Histopathology 2014, 65, 119–131. DOI: 10.1111/his.12381
Focal adhesion protein expression in human diffuse large B-cell lymphoma Rosa Bosch,1 Rebeca Dieguez-Gonzalez,1 Marıa Jose Moreno,1 Alberto Gallardo,2 Silvana Novelli,3,4 I~ nigo Espinosa,4,5 Marıa Virtudes Cespedes,1,6 Miguel Angel Pav on,1 Javier Briones,3,4 Albert Gra~ nena,7,† Jorge Sierra,3,4 Ramon Mangues1,6 & Isolda Casanova1,6 1
Grup d’Oncogenesi i Antitumorals, Institut d’Investigacions Biomediques Sant Pau, Barcelona, Spain, 2Department of Pathology, Clınica Girona, Girona, Spain, 3Department of Haematology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 4Red Tematica de Investigacion Cooperativa en Cancer (RETICC), Barcelona, Spain, 5Department of Pathology, Hospital de la Sant Creu i Sant Pau, Barcelona, Spain, 6CIBER en Bioingenierıa, Biomateriales y Nanomecidicina (CIBER-BBN), Barcelona, Spain, and 7Department of Haematology, Hospital Universitari de Bellvitge, Barcelona, Spain
Date of submission 5 November 2013 Accepted for publication 23 January 2014 Published online Article Accepted 28 January 2014
Briones Bosch R, Dieguez-Gonzalez R, Moreno M J, Gallardo A, Novelli S, Espinosa I, Cespedes M V, Pav on M A, J, Gra~ nena A, Sierra J, Mangues R, Casanova I (2014) Histopathology 65, 119–131
Focal adhesion protein expression in human diffuse large B-cell lymphoma Aims: Focal adhesions have been associated with poor prognosis in multiple cancer types, but their prognostic value in diffuse large B-cell lymphoma (DLBCL) has not been evaluated. The aim of this study was to investigate the expression patterns and the prognostic value of the focal adhesion proteins FAK, Pyk2, p130Cas and HEF1 in DLBCL. Methods and results: Focal adhesion protein expression was examined using immunohistochemistry in normal lymphoid tissues and in 60 DLBCL patient samples. Kaplan–Meier survival and Cox regression analysis were performed to evaluate the correlation of focal adhesion protein expression with patient
prognosis. FAK, Pyk2, p130Cas and HEF1 expression was mostly found in the germinal centres of normal human lymphoid tissues. When assessed in DLBCL samples, FAK, Pyk2, p130Cas and HEF1 were highly expressed in 45%, 34%, 42% and 45% of the samples, respectively. Multivariate Cox analysis revealed that decreased FAK expression was a significant independent predictor of poorer disease outcome. Conclusions: FAK expression is an independent prognostic factor in DLBCL. Our results suggest that the addition of FAK immunostaining to the current immunohistochemical algorithms may facilitate risk stratification of DLBCL patients.
Keywords: diffuse, focal adhesions, large B-cell lymphoma, prognosis
Introduction Focal adhesion (FA) complexes are structures that link the actin cytoskeleton with the extracellular
Address for correspondence: R Mangues, Grup d’Oncogenesi i Antitumorals (GOA), Institut d’Investigacions Biomediques Sant Pau, Sant Antoni Maria Claret 167, Pavell o 11, 2n pis, Barcelona 08028, Spain. e-mail: [email protected] †Deceased. © 2014 John Wiley & Sons Ltd.
matrix. They are formed by transmembrane receptors (integrins), structural proteins (i.e. vinculin, paxillin, and talin), and signalling proteins, including focal adhesion kinase (FAK) and Crk-associated substrate (Cas) family proteins. FAK is a cytoplasmic tyrosine kinase that plays a major role in mediating signal transduction by integrins and growth factor receptors1 in the regulation of cell adhesion, migration, survival, proliferation and differentiation for a variety of cells.1–6 In cancer studies,
120
R Bosch et al.
the clinical implications of FAK have been controversial. In general, FAK overexpression is associated with aggressiveness, migration, invasiveness, metastasis and poor prognosis in many carcinomas.7,8 However, recent studies have shown that weak FAK expression is associated with poor prognosis in cervical cancer,9 extrahepatic bile duct carcinoma,10 and intrahepatic cholangiocarcinoma.11 Pyk2 (also known as RAFTK) is the second member of the FAK cytoplasmic tyrosine kinases.12 It has been shown to interact with proteins that also bind to FAK, such as Src, p130Cas, HEF1, and paxillin.13–16 Pyk2 is mostly expressed in haematopoietic cells, osteoclasts, and the central nervous system,17 and its expression is associated with metastasis and poor outcome in diverse human tumours.18–21 p130Cas, the first member of the Cas family to be identified, is a docking protein involved in multiple intracellular signalling pathways.22,23 Its overexpression contributes to the development of many human cancers, and it is necessary for the transformation initiated by different oncogenes in different cancer types.24–26 In breast cancer, its overexpression is associated with proliferation, metastasis, and poor prognosis.27,28 HEF1 is a Cas family member that is mostly expressed in lymphocytes and epithelial cells,29 and shares many structural and functional characteristics with p130Cas.30 The gene encoding HEF1 has been identified as promoting metastasis in many tumours.31–34 Moreover, all of these FA proteins are involved in normal B-cell development,35–37 suggesting that they can play a role in B-cell transformation. However, their role in haematological malignancies is poorly understood, and their involvement in diffuse large B-cell lymphoma (DLBCL), the most frequent non-Hodgkin lymphoma subtype, has not been assessed before. Our aim in this study was to characterize the expression of the focal adhesion proteins FAK, Pyk2, p130Cas and HEF1 in normal lymphoid tissues and in 60 DLBCL patient samples. We also wanted to correlate our findings with clinicopathological variables and patient survival.
Materials and methods PATIENTS AND TISSUE SAMPLES
We analysed 60 patients with primary nodal DLBCL, diagnosed in our institution according to the criteria of the World Health Organization classification. Availability of adequate histological material and of clinical data were the only criteria for selection of patients. Patients with a follicular lymphoma (including grade 3b) or another type of indolent lymphoma with subsequent transformation into large-cell lymphoma were not included. In addition, patients with primary
central nervous system, mediastinal or HIV-related lymphomas were excluded. All patients received anthracycline-containing chemotherapy. Forty-six (77%) patients were treated with cyclophosphamide, doxorubicin, vincristine, and prednisone, and the remaining 14 received other combination chemotherapies. None of the patients received rituximab, because they were treated before its addition to standard chemotherapy regimes. The clinical features of the patients are shown in Table 1. Approval was obtained from the research committee of Santa Creu i Sant Pau Hospital. Written informed consent was provided according to the declaration of Helsinki. All tissue samples were reviewed to confirm the diagnosis of DLBCL. Two tissue arrays containing the samples were generated with a tissue array device (Beecher Instruments, Silver Spring, MD, USA). Two cylinders (1 mm in largest diameter) from the same tumour were selected. Control tissues (normal lymph node, breast cancer, and ovarian cancer) were also included in the tissue arrays. Additional samples of whole sections of normal lymphoid tissues (lymph nodes, spleen, thymus, and Peyer’s patches) were also evaluated for the expression of FA proteins. IMMUNOHISTOCHEMISTRY
Tissue microarray blocks were sectioned at a thickness of 3 lm, and dried for 16 h at 56°C. They were then dewaxed in xylene, rehydrated through a graded ethanol series, and washed with phosphate-buffered saline. Heat-induced epitope retrieval was performed by immersing the sections in sodium citrate buffer (pH 6.0) and incubating them at 97°C for 20 min using a Dako PTLink (Dako, Carpinteria, CA, USA). The sections were stained using anti-FAK (1:75; Millipore Billerica, MA, USA), anti-Pyk2 (1:100; BD Biosciences, Franklin Lakes, NJ, USA), anti-p130Cas (1:75; Neomarkers, Fremont, CA, USA), and anti-HEF1 (1:200; ABR-Affinity BioReagents, Golden, CO, USA) antibodies. The expression of CD20, CD10, Ki67 and bcl-2 (pre-diluted; Dako) was also assessed to classify the samples into DLBCL subtypes. The reactions were performed in a Dako Autostainer Link48 (Dako), and visualized using the biotin-free EndVision system (Dako) with diaminobenzidine. CD20, CD10, Ki67 and bcl-2 expression was considered to be positive when ≥30% of the tumour cells were reactive, according to other studies.38 We estimated the percentage of positively stained cells (0–100%) for the whole tumour and the intensity of the staining (absent, 0; weak, 1; moderate, 2; and strong, 3) for FAK, Pyk2, p130Cas and HEF1 in DLBCL samples. An expression value for © 2014 John Wiley & Sons Ltd, Histopathology, 65, 119–131.
16 (27)
21 (35)
≥60 (n = 32)
Gender Male (n = 40)
© 2014 John Wiley & Sons Ltd, Histopathology, 65, 119–131.
14 (24)
17 (30)
14 (27)
20 (36)
10 (18)
Yes (n = 19)
B symptoms No (n = 32)
Yes (n = 11)
ECOG status 0–1 (n = 43)
2–4 (n = 12)
9 (21)
High (n = 11)
18 (31)
15 (35)
Serum B2M Normal (n = 32)
Bulky mass No (n = 39)
17 (31)
High (n = 27)
7 (12)
13 (24)
(n = 11)
Serum LDH Normal (n = 28)
Positive
24 (42)
15 (26)
III–IV (n = 27)
Bone marrow Negative (n = 46)
16 (28)
Ann Arbor stage I–II (n = 30)
12 (20)
17 (28)
Age (years)
Focal adhesion protein expression in human diffuse large B-cell lymphoma Rosa Bosch,1 Rebeca Dieguez-Gonzalez,1 Marıa Jose Moreno,1 Alberto Gallardo,2 Silvana Novelli,3,4 I~ nigo Espinosa,4,5 Marıa Virtudes Cespedes,1,6 Miguel Angel Pav on,1 Javier Briones,3,4 Albert Gra~ nena,7,† Jorge Sierra,3,4 Ramon Mangues1,6 & Isolda Casanova1,6 1
Grup d’Oncogenesi i Antitumorals, Institut d’Investigacions Biomediques Sant Pau, Barcelona, Spain, 2Department of Pathology, Clınica Girona, Girona, Spain, 3Department of Haematology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 4Red Tematica de Investigacion Cooperativa en Cancer (RETICC), Barcelona, Spain, 5Department of Pathology, Hospital de la Sant Creu i Sant Pau, Barcelona, Spain, 6CIBER en Bioingenierıa, Biomateriales y Nanomecidicina (CIBER-BBN), Barcelona, Spain, and 7Department of Haematology, Hospital Universitari de Bellvitge, Barcelona, Spain
Date of submission 5 November 2013 Accepted for publication 23 January 2014 Published online Article Accepted 28 January 2014
Briones Bosch R, Dieguez-Gonzalez R, Moreno M J, Gallardo A, Novelli S, Espinosa I, Cespedes M V, Pav on M A, J, Gra~ nena A, Sierra J, Mangues R, Casanova I (2014) Histopathology 65, 119–131
Focal adhesion protein expression in human diffuse large B-cell lymphoma Aims: Focal adhesions have been associated with poor prognosis in multiple cancer types, but their prognostic value in diffuse large B-cell lymphoma (DLBCL) has not been evaluated. The aim of this study was to investigate the expression patterns and the prognostic value of the focal adhesion proteins FAK, Pyk2, p130Cas and HEF1 in DLBCL. Methods and results: Focal adhesion protein expression was examined using immunohistochemistry in normal lymphoid tissues and in 60 DLBCL patient samples. Kaplan–Meier survival and Cox regression analysis were performed to evaluate the correlation of focal adhesion protein expression with patient
prognosis. FAK, Pyk2, p130Cas and HEF1 expression was mostly found in the germinal centres of normal human lymphoid tissues. When assessed in DLBCL samples, FAK, Pyk2, p130Cas and HEF1 were highly expressed in 45%, 34%, 42% and 45% of the samples, respectively. Multivariate Cox analysis revealed that decreased FAK expression was a significant independent predictor of poorer disease outcome. Conclusions: FAK expression is an independent prognostic factor in DLBCL. Our results suggest that the addition of FAK immunostaining to the current immunohistochemical algorithms may facilitate risk stratification of DLBCL patients.
Keywords: diffuse, focal adhesions, large B-cell lymphoma, prognosis
Introduction Focal adhesion (FA) complexes are structures that link the actin cytoskeleton with the extracellular
Address for correspondence: R Mangues, Grup d’Oncogenesi i Antitumorals (GOA), Institut d’Investigacions Biomediques Sant Pau, Sant Antoni Maria Claret 167, Pavell o 11, 2n pis, Barcelona 08028, Spain. e-mail: [email protected] †Deceased. © 2014 John Wiley & Sons Ltd.
matrix. They are formed by transmembrane receptors (integrins), structural proteins (i.e. vinculin, paxillin, and talin), and signalling proteins, including focal adhesion kinase (FAK) and Crk-associated substrate (Cas) family proteins. FAK is a cytoplasmic tyrosine kinase that plays a major role in mediating signal transduction by integrins and growth factor receptors1 in the regulation of cell adhesion, migration, survival, proliferation and differentiation for a variety of cells.1–6 In cancer studies,
120
R Bosch et al.
the clinical implications of FAK have been controversial. In general, FAK overexpression is associated with aggressiveness, migration, invasiveness, metastasis and poor prognosis in many carcinomas.7,8 However, recent studies have shown that weak FAK expression is associated with poor prognosis in cervical cancer,9 extrahepatic bile duct carcinoma,10 and intrahepatic cholangiocarcinoma.11 Pyk2 (also known as RAFTK) is the second member of the FAK cytoplasmic tyrosine kinases.12 It has been shown to interact with proteins that also bind to FAK, such as Src, p130Cas, HEF1, and paxillin.13–16 Pyk2 is mostly expressed in haematopoietic cells, osteoclasts, and the central nervous system,17 and its expression is associated with metastasis and poor outcome in diverse human tumours.18–21 p130Cas, the first member of the Cas family to be identified, is a docking protein involved in multiple intracellular signalling pathways.22,23 Its overexpression contributes to the development of many human cancers, and it is necessary for the transformation initiated by different oncogenes in different cancer types.24–26 In breast cancer, its overexpression is associated with proliferation, metastasis, and poor prognosis.27,28 HEF1 is a Cas family member that is mostly expressed in lymphocytes and epithelial cells,29 and shares many structural and functional characteristics with p130Cas.30 The gene encoding HEF1 has been identified as promoting metastasis in many tumours.31–34 Moreover, all of these FA proteins are involved in normal B-cell development,35–37 suggesting that they can play a role in B-cell transformation. However, their role in haematological malignancies is poorly understood, and their involvement in diffuse large B-cell lymphoma (DLBCL), the most frequent non-Hodgkin lymphoma subtype, has not been assessed before. Our aim in this study was to characterize the expression of the focal adhesion proteins FAK, Pyk2, p130Cas and HEF1 in normal lymphoid tissues and in 60 DLBCL patient samples. We also wanted to correlate our findings with clinicopathological variables and patient survival.
Materials and methods PATIENTS AND TISSUE SAMPLES
We analysed 60 patients with primary nodal DLBCL, diagnosed in our institution according to the criteria of the World Health Organization classification. Availability of adequate histological material and of clinical data were the only criteria for selection of patients. Patients with a follicular lymphoma (including grade 3b) or another type of indolent lymphoma with subsequent transformation into large-cell lymphoma were not included. In addition, patients with primary
central nervous system, mediastinal or HIV-related lymphomas were excluded. All patients received anthracycline-containing chemotherapy. Forty-six (77%) patients were treated with cyclophosphamide, doxorubicin, vincristine, and prednisone, and the remaining 14 received other combination chemotherapies. None of the patients received rituximab, because they were treated before its addition to standard chemotherapy regimes. The clinical features of the patients are shown in Table 1. Approval was obtained from the research committee of Santa Creu i Sant Pau Hospital. Written informed consent was provided according to the declaration of Helsinki. All tissue samples were reviewed to confirm the diagnosis of DLBCL. Two tissue arrays containing the samples were generated with a tissue array device (Beecher Instruments, Silver Spring, MD, USA). Two cylinders (1 mm in largest diameter) from the same tumour were selected. Control tissues (normal lymph node, breast cancer, and ovarian cancer) were also included in the tissue arrays. Additional samples of whole sections of normal lymphoid tissues (lymph nodes, spleen, thymus, and Peyer’s patches) were also evaluated for the expression of FA proteins. IMMUNOHISTOCHEMISTRY
Tissue microarray blocks were sectioned at a thickness of 3 lm, and dried for 16 h at 56°C. They were then dewaxed in xylene, rehydrated through a graded ethanol series, and washed with phosphate-buffered saline. Heat-induced epitope retrieval was performed by immersing the sections in sodium citrate buffer (pH 6.0) and incubating them at 97°C for 20 min using a Dako PTLink (Dako, Carpinteria, CA, USA). The sections were stained using anti-FAK (1:75; Millipore Billerica, MA, USA), anti-Pyk2 (1:100; BD Biosciences, Franklin Lakes, NJ, USA), anti-p130Cas (1:75; Neomarkers, Fremont, CA, USA), and anti-HEF1 (1:200; ABR-Affinity BioReagents, Golden, CO, USA) antibodies. The expression of CD20, CD10, Ki67 and bcl-2 (pre-diluted; Dako) was also assessed to classify the samples into DLBCL subtypes. The reactions were performed in a Dako Autostainer Link48 (Dako), and visualized using the biotin-free EndVision system (Dako) with diaminobenzidine. CD20, CD10, Ki67 and bcl-2 expression was considered to be positive when ≥30% of the tumour cells were reactive, according to other studies.38 We estimated the percentage of positively stained cells (0–100%) for the whole tumour and the intensity of the staining (absent, 0; weak, 1; moderate, 2; and strong, 3) for FAK, Pyk2, p130Cas and HEF1 in DLBCL samples. An expression value for © 2014 John Wiley & Sons Ltd, Histopathology, 65, 119–131.
16 (27)
21 (35)
≥60 (n = 32)
Gender Male (n = 40)
© 2014 John Wiley & Sons Ltd, Histopathology, 65, 119–131.
14 (24)
17 (30)
14 (27)
20 (36)
10 (18)
Yes (n = 19)
B symptoms No (n = 32)
Yes (n = 11)
ECOG status 0–1 (n = 43)
2–4 (n = 12)
9 (21)
High (n = 11)
18 (31)
15 (35)
Serum B2M Normal (n = 32)
Bulky mass No (n = 39)
17 (31)
High (n = 27)
7 (12)
13 (24)
(n = 11)
Serum LDH Normal (n = 28)
Positive
24 (42)
15 (26)
III–IV (n = 27)
Bone marrow Negative (n = 46)
16 (28)
Ann Arbor stage I–II (n = 30)
12 (20)
17 (28)
Age (years)
