SEARCH

SEARCH BY CITATION

Keywords:

  • ATLL;
  • interleukin-5;
  • interleukin-10

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Patients with adult T-cell leukemia/lymphoma (ATLL) are in a severely immunocompromised state. Therefore, it is assumed that ATLL cells either express particular cytokines or induce their expression in host immune cells, disrupting the balanced production of cytokines and causing the host's immune system to break down. We examined the levels of serum cytokines including T helper type 1- (Th1-) associated cytokines [IFN-γ, TNF-α, and interleukin (IL)-2], Th2-associated cytokines (IL-4, -5 and -6) and regulatory T cell-associated cytokines (IL-10 and TGF-β1) in 94 ATLL patients, 39 asymptomatic human T-cell lymphotropic virus type-1 (HTLV-1) carriers and 50 healthy adult volunteers, to clarify whether elevated levels of particular cytokines are associated with the prognosis of ATLL patients. On multivariate analysis, high IL-5 and IL-10 levels were independent and significant unfavorable prognostic factors among the ATLL patients. The IL-10 level significantly increased with disease progression at each step from asymptomatic HTLV-1 carrier to ATLL of the indolent variant (chronic and smoldering subtypes) to ATLL of the aggressive variant (acute and lymphoma subtypes). Furthermore, high IL-10 was significantly associated with high lactate dehydrogenase (LDH), indicating that the IL-10 level reflects the tumor burden. The IL-5 level was not associated with disease progression nor LDH. Among ATLL patients with the aggressive variant, high IL-5, but not high IL-10, was an independent and significant unfavorable prognostic factor on multivariate analysis. Measurement of serum IL-5 and IL-10 levels is useful for predicting the prognosis and for determining a suitable treatment strategy for ATLL patients. © 2006 Wiley-Liss, Inc.

Adult T-cell leukemia/lymphoma (ATLL), which is a peripheral T-cell neoplasm, is most often composed of highly pleomorphic lymphoid cells, and is caused by the human retrovirus, human T-cell lymphotropic virus type-1 (HTLV-1). ATLL has a very poor prognosis, and the median survival time of patients with the acute or lymphoma subtype of ATLL is less than 1 year. ATLL patients are in a severely immunocompromised state, leading to frequent and severe infectious complications and to an unfavorable outcome.1, 2 The human immune system depends on balanced production of cytokines by 2 distinct T helper (Th) cell subsets, Th1 and Th2 cells.3 Th1 cells produce interferon (IFN)-γ, tumor necrosis factor (TNF)-α and interleukin (IL)-2, while Th2 cells produce IL-4, IL-5 and IL-6. Th1 cells play a critical role in cellular immunity, while Th2 cells are involved in humoral immunity. In addition to Th1/Th2 cells, regulatory T (Treg) cells play an important role in controlling the immune balance. The Treg populations described to date include the naturally occurring CD4+CD25+ Treg cells (CD4+CD25+ Treg cells), IL-10-producing Treg cells (IL-10-Treg cells) and transforming growth factor (TGF)-β-producing Treg cells.4, 5, 6, 7 Destruction of the balance among these immune cells leads to various immunodeficiencies; thus, it is suggested that ATLL cells play a role in destroying the immune balance.

It is known that ATLL cells have a CD4+CD25+ surface phenotype,1 and we and other investigators recently reported that the majority of ATLL cells also express CC chemokine receptor 4 (CCR4).8, 9 Among immune cells, CCR4 is selectively expressed on Th2 cells and a subset of Treg cells.10 Thus, the frequent expression of CCR4 on ATLL cells suggests that most ATLL cells originate from Th2 cells or even Treg cells, and HTLV-1-infected Th2 cells or Treg cells may have a long-term selective survival advantage in vivo. Furthermore, functional suppression of the host's normal effector T cells by these HTLV-1-infected Th2 cells or Treg cells can result in a severely immunocompromised state, which is one of the clinical characteristics of patients with ATLL.9, 11 Regarding Treg cells, we and other investigators recently reported that the phenotypic and functional characteristics of ATLL cells and naturally occurring Treg cells are similar.11, 12, 13 On the other hand, several studies on cytokine production by ATLL cells do not support the notion that ATLL cells originate from Th2 cells.14, 15, 16 It is assumed that ATLL cells either express particular cytokines or induce their expression in host immune cells, resulting in disruption of the balanced production of cytokines and causing the host's immune system to break down. It is important to clarify the clinical roles of individual cytokines in ATLL patients so as to develop therapies that overcome the strong immunosuppressive state in these patients.

We examined the serum levels of Th1-associated cytokines including IFN-γ, TNF-α and IL-2, Th2-associated cytokines including IL-4, IL-5 and IL-6, and Treg-associated cytokines including IL-10 and TGF-β1 in 94 ATLL patients, 39 asymptomatic HTLV-1 carriers and 50 healthy adult volunteers, and analyzed the associations between the serum concentration of each cytokine and the clinical characteristics among the ATLL patients. The goal of the present study was to clarify whether elevated levels of particular cytokines are associated with the prognosis of ATLL patients.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Patients and control subjects

This study included 94 ATLL patients and 39 asymptomatic HTLV-1 carriers who were diagnosed between 1989 and 2005 at 2 independent hospitals in Japan (Imamura Bun-in Hospital and Nagoya City University Hospital). The diagnosis and classification of clinical subtypes of ATLL were made according to the criteria proposed by the Japan Lymphoma Study Group.2 The clinical characteristics of the ATLL patients who were analyzed in our study included age, sex, clinical subtype, performance status (PS), presence or absence of B symptoms, white blood cell count (WBC), lymphocyte count, eosinophil count, hemoglobin (Hb) level and platelet (Plt) count in the peripheral blood, serum calcium (Ca) level, serum albumin level, serum lactate dehydrogenase (LDH) level, serum soluble interleukin-2 receptor (sIL-2R) level and involved sites [lymph node, bone marrow, peripheral blood, lung, liver, spleen, gastrointestinal tract, kidney, central nervous system, bone and skin]. It is generally known that ATLL patients with the acute and lymphoma subtypes have an aggressive clinical course, while ATLL patients with the chronic and smoldering subtypes have longer survival.1, 2 Thus, we designated the acute and lymphoma subtypes of ATLL as the aggressive variant, and the chronic and smoldering subtypes of ATLL as the indolent variant in the present study. In the analyses, we used the clinical characteristics at the time the blood sample for analysis of cytokine levels was obtained, and all blood samples were obtained at the time of initial diagnosis of ATLL or HTLV-1 carrier status. Although the treatments administered to the ATLL patients enrolled in this study varied, combination chemotherapy including doxorubicin such as the RCM protocol,17 the LSG15 protocol18 or the cyclophosphamide-doxorubicin-vincristine-prednisone (CHOP) regimen was administered to the majority of the patients. All ATLL patients and asymptomatic HTLV-1 carriers gave written informed consent before the sampling of peripheral blood, according to the Declaration of Helsinki. Control peripheral blood samples were obtained from 50 healthy adult volunteers, who also gave written informed consent before the blood sampling procedure, according to the Declaration of Helsinki.

Measurement of cytokine levels

Peripheral blood samples were transferred to serum-separating tubes and centrifuged at 1000g at 20°C for 20 min after clot formation. The supernatants were carefully harvested, and aliquots were frozen at −80°C until analysis. The concentrations of IFN-γ, TNF-α, IL-2, IL-4, IL-5 and IL-10 in the serum were determined by the Cytometric Bead Array Kit (BD Biosciences, San Jose, CA), according to the manufacturer's instructions. The IL-6 and TGF-β1 concentrations in the serum were determined by enzyme-linked immunosorbent assay (ELISA) using the Human IL-6 or TGF-β1 Immunoassay Kit (R&D Systems, Minneapolis, MN), respectively, according to the manufacturer's instructions.

Statistical analysis

The serum concentrations of each cytokine were compared among the ATLL patients, asymptomatic HTLV-1 carriers and healthy volunteers, using the Kruskal–Wallis test followed by the Bonfferroni-adjusted Mann–Whitney U test. The ATLL patients were divided into 2 groups according to the serum level of each cytokine; cut-off values for each cytokine were set as the mean plus the standard deviation of the serum level of the respective cytokine in the healthy control group. The significance of the difference in each cytokine concentration and clinical characteristics between 2 groups of ATLL patients was assessed by Fisher's exact test. Survival analyses were performed by the Kaplan–Meier method, and survival curves were compared using the log-rank and Breslow–Gehan–Wilcoxon tests. The starting time for survival was the date of blood collection. For survival analysis, all patients who were still alive were censored at the date of last follow-up. Univariate and multivariate analyses were performed with the Cox proportional hazard model. Variables considered for possible inclusion in the regression analysis were those that showed a significant difference in the univariate analysis. All statistical analyses were performed using StatView software (SAS Institute, version 5.0, Cary, NC). In our study, p < 0.05 was considered as significant except in the Bonfferroni-adjusted Mann–Whitney U test. p < 0.05/3 was considered as significant in the Bonfferroni-adjusted Mann–Whitney U test.

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Patient characteristics

Among the 94 ATLL patients, there were 54 males and 40 females with an age range of 38–89 years (median age, 63 years). Their clinical characteristics are summarized in Table I.

Table I. Characteristics of the Patients with ATLL
  • n, number; WBC, white blood cell count; Hb, hemoglobin; Plt, platelet count; Ca, calcium level; LDH, lactate dehydrogenase; sIL-2R, soluble interleukin-2 receptor.–

  • 1

    LDH was expressed as a ratio in which the LDH value in the patient was divided by the upper limit of normal for LDH at the laboratory at the respective hospital.

n94
Age (yr), median (range)63 (38–89)
Sex (male/female)54/40
Clinical Subtype
 Acute57 (60.6%)
 Lymphoma12 (12.8%)
 Chronic17 (18.1%)
 Smoldering8 (8.5%)
Performance status
 026 (27.7%)
 136 (38.3%)
 217 (18.1%)
 310 (10.6%)
 45 (5.3%)
B symptoms
 present24 (25.5%)
 absent70 (74.5%)
WBC (×103/μL), median (range)9.7 (1.3–208.6)
Lymphocyte (×103/μL), median (range)3.4 (0.3–196.1)
Eosinophil (×103/μL), median (range)0.0 (0.0–4.5)
Hb (g/dL), median (range)13.0 (7.7–16.9)
Plt (×103/μL), median (range)198.5 (7.0–1131.0)
Ca (mg/dL), median (range)9.6 (7.6–19.9)
Albumin (g/dL), median (range)3.7 (2.1–4.7)
LDH1 (/upper limit of normal), median (range)1.36 (0.54–10.53)
slL-2R (×103U/mL), median (range)16.6 (0.4–186.7)
Involvement sites
 Lymph node67 (71.3%)
 Bone marrow37 (39.4%)
 Peripheral blood73 (77.7%)
 Lung1 (1.1%)
 Liver14 (14.9%)
 Spleen26 (27.7%)
 Gastrointestinal tract16 (17.0%)
 Kidney1 (1.1%)
 Central nervous system2 (2.1%)
 Bone3 (3.2%)
 Skin22 (23.4%)

Serum cytokine concentrations in ATLL patients, asymptomatic HTLV-1 carriers and healthy adult volunteers

The serum IFN-γ concentration was significantly higher in the ATLL patients and asymptomatic carriers than in the healthy volunteers (p < 0.0001 and p < 0.0001, respectively) (Fig. 1a). The serum TNF-α concentration was significantly lower in the asymptomatic carriers than in the healthy volunteers (p = 0.0328) (Fig. 1b). There were no significant differences in the IFN-γ and TNF- α concentrations between the ATLL patients and the asymptomatic carriers (Fig. 1a and 1b). The median value of serum IFN-γ concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–91.7 pg/mL), 33.0 pg/mL (range, 0.0–178.9 pg/mL) and 23.0 pg/mL (range, 0.0–831.7 pg/mL), respectively. The median value of serum TNF-α concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–45.7 pg/mL), 0.0 pg/mL (range, 0.0–5.8 pg/mL) and 0.0 pg/mL (range, 0.0–8.9 pg/mL), respectively. The IL-6 and IL-10 concentrations were significantly higher in the ATLL patients than in the asymptomatic carriers and healthy volunteers (IL-6, p < 0.0001 and p < 0.0001, respectively; and IL-10, p < 0.0001 and p < 0.0001, respectively), whereas there were no significant differences in the IL-6 and IL-10 concentrations between the asymptomatic carriers and healthy volunteers (Fig. 1c and 1d). The median value of serum IL-6 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–2.7 pg/mL), 0.0 pg/mL (range, 0.0–12.9 pg/mL) and 2.4 pg/mL (range, 0.0–250.4 pg/mL), respectively. The median value of serum IL-10 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 1.3 pg/mL (range, 0.0–3.6 pg/mL), 1.5 pg/mL (range, 0.0–8.9 pg/mL) and 4.2 pg/mL (range, 0.0–268.1 pg/mL), respectively. The TGF-β1 concentration was significantly higher in the ATLL patients and the healthy volunteers than in the asymptomatic HTLV-1 carriers (p = 0.0070, p < 0.0001, respectively) (Fig. 1e). The median value of serum TGF-β1 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 32.7 ng/mL (range, 18.9–52.5 ng/mL), 23.6 ng/mL (range, 3.9–90.5 ng/mL) and 30.5 ng/mL (range, 4.5–101.0 ng/mL), respectively. There were no significant differences in the IL-2, IL-4 (data not shown) and IL-5 (Fig. 1f) concentrations among the ATLL patients, asymptomatic carriers and healthy volunteers. The median value of serum IL-2 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–3.2 pg/mL), 0.0 pg/mL (range, 0.0–12.1 pg/mL) and 0.0 pg/mL (range, 0.0–20.0 pg/mL), respectively (data not shown). The median value of serum IL-4 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–5.9 pg/mL), 1.9 pg/mL (range, 0.0–10.3 pg/mL) and 0.0 pg/mL (range, 0.0–11.9 pg/mL), respectively (data not shown). The median value of serum IL-5 concentrations in the healthy adult volunteers, asymptomatic carriers and ATLL patients was 0.0 pg/mL (range, 0.0–4.0 pg/mL), 1.0 pg/mL (range, 0.0–1.7 pg/mL) and 1.0 pg/mL (range, 0.0–488.9 pg/mL), respectively.

thumbnail image

Figure 1. Comparison of serum cytokine concentrations in healthy adult volunteers, asymptomatic HTLV-1 carriers and patients with ATLL. The concentrations of cytokines in serum samples obtained from 50 healthy volunteers (○), 39 HTLV-1 carriers (•) and 94 ATLL patients (•) are shown. The horizontal white bars represent the median values of the concentration of the indicated cytokine in each group. The dotted lines represent the cut-off values (i.e., the mean value plus the standard deviation in the healthy controls for each cytokine). The p values are indicated in each panel. (n.s., not significant).

Download figure to PowerPoint

Serum cytokine concentrations among the ATLL patients

Among the ATLL patients, the serum IL-6 and IL-10 concentrations were significantly higher in those with the aggressive variant than in those with the indolent variant (p = 0.0002 and p = 0.0019, respectively). There were no significant differences in the IFN-γ, TNF-α, IL-2, IL-4, IL-5 and TGF-β1 concentrations between those with the aggressive variant and those with the indolent variant (data not shown).

Associations between serum cytokine levels and overall survival among the ATLL patients

The overall survival (OS) of all ATLL cases enrolled in our study is shown in Figure 2a. The 50% OS ± standard error was 13.2 ± 1.8 months, which was nearly in agreement with those obtained in previous clinical trials in Japan.19 The OS was significantly shorter in ATLL patients with the aggressive variant than in those with the indolent variant (50% OS, 10.5 versus 42.2 months) (Fig. 2b). The OS was significantly shorter in ATLL patients with a high IL-5 level (IL-5 > 1.7 pg/mL) than in those with a low IL-5 level (50% OS, 7.8 versus 17.2 months) (Fig. 2c); in ATLL patients with a high IL-6 level (IL-6 > 0.7 pg/mL) than in those with a low IL-6 level (50% OS, 9.2 versus 22.0 months) (Fig. 2d) and in ATLL patients with a high IL-10 level (IL-10 > 2.6 pg/mL) than in those with a low IL-10 level (50% OS, 9.5 versus 22.0 months) (Fig. 2e). On the other hand, there were no significant differences in OS between ATLL patients with a low or high level of IFN-γ, IL-2, IL-4 or TGF-β1 (data not shown). The ATLL patients could not be divided into 2 groups according to the serum TNF-α concentration, since the serum TNF-α concentration in all ATLL patients was below the cut-off value.

thumbnail image

Figure 2. Overall survival (OS) of ATLL patients. (a) OS curve of all patients with ATLL enrolled in our study (n = 94). (b) OS curves of the 69 ATLL patients with the aggressive variant and the 25 ATLL patients with the indolent variant. The 50% survival time is 10.5 months and 42.2 months, respectively. The ATLL patients with the aggressive variant showed significantly worse prognosis. (c) OS curves of the ATLL patients according to the serum IL-5 level. The ATLL patients with a high IL-5 level (IL-5 > 1.7 pg/mL) showed significantly worse prognosis. (d) OS curves of the ATLL patients according to the serum IL-6 level. The ATLL patients with a high IL-6 level (IL-6 > 0.7 pg/mL) showed significantly worse prognosis. (e) OS curves of the ATLL patients according to the serum IL-10 level. The ATLL patients with a high IL-10 level (IL-10 > 2.6 pg/mL) showed significantly worse prognosis.

Download figure to PowerPoint

Associations between the serum cytokine levels with each other among the ATLL patients

The associations between the serum cytokine levels among the ATLL patients are summarized in Table II. A high IL-10 level was significantly associated with a high IL-5 level (p = 0.0030) and high IL-6 level (p = 0.0006). A high IL-6 level was significantly associated with a high TGF-β1 level (>40.0 ng/mL) (p = 0.0376). There were no other significant associations between the serum cytokine levels (data not shown).

Table II. Associations Between the Serum Cytokine Levels with Each Other Among the ATLL Patients
CharacteristicAll patients n (%)IL-5 (pg/mL)p-valueIL-6 (pg/mL)p-valueIL-10 (pg/mL)p-value
> 1.7 n (%)≤1.7 n (%)> 0.7 n (%)≤ 0.7 n (%)> 2.6 n (%)≤ 2.6 n (%)
  1. n.s., not significant.

 94 (100)22 (23)72 (77) 61 (65)33 (35) 67 (71)27 (29) 
IFN-γ (pg/mL)
 > 29.043 (46)11 (50)32 (44)n.s.29 (48)14 (42)n.s.30 (45)13 (48)n.s.
 ≤ 29.051 (54)11 (50)40 (56)32 (52)19 (58)37 (55)14 (52)
IL-2 (pg/mL)
 > 1.215 (16)3 (14)12 (17)n.s.11 (18)4 (12)n.s.14 (21)1 (4)n.s.
 ≤ 1.279 (84)19 (86)60 (83)50 (82)29 (88)53 (79)26 (96)
IL-4 (pg/mL)
 > 2.825 (27)3 (14)22 (31)n.s.16 (26)9 (27)n.s.17 (25)8 (30)n.s.
 ≤ 2.869 (73)19 (86)50 (69)45 (74)24 (73)50 (75)19 (70)
IL-5 (pg/mL)
 > 1.722 (23)   17 (28)5 (15)n.s.21 (31)1 (4).0030
 ≤ 1.772 (77)   44 (72)28 (85)46 (69)26 (96)
IL-6 (pg/mL)
 > 0.761 (65)17 (77)44 (61)n.s.   51 (76)10 (37).0006
 ≤ 0.733 (35)5 (23)28 (39)   16 (24)17 (63)
IL-10 (pg/mL)
 > 2.667 (71)21 (95)46 (64).003051 (84)16 (48).0006   
 ≤ 2.627 (29)1 (5)26 (36)10 (16)17 (52)  
TGF-β1 (ng/mL)
 > 40.031 (33)6 (27)25 (35)n.s.25 (41)6 (18).037624 (36)7 (26)n.s.
 ≤ 40.063 (67)16 (73)47 (65)36 (59)27 (82)43 (64)20 (74)

Associations between the serum cytokine levels and clinicalcharacteristics among the ATLL patients

The clinical characteristics of the ATLL patients with a low or high level of IL-5, IL-6 or IL-10 are summarized in Table III. A high IL-5 level was significantly associated with eosinophilia (>500/μL) (p < 0.0001), hypoalbuminemia (<3.0 g/dL) (p = 0.0105) and a high sIL-2R level (>10,000 U/ml) (p = 0.0250). A high IL-6 level was significantly associated with the aggressive variant of ATLL, worse PS from 2 to 4 (P = 0.0058), presence of B symptoms (p = 0.0012), a high LDH level (greater than twice the upper limit of normal) (p = 0.0420) and a high sIL-2R level (p = 0.0008). A high IL-10 level was significantly associated with the aggressive variant of ATLL, worse PS (p = 0.0034), leukocytosis (>10.0 × 103/μL) (p = 0.0414), lymphocytosis (>4.0 × 103/μL) (p = 0.0384), a high LDH level (p = 0.0084), a high sIL-2R level (p < 0.0001) and extranodal involvement of more than 1 site (p = 0.0234). Regarding the involvement sites, a high IL-5 level was significantly associated with favored tumor involvement of the liver (p = 0.0177) and skin (p = 0.0090), and a high IL-10 level was significantly associated with favored tumor involvement of the lymph nodes (p = 0.0443), gastrointestinal tract (p = 0.0338), spleen (p = 0.0239) and bone marrow (p = 0.0022) (data not shown). No other clinical characteristics including age (>60 or ≤60 years), sex (male or female), Hb level in the peripheral blood (<10 or ≥10 g/dL), Plt count in the peripheral blood (<100 × 103 or ≥100 × 103/μL) and serum Ca level (>11.0 or ≤11.0 mg/dL) were associated with a low or high level of IL-5, IL-6 or IL-10 (data not shown).

Table III. Clinical Characteristics of the ATLL Patients According to the IL-5, IL-6, and IL-10 Levels
CharacteristicAll Patients n (%)IL-5 (pg/mL)p-valueIL-6 (pg/mL)p-valueIL-10 (pg/mL)p-value
> 1.7 n (%)≤ 1.7 n (%)> 0.7 n (%)≤ 0.7 n (%)> 2.6 n (%)≤ 2.6 n (%)
  • *

    Aggressive, acute and lymphoma subtypes of ATLL: Indolent, chronic and smoldering subtypes of ATLL.

  • 1

    LDH > 2N, the LDH level is greater than twice the upper limit of normal at the laboratory of the respective hospital.

 94 (100)22 (23)72 (77) 61 (65)33 (35) 67 (71)27 (29) 
Clinical variant*
 Aggressive69 (73)19 (86)50 (69)n.s.50 (82)19 (58).014754 (81)15 (56).0197
 Indolent25 (27)3 (14)22 (31)11 (18)14 (42)13 (19)12 (44)
Performance status
 2–432 (34)11 (50)21 (29)n.s.27 (44)5 (15).005829 (43)3 (11).0034
 0–162 (66)11 (50)51 (71)34 (56)28 (85)38 (57)24 (89)
B symptoms
 Present24 (26)5 (23)19 (26)n.s.22 (36)2 (6).001220 (30)4 (15)n.s.
 Absent70 (74)17 (77)53 (74)39 (64)31 (94)47 (70)23 (85)
WBC (×103/μL)
 > 10.044 (47)12 (55)32 (44)n.s.28 (46)16 (48)n.s.36 (54)8 (30).0414
 ≤ 10.050 (53)10 (45)40 (56)33 (54)17 (52)31 (46)19 (70)
Lymphocyte (×103/μL)
 > 4.041 (44)9 (41)32 (44)n.s.24 (39)17 (52)n.s.34 (51)7 (26).0384
 ≤ 4.053 (56)13 (59)40 (56)37 (61)16 (48)33 (49)20 (74)
Eosinophil (/μL)
 > 5007 (7)7 (32)0 (0)<.00014 (7)3 (9)n.s.7 (10)0 (0)n.s.
 ≤ 50087 (93)15 (68)72 (100)57 (93)30 (91)60 (90)27 (100)
Albumin (g/dL)
 < 3.013 (14)7 (32)6 (8).010511 (18)2 (6)n.s.12 (18)1 (4)n.s.
 ≥ 3.081 (86)15 (68)66 (92)50 (82)31 (94)55 (82)26 (96)
LDH1
 > 2N34 (36)11 (50)23 (32)n.s.27 (44)7 (21).042030 (45)4 (15).0084
 ≤ 2N60 (64)11 (50)49 (68)34 (56)26 (79)37 (55)23 (85)
sIL-2R (U/mL)
 > 10,00057 (61)18 (82)39 (54).025045 (74)12 (36).000851 (76)6 (22)< .0001
 ≤ 10,00037 (39)4 (18)33 (46)16 (26)21 (64)16 (24)21 (78)
Extranodal involvement
 ≥ 2 sites42 (45)14 (64)28 (39)n.s.29 (48)13 (39)n.s.35 (52)7 (26).0234
 < 2 sites52 (55)8 (36)44 (61)32 (52)20 (61)32 (48)20 (74)

With respect to the other cytokines such as IFN-γ, IL-2, IL-4 and TGF-β1, a high IFN-γ level (IFN-γ > 29.0 pg/mL) was significantly associated with anemia (Hb < 10.0 g/dL) (p = 0.0206) and hypoalbuminemia (p = 0.0184), and a low TGF-β1 level was significantly associated with thrombocytopenia (Plt < 100 × 103/μL) (P = 0.0275) and a high TGF-β1 level was significantly associated with leukocytosis (P = 0.0271). No other clinical characteristics were associated with the cytokine status (data not shown).

Prognostic factors in ATLL patients

Univariate Cox proportional hazard analysis identified the following unfavorable prognostic factors, with respect to the patients' survival: a high IL-5 level, a high IL-6 level, a high IL-10 level, the aggressive variant, worse PS, presence of B symptoms, a high LDH level, a high sIL-2R level and extranodal involvement of more than 1 site (Table IV). Multivariate analysis demonstrated that a high IL-5 level, a high IL-10 level, presence of B symptoms and a high LDH level were independent and significant unfavorable prognostic factors (Table IV).

Table IV. Prognostic Factors Affecting Overall Survival Among all ATLL Patients
VariableUnfavorableUnivariateMultivariate
Hazard ratio (95% CI)1p valueHazard ratio (95% CI)1p value
  • 1

    CI, confidence Interval.

IL-5> 1.7 pg/mL3.455 (1.796–6.649).00023.004 (1.520–5.937).0016
IL-6> 0.7 pg/mL2.368 (1.291–4.342).0053 
IL-10> 2.6 pg/mL2.384 (1.190–4.775).01422.165 (1.002–4.678).0494
Clinical variantAggressive2.873 (1.137–7.268).0256 
PS2–42.901 (1.679–5.014).0001 
B symptomPresent1.967 (1.115–3.469).01952.309 (1.265–4.214).0064
LDH> 2N2.419 (1.397–4.191).00162.400 (1.364–4.224).0024
sIL-2R> 10,000 U/mL2.943 (1.563–5.541).0008 
Extranodal involvement≥ 22.061 (1.186–3.581).0103 

Associations between the serum level of each cytokine and OS among the ATLL patients with the aggressive variant

Among the ATLL patients with the aggressive variant, a high IL-5 level, a high IL-6 level and a high IL-10 level were each significantly associated with shorter survival (50% OS, 7.3 versus 13.3 months, 8.9 versus 21.0 months and 8.5 versus 22.0 months, respectively). There were no significant differences in the OS between those with a high or low level of IFN-γ, IL-2, IL-4 or TGF-β1 (data not shown).

Associations between the serum cytokine levels and clinicalcharacteristics among the ATLL patients with the aggressive variant

The clinical characteristics of ATLL patients with the aggressive variant were also analyzed according to the cytokine levels. Among the ATLL patients with the aggressive variant, a high IL-6 level was significantly associated with a high IL-10 level (p = 0.0203). A high IL-5 level was significantly associated with eosinophilia (P = 0.0002) and favored tumor involvement of the skin (p = 0.0083). A high IL-6 level was significantly associated with the presence of B symptoms (p = 0.0205) and a high sIL-2R level (p = 0.0083). A high IL-10 level was significantly associated with worse PS (p = 0.0446), leukocytosis (p = 0.0023), lymphocytosis (p = 0.0126), a high sIL-2R level (p = 0.0005) and favored tumor involvement of the peripheral blood (p = 0.0203) and the bone marrow (p = 0.0446). A high IFN-γ level was significantly associated with anemia (p = 0.0295) and hypoalbuminemia (p = 0.0427), and a low IFN-γ level was significantly associated with favored tumor involvement of the bone marrow (p = 0.0288). A low TGF-β1 level was significantly associated with thrombocytopenia (p = 0.0451), and a high TGF-β1 level was significantly associated with leukocytosis (p = 0.0381). None of the other clinical characteristics were associated with the status of each cytokine among the ATLL patients with the aggressive variant (data not shown).

Prognostic factors for ATLL patients with the aggressive variant

Univariate Cox proportional hazard analysis identified the following unfavorable prognostic factors in ATLL patients with the aggressive variant, with respect to the patients' survival: a high IL-5 level, a high IL-10 level, worse PS, a high LDH level and a high sIL-2R level (Table V). Multivariate analysis demonstrated that a high IL-5 level and a high LDH level were independent and significant unfavorable prognostic factors in ATLL patients with the aggressive variant (Table V).

Table V. Prognostic Factors Affecting Overall Survival Among the ATLL Patients with the Aggressive Variant
VariableUnfavorableUnivariateMultivariate
Hazard ratio (95% CI)p valueHazard ratio (95% CI)p value
IL-5> 1.7 pg/mL3.431 (1.720–6.843).00053.425 (1.707–6.874).0005
IL-10> 2.6 pg/mL2.241 (1.040–4.827).0394 
PS2–42.333 (1.305–4.171).0042 
LDH> 2N2.022 (1.129–3.621).01782.010 (1.119–3.610).0195
sIL-2R> 10,000 U/mL2.543 (1.217–5.314).0131 

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

In the present study, we found that high serum IL-5, IL-6 and IL-10 levels were significantly associated with worse prognosis in ATLL, and multivariate analysis showed that high serum IL-5 and IL-10 levels were independent and significant unfavorable prognostic factors.

A high serum IL-5 level did not seem to directly reflect the tumor burden, since it was not significantly associated with a high LDH level. In addition, the serum IL-5 concentration did not increase with disease progression at each step from asymptomatic HTLV-1 carrier to ATLL patient with the indolent variant to ATLL patient with the aggressive variant. Since it was reported that the ATLL cells of some, but not all, patients with ATLL express IL-5,15, 20, 21 the main source of serum IL-5 in patients with a high serum IL-5 level seems to be ATLL cells. Thus, patients with ATLL cells that express IL-5 seem to have an unfavorable prognosis, regardless of tumor burden. We found that there was a significant relationship between a high serum IL-5 level and eosinophilia among the ATLL patients, and this is in agreement with the study that reported that eosinophilia is induced by overproduction of IL-5.22 Eosinophilia is a significant prognostic factor in other hematological neoplasms such as Hodgkin lymphoma (HL), myelodysplastic syndrome (MDS) and primary cutaneous T-cell lymphoma (CTCL). Tissue eosinophilia in patients with HL was associated with poor prognosis;23 bone marrow eosinophilia was a predictor of poor survival in MDS;24 and circulating eosinophilia was associated with poor disease-free survival in primary CTCL.25 A high IL-5 level seems to be related to eosinophilia and lead to an unfavorable outcome in patients with hematological neoplasms such as HL, MDS and primary CTCL. However, little is known about the unfavorable clinical role of IL-5 in these diseases including ATLL, and further investigations are needed.

A high serum IL-10 level seemed to directly reflect the tumor burden because it was significantly associated with high LDH, high sIL-2R level and extranodal involvement of more than 1 site among the ATLL patients. In addition, the IL-10 concentration significantly increased with disease progression from asymptomatic HTLV-1 carrier to ATLL patient with the indolent variant to ATLL patient with the aggressive variant. This result is consistent with the study that reported that ATLL cells express IL-10.26 IL-10 has strong immunosuppressive effects in that it inhibits the proliferation of T cells and also inhibits the expression of Th1-associated cytokines including IFN-γ and IL-2.27, 28 Thus, excessive production of IL-10 by ATLL cells may impair the host's immune system and give rise to the host's immunosuppressive state, leading to an unfavorable outcome, and simultaneously, offer a selective survival advantage to ATLL cells in the host. Interestingly, the serum IL-10 concentration is also an important prognostic factor in other hematological neoplasms such as HL, non-HL and B-chronic lymphocytic leukemia.29, 30, 31

We found that a high IL-6 level was an unfavorable prognostic factor in ATLL on univariate analysis, in agreement with a previous report.32 However, a high IL-6 level did not remain as an independent prognostic factor in ATLL patients on multivariate analysis. It was reported that ATLL cells express IL-6,33 and the main source of serum IL-6 in ATLL patients seems to be tumor cells. IL-6 is a multifunctional cytokine; for instance, IL-6 induces the production of acute phase protein in hepatocytes, and it also induces the development of B symptoms.34, 35, 36 We found that a high IL-6 level was significantly associated with the presence of B symptoms. Collectively, we surmise that the serum IL-6 level reflects the patient's inflammatory state, which is mainly induced by ATLL cells.

The present study did not reveal any significant associations between the serum IFN-γ, IL-2, IL-4 or TGF-β1 levels and OS among the ATLL patients.

The serum IFN-γ concentration seemed to be associated with HTLV-1 infection, because the IFN-γ concentration was significantly higher in the asymptomatic carriers and ATLL patients than in the healthy volunteers. In the culture supernatants of peripheral blood mononuclear cells (PBMC), the IFN-γ concentration was significantly higher in PBMC obtained from patients with HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) than in PBMC obtained from asymptomatic HTLV-1 carriers.37 These results suggest that IFN-γ is produced by the host's Th1 cells or natural killer (NK) cells as an immune response to HTLV-1-infected cells in the HTLV-1-bearing host.

The serum TNF-α concentration was significantly lower in the asymptomatic HTLV-1 carriers than in the healthy volunteers. It has been shown that HTLV-1 can infect a wide variety of human cell types besides CD4-positive T-lymphocytes, such as NK cells.38, 39, 40, 41, 42, 43 Since it was reported that the effector function of HTLV-1-infected NK cells was impaired,39 the effector function of HTLV-1-infected macrophages including TNF-α production may also be impaired, which may lead to a low serum TNF-α level in HTLV-1-bearing hosts.

In the present study, there were no significant differences in the serum IL-2 and IL-4 levels among the ATLL patients, HTLV-1 carriers and healthy volunteers. This finding is consistent with the study that reported that ATLL cells produce small amounts of IL-2 and IL-4.15

TGF-β1 plays an important role in controlling the immune system of patients with ATLL, because similar to IL-10, TGF-β1 possesses an immunosuppressive function and is produced by ATLL cells.44, 45 However, the serum TGF-β1 level of the ATLL patients did not significantly differ from that of the healthy volunteers, and a low TGF-β1 level was significantly associated with thrombocytopenia among the ATLL patients in the present study. This result suggests that the source of serum TGF-β1 is platelets rather than ATLL cells, as suggested in a previous study.46

The clinical significance of a high serum level of each cytokine among the ATLL patients with the aggressive variant was nearly the same as that of the respective cytokine among all ATLL patients. In ATLL patients with the aggressive variant, univariate analysis showed that high serum IL-5 and IL-10 levels were significantly associated with unfavorable prognosis, and multivariate analysis showed that a high IL-5 level was an independent and significant unfavorable prognostic factor. Collectively, among the various cytokines examined in the present study, the serum IL-5 concentration was the most useful factor for predicting the prognosis of ATLL patients.

In conclusion, ATLL cells that produce Th2-associated cytokines such as IL-6 or Treg-associated cytokines such as IL-10 may disrupt the balanced production of cytokines and cause the host's immune system to break down, placing the host in a highly immunocompromised state. Furthermore, IL-5 produced by ATLL cells may induce a much greater shift in the Th1/Th2 balance in favor of a Th2 response in some ATLL patients, leading to an unfavorable prognosis. These results provide novel insights into the understanding of the immunocompromised state of ATLL patients. In addition, measurement of serum cytokine levels, which can be performed quickly and easily using a small serum sample, is useful for determining a suitable treatment strategy for individual ATLL patients, because it may be possible to predict the prognosis before treatment. IL-5 and IL-10 might be valuable targets for immune therapy as an alternative treatment strategy for patients with ATLL.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Ms. Yasuko Tsuji, Ms. Minako Nakashima and Ms. Akiko Sakaguchi for their excellent secretarial assistance, and also Ms. Chiori Fukuyama for her excellent technical assistance. We also thank Dr. Sadao Suzuki of the Department of Health Promotion and Preventive Medicine, Nagoya City University Graduate School of Medical Sciences for his kind advice on the statistical analyses.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Kikuchi M, Jaffe ES, Ralfkiaer E. Adult T-cell leukaemia/lymphoma. In: JaffeES, HarrisNL, SteinH, VardimanJW, eds. Tumors of haematopoietic and lymphoid tissues. Lyon, France: International Agency for Research on Cancer (IARC), 2001. 2003.
  • 2
    Shimoyama M. Diagnostic criteria and classification of clinical subtypes of adult T-cell leukaemia-lymphoma. A report from the Lymphoma Study Group (1984–87). Br J Haematol 1991; 79: 42837.
  • 3
    Mosmann TR, Sad S. The expanding universe of T-cell subsets: Th1, Th2 and more. Immunol Today 1996; 17: 138.
  • 4
    Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science 2003; 299: 105761.
  • 5
    Vieira PL, Christensen JR, Minaee S, O'Neill EJ, Barrat FJ, Boonstra A, Barthlott T, Stockinger B, Wraith DC, O'Garra A. IL-10-secreting regulatory T cells do not express Foxp3 but have comparable regulatory function to naturally occurring CD4+CD25+ regulatory T cells. J Immunol 2004; 172: 598693.
  • 6
    Chen Y, Kuchroo VK, Inobe J, Hafler DA, Weiner HL. Regulatory T cell clones induced by oral tolerance: suppression of autoimmune encephalomyelitis. Science 1994; 265: 123740.
  • 7
    O'Garra A, Vieira P. Regulatory T cells and mechanisms of immune system control. Nat Med 2004; 10: 8015.
  • 8
    Ishida T, Utsunomiya A, Iida S, Inagaki H, Takatsuka Y, Kusumoto S, Takeuchi G, Shimizu S, Ito M, Komatsu H, Wakita A, Eimoto T, et al. Clinical significance of CCR4 expression in adult T-cell leukemia/lymphoma: its close association with skin involvement and unfavorable outcome. Clin Cancer Res 2003; 9: 362534.
  • 9
    Yoshie O. Expression of CCR4 in adult T-cell leukemia. Leuk Lymphoma 2005; 46: 18590.
  • 10
    Yoshie O, Imai T, Nomiyama H. Chemokines in immunity. Adv Immunol 2001; 78: 57110.
  • 11
    Ishida T, Iida S, Akatsuka Y, Ishii T, Miyazaki M, Komatsu H, Inagaki H, Okada N, Fujita T, Shitara K, Akinaga S, Takahashi T, et al. The CC chemokine receptor 4 as a novel specific molecular target for immunotherapy in adult T-cell leukemia/lymphoma. Clin Cancer Res 2004; 10: 752939.
  • 12
    Matsubara Y, Hori T, Morita R, Sakaguchi S, Uchiyama T. Phenotypic and functional relationship between adult T-cell leukemia cells and regulatory T cells. Leukemia 2005; 19: 4823.
  • 13
    Kohno T, Yamada Y, Akamatsu N, Kamihira S, Imaizumi Y, Tomonaga M, Matsuyama T. Possible origin of adult T-cell leukemia/lymphoma cells from human T lymphotropic virus type-1-infected regulatory T cells. Cancer Sci 2005; 96: 52733.
  • 14
    Yamada Y, Ohmoto Y, Hata T, Yamamura M, Murata K, Tsukasaki K, Kohno T, Chen Y, Kamihira S, Tomonaga M. Features of the cytokines secreted by adult T cell leukemia (ATL) cells. Leuk Lymphoma 1996; 21: 4437.
  • 15
    Noma T, Nakakubo H, Sugita M, Kumagai S, Maeda M, Shimizu A, Honjo T. Expression of different combinations of interleukins by human T cell leukemic cell lines that are clonally related. J Exp Med 1989; 169: 18538.
  • 16
    Hanon E, Goon P, Taylor GP, Hasegawa H, Tanaka Y, Weber JN, Bangham CRM. High production of interferon gamma but not interleukin-2 by human T-lymphotropic virus type I-infected peripheral blood mononuclear cells. Blood 2001; 98: 7216.
  • 17
    Uozumi K, Hanada S, Ohno N, Ishitsuka K, Shimotakahara S, Otsuka M, Chyuman Y, Nakahara K, Takeshita T, Kuwazuru Y. Combination chemotherapy (RCM protocol: response-oriented cyclic multidrug protocol) for the acute or lymphoma type adult T-cell leukemia. Leuk Lymphoma 1995; 18: 31723.
  • 18
    Yamada Y, Tomonaga M, Fukuda H, Hanada S, Utsunomiya A, Tara M, Sano M, Ikeda S, Takatsuki K, Kozuru M, Araki K, Kawano F, et al. A new G-CSF-supported combination chemotherapy, LSG15, for adult T-cell leukaemia-lymphoma: Japan Clinical Oncology Group Study 9303. Br J Haematol 2001; 113: 37582.
  • 19
    Tobinai K, Hotta T. Clinical trials for malignant lymphoma in Japan. Jpn J Clin Oncol 2004; 34: 36978.
  • 20
    Yamagata T, Mitani K, Ueno H, Kanda Y, Yazaki Y, Hirai H. Triple synergism of human T-lymphotropic virus type 1-encoded tax, GATA-binding protein, and AP-1 is required for constitutive expression of the interleukin-5 gene in adult T-cell leukemia cells. Mol Cell Biol 1997; 17: 427281.
  • 21
    Ogata M, Ogata Y, Kohno K, Uno N, Ohno E, Ohtsuka E, Saburi Y, Kamberi P, Nasu M, Kikuchi H. Eosinophilia associated with adult T-cell leukemia: role of interleukin 5 and granulocyte-macrophage colony-stimulating factor. Am J Hematol 1998; 59: 2425.
  • 22
    Clutterbuck EJ, Hirst EM, Sanderson CJ. Human interleukin-5 (IL-5) regulates the production of eosinophils in human bone marrow cultures: comparison and interaction with IL-1, IL-3, IL-6, and GMCSF. Blood 1989; 73: 150412.
  • 23
    von Wasielewski R, Seth S, Franklin J, Fischer R, Hubner K, Hansmann ML, Diehl V, Georgii A. Tissue eosinophilia correlates strongly with poor prognosis in nodular sclerosing Hodgkin's disease, allowing for known prognostic factors. Blood 2000; 95: 120713.
  • 24
    Matsushima T, Handa H, Yokohama A, Nagasaki J, Koiso H, Kin Y, Tanaka Y, Sakura T, Tsukamoto N, Karasawa M, Itoh K, Hirabayashi H, et al. Prevalence and clinical characteristics of myelodysplastic syndrome with bone marrow eosinophilia or basophilia. Blood 2003; 101: 338690.
  • 25
    Tancrede-Bohin E, Ionescu MA, de La Salmoniere P, Dupuy A, Rivet J, Rybojad M, Dubertret L, Bachelez H, Lebbe C, Morel P. Prognostic value of blood eosinophilia in primary cutaneous T-cell lymphomas. Arch Dermatol 2004; 140: 105761.
  • 26
    Mori N, Gill PS, Mougdil T, Murakami S, Eto S, Prager D. Interleukin-10 gene expression in adult T-cell leukemia. Blood 1996; 88: 103545.
  • 27
    Fiorentino DF, Bond MW, Mosmann TR. Two types of mouse T helper cell. IV. Th2 clones secrete a factor that inhibits cytokine production by Th1 clones. J Exp Med 1989; 170: 208195.
  • 28
    Taga K, Mostowski H, Tosato G. Human interleukin-10 can directly inhibit T-cell growth. Blood 1993; 81: 296471.
  • 29
    Vassilakopoulos TP, Nadali G, Angelopoulou MK, Siakantaris MP, Dimopoulou MN, Kontopidou FN, Rassidakis GZ, Doussis-Anagnostopoulou IA, Hatzioannou M, Vaiopoulos G, Kittas C, Sarris AH, et al. Serum interleukin-10 levels are an independent prognostic factor for patients with Hodgkin's lymphoma. Haematologica 2001; 86: 27481.
  • 30
    Blay JY, Burdin N, Rousset F, Lenoir G, Biron P, Philip T, Banchereau J, Favrot MC. Serum interleukin-10 in non-Hodgkin's lymphoma: a prognostic factor. Blood 1993; 82: 216974.
  • 31
    Fayad L, Keating MJ, Reuben JM, O'Brien S, Lee B-N, Lerner S, Kurzrock R. Interleukin-6 and interleukin-10 levels in chronic lymphocytic leukemia: correlation with phenotypic characteristics and outcome. Blood 2001; 97: 25663.
  • 32
    Yamamura M, Yamada Y, Momita S, Kamihira S, Tomonaga M. Circulating interleukin-6 levels are elevated in adult T-cell leukaemia/lymphoma patients and correlate with adverse clinical features and survival. Br J Haematol 1998; 100: 12934.
  • 33
    Mori N, Shirakawa F, Shimizu H, Murakami S, Oda S, Yamamoto K-l, Eto S. Transcriptional regulation of the human interleukin-6 gene promoter in human T-cell leukemia virus type I-infected T-cell lines: evidence for the involvement of NF-kappa B. Blood 1994; 84: 290411.
  • 34
    Castell JV, Gomez-Lechon MJ, David M, Hirano T, Kishimoto T, Heinrich PC. Recombinant human interleukin-6 (IL-6/BSF-2/HSF) regulates the synthesis of acute phase proteins in human hepatocytes. FEBS Lett 1988; 232: 34750.
  • 35
    Weber J, Yang JC, Topalian SL, Parkinson DR, Schwartzentruber DS, Ettinghausen SE, Gunn H, Mixon A, Kim H, Cole D. Phase I trial of subcutaneous interleukin-6 in patients with advanced malignancies. J Clin Oncol 1993; 11: 499506.
  • 36
    Heinrich PC, Castell JV, Andus T. Interleukin-6 and the acute phase response. Biochem J 1990; 265: 62136.
  • 37
    Santos SB, Porto AF, Muniz AL, de Jesus AR, Magalhães E, Melo A, Dutra WO, Gollob KJ, Carvalho EM. Exacerbated inflammatory cellular immune response characteristics of HAM/TSP is observed in a large proportion of HTLV-I asymptomatic carriers. BMC Infect Dis 2004; 4: 7.
  • 38
    Sagara Y, Ishida C, Inoue Y, Shiraki H, Maeda Y. 71-kilodalton heat shock cognate protein acts as a cellular receptor for syncytium formation induced by human T-cell lymphotropic virus type 1. J Virol 1998; 72: 53541.
  • 39
    Lo KMS, Vivier E, Roche N, Dehni G, Levine H, Haseltine WA, Anderson P. Infection of human natural killer (NK) cells with replication-defective human T cell leukemia virus type I provirus. Increased proliferative capacity and prolonged survival of functionally competent NK cells. J Immunol 1992; 149: 41018.
  • 40
    Yoshikura H, Nishida J, Yoshida M, Kitamura Y, Takaku F, Ikeda S. Isolation of HTLV derived from Japanese adult T-cell leukemia patients in human diploid fibroblast strain IMR90 and the biological characters of the infected cells. Int J Cancer 1984; 33: 7459.
  • 41
    Yamamoto N, Matsumoto T, Koyanagi Y, Tanaka Y, Hinuma Y. Unique cell lines harbouring both Epstein-Barr virus and adult T-cell leukaemia virus, established from leukaemia patients. Nature 1982; 299: 3679.
  • 42
    Hoxie JA, Matthews DM, Cines DB. Infection of human endothelial cells by human T-cell leukemia virus type I. Proc Natl Acad Sci U S A. 1984; 81: 75915.
  • 43
    Clapham P, Nagy K, Cheingsong-Popov R, Exley M, Weiss RA. Productive infection and cell-free transmission of human T-cell leukemia virus in a nonlymphoid cell line. Science 1983; 222: 11257.
  • 44
    Niitsu Y, Urushizaki Y, Koshida Y, Terui K, Mahara K, Kohgo Y, Urushizaki I. Expression of TGF-beta gene in adult T cell leukemia. Blood 1988; 71: 2636.
  • 45
    Kim SJ, Kehrl JH, Burton J, Tendler CL, Jeang KT, Danielpour D, Thevenin C, Kim KY, Sporn MB, Roberts AB. Transactivation of the transforming growth factor beta 1 (TGF-beta 1) gene by human T lymphotropic virus type 1 tax: a potential mechanism for the increased production of TGF-beta 1 in adult T cell leukemia. J Exp Med 1990; 172: 1219.
  • 46
    Kropf J, Schurek JO, Wollner A, Gressner AM. Immunological measurement of transforming growth factor-beta 1 (TGF-beta1) in blood; assay development and comparison. Clin Chem 1997; 43: 196574.