Increased Serum Levels of Tumor Necrosis Factor a! Precede Major Complications
of Bone Marrow Transplantation
By E. Holler, H.J. Kolb, A. Moller, J. Kempeni,
S.Liesenfeld, H. Pechumer, W. Lehmacher, G. Ruckdeschel, B. Gleixner,
C. Riedner, G. Ledderose, G. Brehm, J. Mittermuller, and W. Wilmanns
Acute graft-versus-host disease, interstitial pneumonitis,
endothelial leakage syndrome, and veno-occlusive disease
are major complications of bone marrow transplantation.
Though several new regimens for prophylaxis and treatment of these syndromes have been introduced, the overall
incidence has been only slightly reduced over the last few
years. We retrospectively analyzed tumor necrosis factor
alpha (TNFa) serum levels between day -8 and day 100
after bone marrow transplantation in 56 patients transplanted in our unit for a variety of hematological diseases.
In 34 patients with uneventful courses, mean TNFa levels
rose to a maximum of 76 k 29 pg/mL. In contrast, 22 patients with major transplant related complicationsshowed
mean increases of TNFa of 492 k 235 pg/mL ( P < .00Ol).
Increases of TNFa occurred before interstitial pneumonitis
and severe acute graft-versus-host disease with a latency
of 25 to 54 days. Early complications such as endothelial
leakage syndrome and veno-occlusive disease were closely
associated with increases of TNFa serum levels. Our study
suggests two pathways of TNFa release: activation of host
macrophagesand stimulation of donor cells in the course of
acute graft-versus-host disease. Cytokine monitoringshould
be helpful for prediction and earlier treatment of major
transplant related complications.
0 1990 by The American Society of Hematology.
I
CML advanced stages, n = 8; severe aplastic anemia, n = 2;
myelodysplastic syndromes (MDS), n = 6.
Pretransplant conditioning regimens were fractionated total body
irradiation (3 times, 4 Gy/d) in 41 patients or busulphan (4
mg/kg/d over 4 days) in 15 patients, followed by high dose
cyclophosphamid (50 mg/kg on 4 consecutive days). Attendant risks
were fully explained to all patients, donors, and relatives. Written
informed consent was obtained from patients and donors.
N H U M A N BONE MARROW transplantation (BMT),
transplant-related mortality is still a major problem even
in good risk patients. Life-threatening transplant-related
complications (TRC) occur in 25% to 35% of patients, and
the overall incidence has only slightly decreased from 1979 to
1988.’ During the aplastic phase, endothelial leakage syndrome (ELS) and veno-occlusive disease (VOD) are the
major TRC. After engraftment, infections, interstitial pneumonitis (IP), and severe acute graft-versus-host disease
(aGVHD) are the major causes of death during the first 2 to
6 months after BMT. IP, ELS, and VOD occur more often
after allogeneic as compared with autologous BMT, and I P
and ELS are correlated with severity of aGVHD.2-4
Recently, we described severe microangiopathy characterized by intravascular hemolysis, thrombocytopenic purpura,
and increases of factor VIII-related antigen (FVIIIRAg) in
patients with aGVHD and cyclosporin prophylaxis.’ Our
analysis indicated participation of monocyte activation in
endothelial damage. We suggested release of tumor necrosis
factor alpha (TNFa) by monocytes on stimulation with
interferon gamma (IFNT), which in turn is produced by T
lymphocytes involved in graft-versus-host reactions.
We now performed a retrospective analysis of TNFa
serum levels in a consecutive series of 56 patients receiving
BMT in our unit from January 1986 until June 1988. A
strong correlation of major T R C during the first 6 months
after BMT with preceding increases of TNFa serum levels
was found, suggesting a major pathophysiological role of
cytokine release.
MATERIALS AND METHODS
Patients. All patients receiving BMT in our unit from January
1986 until June 1988 were included in this analysis. Fifty-two
patients received allogeneic (49 HLA-identical and three HLAmismatched sibling) transplants; two patients were grafted from
syngeneic donors and two patients had autologous grafts. Median
age of these patients at the time of BMT was 35 (range 18 to 55)
years. Twenty-three recipients were female and 33 were male.
Diagnoses and stages of disease were as follows: acute myeloid
leukemia (AML): first or second complete remission (CR), n = 15,
greater than second CR, n = 3; acute lymphoblastic leukemia
(ALL): first or second CR, n = 8, greater than second CR, n = 2;
chronic myelogenous leukemia (CML): chronic phase, n = 12;
Blood, Vol 75, N o 4 (February 15). 1990: pp 101 1-1016
Prophylaxis and treatment of aGVHD in patients receiving
allogeneic BMT. Prophylaxisof aGVHD was performed by continuous infusion of cyclosporin from day - 1 until day 28 after BMT
followed by oral cyclosporin. Plasma levels were adjusted between
150 and 200 ng/mL in the first 3 to 4 months. After this period,
cyclosporin was gradually reduced in patients without evidence of
aGVHD. In addition, all patients received prophylactic intravenous
methotrexate on days 1, 3, and 6. Acute GVHD was graded
according to standard clinical and biochemical criteria: Patients
who developed aGVHD grade I1 or more were treated with corticosteroids and further weekly injectionsof methotrexate (10 mg/m2) in
the absence of myelosuppression. Patients not responding to this
treatment received 1 or 2 courses of monoclonal antibody (MoAb)
OKT3 (Orthoclone OKT 3R; n = 7) or rabbit antithymocyte
globulin (ATG FreseniusR;n = 4).
Serum samples. Sera of 20 healthy volunteers or bone marrow
donors were used as control. All sera were collected by peripheral
vein puncture or drawing from Hickman catheters and stored
at -6OOC until use. In patients, sera were obtained weekly after
From the Medizinische Klinik III, Klinikum Grosshadern, Universitat Miinchen; Institut f i r Klinische Hamatologie der GSF,
Miinchen; Hauptlaboratorium BASF AG. Ludwigshafen; Knoll
AG. Ludwigshafen; Kinderpoliklinik der Universitat Miinchen;
Medis-Institut der GSF, Neuherberg; and the Institutfur Medizinische Mikrobiologie. Klinikum Grosshadern. Universitat Miinchen,
FRG.
Submitted April 5,1989; accepted October 9, 1989.
Address reprint requests to E . Holler. MD, Medizinische Klinik
III, Klinikum Grosshadern, Marchioninistr.15, 08000 Miinchen
70. FRG.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C.section 1734 solely to
indicate this fact.
0 1990 by The American Society of Hematology.
OO06-4971/90/7504-O012$3.00/0
101 1
1012
HOLLER ET AL
admission to the BMT unit (day -8) until day 21. In patients with
complicated courses, further weekly samples were obtained until
discharge or death. Random samples of patients with uneventful
courses were available after day 21 until day 100. Sera of at least 15
patients were analyzed for each time interval in these patients. In
total, 363 serum samples were included in this study.
Assay of TNFa. TNFa was analyzed by an enzyme-linked
immunosorbent assay (ELISA) using two MoAbs recognizing different epitopes of TNFa.' Specificity of the MoAbs was proven by
absence of crossreactivity with 20 different human tissue specimens.'
Quantitation of recombinant human TNFa (rhTNFa) and natural
human TNFa in supernatants of stimulated mononuclear cell
cultures by ELISA gives identical results as compared with bioassays using L929 cells. The limit of detection is 5 pg/mL as
determined by analysis of rhTNFa. From each sample, at least three
dilutions were analyzed. Mean TNFa levels were calculated and
recorded.
Analysis of bacterial and fungal infections. Surveillance cultures of nose, throat, mouth, skin, and stool were performed twice
weekly. In patients with fever above 38SoC, urine, sputum, and at
least four blood cultures were obtained. Systemic or invasive
infections were diagnosed when bacteria or fungi could be isolated
from blood or tissue.
Analysis of clinical courses and data. Patients were divided in
two groups: patients treated with routine supportive care, antibiotics,
and/or antimycotics and immunosuppressivedrugs were designated
as having no or minor TRC, ie, uneventful courses; and patients
requiring any kind of intensive care support were judged to have
major TRC. Intensive care support included catecholamine treatment due to renal or circulatory impairment and/or oxygen supply/
mechanical ventilation due to respiratory failure.
Major TRC requiring intensive care support in our study were
ELS, VOD, IP, severe microangiopathy resembling thrombotic
thrombocytopenic purpura (MA), and aGVHD, grade IV. ELS was
diagnosed if patients developed increases of body weight (more than
5%) within a few days, due to generalized and/or pulmonary fluid
retention that could not be treated by diuretics alone. VOD and
aGVHD were diagnosed according to clinical riter ria.^^^ IP was
assumed if respiratory failure occurred simultaneously with typical
changes of chest x-rays. Diagnosis of MA included extensive
hemolysis and thrombocytopenic purpura requiring de novo transfusions combined with fluid retention as observed in ELS. Necroscopy
sections were performed in all patients who died. For each patient,
maximal serum levels and time courses of TNFa were recorded.
TNFa levels were compared in the different groups by Wilcoxon
tests. Comparison of overall frequencies was performed by xz
analysis.
(ELS, n = 5; VOD, n = l ) , and 16 patients had later TRC
(IP, n = 9; aGVHD, n = 4; MA, n = 3). TNFa levels
anaylzed at the time of maximal sysmptoms of TRC allowing
clinical diagnosis were pathological in all patients with early
TRC, but only in 3 of 16 patients with IP, M A , and a G V H D
grade IV.
Time courses of TNFa serum levels. Analyses of time
courses of TNFa serum levels, however, revealed significant
increases in all patients with IP, MA, and a G V H D IV before
diagnosis or development of maximal clinical symptoms.
Maximal TNFa serum levels between day - 8 and + 100
after BMT were significantly different between patients with
major TRC and patients with uneventful courses (492 235
pg/mL versus 76 f 29 pg/mL, P<.OOOl, Fig 1). TNFa
maxima preceded clinical diagnosis or maximal symptoms in
patients with IP, aGVHD, and M A with a latency of 24 to 54
days (Table 1). They were closely associated with clinical
symptoms in ELS and VOD. Only 6 of 34 patients with
uneventful courses had increases of TNFa above 100 pg/
mL. In these patients, TNFa maxima were associated with
fever without documented infection (n = 1) or occurred in
the initial phase of minor a G V H D (n = 5). In most patients
with major TRC, TNFa levels were increased for a period of
1300
1200
1100
1000
900
800
700
600
:
500
1-
400
RESULTS
300
TNFa serum levels of healthy controls. TNFa was
detectable in sera of 5 of 20 healthy controls. T h e mean level
was 1 1 pg/mL with a standard deviation of 26 pg/mL (range
10 to 87 pg/mL). Thus, TNFa levels above 100 pg/mL were
considered to be pathological within this study.
TNFa serum levels at the time of diagnosis of TRC. A
total of 22 patients had major TRC in this study. Sixteen
patients had more than one complication in series: in these
patients, early ELS was followed by a G V H D and M A (n =
6) or IP (n = 5), or a combination of both (n = 5 ) . For
statistical analysis, one diagnosis describing the most serious
clinical problem was chosen for each patient: by these
criteria, six patients had TRC occurring before engraftment
200
1
100
I
CONTROLS
N
= 20
$-
P<O.Ol
PTS WITHOUT
COMPLICATIONS
N
=
34
:
P<O.O001
PTS WITH
COMPLICATIONS
N
=
22
Fig 1. Maximal serum levels of TNFa between day 0 and day
100 after BMT. Patients (56) were grouped according to occurrence or absence of TRC end compared with 20 controls. Mean
and individual maxima are shown. Statistical analysis was performed by Wilcoxon tests. Pts, patients.
1013
TNFa RELEASE IN COMPLICATIONS OF EMS
infection, with a mean of 99 f 66 pg/mL, which is lower than
in typical TRC.
Fever and TNFa levels were separately analyzed. During
aplastic phase, there was a weak correlation between absolute TNFa levels and extent of fever (R = .38, not
significant). In this period, 45 patients developed fever
greater than 38SoC, and TNFa levels were pathologic in 23
of them. Moreover, 22 of 23 patients with elevated TNFa
levels suffered from fever greater than 38.5"C. Only one
patient developed fatal aseptic shock and pulmonary ELS at
day +5 after BMT with an acute increase of TNFa to 856
pg/mL in the absence of fever. After engraftment, no
correlation of TNFa levels and fever was observed, as only 5
of 20 patients with pathological increases of TNFa had fever
greater than 38.5"C. In these patients, absence of fever might
be explained by concomitant treatment with corticosteroids
due to aGVHD.
TNFa serum levels and aGVHD. When patients receiving allogeneic transplants were grouped according to severity
of aGVHD, maximal TNFa levels were significantly lower in
patients with absent or grade I aGVHD, as compared with
patients with more severe grades (Table 2). Contribution of
allogeneic stimulation to cytokine release could also be
answered by comparing allogeneic, syngeneic, and autologous BMT. Our study includes only four patients with
syngeneic or autologous grafts not allowing statistical analysis. However, one of these patients developed severe ELS
after autologous BMT, which was associated with a dramatic
increase of TNFa.
Influence of recipient characteristics and increases of
TNFa before BMT on development of TRC. Development
of TRC and increases of TNFa serum levels were not
significantly correlated with age, sex, or diagnosis at the time
of BMT. There was a trend to a higher frequency of
Table 1. Maximal TNFa Serum Levels in Major Transplant
Related Complications Occurring After BMT
Group
n
Maximal
TNF Levels
(pe/mL)
IP
9
431
+ 69
aGVHD
4
462
f
MA
3
472 f 62
ELS
5
558 t 171
1
34
680
76 f 29
VOD
Pts with uneventful courses
Time of
TNF-Maxima.
(Rangel
Time of
Diagnosis.
(Range)
31
( 1 1-00)
20
(4-30)
27
(26-85)
6
(5-12)
5
13
(5-54)
85
(53-150)
45
( 1 2-60)
80
(80-98)
8
(5-15)
8
250
-
Levels were compared with TNFa in patients with uneventful courses.
Mean f SE of TNFa maxima, median and range of time of TNFa-maxima
and time of diagnosis (for IP, VOD, ELS) or maximal clinical symptoms
(for MA, aGVHD) of complications are shown.
Abbreviation: Pts, patients.
*Values are days post-BMT.
2 or more weeks. Therefore, mean TNFa levels were different at any time between day + 8 and day +60 after BMT
from patients with uneventful courses (Fig 2).
Association of TNFa levels with bacterial or fungal
infections and fever. At the time of TNFa maxima, only 10
of 56 patients had systemic or invasive infections. In addition, the overall frequency of infections occurring during the
aplastic phase was analyzed. A total of 20 patients had
documented infections in this period (11 in the group
developing major TRC, 9 in the group with uneventful
courses, difference not significant). Only 8 of these 20
patients had detectable levels of TNFa at the time of
TNF+
Dmll
400
\
300
200
Fig 2. Time course of TNFa
serum levels in 56 patients receiving BMT. Patients were
grouped according to occurrence (x) or absence (0)of TRC.
Mean f SE of all samples analyzed for each time interval are
shown, and differences were
calculated by Wilcoxon tests.
100
-4
0
4
8
12
16
20
40
60
80
100
d
after BMT
HOLLER ET AL
1014
Table 2. Maximal TNFa Serum Levels in Recipients of Allogeneic
BMT: Correlation with Severity of aGVHD
aGVHD
(Grade)
0-1
II
111
IV
n
19
20
7
6
Maximal Serum Levels
of TNFa (+SE, pg/mLI
58
221
306
483
f 22
f
64
f
92
f 185
P
<.02
NS
NS
aGVHD was graded according to clinical criteria as described. Statistical analysis between the groups was performed by Wilcoxon tests.
Abbreviation: NS, not significant.
increased T N F a levels in patients with more advanced
diseases (P = .06). In a total of 51 patients, serum samples
obtained during pretransplant conditioning were available.
Twelve patients had increases of T N F a above 100 pg/mL
before BMT. This pattern of early T N F a release was more
frequently observed in patients with C M L and MDS (38%)
than in patients with other diagnoses (7%, p < 0.01). Increases of T N F a during pretransplant conditioning were
predictive for development of T R C within the first 6 months
after BMT: 10 of these patients (83%) subsequently developed major T R C as compared with 12 of 39 patients (31%)
without increases of TNFa before-BMT (p < 0.01).
Oucome of patients. Only 4 of 22 patients with major
T R C survived beyond 1 year after BMT, while 33 of 34
patients with uneventful courses were alive (P < 0.001).
Causes of death in patients developing major T R C were
ELS/VOD (n = 2), multiorgan failure with adult respiratory distress syndrome (n = 8), or fungal septicemias (n =
6) in the course of TRC, and early relapse after surviving
TRC (n = 2). The only death occurring in patients with
uneventful courses was due to an early relapse.
DISCUSSION
Cytokines have been characterized as the major mediators
of immunity in the last years. Interleukin-1 (IL-1) and
T N F a are predominantly produced by monocytes/macrophages on activation by a variety of stimuli, including
interferon gamma (IFN-7) and bacterial lipopolysaccharid
(LPS).9 Several reports suggest a pathogenetic role of T N F a
in clinical complications such as septic shock” and purpura
fulminans.” A rise of T N F a serum levels has also been
reported before rejection of kidney grafts.’* Here we report a
highly significant association of increases in T N F a serum
levels with major T R C occurring within the first 6 months
after BMT. Differences of T N F a are significant at any time
after BMT when compared with patients with uneventful
courses.
Our data were obtained by a sensitive ELISA using
specific MoAbs. These assays allow detection of immunologically reactive material but frequently lack correlation with
bioassays measuring biologically active ~yt0kines.I~
Determination of cytokines in body fluids such as serum, urine, or
ascites is complicated by the presence of specific inhibitors
for IL-1 l 4 and TNFa.” So far, no systematic studies addressing the influence of inhibitors on cytokine assays in sera have
been reported, and present experiments performed in our
laboratory try to analyze these interactions in sera of BMT
recipients.
A second concern is raised by lability of T N F a , even in
sera cryopreserved a t -7OoC, which has been attributed to
pr0tea~es.I~
We recently started a prospective monitoring of
cytokine levels in fresh sera of BMT recipients: so far, 20
patients have been analyzed between day - 8 and + 100 after
BMT. Absolute T N F a levels were significantly higher in
fresh sera, indicating some loss of activity by cryopreservation, but again we observed a significant correlation of
increases of T N F a with aGVHD and occurrence of TRC.
Maximal T N F a levels were 56 f 26 pg/mL in patients
without aGVHD (n = 6), 480 210 pg/mL in patients with
aGVHD grade I (n = 7), and 1,672 f 746 pg/mL in patients
with aGVHD grade I1 or more (n = 7, P < .05). In addition,
increases of T N F a levels before BMT again proved to be
predictive for development of T R C within the first 4 months
after BMT (P< .01). Thus, loss of T N F a activity by
cryopreservation seems to be a random process and does not
interfere with results obtained in the retrospective analysis.
As most clinical studies reported raised levels of T N F a in
association with microbial or parasitic infections,16 we analyzed documented infections in our patients. However, only
21% of patients had documented infections at the time of
T N F a maxima, and vice versa, TNFa serum levels increased
in only 40%of patients with septicemia without further signs
of TRC. Therefore, infections are unlikely to be the major
cause of cytokine release observed in our study.
In our analysis of microangiopathy occurring in patients
with aGVHD receiving cyclosporin prophylaxis, we postulated cytokine release by monocytes activated via IFN--, in
the course of T cell stimulation. Thus specific immunologic
activation in aGVHD might be augmented by monocyte
activation with subsequent release of cytokines. Other groups
also suggested a role of cytokines as effectors in lesions due to
aGVHD. Piguet showed that histological changes of skin and
gut in the course of aGVHD could be prevented by prophylactic use of a polyclonal antibody neutralizing TNFa in a
mouse model.” In addition, this regimen reduced aGVHDrelated mortality from 100% to 30%without further immunosuppression.
When T N F a levels were analyzed in patients with aGVHD
by other groups, they failed to detect significant levels in
more than 50% of patients.” This is also confirmed by our
study as only 3 of 16 patients with aGVHD or associated
complications had pathological levels a t the time of diagnosis
or maximal symptoms. However, occurrence of raised T N F a
levels in all patients before clinical symptoms, as seen in our
patients, suggests cytokine release to be an early event during
induction of clinical changes. This is further supported by the
close correlation of TNFa maxima with clinical grades of
aGVHD.
IP is a typical complication of the first 2 to 6 months after
allogeneic BMT. So far toxicity (total body irradiation) and
viral (cytomegalovirus) infections have been discussed as the
major pathogenetic mechanisms. Cytokine activity before
occurrence of respiratory changes as observed in our study
might indicate local activation of pulmonary macrophages.
TNFa RELEASE IN COMPLICATIONSOF BMT
1015
Again, Piguet reported a necrotizing alveolitis associated
with increased local production of TNFa in mice with
aGVHD.” A pathophysiological role of TNFa for development of IP would also fit with the data of Niederwieser, who
detected excessive IFN-y release in the course of IP and
cytomegalovirus infections after human BMT.”
Thus, release of TNFa in patients with IP and MA could
be explained by activation of donor cells as a consequence of
aGVHD-associated and/or viral stimulation. Our study,
however, also revealed increases of TNFa early after BMT
while patients are still aplastic, and a close correlation of
early complications as ELS and VOD with cytokine release
was observed. So far, toxic mechanisms have been discussed
in the development of ELS as cyclosporin for pulmonary
capillary leakage ~yndrome,~
while VOD was observed more
frequently in patients with pre-existing liver disease! As in
MA, endothelial cells are the targets of pathogenetic mechanisms in both syndromes. Contribution of TNFa, as suggested by our data, would fit with experimental results
showing TNFa as the predominant cytokine modulating
endothelial functions.21In VOD, local activation of Kupffer
cells might result in damage of small hepatic venules, which
is the characteristic histopathological lesion in this disease.
As only one patient in our study developed VOD, the
association with release of cytokines clearly needs confirmation by a systematic analysis of a larger series of patients.
However, this observation suggests stimulation of host macrophages to be a further pathway of cytokine release in the
setting of BMT. Tissue macrophages and related cells are not
eliminated by pretransplant conditioning and persist up to 4
months after BMT, as demonstrated for alveolar macrophages22and Langerhans cells.23In addition, macrophages
retain their ability to release cytokines after differentiation
from blood monocytes.24Participation of host macrophages
in cytokine release and induction of TRC is strongly supported by a significantly higher risk of TRC in patients with
increases of TNFa before BMT, as seen in our analysis.
As cytokine activation is independent of HLA-restriction,
release of TNFa indicated by positive serum levels might
result from a complex interaction of donor cells and recipient
macrophages. Host macrophages might be stimulated by
pretransplant conditioning or endotoxin as mucosal barriers
are most severely damaged in the early aplastic phase. TNFa
release by host macrophages should change reactivity of
donor cells directly and by induction of HLA-antigens in
recipient tissues. Vice versa, IFN-y and IL-2 released by
activated donor cells might also stimulate recipient macrophages in the course of aGVHD and thus potentiate clinical
changes. These interactions are of particular interest in
patients with CML and MDS. Some BMT centers observed
an increased incidence of TRC in patients transplanted for
CML, and in our experience, the risk of TRC is increased in
patients with MDS. As increases of TNFa before BMT
indicating activation of host macrophages were predominantly observed in patients with these diseases, our data
suggest abnormal reactivity of host macrophages, which
might contribute to an increased cytokine release after BMT.
Finally, the close association of all major TRC with
increases of TNFa raises further questions concerning the
role of endothelial cells in these syndromes. They might be
the primary targets in complicationsassociated with aGVHD,
as they are in early complications occurring during aplastic
phase. ELS resembles acute endothelial damage observed
during immunotherapy with interleukin-225and might result
from direct toxicity of cytokines. Delay between TNFa
maxima and complications occurring after engraftment indicates a more complex interaction of cytokines and target cells
that might involve enhanced turnover of endothelial cells and
macrophages, as well as induction of organ-specific immune
reactions. There is increasing evidence for the central role of
endothelial cells in initiating local immune responses by
expression of various adhesion molecules and production of a
variety of further cytokines.26
In summary, our data indicate a major pathophysiological
role of TNFa in complications occurring during the first 6
months after BMT. Further studies are needed to reveal the
exact pathways of cytokine release with regard to participating cells and involvement of related cytokines. Cytokine
monitoring should help in early detection of major TRC and
allow an earlier and more differentiated treatment in the
future.
ACKNOWLEDGMENT
The authors thank G. Chmel for skillful technical assistance in
performing enzyme immunoassays and thank the nurses of the BMT
unit for their cooperation in collection of serum samples.
REFERENCES
1. Zwaan FE, Gratwohl A, Hermans J, Lyklema A: Transplant
related mortality in leukemia. Report from the Leukemia Working
Party 1988. Bone Marrow Transplantation 3,1:153, 1988 (suppl)
2. Weiner RS, Bortin MM, Gale RP, Gluckman E, Kay HEM,
Kolb HJ, Hartz AJ, Rimm AA: Interstitial pneumonitis after bone
marrow transplantation: Assessment of risk factors. Ann Intern Med
104168,1986
3. Powles R, Pedrazzini A, Crofts M, Clink H, Millar J, Batthia
G, Perez D: Mismatched family bone marrow transplantation.
Semin Hematol21:182, 1984
4. Jones RJ, Lee SK, Beschorner WE: Veno-occlusive disease of
the liver following bone marrow transplantation. Transplantation
44:778, 1987
5. Holler E, Kolb HJ, Hiller E, Mraz W, Lehmacher W, Gleixner
B, Seeber Ch, Jehn U, Gerhartz HH, Brehm G, Wilmanns W:
Microangiopathy in patients on cyclosporin prophylaxis who developed acute graft-versus-host disease after HLA-identical bone
marrow transplantation. Blood 73:2018, 1989
6. Storb R, Thomas ED: Graft-versus-host disease in dog and
man: The Seattle experience. Immunol Rev 88215, 1985
7. Kapp A, Textor A, Krutmann J, Moller A: Immuno-modulating cytokines in atopic dermatitis and psoriasis: Production of tumor
necrosis factor/cachectin and lymphotoxin by mononuclear cells in
vitro. Br J Dermatol (in press)
8. Moller P, Mielke B, Heilig B, Dorken B, Moiler A Tumor
necrosis factor alpha and lymphotoxin, physiologically produced by
a subset of mantel zone lymphocytes, is detectable in a subset of
B-cell neoplasias including hairy cell leukemia. Blood (in press)
1016
9. Gifford GE, Flick DA: Natural production and release of
tumour necrosis factor, in Bock G, Marsh J (eds): Tumour Necrosis
Factor and Related Cytokines, Ciba Foundation Symposium 131.
New York, NY, Wiley, 1987, p 3
10. Tracey KJ, Fong Y, Hesse DG, Manogue KR, Lee AT, Kuo
GC, Lowry SF, Cerami A: Anticachectin/TNF monoclonal antibodies prevent septic shock during lethal bacteremia. Nature 330:662,
1987
11. Girardin E, Grau GE, Dayer JM, Roux-Lombard P, The J5
study group, Lambert PH: Tumour necrosis factor and interleukin 1
in the serum of children with severe infectious purpura. N Engl J
Med 319:397,1988
12. Maury CPJ, Teppo AM: Raised serum levels of cachectin/
tumor necrosis factor alpha in renal allograft rejection. J Exp Med
166:1132,1987
13. Balkwill F, Burke F, Talbot D, Tavernier J, Osborne R,
Naylor S, Durbin H, Fiers W: Evidence for tumour necrosis
factor/cachectin production in cancer. Lancet 2:1229,1987
14. Larrick JW: Native interleukin 1 inhibitors. lmmunol Today
10:61, 1989
15. Olsson I, Lantz M, Nilsson E, Peetre C, Thysell H, Grubb A,
Adolf G: Isolation and characterization of a tumor necrosis factor
binding protein from urine. Eur J Haematol42:270,1989
16. Scudari P, Sterling KE, Lam KS, Finley PR, Ryan KJ, Ray
CG, Petersen E, Slymen DJ, Salmon SE: Raised serum levels of
tumor necrosis factor in parasitic infections. Lancet 1:1364, 1986
17. 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-versus-host disease. J Exp Med 166:1280, 1987
18. lrle C, Salamin AF, Bacigalupo A, Chapuis B, Gratwohl A,
HOLLER ET AL
Hows J, Piguet P, Grau G: Serum T N F levels during graft-versushost disease after allogeneic bone marrow transplantation. Bone
Marrow Transplantation. 1:127, 1988 (suppl3)
19. Piguet PF, Grau GE, Collart M, Vassalli P, Kapanci Y:
GvHR induce an alveolitis involving TNFalpha. Bone Marrow
Transplantation. 1:111, 1988 (suppl3)
20. Niederwieser D, Auback J, Hajek-Rosenmayr A, Schonitzer
D, Schendl D, Huber CH: The role of cytokines and MHC antigens
in human GvHD: Analysis of T-cell specificities and endogenous
cytokine levels. Bone Marrow Transplantation 2:84, 1989 (suppl 4)
21. Pober JS: Effects of tumor necrosis factor and related
cytokines on vascular endothelial cells, in Bock G, Marsh J (eds):
Tumor Necrosis Factor and Related Cytokines, Ciba Foundation
Symposium 131. New York, NY, Wiley, 1987, p 170
22. Thomas ED, Ramberg RE, Sale G E Direct evidence for a
bone marrow origin of the alveolar macrophage in man. Science
192:1016,1976
23. Perreault C, Pelletier M, Belanger R, Boileau J, Bonny Y,
David M, Gyger M, Landry D, Montplaisir S: Persistence of host
Langerhans cells following allogeneic bone marrow transplantation:
Possible relationship with acute graft-versus-host disease. Br J
Haematol60253,1985
24. Johnston RB: Current concepts: Immunology-Monocytes
and Macrophages. N Engl J Med 318,12:747,1988
25. Cotran RS, Pober JS, Gimbrone MA, Springer TA, Wiebke
EA, Gaspari A, Rosenberg SA, Lotze MT: Endothelial activation
during interleukin 2 immunotherapy: A possible mechanism for the
vascular leak syndrome. J Immunol 193: 1883, 1987
26. Cotran RS: New roles for the endothelium in inflammation
and immunity. Am J Pathol 129:407, 1987