Abstract

Background: Soluble interleukin-2 receptor (SIL-2R) is known to be a prognostic parameter in patients with diffuse large B-cell lymphoma (DLBCL) receiving cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) therapy. However, its prognostic value has not been well known since the introduction of rituximab.

Patients and methods: We retrospectively evaluated the prognostic impact of SIL-2R in 228 DLBCL patients, comparing 141 rituximab-combined CHOP (RCHOP)-treated patients with 87 CHOP-treated patients as a historical control.

Results: Patients with high serum SIL-2R showed significantly poorer event-free survival (EFS) and overall survival (OS) than patients with low SIL-2R in both the RCHOP group (2-year EFS, 66% versus 92%, P < 0.001; OS, 82% versus 95%, P = 0.005) and the CHOP group (2-year EFS, 40% versus 82%; OS, 61% versus 90%, both P < 0.001). Multivariate analysis including the five parameters of International Prognostic Index (IPI) and two-categorized IPI revealed that SIL-2R was an independent prognostic factor for EFS and OS in the RCHOP group as well as in the CHOP group.

Conclusions: Our results demonstrate that SIL-2R retains its prognostic value in the rituximab era. The prognostic value of SIL-2R in DLBCL patients receiving rituximab-combined chemotherapy should be reassessed on a larger scale and by long-term follow-up.

introduction

Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin's lymphoma [1]. It takes an aggressive clinical course and comprises a heterogeneous group of lymphomas in terms of morphology, phenotype, molecular biology and clinical behavior. Up to now, the International Prognostic Index (IPI) has been the most widely used predictive model for patients with DLBCL treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) [2]. On the other hand, soluble interleukin-2 receptor (SIL-2R) has also been investigated as a prognostic factor, and several studies have demonstrated that a high level of SIL-2R before treatment is associated with both a low remission rate and poor prognosis [3–8].

SIL-2R is the soluble form of interleukin-2 receptor (IL-2R). IL-2R is expressed on the cell membrane of lymphocytes and plays important roles in their activation and proliferation [9]. It is composed of at least three glycoprotein chains: α (55 kDa), β (75 kDa) and γ (64 kDa). Each subunit is able to bind to the ligand independently with either low (IL-2Rα) or intermediate (IL-2Rβ and γ) affinity. It is now possible to examine the expression of the soluble-type α subunit [10]. The soluble IL-2Rα chain is induced and expressed only after mononuclear cell (T cell, B cell, monocyte, and natural killer cell) activation [11, 12]. Therefore, activated T and B cells have elevated levels of SIL-2R.

Although the CHOP regimen has been the mainstay of treatment for aggressive lymphomas for several decades [13], treatment outcome has significantly improved with the introduction of rituximab (an anti-CD20 chimeric antibody) in both young and elderly patients [14–17]. Since the introduction of rituximab, several prognostic factors have been reevaluated. Sehn et al. [18] recently reevaluated five prognostic factors and demonstrated that the IPI remained predictive; they proposed a revised IPI in which DLBCL patients are classified into very good (no IPI risk factors), good (one to two risk factors) and poor (three to five risk factors) categories. In contrast, BCL2, BCL6 and immunohistochemically defined germinal center (GC) phenotype have been reported to have no prognostic value when rituximab is added to chemotherapy [19–24]. Other clinical factors or biomarkers identified in patients receiving CHOP therefore need to be reassessed in patients treated with CHOP combined with rituximab.

Up to now, the prognostic value of SIL-2R in RCHOP has not been investigated. The aim of the present study was to retrospectively reassess the prognostic value of SIL-2R in DLBCL patients receiving RCHOP as compared with CHOP alone and to investigate whether or not this factor still influences the outcome of DLBCL.

patients and methods

patient characteristics

In the present study, we reviewed the medical records of patients with CD20-positive DLBCL who received CHOP with or without rituximab as a first-line therapy at the Cancer Institute Hospital from January 2000 to December 2006 and were followed until January 2008. The study protocol and sampling were approved by the Institutional Review Board of the Cancer Institute Hospital. Informed consent for retrospective analysis and additional immunophenotypic analysis and gene rearrangement studies was obtained.

Patients were analyzed if they were older than 18 years and had a performance status (PS) of zero to three according to the criteria of the European Cooperative Oncology Group. Patients were excluded if they had clinically relevant cardiac diseases or positivity for antibodies against HIV-1 or 2. Patients with primary mediastinal large B-cell lymphoma, primary CNS lymphoma and primary testicular lymphoma were also not included in this study.

The disease stage was evaluated according to the Ann Arbor staging system. All patients had undergone staging investigations, including physical examinations, blood and serum analysis, bone marrow aspiration and biopsy and computed tomography of the neck, chest, abdomen and pelvis. Magnetic resonance imaging was used for evaluation of involved organs in the head and neck. The following clinical and laboratory data were available at the time of diagnosis: age, sex, serum lactate dehydrogenase level, PS, presence of B symptoms, clinical stage and number of extranodal sites. This allowed the IPI scores to be determined in the studied patients. Patients were categorized into either a low-risk group (IPI score, 0–2) or a high-risk group (IPI score, 3–5). Response to initial therapy was evaluated according to the Cheson criteria [25].

treatment

In both the CHOP and RCHOP groups, CHOP chemotherapy was given triweekly at a standard dose. Patients with stages IB–IV received six cycles, and patients with stage IA three cycles, of CHOP chemotherapy followed by radiotherapy for the involved field. After incorporation of rituximab into the CHOP regimen in February 2004, patients were treated with RCHOP regimen, in which rituximab was administered at a standard dose of 375 mg/m2 once weekly for 8 weeks concurrently with triweekly CHOP, as described previously [26].

chemical studies

The serum SIL-2R levels were determined using a sandwich enzyme-linked immunosorbent assay kit (Cell-free Interleukin-2 Receptor Test Kit, T Cell Science, Cambridge, MA) using two mAbs against distinct two different epitopes of the p55 alpha-chain of the IL-2R complex. Serum SIL-2R was considered ‘high’ when higher than the median and ‘low’ when lower than the median.

pathological studies

Biopsy samples collected before treatment were fixed in formalin, embedded in paraffin, sliced and stained with hematoxylin and eosin for morphological analysis. For diagnosis of DLBCL, immunohistochemical analysis was carried out using the dextran-polymer method (EnVision+; Dako, Glostrup, Denmark) with mAbs against CD5, CD10, CD20, Ki67, BCL2, BCL6 and MUM1 in most cases and with CyclinD1 to exclude the possibility of a pleomorphic variant of mantle cell lymphoma when the lymphoma was CD5 positive. Patients with a small-cell component implying transformation from low-grade/indolent B-cell lymphoma were excluded. All the samples were reviewed by an expert hematopathologist (KT).

statistical analysis

Basic characteristics of the CHOP group and RCHOP group were compared by Fisher's exact test. Event-free survival (EFS) was calculated from the date of diagnosis to the date of documented disease progression, relapse or death from any cause or to the stopping date. Overall survival (OS) was calculated from the date of diagnosis until death from any cause or the last follow-up. If the stopping date was not reached, the data were censored at the date of the last follow-up evaluation. Survival curves were estimated by the Kaplan–Meier method, and overall differences were compared by the log-rank test. Log-rank test was carried out according to SIL-2R, two-categorized IPI for the two treatment groups. To estimate the unbiased prognostic impacts of SIL-2R on EFS and OS, Cox proportional hazards analysis was applied. First, we conducted univariate Cox analysis for SIL-2L, all IPI factors and dichotomized IPI and then we carried out multivariate Cox analysis adjusted for SIL-2R and each of the IPI risk factors, with final adjustment for SIL-2R and dichotomized IPI. Only factors that were associated with at least a trend toward significance in the univariate analysis (unadjusted P value <0.20) were evaluated in the multivariate model. We set P <0.05 as the level of statistical significance. Data were analyzed using SPSS software version 11.0 for Windows (SPSS, Chicago, IL).

results

patient characteristics

A total of 228 patients were analyzed, of whom 87 (38.2%) were given CHOP and 141 (61.8%) were given RCHOP. The median SIL-2R was 1005.5 mg/dl (range 220–35 600), and high SIL-2R was observed in 114 (50.0%) patients: 40 of 87 (46.0%) in the CHOP group and 74 of 141 (52.5%) in the RCHOP group. There was no significant difference in the proportion of high SIL-2R patients between the two treatment groups. The characteristics of the patients are listed in Table 1. Patient and disease characteristics were well balanced between the groups.

Table 1.

Patients’ characteristics according to serum SIL-2R level for CHOP and RCHOP group

Characteristics CHOP group
 
RCHOP group
 
P, all 
All Low SIL2R High SIL2R All Low SIL2R High SIL2R 
No. of patients (%) 87(100) 47 (54) 40 (46) 141 (100) 67 (48) 74 (52)  
Sex, no. (%)       0.41 
    Male 50 (57) 27 (57) 23 (58) 72 (51) 27 (40) 45 (61)  
    Female 37 (43) 20 (43) 17 (42) 69 (49) 40 (60) 29 (39)  
Age, no. (%)       0.52 
    ≤60 24 (28) 13 (28) 11 (28) 45 (32) 29 (43) 16 (22)  
    >60 63 (72) 34 (72) 29 (72) 96 (68) 38 (57) 58 (78)  
LDH, no. (%)       0.54 
    Normal 29 (32) 22 (35) 7 (17) 68 (48) 45 (67) 23 (31)  
    High 58 (68) 25 (65) 33 (83) 73 (52) 22 (33) 51 (69)  
PS, no. (%)       0.81 
    0–1 77 (89) 44 (94) 33 (83) 127 (90) 66 (98) 61 (82)  
    2–3 10 (11) 3 (6) 7 (17) 14 (10) 1 (2) 13 (18)  
Stage, no. (%)       0.73 
    I, II 55 (63) 40 (85) 15 (38) 93 (66) 57 (85) 36 (49)  
    III, IV 32 (37) 7 (15) 25 (72) 48 (34) 10 (15) 38 (51)  
Extranodal sites, no. (%)       0.84 
    0, 1 67 (77) 43 (91) 24 (60) 106 (75) 63 (94) 43 (57)  
    ≤2 20 (23) 4 (9) 16 (40) 35 (25) 4 (6) 31 (43)  
IPI, no. (%)       0.86 
    L/L-I 60 (69) 40 (85) 20 (50) 100 (71) 63 (94) 37 (50)  
    H/H-I 27 (31) 7 (15) 20 (50) 41 (29) 4 (6) 37 (50)  
Characteristics CHOP group
 
RCHOP group
 
P, all 
All Low SIL2R High SIL2R All Low SIL2R High SIL2R 
No. of patients (%) 87(100) 47 (54) 40 (46) 141 (100) 67 (48) 74 (52)  
Sex, no. (%)       0.41 
    Male 50 (57) 27 (57) 23 (58) 72 (51) 27 (40) 45 (61)  
    Female 37 (43) 20 (43) 17 (42) 69 (49) 40 (60) 29 (39)  
Age, no. (%)       0.52 
    ≤60 24 (28) 13 (28) 11 (28) 45 (32) 29 (43) 16 (22)  
    >60 63 (72) 34 (72) 29 (72) 96 (68) 38 (57) 58 (78)  
LDH, no. (%)       0.54 
    Normal 29 (32) 22 (35) 7 (17) 68 (48) 45 (67) 23 (31)  
    High 58 (68) 25 (65) 33 (83) 73 (52) 22 (33) 51 (69)  
PS, no. (%)       0.81 
    0–1 77 (89) 44 (94) 33 (83) 127 (90) 66 (98) 61 (82)  
    2–3 10 (11) 3 (6) 7 (17) 14 (10) 1 (2) 13 (18)  
Stage, no. (%)       0.73 
    I, II 55 (63) 40 (85) 15 (38) 93 (66) 57 (85) 36 (49)  
    III, IV 32 (37) 7 (15) 25 (72) 48 (34) 10 (15) 38 (51)  
Extranodal sites, no. (%)       0.84 
    0, 1 67 (77) 43 (91) 24 (60) 106 (75) 63 (94) 43 (57)  
    ≤2 20 (23) 4 (9) 16 (40) 35 (25) 4 (6) 31 (43)  
IPI, no. (%)       0.86 
    L/L-I 60 (69) 40 (85) 20 (50) 100 (71) 63 (94) 37 (50)  
    H/H-I 27 (31) 7 (15) 20 (50) 41 (29) 4 (6) 37 (50)  

SIL2R, soluble interleukin-2 receptor; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; RCHOP, rituximab-combined CHOP; LDH, lactate dehydrogenase; PS, performance status; IPI, International Prognostic Index; L/L-I, low or low-intermediate; H/H-I, high or high-intermediate; high SIL-2R; SIL-2R >1000 U/ml, low SIL-2R; SIL-2R ≤1000 U/ml.

survival analysis

With median follow-up periods of 30 months in the RCHOP group and 44 months in the CHOP group, EFS rates at 2 years were 78% and 65%, respectively (P = 0.030), and OS rates at 2 years were 89% and 81%, respectively (P = 0.040).

For CHOP therapy, the EFS and OS rates at 2 years were 82% and 93% for low SIL-2R and 43% and 65% for high SIL-2R, respectively. The differences in both the EFS and OS rates between the two SIL-2R levels were significant (both P < 0.001) (Figure 1A and B). In the RCHOP group, the EFS and OS rates at 2 years were 90% and 95% for low SIL-2R and 66% and 84% for high SIL-2R, respectively. The differences in both EFS and OS rates between the two SIL-2R levels were significant (EFS, P < 0.001; OS P = 0.005) (Figure 1C and D).

Figure 1.

Event-free survival (EFS) and overall survival (OS) curves for diffuse large B-cell lymphoma patients treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) and rituximab-combined CHOP (RCHOP) in relation to soluble interleukin-2 receptor (SIL-2R). EFS (A) and OS (B) curves according to low (n = 47) versus high (n = 40) SIL-2R in the CHOP group. EFS (C) and OS (D) curves according to low (n = 67) versus high (n = 74) SIL-2R in the RCHOP group.

Figure 1.

Event-free survival (EFS) and overall survival (OS) curves for diffuse large B-cell lymphoma patients treated with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) and rituximab-combined CHOP (RCHOP) in relation to soluble interleukin-2 receptor (SIL-2R). EFS (A) and OS (B) curves according to low (n = 47) versus high (n = 40) SIL-2R in the CHOP group. EFS (C) and OS (D) curves according to low (n = 67) versus high (n = 74) SIL-2R in the RCHOP group.

To study the impact of rituximab on the predictive value, we examined the clinical outcome according to treatment in the SIL-2R low and high groups. The patients with high SIL-2R who received RCHOP therapy had a significantly better OS at 2 years than patients treated with CHOP alone (84% versus 65%, P = 0.020). The EFS at 2 years was estimated to be 66% for the RCHOP group and 43% for the CHOP group (P = 0.010). For the patients with low SIL-2R, the influence of rituximab on OS and EFS was not significant (OS, 93% versus 95%, P = 0.310; EFS, 82% versus 90%, P = 0.160) (Table 2).

Table 2.

Analysis of 2-year survival rate according to CHOP and RCHOP therapy in both SIL-2R groups

Clinical outcome Low SIL-2R
 
High SIL-2R
 
CHOP RCHOP P CHOP RCHOP P 
2-year survival       
EFS (%) 82 90 0.160 43 66 0.0010 
OS (%) 93 95 0.310 65 84 0.0020 
Clinical outcome Low SIL-2R
 
High SIL-2R
 
CHOP RCHOP P CHOP RCHOP P 
2-year survival       
EFS (%) 82 90 0.160 43 66 0.0010 
OS (%) 93 95 0.310 65 84 0.0020 

CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; RCHOP, rituximab-combined CHOP; SIL2R, soluble interleukin-2 receptor; EFS, event-free survival; OS, overall survival.

For comparison with this parameter, we analyzed the survival curves according to the IPI in both treatment groups. The EFS and OS rates at 2 years were 35% and 59% for high or high-intermediate IPI and 77% and 91% for low or low-intermediate IPI, respectively, in the CHOP group. The differences in both EFS and OS rates between the two IPI groups were significant (both P < 0.001). Similarly, the EFS and OS rates were 58% and 80% for high or high-intermediate IPI and 86% and 94% for low or low-intermediate IPI, respectively, in the RCHOP group. Again, the differences in the EFS and OS rates were significant (P < 0.001 and P = 0.004, respectively).

To estimate unbiased prognostic impacts, Cox univariate analysis showed that a high SIL-2R level, high PS, advanced stage, multiple extranodal sites and high or high-intermediate risk of IPI were associated with poor EFS and OS in both treatment groups (Table 3). In the second step, Cox multivariate analysis showed that only SIL-2R was significantly associated with a higher risk of event and that SIL-2R and PS were independently associated with poor OS in both treatment groups (Table 4). Finally, SIL-2R was a significant risk factor for EFS and a borderline risk factor for OS in both the CHOP and RCHOP groups (P = 0.060 and 0.070, respectively), whereas IPI was a significant risk factor for EFS and OS in the CHOP group and a borderline significant risk factor for EFS and OS (P = 0.070 and 0.080, respectively) in the RCHOP group (Table 5).

Table 3.

The effects of clinical factors on EFS and OS in CHOP by univariate analysis using Cox proportional hazard model

 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 4.30 1.94–9.74 <0.001 
Age    
≤60 1.00   
>60 1.48 0.64–3.46 0.36 
LDH    
Normal 1.00   
High 2.65 0.80–8.74 0.11 
PS    
0–1 1.00   
2–3 4.10 1.64–10.25 0.003 
Stage    
I, II 1.00   
III, IV 3.75 1.79–7.83 <0.001 
Extranodal sites    
0–1 1.00   
≤2 3.24 1.54–6.81 0.002 
IPI    
L/L-I 1.00   
H/H-I 3.97 1.91–8.25 <0.001 
OS SIL-2R    
Low 1.00   
High 5.64 1.84–17.23 0.002 
Age    
≤60 1.00   
>60 2.20 0.64–7.61 0.21 
LDH    
Normal 1.00   
High 4.71 0.63–35.41 0.13 
PS    
0–1 1.00   
2–3 7.18 2.44–21.13 <0.001 
Stage    
I, II 1.00   
III, IV 5.15 1.92–13.81 0.001 
Extranodal sites    
0-1 1.00   
≤2 4.24 1.64–10.98 0.003 
IPI    
L/L-I 1.00   
H/H-I 7.28 2.69–19.67 <0.001 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 4.20 1.72–10.33 0.002 
Age    
≤60 1.00   
>60 1.38 0.61–3.11 0.44 
LDH    
Normal 1.00   
High 1.41 0.68–2.93 0.35 
PS    
0–1 1.00   
2–3 3.62 1.35–8.46 0.003 
Stage    
I, II 1.00   
III, IV 3.42 1.64–7.11 0.001 
Extranodal sites    
0–1 1.00   
≤2 3.43 1.67–7.03 <0.001 
IPI    
L/L-I 1.00   
H/H-I 3.40 1.82–6.25 <0.001 
OS SIL-2R    
Low 1.00   
High 6.42 1.45–28.45 0.01 
Age    
≤60 1.00   
>60 3.50 0.79–15.52 0.10 
LDH    
Normal 1.00   
High 1.69 0.58–4.97 0.34 
PS    
0–1 1.00   
2–3 5.97 2.03–17.54 0.001 
Stage    
I, II 1.00   
III, IV 2.46 0.89–6.79 0.08 
Extranodal sites    
0–1 1.00   
≤2 2.65 0.96–7.33 0.06 
IPI    
L/L-I 1.00   
H/H-I 4.03 1.43–11.34 0.08 
 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 4.30 1.94–9.74 <0.001 
Age    
≤60 1.00   
>60 1.48 0.64–3.46 0.36 
LDH    
Normal 1.00   
High 2.65 0.80–8.74 0.11 
PS    
0–1 1.00   
2–3 4.10 1.64–10.25 0.003 
Stage    
I, II 1.00   
III, IV 3.75 1.79–7.83 <0.001 
Extranodal sites    
0–1 1.00   
≤2 3.24 1.54–6.81 0.002 
IPI    
L/L-I 1.00   
H/H-I 3.97 1.91–8.25 <0.001 
OS SIL-2R    
Low 1.00   
High 5.64 1.84–17.23 0.002 
Age    
≤60 1.00   
>60 2.20 0.64–7.61 0.21 
LDH    
Normal 1.00   
High 4.71 0.63–35.41 0.13 
PS    
0–1 1.00   
2–3 7.18 2.44–21.13 <0.001 
Stage    
I, II 1.00   
III, IV 5.15 1.92–13.81 0.001 
Extranodal sites    
0-1 1.00   
≤2 4.24 1.64–10.98 0.003 
IPI    
L/L-I 1.00   
H/H-I 7.28 2.69–19.67 <0.001 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 4.20 1.72–10.33 0.002 
Age    
≤60 1.00   
>60 1.38 0.61–3.11 0.44 
LDH    
Normal 1.00   
High 1.41 0.68–2.93 0.35 
PS    
0–1 1.00   
2–3 3.62 1.35–8.46 0.003 
Stage    
I, II 1.00   
III, IV 3.42 1.64–7.11 0.001 
Extranodal sites    
0–1 1.00   
≤2 3.43 1.67–7.03 <0.001 
IPI    
L/L-I 1.00   
H/H-I 3.40 1.82–6.25 <0.001 
OS SIL-2R    
Low 1.00   
High 6.42 1.45–28.45 0.01 
Age    
≤60 1.00   
>60 3.50 0.79–15.52 0.10 
LDH    
Normal 1.00   
High 1.69 0.58–4.97 0.34 
PS    
0–1 1.00   
2–3 5.97 2.03–17.54 0.001 
Stage    
I, II 1.00   
III, IV 2.46 0.89–6.79 0.08 
Extranodal sites    
0–1 1.00   
≤2 2.65 0.96–7.33 0.06 
IPI    
L/L-I 1.00   
H/H-I 4.03 1.43–11.34 0.08 

EFS, event-free survival; OS, overall survival; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; RCHOP, rituximab-combined CHOP; HR, hazard ratio; CI, confidential interval; SIL2R, soluble interleukin-2 receptor; LDH, lactate dehydrogenase; PS, performance status; IPI, International Prognostic Index; L/L-I, low or low-intermediate; H/H-I, high or high-intermediate.

Table 4.

Multivariate Cox proportional hazard regression analysis for SIL-2R and IPI risk factors in both treatment groups

 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 2.74 1.05–7.14 0.04 
PS    
0–1 1.00   
2–3 2.12 0.73–6.14 0.17 
Stage    
I, II 1.00   
III, IV 1.82 0.65–5.09 0.25 
Extranodal sites    
0–1 1.00   
≤2 1.09 0.38–3.10 0.87 
OS SIL-2R    
Low 1.00   
High 3.53 1.03–12.95 0.05 
LDH    
Normal 1.00   
High 3.21 0.40–25.51 0.27 
PS    
0–1 1.00   
2–3 3.60 0.98–13.20 0.05 
Stage    
I, II 1.00   
III, IV 2.00 0.51–7.87 0.32 
Extranodal sites    
0–1 1.00   
≤2 0.86 0.22-0.83 0.83 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 2.65 1.01–7.30 0.05 
PS    
0–1 1.00   
2–3 1.66 0.62–4.42 0.31 
Stage    
I, II 1.00   
III, IV 1.69 0.65–4.42 0.28 
Extranodal sites    
0–1 1.00   
≤2 1.36 0.50–3.67 0.55 
OS SIL-2R    
Low 1.00   
High 5.09 1.00–25.88 0.05 
Age    
≤60 1.00   
>60 2.45 0.54–11.17 0.24 
PS    
0–1 1.00   
2–3 4.49 1.15–17.45 0.03 
Stage    
I, II 1.00   
III, IV 1.02 0.23–4.45 0.98 
Extranodal sites    
0–1 1.00   
≤2 0.70 0.15–3.39 0.66 
 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 2.74 1.05–7.14 0.04 
PS    
0–1 1.00   
2–3 2.12 0.73–6.14 0.17 
Stage    
I, II 1.00   
III, IV 1.82 0.65–5.09 0.25 
Extranodal sites    
0–1 1.00   
≤2 1.09 0.38–3.10 0.87 
OS SIL-2R    
Low 1.00   
High 3.53 1.03–12.95 0.05 
LDH    
Normal 1.00   
High 3.21 0.40–25.51 0.27 
PS    
0–1 1.00   
2–3 3.60 0.98–13.20 0.05 
Stage    
I, II 1.00   
III, IV 2.00 0.51–7.87 0.32 
Extranodal sites    
0–1 1.00   
≤2 0.86 0.22-0.83 0.83 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 2.65 1.01–7.30 0.05 
PS    
0–1 1.00   
2–3 1.66 0.62–4.42 0.31 
Stage    
I, II 1.00   
III, IV 1.69 0.65–4.42 0.28 
Extranodal sites    
0–1 1.00   
≤2 1.36 0.50–3.67 0.55 
OS SIL-2R    
Low 1.00   
High 5.09 1.00–25.88 0.05 
Age    
≤60 1.00   
>60 2.45 0.54–11.17 0.24 
PS    
0–1 1.00   
2–3 4.49 1.15–17.45 0.03 
Stage    
I, II 1.00   
III, IV 1.02 0.23–4.45 0.98 
Extranodal sites    
0–1 1.00   
≤2 0.70 0.15–3.39 0.66 

SIL2R, soluble interleukin-2 receptor; IPI, International Prognostic Index; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; HR, hazard ratio; CI, confidential interval; EFS, event-free survival; PS, performance status; OS, overall survival; LDH, lactate dehydrogenase; RCHOP, rituximab-combined CHOP.

Table 5.

Multivariate Cox proportional hazard analysis for SIL-2R and categorized IPI in both treatment groups

 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 2.98 1.22–7.29 0.01 
IPI    
L/L-I 1.00   
H/H-I 2.47 1.11–5.47 0.02 
OS SIL-2R    
Low 1.00   
High 3.12 0.93–10.41 0.06 
IPI    
L/L-I 1.00   
H/H-I 4.66 1.60–13.58 0.005 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 3.00 1.12–8.07 0.02 
IPI    
L/L-I 1.00   
H/H-I 2.06 0.93–4.57 0.07 
OS SIL-2R    
Low 1.00   
High 4.30 0.85–21.91 0.07 
IPI    
L/L-I 1.00   
H/H-I 4.16 0.80–6.69 0.08 
 Variable HR 95% CI P value 
CHOP     
EFS SIL-2R    
Low 1.00   
High 2.98 1.22–7.29 0.01 
IPI    
L/L-I 1.00   
H/H-I 2.47 1.11–5.47 0.02 
OS SIL-2R    
Low 1.00   
High 3.12 0.93–10.41 0.06 
IPI    
L/L-I 1.00   
H/H-I 4.66 1.60–13.58 0.005 
RCHOP     
EFS SIL-2R    
Low 1.00   
High 3.00 1.12–8.07 0.02 
IPI    
L/L-I 1.00   
H/H-I 2.06 0.93–4.57 0.07 
OS SIL-2R    
Low 1.00   
High 4.30 0.85–21.91 0.07 
IPI    
L/L-I 1.00   
H/H-I 4.16 0.80–6.69 0.08 

SIL2R, soluble interleukin-2 receptor; IPI, International Prognostic Index; CHOP, cyclophosphamide, doxorubicin, vincristine and prednisone; HR, hazard ratio; CI, confidential interval; EFS, event-free survival; OS, overall survival; LDH, lactate dehydrogenase; RCHOP, rituximab-combined CHOP; L/L-I, low or low-intermediate; H/H-I, high or high-intermediate.

discussion

Although SIL-2R is easy to measure, its prognostic value has been underestimated due to its evaluation in smaller populations than those for other parameters, such as IPI [2]. The SIL-2R level was reported to be significantly high in highly aggressive lymphomas [6] and subsequently was recognized to reflect tumor burden and poor outcome [3–8]. However, these results were obtained in patients receiving chemotherapy, and the prognostic value of SIL-2R has not been assessed in rituximab-combined treatment.

In the present study, univariate analysis showed that SIL-2R retained its prognostic value in DLBCL patients treated with RCHOP, as well as in those receiving CHOP alone. Multivariate analysis also showed that SIL-2R was an independent significant prognostic factor after adjustment for IPI risk factors and independently associated with significantly decreased EFS and moderately decreased OS after adjustment by two-categorized IPI in both the CHOP and RCHOP groups. On the other hand, the clinical outcome of patients with high SIL-2R was significantly improved by addition of rituximab to the chemotherapy, in contrast to the lack of any difference in the patients with low SIL-2R. To our knowledge, this is the first report to demonstrate the prognostic value of SIL-2R in DLBCL patients treated with rituximab-combined chemotherapy.

Although the present study was not a randomized prospective one, and possibly biased by factors other than IPI and SIL-2R, the distribution of baseline characteristics, including IPI factors, was similar between the two treatment groups. On the other hand, the population employed in the present analysis had more limited disease and a favorable IPI score compared with those in previous studies of DLBCL [13–17]. This might account for the better outcome of our patients than for those in previous reports such as that by Coiffier et al. [14] who observed 2-year survival rates of 70% and 57% in elderly patients treated with RCHOP and CHOP, respectively. Even with the excellent outcome we observed, however, the prognostic value of SIL-2R was significant and greater than that of other IPI risk factors. To allow our present results to be generalized to routine patient care, these findings should be validated in a variety of patient populations.

A number of prognostic markers have been identified in patients with DLBCL treated by chemotherapy alone [19–21], some of which have been reassessed and shown not to be associated with prognosis in patients receiving rituximab-combined chemotherapy [22–24]. BCL2 overexpression was reported to be associated with poorer survival in patients treated with CHOP-like regimens [19], but its prognostic value was not confirmed in patients receiving rituximab-combined chemotherapy in several studies, indicating that addition of rituximab overcomes the negative influence of BCL2 overexpression [24]. BCL6, a marker of germinal center derivation, has been identified as an indicator of favorable outcome in DLBCL [20], although outcome in patients receiving immunochemotherapy was reported to be uninfluenced by BCL6 status [22]. Similarly, no correlation between immunohistochemically defined GC phenotype and survival rate was observed in patients receiving immunochemotherapy [21], in contrast to previous findings of inferior outcomes in non-GC patients relative to GC patients in the prerituximab era [23]. Up to now, no marker other than IPI has been found to be of prognostic relevance since the clinical introduction of rituximab.

The mechanism by which rituximab added to chemotherapy improves outcome in relation to biological features has been evaluated in several studies. They showed that rituximab may suppress the constitutively active nuclear factor-kappa B pathway in non-GC phenotype DLBCL or downregulate Bcl-2-related antiapoptotic proteins, thereby increasing the sensitivity of lymphoma cells to chemotherapy [27–29]. These effects of rituximab may reduce the prognostic significance of the non-GC phenotype and BCL2. Although the mechanism by which SIL-2R retains its prognostic value after addition of rituximab to chemotherapy is unknown, SIL-2R may directly represent the tumor burden [7, 8].

To date, several studies including the present one have demonstrated that IPI score remains predictive in the rituximab era, in contrast to biomarkers [18, 22, 23]. In the present study, the IPI system identified only two risk groups instead of four among our patients—a low and low-intermediate group and a high and high-intermediate group—as reported in previous studies [22, 23].

The addition of rituximab to chemotherapy has improved the outcome of patients. We and others have shown that OS now exceeds 50% even in the groups with unfavorable indicators [14–18, 22, 23], although some patients still have a very poor outcome. Therefore, other predictive factors must be characterized in order to identify patients who should receive alternative initial therapy. A number of molecular prognostic markers have already been identified in patients with DLBCL [30]. These markers now need to be reevaluated in the rituximab era to identify patients with unfavorable prognostic factors and to devise adequate treatment strategies.

In conclusion, we have demonstrated that a high serum SIL-2R level is an indicator of poor prognosis in DLBCL patients receiving rituximab combination chemotherapy. To accurately confirm whether serum SIL-2R influences the outcome of patients receiving rituximab combination chemotherapy, prospective investigation with long-term follow-up will be required.

The authors are grateful to the members of the Ganken Adult Lymphoma Study Group, including Makoto Kodaira, Shuhei Yamada, Kyoko Ueda and Tomohiro Myojo, for treating the patients at the Cancer Institute Hospital and to Daigo Shoji, Chie Watanabe, Chizuru Suitsu, Ayako Nishito and Michiko Ennishi for collecting the clinical data. Contributions: DE designed the study, treated the patients, collected clinical data and wrote the paper; KT scored the immunohistochemical staining, designed the study and wrote the paper; MY assisted in designing the study and writing the paper; HA treated the patients and collected clinical data; SS, YM, YT and ST treated the patients and assisted in writing the paper; HK, KI and MT supervised the paper and KH designed the study, supervised all aspects of the research and analyses and wrote the paper. Conflict of interest disclosure: The authors declare no competing financial interests.

References

1.
Jaffe
ES
Harris
NL
Stein
H
Vardiman
JW
World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissue. IARC press
 , 
2001
2.
TIN-HsLPF
A predictive model for aggressive non-Hodgkin's lymphoma. The International Non-Hodgkin's Lymphoma Prognostic Factors Project
N Engl J Med
 , 
1993
, vol. 
329
 (pg. 
987
-
994
)
3.
Gause
A
Jung
W
Schmits
R
, et al.  . 
Soluble CD8, CD25 and CD30 antigens as prognostic markers in patients with untreated Hodgkin's lymphoma
Ann Oncol
 , 
1992
, vol. 
3
 
Suppl 4
(pg. 
49
-
52
)
4.
Chilosi
M
Semenzato
G
Cetto
G
, et al.  . 
Soluble interleukin-2 receptors in the sera of patients with hairy cell leukemia: relationship with the effect of recombinant alpha-interferon therapy on clinical parameters and natural killer in vitro activity
Blood
 , 
1987
, vol. 
70
 (pg. 
1530
-
1535
)
5.
Wagner
DK
Kiwanuka
J
Edwards
BK
, et al.  . 
Soluble interleukin-2 receptor levels in patients with undifferentiated and lymphoblastic lymphomas: correlation with survival
J Clin Oncol
 , 
1987
, vol. 
5
 (pg. 
1262
-
1274
)
6.
Kamihira
S
Atogami
S
Sohda
H
, et al.  . 
Significance of soluble interleukin-2 receptor levels for evaluation of the progression of adult T-cell leukemia
Cancer
 , 
1994
, vol. 
73
 (pg. 
2753
-
2758
)
7.
Niitsu
N
Iijima
K
Chizuka
A
A high serum-soluble interleukin-2 receptor level is associated with a poor outcome of aggressive non-Hodgkin's lymphoma
Eur J Haematol
 , 
2001
, vol. 
66
 (pg. 
24
-
30
)
8.
Nakase
K
Tsuji
K
Tamaki
S
, et al.  . 
Elevated levels of soluble interleukin-2 receptor in serum of patients with hematological or non-hematological malignancies
Cancer Detect Prev
 , 
2005
, vol. 
29
 (pg. 
256
-
259
)
9.
Takeshita
T
Asao
H
Ohtani
K
, et al.  . 
Cloning of the gamma chain of the human IL-2 receptor
Science
 , 
1992
, vol. 
257
 (pg. 
379
-
382
)
10.
Rubin
LA
Nelson
DL
The soluble interleukin-2 receptor: biology, function, and clinical application
Ann Intern Med
 , 
1990
, vol. 
113
 (pg. 
619
-
627
)
11.
Smith
KA
Interleukin-2: inception, impact, and implications
Science
 , 
1988
, vol. 
240
 (pg. 
1169
-
1176
)
12.
Voss
SD
Sondel
PM
Robb
RJ
Characterization of the interleukin 2 receptors (IL-2R) expressed on human natural killer cells activated in vivo by IL-2: association of the p64 IL-2R gamma chain with the IL-2R beta chain in functional intermediate-affinity IL-2R
J Exp Med
 , 
1992
, vol. 
176
 (pg. 
531
-
541
)
13.
Fisher
RI
Gaynor
ER
Dahlberg
S
, et al.  . 
Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin's lymphoma
N Engl J Med
 , 
1993
, vol. 
328
 (pg. 
1002
-
1006
)
14.
Coiffier
B
Lepage
E
Briere
J
, et al.  . 
CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma
N Engl J Med
 , 
2002
, vol. 
346
 (pg. 
235
-
242
)
15.
Habermann
TM
Weller
EA
Morrison
VA
, et al.  . 
Rituximab-CHOP versus CHOP alone or with maintenance rituximab in older patients with diffuse large B-cell lymphoma
J Clin Oncol
 , 
2006
, vol. 
24
 (pg. 
3121
-
3127
)
16.
Pfreundschuh
M
Trumper
L
Osterborg
A
, et al.  . 
CHOP-like chemotherapy plus rituximab versus CHOP-like chemotherapy alone in young patients with good-prognosis diffuse large-B-cell lymphoma: a randomised controlled trial by the MabThera International Trial (MInT) Group
Lancet Oncol
 , 
2006
, vol. 
7
 (pg. 
379
-
391
)
17.
Sehn
LH
Donaldson
J
Chhanabhai
M
, et al.  . 
Introduction of combined CHOP plus rituximab therapy dramatically improved outcome of diffuse large B-cell lymphoma in British Columbia
J Clin Oncol
 , 
2005
, vol. 
23
 (pg. 
5027
-
5033
)
18.
Sehn
LH
Berry
B
Chhanabhai
M
, et al.  . 
The revised International Prognostic Index (R-IPI) is a better predictor of outcome than the standard IPI for patients with diffuse large B-cell lymphoma treated with R-CHOP
Blood
 , 
2007
, vol. 
109
 (pg. 
1857
-
1861
)
19.
Gascoyne
RD
Adomat
SA
Krajewski
S
, et al.  . 
Prognostic significance of Bcl-2 protein expression and Bcl-2 gene rearrangement in diffuse aggressive non-Hodgkin's lymphoma
Blood
 , 
1997
, vol. 
90
 (pg. 
244
-
251
)
20.
Barrans
SL
O'Connor
SJ
Evans
PA
, et al.  . 
Rearrangement of the BCL6 locus at 3q27 is an independent poor prognostic factor in nodal diffuse large B-cell lymphoma
Br J Haematol
 , 
2002
, vol. 
117
 (pg. 
322
-
332
)
21.
Hans
CP
Weisenburger
DD
Greiner
TC
, et al.  . 
Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray
Blood
 , 
2004
, vol. 
103
 (pg. 
275
-
282
)
22.
Winter
JN
Weller
EA
Horning
SJ
, et al.  . 
Prognostic significance of Bcl-6 protein expression in DLBCL treated with CHOP or R-CHOP: a prospective correlative study
Blood
 , 
2006
, vol. 
107
 (pg. 
4207
-
4213
)
23.
Nyman
H
Adde
M
Karjalainen-Lindsberg
ML
, et al.  . 
Prognostic impact of immunohistochemically defined germinal center phenotype in diffuse large B-cell lymphoma patients treated with immunochemotherapy
Blood
 , 
2007
, vol. 
109
 (pg. 
4930
-
4935
)
24.
Wilson
H
Pittaluga
S
O'Connor
P
Rituximab may overcome Bcl-2-associated chemotherapy resistance in untreated diffuse large B-cell lymphomas [abstract]
Blood
 , 
2001
, vol. 
98
 pg. 
343a
 
25.
Cheson
BD
Horning
SJ
Coiffier
B
, et al.  . 
Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group
J Clin Oncol
 , 
1999
, vol. 
17
 pg. 
1244
 
26.
Ogura
M
Morishima
Y
Kagami
Y
, et al.  . 
Randomized phase II study of concurrent and sequential rituximab and CHOP chemotherapy in untreated indolent B-cell lymphoma
Cancer Sci
 , 
2006
, vol. 
97
 (pg. 
305
-
312
)
27.
Davis
RE
Brown
KD
Siebenlist
U
Staudt
LM
Constitutive nuclear factor kappaB activity is required for survival of activated B cell-like diffuse large B cell lymphoma cells
J Exp Med
 , 
2001
, vol. 
194
 (pg. 
1861
-
1874
)
28.
Jazirehi
AR
Bonavida
B
Cellular and molecular signal transduction pathways modulated by rituximab (rituxan, anti-CD20 mAb) in non-Hodgkin's lymphoma: implications in chemosensitization and therapeutic intervention
Oncogene
 , 
2005
, vol. 
24
 (pg. 
2121
-
2143
)
29.
Jazirehi
AR
Huerta-Yepez
S
Cheng
G
Bonavida
B
Rituximab (chimeric anti-CD20 monoclonal antibody) inhibits the constitutive nuclear factor-{kappa}B signaling pathway in non-Hodgkin's lymphoma B-cell lines: role in sensitization to chemotherapeutic drug-induced apoptosis
Cancer Res
 , 
2005
, vol. 
65
 (pg. 
264
-
276
)
30.
Lossos
IS
Morgensztern
D
Prognostic biomarkers in diffuse large B-cell lymphoma
J Clin Oncol
 , 
2006
, vol. 
24
 (pg. 
995
-
1007
)