Lung transplantation has become a viable treatment option for patients with end-stage lung disease. Despite advances in immunosuppressive therapies, however, the majority of lung transplant patients surviving more than 5 years develop obliterative bronchiolitis (OB), a form of chronic allograft rejection in the lung (1). OB is characterized histopathologically by inflammation of subepithelial structures and epithelial cell injury of the small airways, leading to excessive fibroproliferation and airway obliteration (1). Activated cells in OB lesions secrete various cytokines that mediate and augment the inflammatory and fibroproliferative responses (2,3). Tumor necrosis factor (TNF)-α is a macrophage and T-cell–derived cytokine with numerous proinflammatory effects (4). TNF-α induces the release of other inflammatory mediators and of angiogenic factors and chemokines. Adhesion molecule expression in endothelial cells is increased by TNF-α, enabling leukocyte extravasation, and TNF-α is also a potent fibrogenic cytokine affecting fibroblast proliferation and chemotaxis (5). In the context of lung transplantation, TNF-α has been suggested to contribute to acute and chronic forms of lung allograft rejection (6–11). Blockade of the proinflammatory effects of TNF-α by anti–TNF-α antibodies or soluble TNF-α receptor fusion protein has proven useful in the treatment of diseases with chronic inflammation as a key feature in their pathophysiology (12). A few studies performed in rodent lung transplantation models revealed that inhibition of TNF-α attenuated acute (6,7) and chronic rejection (10,11). In development of chronic rejection, the role of TNF-α has not, however, been previously studied in the small airways. Our group has developed a porcine heterotopic bronchial allograft model, in which we transplant small bronchi close to the bronchioles, the actual site of OB. The main differences from the bronchioles are the presence of cartilaginous structures and submucosal glands in bronchi. In our model, heterotopic bronchial allografts exhibit changes similar to those in human OB (13). These changes can be delayed or prevented with immunosuppression (14). In this study, we investigated TNF-α protein expression in the development of OB and evaluated the potential of TNF-α inhibition with infliximab, a monoclonal anti–TNF-α antibody, to ameliorate the inflammatory changes seen in the heterotopic bronchial allografts.
Porcine Heterotopic Bronchial Transplantation Model
The animals were nonrelated, random-bred, domestic pigs, each weighing approximately 20 kg. They received humane care in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (publication No. 86-23, revised 1985). The study protocol was accepted by the institutional committee for animal research and by the Uusimaa Provincial Administration, Finland. Special attention was given to anesthesia and pain relief during surgical procedures.
Anesthesia was induced with intramuscular ketamine (10–15 mg/kg) (Pfizer, Brooklyn, NY) azaperone (10–15 mg/kg) (Janssen Pharmaceutica, Beerse, Belgium), atropine (0.05 mg/kg) (Leiras, Turku, Finland), and intravenous sodium pentobarbital (6–12 mg/kg) (Orion, Turku, Finland). Intravenous diazepam (0.25 mg/kg) (Orion Pharma, Espoo, Finland) was given before intubation, and after intubation, muscle relaxation was maintained with intravenous pancuronium bromide (2–4 mg) (Organon, Oss, Holland). During the surgical procedure, the animals were ventilated with 40% oxygen and enflurane (Abbott, Espoo, Finland).
Left thoracotomy was performed for removal of the caudal lobe. Peripheral bronchial segments (1–2 cm in length and 1–2 mm in diameter) were dissected free of the surrounding lung parenchyma. Bronchial implants were transplanted subcutaneously into the ventral side of each recipient, each serving as both donor and recipient. The donor and recipient animals were crossed within their own groups.
Postoperative pain was controlled with diclofenac acid 37.5 mg (Novartis, Basel, Switzerland) intramuscularly, and ceftriaxone 500 mg (Roche, Basel, Switzerland) was given intramuscularly for infection control for 3 days. For ulcer prophylaxis, perioperatively administered intravenous ranitidine 50 mg (Orion Pharma) was continued at a daily oral dose of 150 mg for 3 weeks.
Three groups were formed (n=4 each): autografts, nontreated allografts, and allografts treated with one preoperatively (after induction of anesthesia) administered intravenous 5-mg/kg dose of a chimeric human-mouse anti–TNF-α monoclonal antibody, infliximab (Schering-Plough, Kenilworth, NJ). Grafts were removed on postoperative days 2, 4, 7, 11, 14, 21, and 28. At each time point, three parallel samples were removed from each animal. The total number of samples in this study was 252. For the harvesting procedure, the pigs were anesthetized with 6 mg/kg ketamine and 6 mg/kg azaperone intramuscularly. At the end of follow-up, they were killed with intravenous sodium pentobarbital.
Luminal obliteration, epithelial destruction, bronchial wall inflammation, fibrosis and necrosis, cartilage necrosis and proliferation, and pericartilaginous inflammation and fibrosis were semiquantitatively graded from 0 to 3 in hematoxylin-eosin–stained sections as follows: 0, no alteration; 1, mild alteration, including a minor portion of the observed area; 2, moderate (pathologic areas equal in area to normal tissue); and 3, severe alterations (pathologic changes as predisposing component). Percentage of normal epithelial cells, epithelial cells with metaplastic atypia (ciliated respiratory cells changing into squamous cells), and the pure basal cell layer of the remaining epithelium were also evaluated. From each sample, three separate bronchi (if present) were analyzed. The samples were reviewed in a blinded fashion.
For immunohistochemical analysis, part of each sample was snap-frozen in liquid nitrogen and stored at −70°C; 4-μm-thick sections were cut, and the frozen sections were placed on glass slides, (SuperFrost Plus, Menzel-Gläser, Germany), air-dried, and fixed in acetone at −20°C for 20 min, and stored at −20°C until used. A three-layer indirect immunoperoxidase method was used. To avoid nonspecific staining resulting from endogenous peroxidase, before staining, the slides had been refixed in chloroform for 30 min at room temperature and then air-dried. The use of chloroform is based on its ability to reduce enzyme activity, and on previous experience (13,14). Mouse monoclonal antibodies against pig CD4+ and CD8+ cells (clones 74-12-4 and PT81B) (VMRD Inc., Dullman, WA) (13,14), against human monocytes-macrophages, also reactive with porcine macrophages (clone MAC387) (Serotec, Oxford, UK) (13,14), and against pig TNF-α (clone 9B4) (Endogen, Woburn, MA) (15) were used. The primary antibodies were diluted in 0.1% bovine serum albumin (Sigma, St. Louis, MO) in Tris-NaCl buffer (pH 7.4). The frozen sections were first incubated with the primary antibody (against CD4, CD8, and macrophages for 30 min, at room temperature, at dilutions of 1:200, 1:1,000, and 1:200; and against TNF-α for 18 hr at 4°C, at 1:100); then with peroxidase-conjugated, rabbit anti-mouse immunoglobulin (Ig) antibody (Dako A/S, Denmark), diluted at 1:10 for 30 min at room temperature; and thereafter with a peroxidase-conjugated, goat anti-rabbit Ig antibody (Zymed, San Francisco, CA), diluted at 1:50, for 30 min at room temperature. The secondary and tertiary antibodies were diluted in 50% normal pig serum in Tris-NaCl buffer. Between incubations, the slides were washed with Tris-NaCl buffer. The reaction was revealed by 3-amino-9-ethyl carbazole (Sigma) solution containing hydrogen peroxidase. The sections were counterstained with Mayer’s hemalun, and coverslips were mounted (Aquamount; Gurr, Poole, UK). Controls for immunostaining were performed with mouse monoclonal IgG1 (X931; Dako A/S) as a primary antibody. None of the controls showed positivity.
The number of cells positive for CD4, CD8, and TNF-α, and macrophages were counted using oil immersion at 100× objective magnification, separately in the epithelium, in the bronchial wall, and in the obliterative plug. The number of positive cells was counted in 10 randomly chosen microscopic fields in the epithelium and in five fields in the bronchial wall and obliterative plug. The number of TNF-α–positive epithelial cells and inflammatory cells among the epithelium were calculated separately. In addition, TNF-α–positive epithelial cells were divided according to weak or strong staining. From each sample, three separate bronchi (if present) were analyzed. The samples were reviewed in a blinded fashion (by H.S.A. and U.-S.S.) Interobserver variability was minimal (H.S.A., P.K.M., and U.-S.S interpreted the results).
All data are expressed as mean+SEM. Variations between the allograft groups were analyzed with the Mann-Whitney U test, and variations between three groups were calculated by the nonparametric Kruskal-Wallis one-way analysis by ranks. The rank sums were then used for Dunn’s test at a significance level of 5% (Medstat; Astra Group A/S, Copenhagen, Denmark). Values of P <0.05 were considered significant. For correlation analysis, Spearman’s rank correlation (Statistica version 5; StatSoft Inc., Tulsa, OK) was used.
No side effects appeared during infliximab infusion or thereafter.
Epithelium and Obliteration
Autografts maintained virtually normal bronchial epithelium and patent bronchial lumens (P <0.01 from day 7 and from day 11 onward, respectively, in comparison with allograft groups) (Fig. 1). Nontreated allografts lost their epithelial covering almost completely by day 7. Inhibition of TNF-α significantly slowed epithelial loss early in the disease process (P <0.05 on days 7 and 11 between treated and nontreated allografts) (Figs. 1 and 2). In treated allografts, on days 4, 7, and 11 nearly half the epithelium was of normal type, whereas in nontreated allografts, one third was normal on day 4; on day 7, mainly only basal cells existed (data not shown).
Epithelial cells, and inflammatory cells among the epithelium, showed positivity for TNF-α. In allografts, inhibition of TNF-α reduced the number of TNF-α–positive epithelial cells to the level of autografts, and on day 4 this number was significantly lower in treated than in nontreated allografts (P <0.01) (Table 1 and Fig. 3A and B). Proportions of cells with intense staining were greater in nontreated than in the treated allografts on day 4 (60% vs. 50%) and on day 7 (100% vs. 0%).
Recruitment of CD8+ lymphocytes in the epithelium was significantly more diminished on days 2, 4, and 7 in treated than in nontreated allografts, and numbers of macrophages and TNF-α–expressing inflammatory cells in the epithelium were significantly lower on day 4 (P <0.05) (Tables 1 and 2 and Fig. 3A–F). No significant differences existed in CD4+ counts between allograft groups (Table 2 and Fig. 3G and H). In autografts, the number of CD8+ cells was close to the same at early observation points as in treated allografts, and numbers of other inflammatory cells and TNF-α–expressing cells remained low through follow-up (Tables 1 and 2).
From day 11 onward, the degree of epithelial loss increased in the treatment group; on day 21, it was complete (Fig. 1). The process of luminal obliteration was slower in the treated group, with a significant difference from the nontreated allografts on day 11 (P <0.05) (Fig. 1). In the obliterative plug, the numbers of TNF-α–positive cells, CD4+ and CD8+ cells, and macrophages in both allograft groups showed no significant differences (Tables 1 and 2).
Throughout follow-up, the degree of bronchial wall inflammation in the treated allografts was milder than in nontreated allografts, and this difference reached statistical significance on days 4, 7, and 21 (P <0.01) (Figs. 1 and 2). In the bronchial wall, macrophages; lymphocytes; and occasional endothelial cells, smooth muscle cells, and fibroblasts expressed TNF-α. TNF-α expression was increased in the allografts in comparison with autografts (P <0.05 from day 7 onward), but no significant differences appeared in the number of TNF-α–positive cells between allograft groups (Table 1).
Inhibition of TNF-α had the same effect on the invasion of CD8+ cells in the bronchial wall as in the epithelium; the number of CD8+ lymphocytes was significantly reduced in the beginning: on days 2, 4, and 7 (P <0.05, in comparison with nontreated allografts) (Table 2 and Fig. 3C and D). In addition, the numbers of macrophages were significantly lower on day 7 in the treated allografts than in nontreated allografts (P <0.05) (Table 2). No clear differences were observable in the CD4+ lymphocyte counts between allograft groups (Table 2). In autografts, invasion of CD8+ cells was significantly less than in the allograft groups (P <0.05 through the follow-up), and CD4+ lymphocytes and macrophages existed in only small amounts (Table 2).
Fibrosis in the bronchial wall was reduced in the treated group early on, significantly on days 11 and 14 (P <0.05 in comparison with nontreated allografts) (Table 3); and bronchial wall necrosis was milder in the treated group, significantly on day 7 (P <0.05) (Table 3). In autografts, no necrotic areas were evident after 1 week, and inflammation and fibrosis remained mild (P <0.01 from day 4 onward, P <0.01 from day 11 onward, in comparison with allograft groups) (Fig. 1 and Table 3).
TNF-α inhibition diminished inflammatory cell infiltrates around the cartilage, with a significant difference between treated and nontreated allografts on days 4, 7, and 11 (P <0.05). Cartilaginous necrosis and development of pericartilaginous fibrosis were slightly milder in the treatment group than in nontreated allografts. Some difference in proliferation of cartilage existed at late observation points, when it was more pronounced in treated than in nontreated allografts. In autografts, pericartilaginous inflammation was mild (P <0.01 in comparison with allograft groups), and cartilage remained viable (data on cartilage are listed in Table 3).
When all groups were analyzed together, the degree of epithelial destruction from day 4 onward correlated with that same day’s obliteration (r =0.42–0.99, P <0.001) and future obliteration (r =0.67–0.96, P <0.001). The number of TNF-α–positive epithelial cells on days 4, 7, and 11 correlated with epithelial damage (r =0.62, P <0.001) and obliteration (r =0.67, P <0.001) on the same day or at future time points (r =0.39–0.74, P <0.05) (r =0.33–0.86, P <0.05). In the bronchial wall, the number of TNF-α–positive cells on days 7, 11, 14, and 21 correlated with same-day (r =0.43–0.62, P <0.05) or future obliteration (r =0.49–0.82, P <0.01).
In OB, inflammatory cells and their products play a key role (2,3,14,16,17). TNF-α is involved in inflammation in multiple steps, its action being considered crucial in the initiation and maintenance of this process (18). TNF-α is an important cytokine involved in the development of cytotoxic lymphocytes, enhancing T-cell responses and regulating leukocyte movement, and it shows direct cytotoxic activity (19,20). In our model, the pathologic changes of OB occur at an accelerated rate; an influx of inflammatory cells is followed by rapid fibroproliferation, leading to total luminal occlusion within 3 weeks (13,14). By blocking TNF-α, we observed a significant reduction in inflammatory cell infiltrates, especially CD8+ lymphocytes early in the disease process. As a consequence of diminished inflammation, epithelium was preserved longer in our treated grafts than in nontreated ones. Inhibition of TNF-α reduced TNF-α expression in epithelial cells and inflammatory cells among the epithelium and led to milder necrosis of the bronchial wall structures.
Of the agents targeting TNF-α, the neutralizing mouse-human chimeric anti–TNF-α monoclonal antibody infliximab is the most intensively investigated and has the widest indications, and is now an approved therapy for rheumatoid arthritis and chronic inflammatory bowel disease (12,21,22). We chose to study the effect of this drug in our model because protein coding sequences of human and pig TNF-α share a considerable degree of homology (86%) (23). Furthermore, in previous experiments in different pig and rat models, infliximab has shown anti-inflammatory effects (24,25). In lung diseases, infliximab has been successfully tried in individual patient cases in rheumatoid arthritis-associated pulmonary fibrosis (26). Infliximab targets both soluble and transmembrane forms of TNF-α (27). In studies of rheumatoid arthritis, infliximab has been shown to reduce leukocyte trafficking into the joints by diminishing adhesion molecule and chemokine expression (28) and angiogenesis (29) and to inhibit the TNF-α–controlled proinflammatory cytokine network (30). In this study, a likely explanation for the attenuated rejection in the treated grafts was the ability of infliximab to decrease chemokines and adhesion molecules and directly inhibit the lymphoproliferative and cytotoxic actions of TNF-α. By binding in transmembrane TNF-α on the cell surface, infliximab caused cell destruction. As a result, this led to a decreased number of TNF-α–expressing cells and to elimination of their action in augmenting the inflammatory reaction.
Inhibition of TNF-α had the most prominent influence on the number of CD8+ lymphocytes and to some extent on macrophages, but CD4+ cell counts remained at the same level between allograft groups. This may have occurred because the rapid influx of inflammatory cells was largely attributable to CD8+ lymphocytes, and the other cells were relatively scarce in the early inflammatory phase. In addition to decreased inflammation, a decreased bronchial wall fibrosis was evident in our anti–TNF-α antibody-treated allografts. The role of TNF-α as a profibrotic cytokine involved in the development of lung fibrosis is characterized (31), and in OB, TNF-α has been suggested to increase the expression of profibrotic growth factors (11). Our findings support the profibrotic role of TNF-α in OB development.
In the context of experimental lung transplantation, two studies have explored the potential of inhibiting TNF-α in the treatment of acute rejection with polyclonal antibodies (6) or with neutralizing TNF-α antisera (7). Results were contradictory, with the former study showing no benefit from TNF-α inhibition, whereas the latter found significant attenuation of acute rejection in the treatment group. In studies using a mouse tracheal transplant model, treatment with serial doses of neutralizing antibodies against TNF-α (10) or TNF-α–soluble receptor (11) prevented or delayed the development of chronic rejection. Our study in a large-animal model in which small airways were transplanted demonstrated that inhibition of TNF-α with infliximab, an anti–TNF-α antibody widely used in clinical practice, has a similar impact. The limitations of our model are the lack of air-epithelial interface and initial avascularity. However, neovascularization occurs in our model within only a few days (13). In addition, the drug infusion was given before removal of the left lung lobe, so infliximab was already present in the tissue at the time of subcutaneous implantation.
At a 5-mg/kg dose, the median half-life of infliximab in humans is 9.5 days, and infliximab is detected in the serum through 8 to 12 weeks (32,33). In our model, infliximab treatment significantly delayed but, as expected, did not prevent the development of OB. This may be attributable to single-infusion therapy. Furthermore, various cytokines participate in the rejection process (2,3). It is most likely that even in the presence of anti–TNF-α antibody, inhibition of a single cytokine is not enough to totally prevent the progression of the obliterating process in our large-animal model.
This study showed a correlation between epithelial destruction and obliteration. In addition, expression of TNF-α in epithelial cells and in bronchial wall cells correlated significantly with epithelial injury and obliteration. Inhibition of TNF-α delayed onset of OB by reducing inflammation and reducing especially the CD8+ lymphocytes, and slowed the process of epithelial destruction and luminal occlusion. These results further confirm that TNF-α plays a role in the development of OB by promoting inflammation and epithelial injury. Furthermore, our results suggest that in lung transplant patients, inhibition of TNF-α may be beneficial; however, taking into consideration the insidious pathogenesis of OB, further studies should help assess the timing for this intervention.
We thank Sisko Litmanen for excellent technical assistance and Irmeli Lautenschlager, M.D, Ph.D., for helpful advice regarding TNF immunohistochemistry.
1. Estenne M, Hertz MI. Bronchiolitis obliterans after human lung transplantation. Am J Respir Crit Care Med 2002; 166: 440.
2. Boehler A, Bai XH, Liu M, et al. Upregulation of T-helper 1 cytokines and chemokine expression in post-transplant airway obliteration. Am J Respir Crit Care Med 1999; 159: 1910.
3. Neuringer IP, Walsh SP, Mannon RP, et al. Enhanced T cell cytokine gene expression in mouse airway obliterative bronchiolitis. Transplantation 2000; 69: 399.
4. Beutler B, Cerami A. Tumor necrosis, cachexia, shock, and inflammation: A common mediator. Annu Rev Biochem 1988; 57: 505.
5. Strieter RM, Lukacs NW, Kunkel SL. Role of adhesion molecules and cytokines in lung transplantation. In: Ward PA, Fantone JC, eds. Adhesion molecules and the lung. New York, Marcel Dekker 1996: 279.
6. Saito R, Prehn J, Zuo XJ, et al. The participation of tumor necrosis factor in the pathogenesis of lung allograft rejection in the rat. Transplantation 1993: 55; 967.
7. DeMeester SR, Rolfe MW, Kunkel SL, et al. The bimodal expression of tumor necrosis factor-alpha in association with rat lung reimplantation and allograft rejection. J Immunol 1993: 150; 2494.
8. Magnan A, Mege JL, Reynaud M, et al. Monitoring of alveolar macrophage production of tumor necrosis factor-alpha and interleukin-6 in lung transplant recipients: Marseille and Montreal lung transplantation group. Am J Respir Crit Care Med 1994: 150; 684.
9. Sekine Y, Bowen LK, Heidler KM, et al. Role of passenger leukocytes in allograft rejection: Effect of depletion of donor alveolar macrophages on the local production of TNF alpha, T helper 1/T helper 2 cytokines, IgG subclasses, and pathology in a rat model of lung transplantation. J Immunol 1997; 159: 4084.
10. Smith CR, Jaramillo A, Lu KC, et al. Prevention of obliterative airway disease in HLA-A2-transgenic tracheal allografts by neutralizing of tumor necrosis factor. Transplantation 2001: 72; 1512.
11. Aris RM, Walsh S, Chalermskulrat W, et al. Growth factor upregulation during obliterative bronchiolitis in the mouse model. Am J Respir Crit Care Med 2002: 166; 417.
12. Feldmann M, Maini RN. Anti-TNF alpha therapy of rheumatoid arthritis: What have we learned? Annu Rev Immunol 2001: 19; 163.
13. Uusitalo MH, Salminen US, Ikonen TS, et al. Alloimmune injury preceding airway obliteration in porcine heterotopic lung implants: A histologic and immunohistologic study. Transplantation 1999; 68: 970.
14. Maasilta PK, Salminen US, Lautenschlager IT, et al. Immune cells and immunosuppression in a porcine bronchial model of obliterative bronchiolitis. Transplantation 2001; 72: 998.
15. Su X, Prestwood AK, Otto CM, et al. Monoclonal antibodies to porcine tumor necrosis factor alpha: Development of an enzyme-linked immunosorbent assay. J Immunoassay 1992; 13: 253.
16. Reinsmoen NL, Bolman RM, Savik K, et al. Are multiple immunopathogenetic events occurring during the development of obliterative bronchiolitis and acute rejection. Transplantation 1993; 55: 1040.
17. Neuringer IP, Mannon RB, Coffman TM, et al. Immune cells in a mouse model of obliterative bronchiolitis. Am J Respir Cell Mol Biol 1998; 19: 379.
18. Kollias G, Douni E, Kassiotis G, et al. On the role of tumor necrosis factor and receptors in models of multiorgan failure, rheumatoid arthritis, multiple sclerosis and inflammatory bowel disease. Immunol Rev 1999; 169: 175.
19. Scheurich P, Thoma B, Ucer U, et al. Immunoregulatory activity of recombinant human tumor necrosis factor (TNF)-alpha: Induction of TNF receptors on human T cells and TNF-alpha-mediated enhancement of T cell responses. J Immunol 1987; 138: 1786.
20. Sedgwick JD, Riminton DS, Cyster JG, et al. Tumor necrosis factor: A master regulator of leukocyte movement. Immunol Today 2000; 21: 110.
21. Lipsky PE, van der Heijde DM, St Clair EW, et al. Infliximab and methotrexate in the treatment of rheumatoid arthritis: Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. N Engl J Med 2000: 343; 1594.
22. Hanauer SB, Feagan BG, Lichtenstein GR, et al. Maintenance infliximab for Crohn’s disease: the ACCENT I randomised trial. Lancet 2002: 359; 1541.
23. Kuhnert P, Wüthrich C, Peterhans E, et al. The porcine tumor necrosis factor encoding genes: Sequence and comparative analysis. Gene 1991; 102: 171.
24. Olmarker K, Rydevik B. Selective inhibition of tumor necrosis factor-α prevents nucleus pulposus-induced thrombus formation, intraneural edema, and reduction of nerve conduction velocity. Spine 2001; 26: 863.
25. Kulmatycki KM, Abouchehade K, Sattari S, et al. Drug-disease interaction: Reduced β-adrenergic and potassium channel antagonist activities of sotalol in the presence of acute and chronic inflammatory conditions in the rat. Br J Pharm 2001; 133: 286.
26. Vassallo R, Matteson E, Thomas CF Jr. Clinical response of rheumatoid arthritis-associated pulmonary fibrosis to tumor necrosis factor-alpha inhibition. Chest 2002; 122: 1093.
27. Scallon BJ, Moore MA, Trinh H, et al. Chimeric anti-TNF-alpha monoclonal antibody cA2 binds recombinant transmembrane TNF-alpha and activates immune effector functions. Cytokine 1995; 7: 251.
28. Taylor PC, Peters AM, Paleolog E, et al. Reduction of chemokine levels and leukocyte traffic to joints by tumor necrosis factor alpha blockade in patients with rheumatoid arthritis. Arthritis Rheum 2000; 43: 38.
29. Maini RN, Taylor PC, Paleolog E, et al. Anti-tumour necrosis factor specific antibody (infliximab) treatment provides insights into the pathophysiology of rheumatoid arthritis. Ann Rheum Dis 1999; 58 (suppl): 156.
30. Charles P, Elliott MJ, Davis D, et al. Regulation of cytokines, cytokine inhibitors, and acute-phase proteins following anti-TNF-alpha-therapy in rheumatoid arthritis. J Immunol 1999; 163: 1521.
31. Brass DM, Hoyle GW, Poovey HG, et al. Reduced tumor necrosis factor-alpha and transforming growth factor-beta1 expression in the lungs of inbred mice that fail to develop fibroproliferative lesions consequent to asbestos exposure. Am J Pathol 1999; 154: 853.
32. Cornillie F, Shealy D, Haens GD, et al. Infliximab induces potent anti-inflammatory and local immunomodulatory activity but no systemic immune suppression in patients with Crohn’s disease. Aliment Pharmacol Ther 2001; 15: 463.
33. Kavanaugh A, St Clair EW, McCune WJ, et al. Chimeric anti-tumor necrosis factor-alpha monoclonal antibody treatment of patients with rheumatoid arthritis receiving methotrexate therapy. J Rheumatol 2000; 27: 841.