|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2006.08.1331 on March 26 2007 © 2007 American Society of Clinical Oncology. Plasma Cytokine and Soluble Receptor Signature Predicts Outcome of Patients With Classical Hodgkin's Lymphoma: A Study From the Groupe d'Etude des Lymphomes de l'Adulte
From the Hôpital Le Bocage, Dijon; Hôpital St-Louis; Hôpital Pitié Salpétrière, Paris; Hôpital Henri Mondor, Créteil; Centre Hospitalier Universitaire (CHU) Lille, Lille; Centre Léon Bérard, Lyon; Hôpital Jean Minjoz, Besançon; CHU Nancy, Nancy; Centre Henri Becquerel, Rouen; Hospices Civils de Lyon and Université Claude Bernard, Lyon; Institut Gustave Roussy, Villejuif, France Address reprint requests to René-Olivier Casasnovas, MD, Service d'hématologie clinique, Hôpital Le Bocage, Centre Hospitalier Universitaire Dijon, Bd de Lattre de Tassigny, 21034 Dijon Cedex, France; e-mail: olivier.casasnovas{at}chu-dijon.fr
Purpose: Approximately 15% of patients with localized and 30% with disseminated classical Hodgkin's lymphoma fail to respond or relapse after first-line treatment. Usual prognosis scoring systems are actually unable to identify this small subset of patients with good confidence, pointing out the need for additional prognostic biomarkers. Patients and Methods: We prospectively analyzed the prognosis value of plasma levels of tumor necrosis factor (TNF), its soluble receptors TNF-R1 and TNF-R2, IL-10, IL1-RA, IL-6, and soluble CD30 (sCD30) when taken before any treatment in 519 consecutive patients with a first diagnosis of classical Hodgkin's lymphoma.
Results: Levels of TNF Conclusion: Plasma cytokine signature is sufficient to predict disease-related outcome in classical Hodgkin's lymphoma, and allows the identification of patients with very high risk of treatment failure.
More than 80% of patients with classical Hodgkin's lymphoma are virtually cured by conventional therapeutic strategies. Due in part to the overall low number of treatment failures, the prognostic models based on normal clinical and biological parameters are limited in their recognition of patients at high risk for treatment failure. Among Ann Arbor stages I and II which account for two thirds of patients with Hodgkin's lymphoma, three prognostic scoring systems using some similar clinical and biological parameters have been independently developed by three large cooperative groups—European Organisation for Research and Treatment of Cancer (EORTC),1 GHSG,2 and Canada–Eastern Cooperative Oncology Group (ECOG).3 These scoring systems are currently used to stratify the treatment of localized Hodgkin's lymphoma and are useful to define subgroups of patients with good or very good outcome who are suitable for a decrease of treatment intensity. Conversely, the prognostic subgroup identified with the worst outcome achieves a continuous complete remission in more than 80% of patients4 and represents only 2% of all patients with Hodgkin's lymphoma. In advanced diseases, in which the failure rate is significantly higher and reaches 30% of all patients, a seven-factored international prognostic score (IPS) was developed.5 An IPS score of 5 or higher identifies a poor prognosis subset of patients with a 47% probability of 5-year failure-free survival, which represents only 7% of this population. Recently, localized Hodgkin's lymphomas were also found to respond to a prognostic score derived from the IPS,4 but which failed to recognize a high risk subset. In this setting, there is evidence that usual clinical and biologic parameters are unable to identify with patients having both good sensitivity and specificity but with very high risk of failure, and that new prognosis factors are needed.
Among the biologic factors with a putative prognosis influence in Hodgkin's lymphoma, cytokines produced by both the tumor and surrounding reactive cells are thought to contribute to the pathogenesis of the disease6 and may be relevant candidates. Indeed, Reed-Sternberg cells are known to produce Th2 cytokines such as IL-137 or IL-6 acting as autocrine growth factors, or cytokines such as IL-10 that inhibit the T-cell–mediated response.6 Moreover, tumor cells commonly express tumor necrosis factor (TNF) receptor family members,6 such as p55 and p75 TNF
In this setting, the Groupe d'Etude des Lymphomes de l'Adulte (GELA) designed a prospective multicentric study to assess the prognosis value of plasma levels of TNF
Patients Five hundred nineteen consecutive patients with a first diagnosis of classical Hodgkin's lymphoma, excluding nodular lymphocyte predominance lymphoma, were prospectively enrolled from 17 institutions from October 1998 to June 2002. The patient median age was 33 years (range, 15 to 93 years). All patients had negative serology for HIV and signed an informed consent form approved by the ethics committee of the University Hospital of Dijon (Dijon, France) who reviewed the study. The diagnosis of classical Hodgkin's lymphoma was mainly based on lymph node histology. The pathologic features of 88% of patients were reviewed by two independent GELA pathologists and were classified according to the 2001 WHO classification of hematologic malignancies.10 Hodgkin's lymphoma was assumed unclassified when only a biopsy of extranodal tissue was available. The extent of disease and the presence of B symptoms were staged in accordance with Ann Arbor classification. Patient characteristics are listed in Table 1.
Treatment and Patient Outcome Patients were treated either in EORTC-GELA Hodgkin's lymphoma trials or according to GELA recommendations: patients with stage I and II disease received four to six cycles of an anthracycline-based chemotherapy regimen, followed by 20 to 36 Gy of involved-field radiotherapy; patients with stage III and IV disease received eight cycles of anthracycline-based chemotherapy. Response was assessed using Cheson criteria11 at the end of the first line of treatment, except for patients with progressive disease who were evaluated at time of progression. Three hundred twelve patients (60%) achieved a complete response, 151 patients (29%) achieved an unconfirmed complete response, and 20 patients (4%) achieved a partial response. Thirty-six patients had stable (n = 11; 2%) or progressive disease (n = 25; 5%). Forty patients (8%) relapsed, with a median time to relapse of 10.5 months (range, 1.6 to 56.0 months). Forty-one patients (8%) died: 25 from Hodgkin's lymphoma progression, six from secondary cancer, nine from cardiac or infectious treatment-related toxicities, and one patient from suicide. The median follow-up time for patients still alive was 49 months (range, 4 to 64 months).
Samples and Cytokine Plasma Levels Assessment The cytokine and soluble receptor plasma levels were also evaluated in a group of 32 healthy volunteers who submitted informed consent, including 17 men and 15 women with a median age of 32 years (range, 21 to 60 years).
Statistical Analysis Overall survival was calculated from the date of enrollment onto the study to the date of death as a result of any cause or the date of last follow-up. Event-free survival was measured from the date of enrollment to either the date of treatment failure (progression or relapse), the date of death as a result of any cause, or the date of last follow-up. Survival functions of patient subgroups defined by the cytokine or receptor circulating levels were estimated using the Kaplan-Meier product limit method and compared using the log-rank test. To construct a model for the prediction of event-free survival, a Cox proportional hazards regression model was set up, including the cytokine or receptor circulating levels as explanatory variables.13 A step by step backward elimination of the nonsignificant parameters was then performed until the best model was obtained. However, to set up a cytokine prognosis index that would be easy to use in routine practice, we decided to select only the three variables that produced the smallest loss of discriminating power. To choose the most accurate model, we tested all the candidate models with three variables and classified them according to their Akaike's information criterion.14 The final three variables' model had the better Akaike's information criterion and was the closest to those obtained for the four variables' model. To validate the selected model, we used the bootstrap method, which allows resampling with replacement to be generated from the original population.15 When backward selection was applied separately to each bootstrap resample (1,000 replications), it allowed us to estimate parameters and generate robust confidence intervals without basing our analysis on a single selected model. To compare the relative influence on the event-free survival of the related cytokine index with previously existing prognostic scoring systems, successive Cox proportional hazards regression models were set up that included the cytokine index and either the EORTC, Canada-ECOG, GHSG, or IPS scores as explanatory variables. Two-sided P values less than .05 were considered statistically significant. All statistical analysis was performed using SAS software version 9.13 (SAS Institute, Cary, NC).
Plasma Cytokine and Receptor Levels' Prognostic-Based Model Patients had significantly higher TNF  , TNF-R1, TNF-R2, IL1-RA, IL-10, IL-6, and sCD30 plasma levels than normal healthy participants (P < .001 for all parameters; Appendix Table A1, online-only). TNF, TNF-R1, TNF-R2, and IL1-RA were detectable in all patients' plasma samples, whereas IL-10, IL-6, and sCD30 were undetectable in 195 patients (38%), 31 patients (6%), and one patient (0.2%), respectively. Patients had similar cytokine or receptor levels regardless of the histologic subtype of Hodgkin's lymphoma. High levels of cytokine or receptor were related to advanced-stage disease, but not to the number of nodal sites involved. Except for IL-10, plasma levels of tested molecules were higher in poor-prognosis patients' subsets as defined by routinely used prognostic factors (Table 1) compared with patients who presented no pejorative factors.
To analyze the prognostic value of a plasma cytokine or soluble receptor, we determined a cutoff value for each parameter, based on its ability to predict failure to treatment with the best sensitivity and specificity (Appendix Table A2, online-only). The resulting threshold levels were 46 pg/mL for TNF
Cytokine Prognostic Index The relative risk of failure for the four significant parameters were similar, and the three most discriminating ones were selected to establish a simple prognostic index: sCD30, IL-6, and IL-RA (Table 3). A score from 0 to 3 was then calculated for each individual patient, assigning the 519 patients to four prognostic groups (Table 4 and Fig 1).
Multivariate Analysis of Prognosis Scoring Systems To assess the clinical value of the cytokine signature, we compared its prognosis powerfulness to previously validated scoring systems. First, we analyzed the prognostic value of the EORTC, GHSG, Canada-ECOG, and IPS scoring systems. The EORTC, GHSG, and Canada-ECOG scores, previously developed for limited-stage Hodgkin's lymphoma,1-3 were found to influence the probability of event-free survival of the 385 patients with Ann Arbor stages I or II disease (P < .02, P < .035, and P < .002, respectively), but also influenced the event-free survival of the entire population of 519 patients (P < .002). The IPS initially designed for advanced diseases5 had a significant impact on the duration of event-free survival for both the 154 patients with Ann Arbor stage III or IV disease (P < .001) and the whole population (P < .001). Then, the new index was tested using multivariate analysis against each classical scoring system. The cytokine signature remained the unique independent prognosis model for event-free survival when comparing the EORTC, GHSG, or Canada-ECOG scoring systems, as well in the whole population (P < .001 in each case) as in the group of patients with Ann Arbor stage I or II (P < .005 in each case). The regression analysis using IPS and cytokine signature as covariates showed that both IPS and the cytokine index were strong independent predictors for event-free survival in both the group of patients with advanced disease (IPS: P = .012; relative risk [RR] = 1.57; cytokine index, P = .0005; RR = 2.55) and in the whole population (IPS: P < .001; RR = 1.4; cytokine index: P < .001; RR = 1.8). To further assess the added value of the cytokine model, we applied the cytokine index to the cohort of 519 patients who were grouped according to the IPS (Fig 2). We identified patients with an especially high probability of early treatment failure in both the subset of patients with an IPS of 2 or lower and in the subset of patients with an IPS higher than 2.
We identified a plasma cytokine signature that is able to predict the outcome of patients with classical Hodgkin's lymphoma independently of previous prognosis scoring systems and that is able to recognize a subset of patients with a particularly high risk of failure, reaching 85% at 5 years. Some parameters constituting this cytokine signature were studied previously as single prognosis factors. While using plasma samples, we corroborated recurrent data that underlined the poorest outcome of patients exhibiting high sera levels of sCD30 at diagnosis.16-21 High sCD30 levels are clearly related to advanced-stage disease, suggesting that sCD30 levels could be related to the overexpression of tumor cell CD30 and, subsequently, to the tumor burden. Biologic properties of sCD30 could also influence patients' prognosis, since high amounts of monomeric sCD30 might link the CD30 ligand with high affinity and inhibit both T-cell activation22 and tumor cell death induced by CD30-CD30L interaction.23 However, sCD30 alone cannot explain the clinical outcome of numerous Hodgkin's lymphoma patients, since only 40% of patients with high sCD30 levels will relapse. The remaining TNF receptor family members previously found to influence patients outcome24 add less prognosis information to the cytokine signature than sCD30, suggesting that sCD30 best reflects the balance between the host immune system effectors and the tumor cells. Inversely, the biologic consequences of the overexpression of tumor cell membrane CD30 are probably poorly related to its circulating levels. Membrane CD30 overexpression strongly drives ligand-independent activation of NF-kB, and leads to constitutive cytokine secretion by tumor cells,25 suggesting that cytokine levels could be a better surrogate marker of the functional activity of NF-kB26 than sCD30; specifically IL-6, whose genetically driven long-term hypersecretion is a susceptibility factor to develop Hodgkin's lymphoma.27 Our results confirm a major prognostic role of IL-6,17,28 which is independent of classical prognosis factors, and of TNF receptor family members levels, including sCD30. The natural competitive inhibitor of IL-1, IL1-RA, which is also found at high levels in Hodgkin's lymphoma patients,29 is related to B symptoms and poor outcome, indicating that its anti-inflammatory properties do not prevent the occurrence of clinical inflammatory symptoms in Hodgkin's lymphoma. Interestingly, IL1-RA secretion is partly dependent on endogenous IL-6 secretion and isenhanced by IL-1330 and IL-10,31 and may constitute an indirect marker of these Th2 cytokines' production. Indeed, IL1-RA circulating level seems to be a better indicator of patient outcome than plasma IL-10 level,32-34 which did not retained prognosis significance in multivariate analysis. So, despite the complexity of the biologic events driving the relationships between the tumor cells and the panel of surrounding cells, the combination of selected soluble effectors and receptors released in the plasma is able to provide relevant prognostic insights into patients with Hodgkin's lymphoma. To date, only indirect markers of cytokine secretion, such as B symptoms, erythrocyte sedimentation rate, lymphocyte count, hemoglobin, or albumin levels were included in prognosis scoring systems. Here, we show that whatever the disease stage, the prognosis information provided by the plasma cytokine signature is more relevant than classical scores such as the EORTC, GHSG, and Canada-ECOG scoring systems, and adds significant prognosis information to the seven-factored IPS to determine patients with very poor event-free survival. So the plasma cytokine signature more accurately reflects the biologic properties of the tumor, and seems more powerful in identifying patients who would escape conventional therapy. Then, a cytokine signature of 0 identifies patients at low risk of failure reaching 5-year event-free survival at a probability higher than 90%. Conversely, a cytokine signature of 3 is related to high risk of chemotherapy-resistant Hodgkin's lymphoma (Table 5), with relapses occurring early with a median time of 14 months from diagnosis. This latter group represents only 4% of the whole population, but with 14% of patients displaying an IPS higher than 2, it constitutes a sizeable target population eligible for alternative therapies.
Finally, the cytokine signature evaluation on plasma samples at diagnosis seems to be widely feasible in a large hospital network. Enzyme-linked immunosorbent assay (ELISA) techniques are highly reproducible and may be developed as easy-to-use kits to determine this index in individual patients in routine practice. Furthermore, the introduction of multiplex techniques that allow several cytokine measurements from the same sample will greatly contribute to a better feasibility of such a profile approach. Overall, these results, which require further validation in an independent patient population, demonstrate that plasma cytokine signature could play a key role in the risk assessment of front-line treatment failure in order to propose a better risk-adapted therapeutic strategy in newly diagnosed patients with classical Hodgkin's lymphoma.
The authors indicated no potential conflicts of interest.
Conception and design: René-Olivier Casasnovas, Jean-Yves Blay, Jacques Bienvenu, Michel Guiguet, Liliane Intrator, Christophe Ferme, Gilles Salles Administrative support: Monique Grandjean Provision of study materials or patients: René-Olivier Casasnovas, Pauline Brice, Marine Divine, Franck Morschhauser, Jean Gabarre, Jean-Yves Blay, Laurent Voillat, Pierre Lederlin, Aspasia Stamatoullas, Jacques Bienvenu, Michel Guiguet, Liliane Intrator, Josette Brière, Christophe Ferme, Gilles Salles Collection and assembly of data: René-Olivier Casasnovas, Jacques Bienvenu, Michel Guiguet, Liliane Intrator, Monique Grandjean, Josette Brière Data analysis and interpretation: René-Olivier Casasnovas, Nicolas Mounier, Gilles Salles Manuscript writing: René-Olivier Casasnovas Final approval of manuscript: René-Olivier Casasnovas, Nicolas Mounier, Christophe Ferme, Gilles Salles
The following persons and institutions participated in this Groupe d'Etude des Lymphomes de l'Adulte (GELA) study. Biologic committee: J. Bienvenu, G. Salles, M. Guiguet, M.F. Frigère, R.O. Casasnovas, J.Y. Blay, L. Intrator, J.P. Farcet; pathologic review committee: J. Brière, J. Audouin; statistics: N. Mounier, N. Nio, K. Dordonne; GELA Clinical research: A. Salis; Centre Hospitalier Universitaire Dijon Clinical research: M. Grandjean; study centers: Hôpital Le Bocage, Dijon–R.O. Casasnovas, D. Caillot; Hôpital Saint-Louis, Paris–P. Brice, C. Gisselbrecht; Hôpital Henri Mondor, Créteil–M. Diviné; Centre Hospitalier de Brabois, Nancy–P. Lederlin; Centre Hospitalier universitaire de Lille, Lille–F. Morschhauser; Hôpital Pitié-Salpétrière, Paris–J. Gabarre; Centre Léon Bérard, Lyon–C. Sebban, J.Y. Blay; Centre Hospitalier Lyon Sud, Lyon–G. Salles, C. Thieblemont, B. Coiffier; Hôpital Jean Minjoz–L. Voillat, A. Brion, J. Vuillez; Centre Becquerel, Rouen–A. Stamatoullas, S. Lepretre, H. Tilly; Hôpital Necker, Paris–C. Belanger, O. Hermine; Centre Hospitalier de Chambéry, Chambéry–M. Blanc; Centre Hospitalier universitaire de Caen, Caen–O. Reman; Centre Hospitalier de Lens, Lens–P. Morel; Hôtel Dieu, Paris–A. Delmer; Institut Gustave Roussy, Villejuif–C. Fermé, P. Carde; Hôpital Inter Armées Percy, Clamart–G. Nedellec.
published online ahead of print at www.jco.org on March 26, 2007. Supported by Grant No. PHRC 1998 from the French government. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Carde P, Hagenbeek A, Hayat M, et al: Clinical staging versus laparotomy and combined modality with MOPP versus ABVD in early-stage Hodgkin's disease: The H6 twin randomized trials from the European Organization for Research and Treatment of Cancer Lymphoma Cooperative Group. J Clin Oncol 11:2258-2272, 1993 2. Duhmke E, Diehl V, Loeffler M, et al: Randomized trial with early-stage Hodgkin's disease testing 30 Gy vs. 40 Gy extended field radiotherapy alone. Int J Radiat Oncol Biol Phys 36:305-310, 1996[Medline] 3. Gospodarowicz MK, Sutcliffe SB, Bergsagel DE, et al: Radiation therapy in clinical stage I and II Hodgkin's disease: The Princess Margaret Hospital Lymphoma Group. Eur J Cancer 28A:1841-1846, 1992[CrossRef] 4. Gisselbrecht C, Mounier N, Andre M, et al: How to define intermediate stage in Hodgkin's lymphoma? Eur J Haematol Suppl 111-114, 2005 5. Hasenclever D, Diehl V: A prognostic score for advanced Hodgkin's disease: International Prognostic Factors Project on Advanced Hodgkin's Disease. N Engl J Med 339:1506-1514, 1998 6. Skinnider BF, Mak TW: The role of cytokines in classical Hodgkin lymphoma. Blood 99:4283-4297, 2002 7. Skinnider BF, Elia AJ, Gascoyne RD, et al: Interleukin 13 and interleukin 13 receptor are frequently expressed by Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma. Blood 97:250-255, 2001 8. Hooper NM, Karran EH, Turner AJ: Membrane protein secretases. Biochem J 321:265-279, 1997 (part 2)[Medline] 9. Hansen HP, Dietrich S, Kisseleva T, et al: CD30 shedding from Karpas 299 lymphoma cells is mediated by TNF-alpha-converting enzyme. J Immunol 165:6703-6709, 2000 10. Stein H, Delsol G, Pileri S: Classical Hodgkin lymphoma pathology and genetics of tumours of haematopoietic and lymphoid tissues. Lyon, France, IARC Press, 2001, pp 244-253 11. 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 17:1244, 1999 12. Zweig MH, Campbell G: Receiver-operating characteristic (ROC) plots: A fundamental evaluation tool in clinical medicine. Clin Chem 39:561-577, 1993 13. Cox D: Regression model and life tables. Stat Soc B 34:187-220, 1972 14. Akaike H: A new look at the statistical model identification. Automatic Control, IEEE Transactions 19:716-723, 1974[CrossRef] 15. Efron B: Bootstrap methods: Another look at the jackknife. Ann Stat 1979:1-26, 1979 16. Zanotti R, Trolese A, Ambrosetti A, et al: Serum levels of soluble CD30 improve International Prognostic Score in predicting the outcome of advanced Hodgkin's lymphoma. Ann Oncol 13:1908-1914, 2002 17. Vener C, Guffanti A, Pomati M, et al: Soluble cytokine levels correlate with the activity and clinical stage of Hodgkin's disease at diagnosis. Leuk Lymphoma 37:333-339, 2000[Medline] 18. Axdorph U, Sjoberg J, Grimfors G, et al: Biological markers may add to prediction of outcome achieved by the International Prognostic Score in Hodgkin's disease. Ann Oncol 11:1405-1411, 2000 19. Nadali G, Tavecchia L, Zanolin E, et al: Serum level of the soluble form of the CD30 molecule identifies patients with Hodgkin's disease at high risk of unfavorable outcome. Blood 91:3011-3016, 1998 20. Gause A, Pohl C, Tschiersch A, et al: Clinical significance of soluble CD30 antigen in the sera of patients with untreated Hodgkin's disease. Blood 77:1983-1988, 1991 21. Nadali G, Vinante F, Ambrosetti A, et al: Serum levels of soluble CD30 are elevated in the majority of untreated patients with Hodgkin's disease and correlate with clinical features and prognosis. J Clin Oncol 12:793-797, 1994[Abstract] 22. Su CC, Chiu HH, Chang CC, et al: CD30 is involved in inhibition of T-cell proliferation by Hodgkin's Reed-Sternberg cells. Cancer Res 64:2148-2152, 2004 23. Hargreaves PG, Al-Shamkhani A: Soluble CD30 binds to CD153 with high affinity and blocks transmembrane signaling by CD30. Eur J Immunol 32:163-173, 2002[CrossRef][Medline] 24. Warzocha K, Bienvenu J, Ribeiro P, et al: Plasma levels of tumour necrosis factor and its soluble receptors correlate with clinical features and outcome of Hodgkin's disease patients. Br J Cancer 77:2357-2362, 1998[Medline] 25. Horie R, Watanabe T, Morishita Y, et al: Ligand-independent signaling by overexpressed CD30 drives NF-kappaB activation in Hodgkin-Reed-Sternberg cells. Oncogene 21:2493-2503, 2002[CrossRef][Medline] 26. Hinz M, Lemke P, Anagnostopoulos I, et al: Nuclear factor kappaB-dependent gene expression profiling of Hodgkin's disease tumor cells, pathogenetic significance, and link to constitutive signal transducer and activator of transcription 5a activity. J Exp Med 196:605-617, 2002 27. Cozen W, Gill PS, Ingles SA, et al: IL-6 levels and genotype are associated with risk of young adult Hodgkin lymphoma. Blood 103:3216-3221, 2004 28. Seymour JF, Talpaz M, Hagemeister FB, et al: Clinical correlates of elevated serum levels of interleukin 6 in patients with untreated Hodgkin's disease. Am J Med 102:21-28, 1997[Medline] 29. Gruss HJ, Dolken G, Brach MA, et al: High concentrations of the interleukin-1 receptor antagonist in serum of patients with Hodgkin's disease. Lancet 340:968, 1992[Medline] 30. Gabay C, Porter B, Guenette D, et al: Interleukin-4 (IL-4) and IL-13 enhance the effect of IL-1beta on production of IL-1 receptor antagonist by human primary hepatocytes and hepatoma HepG2 cells: Differential effect on C-reactive protein production. Blood 93:1299-1307, 1999 31. Crepaldi L, Silveri L, Calzetti F, et al: Molecular basis of the synergistic production of IL-1 receptor antagonist by human neutrophils stimulated with IL-4 and IL-10. Int Immunol 14:1145-1153, 2002 32. Sarris AH, Kliche KO, Pethambaram P, et al: Interleukin-10 levels are often elevated in serum of adults with Hodgkin's disease and are associated with inferior failure-free survival. Ann Oncol 10:433-440, 1999 33. Vassilakopoulos TP, Nadali G, Angelopoulou MK, et al: Serum interleukin-10 levels are an independent prognostic factor for patients with Hodgkin's lymphoma. Haematologica 86:274-281, 2001 34. Visco C, Vassilakopoulos TP, Kliche KO, et al: Elevated serum levels of IL-10 are associated with inferior progression-free survival in patients with Hodgkin's disease treated with radiotherapy. Leuk Lymphoma 45:2085-2092, 2004[CrossRef][Medline] Submitted July 3, 2006; accepted February 6, 2007. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 and (B) patients with IPS more than 2, after subdivision into three risk groups on the basis of the cytokine signature (score of 0, 1 to 2, and 3). According to maximum likelihood estimates: P < .001 for both patient subgroups with IPS 
