Leukemia (1999) 13, 1434–1440
 1999 Stockton Press All rights reserved 0887-6924/99 $15.00
http://www.stockton-press.co.uk/leu
A high serum soluble Fas/APO-1 level is associated with a poor outcome of
aggressive non-Hodgkin’s lymphoma
N Niitsu, K Sasaki and M Umeda
First Department of Internal Medicine, Toho University School of Medicine, 6-11-1 Omori-Nishi, Ota-ku, Tokyo 143-0015, Japan
Soluble Fas (sFas) in the serum is believed to be able to inhibit
apoptosis of lymphocytes. It has been reported that the serum
sFas level is increased in various diseases, including malignant lymphoma and systemic lupus erythematosus. We studied
the association between sFas and the prognosis of patients
with aggressive non-Hodgkin’s lymphoma (NHL). Compared
with normal controls, the serum sFas level was increased significantly in patients with aggressive NHL and adult T cell
leukemia/lymphoma. Among patients with aggressive NHL, the
complete remission rate was significantly decreased in the
subgroup having high serum sFas levels. Both the overall survival rate and the disease-free survival (DFS) rate were significantly lower for this subgroup than for patients with low serum
sFas levels. The 5-year survival rates estimated by the Kaplan–
Meier method for patients with high and low serum sFas levels
were 27.6% and 68.3%, respectively (P ⴝ 0.0001). The 5-year
DFS rates estimated for patients with high and low serum sFas
levels were 44.7% and 71.9%, respectively (log-rank test:
P ⴝ 0.0023, and generalized Wilcoxon test: P ⴝ 0.0014). Among
patients with a low and low–intermediate risk group according
to the International Prognostic Index (IPI), the 5-year survival
rates for low and high serum sFas subgroups were 72.8% and
42.0%, respectively, showing a significant difference (Wilcoxon
test: P ⴝ 0.0163, log-rank test: P ⴝ 0.0115). Among patients
with a high–intermediate and high risk group, the 5-year survival rates for low and high serum sFas subgroups were 51.6%
and 17.4%, respectively, again showing a significant difference
(Wilcoxon test: P ⴝ 0.0001, log-rank test: P ⴝ 0.0002). Multivariate analysis of a series of prognostic factors, including the five
used to calculate the IPI, showed that the serum sFas level was
an independent prognostic factor for the overall survival.
Based on these results, a serum sFas level of 10 ng/ml or more
can be considered to indicate a poor prognosis in patients with
advanced NHL, and this finding may be useful for developing
strategies for further treatment.
Keywords: soluble Fas; non-Hodgkin’s lymphoma; prognostic factor; COP-BLAM therapy
Introduction
The cell-surface protein Fas/APO-1 is a member of the nerve
growth factor (NGF)/tumor necrosis factor (TNF) receptor
superfamily. Fas/APO-1 is expressed by a variety of human B
and T cell lines, as well as well as by many different tumor
cells and various normal human tissues. Fas/APO-1 is a particularly important receptor, since its triggering by antibodies or
Fas-ligand may cause apoptosis in susceptible cells.1–4 Soluble
isoforms of the Fas molecule (sFas) have been identified5 and
Cheng et al5 have reported an increased serum concentration
of sFas in patients with systemic lupus erythematosus (SLE).
The sFas molecule is considered to inhibit apoptosis of lymphocytes, and it has been reported that its serum level is elevated in patients with malignant lymphoma and lymphocytic
leukemia.6
In recent years, treatment for non-Hodgkin’s lymphoma
Correspondence: N Niitsu: Fax: 03–3763–8298
Received 9 March 1999; accepted 25 May 1999
(NHL), particularly intermediate- and high-grade NHL, has
been chosen depending on the level of risk as defined by the
International Prognostic Index (IPI), calculation of which is
based on five independent prognostic factors (age, performance status (PS), number of extranodal lesions, Ann Arbor
stage, and serum lactate dehydrogenase level).7 Standard therapy, in which the CHOP (cyclophosphamide, doxorubicin,
vincristine, prednisone) regimen plays a central role, is indicated for patients having an index of 0 or 1 (the low (L) risk
group) and an index of 2 (the low–intermediate (L–I) risk
group). For patients in the high–intermediate (H–l) and high
(H) risk groups, who have three or more of the five IPI risk
factors, high-dose chemotherapy is administered in combination with peripheral blood stem cell transplantation. Even
if NHL is treated in accordance with these criteria, however,
therapy is not effective for some patients in the L and L–I risk
groups, but is effective for some patients in the H–l and H risk
groups. Consequently, it is doubtful whether all of the patients
in the H–l and H risk groups require such high-dose chemotherapy.
In the present study, we determined serum sFas levels in all
patients with aggressive NHL, and the association with various
outcomes was investigated. We found that a high serum sFas
level is related to a poor prognosis, particularly in the H–l and
H risk groups.
Materials and methods
Patients
sFas was measured in 175 consecutive untreated patients who
were managed at the 1st Department of Internal Medicine,
Toho University School of Medicine (Tokyo, Japan) from 1987
to 1996. Of the 175 patients, 158 had NHL. Twenty-seven
patients had indolent type, 131 had aggressive lymphoma
according to the REAL schema.8 Indolent lymphoma was of
the B cell lineage in all patients, including two with lymphoplasmacytic lymphoma, two with splenic marginal zone lymphoma, seven with marginal zone B cell lymphoma, and 16
with follicle center lymphoma (grade I/II). Aggressive lymphoma was of the B cell lineage in 99 patients and of the T
cell lineage in 32. The aggressive B cell lymphoma was
mantle cell lymphoma in six patients, follicle center lymphoma (grade III) in 18, and diffuse large B cell lymphoma in
75. Aggressive T cell type was peripheral T cell lymphoma
in 25 patients, angioimmunoblastic lymphoma (AIL) in two,
angiocentric lymphoma in one, and anaplastic large cell (T
and null cell types) in four (two were ALK-positive and two
were ALK-negative). Among patients with aggressive and
highly aggressive lymphoma, those receiving COP-BLAM
therapy were scheduled to be included in this study, but no
patient with highly aggressive lymphoma was actually
enrolled.
In addition, nine patients had Hodgkin’s disease and eight
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
others had adult T cell leukemia/lymphoma (ATLL), excluding
the lymphoma type of disease. The median age was 64 years
(range: 18–80 years). Clinical staging was performed according to the Ann Arbor classification system.9 Evaluation
included a complete history and physical examination: chest
roentgenography; bone marrow aspiration and biopsy; computed tomography of the chest, abdomen and pelvis; hemogram and differential counts; and routine biochemistry tests.
Laparotomy was not performed for staging. According to the
IPI, the risk was L in 34, L–l in 55, H–l in 30, and H in 12
patients. The 175 patients were treated with combination
chemotherapy. Patients with aggressive lymphoma and disseminated disease were usually treated with COP-BLAM10
(cyclophosphamide, vincristine, prednisone, bleomycin,
doxorubicin and procarbazine), or biweekly COP-BLAM with
granulocyte colony-stimulating factor (G-CSF).11 Indolent lymphoma was treated with COP (cyclophosphamide, vincristine,
prednisone) or COP-BLAM regimen. All patients were followed-up at intervals of a few months. Re-evaluation included
physical examination, hemogram and differential counts, biochemistry tests, and computed tomography of the chest, abdomen and pelvis. The median follow-up time was 66 months
(range: 23–122 months). Of the 131 patients with aggressive
NHL, 99 are still alive after follow-up for 88–120 months
(median: 68 months). The other 32 patients died between 1
and 34 months (median: 12 months) after diagnosis.
Plasma samples were obtained from 34 healthy volunteers
with a mean age of 62 years (range: 19–84 years) for comparison. Samples were collected if the controls had not had fever
within 1 week, were not receiving any medication, were not
known to be pregnant, and did not have a history of any
chronic or acute illnesses.
Enzyme-linked immunosorbent assay (ELISA) for sFas
A sandwich enzyme immunoassay (sFas ELISA kit; MBL,
Nagoya, Japan) was used to determine the serum sFas level.
Briefly, 10 ␮l of serum was incubated with a biotin-labeled
monoclonal antibody against sFas for 2 h in precoated
microwells. After washing, streptavidin–horseradish peroxidase was added for 1 h and development was done with TMB
substrate solution. The reaction was then stopped and the
absorbance determined at 450 nm. The sFas level (␮g/ml) was
calculated from a standard curve. Data are shown as the mean
values of triplicate samples.
Soluble interleukin-2 receptor (sIL-2R) levels were also measured with a sandwich ELISA, as described elsewhere.12
eralized Wilcoxon’s tests.14 Differences between groups were
evaluated by the Mann–Whitney’s U test (nonparametric
analysis),15 and P ⬍ 0.05 was taken to indicate significance.
Multivariate analysis of the prognosis was performed using
Cox’s proportional-hazards regression model.16 All calculations were performed with SAS software (version 6.10; SAS
Institute, Cary, NC, USA).
Results
Serum sFas level in patients and healthy controls
The mean (± s.d.) serum sFas level of the healthy controls was
1.88 ± 0.87 ng/ml. The mean level for the patients with
aggressive NHL and ATLL was 7.73 ± 6.20 ng/ml
(P = 0.00001) and 14.29 ± 3.24 ng/ml (P = 0.00001), respectively, and these levels were higher than the control value. The
mean sFas level for patients with Hodgkin’s disease and indolent NHL was 4.74 ± 1.73 ng/ml and 5.67 ± 3.91 ng/ml,
respectively, and these levels were not significantly different
from the control value (Figure 1). Serum sFas levels were
significantly
correlated
with
serum
LDH
levels
(Y = 6.722 + 0.0014X, r = 0.2943, P = 0.00065) and sIL-2R
levels (Y = 6.642 + 0.0001X, r = 0.3664, P = 0.000017)
(Figure 2).
sFas and the clinical characteristics of aggressive NHL
(Table 1)
Table 1 shows the clinical characteristics of the 131 patients
with aggressive NHL. An elevated serum sFas level before
treatment was correlated with poor prognostic features, such
as a poor performance status, an elevated LDH levels, Ann
Arbor stage III or IV, H–l and H risk groups (IPI), and an elevated serum (sIL-2R) levels (P ⭐ 0.01 for all comparisons,
Mann–Whitney U test) (Table 1).
CR was achieved in 110 (84%) of the 131 patients. These
110 patients had an sFas level of 6.73 ± 5.17 ng/ml at diagnosis, which was significantly lower than the mean level
(12.99 ± 8.35 ng/ml) of those who showed a partial response
Statistical analysis
The response to treatment was documented on completion of
chemotherapy. Complete remission (CR) was defined as the
absence of detectable disease based on clinical, radiologic
and histologic criteria. Partial remission (PR) required a ⬎50%
reduction of tumor volume. The duration of CR was calculated
from the completion of chemotherapy to relapse or last follow-up. Overall survival (OS) was the interval from initiation
of therapy to the time of death or last follow-up. Survival
analysis was performed according to the Kaplan–Meier
method.13 Disease-free survival (DFS) was defined as the
absence of any recurrence, so patients who never reached CR
had a DFS of 0 months. The statistical significance of differences in survival was determined by the log-rank and gen-
Figure 1
Serum levels of soluble Fas protein in malignant lymphoma (n = 175) and healthy controls (n = 34) (P = 0.0001; Mann–
Whitney U test). Horizontal bars indicate mean values. NHL, nonHodgkin’s lymphoma; ATLL, adult T cell leukemia/lymphoma.
1435
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
1436
Table 1
Correlation of serum levels of sFas with clinical and prognostic characteristics in aggressive non-Hodgkin’s lymphoma
Characteristic
Figure 2
(a) Relationship between sFas and LDH levels. (b)
Relationship between sFas and sIL-2R levels.
or failed to respond (P = 0.003). There was no correlation
between the serum sFas level and any histologic type in the
REAL, nor was there any difference of the serum sFas level
between T cell and B cell NHL.
Assessment of the correlation of sFas with overall
survival and DFS in aggressive NHL
The 131 patients with aggressive NHL were divided into
groups having different sFas levels at a cut-off value of
10 ng/ml. Both overall and DFS rates were calculated for
patients falling into each 1 ng/ml interval from an sFas concentration of 3 ng/ml upward. Both rates were significantly different for patients having sFas levels higher than 10 ng/ml
from those for patients with levels of 10 ng/ml or less. Consequently, an sFas concentration of 10 ng/ml was taken as the
cut-off value.
The 5-year survival rates for the high and low sFas groups
were 27.6% and 68.3%, respectively (P = 0.0001 for both logrank test and generalized Wilcoxon test), with DFS being
44.7% and 71.9% (log-rank test: P = 0.0023, and generalized
Wilcoxon test: P = 0.0014) (Figures 3a, 4a). Patients were next
divided into two groups by the IPI score (L plus L–I and H–I
plus H risk groups), and overall survival and DFS were calculated based on the sFas level in each group.
Among patients who were from the L + L–I risk groups
according to the IPI, the 5-year survival rates for the low and
high serum sFas level subgroups were 72.8% and 42.0%,
respectively, and the difference was significant (Wilcoxon test:
P = 0.0163, log-rank test: P = 0.0115) (Figure 3b). Similarly,
among patients from the H–I + H risk group, the 5-year sur-
Age at diagnosis
⬍60 years
⭓60 years
Gender
Male
Female
WHO performance status
0, 1
2苲4
Serum lactate dehydrogenase
level
Normal
⬎Normal
Ann Arbor stage
I, II
III, IV
Extranodal sites
⭐1
⬎1
International Prognostic Index
low/low-intermediate
high–intermediate/high
Immunophenotype
T
B
Soluble interleukin-2 receptor
level
⬍1000 U/ml
⭓1000 U/ml
Bulky mass
Present
Absent
Response to treatment
Complete remission
No complete remission
a
No. of sFas (ng/ml) P valuea
patients mean ± s.d.
51
80
6.51 ± 5.08 0.0569
8.51 ± 6.73
75
56
8.25 ± 6.75 0.2732
7.04 ± 5.37
117
14
6.92 ± 5.41 0.0046
14.51 ± 8.27
35
96
5.10 ± 3.81 0.0002
8.69 ± 6.63
54
77
6.18 ± 5.83 0.0157
8.82 ± 6.26
123
8
7.67 ± 6.16 0.6425
8.72 ± 7.20
89
42
6.56 ± 5.73 0.0015
10.20 ± 6.50
32
99
8.97 ± 6.20 0.1951
7.33 ± 6.18
71
60
5.71 ± 4.97 0.0001
10.12 ± 6.68
21
110
8.33 ± 7.36 0.6328
7.62 ± 5.99
110
21
6.73 ± 5.17 0.0030
12.99 ± 8.35
Mann–Whitney test.
vival rates for the low and high serum sFas level subgroups
were 51.6% and 17.4%, respectively (Wilcoxon test:
P = 0.0001, log-rank test: P = 0.0002), again showing a significant difference (Figure 3c). Among patients from the L + L–
I risk group, the 5-year DFS rates for the low and high serum
sFas level subgroups were 74.6% and 54.7%, respectively,
without any significant difference (Figure 4b). Among patients
from the H–I + H risk group, however, the 5-year DFS rate for
the low serum sFas level subgroup was 60.8%, which was
significantly higher than that of the high serum sFas level subgroup (33.3%) (Wilcoxon test: P = 0.0156 log-rank test:
P = 0.0282) (Figure 4c).
Univariate and multivariate analyses of overall and
DFS in patients with aggressive NHL
The overall survival was significantly worse for patients who
were 60 years or older and had a WHO PS of 2苲4, a serum
LDH level that was normal or raised, belonged to the IPI H–
I + H risk group, had T cell NHL, had an sIL-2R of 1000 U/ml
or higher, and had a serum sFas level of 10 ng/ml or higher.
The DFS rate was significantly lower in patients with a PS of
2–4, normal or raised serum LDH levels, T cell NHL, and an
sFas level of 10 ng/ml or higher (Table 2). Multivariate analysis
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
1437
Figure 3
Overall survival curves of patients with intermediate- and
high-grade non-Hodgkin’s lymphoma. (a) All patients; (b) low and
low–intermediate risk group; (c) high–intermediate and high risk
group.
of serum sFas and the five factors used to calculate the IPI
was performed with the Cox proportional hazards regression
model and showed that the Fas level was an independent
prognostic factor for both overall survival and DFS (Table 3).
Discussion
Fas antigen is a member of the TNF receptor superfamily, and
can induce apoptosis in cells after binding by its natural ligand
(Fas L) or by anti-Fas antibody. There are two subsets of Fas
antigen, membrane Fas occurring on the surface of cells and
soluble Fas that lacks the transmembrane domain. These two
Fas molecules are synthesized by corresponding mRNAs that
differ from each other in splicing. It is known that the serum
sFas level is increased in hepatitis B virus infection, SLE,5
Figure 4
Disease-free survival curves of patients with intermediate- and high-grade non-Hodgkin’s lymphoma. (a) All patients; (b) low
and low–intermediate risk group; (c) high–intermediate and high
risk group.
acute myelogenous leukemia,17 B and T cell leukemia, and
lymphoma.6
The sFas level was reported to be increased in AML with
active disease, and was not associated with any other laboratory parameter. In AML, accordingly, sFas may be secreted or
shed by reactive or stromal cells rather than being produced
by leukemic cells.17
Among NHL patients, elevation of serum sFas has been
reported in centroblastic and centrocytic B cell lymphoma,6
but was not correlated with the karyotype or with the level of
anti-apoptotic bcl-2 protein. NHL is considered to be resistant
to chemotherapy when the serum sFas level is high. Kondo et
al18 have reported that both bcl-2 and an apoptosis-inducing
receptor (CD95) are expressed by many low-grade lymphomas. Overexpression of bcl-2 is considered to be a prog-
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
1438
Table 2
Univariate analysis of prognostic factors in aggressive non-Hodgkin’s lymphoma
Characteristic
No. of
patients
5-year
survival (%)
P value
5-year DFS
(%)
Wilcoxon test Log-rank test
Age at diagnosis
⬍60 years
⭓60 years
Gender
Male
Female
WHO performance status
0, 1
2苲4
Serum LDH level
Normal
⬎Normal
Ann Arbor stage
I, II
III, IV
Extranodal sites
⭐1
⬎1
International Prognostic Index
low/low-intermediate
high-intermediate/high
Immunophenotype
T
B
Soluble interleukin-2 receptor
level
⬍1000 U/ml
⭓1000 U/ml
Soluble Fas level
⬍10 ng/ml
⭓10 ng/ml
Overall survival
Performance status (0, 1/2苲4)
Lactate dehydrogenase
(normal/⬎normal)
sFas (⬍10/⭓10)
Age (⬍60/⭓60 yr)
Extranodal disease (⬎1 site)
Stage (I, II/III, IV)
Disease-free survival
Performance status (0, 1/2苲4)
Lactate dehydrogenase
(normal/⬎normal)
Extranodal disease (⬎1 site)
sFas (⬍10/⭓10)
Stage (I, II/III, IV)
Age (⬍60/⭓60 yr)
Wilcoxon test Log-rank test
51
80
69.2
51.9
0.0238
0.0247
69.6
64.0
0.5633
0.5176
75
56
55.4
62.6
0.9333
0.8387
63.5
70.0
0.5202
0.5379
117
14
62.7
21.8
0.0001
0.0001
69.3
21.4
0.0003
0.0002
35
96
83.9
48.7
0.0093
0.0047
87.0
57.7
0.0284
0.0153
54
77
62.3
56.4
0.0890
0.1607
71.2
62.2
0.0613
0.1345
123
8
59.1
51.4
0.2471
0.3575
66.1
75.0
0.9646
0.9233
89
42
68.0
35.6
0.0006
0.0004
71.4
51.1
0.0924
0.0654
32
99
31.1
65.8
0.0096
0.0024
43.4
71.7
0.0032
0.0057
71
60
73.0
37.4
0.0005
0.0006
73.1
54.2
0.0718
0.0697
97
34
68.3
27.6
0.0001
0.0001
71.9
44.7
0.0014
0.0023
Table 3
Cox’s proportional hazards regression model for overall
survival and disease-free survival with sFas and factors identified by
the International Prognostic Index
Factor
P value
Hazards
ratio
P value
3.710
2.927
0.0021
0.0295
2.885
2.072
2.016
1.419
0.0020
0.0448
0.2765
0.2846
4.491
2.908
0.0046
0.0541
2.468
2.069
1.653
1.343
0.3944
0.0757
0.1930
0.4352
nostic determinant for low-grade lymphoma, because overexpression partially blocks apoptosis induced by Fas.19
The present study demonstrated that the serum sFas level
was higher in patients with aggressive NHL and ATLL than in
healthy controls. In aggressive NHL, the sFas level varied
widely, ranging from 1 to 32.25 ng/ml (mean ± s.d.:
7.73 ± 6.2 ng/ml). Thus, sFas remained within the normal
range in some patients and was elevated in others. In ATLL,
in contrast, the sFas level ranged from 8.6 to 18.1 ng/ml
(mean ± s.d.: 14.29 ± 3.24 ng/ml) and was elevated in all
patients. Kamihira et al20 studied serum sFas levels in human
retroviral infection and diseases associated with such infection, and they reported that the level in carriers of human T
cell leukemia virus type-I (HTLV-I) was comparable to that in
healthy controls, but sFas level was elevated in patients with
ATL and AIDS (P ⬍ 0.01). These findings are consistent with
our data. All patients with ATLL died within 2 years. Recently,
it has been reported that sFas was increased in angioimmunoblastic T cell lymphoma (AIL).21 The serum sFas levels of
two patients with AIL included in the present study were 10.2
and 14.6 ng/ml (data not shown). In addition, it has been
reported that the serum sFas level was higher in men than in
women, and that it increased with age. In our patients with
aggressive NHL, however, the sFas level did not depend on
either age or sex,22 but there was a significant correlation
between serum sFas and sIL-2R levels. These two receptors
are probably produced by T cells and by lymphoma cells that
have undergone the same activation process.
In recent years, various prognostic factors for NHL have
been studied. The IPI is a typical one. In addition, serum cytokines such as soluble IL-2R,12 IL-623 and IL-10,24 and adhesion
molecules have been studied. Soluble ICAM-125 has been
reported to be related to the tumor burden and disease
activity, whereas soluble CD4426 is associated with the prog-
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
nosis of diffuse large cell lymphoma. Mutations of the p53
gene have been shown to be associated with a poor prognosis
of aggressive B cell lymphoma.27 However, p53 mutations
does not seem to have any prognostic value in patients from
the IPI H–I and H risk groups.
In the present study, we found that many patients with
aggressive lymphoma had elevated pretreatment sFas levels.
The sFas level was associated with the response to treatment,
and patients with a low pretreatment sFas level often achieved
CR. In contrast, a high pretreatment sFas level was associated
with poor survival among our unselected patients with
aggressive NHL given the COP-BLAM regimen. A high sFas
level was particularly associated with a poor performance
status and a high Ann Arbor stage.
The sFas level was slightly higher in bulky than in nonbulky lymphoma, but the difference was not significant.
Consequently, sFas seems to be associated with tumor dissemination rather than tumor burden. The prognosis of T cell
lymphoma was worse than that of B cell lymphoma, but there
was no significant difference of the sFas level between them,
indicating that factors other than sFas were involved in
determining the prognosis. Munker et al28 reported that the
sCD95 level was higher in patients with low- and intermediate-grade NHL than in normal controls (P ⬍ 0.01), and that
stratifying patients based on a cut-off value of 75 U/ml had
no impact on progression coupled with the high incidence of
recurrence considered to account for the comparable progression-free survival rates.
The association of overall and DFS with the serum sFas
level was assessed according to IPI risk category. Among the
89 patients in the L and L–I risk groups, the serum sFas level
was 10 ng/ml or higher in 15. The overall survival rate of these
15 patients was significantly lower than that of the others, but
there was no significant difference in DFS between the two
groups. The serum sFas level was 10 ng/ml or higher in 19 of
the 42 patients from the H–I and H risk group. Both overall
and DFS rates were significantly lower for these 19 patients
than for the others.
According to Shipp et al,7 the 5-year survival rate for the IPI
L risk group was 73% after treatment with the CHOP regimen
(cyclophosphamide,
doxorubicin,
vincristine,
and
prednisone), whereas the rate for the L–I risk group was 51%.
The rates for the H–I and H risk groups were only 43% and
26%, respectively.7 Therefore, they recommended CHOP as
standard therapy for the L and L–I risk groups, but this regimen
was not necessarily regarded as adequate for the IPI H–I and
H groups. Accordingly, further investigation is necessary to
develop strategies for the treatment of patients in these risk
groups. Clinical studies have been conducted on high-dose
chemotherapy administered in combination with peripheral
blood stem cell transplantation (PBSCT), particularly in young
patients, and it has been reported that such therapy was effective.29,30 Among patients in the H–I and H risk groups, however, some do not relapse even after administration of CHOP
or COP-BLAM alone. This raises the problem of whether all
these patients actually need high-dose induction therapy. The
present study showed that a serum sFas level of 10 ng/ml or
higher may predict a poor prognosis for overall and DFS. With
respect to overall survival, the duration was decreased in
patients from both the L + L–I and H–I + H risk groups when
the serum sFas level was 10 ng/ml or higher. Even when the
serum sFas level was elevated, DFS was not significantly
worse in the L + L–I risk group. However, the DFS was significantly poorer when the serum sFas level exceeded 10 ng/ml
in the H–I + H risk group.
Based on these findings, if the serum sFas level exceeds
10 ng/ml in the H–I + H risk group, high-dose chemotherapy
in combination with PBSCT may be indicated, because the
prognosis is significantly worse.
We performed multivariate analysis using sFas levels and
the prognostic factors used to calculate the IPI in relation to
the overall survival and DFS. These analyses showed that a
high serum sFas level was an independent prognostic factor
for both overall survival and DFS (Table 3).
Consequently, if the sFas is used to estimate the outcome
in new patients, more appropriate treatment for NHL may
be selected.
References
1 Oehm A, Behrmann I, Falk W, Pawlita M, Maier G, Klas C, LiWeber M, Richards S, Dhein J, Trauth BC. Purification and molecular cloning of the APO-1 cell surface antigen, a member of
the tumor necrosis factor/nerve growth factor receptor superfamily.
Sequence identity with the sFas antigen. J Biol Chem 1992; 267:
10709–10715.
2 Leithauser F, Dhein J, Mechtersheimer G, Koretz K, Bruderlein S,
Henne C, Schmidt A, Debatin KM, Krammer PH, Moller P. Constitutive and induced expression of APO-1, a new member of the
nerve growth factor/tumor necrosis factor receptor superfamily, in
normal and neoplastic cells. Lab Invest 1993; 69: 415–429.
3 Debatin K-M, Goldman CK, Waldman TA, Krammer PH. APO-1induced apoptosis of leukemia cells from patients with adult Tcell leukemia. Blood 1993; 81: 2972–2977.
4 Trauth BC, Klas C, Peter AM, Matzku S, Moller P, Falk W, Debatin
KM, Krammer PH. Monoclonal antibody-mediated tumor
regression by induction of apoptosis. Science 1989; 245: 301–
305.
5 Cheng J, Zhou T, Liu C, Shapiro JP, Brauer MJ, Kiefer MC, Barr
PJ, Mountz JD. Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule. Science 1994; 263: 1759–1762.
6 Knipping E, Debatin KM, Stricker K, Heilig B, Eder A, Krammer
PH. Identification of soluble APO-1 in supernatants of human Band T-cell lines and increased serum levels in B-and T-cell leukemias. Blood 1995; 85: 1562–1569.
7 Shipp HA et al. A predictive model for aggressive non-Hodgkin’s
lymphomas. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. New Engl J Med 1993; 329: 987–994.
8 Hiddemann W, Coiffer B, Fisher RI, Cabanillas F, Cavali F, Nadler
LM, De Vita VT, Lister TA, Armitage JO. Lymphoma classification:
the gap between biology and clinical management is closing.
Blood 1996; 88: 4085–4089.
9 Carbone PP, Kaplan HS, Musshoff K, Smithers DW, Tubiana M.
Report of the Committee on Hodgkin’s Disease Staging Classification. Cancer Res 1971; 31: 1860–1861.
10 Niitsu N, Umeda M. COP-BLAM regimen combined with granulocyte colony-stimulating factor (G-CSF) and high-grade non-Hodgkin’s lymphoma. Eur J Haematol 1995; 55: 88–92.
11 Niitsu N, Umeda M. Biweekly COP-BLAM (cyclophosphamide,
vincristine,
prednisone,
bleomycin,
doxorubicin
and
procarbazine) regimen combined with granulocyte colonystimulating factor (G-CSF) for intermediate- and high-grade nonHodgkin’s lymphoma. Eur J Haematol 1996; 56: 163–167.
12 Pui C-H, lp SH, Kung P, Dodge RK, Berard CW, Crist WM, Murphy SB. High serum interleukin-2 receptor levels are related to
advanced disease and a poor outcome in childhood nonHodgkin’s lymphoma. Blood 1987; 70: 624–628.
13 Kaplan EL, Meier P. Nonparametric estimation from incomplete
observations. J Am Stat Assoc 1958; 53: 457–481.
14 Wilcoxon F. Individual comparisons by ranking methods. Biometrics Bull 1945; 1: 80–83.
15 Mantel N. Evaluation of survival data and two new rank order
statistics arising in its consideration. Cancer Chemother Rep 1966;
50: 163–170.
16 Cox DR. Regression models and life tables. J Stat Soc Brit 1972;
34: 187–202.
17 Munker R, Midis G, Owen-Schanb L, Andreff M. Soluble FAS
1439
High serum soluble Fas/APO-1 level and poor outcome of aggressive NHL
N Niitsu et al
1440
18
19
20
21
22
23
24
25
(CD95) is not elevated in the serum of patients with myeloid leukemias, myeloproliferative and myelodysplastic syndromes. Leukemia 1996; 10: 1531–1533.
Kondo E, Yoshino T, Yamadori I, Matsuo Y, Kawasaki N,
Minowada J, Akagi T. Expression of bcl-2 protein and Fas antigen
in non-Hodgkin’s lymphoma. Am J Pathol 1994; 145: 330–337.
Reed JC. Bcl-2: prevention of apoptosis as a mechanism of drug
resistance. Hematol Oncol Clin N Am 1995; 9: 451–473.
Kamihira S, Yamada Y, Hiragata Y, Yamaguchi J, Izumikawa K,
Sugahara K, Tsuruta K, Atogami S, Tsukasaki K, Maeda T, Tomonaga M. Serum levels of soluble Fas/APO-1 receptor in human
retroviral infection and associated diseases. Intern Med 1997; 36:
166–170.
Yufu Y, Choi I, Itirase N, Tokor A, Noguchi Y, Goto T, Uike N,
Kozuru M. Soluble Fas in the serum of patients with non-Hodgkin’s lymphoma: higher concentrations in angioimmunoblastic
T-cell lymphoma. Am J Hematol 1998; 58: 334–336.
Seishima M, Takemura M, Saito K, Sano H, Minatoguchi S, Fujiwara H, Hachiya T, Noma A. Highly sensitive ELISA for soluble
Fas in serum: increased soluble Fas in the elderly. Clin Chem
1996; 42: 1911–1914.
Preti HA, Cabanilas F, Talpaz M, Tucker SL, Seymour JF, Kurzrock
R. Prognostic value of serum interleukin-6 in diffuse large-cell
lymphoma. Ann Intern Med 1997; 127: 186–194.
Cortes J, Kurzrock R. Interleukin-10 in non-Hodgkin’s lymphoma.
Leuk Lymphoma 1997; 26: 251–259.
Christiansen I, Gidlof C, Kalkner KM, Hagberg H, Bennmarker H,
Totterman T. Elevated serum levels of soluble ICAM-1 in non-
26
27
28
29
30
Hodgkin’s lymphomas correlate with tumour burden, disease
activity and other prognostic markers. Br J Haematol 1996; 92:
639–646.
Ristamaki R, Joensuu H, Lappalainen K, Teerenhovi L, Jalkanen S.
Elevated serum CD44 level is associated with unfavorable outcome in non-Hodgkin’s lymphoma. Blood 1997; 90: 4039–4045.
Ichikawa A, Kinoshita T, Watanabe T, Kato H, Nagai H, Tsushita
K, Saito H, Hotta T. Mutations of the p53 gene as a prognostic
factor in aggressive B-cell lymphoma. New Engl J Med 1997; 337:
529–534.
Munker R, Younes A, Cabanullas F, Andreeff M. Soluble CD95 in
the serum of patients with low and intermediate grade malignant
lymphomas: absence of prognostic correlations. Leuk Lymphoma
1997; 27: 517–521.
Gianni AM, Bregni M, Siena S, Brambilla C, Di Nicola M, Lombardi F, Gandola L, Tarella C, Pileri A, Ravagnani F, Valagussa
P, Bonadonna G. High-dose chemotherapy and autologous bone
marrow transplantation compared with MACOP-B in aggressive
B-cell lymphoma. New Engl J Med 1997; 336: 1290–1297.
Nademanee A, Molina A, O’Donnell MR, Dagis A, Snyder DS,
Parker P, Stein A, Smith E, Planas I, Kashyap A, Somlo G, Spielberger R, Fung E, Wong KK, Margolin K, Chow W, Sniccinski I,
Vora N, Blume KG, Niland J, Forman SJ. Results of high-dose therapy and autologous bone marrow/stem cell transplantation during
remission in poor-risk intermediate- and high-grade lymphoma:
International index high and high-intermediate risk group. Blood
1997; 90: 3844–3852.