Business & Finance

Phosphorylated STAT5 Represents a New Possible Prognostic Marker in Hodgkin Lymphoma

Description
An important pathogenetic mechanism in Hodgkin lymphoma (HL) is the interaction between the neoplastic and reactive cells mediated by a complex network of cytokines with activation of cytokine signal transduction (STAT) pathways. We studied the
Published
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  472   Am J Clin Pathol   2008;129:472-477 472   DOI: 10.1309/63H1A83HRTBQ07DB  © American Society for Clinical Pathology Hematopathology    / P STAT5 IN  H ODGKIN  L   YMPHOMA  Phosphorylated STAT5 Represents a New Possible Prognostic Marker in Hodgkin Lymphoma Maurizio Martini, MD, PhD, 1  Stefan Hohaus, MD, PhD, 2  Giovanna Petrucci, PhD, 1  Tonia Cenci, PhD, 1  Francesco Pierconti, MD, PhD, 1  Giuseppina Massini, MD, 2  Luciana Teofili, MD, PhD, 2  Giuseppe Leone, MD, 2  and Luigi M. Larocca, MD 1 Key Words: STAT5; Hodgkin lymphoma; Prognosis DOI: 10.1309/63H1A83HRTBQ07DB A b s t r a c t   An important pathogenetic mechanism in Hodgkin lymphoma (HL) is the interaction between the neoplastic and reactive cells mediated by a complex network of cytokines with activation of cytokine signal transduction (STAT) pathways. We studied the  prognostic impact of the phosphorylation status of STAT5 in HL. By using immunohistochemical analysis, we found  phosphorylated STAT5 (pSTAT5) in 35 (38%) of 93 lymph node biopsy specimens of patients with HL. The detection of pSTAT5 in Hodgkin and Reed-Sternberg (HRS) cells in classical HL (cHL) was not associated with any clinical and biologic features evaluated, including Epstein-Barr virus status. The primary end  point for analysis of clinical outcome was freedom from treatment failure (FFTF). At a median follow-up of 5  years, pSTAT5+ patients with cHL had a better FFTF than pSTAT5– patients (77% vs 56%; P = .03), which translated into a reduced risk for failure for pSTAT5+  patients with a hazard ratio of 0.2 (95% confidence interval, 0.06-0.73; P = .015).Our data suggest that the phosphorylation status of STAT5 of HRS cells in cHL could be a prognostic marker in HL. Hodgkin lymphoma (HL) is a lymphoid malignancy characterized by the presence of rare neoplastic cells (Hodgkin and Reed-Sternberg [HRS] cells in classical HL [cHL] and the lymphocytic and histiocytic [L&H] cells in nodular lymphocyte-predominant HL [NLPHL]) surrounded by a mixed inflammatory infiltrate composed of lympho-cytes, eosinophils, macrophages, plasma cells, and fibro-blasts. The interactions between neoplastic and reactive cells are mediated by a complex network of cytokines that have a key role in determining the clinical and pathologic features of HL. The effects of cytokine production are bidirectional: cytokines produced by HRS cells not only act as autocrine growth factors but also induce the formation of the reactive cell infiltrate. In turn, cytokines produced by the surround-ing reactive cells sustain the proliferation and survival of the HRS cells themselves. 1,2 The binding of cytokines to the respective receptors on the cellular surface activates several intracellular signaling pathways such as the phosphoinositide-3 kinase and MEK/ ERK pathways, Notch1, and signal transducers and activators of transcription (STATs). 3-7  It has been demonstrated that the constitutive activation of these pathways promotes survival and proliferation of HRS cells. 7,8 Many cytokine receptors are associated with cytoplasmic protein kinases belonging to the Janus protein tyrosine kinase (JAK) family: signal transduction results from the phosphoryla-tion of the receptor and associated JAK, which, in turn, recruits and phosphorylates substrate molecules, including STATs. Only in this “activated” form do the phosphorylated STATs (pSTATs) dimerize and translocate to the nucleus where they bind to specific regulatory sequences of target genes. 9,10  Activated STATs can be detected by immunohistochemical  Am J Clin Pathol   2008;129:472-477 473473   DOI: 10.1309/63H1A83HRTBQ07DB   473 © American Society for Clinical Pathology Hematopathology    / O RIGINAL   A  RTICLE analysis using antibodies recognizing specifically the tyrosine-phosphorylated form of a given STAT protein.Several phosphorylated members of the STAT family can be detected in HRS cells. STAT3 and STAT6 are com-monly activated in HL, being present in about 75% to 90% of HL cell lines and primary HRS cells, 11,12  whereas pSTAT5 is expressed in 25% to 30% of cHL cases and is absent in the L&H cells from NLPHL samples. 12,13  In particular, several cytok-ines produced by HL cell lines or expressed in HL-involved tissues, including interleukin (IL)-5, IL-13, IL-7, and IL-9, induce or could determine STAT5 phosphorylation. 14-17  It is interesting that pSTAT5 up-regulates BCL-xL and cyclin D1 in normal and tumor cells, indeed promoting their survival and proliferation. 18-21  In addition, recent studies revealed that Epstein-Barr virus (EBV) infection may increase the phosphorylated forms of several STATs, including STAT5, directly by latent membrane protein (LMP)-1 and indirectly by inducing the expression of IL-6. 22,23 Considering that a significant association between phos-phorylation of STATs and prognosis has been demonstrated in breast and prostate cancer, 24,25  we studied the phosphoryla-tion status of STAT5 in lymph nodes of patients with HL by using immunostaining. Results obtained were correlated with various clinical and biologic findings, including EBV status. Materials and Methods Study Population Characteristics The phosphorylation status of STAT5 was evaluated in HL+ lymph node biopsy specimens obtained at the Department of Pathology, Catholic University, Rome, Italy, between 1980 and 2005. All samples were obtained at diagno-sis. The study population consisted of 93 patients (38 females and 55 males; median age, 32 years, range, 14-71 years); 79 cases were diagnosed after 1993. Clinical findings are shown in z Table 1 z . Information on the type of chemotherapy regi-men was available for 90 patients. Chemotherapy consisted of doxorubicin (Adriamycin), bleomycin, vinblastine, and dacarbazine in 59 patients; 8 patients were treated with mech-lorethamine, vincristine (Oncovin), procarbazine, and pred-nisone–containing regimens. 26  Patients with advanced-stage disease (stage IIB with bulky disease through stage IV) and younger than 60 years were treated between 1999 and 2001 with a modified Stanford V regimen (cyclophosphamide, vincristine [Oncovin], vinblastine, doxorubicin [Adriamycin], bleomycin, etoposide, and prednisone; 15 patients) and from October 2001 with bleomycin, etoposide, doxorubicin, cyclo-phosphamide, vincristine, procarbazine, and prednisone (8 patients). 26  Radiotherapy was included for consolidation in patients with limited-stage disease and initial bulky disease. Informed consent was obtained from all patients, according to institutional guidelines. Immunohistochemical Analysis For immunophenotypic studies and lineage assignment of HRS, the avidin-biotin-peroxidase complex method was performed on paraffin sections using a commercially avail-able kit (DAKO LSAB 2, Dakopatts, Glostrup, Denmark) and the following commercially available monoclonal antibodies: CD3, CD15, CD20, CD30, CD45, PAX-5 and bcl-6. The CD30+, CD15+, and CD45– diagnostic profile was required for the diagnosis of classical HL.Immunostaining for pSTAT5 was performed on 3- to 4-µm-thick sections from paraffin-embedded tissue of 93 HL samples using a primary rabbit monoclonal antibody directed against phospho-STAT5 (Y694) (Epitomics, Burlingame, CA), as previously described. 27  Paraffin sections were depar-affinized in xylene, hydrated in a series of graded alcohols, z Table 1 z Characteristics of 93 Patients With Hodgkin Lymphoma *  pSTAT5+ pSTAT5– Variable   Total   (n = 35) †   (n = 53) † Age (y)   <45 68 (73) 24 (69) 40 (75) >45 25 (27) 11 (31) 13 (25)Sex Female 38 (41) 15 (43) 23 (43) Male 55 (59) 20 (57) 30 (57)Histologic type NLP   5 (5) — — Nodular sclerosis 71 (76) 26 (74) 45 (85) Mixed cellularity 9 (10) 3 (9) 6 (11) Lymphocyte-depleted 3 (3) 3 (9) 0 (0) Not specified 5 (5) 3 (9) 2 (4)EBV (n = 81) Negative   58 (72) 23/30 (77) 32/46 (70) Positive 23 (28) 7/30 (23) 14/46 (30)Stage I-IIA   37 (40) 13 (37) 21 (40) IIB-IV   56 (60) 22 (63) 32 (60)B symptoms (n = 90)   No   47 (52) 18/34 (53) 25/52 (48) Yes   43 (48) 16/34 (47) 27/52 (52)IPS (n = 84)   <2   66 (79) 22/31 (71) 42/50 (84)  ≥ 2   18 (21) 9/31 (29) 8/50 (16)Therapy (n = 90)   ABVD   59 (66)   21 (60)   35/51 (69) MOPP-containing   8 (9)   5 (14)   2/51 (4) Stanford V   15 (17)   5 (14)   10/51 (20) BEACOPP   8 (9)   4 (11)   4/51 (8) ABVD, doxorubicin (Adriamycin), bleomycin, vinblastine, and dacarbazine; BEACOPP, bleomycin, etoposide, doxorubicin, cyclophosphamide,   vincristine (Oncovin), procarbazine, and prednisone; EBV, Epstein-Barr virus; IPS, International Prognostic Score; MOPP, mechlorethamine, vincristine (Oncovin), procarbazine, and prednisone; NLP, nodular lymphocyte predominant; Stanford V, cyclophosphamide, vincristine (Oncovin), vinblastine, doxorubicin (Adriamycin), bleomycin, etoposide, and prednisone. *  Data are given as number (percentage) or number/total (percentage). Numbers of cases differing from 93 or the column totals for pSTAT5 are given. †  Data refer to classical Hodgkin lymphoma (n = 88).  474   Am J Clin Pathol   2008;129:472-477 474   DOI: 10.1309/63H1A83HRTBQ07DB  © American Society for Clinical Pathology Martini et al    / P STAT5 IN  H ODGKIN  L   YMPHOMA  and subjected to heat-induced antigen retrieval by microwave treatment in 0.01 mol/L of citrate buffer, pH 6. Tissue sections were treated with freshly made hydrogen peroxide for endog-enous peroxidase inhibition and then incubated 1 hour with diluted (1:50) primary antibody. After incubation with bio-tinylated antirabbit secondary antibody (ScyTek, Logan, UT), immunodetection was performed using an avidin-biotin-per-oxidase complex solution (ScyTek), 3,3 9 -diaminobenzidine as the chromogen, and Mayer hematoxylin as the counterstain.The specificity of the pSTAT5 antibody has been dem-onstrated on paraffin-embedded breast carcinoma tissue and bone marrow biopsy specimens of patients with myeloprolif-erative diseases, 27  in which the phosphorylation state changes observed in immunohistochemical studies were confirmed on Western blots. A further antibody specificity control was assessed on consecutive sections of 2 pSTAT5+ HL cases (HRS and reactive lymphocytes) that show a complete loss of immunoreactivity after pretreatment with alkaline phosphatase. 28  Moreover, we exposed paraffin sections to heat-induced antigen retrieval by microwave to improve phosphoepitopes and to avoid false-negative reactions for tyrosine phosphorylated proteins in the immunohistochemical analysis. 28  We included only cases with pSTAT5+ results in the surrounding small lymphocytes, which served as internal positive control samples z Figure 1 z . Cases were evaluated independently by 2 of us (M.M. and F.P.). In Situ Hybridization for EBV EBV infection was evaluated in 81 of 93 HL patients by in situ hybridization of EBV-encoded small RNAs (EBERs) on formalin-fixed, paraffin-embedded tissue sections. In situ hybridization analysis was performed using a cocktail of fluo-rescein isothiocyanate–labeled oligonucleotides complemen-tary to the nuclear EBERs (1/2), following the manufacturer’s instructions (DAKO; Dakopatts, Glostrup, Denmark) and as previously described. 29 Statistical Analysis The Fisher exact test was used to examine the associations of STAT5 phosphorylation status with patient characteristics. The primary end point of survival analysis was freedom from treatment failure, with progression during treatment, lack of complete remission at the end of first-line treatment, relapse, and death of any cause counted as adverse events. Survival curves were estimated by using the Kaplan-Meier product limit meth-od. The Kaplan-Meier plot presented is an unadjusted survival curve. The log-rank test was used for analysis of differences in freedom from treatment failure and the Cox proportional haz-ards model for relative risk of failure. Because patients had been treated with different regimens, the analysis was adjusted for the type of chemotherapy regimen. Computations were performed using Stata 6.0 software (Stata, College Station, TX). Results Phosphorylation of STAT5 in HL The phosphorylation status of STAT5 was assessed by immunohistochemical analysis in 93 lymph node biopsy specimens from patients with HL. Positive staining of reac-tive lymphocytes in the inflammatory infiltrate surrounding the HRS cells served as an internal positive control result. Staining of small lymphocytes resulted in a variable percent-age between 40% and 80%. Of 93 evaluated lymph nodes, 35 (38%) showed pSTAT5 in HRS cells z Image 1 z  (Table 1). In the positive cases, more than 90% of the HRS cells stained positively for pSTAT5. pSTAT5 was not detected in the L&H cells of 5 patients with the NLPHL variant, whereas no differ-ences were observed in pSTAT5 detection among the various histologic subtypes of classical HL. Association Between STAT5 Phosphorylation and EBV Status EBV status was determined in 81 cases (Table 1). HRS cells and reactive inflammatory lymphocytes were positive for EBER in 23 cases (28%) and 20 cases (25%), respectively. All samples with EBER+ staining in reactive lymphocytes also showed EBER+ staining in HRS cells. The presence of EBV and STAT5 phosphorylation in HRS cells was not significantly associated ( P = .6). Nevertheless, by restricting the analysis to the group of patients with the nodular sclerosis histologic type, cases with EBV– HL more frequently expressed pSTAT5 (20/45 [44%]) than did EBV+ cases (3/16 [19%]), although the difference was not statistically significant ( P = .08). 0.000.250.500.751.002    P  r  o   b  a   b   i   l   i   t  y  o   f   F   F   T   F Time (y)pSTAT5+pSTAT5 –4 6 8 10 12 14 16 18 20 22 z Figure 1 z  Unadjusted Kaplan-Meier plot for freedom from treatment failure (FFTF) in 86 patients with classical Hodgkin lymphoma according to the expression of phosphorylated STAT5 (pSTAT5).  Am J Clin Pathol   2008;129:472-477 475475   DOI: 10.1309/63H1A83HRTBQ07DB   475 © American Society for Clinical Pathology Hematopathology    / O RIGINAL   A  RTICLE Association Between STAT5 Phosphorylation Status and Clinical Characteristics in HL STAT5 phosphorylation was found only in cHL (Table 1). The association between pSTAT5 and clinical findings and outcome was analyzed only in this group of 88 patients. No associations between pSTAT5 and clinical characteris-tics such as age, sex, disease stage, presence of B symptoms, or International Prognostic Score (Hasenclever score) were detected. Prognostic Impact of STAT5 Phosphorylation Status Follow-up data were available for 86 patients with cHL; 2 patients were lost to follow-up. A total of 30 events were observed. The probability of freedom from treatment failure at a median follow-up of 5 years was 67% (95% confidence interval [CI], 56%-76%). The adjusted probability of 5-year-failure-free survival for patients with pSTAT5 in the HRS cells was 77%, whereas it was 56% for pSTAT5– patients ( P = .03; log-rank test; Figure 1). After adjusting for age, sex, ABCD z Image 1 z  Expression of phosphorylated STAT5 (pSTAT5). A , B , and D , Classical nodular sclerosis Hodgkin lymphoma (HL; A , H&E, ×250) that shows pSTAT5 nuclear staining in Hodgkin–Reed Sternberg (HRS) cells and reactive lymphocytes ( B , avidin-biotin-peroxidase complex method lightly counterstained with hematoxylin, ×250; inset, ×400). D , pSTAT5 nuclear staining in HRS cells and reactive lymphocytes was lost after section pretreatment with alkaline phosphatase (avidin-biotin-peroxidase complex method lightly counterstained with hematoxylin, ×250). C , Classical HL with pSTAT5– HRS cells and several pSTAT5+ reactive lymphocytes (avidin-biotin-peroxidase complex method lightly counterstained with hematoxylin, ×250).  476   Am J Clin Pathol   2008;129:472-477 476   DOI: 10.1309/63H1A83HRTBQ07DB  © American Society for Clinical Pathology Martini et al    / P STAT5 IN  H ODGKIN  L   YMPHOMA  stage, and type of chemotherapy, the hazard ratio for failure in pSTAT5+ patients was 0.2 (95% CI, 0.06-0.73; P = .015). Discussion Activation of STAT proteins is a common feature in HL. Recent studies show that STAT3 and STAT6 are frequently phosphorylated in classic HL and in primary HRS cells, whereas STAT5 phosphorylation occurs more rarely. 11,13  Our study demonstrates that pSTAT5 is expressed in 38% (35/93) of cHL cases. Moreover, our observations confirm previous data showing that STAT5 is not activated in the NLPHL. 13 Previous studies suggested an association between JAK-STAT signaling and EBV. STAT3, STAT5, and STAT1 can up-regulate LMP-1 expression by binding to specific functional sites in the LMP-1 promoter. Conversely, LMP-1 may activate STAT3 and STAT5 by induction of expression of nuclear factor κ B and IL-6. 22,23  However, the association between pSTAT5 and EBV was observed only in nasopha-ryngeal carcinoma and in epithelial cells latently infected by EBV, but not in EBV+ malignant cells of HL. 22  In line with this report, we found no association between EBV and pSTAT5 in cHL. Therefore, in cHL, STAT5 activation may be a consequence of other molecular alterations.Recently, Weniger et al 30  described mutations in the gene coding for the suppressor of cytokine signaling (SOCS)-1 in about 40% of laser-microdissected HRS cells of cHL and in 3 of 5 classic HL cell lines. In addition, these mutations were significantly associated with accumulation of nuclear pSTAT5 in HRS cells of tumor tissues. 30  Hypermethylation or muta-tions resulting in reduced, absent, or nonfunctional SOCS-1 proteins might account for high levels of pSTAT5. 30 The major finding of our study is that patients with activated STAT5 had a better outcome than patients without pSTAT5. To our knowledge, this is the first report on an asso-ciation between pSTAT5 and outcome in HL. This finding is limited by the retrospective nature of the study, resulting in heterogeneity of treatment. We therefore adjusted the survival analysis for the type of chemotherapy and stratified for patient characteristics. Larger, prospective studies will be needed to confirm our findings.The biologic mechanisms that explain the favorable effect of pSTAT5 on clinical outcome remain to be deter-mined. STAT5 activation was not associated with any patient characteristic having a prognostic value. Because pSTAT5 is supposed to be involved in oncogenesis, 18  the more favorable prognosis of patients with activated STAT5 seems, at first, surprising. However, our findings agree with data obtained from patients with breast cancer in whom pSTAT5 identi-fied patients with favorable prognosis. 24  Actually, STAT5 promotes differentiation in many cell systems, including hematopoietic and mammary cells. 31,32  Moreover, STAT5 may suppress dedifferentiation of breast epithelial cells, which is associated with breast cancer invasion and metasta-sis. 24  A similar mechanism may be involved in HL. Recently, STAT5 was demonstrated as a negative regulator of bcl-6 function, directly repressing the expression of this protein in B-lymphoma cells and other hematopoietic cell lines. This results in a loss of repression of bcl-6 target genes, thereby allowing B cells to differentiate. 33 Recent studies described STATs as a new target in the therapeutic strategies against tumors. Holtick et al 34  and Desrivieres et al 35  demonstrated that inhibition of STAT3 phosphorylation using the kinase inhibitor tyrphostin increased the sensitivity of classical HL cell lines to drug-induced apop-tosis. In addition, our study suggests that pSTAT5 induction could be useful in HL. The study of STAT phosphorylation status in HL may identify patients who might be candidates for these types of targeted therapy in the future. From the Departments of 1 Pathology and 2  Hematology, Catholic University of Rome, Rome, Italy.Supported in part by Fondi d’Ateneo, Progetti D1 2006, Università Cattolica Sacro Cuore, Roma, Italy. Address reprint requests to Dr Larocca: Dept of Pathology, Catholic University, Largo A. Gemelli No. 8, 00168, Roma, Italy. Acknowledgments: We thank Giuseppe Luongo and Pierpaolo Occhilupo for technical support. References  1. Skinnider BF, Mak TW. The role of cytokines in classical Hodgkin lymphoma. Blood. 2002;99:4283-4297. 2. Khan G. Epstein-Barr virus, cytokines, and inflammation: a cocktail for the pathogenesis of Hodgkin’s lymphoma? Exp Hematol. 2006;34:399-406. 3. Dutton A, Reynolds GM, Dawson CW, et al. Constitutive activation of phosphatidyl-inositide 3 kinase contributes to the survival of Hodgkin’s lymphoma cells through a mechanism involving Akt kinase and mTOR.  J Pathol. 2005;205:498-506. 4. Zheng B, Fiumara P, Li YV, et al. MEK/ERK pathway is aberrantly active in Hodgkin’s disease: a signaling pathway shared by CD30, CD40, and RANK that regulates cell proliferation and survival. Blood. 2003;102:1019-1027. 5. Jundt F, Anagnostopoulos I, Forster R, et al. Activated Notch1 signaling promotes tumor cell proliferation and survival in Hodgkin’s and anaplastic large cell lymphoma. Blood. 2002;99:3398-3403. 6. Renné C, Willenbrock K, Küppers R, et al. Autocrine- and paracrine-activated receptor tyrosine kinases in classic Hodgkin’s lymphoma. Blood. 2005;105:4051-4059. 7. Brauninger A, Schmitz R, Bechtel D, et al. Molecular biology of Hodgkin’s and Reed/Sternberg cells in Hodgkin’s lymphoma. Int J Cancer. 2006;118:1853-1861. 8. Re D, Kuppers R, Diehl V. Molecular pathogenesis of Hodgkin’s lymphoma.  J Clin Oncol. 2005;23:6379-6386. 9. Rawlings JS, Rosler KM, Harrison DA. The JAK/STAT signaling pathway.  J Cell Sci. 2004;117:1281-1283.
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks