⴙ or CD8ⴙ donor T cells. The results indicated that the treatment had only a moderate effect on both a CD8ⴙ T cell–mediated major histocompatibility complex–matched GVHD model involving multiple minor histocompatibility antigens and a p3 F1 acute GVHD model directed against a haplo-mismatched major histocompatibility complex barrier involving both CD4ⴙ and CD8ⴙ T cells. In contrast, treatment with the anti–TNF-␣ antibody had a highly significant effect (100% survival rate) on the CD4ⴙ T cell–mediated component of this latter model. Importantly, anti–TNF-␣ antibody did not block the development of a graft-versus-leukemia effect against a murine myeloid leukemia challenge in either a syngeneic or allogeneic p3 F1 setting. This suggests that the inhibition of TNF-␣ during allogeneic hematopoietic cell transplantation may be able to diminish the inflammatory GVHD reaction without hindering effective graft-versus-leukemia responses. © 2003 American Society for Blood and Marrow Transplantation
KEY WORDS Allogeneic hematopoietic cell transplantation versus-leukemia (GVL) ● Anti–TNF-␣
INTRODUCTION Allogeneic hematopoietic cell transplantation (HCT), using bone marrow, umbilical cord blood, or mobilized peripheral blood as the source of stem cells is a clinical treatment for several malignant and nonmalignant diseases, including acute and chronic leukemias, myelomas, and severe immunodeficiencies [1-3]. In those patients being treated for leukemia, the success of HCT is significantly limited by the complications of graft-versus-host disease (GVHD) and leukemia relapse. Depletion of mature T cells from the donor inoculum reduces the incidence and severity of acute GVHD. However, T cell depletion also increases the risk of leukemia relapse, suggesting that alloreactive donor T cells may mediate both GVHD and the graft-versus-leukemia (GVL) effect [4-6]. Thus, the development of approaches to separate 292
●
Graft-versus-host disease (GVHD)
●
Graft-
GVL responses from GVHD immunopathogenesis is a major focus of research in the allogeneic HCT field. It is now well accepted that a cascade of cytokines is involved in the induction and maintenance of GVHD [7]. The inflammatory cytokine, tumor necrosis factor–alpha (TNF-␣) can be involved in at least 3 phases of acute GVHD. First, during intensive transplant conditioning, host endothelium and epithelium in the intestine, skin, and liver can be damaged. This damage can lead to activated host cells that secrete TNF-␣ and interleukin (IL)-1. These cytokines, in turn, can upregulate major histocompatibility complex (MHC) antigens and adhesion molecules that will enhance the recognition of host MHC or minor histocompatibility antigens (miHA) by mature donor T cells present in the donor HCT inoculum [8,9]. In the second phase, the activated donor T cells begin to differentiate and proliferate [7]. As they do,
TNF-␣ in GVHD
cytokines (predominantly the TH1-related interferon gamma [IFN-␥] and IL-2) act to amplify the response even further by inducing cytotoxic T lymphocyte (CTL) and natural killer (NK) cell responses and by priming donor and residual host macrophages to produce TNF-␣. In turn, TNF-␣ released from both T cells and activated macrophages induce the production of chemokines, such as macrophage inflammatory protein–1 and monocyte chemoattractant protein–1, that function to recruit cells to inflammatory sites. TNF-␣ also upregulates the expression of intercellular adhesion molecules–1 on vascular endothelium, which enhances the ability of T cells and other leukocytes to extravasate into GVHD target tissues [10,11]. During the third phase of acute GVHD, massive tissue destruction is found but is not only mediated by CTL and NK cells. Monocytes may encounter endotoxin (lipopolysaccharide) that has leaked through the conditioning regimen– damaged intestinal wall and may be triggered to secrete large amounts of TNF-␣. TNF-␣ can then directly contribute to tissue damage by inducing apoptosis, or even necrosis, of target cells. TNF-␣ may synergize with the cellular damage caused by CTL and NK cells and further amplify tissue destruction and the inflammatory response [10,12]. These combined pathologic effects could then result in the clinical symptoms of GVHD. Anti-TNF therapies have proved successful in the treatment of inflammatory diseases, such as rheumatoid arthritis [13] and Crohn’s disease [14]. These approaches seek to bind TNF-␣ or to block the interaction of TNF-␣ with its receptors thereby inhibiting its biological effects. In this regard, infliximab, a mouse/human Fc chimeric IgG1 monoclonal antibody (mAb) specific for human TNF-␣, has exhibited clinical efficacy as an inflammatory inhibitor [15], and its applicability to the treatment or prevention of GVHD currently is under consideration. Early evidence from clinical studies in HCT indicated that the systemic release of TNF-␣ during pretransplant conditioning correlated with poor outcomes [16]. In other studies examining the posttransplant period, it was found that TNF-␣ significantly increased in the second to third week, and the high levels correlated with developing acute GVHD as well as with other acute endothelial complications [17-20]. Based on these findings, clinical trials were initiated with anti–TNF-␣&rtf-space; mAb posttransplant and resulted in some success in the inhibition of GVHD pathology, particularly in the skin and gut, but overall they were limited by shortterm administration and problems with the immunogenicity of the reagents [21-23]. Holler et al. [24] also tested prophylactic neutralization of TNF-␣ during the course of pretransplant conditioning and found
BB&MT
that it could reduce the levels of TNF-␣, and subsequently the severity and onset of acute GVHD. Over the last 15 years, the effect of anti–TNF-␣ antibodies on development of GVHD in experimental models has been somewhat varied, partially because of the varied reagents involved in the studies. Piguet et al. [25] were the first to show the involvement of TNF-␣ in GVHD, using a P3 F1 model of MHCincompatible HCT. Lethally irradiated mice were injected with T cell– depleted bone marrow and lymph node T cells to induce GVHD, and it was found that a weekly treatment with a neutralizing rabbit antimouse recombinant TNF-␣ IgG polyclonal antibody reduced the incidence of skin and gut lesions associated with GVHD and decreased mortality rate. Other groups have confirmed these observations by administering various types of anti–TNF-␣ antibody preparations or inhibitors, starting either before HCT or within the first week after transplant [26-32]. However, in some models, TNF-␣ blockade did not seem to have much of an effect [33]. In the current study, we have sought to investigate the potential of a recently generated rat/mouse Fc chimeric IgG2a mAb (CNTO2213) equivalent of infliximab, directed against murine TNF-␣, to inhibit GVHD. Of particular interest, we wanted to test its effectiveness in both CD4- and CD8-mediated GVHD development and whether it was capable of inhibiting GVHD while retaining a GVL effect against a myeloid leukemia challenge. The results indicated that anti–TNF-␣ mAb treatment from day ⫺1 to day 12 post-HCT had a moderate effect in a CD8mediated GVHD model involving multiple miHA differences and an equally moderate effect in a model directed across an MHC barrier when both CD4⫹ and CD8⫹ donor T cells were transplanted. However, the anti–TNF-␣ mAb treatment had a highly significant effect (100% survival rate) on the isolated CD4-mediated GVHD component in this latter MHC disparate model. In addition, treatment with anti–TNF-␣ mAb within the first 3 weeks of a syngeneic or allogeneic donor transplant of bone marrow and T cells did not block the development of a GVL effect against a myeloid leukemia challenge. These results suggest that controlling the levels of TNF-␣ during allogeneic HCT may be able to limit the development of GVHD while still allowing the generation of GVL responses.
MATERIALS AND METHODS Mice
Male C3H/HeJ (C3H; H2k), (B6xC3H)F1 (H2bxk), (B6xDBA/2)F1 (H2bxd), C57BL/6 (B6; H2b), and CBA/JCr (CBA; H2k) mice were purchased from the National Cancer Institute Research and Development Center (Frederick, MD). B10.BR (H2k) mice 293
R. Korngold et al.
were purchased from the Jackson Laboratories (Bar Harbour, ME). Mice used for experiments were between 6 and 10 weeks of age and were housed in sterile microisolator cages within a specific pathogen-free facility, receiving autoclaved food and water ad libitum. Cell Line
MMB3.19, a c-myc–transformed myeloid leukemia line, was originally cloned from the ascites of a B6 mouse that had been injected with a myc-encoding Moloney murine leukemia virus construct as previously described [34,35]. Media
Phosphate-buffered saline (PBS) supplemented with 0.1% bovine serum albumin (BSA) (Sigma Chemical Co, St Louis, MO) was used for all in vitro manipulations of the donor bone marrow and lymphocytes. Immediately before injection, the cells were washed and resuspended in PBS alone. For maintaining cell lines and for in vitro assays, RPMI 1640 medium (Mediatech, Herndon, VA) was used, supplemented with 10% fetal bovine serum (FBS; GIBCO, Grand Island, NY), 2 mmol/L L-glutamine, 50 IU/mL penicillin,and 50 g/mL streptomycin. Antibodies
The CNTO2213 mAb, a rat/mouse Fc chimeric IgG2a construct with rat (Fab)2 units specific for murine TNF-␣ [36] (tissue culture derived and purified with protein A) and its isotype control CNT0412, a human anti-CD4 mAb, were provided by Centocor (Malvern, PA). Ascites fluid containing mAb (5 to 10 mg/mL) were generated from hybridoma lines specific for Thy-1.2 (J1j; ATTC TIB-184), CD4 (RL172; [37]), or CD8 (3.168; [38]) proteins, and were used for the preparation of cellular grafts. Affinity-purified goat antimouse IgG (Cappel, Cosa Mesa, CA) was used for B cell depletion. Guinea pig complement was purchased from Rockland Immunochemicals (Gilbertsville, PA). Anti-CD3, anti-CD4, anti-CD8, antiB220, and isotype control mAb, all coupled to phycoerythrin, were all purchased from BD Biosciences (Palo Alto, CA). Bone Marrow Transplantation
As previously described [39], bone marrow was harvested from the tibia and femurs of donor mice by flushing with PBS containing 0.01% BSA (PBS/BSA). Bone marrow cells were depleted of T cells using an anti-Thy 1.2 mAb (J1j; ATCC) at a 1:100 dilution and guinea pig complement (Rockland Immunochemicals, Gilbertsville, PA) at a dilution of 1:6 for 45 minutes at 37°C. Lymphocytes were isolated from spleens and lymph nodes of donor mice. Splenocytes were treated 294
with Gey’s balanced salt lysing solution containing 0.7% ammonium chloride (NH4Cl) to remove red blood cells. After red blood cell depletion, spleen and lymph node cells were pooled and depleted of B cells by panning on a plastic petri dish, precoated with a 5 g/mL dilution of goat anti-mouse IgG for 1 hour at 4°C. These treatments resulted in donor populations of 90% to 95% CD3⫹ cells, as quantitated by fluorescent flow cytometry. T cells subsets were then isolated via negative selection using either anti-CD8 (3.168, 1:50 dilution) or anti-CD4 mAb (RL172, 1:100 dilution) and complement. These treatments reduced the targeted T cell subset populations to background levels as determined by flow cytometric analysis. Recipient mice were exposed to 13 Gy whole body irradiation from a Mark-I-68A 137Cs source (JL Shepherd, San Fernando, CA) at 1.43 Gy/min, delivered in a split dose of 6.5 Gy each, separated by 3 hours. These mice were then transplanted with 2 ⫻ 106 anti-Thy 1.2–treated bone marrow cells (ATBM; T cell-depleted) along with the indicated number of appropriate T cells (donor CD4 or CD8 enriched T cells), intravenously (iv) via the tail vein. Mice were treated with CNTO2213 anti–TNF-␣ or isotype control CNTO412 mAbs (1 mg; intraperitoneally [ip]) one day before transplantation and again on days 0, 4, 8, and 12 (all at 0.5 mg; ip). For GVL experiments, B6 recipient mice were challenged with an injection of MMB3.19 cells (1 ⫻ 105in 0.5 mL PBS; ip) one day before transplantation of donor ATBM and T cells, with a similar schedule of anti–TNF-␣ mAb treatment. In both GVHD and GVL experiments, the mice were checked daily for morbidity and mortality until completion. As indicated, the data were pooled from 2 to 3 separate experiments, and median survival times (MST) were determined as the interpolated 50% survival point of a linear regression through all of the day of death data points and including zero. Statistical comparisons for survival between experimental groups were performed by the nonparametric Wilcoxon signed rank test. Significance for weight comparisons was determined by the t test at individual time-points. Flow Cytometry
Appropriate mAbs in volumes of 25 L were incubated with 2 to 5 ⫻ 105 cells in the wells of a 96-well U-bottom microplate at 4°C for 30 minutes, centrifuged at 1500 rpm for 3 minutes, and washed with PBS containing 0.1% BSA and 0.01% sodium azide (wash buffer). The fluorescence analysis was performed on an EPICS Profile II analyzer (Coulter, Hialeah, FL) in the Kimmel Cancer Institute Flow Cytometry Facility, at Thomas Jeffeson University in Philadelphia. The percentage of positive cells and the
TNF-␣ in GVHD
arithmetic mean fluorescence intensity were calculated for each sample. Serum Cytokine Measurements
Blood was collected via the tail vein, allowed to clot overnight at 4°C, and centrifuged at 10,000 rpm for 30 minutes at 4°C. Serum was collected and pooled from groups of 5 mice in each treatment group. The Cytometric Bead Array (BD Biosciences) was used to measure serum levels of TNF-␣, IFN-␥, and IL-5. Briefly, polystyrene beads, dyed to emit unique fluorescent intensities and coupled to specific mAb against each of the 3 cytokines, served as capture reagents. Cytokines were detected simultaneously in a mixture via direct immunoassay containing 50 l of test sample, 10 l of each mouse cytokine capture mAb-bead reagent and 50 l of cytokine-specific phycoerythrin -coupled mAb used as a detection reagent. Standard curves were obtained using sets of calibrators ranging from 0 to 5000 pg/mL for each cytokine in the mixture. The mixtures were incubated for 2 hours in the dark at room temperature and washed to remove unbound phycoerythrin -detection reagent before data acquisition using flow cytometry. The concentration of IL-6 in serum was determined via a commercial ELISA kit (R & D Systems, Inc, Minneapolis, MN), calibrated using an IL-6 standard provided in the kit, and performed on triplicate samples.
ated GVHD, the MHC-matched, miHA-disparate B10.BR3 CBA GVHD model was utilized with a well established etiology [41]. CBA mice were irradiated lethally (13 Gy, split dose) and transplanted with B10.BR ATBM cells (2 ⫻ 106), alone, or in addition to a highly enriched population (95%) of CD8⫹ T cells (3 ⫻ 106). Mice were left either untreated or treated with the isotype-matched control CNTO412 or anti– TNF-␣ mAbs on day ⫺1 (1 mg, ip) and days 0, 4, 8, and 12 of transplant (0.5 mg; ip). Whereas all recipients of ATBM cells alone survived for at least 70 days, as expected, mice transplanted with donor T cells and left untreated or treated with control CNTO412 mAb all succumbed to GVHD with similar MST values of 21.5 and 29.1 days, respectively (P ⬍ 0.59; Figure 1A). In contrast, CBA recipients of donor T cells that were administered anti–TNF-␣ mAb, exhibited a 42.9% survival rate with an MST of 48.4 days (P ⬍ 0.018 in comparison with the CNTO412 control group). In
Pathologic Analysis
Full-thickness ear biopsies were sampled from each mouse of the various treatment groups and immediately placed in 10% phosphate buffered formalin and fixed for at least 24 hours before processing. The samples were then dehydrated through graded alcohols and xylene and embedded in paraffin. Sections were cut at 6 m and stained with H & E. Histopathologic examination was performed using a standard light microscope with a 40X objective and a 10X ocular lens. The analysis was performed under blinded conditions as to treatment groups. The mean number of dyskeratotic epidermal cells per linear millimeter was determined, as previously [40], by counting the dyskeratotic cells (ie, pyknotic nucleus surrounded by an eosinophilic cytoplasm) and total epidermal length in millimeters (minimum 10 mm per tissue section), measured for 3 to 4 separate tissue sections per animal and then averaged. The mean value was then calculated from these averages with 3 animals per group. RESULTS Effect of Anti–TNF-␣ mAb on CD8 T Cell-Mediated GVHD
To determine if anti–TNF-␣ mAb treatment could affect the development of CD8⫹ T cell–medi-
BB&MT
Figure 1. Effect of anti–TNF-␣ mAb on CD8 T cell–mediated GVHD. CBA mice were lethally irradiated (13 Gy, split dose) and transplanted with B10.BR ATBM cells (2 ⫻ 106), alone or in addition to a highly enriched population (95%) of CD8⫹ T cells (3 ⫻ 106). Mice were either left untreated, treated with control CNTO412 mAb, or the anti–TNF-␣ mAb on day ⫺1 (1 mg, ip) and days 0, 4, 8, and 12 of transplant (0.5 mg; ip). A, percent survival posttransplant. MST values (days) were untreated (21.5; n ⫽ 3); CNTO412 (29.1; n ⫽ 7); and anti–TNF-␣ mAb (48.4; n ⫽ 8; P ⬍ .018 v CNTO412). Results represent pooled data from 2 similar experiments. B, mean ⫾ SEM % initial body weight; pooled data (same as in A). 295
R. Korngold et al.
addition, surviving anti–TNF-␣ mAb-treated mice did not display evident symptoms of GVHD (eg, ruffled fur, skin lesions, hunched posture, or diarrhea), and their body weights were at a relatively constant level ranging 5% to 12% below that of the control ATBM transplanted group (with no significant difference; P⬍ .17, on days 6 to 32, and P ⫽ .05 on day 49; Figure 1B). Mice that did develop fatal GVHD in the presence of anti–TNF-␣ mAb seemed to do so with slower kinetics than the untreated or CNTO412treated groups. Effect of Anti–TNF-␣ mAb on GVHD across an MHC Barrier
The haploidentical C3H3(B6xC3H)F1 mouse model was utilized to determine whether the neutralization of TNF-␣ by mAb treatment could affect the course of GVHD across a full MHC barrier. C3H T cells (both CD4⫹ and CD8⫹; 5 ⫻ 106) and ATBM cells (2 ⫻x 106) were transplanted iv into lethally irradiated (13 Gy, split dose) (B6xC3H)F1 mice, which induced a rapid acute GVHD response characterized by severe weight loss and early fatality (MST of 5.3 days; Figures 2A and B). Similar results were obtained in recipients treated with control CNTO412 mAb, but those mice treated with anti–TNF-␣ mAb (1 mg ip on day ⫺1 and 0.5 mg on days 0, 4, 8, and 12) exhibited a 40% long-term survival rate with an MST of 40.9 days (P ⱕ .001 compared with either untreated or the CNTO412 control groups). In terms of weight loss, after an initial slight drop in the first few days caused by the irradiation conditioning, the control ATBM mice steadily gained weight throughout the remainder of the experiment (Figure 2B). However, the untreated and CNTO412treated groups transplanted with donor T cells never recovered from the initial drop and instead continued to rapidly lose weight until their death, consistent with severe GVHD. However, the anti–TNF-␣ mAbtreated mice began to recover by day 9, and surviving animals after day 37 continued to gain weight during the remaining course of the experiment, tracking approximately 6% to 12% below the ATBM group (P ⱕ .02 at all time points). Effect of Anti–TNF-␣ mAb on CD4ⴙ T-Cell–Mediated GVHD
Because donor CD4⫹ T cell responses tend to dominate the development of GVHD in the C3H3(B6xC3H)F1model [35], and in light of the initial observation of a moderate effect of anti–TNF-␣ mAb treatment when a complete donor T cell inoculum was transplanted, we focused our attention on the CD4-mediated GVHD component. The injection of 3 ⫻ 106 C3H CD4⫹ T cells together with 2 ⫻ 106 ATBM cells into irradiated (13 Gy, split dose) 296
Figure 2. Effect of anti–TNF-␣ mAb on GVHD across an MHC barrier. (B6xC3H)F1 mice were lethally irradiated (13 Gy, split dose) and transplanted iv with C3H ATBM cells (2 ⫻ 106), alone, or in addition to T cells (5 ⫻ 106). Mice were either left untreated, treated with control CNTO412 mAb, or the anti–TNF-␣ mAb on day ⫺1 (1 mg, ip) and days 0, 4, 8, and 12 of transplant (0.5 mg; ip). A, percent survival posttransplant. MST values (days) were untreated (5.3); CNTO412 (5.8); and anti–TNF-␣ mAb (40.9; P ⱕ .001 v CNTO412 or untreated). Results represent pooled data (n ⫽ 15 per group) from 3 similar experiments. B, mean ⫾ SEM % initial body weight; pooled data (same as in A).
(B6xC3H)F1 mice resulted in the majority of the untreated (75%; MST of 26.5 days) and control CNTO412-treated (81.3%; MST of 13.5 days) mice succumbing to severe acute GVHD (Figure 3A). In contrast, 100% of the mice treated with the anti– TNF-␣ mAb (1 mg ip on day ⫺1 and 0.5 mg on days 0, 4, 8, and 12) survived beyond 60 days (P ⬍ .001). These mice did not exhibit any visible symptoms of GVHD and rapidly recovered from their initial body weight loss after irradiation and continued to gain weight until the end of the experiment in parallel to the ATBM control group (no significant difference by day 58; P ⱖ .12; Figure 3B). The highly significant effect of anti–TNF-␣ mAb treatment on survival in the CD4-mediated GVHD suggested that the more modest effect observed previously with transfer of a whole donor T cell inoculum (Figure 1A) was likely owing to less inhibition of CD8-mediated anti-MHC
TNF-␣ in GVHD
lial tissue between the CNTO412-treated and the anti–TNF-␣; mAb-treated groups (0.95 v 0.16 cells per millimeter, respectively; P ⬍ .001; Figure 4). Representative epithelial tissue samples from these 2 groups showed the discord in dyskeratotic cell numbers (Figures 5A and B), with almost none observed in any given field of the anti–TNF-␣; mAb-treated tissue. Also notable was that there was no significant difference between the anti–TNF-␣; mAb-treated and the control ATBM groups on day 6 (0.16 v 0.11 dyskeratotic cells per millimetre; P ⱖ .11), and this lack of significance persisted on days 15 and 30 posttransplantation (P ⱖ .08; Figure 4). On day 45, although both groups were still declining in number, the anti–TNF-␣ mAb-treated group was found to be significantly higher than the ATBM control (P ⱕ .001). No comparison could be made with the control CNTO412-treated group at these later time-points because those mice had all succumbed to GVHD before day 15. Serum Cytokine Concentrations after Anti–TNF-␣ mAb Treatment
Figure 3. Effect of anti–TNF-␣ mAb on CD4⫹ T cell–mediated GVHD. (B6xC3H)F1 mice were lethally irradiated (13 Gy, split dose) and transplanted iv with C3H ATBM cells (2 ⫻ 106), alone, or in addition to a highly enriched population (⬎98%) of CD4⫹ T cells (5 ⫻ 106). Mice were left untreated or treated with control CNTO412 or the anti–TNF-␣ mAb on day ⫺1 (1 mg, ip) and days 0, 4, 8, and 12 of transplant (0.5 mg; ip). A, percent survival posttransplant. MST values (days) were untreated (26.5); CNTO412 (13.5); and anti–TNF-␣ mAb (⬎60; P ⬍ .001 v CNTO412 or untreated). Results represent pooled data (n ⫽ 15 per group) from 2 similar experiments. B, mean ⫾ SEM % initial body weight; pooled data (same as in A).
The concentrations of TNF-␣, IFN-␥, IL-5 and IL-6 were measured in the pooled sera of mice at various time points (days ⫺1 to 11) during the course of GVHD in the CD4 T cell–mediated C3H3(B6xC3H)F1 model. Mice were treated with anti–TNF-␣ or CNTO412 mAbs, as per the schedule used previously, with 1 mg ip on day ⫺1 and 0.5 mg on days 0, 4, and 8 (the only difference being the lack of injection on day 12). Anti-TNF mAb-treated mice exhibited a significant reduction in circulating TNF-␣ on days 6 and 7 (76 and 83 pg/mL, respectively)
class I responses. However, this could not be tested directly in this model, because purified C3H CD8⫹ T cells were unable to mediate lethal GVHD on their own without the presence of CD4 T cells. Pathologic Analysis after Anti–TNF-␣ mAb Treatment
Sequential pathologic samples were prepared from the ear skin of mice in the CD4⫹ T cell–mediated C3H3(B6xC3H)F1 model, treated with anti–TNF-␣; or CNTO412 mAbs, as above. Dyskeratotic cells (apoptotic keratinocytes) are a useful diagnostic measure of the severity of GVHD in that their increased presence in epithelial layers correlates well with the severity of disease [40]. In the initial stage of GVHD (day 6), there was a significant difference in the mean number of dyskeratotic cells per millimeter of epithe-
BB&MT
Figure 4. Summary of cutaneous GVHD pathologic effects after anti–TNF-␣ mAb treatment. Sequential pathologic samples were prepared from the ear skin of mice in the CD4⫹ T cell–-mediated C3H3(B6xC3H)F1 model, treated with anti–TNF-␣ or CNTO412 mAb, as described in Figure 3. The mean ⫾ SEM number of dyskeratotic cells per millimeter of epithelial tissue were calculated as described in Materials and Methods for 3 samples each group. 297
R. Korngold et al.
compared with the control CNTO412-treated group (224 and 365 pg/mL, respectively; Figure 6A). TNF levels had been decreased to nearly background levels on these days when compared with the control ATBM group and remained that way even on day 11 posttransplantation, by which time all CNTO412-treated mice had died. There was also a significant reduction in the levels of circulating IFN-␥ in the anti–TNF-␣ mAb-treated group (ranging from 55 to 70 pg/mL) as seen on days 5 through 7 posttransplantation, compared with 154 to 331 pg/mL observed in the CNTO412-treated group (Figure 6B). IFN-␥ levels were, however, still highly increased over that seen in the control ATBM mice. In contrast, IL-6 levels were only slightly decreased in comparison with the CNTO412-treated group on days 6 and 7 of GVHD but were clearly elevated (370 to 390 pg/mL) above the background levels (75 to 80 pg/mL) obtained in the ATBM mice (Figure 6C). IL-5 levels did not seem to be affected by treatment with anti–TNF-␣ mAb and were equivalent to the elevated levels (44 to 52 pg/mL) found in the CNTO412-treated mice (Figure 6D).
Figure 6. Serum cytokine concentrations after anti–TNF-␣ mAb treatment. The concentrations of TNF-␣, IFN-␥, IL-5, and IL-6 were measured in the pooled sera of mice at various time points (days ⫺1 to 11) during the course of GVHD in the CD4 T cell–mediated C3H3(B6xC3H)F1 model. Mice were treated with anti–TNF-␣ or CNTO412 mAbs; 1 mg ip on day ⫺1 and 0.5 mg on days 0, 4, and 8. Serum from 5 mice per group were pooled and analyzed for the following cytokines: A, TNF-␣ levels; B, IFN-␥; C, IL-6; D, IL-5. ■, ATBM; , CNT0412; , Anti-TNF mAb.
Effect of Anti–TNF-␣ mAb Treatment on Syngeneic GVL Activity
Figure 5. Histologic identification of apoptotic target epidermal cells in murine ear skin 6 days posttransplant. Tissue was sampled from mice described in Figure 3. As compared with CNTO412 mAb-treated recipients of C3H CD4⫹ T cells (A; arrows indicate apoptotic basal cells), skin of anti–TNF-␣; mAb-treated recipients (B) contained only rare apoptotic cells and resembled skin of ATBM alone recipients. (Original magnification x40.) 298
The effectiveness of the anti–TNF-␣ mAb treatment in CD4-mediated GVHD raised the question of whether the neutralization of this inflammatory cytokine would also diminish the capacity of donor T cells to mediate a GVL response. To address this issue, we initially investigated a syngeneic B6 GVL bone marrow transplant model involving a mouse myeloid leukemia line (MMB3.19) that had been previously used in our laboratory [35]. B6 mice were irradiated (10 Gy) and transplanted with B6 ATBM cells (2 ⫻ 106) alone, or in combination with 2 ⫻ 107 CD4⫹ T cells from donor B6 mice that had been presensitized to MMB3.19 (ip immunization with 30 Gy–treated tumor cells 3 weeks before transplantation). These mice were then challenged with an ip injection of 7.5 – 104 MMB3.19 leukemia cells on day one. Next, mice were either left untreated or treated with anti–TNF-␣ or control CNTO412 mAbs (1 mg on day ⫺1 ip; 0.5 mg on days 0 to 24, every 4th day). We extended the period of treatment to ensure that it would overlap with the GVL response. Mice given only ATBM and challenged with the MMB3.19 cells all succumbed to tumor burden with most fatalities occurring by day 21 posttransplantation with an MST of 16.7 days (Figure 7). However, mice given the donor CD4⫹ T cell inoculum and left untreated or treated with
TNF-␣ in GVHD
Figure 7. Effect of anti–TNF-␣ mAb treatment on syngeneic GVL activity. B6 mice were irradiated (10 Gy) and transplanted with B6 ATBM cells (2 ⫻ 106) alone, or in combination with 2 ⫻ 107 CD4⫹ T cells from donor B6 mice that had been presensitized to MMB3.19. These mice were then challenged with an ip injection of 7.5 ⫻ 104 MMB3.19 cells on day one. Mice were either left untreated or treated with anti–TNF-␣ or control CNTO412 mAbs (1mg on day ⫺1 ip; 0.5 mg on days 0 to 24, every 4th day). MST values (days) were: ATBM plus MMB3.19 (16.7); ATBM plus CD4⫹ plus MMB3.19 (25.3); ATBM plus CD4⫹ plus MMB3.19 plus CNTO412 (25.2); ATBM plus CD4⫹ plus MMB3.19 plus anti–TNF-␣ mAb (23.8; P ⱕ .029 v ATBM plus MMB3.19; P ⬎ .72 v CNTO412). The results represent pooled data (n ⫽ 16 per group) from 3 similar experiments.
CNTO412 mAb were able to mediate a GVL effect and prolong survival with MSTs of 25.3 and 25.2 days, respectively. Of most importance, mice treated with anti–TNF-␣ mAb displayed a significant increase in survival time because of their syngeneic GVL capacity compared with the ATBM plus MMB3.19 control group (P ⬍ .03), with an MST of 23.8 days, equivalent to that observed in the untreated (P ⬍ .60) and CNTO412-treated (P ⬎ 0.72) CD4 T cell groups. Thus, the same anti–TNF-␣ mAb treatment that effectively blocked the development of CD4⫹ T cell– mediated GVHD (as shown above) did not seem to affect the capacity of CD4⫹ T cells to mediate antimyeloid leukemia responses in an environment that lacked alloreactivity. Effect of Anti–TNF-␣ mAb Treatment on Allogeneic GVL Activity
The most important question was whether the CNTO2213 anti–TNF-␣ mAb treatment in the CD4-mediated GVHD model could block disease development while still being permissive for the GVL response in the allogeneic setting. As in previous experiments, (B6xC3H)F1 mice were irradiated (13 Gy, split dose) and transplanted with C3H ATBM cells (2 ⫻ 106) alone, or in combination with 3 ⫻ 106 naı¨ve
BB&MT
C3H CD4⫹ T cells. These mice were then challenged with an ip injection of 7.5 ⫻ 104 MMB3.19 leukemia cells (completely allogeneic to the donor C3H cells) on day one, and mice were either left untreated or treated with anti–TNF-␣ or control CNTO412 mAbs (1mg on day ⫺1 ip; 0.5 mg on days 0 through 24, every 4th day). All of the mice given only ATBM and challenged with the MMB3.19 cells succumbed to tumor burden, with an MST of 24.2 days (Figure 8). Mice transplanted with ATBM and donor CD4⫹ T cells without MMB3.19 challenge succumbed to acute GVHD with an MST of 6.8 days. Similarly, transplanted mice with MMB3.19, and left untreated or treated with CNTO412 mAb, also died of GVHD with MSTs of 10.7 and 6.7 days, respectively, as would still be expected because of the antihost alloreactivity. In contrast, those mice receiving ATBM, CD4⫹ T cells, and challenge with MMB3.19 leukemia, followed by anti–TNF-␣ mAb treatments, avoided the early GVHD fatality and were able to mediate a GVL effect. Survival of this group was prolonged with an MST of 32.1 days and was significantly longer than the ATBM mice given MMB3.19 cells alone (P ⱕ .002). Mice mortality in the anti–TNF-␣ mAb group appeared to be caused by tumor burden, because they did not exhibit symptoms of GVHD, and leukemia cells were evident in the peritoneum at autopsy. In an
Figure 8. Effect of anti–TNF-␣ mAb treatment on allogeneic GVL activity. (B6xC3H)F1 mice were irradiated (13 Gy, split dose) and transplanted with C3H ATBM cells (2 ⫻ 106) alone, or in combination with 3 ⫻ 106 naı¨ve C3H CD4⫹ T cells. These mice were then challenged with an ip injection of 7.5 ⫻ 104 MMB3.19 cells on day 1, and mice were either left untreated or treated with anti– TNF-␣ or control CNTO412 mAbs (1mg on day ⫺1 ip; 0.5 mg on days 0 to 24, every 4th day). MST values (days) were: ATBM plus MMB3.19 (24.2); ATBM plus CD4 (6.8); ATBM plus CD4 plus MMB3.19 (10.7); ATBM plus CD4 plus MMB3.19 plus CNTO412 (6.7); ATBM plus CD4 plus MMB3.19 plus anti–TNF-␣ mAb (32.1; P ⱕ .002 v ATBM plus MMB3.19). The results represent pooled data (n ⫽ 16 per group) from 2 similar experiments. 299
R. Korngold et al.
allogeneic system, anti–TNF-␣ mAb treatment effectively blocked the development of CD4⫹ T cell–mediated GVHD, but allowed a significant level of GVL activity to be generated.
DISCUSSION In 1987, Piguet et al. [25] were the first to show the involvement of TNF-␣ in GVHD, using a P3 F1 model of MHC-incompatible HCT . They found that injection of a neutralizing rabbit antimouse recombinant TNF-␣ IgG polyclonal antibody could significantly prevent skin and gut lesions, decrease mortality rate (70% survival at day 80), and prevent weight loss (a systemic indicator of GVHD severity). Others have confirmed Piguet’s observations by administering anti–TNF-␣ antibodies starting either before HCT or as many as 7 days after [26-31]. In many of these studies, the effectiveness of the treatment varied widely and was complicated by varying sources of neutralizing antibodies to TNF-␣ and varying strain combinations for the models themselves. In our own earlier experience, it was shown that the protective effects of a polyclonal anti–TNF-␣ antibody were restricted to acute GVHD induced by CD4⫹ T cells and were not observed in a CD8-dependent model of GVHD [28]. This latter model, the B10.BR3 CBA strain combination, which is MHC compatible but disparate for multiple miHA, is the same as used in our current studies, although we were now able to observe moderate efficacy on GVHD development with the CNTO2213 anti–TNF-␣ mAb (Figure 1). The difference in the observed results of this CD8-mediated GVHD model is likely caused by the improved potency of the mAb preparation and the enhanced pharmacokinetic profile of the chimeric rat/mouse antibody itself. Yet the earlier findings were still consistent in regard to the relatively better effect of anti–TNF-␣ antibodies for CD4-mediated GVHD, which is also supported by the current results in the C3H3 (B6xC3H)F1 model (Figures 2 and 3). Treatment with CNTO2213 had a much more dramatic effect on survival when only donor CD4⫹ T cells were transplanted than with a donor inoculum of both CD4 and CD8 T cells. A key question in relation to the effects of anti– TNF-␣ mAb treatment on GVHD development is the level at which the cytokine blockade is having the most impact. Three possibilities to consider are whether (1) it is effective at blocking the initial conditioning-related endothelial activation and upregulation of adhesion molecules necessary for later T cell homing into the target tissues; (2) it blocks the TNF-␣ produced by alloreactive donor T cells in the peripheral lymphoid organs, which helps activate antigen-presenting cells, which via IL-12 and more 300
TNF-␣ production can then expand the initial stimulation of more T cells and their orientation towards Th1 cytokine production; or (3) it blocks the cytolytic effects on target tissue of TNF-␣ released locally by either T cells, macrophages, or mast cells [28] participating in the inflammatory process. Of course, all 3 effects of TNF-␣ blockade may be occurring, because the treatment regimen spans the time over which they each might be critical. In this regard, evidence provided by the use of p55 TNF-␣ receptor– deficient donor T cells in a GVHD model system suggested the importance of TNF-␣ in the alloreactive T cell response [42]. In addition, recent observations in the study of intestinal GVHD have indicated the importance of TNF-TNFR2 interactions for the development of MHC class II– disparate disease mediated by intestinal CD4⫹ Th1 cytokine production [43]. However, TNF-␣ did not seem to be critical for donor T cell activation in the spleen. Further investigations utilizing appropriate TNF-receptor– deficient or TNF-␣–nonproducing T cells and macrophages in GVHD models could shed additional light on the subject. The above-mentioned potential effects of TNF-␣ blockade also can be critical to the development of GVL activity, but conditions for successful development of these responses may also be dependent on the precise nature of the leukemia cells involved, where they are located, and the effector mechanisms by which the T cells are able to respond to them. First, certain types of leukemia cells may lack MHC class II expression (eg, many T cell leukemias) [44-46], which would likely make them a poorer target of CD4mediated GVL effects, although indirect presentation pathways can still operate for activation [47]. In murine models utilizing the MHC class II–negative mastocytoma, P815, both Hill et al. [48] and Tsukada et al. [49] have shown the importance of TNF-␣ to what is likely a CD8-mediated GVL response. In both of these cases, the tumor cells were administered iv and, therefore, the sites of tumor growth were likely to be accessible to the donor lymphocytes as part of their normal migration patterns through the lymphoid system. This would suggest that TNF-␣ was not required for increased extravasation of effector CD8 T cells through endothelial surfaces to reach the tumor cells. Thus, TNF-␣ may have been critical for CD8 activation, either directly or indirectly via antigen-presenting cell activation, or it was directly involved in the cytolysis of tumor cells. Unlike P815 tumor cells, the MMB3.19 leukemia cells express MHC class II and, therefore, can directly interact with responding CD4 T cells. Previous studies with MMB3.19 and similar myeloid leukemia models have shown that GVL responses can be mediated by either CD4⫹ or CD8⫹ donor T cells and that the former subset can effec-
TNF-␣ in GVHD
tively utilize either perforin- or FasL-dependent cytolytic pathways for the effect [50,51]. A second reason why TNF-␣ may not be critical to the CD4⫹ T cell–mediated GVL response is that in the case of many leukemia cells of myeloid origin (particularly monocytic and dendritic cell forms), they have the capability of producing and secreting their own TNF-␣, so they are less likely to be susceptible to the cytolytic effects of the molecule. This is the situation for several of the murine myeloid leukemia cell lines with which we have been working in our laboratory, including the MMB3.19 cell line, and they are completely resistant to any TNF-␣–mediated cytolysis [52] (R.K., unpublished data). Furthermore, returning to the important aspect of accessibility of the GVL-mediating CD4⫹ T cells to the MMB3.19 leukemia cells, the tumor challenge was ip, and, thus, T cell extravasation may be somewhat different than that required for the immunopathogenesis of GVHD lesions in target organs. In this regard, in other murine models, IL-15 has been found to be the dominant promoter of endothelial cell changes that enhance T cell entry into the peritoneal extralymphoid tissues [53]. As mentioned earlier, the role of TNF-␣ in the development of GVHD appeared to be more critical for CD4- rather than CD8-mediated disease in the models that were tested. This difference may be related to the underlying type of systemic disease that develops. The kinetics of the CD4-mediated GVHD response in the C3H3(B6xC3H)F1model is much more rapid (the majority of the mice succumbed by 11 days) than the CD8-mediated disease in the B10.BR3 CBA strain combination (mice died between 13 and 40 days). The CD4-mediated form of disease presents as a strong intestinal GVHD reaction, and the observed effect of TNF-␣ inhibition in this model is consistent with previous observations related to the dependence on TNF-␣ interactions for intestinal GVHD in the B63bm12 MHC class II– disparate model [43], and lipopolysaccharide–induced TNF-␣ release in a similar C3H3(B6xC3H)F1model [54]. However, the CD8-mediated B10.BR3 CBA GVHD response involves a much more widespread distribution of target tissue–related inflammation and injury with marked immunopathology in the liver, gut, skin, and lymphoid organs developing over the first several weeks after transplantation [28]. Therefore, it would seem that either some of these pathologic effects are independent of TNF-␣ or that enough fatal injury was still able to develop in some mice after cessation of anti–TNF-␣ mAb treatments on day 12 posttransplantation. Further pathologic analysis is necessary in this model to discern the effects of the TNF-␣ blockade on GVHD at the level of the target organs. In the allogeneic C3H3(B6xC3H)F1model, the
BB&MT
anti–TNF-␣ mAb treatment was clearly effective in blocking the CD4-mediated GVHD, while still allowing a significant GVL response to develop against the MMB3.19 leukemia cells (Figure 8). However, the GVL effect in this situation was not able to sustain meaningful long-term survival of the recipients, with all animals eventually succumbing to tumor burden. Although transplantation of higher dosages of donor T cells would likely increase the long-term survival rate, the model is counterbalanced by the limitation that increasing the donor T cell number would also significantly increase the GVHD potential of those cells when given at time of marrow transplant, beyond the level that can be controlled by the anti–TNF-␣ mAb treatment. The possible solution to this problem may be in the use of a delayed donor lymphocyte infusion (DLI) approach. In the clinical situation, DLI has been found to be effective in preventing or reversing leukemic relapse and EBV-induced lymphoproliferative disease in patients with poor reconstitution of donor T cells [55-57]. In murine models, it has been found that delaying the infusion of donor T cells to at least 2 weeks after bone marrow transplantation helps avoid the “cytokine storm” associated with the preconditioning irradiation exposure of the recipients, which is inductive for GVHD [58,59]. Delay also allows for both the loss of host antigen-presenting cells [60] and the de novo development of regulatory CD25⫹ T cells [61], both of which can limit the induction of GVHD. Therefore, higher doses of donor T cells can potentially be administered as a DLI with less risk of GVHD, although GVHD is still clearly the most dominant complication of the approach [55,56,62]. It remains to be seen whether TNF-␣ blockade will affect a DLI-mediated GVL effect; this is currently under study. We have found that a chimeric rat/mouse anti– TNF-␣ mAb could reduce the development of GVHD in murine models, with a more significant effect in CD4- rather than CD8-mediated disease. However, in the CD4 model, the development of GVL responses was unhindered by anti–TNF-␣ mAb treatment, and, thus, this approach may serve as a potential means for separating out deleterious from beneficial effects of donor T cell reactive capabilities.
ACKNOWLEDGMENTS The authors thank Dana Telem, Allison Lewandowski, Jeffrey Vakil, Danielle Castor, and Diana Whitaker-Menezes for their expert technical assistance at various points in the study. This work was funded through a research agreement sponsored by Centocor, Inc, Malvern, PA. 301
R. Korngold et al.
REFERENCES 1. Horowitz MM. Uses and growth of hematopoietic cell transplantation. In: Thomas ED, Blume KG, Forman SJ, eds: Hematopoietic Cell Transplantation. 2nd ed. Malden, MA. Blackwell Science; 1999:12-18. 2. Bensinger WI, Maloney D, Storb R. Allogeneic hematopoietic cell transplantation for multiple myeloma. Sem Hematol. 2001; 38:243-249. 3. Saba N, Flaig T. Bone marrow transplantation for nonmalignant diseases. J Hematother Stem Cell Res. 2002;11:377-387. 4. Truitt RL, Johnson BD. Principles of graft-vs-leukemia reactivity. Biol Blood Marrow Transplant. 1995;1:61-68. 5. Nash RA, Storb R. Graft-versus-host effect after allogeneic hematopoietic stem cell transplantation: GVHD and GVL. Curr Opin Immunol. 1996;8:674-680. 6. Shimoni A, Giralt S, Khouri I, Champlin R. Allogeneic hematopoietic transplantation for acute and chronic myeloid leukemia: non-myeloablative preparative regimens and induction of the graft-versus-leukemia effect. Curr Oncol Rep. 2000;2:132139. 7. Antin JH, Ferrara JLM. Cytokine dysregulation and acute graft-versus-host disease. Blood. 1992;80:2964-2968. 8. Norton J, Sloane JP. ICAM-1 expression on epidermal keratinocytes in cutaneous graft-versus-host disease. Transplant. 1991;51:1203-1206. 9. Johnson DR, Pober JS. Tumor necrosis factor regulation of major histocompatibility complex gene expression. Immunol Res. 1991;10:141-155. 10. Pober JS. Effects of tumour necrosis factor and related cytokines on vascular endothelial cells. Ciba Found Symp. 1987;131: 170-184. 11. Shanley TP, Warner RL, Ward PA. The role of cytokines and adhesion molecules in the development of inflammatory injury. Mol Med Today. 1995;1:40-45. 12. Ghayur T, Seemayer TA, Kongshawn PAL, Gartner JS, Lapp WS. Graft-versus-host (GVH) reactions in the beige mouse: an investigation of the role of host and donor natural killer cells in the pathogenesis of GVH disease. Transplantation. 1987;44: 261-267. 13. Shanahan JC, Clair W St. Tumor necrosis factor-alpha blockade: a novel therapy for rheumatic disease. Clin Immunol. 2002; 103:231-242. 14. Kam LY, Targan SR. TNF-alpha antagonists for the treatment of Crohn’s disease. Expert Opin Pharmacother. 2000;1:615-622. 15. Keating GM, Perry CM. Infliximab: An updated review of its use in Crohn’s disease and rheumatoid arthritis. Bio Drugs. 2002;16:111-148. 16. Deeg H., Spitzer TR, Cottler-Fox M, Cahill R, Pickle LW. Conditioning-related toxicity and acute graft-versus-host disease in patients given methotrexate/ cyclosporine prophylaxis. Bone Marrow Transplant. 1991;7:193-198. 17. Min CK, Lee WY, Min DJ, et al. The kinetics of circulating cytokines including IL-6, TNF-alpha, IL-8 and IL-10 following allogeneic hematopoietic stem cell transplantation. Bone Marrow Transplant. 2001;28:935-940. 18. Holler E, Kolb HJ, Moller A, et al. Increased serum levels of tumor necrosis factor alpha precede major complications of bone marrow transplantation. Blood. 1990;75:1011-1016. 19. Tanaka J, Imamura M, Kasai M, et al. Rapid analysis of tumor necrosis factor-alpha mRNA expression during venooclusive 302
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
disease of the liver after allogeneic bone marrow transplantation. Transplantation. 1993;55:430-432. Tanaka J, Imamura M, Kasai M, et al. Cytokine gene expression in peripheral blood mononuclear cells during graft-versus-host disease after allogeneic bone marrow transplantation. Br J Haematol. 1993;85:558-565. Herve P, Flesch M, Tiberghien P, et al. Phase I-II trial of a monoclonal anti-tumor necrosis factor alpha antibody for the treatment of refractory severe acute graft-versus-host disease. Blood. 1992;79:3362-3368. Holler E, Kolb HJ, Hintermeier-Knabe R, Kempeni J, Moller A, Wilmanns W. Systemic release of tumor necrosis factor alpha in human allogeneic bone marrow transplantation: Clinical risk factors, prognostic significance and therapeutic approaches. In: Link Freund, Schmidt Wlete, eds: Cytokines in Hematopoiesis, Oncology and AIDS II. Berlin-Heidelberg: Springer; 1992;435-442. Holler E, Kolb HJ, Hintermeier-Knabe R, et al. Role of tumor necrosis factor alpha in acute graft-versus-host disease and complications following allogeneic bone marrow transplantation. Transplant Proc. 1993;25:1234-1236. Holler E, Kolb HJ, Mittermuller J, et al. Modulation of acute graft-versus-host disease after allogeneic bone marrow transplantation by tumor necrosis factor ␣ (TNF␣) release in the course of pretransplant conditioning: Role of conditioning regimens and prophylactic application of a monoclonal antibody neutralizing human TNF␣ (MAK 195F). Blood. 1995;86:890899. Piguet PF, Grau GE, Allet B, Vassalli P. Tumor necrosis factor/cachectin is an effector of skin and gut lesions of the acute phase of graft-vs-host disease. J Exp Med. 1987;166:12801289. Shalaby MR, Fendly B, Sheehan KC, Schreiber RD, Ammann AJ. Prevention of the graft-versus-host reaction in newborn mice by antibodies to tumor necrosis factor-alpha. Transplantation. 1989;47:1057-1061. Holler E, Kolb HJ, Mittermuller J, et al. Modulation of acute graft-versus-host-disease after allogeneic bone marrow transplantation by tumor necrosis factor alpha (TNF alpha) release in the course of pretransplant conditioning: role of conditioning regimens and prophylactic application of a monoclonal antibody neutralizing human TNF alpha (MAK 195F). Blood. 1995;86:890-899. Murphy GF, Sueki H, Teuscher C, Whitaker D, Korngold R. Role of mast cells in early epithelial target cell injury in experimental acute graft-versus-host disease. J Invest Dermatol. 1994; 102:451-461. Wall DA, Sheehan KC. The role of tumor necrosis factor and interferon gamma in graft-versus-host disease and related immunodeficiency. Transplantation. 1994;57:273-279. Hattori K, Hirano T, Miyajima H, et al. Differential effects of anti-Fas ligand and anti-tumor necrosis factor alpha antibodies on acute graft-versus-host disease pathologies. Blood. 1998;91: 4051-4055. Holler E, Thierfelder S, Behrends U, et al. Anti-TNF-alpha and pentoxifylline for prophylaxis of a GVHD in murine allogeneic bone marrow transplantation. Oncologie. 1992;15:31-35. Xun CQ, Thompson JS, Jennings CD, Brown SA, Widmer MB. Effect of total body irradiation, busulfan-cyclophosphamide, or cyclophosphamide conditioning on inflammatory cytokine release and development of acute and chronic graft-
TNF-␣ in GVHD
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
versus-host disease in H-2-incompatible transplanted SCID mice. Blood. 1994;83:2360-2367. Vallera DA, Taylor PA, Vannice JL, Panoskaltsis-Mortari A, Blazar BR. Interleukin-1 or tumor necrosis factor-alpha antagonists do not inhibit graft-versus-host disease induced across the major histocompatibility barrier in mice. Transplantation. 1995;60:1371-1374. Baumbach WR, Keath EJ, Cole MD. A mouse c-myc retrovirus transforms established fibroblast lines in vitro and induces monocyte-macrophage tumors in vivo. J Virol. 1986;59:276283. Korngold R, Leighton C, Manser T. Graft-versus-myeloid leukemia responses following syngeneic and allogeneic bone marrow transplantation. Transplantation. 1994;58:278-287. Kadokami T, Frye C, Lemster B, Wagner CL, Feldman AM, McTiernan CF. Anti-tumor necrosis factor-alpha antibody limits heart failure in a transgenic model. Circulation. 2001;104: 1094-1097. Ceredig R, Lowenthal JW, Nabholz M, MacDonald HR. Expression of interleukin-2 receptors as a differentiation marker on intrathymic stem cells. Nature. 1985;314:98-100. Sarmiento M, Glasebrook AL, Fitch FWIgG or IgM monoclonal antibodies reactive with different determinants on the molecular complex bearing Lyt-2 antigen block T-cell mediated cytolysis in the absence of complement. J Immunol. 1980;125: 2665-2672. Hsieh MH, Patterson AE, Korngold RT cell subsets mediate graft-versus-myeloid leukemia responses via differential utilization of cytotoxic mechanisms. Biol Blood Marrow Transplant. 2000;6:231-240. Ferrara J, Guillen FJ, Sleckman B, Burakoff SJ, Murphy GF. Cutaneous acute graft-versus-host disease to minor histocompatibility antigens in a murine model: histologic analysis and correlation to clinical disease. J Invest Dermatol. 1986;86:371375. Korngold B, Sprent J. Lethal graft-versus-host disease after bone marrow transplantation across minor histocompatibility barriers in mice: Prevention by removing mature T cells from marrow. J Exp Med. 1978;148:1687-1698. Hill GR, Teshima T, Rebel VI, et al. The p55 TNF-alpha receptor plays a critical role in T cell alloreactivity. J Immunol. 2000;164:656-663. Brown GR, Lee E, Thiele DL. TNF-TNFR2 interactions are critical for the development of intestinal graft-versus-host disease in MHC class II-disparate (C57BL/6J–C57BL/6J x bm12)F1 mice. J Immunol. 2002;168:3065-3071. Liu A, Takahashi M, Toba K, et al. Regulation of the expression of MHC class I and II by class II transactivator (CIITA) in hematopoietic cells. Hematol Oncol. 1999;17:149-160. Yunis JJ, Band H, Bonneville F, Yunis EJ. Differential expression of MHC class II antigens in myelomonocytic leukemia cell lines. Blood. 1989;73:931-937. Lecchi M, Lovisone E, Genetta C, Peruccio D, Resegotti L, Richiardi P. Gamma-IFN induces a differential expression of HLA-DR, DQ and DP antigens on peripheral blood myeloid leukemic blasts at various stages of differentiation. Leuk Res. 1989;13:221-226.
BB&MT
47. Ossendorp F, Mengede E, Camps M, Filius R, Melief CJ. Specific T helper cell requirement for optimal induction of cytotoxic T lymphocytes against major histocompatibility complex class II negative tumors. J Exp Med. 1998;187:693-702. 48. Hill GR, Teshima T, Gerbitz A, et al. Differential roles of IL-1 and TNF-alpha on graft-versus-host disease and graft versus leukemia. J Clin Invest. 1999;104:459-467. 49. Tsukada N, Kobata T, Aizawa Y, Yagita H, Okumura K. Graft-versus-leukemia effect and graft-versus-host disease can be differentiated by cytotoxic mechanisms in a murine model of allogeneic bone marrow transplantation. Blood. 1999;93:27382747. 50. Hsieh MH, Korngold R. Differential use of FasL- and perforin-mediated cytolytic mechanisms by T-cell subsets involved in graft-versus-myeloid leukemia responses. Blood. 2000;96: 1047-1055. 51. Hsieh MH, Patterson AE, Korngold R. T-cell subsets mediate graft-versus-myeloid leukemia responses via different cytotoxic mechanisms. Biol Blood Marrow Transplant. 2000;6:231-240. 52. Hsieh MH, Varadi G, Flomenberg N, Korngold R. Leucylleucine methyl ester-treated haploidentical donor lymphocyte infusions can mediate graft-versus-leukemia activity with minimal graft-versus-host disease risk. Biol Blood Marrow Transplant. 2002;8:303-315. 53. Estess P, Nandi A, Mohamadzadeh M, Siegelman MH. Interleukin 15 induces endothelial hyaluronan expression in vitro and promotes activated T cell extravasation through a CD44dependent pathway in vivo. J Exp Med. 1999;190:9-19. 54. Cooke KR, Hill GR, Crawford JM, et al. Tumor necrosis factor-alpha production to lipopolysaccharide stimulation by donor cells predicts the severity of experimental acute graftversus-host disease. J Clin Invest. 1998;102:1882-1891. 55. Porter DL, Antin JH. The graft-versus-leukemia effects of allogeneic cell therapy. Ann Rev Med. 1999;50:369-386. 56. Schleuning M. Adoptive allogeneic immunotherapy— history and future perspectives. Transfus Sci. 2000;23:133-150. 57. Guglielmi C, Arcese W, Dazzi F, et al. Donor lymphocyte infusion for relapsed chronic myelogenous leukemia: Prognostic relevance of the initial cell dose. Blood. 2002;100:397-405. 58. Ferrara JL. Cytokine dysregulation as a mechanism of graft versus host disease. Curr Opin Immunol. 1993;5:794-799. 59. Johnson BD, Hanke CA, Truitt RL. The graft-versus-leukemia effect of post-transplant donor leukocyte infusion. Leuk Lymphoma. 1996;23:1-9. 60. Zhang Y, Louboutin JP, Zhu J, Rivera AJ, Emerson SG. Preterminal host dendritic cells in irradiated mice prime CD8⫹ T cell-mediated acute graft-versus-host disease. J Clin Invest. 2002;109:1335-1344. 61. Johnson BD, Becker EE, LaBelle JL, Truitt RL. Role of immunoregulatory donor T cells in suppression of graft-versushost disease following donor leukocyte infusion therapy. J Immunol. 1999;163:6479-6487. 62. Luznik L, Fuchs EJ. Donor lymphocyte infusions to treat hematologic malignancies in relapse after allogeneic blood or marrow transplantation. Cancer Control. 2002;9:123-137.
303
