High Interleukin-10 Serum Levels Are Associated with Fatal Outcome in Patients After Bone Marrow Transplantation

High interleukin-10 serum levels are associated with fatal outcome in patients after bone marrow transplantation

L Hempel1,DKo¨rholz2, P Nußbaum3,HBo¨nig2, S Burdach2,4 and F Zintl1

1Department of Pediatric Hematology and Oncology, Friedrich-Schiller University Medical Center, Jena; 2Department of Pediatric Hematology and Oncology, Heinrich-Heine University Medical Center, Du¨sseldorf; 3Institute of Clinical Pathology, Osnabru¨ck; and 4Biomedical Research Center, Heinrich-Heine University, Du¨sseldorf, Germany

Summary: production.1–3 Experiences with IL-10 knock-out mice show that IL-10 may contribute to the control of inflamma- IL-10 plays an important role in the control of immune tory reactions because these animals showed severe gastro- reactions during systemic infection. Here, IL-10 serum intestinal inflammation.4 In mouse models, elevation of IL- levels were investigated in patients after BMT. The IL- 10 during systemic infection protected the animals from 10 levels correlated with the clinical course of the an overwhelming immune reaction. Furthermore, mice with patients and with serum levels of C-reactive protein high endogenous or exogenous IL-10 levels survived sys- (CRP) and neopterin (NP). A total of 26 patients with temic shock syndromes, while application of anti-IL-10 led AML (7), ALL (12), CML (2), NHL (3) and multifocal to a fatal reaction.5–7 In contrast, in humans elevated IL-10 Ewing’s sarcoma (2) had received autologous (10) or levels during systemic shock correlated with an increased allogeneic (16) BMT from related (9) or unrelated TNF-␣ production, increased severity of septic shock and donors (7). Routine serum samples were obtained prior a high risk of disseminated intravascular coagulation.8–10 to BMT and at days 46 and 100 after BMT. However, The association between high IL-10 levels and fatality of in patients with severe complications additional samples septic shock was explained by the fact that IL-10 might not were drawn at individual points in time. Prior to BMT, be able to inhibit production of proinflammatory cytokines, IL-10 serum levels were not detectable in 24/24 patients. while suppressing T cell reactions which might otherwise Post-BMT, 11 patients developed elevated IL-10 levels, have supported pathogen elimination. In this study the role of these eight died of complications (DOC), whereas of IL-10 in patients after BMT was investigated under the only one of 15 patients with undetectable IL-10 died of hypothesis that in post-BMT patients, as in sepsis, complications, indicating that high IL-10 levels were sig- increased IL-10 might be correlated to poor outcome. IL- nificantly correlated with severe life-threatening compli- 10 serum levels correlated with the clinical course of the cations (␹2, P Ͻ 0.01). To determine the pathomechan- disease, the severity of acute phase reactions and macro- ism and role of the increased IL-10 levels, they were phage activation. correlated to the respective NP and CRP serum concentrations. CRP and NP concentrations were found significantly elevated in patients with detectable IL-10, Patients and methods indicating a severe acute phase reaction associated with Patients macrophage activation. In conclusion, high IL-10 serum levels in patients after BMT were significantly associa- Twenty-six children and adolescents were enrolled into this ted with fatal outcome. Since IL-10 is a strong sup- study. Patients’ ages ranged from 4.0 to 22.4 years (median: pressor of T cell immunity, high IL-10 production in 13.4 years). The male to female ratio was 18 to 8. Patients patients with severe complications such as septic shock had previously been treated for ALL (12), AML (7), CML or GVHD Ͼgrade II after BMT might lead to functional (2), NHL (3) and Ewing’s sarcoma (2). Five patients had immunodeficiency contributing to the poor prognosis of received an allogeneic transplant from a matched-related these patients. donor, four patients had been transplanted with bone mar- Keywords: sepsis; bone marrow transplantation; inter- row from an haploidentical related donor and seven patients leukin 10; neopterin had had unrelated donors. In addition, 10 patients had received autologous bone marrow or peripheral blood stem cell transplants. Blood was drawn from patients prior to BMT and at days 46 and 100 after BMT. However, in IL-10 is a potent macrophage and T cell-deactivating patients with multi-organ failure (MOF) additional blood cytokine, leading to a significant suppression of cytokine samples were taken during these complications. At the time of study, all patients with allogeneic transplants received immunosuppressive treatment. Acute GVHD was seen in Correspondence: L Hempel, Department of Pediatric Hematology and Oncology, Friedrich-Schiller University Medical Center, Kochstr 2, D- only six of 16 patients (Table 1). However, six of 10 07740 Jena, Germany patients without acute GVHD died in the early post-trans- Received 21 March 1997; accepted 15 May 1997 plant period. Interleukin-10 and bone marrow transplantation L Hempel et al 366 Table 1 Patients characteristics

No. Initials Sex Age Diagnosis Status at BMT Type of transplant Acute GVHD Status

1 DH F 17.2 ALL CR matched related GVHD II CCR 2 AK M 14.6 CML PR matched related none CCR 3 DS M 14.7 ALL CR matched related none DOC, day +67 4 KS M 4.0 ALL CR matched related none CCR 5 FT M 11.5 AML CR matched related none relapse day +126, died day +183 6 CK M 14.6 CML PR matched unrelated none DOC, day +34 7 RK M 8.7 AML relapse matched unrelated none DOC, day +20 8 CK M 10.4 NHL PR matched unrelated none DOC, day +26 9 SM F 7.1 ALL CR matched unrelated GVHD I–II CCR 10 KR F 22.4 ALL relapse matched unrelated none DOC, day +27 11 SR M 18.1 AML CR matched unrelated GVHD III DOC, day +103 12 FZ M 17.4 AML relapse matched unrelated none DOC, day +21 13 TB F 11.2 ALL CR mismatched related GVHD II CCR 14 CL F 10.8 ALL CR mismatched related GVHD II CCR 15 NS M 13.7 ALL relapse mismatched related GVHD II–III DOC, day +83 16 CS F 14.3 ALL relapse mismatched related none DOC, day +19 17 RB M 12.8 AML CR autologous none relapse day +293, alive 18 RF M 19.9 ALL CR autologous none relapse day +192, died day +220 19 EG M 17.7 AML CR autologous none CCR 20 MG M 17.9 NHL CR autologous none CCR 21 BH F 7.5 ALL CR autologous none CCR 22 HH M 17.5 AML CR autologous none CCR 23 DS M 15.4 EWS PR autologous none relapse day +161, alive 24 JS F 11.0 EWS PR autologous none CCR 25 NS M 7.3 NHL CR autologous none CCR 26 PW M 12.8 ALL CR autologous none CCR

The study was performed in accordance with the ethical Statistical analysis of data

standards laid down in the 1964 Declaration of Helsinki. 2 Patients or their parents gave their consent. The study was The Wilcoxon and ␹ tests were used for statistical analysis. authorized by the local ethics committee.

Results Cytokine analysis Concentrations of IL-10 were assessed by commercially To determine the role of IL-10 in patients after BMT, IL- available ultrasensitive ELISA kits (BioSource, Camarillo, 10 serum levels were correlated with the clinical course of CA, USA) as described previously.3 The lowest concen- the patients. Prior to BMT (‘day 0’) serum levels of IL- tration detected by this assay was 200 fg/ml. 10 were undetectable in all patients. Post-BMT 11 patients developed detectable IL-10 at various points in time. Of these, eight patients died of complications within 1–14 Neopterin (NP) determination (median 4) days after an elevated IL-10 had been found. The three surviving patients with elevated IL-10 serum lev- Serum neopterin levels were measured by HPLC. Briefly, els presented with acute GVHD grade II or severe CMV 200 ␮l samples or standards were added to a mobile phase infection. One patient received systemic IL-2 treatment at containing 0.1 m KH2PO4, (Merck, Darmstadt, Germany), the time of IL-10 determination. In 15/26 patients IL-10 0.005 m octansulfonic acid (Fluka, Bux, Switzerland; pH remained undetectable throughout the observation period. was adjusted to 3.0 with 1 N HCl) and 50 ␮l of 3.4 m Of these, only one succumbed to complications. The last HClO4. After vortexing the mixtures for 30 s, samples were IL-10 measurement in this patient had been done 21 days centrifuged at 11 500 r.p.m. for 5 min. Supernatants (50 ␮l) prior to his death so that an increased IL-10 may have been were applied to a HPLC (Li-Chros-Pher 100 RP 8E; missed. Thus, we found detectable IL-10 levels to be sig- Merck). Neopterin concentrations were determined at Ex nificantly correlated with DOC (Figure 1; Table 2; ␹2, wavelength of 360 nm compared with Em wavelength of P Ͻ 0.01). 430 nm. Next, the mechanism of in vivo IL-10 production was investigated. It was found that in patients with detectable CRP determination IL-10, CRP levels (median: 8.57 mg/dl) were significantly elevated as compared to patients with undetectable IL-10 CRP values were determined by particle-enhanced immu- (median: 0 mg/l; Figure 2, Wilcoxon test, P Ͻ 0.001). nonephelometry. 40 ␮l serum were applied to a fully auto- Thus, in our patient population the IL-10 serum levels were matic nephelometer (Behringwerke, Marburg, Germany). associated with a severe acute phase reaction. Interleukin-10 and bone marrow transplantation L Hempel et al 367 70 Since IL-10 is produced by both T cells and macro- DOC phages,11–13 we determined the serum concentrations of 60 no DOC neopterin as an indicator of macrophage activation.14–16 It was shown that high IL-10 levels were positively correlated 50 with increased neopterin levels (Figure 3, Wilcoxon test, P Ͻ 0.01). Thus, increased IL-10 levels in post-BMT 40 patients might be produced by macrophage activation during an acute phase reaction. 30

20 Discussion

IL-10 serum levels (ng/1) 10 After bone marrow transplantation, different parts of the immune system recover at different times. One of the earl- 0 n=9 17 3 17 17 iest detectable functions is the production of macrophage- derived cytokines.17 Increased macrophage activation might day 0 46 100 be observed at the time of engraftment.14 Since most of the Figure 1 IL-10 serum levels in patients after BMT. IL-10 serum levels patients with increased IL-10 levels in this study were were determined in 26 patients prior to and post-BMT. Figure 1 shows found within the first month after BMT, it is most likely time after BMT on the x-axis, and IL-10 concentrations on the y-axis. that the IL-10 was derived from activated monocytes rather Patients who died of complications (DOC) are depicted as open squares, patients who did not succumb to complications as solid circles. than from T cells. Additional evidence for this point of view is presented by the observation that increased IL-10 levels were associated with a significantly increased production of neopterin, a protein released from macrophages Table 2 Clinical course of patients with elevated IL-10 levels upon activation.14–16 No. IL-10 level Complications Status In patients after BMT, elevated neopterin and an abnor- mal monocyte activation has been observed from the time 14 6 46.6 sepsis, MOF, VOD, ARF DOC, day 34 of engraftment until 6 weeks after BMT. Since macro- 7 64.4 sepsis, MOF, VOD, ARF DOC, day 20 phage activation did not correlate with complications such 8 8.5 sepsis, VOD DOC, day 26 as GVHD or veno-occulusive disease (VOD), it was sug- 10 8.9 VOD, ARF DOC, day 27 gested that the monocyte activation might predispose to, 11 21.0 GVHD III–IV DOC, day 103 rather than result from, the development of these compli- 12 62.2 sepsis, MOF, VOD, ARF DOC, day 21 15 35.5 sepsis, MOF, GVHD III, ARF DOC, day 83 cations. In our study we could demonstrate macrophage 16 62.0 sepsis, MOF, VOD, ARF DOC, day 19 activation in patients with life-threatening complications 9 40.7 CMV infection CCR after BMT, which correlated with an increased release of 14 2.0 GVHD II CCR IL-10. Since increased IL-10 production has been reported 23 7.1 systemic IL-2 treatment relapsed, alive to lead to a suppression of hematopoiesis18 as well as antigen-specific T cell activation,1,19 the increased IL-10 pro-

Patients without Patients with Patients without Patients with elevated IL-10 elevated IL-10 elevated IL-10 elevated IL-10 30 40

µ 30 20

15 20

Neopterin serum levels ( g/l) 5 CRP serum levels (mg/ dl)

0 0 n=15 n=11 n=15 n=11 Figure 3 Correlation between IL-10 and neopterin serum levels in Figure 2 Correlation between IL-10 and CRP levels in patients after patients after BMT. Neopterin serum levels were determined in 26 patients BMT. CRP serum levels were determined in 26 patients after BMT on after BMT on the same days that IL-10 was measured. Elevated neopterin the same days that IL-10 was measured. Elevated CRP was significantly was significantly associated with detectable IL-10 (Wilcoxon test, associated with detectable IL-10 (Wilcoxon test, P Ͻ 0.001). P Ͻ 0.01). Interleukin-10 and bone marrow transplantation L Hempel et al 368 duction in these critically ill patients might have led to an interleukin 10 in serum associated with fatality in meningo- immune-paralysis. This has previously been reported for coccal disease. Infect Immunol 1995; 63: 2109–2112. patients with septic shock syndromes.8–10 In addition, 9 Marchant A, Alegre ML, Hakim A et al. Clinical and biologi- enhanced stimulation of IL-10 increases susceptibility cal significance of interleukin-10 plasma levels in patients towards certain pathogens such as candida.20 Thus, with septic shock. J Clin Immunol 1995; 15: 266–273. 10 van Deuren M, van der Ven-Jongekrijg J, Bartelink AK et although it could not be determined from our results al. Correlation between proinflammatory cytokines and anti- whether high IL-10 levels were preceding the development inflammatory mediators and the severity of disease in menin- of severe complications and thereby could be predictive of gococcal infections. J Infect Dis 1995; 172: 433–439. fatal outcome, it might be conceivable that IL-10 prevented 11 Del Prete G, De Carli M, Almerigogna F et al. Human IL-10 patients from recovery by inhibiting immune-mediated is produced by both type 1 helper (Th1) and type 2 helper elimination of pathogens. Therefore, future studies should (Th2) T cell clones and inhibits their antigen-specific prolifer- investigate whether neutralization of IL-10 might help ation and cytokine production. J Immunol 1993; 150: 353– patients with severe GVHD or sepsis early after BMT. 360. 12 De Waal Malefyt R, Abrams J, Bennett B et al. Interleukin- 10 inhibits cytokine synthesis by human monocytes: an auto- regulatory role of IL-10 produced by monocytes. J Exp Med Acknowledgements 1991; 174: 1209–1220. 13 Fiorentino DF, Bond MW, Mosmann TR. Two types of mouse This study was supported by DFG grants Ko 971/3-2 and He helper T cell. IV Th2 clones secrete a factor that inhibits cyto- 2202/2-2. kine production by Th1 clones. J Exp Med 1989; 170: 2081–2095. 14 Castenskiold EC, Kelsey SM, Collins PW et al. Functional hyperactivity of monocytes after bone marrow transplan- References tation – possible relevance for the development of post- transplant complications or relapse. Bone Marrow Transplant 1 De Waal Malefyt R, Haanen J, Spits H et al. Interleukin-10 1995; 15: 879–884. and viral IL-10 strongly reduce antigen-specific human T cell 15 Gomez-Jimenez J, Martin MC, Sauri R et al. Interleukin-10 proliferation by diminishing the antigen-presenting capacity of and the monocyte/macrophage-induced inflammatory response monocytes via down regulation of class II major histocompat- in septic shock. J Infect Dis 1995; 171: 472–475. ibility complex expression. J Exp Med 1991; 174: 915–924. 16 Haupt W, Hohenberger W, Klein P, Christou NV. Detection 2 De Waal Malefyt R, Yssel H, De Vries JE. Direct effects of of neopterin, interleukin-6 and acute-phase proteins as para- IL-10 on subsets of human CD4+ T cell clones and resting T meters of potential monocyte activation in preoperative cells. J Immunol 1993; 150: 4754–4765. patients. Infection 1995; 23: 263–266. 3 Hempel L, Ko¨rholz D, Burdach S et al. Interleukin-10 directly 17 Pechumer H, Leinisch E, Bender-Go¨tze C, Ziegler-Heitbrock inhibits interleukin-6 production by T cells. Scand J Immunol HW. Recovery of monocytes after bone marrow transplan- 1995; 41: 462–466. tation – rapid reappearance of tumor necrosis factor alpha and 4Ku¨hn R, Lohler J, Rennick D et al. Interleukin-10-deficient interleukin 6 production. Transplantation 1991; 52: 698–704. mice develop chronic enterocolitis. Cell 1993; 75: 263–274. 18 Oehler L, Foedinger M, Koeller M et al. Interleukin-10 5 Berg DJ, Kuhn R, Rajewsky K et al. Interleukin-10 is a central inhibits spontaneous colony-forming unit granulocyte–macro- regulator of the response to LPS in murine models of endo- phage growth from human peripheral blood mononuclear cells toxin shock and the Schwartzman reaction but not endotoxin by suppression fo endogenous GM-CSF release. Blood 1997; tolerance. J Clin Invest 1995; 96: 2339–2347. 89: 1147–1153. 6 Bourrie B, Bouaboula M, Benoit JM et al. Enhancement of 19 Ko¨rholz D, Schneider M, Giesecke E et al. Induction of endotoxin-induced interleukin-10 production by SR 31737A, remission by cyclosporin A in a patient with a hemophago- a sigma ligand. Eur J Immunol 1995; 25: 2882–2887. cytic syndrome. Eur J Paediat 1995; 154: 778 (Abstr.). 7 van der Poll T, Marchant A, Buurman WA et al. Endogenous 20 Romani L, Puccetti P, Menacci A et al. Neutralization of IL- IL-10 protects mice from death during septic peritonitis. J 10 up-regulates nitric oxide production and protects suscep- Immunol 1995; 155: 5397–5401. tible mice from challenge with Candida albicans. J Immunol 8 Lehmann AK, Halstensen A, Sornes S et al. High levels of 1994; 152: 3514–3521.