Serum markers of graft-versus-host disease after bone marrow transplantation

Size: px

Start display at page:

Download "Serum markers of graft-versus-host disease after bone marrow transplantation"

Edgar Casey
6 years ago
Views:

1 Serum markers of graft-versus-host disease after bone marrow transplantation Hiroyuki Kayaba, MD, PhD, a Makoto Hirokawa, PhD, b Arata Watanabe, PhD, c Norihiro Saitoh, MD, a Cui Changhao, MD, a Yoshiyuki Yamada, MD, a Kohei Honda, MD, PhD, a Yoshimi Kobayashi, MD, a Osamu Urayama, MD, PhD, a and Junichi Chihara, MD, PhD a Akita, Japan Background: Graft-versus-host disease is one of the major complications after allogenic bone marrow transplantation, but it is not easy to anticipate the onset. Objectives: The purpose of this study was to determine clinically useful markers of acute graft-versus-host disease. Methods: We measured the serum levels of tumor necrosis factor-α, soluble tumor necrosis factor receptor 1, soluble c-kit, soluble Fas, soluble intercellular adhesion molecule-1, growthrelated oncogene protein-α, thrombomodurin, and interleukin- 16 in 13 patients at 1 to 7 weeks after allogenic bone marrow transplantation. Results: The patients with acute graft-versus-host disease showed a significant increase of tumor necrosis factor, soluble tumor necrosis factor receptor 1, soluble Fas, soluble intercellular adhesion molecule-1, and growth-related oncogene protein-α, although there was a decrease of soluble c-kit. The increases of serum soluble tumor necrosis factor receptor 1, intercellular adhesion molecule-1, and growth-related oncogene protein-α were preceded by the elevation of soluble Fas. Conclusion: The patients with acute graft-versus-host disease had increased serum levels of tumor necrosis factor-α, soluble tumor necrosis factor receptor 1, soluble Fas, and soluble intercellular adhesion molecule 1 and a decreased soluble c-kit level. Tumor necrosis factor-α and soluble c-kit were shown to be sensitive and specific parameters for graft-versus-host disease after bone marrow transplantation, and soluble Fas was shown to be a predictor of acute graft-versus-host disease after bone marrow transplantation. (J Allergy Clin Immunol 2000;106:S40-4.) Key words: Graft-versus-host disease, bone marrow transplantation, tumor necrosis factor, tumor necrosis factor-receptor, Fas, intercellular adhesion molecule-1, c-kit, growth-related oncogene protein-α Allogenic bone marrow transplantation (BMT) is usually performed for the treatment of hematopoietic diseases associated with abnormal or malignant marrow and From a the Department of Clinical and Laboratory Medicine, b the 3rd Department of Internal Medicine, and c the Department of Pediatrics, Akita University School of Medicine. Reprint requests: Hiroyuki Kayaba, PhD, Department of Clinical and Laboratory Medicine, Akita University School of Medicine, Hondo, Akita , Japan. kayaba@yukari.hos.akita-u.ac.jp. Copyright 2000 by Mosby, Inc /2000 $ /0/ doi: /mai S40 Abbreviations used BMT: Bone marrow transplantation GROα: Growth-related oncogene protein-α GVHD: Graft-versus-host disease sc-kit: Soluble c-kit sfas: Soluble Fas sicam-1: Soluble intercellular adhesion molecule-1 stnfr1: Soluble TNF receptor 1 IFN: Interferon during intensive therapy for solid malignant tumors in combination with chemotherapy and/or radiation therapy. Graft-versus-host disease (GVHD) is one of the lifethreatening complications of BMT. The onset of acute GVHD is usually within 8 weeks after transplantation, and it manifests with skin rash, liver dysfunction, and diarrhea. Great attention is paid to the prevention of GVHD in management before and after BMT. Some serum markers (such as the levels of TNF-α, 1,2 IFN-γ, 3 and soluble c-kit [sc-kit] 4 ) have been reported to be useful for the prediction or early detection of acute GVHD and for the assessment of the response to treatment. However, there still remain many problems with attempts to use these parameters as reliable and sensitive markers of acute GVHD. In the early period after BMT, complex interactions between graft and host occur that are mediated by various cytokines (including TNF-α and IFN-γ) produced by activated T cells and natural killer cells. In the present study, we selected soluble Fas (sfas), soluble ICAM-1 (sicam-1), TNF-α, soluble TNF receptor 1 (stnfr1), sc-kit, growth-related oncogene protein-α (GRO-α), thrombomodurin, and IL-16 as candidate markers for acute GVHD and measured serum levels of in 13 patients 50 days after BMT. Our objective was to determine which of these markers were clinically useful for the early detection and evaluation of acute GVHD. MATERIAL AND METHODS Patients This study included 6 male and 7 female patients, aged 6 to 49 years (mean, 30.1 years) who underwent allogenic BMT. The underlying diseases were acute lymphocytic leukemia (n = 5 patients), acute myelocytic leukemia (2 patients), chronic myelocytic leukemia (2 patients), myelodysplastic syndrome (2 patients),

2 J ALLERGY CLIN IMMUNOL VOLUME 106, NUMBER 1, PART 2 Kayaba et al S41 TABLE I. Clinical features of patients Patient No. Age (yr) Gender Underlying disease Type of transplant GVHD 1 10 F ALL Allo BMT Grade M ALL Allo BMT Grade M CML Allo BMT Grade M NHL Allo BMT Grade F CML Allo BMT Grade F ALL Allo BMT Grade F AA Allo BMT Grade M ALL Allo BMT Grade M ANLL Allo BMT Grade F MDS Allo BMT Grade F ANLL Allo BMT Grade F ALL Allo BMT Grade M MDS Allo BMT Grade 3 ALL, Acute lymphocytic leukemia; CML, chronic myelocytic leukemia; NHL, non-hodgkin lymphoma; AA, aplastic anemia; ANLL, acute nonlymphocytic leukemia; MDS, myelodysplastic syndrome; Allo BMT, allogenic bone marrow transplantation. aplastic anemia (1 patient), and non-hodgkin s lymphoma (1 patient). Five of the 13 patients experienced the development of GVHD. The disease distribution in the patients with GVHD were acute lymphocytic leukemia (2 patients), chronic myelocytic leukemia (1 patient), non-hodgkin s lymphoma (1 patient), and myelodysplastic syndrome (1 patient). The mean ages of the patients with and without GVHD were 29.2 ± 5.6 years and 30.7 ± 5.2 years, respectively (Table I). Collection of serum Serum was collected from the patients with BMT after informed consent was obtained. Blood samples were collected before and 3, 7, 14, 21, 28, 35, 42, and 49 days after BMT; the serum was stored at 80 C until assay. The levels of TNF-α, stnfr1, sc-kit, sfas, sicam-1, GRO-α, thrombomodulin, and IL-16 in the serum samples were measured by ELISA. The patients were divided into 2 groups according to the presence or absence of GVHD. Serum from healthy volunteers matched for age and sex was used as the control. The serum concentration of each parameter was compared between the patients with or without GVHD and the normal control subjects. Statistical analysis One factor analysis of variance with Scheffe s F-test was used for comparison of mean values. A probability value of less than.05 was defined as statistically significant. RESULTS TNF-α The serum TNF-α level of the patients with acute GVHD (3.29 ± 2.3 pg/ml) was significantly higher than that of the patients without acute GVHD (1.1 ± 0.78 pg/ml). Moreover, the serum TNF-α level of the patients without acute GVHD was significantly lower than that of the control subjects (4.44 ± 1.88 pg/ml; Fig 1). Three of 5 patients with acute GVHD had a peak TNF-α level exceeding 5 pg/ml, but 1 patient with grade IV GVHD showed no marked increase of TNF-α level during his clinical course. No relationship was found between the severity of GVHD and the serum TNF-α level. The positive and negative predictive values, when the cut-off value was settled at 5 pg/ml, were 1.0 and 0.8, respectively. stnfr1 A significant increase of the serum stnfr1 level was observed in the patients both with and without acute GVHD ( ± ng/ml; P <.01; and ± ng/ml; P <.05, respectively) when compared with the control subjects (810 ± ng/ml; Fig 2). Four of 5 patients with GVHD had a peak serum stnfr1 level of 1500 ng/ml or more, and the positive and negative predictive values were 0.45 and 0.75, respectively, with the cut-off value of 1500 ng/ml. sc-kit The mean serum level of sc-kit was decreased after BMT, irrespective of the presence or absence of acute GVHD when compared with that in control subjects (266.5 ± AU/mL; P <.01). Moreover, the patients with acute GVHD had significantly lower serum sc-kit levels than those patients without acute GVHD (GVHD(+), ± 87.5 AU/mL vs GVHD(-), 196 ± 65.4 AU/mL; P <.01; Fig 3). When the cut-off value was settled at 100 AU/mL (namely, when the patients with serum sc-kit levels less than 100 AU/ml during 7 weeks after BMT were judged as abnormal), the positive predictive value was 0.8 and the negative predictive value was sfas sfas levels in the patients with acute GVHD (4168 ± pg/ml) showed a significant increase when compared with those levels in the patients without acute GVHD ( ± pg/ml) or in the control subjects (1420 ± 290 pg/ml; Fig 4). Three of 5 patients with acute GVHD had sfas levels of more than 5000 pg/ml on the day of BMT, and sfas rose before the serum TNF-α level increased in those patients. The patients without acute GVHD had no significant increase of sfas. The positive and negative predictive value were 0.67 and 0.86, respectively, with the cut-off value of 5000 pg/ml. sicam-1 The serum sicam-1 level increased gradually after

S42 Kayaba et al J ALLERGY CLIN IMMUNOL JULY 2000 FIG 1. Serum TFN-α levels during the 7 weeks after BMT.

The horizontal bars represent the 10th, 25th, 50th, 75th, and 90th percentiles from bottom to top, in that order.

subjects. FIG 2. Serum stnfri levels during the 7 weeks after BMT.

control subjects. FIG 3. Serum sc-kit levels during 7 weeks after BMT.

The shaded area represents the normal range (mean ± 2SD for the control subjects). FIG 4.

There was a significant increase of sfas levels in the patients with acute GVHD compared with those levels in patients without acute

3 S42 Kayaba et al J ALLERGY CLIN IMMUNOL JULY 2000 FIG 1. Serum TFN-α levels during the 7 weeks after BMT. The serum TNF-α levels are presented as box plots. The horizontal bars represent the 10th, 25th, 50th, 75th, and 90th percentiles from bottom to top, in that order. The serum TNF-α level of patients without acute GVHD was significantly lower than that of patients with acute GVHD or the control subjects. FIG 2. Serum stnfri levels during the 7 weeks after BMT. A significant increase of the serum stnfr1 level was observed in the patients both with and without acute GVHD when compared with the control subjects. FIG 3. Serum sc-kit levels during 7 weeks after BMT. The patients with acute GVHD had significantly lower serum sc-kit levels than those patients without acute GVHD. The shaded area represents the normal range (mean ± 2SD for the control subjects). FIG 4. Serum sfas levels during the 7 weeks after BMT. There was a significant increase of sfas levels in the patients with acute GVHD compared with those levels in patients without acute GVHD or the control subjects. FIG 5. Serum sicam-1 levels at 3 to 7 weeks after BMT. In the period from 3 to 7 weeks after BMT, the patients with acute GVHD showed a significant increase of serum sicam-1 levels when compared with the control subjects. Serum sicam-1 levels in the patients without acute GVHD showed no significant difference from the control values. FIG 6. Serum GRO-α levels over 7 weeks after BMT. The mean serum GRO-α level in the patients without acute GVHD was significantly lower than that in the patients with acute GVHD.

4 J ALLERGY CLIN IMMUNOL VOLUME 106, NUMBER 1, PART 2 Kayaba et al S43 BMT in patients both with or without GVHD. For 2 weeks after BMT, there was no significant difference of sicam-1 between the patients with or without GVHD and the control subjects. From 3 to 7 weeks after BMT, however, the patients with acute GVHD showed a significant (P <.05) increase of serum sicam-1 (468 ± ng/ml) when compared with the control subjects ( ± ng/ml). In contrast, the serum sicam-1 levels of patients without acute GVHD (339.9 ± ng/ml) showed no marked difference from those in control subjects (Fig 5). One of the patients with an increase of serum sicam-1 eventually experienced the development of chronic GVHD. GRO-α The mean serum GRO-α level was ± 8.6 pg/ml in the patients without acute GVHD and was significantly lower than the levels in patients with acute GVHD ( ± pg/ml; Fig 6). All the patients with acute GVHD had serum GRO-α levels higher than 100 ng/ml after BMT, although none of the patients without acute GVHD had GRO-α levels exceeding 100 ng/ml. The increase of the serum GRO-α was preceded by an increase of sfas or TNF-α. Thrombomodurin The serum thrombomodurin level increased gradually after BMT and reached a plateau that was close to the level in the control subjects (20.6 ± 4.4 U/mL) by day 50 after BMT. There was no marked difference between patients with (19.7 ± 7.7 U/mL) or without (18.55 ± 10 U/mL) acute GVHD. IL-16. There were no significant differences of IL-16 levels between the 3 groups. The mean IL-16 values for the patients with or without acute GVHD and the control subjects were ± 73.5 pg/ml, 85.4 ± 69.3 pg/ml, and ± 36.1 pg/ml, respectively. DISCUSSION Acute GVHD is one of the life-threatening complications of BMT. It usually develops within 8 weeks after transplantation; the skin, liver, and intestine are the main sites involved. Because BMT has become a common treatment for hematopoetic disease, malignancy, and some metabolic diseases, the clinical importance of reliable markers for acute GVHD has increased. In 1975, TNF-α was discovered as a factor that caused tumor necrosis in mice, and effects related to inflammatory reactions and hematopoietic cells were subsequently shown. TNF-α may also play an important role in GVHD, 5 and its serum level is reported to increase in patients with GVHD. 1,2 Moreover, anti TNF-α and anti- Fas monoclonal antibodies are reported to be effective against lethal acute GVHD in mice. 6 In our series, the patients with acute GVHD had significantly higher serum TNF-α levels compared with those patients without acute GVHD. Moreover, the positive predictive value of serum TNF-α levels was the highest among the markers examined; however, the rise of the serum level was preceded by sfas. These observations may suggest that the serum TNF-α level is a good parameter for the evaluation of the therapy for GVHD. The fact that an anti TNF-α monoclonal antibody is reported to improve the symptoms and signs of GVHD may suggest the critical role of TNF-α. TNFR-deficient mice are reported to have far less GVHD than normal mice. 7 Serum stnfr-1 levels were elevated in both groups of patients after BMT in our series, and the peak value was significantly greater in patients with GVHD. The fact that the patients with acute GVHD showed a significant increase of stnfr-1, which neutralizes TNF-α, may mean there is excessive TNF-α production in patients with acute GVHD. The increase of stnfr-1 was not seen earlier than that of other parameters in the present study, but stnfr-1 has been reported to increase in advance of transplantation-related complications. 8 The decrease of the serum sc-kit level is reported to be an indicator of acute GVHD after BMT. 4 Because sc-kit is expressed by immature hematopoietic cells, sckit levels may be an indicator the number of hematopoietic cells. In this study, serum sc-kit levels decreased after BMT, and the decline was more pronounced in the patients with GVHD, irrespective of the severity of GVHD. The negative predictive value of serum sc-kit level was the highest (0.88), with the second highest positive predictive value (0.80) among the parameters examined. The serum sc-kit may be useful not only for identifying the patients with GVHD but also for ruling out GVHD in patients with sufficient serum levels of sc-kit. The serum sfas level showed a marked increase in our patients with acute GVHD when compared with the control subjects. SFas has already been reported to increase in patients with acute GVHD. 6,9 In a murine experimental model, an anti-fas monoclonal antibody (Fas ligand) is reported to be effective for improving acute GVHD, especially for hepatic involvement. 6 In the present study, the increase of serum sfas was observed within 2 weeks after BMT and preceded the increase of TNF-α. These results suggest sfas may be useful as an early indicator of acute GVHD; however, the positive predictive value was lower than those of TNF-α or sc-kit. The mechanism and the source of the elevation of sfas in patients with acute GVHD are not known. It is reported that an injection of sfas into mice resulted in increase of lymphocytes, 10 which suggests an inhibitory effect of sfas on cell death. It is possible that the elevated level of serum sfas in patients with acute GVHD may prevent the transplanted blood cells from apoptosis. ICAM-1 plays an important role in the migration and activation of inflammatory cells in patients with inflammatory and immunologic diseases. An anti ICAM-1 monoclonal antibody has been reported to have a beneficial effect on GVHD after small bowel transplantation in animal experiments. 11 Furthermore, elevation of sicam-1 was reported in patients with chronic GVHD after BMT. 12 It was not until 2 weeks after BMT that the serum sicam-1 level of our patients

5 S44 Kayaba et al J ALLERGY CLIN IMMUNOL JULY 2000 with acute GVHD showed a significant increase compared with that in the patients without acute GVHD or the control subjects, and one patient with an increase of sicam-1 experienced the development of chronic GVHD. These results indicate that sicam-1 may be an indicator of chronic GVHD rather than acute GVHD after BMT. GRO-α is a chemokine that is thought to be involved in the moderation of the acute inflammatory response and has a stimulatory effect on granulocytes. It is released by granulocytes after stimulation with TNFα. 13 We found a significant decrease of the serum GROα level in the patients without acute GVHD compared with the patients with acute GVHD or the control subjects. The decrease of serum GRO-α in patients without acute GVHD may have reflected the low serum TNF-α level than in patients with acute GVHD. The source of GRO-α in our patients with acute GVHD was unclear. IL- 16 is produced by CD8 + cells and has chemoattractant and proliferative effects on CD4 + cells and is thought to play an important role in inflammatory reactions. 14 However, the serum level of IL-16 showed a wide range in this study, and there were no differences among the 3 groups studied. Serum thrombomodurin levels were also not helpful as an indicator of acute GVHD, although an increase of thrombomodurin has been reported in microangiopathy associated with grade II to IV GVHD. 15 In summary, we demonstrated that patients with acute GVHD had increased serum levels of TNF-α, stnfr1, sfas, and sicam-1 and a decreased sc-kit level. Judging from the positive and negative predictive values, TNF-α and sc-kit were shown to be sensitive and specific parameters for GVHD after BMT. sfas was shown to be a predictor of acute GVHD after BMT; however, the positive predictive value was lower than the predictive value of TNF-α and sc-kit. REFERENCES 1. Nagler A, Bishara A, Brautbar C, Barak V. Dysregulation of inflammatory cytokines in unrelated bone marrow transplantation. Cytokines Cell Mol Ther 1998;4: Miyamoto T, Akishi K, Hayashi S, Gondo H, Murakawa M, Tanimoto K, et al. Serum concentration of the soluble interleukin-2 receptor for monitoring acute graft-versus-host disease. Bone Marrow Transplant 1996;17: Toren A, Barak V, Novick D, Nagler A. Soluble interferon-gamma receptor and interferon-gamma in patients undergoing allogenic bone marrow transplantation for hematological malignancies. Cytokines Cell Mol Ther 1997;3: Hashino S, Imamura M, Kobayashi S, Tanaka J, Kasai M, Sakuraba K, et al. Soluble c-kit in acute GVHD after allogenic bone marrow transplantation. Br J Haematol 1995;89: Facon T, Jouet JP, Noel-Walter MP, Bloget F, Bauters F, Janin A. Involvement of TNF-alpha secreting macrophages in lethal forms of human graft-versus-host disease. Bone Marrow Transplant 1997;20: Hattori K, Hirano T, Miyajima H, Yamakawa N, Tateno M, Oshimi K, 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: Speiser DE, Bachmann MF, Frick TW, McKall-Faienza K, Griffiths E, Pfeffer K, et al. TNF receptor p55 controls early acute graft-versus-host disease. J Immunol 1997;158: Or R, Kalinkovich A, Nagler A, Weisman Z, Naparstek E, Weiss L, et al. Soluble tumor necrosis factor (stnf) receptors: a possible prognostic marker for bone marrow transplantation-related complications. Cytokines Mol Ther 1996;2: Liem LM, van Lopik T, van Nieuwenhuijze AE, van Houwelingen HC, Aarden L, Goulmy E. Soluble Fas levels in sera of bone marrow transplantation recipients are increased during acute graft-versus-host disease but not during infections. Blood 1998;91: Knipping E, Debatin KM, Stricker K, Heiling B, Eder A, Krammer PH. Identification of soluble APO-1 in supernatants of human B- and T-cell lines and increased serum levels in B- and T-cell leukemias. Blood 1995;85: Poritz LS, Page MJ, Tilberg AF, Koltun WA. Amelioration of graft versus host disease with anti-icam-1 therapy. J Surg Res 1998;80: Maeda Y, Chihara J, Sumimoto Y, Yamada H, Miyatake J, Matsuda M, et al. Soluble intracellular adhesion molecule-1 levels of patients with acute myeloid leukemia after allogenic bone marrow transplantation. J Allergy Clin Immunol 1997;100:S Gasperini S, Calzetti F, Russo MP, De Gironcoli M, Cassatella MA. Regulation of GRO alpha production in human granulocytes. J Inflamm 1995;45: Cruikshank WW, Kornfeld H, Center DM. Signaling and functional properties of interleukin-16. Int Rev Immunol 1998;16: Kanamori H, Maruta A, Sasaki S, Yamazaki E, Ueda S, Katoh K, et al. Diagnostic value of hemostatic parameters in bone marrow transplantassociated thrombotic microangiopathy. Bone Marrow Transplant 1998;21:705-9.

Cytokines, adhesion molecules and apoptosis markers. A comprehensive product line for human and veterinary ELISAs

Cytokines, adhesion molecules and apoptosis markers A comprehensive product line for human and veterinary ELISAs IBL International s cytokine product line... is extremely comprehensive. The assays are