Downloaded from http://jcp.bmj.com/ on April 6, 2015 - Published by group.bmj.com
Original article
ERBB3 as an independent prognostic marker for nasopharyngeal carcinoma Warut Tulalamba,1 Noppadol Larbcharoensub,2 Tavan Janvilisri3 ▸ Additional material is published online only. To view please visit the journal online (http://dx.doi.org/10.1136/ jclinpath-2013-202154). 1
Graduate Programme in Molecular Medicine, Mahidol University, Bangkok, Thailand 2 Department of Pathology, Faculty of Medicine Ramathibodi Hospital Mahidol University, Bangkok, Thailand 3 Department of Biochemistry, Faculty of Science Mahidol University, Bangkok, Thailand Correspondence to Dr Tavan Janvilisri, Department of Biochemistry, Faculty of Science, Mahidol University, 272 Rama VI Road, Phayathai, Rajdhevi, Bangkok 10400, Thailand; [email protected] Received 20 December 2013 Revised 30 March 2014 Accepted 21 April 2014 Published Online First 13 May 2014
ABSTRACT Aim Although the ERBB proteins have been shown to be associated in many types of human tumours and serve as important cancer therapeutic targets, however, data regarding the expression and clinical relevance of ERBBs in nasopharyngeal carcinoma (NPC) are still conflicting. The aim of this study is to investigate the expression pattern of all ERBB members simultaneously in NPC tissues using immunohistochemistry and determine their clinical relevance. Methods The expression of all members of ERBB proteins was evaluated using immunohistochemistry on 82 NPC tissue samples. Relationship between the ERBB protein expression, clinicopathological parameters and patient outcome was assessed using univariate and multivariate analyses. Results We found that ERBB1, ERBB2 and ERBB3 were strongly expressed in the normal nasopharyngeal epithelial cells. A marked reduction of ERBB1 and ERBB2 expression in NPC was observed compared with the non-cancerous tissues. 76 of 82 (92.7%) cases were ERBB3-positive, while ERBB4 was not expressed in both normal and NPC. The univariate log-rank analysis showed that regional lymph node metastasis, systemic metastasis, recurrence and ERBB3 expression were associated with patient survival. The ERBB3 expression was not correlated to other clinicopathological factors. Furthermore, multivariate analysis revealed that ERBB3 expression was an independent prognostic factor influencing patient survival. Conclusions Our results suggested that the expression of ERBB3 is associated with patient survival and could serve as a novel and valuable predictor for prognostic evaluation of patients with NPC.
INTRODUCTION
To cite: Tulalamba W, Larbcharoensub N, Janvilisri T. J Clin Pathol 2014;67:667–672.
The ERBB protein family, whose members belong to type 1 tyrosine kinase receptors including ERBB1 (also known as EGFR/HER1), ERBB2 (HER2/Neu), ERBB3 (HER3) and ERBB4 (HER4), represents the key control of fundamental cellular processes, including proliferation, cell migration, metabolism and survival under normal physiological conditions.1 2 Binding of the ligands to ERBB triggers homodimerisation or heterodimerisation of ERBB, resulting in autophosphorylation of the receptors and initiates downstream signalling events. The activated ERBB proteins recruit various signalling complexes such as those in MAPK, PI3K/ Akt, STATs, Src kinase and mTOR pathways.3 ERBB receptors have been closely associated with pathogenesis and progression of several cancers, including breast, colon, pancreas, ovary, brain, lung, and head and neck cancer.4 5 Alterations of ERBB in human tumours include gene amplification leading to
Tulalamba W, et al. J Clin Pathol 2014;67:667–672. doi:10.1136/jclinpath-2013-202154
aberrant receptor expression and gain-of-function mutations that promote constitutive receptor activation and subsequent induction of specific intracellular signalling cascades.3 5 Expression of ERBB receptors in many types of cancer has therefore been extensively studied as diagnostic and prognostic biomarkers as well as potential targets for selective therapy. Nasopharyngeal carcinoma (NPC) is a malignant cancer arising from the epithelial surface of the nasopharynx. The incidence is high in East and Southeast Asia including Thailand.6 The aetiological factors of NPC include the Epstein–Barr virus (EBV) infection, ethnic background, genetic susceptibility, environmental factors and dietary consumption.7 Most of the patients with NPC appear in advanced stages, with locally advanced or metastatic disease at presentation, due to the silent and painless area of primary anatomical site of tumour growth. Moreover, NPC in advanced stages exhibit higher metastatic potential than other head and neck cancers.8 NPC has a poor prognosis because of late presentation of lesions, poor understanding of molecular pathogenesis, lack of suitable biomarkers for early detection and poor response to available therapies.7 9 Although several studies evaluated the prognostic values of ERBB gene and protein expression in NPC, however, a discrepancy exists among them. While overexpression of ERBB1 was observed in more than 60% of NPC patients and was correlated to poor prognosis in one study,10 others reported 40–70% ERBB1-positive NPC tissues without prognostic value.11–13 Similar to ERBB1, the ErbB2 gene amplification and protein overexpression of the corresponding protein in NPC were reported and linked to poor prognosis.13 14 The contradicting data on gene amplification exist,15 and a range of ∼15– 50% ERBB2 expression has been observed in NPC tissues.11–13 15–17 Interestingly, a number of studies revealed the absence of ERBB2 in most NPC tissues.11 18 19 The information for ERBB3 and ERBB4 in NPC is still lacking. Due to this disparity, we made the first attempt to investigate the expression pattern of all ERBB members in 82 NPC patient biopsies using immunohistochemistry. The relationship between the expression of ERBB, clinicopathological data and patient outcome was also evaluated.
MATERIALS AND METHODS Tissue specimens, patients and treatment Eighty-two paraffin-embedded tissue samples from NPC patients that were diagnosed, pathologically confirmed and treated at Ramathibodi Hospital between 1998 and 2007 were included in this 667
Downloaded from http://jcp.bmj.com/ on April 6, 2015 - Published by group.bmj.com
Original article Table 1 Clinicopathological characteristics of nasopharyngeal carcinoma patients in this study Characteristics Age Mean Median Range Gender Male Female WHO classification Type 2a Type 2b AJCC staging Stage I Stage II Stage III Stage IV T stage T1 T2 T3 T4 Regional lymph node metastasis No Yes (N1, N2, N3) Systemic metastasis No Yes (M1) Recurrence No Yes N/A Total
No. of patients
48.67 48 14–79 57 25 43 39 3 13 27 39 10 24 25 23 15 67 77 5 35 42 5 82
AJCC, American Joint Committee on Cancer.
Years Years Years % 69.5 30.5 % 52.4 47.6 % 3.7 15.9 32.9 47.6 % 12.2 29.3 30.5 28.0 % 18.3 81.7 % 93.9 6.1 % 42.7 51.2 6.1 100.00
study. All specimens were taken at the time of diagnosis for NPC before the therapeutic treatment. The detailed patient characteristics are outlined in table 1. Clinical stages and histological grades were evaluated according to the International Union against Cancer/American Joint Committee on Cancer (UICC/AJCC) staging system and the WHO classification, respectively. The patients were given radiation to primary tumour at 18–20 Gy daily, five times/week, with a total dose ∼70 Gy. Three cycles of chemotherapy (cisplatin 100 mg/m2) were given concurrently during weeks 1, 4, 7, 10, 13 and 16 of radiotherapy. Subsequent to the completion of concurrent chemoradiotherapy, the patients received the chemotherapy consisting of cisplatin (80 mg/m2) at day 1 and 5-fluorouracil (1000 mg/m2/day) during days 1–5 every 4 weeks for three cycles. All patients were followed up every 3–6 months for at least 5 years.
Immunohistochemistry The paraffin-embedded specimens were cut into 4 mm sections and incubated at 65°C for 1 h. All sections were deparaffinised in xylene, were rehydrated in a series of reducing concentrations of ethanol and were submerged into Tris-EDTA buffer pH 9.0 with three 5 min cycles of microwave. The sections were then incubated in 3% H2O2 for 30 min to quench endogenous peroxidase activity and were treated with 5% normal goat serum (Cell Signaling Technology, Massachusetts, USA) in Tris buffer saline (TBS) for 1 h to reduce non-specific binding. Rabbit anti-ERBB1, ERBB2, ERBB3 or ERBB4 monoclonal antibodies (1:50; Cell Signaling Technology) was incubated with the sections for 12 h at 4°C in a humidified chamber. The antibody was replaced by normal goat serum for negative controls. The samples were washed with TBS and were subjected to SignalStain Boost IHC Detection Reagent (Cell Signaling Technology) for 1 h. The immunoreactions were visualised with 3-amino-9-ethylcarbazole peroxidase substrate kit (Vector Laboratory, California, USA) with haematoxylin counterstaining.
Figure 1 ERBB protein expression in nasopharyngeal carcinoma. Shown are representative photomicrographs at magnification ×400 of normal nasopharynx and cancerous nasopharyngeal tissues that were subjected to immunostaining of ERBB1 (A and B), ERBB2 (D and E), ERBB3 (G and H) and ERBB4 ( J and K). Human liver tissues were used as positive controls (C, F, I and L).
668
Tulalamba W, et al. J Clin Pathol 2014;67:667–672. doi:10.1136/jclinpath-2013-202154
Downloaded from http://jcp.bmj.com/ on April 6, 2015 - Published by group.bmj.com
Original article Ten normal nasopharyngeal tissues and human liver tissues were included as positive controls. The immunoreactivities were classified as a continuum from undetected level (0%) to diffuse and homogenous, strong staining (100%). The levels of ERBB proteins are based on the intensity of staining compared with internal controls within the same sample. The positive staining is defined when the reactivity is greater in the tumour cells than in the control cells and the staining is observed in more than 10% of tumour cells.
Statistical analysis A comparison between the clinicopathological profiles and survival was evaluated using χ2 test. Univariate analyses were performed to determine the prognostic factors associated with 5-year survival. Univariate and multivariable Cox regression analyses were also performed using all variables. Survival curves were constructed using the Kaplan–Meier plots. Two-tailed p values of
Original article
ERBB3 as an independent prognostic marker for nasopharyngeal carcinoma Warut Tulalamba,1 Noppadol Larbcharoensub,2 Tavan Janvilisri3 ▸ Additional material is published online only. To view please visit the journal online (http://dx.doi.org/10.1136/ jclinpath-2013-202154). 1
Graduate Programme in Molecular Medicine, Mahidol University, Bangkok, Thailand 2 Department of Pathology, Faculty of Medicine Ramathibodi Hospital Mahidol University, Bangkok, Thailand 3 Department of Biochemistry, Faculty of Science Mahidol University, Bangkok, Thailand Correspondence to Dr Tavan Janvilisri, Department of Biochemistry, Faculty of Science, Mahidol University, 272 Rama VI Road, Phayathai, Rajdhevi, Bangkok 10400, Thailand; [email protected] Received 20 December 2013 Revised 30 March 2014 Accepted 21 April 2014 Published Online First 13 May 2014
ABSTRACT Aim Although the ERBB proteins have been shown to be associated in many types of human tumours and serve as important cancer therapeutic targets, however, data regarding the expression and clinical relevance of ERBBs in nasopharyngeal carcinoma (NPC) are still conflicting. The aim of this study is to investigate the expression pattern of all ERBB members simultaneously in NPC tissues using immunohistochemistry and determine their clinical relevance. Methods The expression of all members of ERBB proteins was evaluated using immunohistochemistry on 82 NPC tissue samples. Relationship between the ERBB protein expression, clinicopathological parameters and patient outcome was assessed using univariate and multivariate analyses. Results We found that ERBB1, ERBB2 and ERBB3 were strongly expressed in the normal nasopharyngeal epithelial cells. A marked reduction of ERBB1 and ERBB2 expression in NPC was observed compared with the non-cancerous tissues. 76 of 82 (92.7%) cases were ERBB3-positive, while ERBB4 was not expressed in both normal and NPC. The univariate log-rank analysis showed that regional lymph node metastasis, systemic metastasis, recurrence and ERBB3 expression were associated with patient survival. The ERBB3 expression was not correlated to other clinicopathological factors. Furthermore, multivariate analysis revealed that ERBB3 expression was an independent prognostic factor influencing patient survival. Conclusions Our results suggested that the expression of ERBB3 is associated with patient survival and could serve as a novel and valuable predictor for prognostic evaluation of patients with NPC.
INTRODUCTION
To cite: Tulalamba W, Larbcharoensub N, Janvilisri T. J Clin Pathol 2014;67:667–672.
The ERBB protein family, whose members belong to type 1 tyrosine kinase receptors including ERBB1 (also known as EGFR/HER1), ERBB2 (HER2/Neu), ERBB3 (HER3) and ERBB4 (HER4), represents the key control of fundamental cellular processes, including proliferation, cell migration, metabolism and survival under normal physiological conditions.1 2 Binding of the ligands to ERBB triggers homodimerisation or heterodimerisation of ERBB, resulting in autophosphorylation of the receptors and initiates downstream signalling events. The activated ERBB proteins recruit various signalling complexes such as those in MAPK, PI3K/ Akt, STATs, Src kinase and mTOR pathways.3 ERBB receptors have been closely associated with pathogenesis and progression of several cancers, including breast, colon, pancreas, ovary, brain, lung, and head and neck cancer.4 5 Alterations of ERBB in human tumours include gene amplification leading to
Tulalamba W, et al. J Clin Pathol 2014;67:667–672. doi:10.1136/jclinpath-2013-202154
aberrant receptor expression and gain-of-function mutations that promote constitutive receptor activation and subsequent induction of specific intracellular signalling cascades.3 5 Expression of ERBB receptors in many types of cancer has therefore been extensively studied as diagnostic and prognostic biomarkers as well as potential targets for selective therapy. Nasopharyngeal carcinoma (NPC) is a malignant cancer arising from the epithelial surface of the nasopharynx. The incidence is high in East and Southeast Asia including Thailand.6 The aetiological factors of NPC include the Epstein–Barr virus (EBV) infection, ethnic background, genetic susceptibility, environmental factors and dietary consumption.7 Most of the patients with NPC appear in advanced stages, with locally advanced or metastatic disease at presentation, due to the silent and painless area of primary anatomical site of tumour growth. Moreover, NPC in advanced stages exhibit higher metastatic potential than other head and neck cancers.8 NPC has a poor prognosis because of late presentation of lesions, poor understanding of molecular pathogenesis, lack of suitable biomarkers for early detection and poor response to available therapies.7 9 Although several studies evaluated the prognostic values of ERBB gene and protein expression in NPC, however, a discrepancy exists among them. While overexpression of ERBB1 was observed in more than 60% of NPC patients and was correlated to poor prognosis in one study,10 others reported 40–70% ERBB1-positive NPC tissues without prognostic value.11–13 Similar to ERBB1, the ErbB2 gene amplification and protein overexpression of the corresponding protein in NPC were reported and linked to poor prognosis.13 14 The contradicting data on gene amplification exist,15 and a range of ∼15– 50% ERBB2 expression has been observed in NPC tissues.11–13 15–17 Interestingly, a number of studies revealed the absence of ERBB2 in most NPC tissues.11 18 19 The information for ERBB3 and ERBB4 in NPC is still lacking. Due to this disparity, we made the first attempt to investigate the expression pattern of all ERBB members in 82 NPC patient biopsies using immunohistochemistry. The relationship between the expression of ERBB, clinicopathological data and patient outcome was also evaluated.
MATERIALS AND METHODS Tissue specimens, patients and treatment Eighty-two paraffin-embedded tissue samples from NPC patients that were diagnosed, pathologically confirmed and treated at Ramathibodi Hospital between 1998 and 2007 were included in this 667
Downloaded from http://jcp.bmj.com/ on April 6, 2015 - Published by group.bmj.com
Original article Table 1 Clinicopathological characteristics of nasopharyngeal carcinoma patients in this study Characteristics Age Mean Median Range Gender Male Female WHO classification Type 2a Type 2b AJCC staging Stage I Stage II Stage III Stage IV T stage T1 T2 T3 T4 Regional lymph node metastasis No Yes (N1, N2, N3) Systemic metastasis No Yes (M1) Recurrence No Yes N/A Total
No. of patients
48.67 48 14–79 57 25 43 39 3 13 27 39 10 24 25 23 15 67 77 5 35 42 5 82
AJCC, American Joint Committee on Cancer.
Years Years Years % 69.5 30.5 % 52.4 47.6 % 3.7 15.9 32.9 47.6 % 12.2 29.3 30.5 28.0 % 18.3 81.7 % 93.9 6.1 % 42.7 51.2 6.1 100.00
study. All specimens were taken at the time of diagnosis for NPC before the therapeutic treatment. The detailed patient characteristics are outlined in table 1. Clinical stages and histological grades were evaluated according to the International Union against Cancer/American Joint Committee on Cancer (UICC/AJCC) staging system and the WHO classification, respectively. The patients were given radiation to primary tumour at 18–20 Gy daily, five times/week, with a total dose ∼70 Gy. Three cycles of chemotherapy (cisplatin 100 mg/m2) were given concurrently during weeks 1, 4, 7, 10, 13 and 16 of radiotherapy. Subsequent to the completion of concurrent chemoradiotherapy, the patients received the chemotherapy consisting of cisplatin (80 mg/m2) at day 1 and 5-fluorouracil (1000 mg/m2/day) during days 1–5 every 4 weeks for three cycles. All patients were followed up every 3–6 months for at least 5 years.
Immunohistochemistry The paraffin-embedded specimens were cut into 4 mm sections and incubated at 65°C for 1 h. All sections were deparaffinised in xylene, were rehydrated in a series of reducing concentrations of ethanol and were submerged into Tris-EDTA buffer pH 9.0 with three 5 min cycles of microwave. The sections were then incubated in 3% H2O2 for 30 min to quench endogenous peroxidase activity and were treated with 5% normal goat serum (Cell Signaling Technology, Massachusetts, USA) in Tris buffer saline (TBS) for 1 h to reduce non-specific binding. Rabbit anti-ERBB1, ERBB2, ERBB3 or ERBB4 monoclonal antibodies (1:50; Cell Signaling Technology) was incubated with the sections for 12 h at 4°C in a humidified chamber. The antibody was replaced by normal goat serum for negative controls. The samples were washed with TBS and were subjected to SignalStain Boost IHC Detection Reagent (Cell Signaling Technology) for 1 h. The immunoreactions were visualised with 3-amino-9-ethylcarbazole peroxidase substrate kit (Vector Laboratory, California, USA) with haematoxylin counterstaining.
Figure 1 ERBB protein expression in nasopharyngeal carcinoma. Shown are representative photomicrographs at magnification ×400 of normal nasopharynx and cancerous nasopharyngeal tissues that were subjected to immunostaining of ERBB1 (A and B), ERBB2 (D and E), ERBB3 (G and H) and ERBB4 ( J and K). Human liver tissues were used as positive controls (C, F, I and L).
668
Tulalamba W, et al. J Clin Pathol 2014;67:667–672. doi:10.1136/jclinpath-2013-202154
Downloaded from http://jcp.bmj.com/ on April 6, 2015 - Published by group.bmj.com
Original article Ten normal nasopharyngeal tissues and human liver tissues were included as positive controls. The immunoreactivities were classified as a continuum from undetected level (0%) to diffuse and homogenous, strong staining (100%). The levels of ERBB proteins are based on the intensity of staining compared with internal controls within the same sample. The positive staining is defined when the reactivity is greater in the tumour cells than in the control cells and the staining is observed in more than 10% of tumour cells.
Statistical analysis A comparison between the clinicopathological profiles and survival was evaluated using χ2 test. Univariate analyses were performed to determine the prognostic factors associated with 5-year survival. Univariate and multivariable Cox regression analyses were also performed using all variables. Survival curves were constructed using the Kaplan–Meier plots. Two-tailed p values of
