Soluble intercellular adhesion molecule-1 (s-ICAM-1/s-CD54) in diffuse large B-cell lymphoma: association with clinical characteristics and outcome

M. J. Terol1,+, M. Tormo1, J. A. Martinez-Climent1, I. Marugan1, I. Benet1, A. Ferrandez2, A. Teruel1, R. Ferrer3 and J. García-Conde1

Departments of 1 Hematology and Medical Oncology and 2 Pathology, Hospital Clínico Universitario, University of Valencia, Valencia 3 Hematology Section, Hospital Francesc de Borja, Gandia, Spain

Received 8 April 2002; revised 26 July 2002; accepted 9 September 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

High serum levels of soluble intercellular adhesion molecule-1(s-ICAM-1/s-CD54) have been associated with adverse clinical features and poor outcome in chronic lymphocytic leukemia, Hodgkin’s disease and non-Hodgkin’s lymphoma, but their value in the different subtypes of non-Hodgkin’s lymphoma has not been well addressed.

Patients and methods:

Our aim was to study the serum levels of s-ICAM-1 in diffuse large B-cell lymphoma (DLBCL) and to correlate them with clinical characteristics and outcome. We analyzed the serum levels of s-ICAM-1 in a series of 55 patients with DLBCL diagnosed in a single institution. s-ICAM-1 levels were quantified by an immunoenzymatic assay. Median age was 62 years (range 22–96); 29 (53%) were male. Twenty-eight (51%) presented with advanced clinical stage (III/IV), 32 (58%) had extranodal involvement, 28 (51%) had high serum lactate dehydrogenase (LDH) and 23 (43%) had high ß2-microglobulin levels. All patients received anthracycline-containing regimens. Correlation between clinical variables and s-ICAM-1 levels were tested with the Mann–Whitney U-test and survival was plotted by the Kaplan–Meier method, and curves compared with the log-rank test.

Results:

Serum levels of s-ICAM-1 were significantly increased in patients with DLBCL compared with normal controls (589 ± 487 versus 279 ± 65 ng/ml, respectively; P <0.001). Higher levels of s-ICAM-1 were present in patients with B symptoms, advanced stage and increased LDH and ß2-microglobulin. s-ICAM-1 levels also correlated with achievement of a complete response. Patients with s-ICAM-1 over 668 ng/ml had a shorter time to treatment failure (TTF) (3-year TTF, 59% versus 20%, respectively; P = 0.01) and overall survival (OS) (3-year OS, 58% versus 22%, respectively; P = 0.04) than the remainders. When only low and low–intermediate risk patients in the international prognostic index score were considered, those with s-ICAM-1 over 668 ng/ml also had worse TTF and OS.

Conclusions:

In DLBCL, s-ICAM-1 levels correlated with high tumor burden and lymphoma dissemination and may contribute to assessment of prognosis.

Key words: diffuse large B-cell lymphoma, prognosis, soluble intercellular adhesion molecule-1 (CD54)


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Intercellular adhesion molecule-1 (ICAM-1; CD54) is a cell-surface receptor from the immunoglobulin superfamily that is involved in several physiological processes through the adhesion to its ligand, the ß2-integrin, LFA-1, present on lymphocytes [1]. ICAM-1 is expressed at low levels on lymphocytes and endothelium and its expression is increased on cytokine stimulation [2]. In the immune response, ICAM-1 participates as an adhesion molecule as well as a co-stimulatory molecule to improve both antigen recognition by the T-cell receptor complex and subsequent T-cell activation [3]. ICAM-1 is also involved in lymphoid trafficking and extravasation through lymphocyte/endothelial interactions [4]. Finally, in normal B cells it also mediates homotypic adhesions.

There is some evidence that ICAM-1 is involved in neoplastic dissemination. It is expressed at high levels in solid tumors, such as hepatocellular carcinoma [5], melanoma [6] and bladder cancer [7], with this expression correlating with metastatic dissemination in these tumors. In B-cell lymphoproliferative disorders the tumoral expression of ICAM-1 is closely related to the degree of cell maturation [8, 9]. Thus, mantle-cell-derived lymphoproliferative diseases, such as chronic lymphocytic leukemia (CLL) or mantle-cell lymphoma, are often negative or weakly positive for ICAM-1, whereas ICAM-1 expression is more heterogeneous in germinal center cell lymphomas such as follicular or diffuse large-cell subtypes. It has been suggested that reduction of this molecule on the neoplastic lymphoid cells could impair the T-cell recognition with this contributing to neoplastic dissemination through a defective antitumor response [10]. In accord with this, we previously described a positive correlation between low expression of ICAM-1 and advanced stage, extranodal involvement, bone marrow infiltration, poor response to treatment and worse survival in patients with non-Hodgkin’s lymphoma, especially in the aggressive subgroup [11]. Recently, loss of ICAM-1 expression on lymphoma cells has been associated with decreased tumor-infiltrating T lymphocytes in diffuse large B-cell lymphoma (DLBCL) [12].

ICAM-1 can be shed from the cellular surface into serum, resulting in a soluble form (s-ICAM-1) which is also capable of recognizing its ligand LFA-1. Recently, several studies have pointed out the important prognostic value of serum levels of this molecule in Hodgkin’s disease [13], CLL [14] and non-Hodgkin’s lymphoma [15]. In the last setting, high ICAM-1 levels have been associated with adverse prognostic factors such as advanced stage, high ß2-microglobulin levels and poor response to treatment, specially in high-grade subgroups. However, most series contain several histological subtypes making the interpretation of results difficult.

The aim of the present study was to determine the serum levels of ICAM-1 at diagnosis in a series of patients consecutively diagnosed with DLBCL in our institution, and to correlate them with biological and clinical features, as well as response to treatment and outcome.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients: clinical data and treatment
Fifty-five patients consecutively diagnosed with DLBCL in a single institution between 1993 and 2000 were included in the present study. The availability of frozen serum obtained at diagnosis was the only criterion for inclusion. Patients who tested positive for the human immunodeficiency virus were excluded from the study.

Median age of the series was 62 years (range 22–96). Twenty-nine patients (53%) were male. In all cases the diagnosis was based on histological criteria according to the REAL/WHO classification. The main initial characteristics of the patients are listed in Table 1. Advanced Ann Arbor stage was observed in 28 patients (51%) and extranodal involvement in 32 patients (58%), including bone marrow infiltration in five patients (9%). Nineteen cases had high or high–intermediate risk according to the international prognostic index (IPI) [16]. The median follow-up of the alive patients was 40 months (range 14–113).


View this table:
[in this window]
[in a new window]
 
Table 1. Main clinical characteristics of the 55 patients with diffuse large B-cell lymphoma
 
Staging procedures included thoracic, abdominal and pelvic computed tomography scans and bone marrow biopsy in all cases. All patients were treated with anthracycline-containing chemotherapies; the regimen administered depended on the risk according to the tumor score system [17] and IPI [16]. Those patients who scored <3 points received six cycles of CHOP (cyclophosphamide, doxorubicin, vincristine and prednisone), whereas those with >=3 points received the alternating triple therapy schedule [18] for a total number of nine cycles. Post-therapy restaging consisted of repetition of tests and/or biopsies that had previously had abnormal results. Complete response (CR) was defined as the total disappearance of tumor masses and disease-related symptoms and normalization of abnormal test results for at least 2 months. Partial response was defined as a decrease in tumor mass or organ infiltration of at least 50%, along with the disappearance of disease-related symptoms. Patients not included in these categories and early deaths were considered non-responders.

Determination of serum ICAM-1 levels
Serum samples from patients with DLBCL were obtained at diagnosis before treatment. Moreover, serum samples from 30 healthy donors were obtained in order to have a normal value. The soluble ICAM-1 (s-ICAM-1) levels in serum were determined by enzyme immunoassay kits purchased from R&D Systems (Abingdon, UK). This assay is based on the simultaneous reaction of s-ICAM-1 present in the sample or standard to two antibodies directed against different epitopes on the s-ICAM-1 molecule. The capture antibody is coated onto the walls of the microtiter wells and the other is conjugated to the enzyme horseradish peroxidase. s-ICAM-1 present in the serum forms a bridge between the two antibodies. After removal of unbound material by aspiration and washing, the amount of conjugate bound to the well is detected by reaction with a substrate (tetramethylbenzidine) specific for the enzyme, which yields a colored product proportional to the amount of conjugate, and thus s-ICAM-1 present in the serum. The colored product is quantified using a spectrophotometer set a 450 nm with a correction wavelength of 620 nm. By analyzing standards of known s-ICAM-1 concentration coincident with samples and plotting a curve of signal versus concentration, the concentration of unknown samples could be determined. The sensitivity of the assay was <0.35 ng/ml, the inter-assay coefficient of variation (CV) was <8%, and the intra-assay CV varied from 3.3% to 4.8%.

Statistical analysis
s-ICAM-1 levels are presented as mean ± standard deviation (SD). Comparison of s-ICAM-1 levels and clinical and biological variables were made by the Mann–Whitney U-test. Survival curves were built by the Kaplan–Meier method [19], and the curves compared with the long-rank test [20]. Univariate and multivariate analysis for response and survival were made by a logistic regression and the Cox regression model [21], respectively. A two-sided P value of <=0.05 was considered to be statistically significant.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
s-ICAM-1 in patients versus healthy controls
The serum levels of s-ICAM-1 (mean ± SD) in the 30 healthy donors were 279 ± 65 ng/ml, with no patient showing a serum value higher than 416 ng/ml. In the 55 patients with DLBCL such levels were significantly higher (589 ± 487 ng/ml; P <0.001) compared with the control values.

s-ICAM-1 and clinical features
The correlations between s-ICAM-1 levels and the main clinical and biological features are detailed in Table 2. s-ICAM-1 levels correlated with some factors depending on the patient: those with B symptoms had significantly higher levels of s-ICAM-1 than the others (P = 0.034). Those with non-ambulatory performance status also had higher levels of s-ICAM-1, although this difference did not reach statistical significance (P = 0.09). No correlation was observed between s-ICAM-1 levels and age.


View this table:
[in this window]
[in a new window]
 
Table 2. Correlation between soluble intercellular adhesion molecule-1 (s-ICAM-1) levels and the main clinical features
 
s-ICAM-1 levels were also associated with tumor burden and dissemination. Patients with advanced stage (III/IV), high lactate dehydrogenase (LDH) and high ß2-microglobulin levels presented with significantly higher levels of s-ICAM-1 compared with the remainders. We did not observe any difference in s-ICAM-1 levels between patients with extranodal involvement compared with those with nodal disease, or between those with two or more extranodal sites versus the others. Finally, no relationship was found between s-ICAM-1 levels and the IPI score.

s-ICAM-1, response to treatment and outcome
In the univariate analysis, there was a significant negative correlation between the levels of s-ICAM-1 and achievement of a CR as either a continuous or dichotomized variable. Patients with low levels of s-ICAM-1 reached CR more frequently than those with normal s-ICAM-1 (CR rates 82% versus 42%, respectively; P = 0.023). When a logistic regression for response achieved was performed including IPI, ß2-microglobulin and s-ICAM-1 levels, only s-ICAM-1, as both a continuous or dichotomized variable, was predictive for response [relative risk (RR) 2.37, 95% confidence interval (CI) 1.03–5.48; P = 0.008].

Three-year time to treatment failure (TTF) and survival of the whole series were 50% and 49%, respectively. As both a continuous or dichotomized variable, s-ICAM-1 levels correlated with outcome. Patients with levels of s-ICAM-1 over 668 ng/ml (percentile 75) had a shorter TTF than the others (3-year TTF 59% versus 20%, respectively; P = 0.01) (Figure 1A). Moreover, patients with low s-ICAM-1 levels showed longer overall survival (OS) then the remainders (3-year OS, 58% versus 22%, respectively; P = 0.04) (Figure 1B). When only patients with low IPI scores were selected, patients with s-ICAM-1 levels over 668 ng/ml also had a shorter TTF (3-year TTF 67% versus 14%; P = 0.0002) (Figure 2A) and survival (3-year OS 68% versus 28%; P = 0.006) than the remainders (Figure 2B). In the univariate analysis, additional variables with prognostic value for survival were: advanced stage, LDH, ß2-microglobulin, performance status, extranodal involvement and the IPI. Differences in CR rate, 3-year TTF and 3-year OS based on each parameter of the IPI are detailed in Table 3. In the Cox regression analysis performed by pairs, s-ICAM-1 retained its prognostic significance apart from advanced stage, LDH, ß2-microgobulin and the IPI (low/low–intermediate versus high/high–intermediate). When this test was performed taking into account IPI score (low/low–intermediate risk versus high/high–intermediate), ß2-microglobulin and s-ICAM-1(<668 versus >668 ng/ml), only s-ICAM-1 retained its prognostic significance (RR 4.45, 95% CI 1.381–14.28; P = 0.01).



View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Time to treatment failure (A) and overall survival (B) of the global series of patients based on serum levels of soluble intercellular adhesion molecule-1 (s-ICAM-1).

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 2. Time to treatment failure (A) and overall survival (B) of the low/low–intermediate risk patients in the international prognostic index based on serum levels of soluble intercellular adhesion molecule-1 (s-ICAM-1).

 

View this table:
[in this window]
[in a new window]
 
Table 3. Impact on response, overall survival (OS) and time to treatment failure (TTF) of the different variables included in the international prognostic index
 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
DLBCL is a very heterogeneous disease from both the clinical and biological point of view. A large number of clinical prognostic factors have been identified over the last 20 years and some of them, such as those included in the IPI [16], contribute to the clinical management of these patients, by providing a clinical risk profile of each patient.

More recently, a large number of biological factors have also been described. A high proliferation rate evaluated by the expression of Ki-67 antigen on tumor cells identified a group of patients who had a very poor outcome [22]. Bcl-2 protein expression also impacts negatively on survival [23]. Among the soluble adhesion molecules, patients with high s-CD44 levels showed a significantly decreased overall and progression-free survival [24] compared with those with low levels. All these factors contribute to the assessment of prognosis in these patients, as they reflect biological aspects intrinsic to tumor biology, such as growth rate and sensitivity to chemotherapy, as well as other more complex aspects such as dissemination.

ICAM-1 is an adhesion molecule implicated in many physiological functions: it participates in the lymphocyte recirculation and homing and also play a role in the immune response as an adhesive and co-stimulator molecule. Low expression of ICAM-1 has been correlated with dissemination in low-grade lymphoproliferative disorders [25]. In our previous study, its absence or low positivity was associated with advanced stage, bone marrow infiltration and an adverse clinical course in diffuse large cell lymphoma [11]. A soluble form of this molecule has been identified and high levels of serum s-ICAM-1 correlated with adverse prognostic factors in non-Hodgkin’s lymphoma [2629], CLL [13] or Hodgkin’s disease [15]. In the last, s-CD54 was identified as an independent prognostic factor from other well-known factors in this entity. In high-grade non-Hodgkin’s lymphoma, Christiansen et al. [14] found a correlation between s-CD54 levels and disease activity as well as tumor burden and dissemination. Nevertheless, this group included patients of different histologies and phenotypes. In agreement with this, but limited to the DLBCL subtype, we report on a significant correlation between high levels of s-ICAM-1 and B symptoms, advanced stage, and increased levels of LDH and ß2-microglobulin. Also in agreement with Christiansen et al., we did not find any association with age or extranodal involvement.

In non-Hodgkin’s lymphoma, as a dichotomized variable, patients with increased levels of s-ICAM-1 had a shorter survival compared with those with low levels [14, 29], but these data have not been confirmed either as a continuous variable or in the multivariate analysis [14]. On the other hand, in primary extranodal lymphomas [25], high serum s-ICAM-1 associated significantly with shorter disease-free survival and OS, in both univariate and multivariate analysis. In our series, patients with levels of s-ICAM-1 higher than the 75th percentile had a worse prognosis in terms of TTF and survival. In DLBCL, serum ß2-microglobulin adds prognostic information to the IPI system [30], but it is not always determined in all institutions. In our series, in the multivariate analysis, s-ICAM-1 levels retained its prognostic significance apart from the IPI and ß2-microglobulin, although these data need to be confirmed in larger series due to the small number of patients. As has been suggested by other authors [31, 32], the performance status, the serum LDH value and extranodal involvement (two or more) are the main discriminating factors in terms of CRs and 3-year TTF and OS. Furthermore, s-ICAM-1 seemed to identify a group of patients with a worse outcome among those with low risk in the IPI system. All, these data suggest that this molecule adds prognostic significance, probably dependent on biological aspects of the disease.

The origin of s-ICAM-1 and its physiological role are not completely known. It is probably shed from the cell surface of activated endothelium and hematopoietic cells and retains its functional capability to recognize its ligand, LFA-1, present on the lymphocyte membrane. High levels of s-ICAM-1 may adhere to lymphocyte LFA-1, making neoplastic cells capable of avoiding the control of the immunological system. On the other hand, adhesion to bone marrow or lymph node extracellular matrix through ß1- and ß2-integrins impairs the apoptotic process, prolonging cell survival and contributing to chemoresistance in several lymphoid malignancies such as multiple myeloma [33] and CLL [34, 35]. This could also occur in diffuse large cell lymphoma, as it has been suggested by Fornarini et al. [36], although it has not been fully demonstrated.

In conclusion, in DLBCL, high serum levels of s-ICAM-1 correlate with disease activity, tumor burden and dissemination, low rate of response and adverse prognosis in terms of TTF and OS. Quantification of s-ICAM-1 levels may identify a subgroup of patients with worse prognosis in both the whole group and the low/low–intermediate risk group of the IPI.


    Acknowledgements
 
We are very grateful to A. López-Guillermo for his instructive comments during the preparation of this manuscript. This work was supported by the grant from the Fondo de Investigacion Sanitaria de la Seguridad Social (FIS 98/0491) from the Spanish Ministry of Health.


    Footnotes
 
+ Correspondence to: Dr M. J. Terol, Department of Hematology and Medical Oncology, Hospital Clínico Universitario, Avda Blasco Ibañez 17, Valencia 46010, Spain. Tel: +34-96-3862625; Fax: +34-96-3622238; E-mail: terol39{at}mail.ono.es Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Malgoba MW, Sanders ME, Ginther Luce GE et al. ICAM-1: a ligand for LFA-1 dependent adhesion of B, T and myeloid cells. Nature 1988; 331: 86–88.[CrossRef][ISI][Medline]

2. Voraberger G, Schafer R, Stratowa C. Cloning of the human gene for intercellular adhesion molecule-1 and analysis of its 5'-regulatory region. Induction by cytokines and phorbol ester. J Immunol 1991; 147: 2777–2786.[Abstract/Free Full Text]

3. Kuhlman P, Moy VT, Lollo BA, Brian AA. The accessory function of murine ICAM-1 in T lymphocyte activation. Contributions of adhesion and co-activation. J Immunol 1991; 146: 1773–1782.[Abstract/Free Full Text]

4. Wawryck SO, Novotny JR, Wicks IP et al. The role of LFA-1/ICAM-1 interaction in human leukocyte homing and adhesion. Immunol Rev 1989; 108: 135–161.[ISI][Medline]

5. Momosaki S, Yano H, Ogasawara S et al. Expression of intercellular adhesion molecule 1 in human hepatocellular carcinoma. Hepatology 1995; 22: 1708–1713.[CrossRef][ISI][Medline]

6. Johnson JP, Stade BG, Holzmann B et al. De novo expression of intercellular-adhesion molecule 1 in melanoma correlates with increased risk of metastasis. Proc Natl Acad Sci USA 1989; 86: 641–644.[Abstract]

7. Tomita Y, Watanabe H, Kobayashi H et al. Expression of intercellular adhesion molecule-1 on transitional cell cancer. Am J Pathol 1993; 143: 191–198.[Abstract]

8. Horst E, Radaszkiewicz T, Hofftman-den Otter A et al. Expression of the leukocyte integrin LFA-1 (CD11a/CD18) and its ligand ICAM-1 (CD54) in lymphoid malignancies is related to lineage derivation and stage of differentiation but not to tumor grade. Leukemia 1991; 5: 848–853.[ISI][Medline]

9. Boyd AW, Dunn SM, Fecondo JV et al. Regulation of expression of a human intercellular adhesion molecule (ICAM-1) during lymphohematopoietic differentiation. Blood 1989; 73: 1896–1903.[Abstract]

10. Cavallo F, Martin-Fontecha A, Bellone M et al. Co-expression of B7-1 and ICAM-1 on tumors is required for rejection and the establishment of a memory response. Eur J Immunol 1995; 25: 1154–1162.[ISI][Medline]

11. Terol MJ, López-Guillermo A, Bosch F et al. Expression of the adhesion molecule ICAM-1 in non-Hodgkin’s lymphoma: relationship with tumor dissemination and prognostic importance. J Clin Oncol 1998; 16: 35–40.[Abstract]

12. Stopeck AT, Gessner A, Miller TP et al. Loss of B7.2 (CD86) and intracellular adhesion molecule 1 (CD54) expression is associated with decreased tumor-infiltrating T lymphocytes in diffuse B-cell large-cell lymphoma. Clin Cancer Res 2000; 6: 3904–3909.[Abstract/Free Full Text]

13. Christiansen I, Gidlöf C, Wallgren AC et al. Serum levels of soluble intercellular adhesion molecule-1 are increased in chronic B-lymphocyte leukemia and correlate with clinical stage and prognostic markers. Blood 1994; 84: 3010–3016[Abstract/Free Full Text]

14. Christiansen I, Gidlöf C, Kälkner KM et al. Elevated serum levels of soluble ICAM-1 in non-Hodgkin’s lymphomas correlate with tumor burden, disease activity and other prognostic markers. Br J Haematol 1996; 92: 639–646.[ISI][Medline]

15. Christiansen I, Enblad G, Kälkner KM et al. Soluble ICAM-1 in Hodgkin’s disease: a promising independent predictive marker for survival. Leuk Lymphoma 1995; 19: 243–251.[ISI][Medline]

16. The international non-Hodgkin’s lymphoma prognostic factors project. A predictive model for aggressive non-Hodgkin’s lymphoma. N Engl J Med 1993; 329: 987–994.[Abstract/Free Full Text]

17. Rodriguez J, Cabanillas F, McLaughlin et al. A proposal for a simple staging system for intermediate grade lymphoma and immunoblastic lymphoma based on the tumor score. Ann Oncol 1992; 3: 711–717.[Abstract]

18. Cabanillas F, Rodriguez-Diaz Pavon J, Hagemeister FB et al. Alternating triple therapy for the treatment of intermediate grade and immunoblastic lymphoma. Ann Oncol 1998; 9: 511–518.[Abstract]

19. Kaplan GL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

20. Peto R, Pike MC. Conservation of the approximation {Sigma}(O-E)2/E in the log-rank test for survival data or tumor incidence data. Biometrics 1973; 29: 759–784.

21. Cox DR. Regression model and life tables. J R Stat Assoc 1972; 34: 187–220.

22. Miller TP, Grogan TM, Dahlberg S et al. Prognostic significance of the Ki-67-associated proliferative antigen in aggressive non-Hodgkin’s lymphomas: a prospective Southwest Oncology Group trial. Blood 1994; 83: 1460–1466.[Abstract/Free Full Text]

23. 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; 90: 244–251.[Abstract/Free Full Text]

24. Niitsu N, Iijima K. High serum soluble CD44 is correlated with a poor outcome aggressive non-Hodgkin’s lymphoma. Leuk Res 2002; 26: 241–248.[CrossRef][ISI][Medline]

25. Stauder R, Greil R, Schultz TF et al. Expression of leukocyte function-associated antigen-1 and 7F7-antigen, an adhesion molecule related to intercellular adhesion molecule-1 (ICAM-1) in non-Hodgkin’s lymphoma and leukemias: possible influence on growth pattern and leukaemic behaviour. Clin Exp Immunol 1989; 77: 234–238.[ISI][Medline]

26. Perez-Encinas M, Quintas A, Bendana A et al. Correlation and prognostic value of serum soluble ICAM-1, beta-1 microglobulin, and IL-2alphaR levels in non-Hodgkin’s lymphoma. Leuk Lymphoma 1999; 33: 551–558.[ISI][Medline]

27. Harabuchi Y, Kataura A, Imai K. Circulation intercellular adhesion molecule-1 and its cellular expression in head and neck non-Hodgkin’s lymphomas, including lethal midline granuloma. Ann Otol Rhinol Laryngol 1996; 105: 634–642.[ISI][Medline]

28. Lei KI, Johnson PJ. The prognostic significance of serum levels of soluble intercellular adhesion molecule-1 in patients with primary extranodal non-Hodgkin’s lymphoma. Cancer 2000; 89: 1387–1395.[CrossRef][ISI][Medline]

29. Tacyildiz N, Yavuz G, Gozdasoglu S et al. Serum levels and differential expression of intercellular adhesion molecule-1 in childhood leukemia and malignant lymphoma: prognostic importance and relationship with survival. Pediatr Hematol Oncol 1999; 16: 149–158.[CrossRef][ISI][Medline]

30. Conconi A, Zucca E, Roggero E et al. Prognostic models for diffuse large B-cell lymphoma. Hematol Oncol 2000; 18: 61–73.[CrossRef][ISI][Medline]

31. Mounier N, Morel P, Haioun C et al. A multivariate analysis of the survival of patients with aggressive lymphoma: variations in the predictive value of prognostic factors during the course of the disease. Groupe d’Etudes des Lymphomes de L’Adulte. Cancer 1998; 82: 1952–1962.[CrossRef][ISI][Medline]

32. Dumontet C, Mounier N, Munck JN et al. Factors predictive of early death in patients receiving high-dose CHOP (ACVB regimen) for aggressive non-Hodgkin’s lymphoma: a GELA study. Br J Haematol 2002; 118: 210–217.[CrossRef][ISI][Medline]

33. Hazlehurst LA, Damiano JS, Buyuksal I et al. Adhesion to fibronectin via beta 1 integrins regulates p27kip1 levels and contributes to cell adhesion mediated drug resistance (CAM-DR). Oncogene 2000; 19: 4319–4327.[CrossRef][ISI][Medline]

34. Lagneaux L, Delforge A, Bron D et al. Chronic lymphocytic leukemic B cells but not normal B cells are rescued from apoptosis by contact with normal bone marrow stromal cells. Blood 1998; 91: 2387–2396.[Abstract/Free Full Text]

35. De la Fuente MT, Casanova B, García-Gila M et al. Fibronectin interaction with alpha4beta1 integrin prevents apoptosis in B-cell chronic lymphocytic leukemia: correlation with Bcl-2 and Bax. Leukemia 1999; 13: 266–274.[CrossRef][ISI][Medline]

36. Fornarini B, D’ambrosio C, Natoli C et al. Adhesion to 90K (mac-2 BP) as a mechanism for lymphoma drug resistance in vivo. Blood 2000; 96: 3282–3285.[Abstract/Free Full Text]