RESEARCH ARTICLE
Breast Cancer Molecular Subtypes and Oxidative DNA Damage Danica Jakovcevic, MD,* Natalija Dedic-Plavetic, MD, PhD,w Damir Vrbanec, MD, PhD,w Antonia Jakovcevic, MD,z and Jasminka Jakic-Razumovic, MD, PhDz
Background: Oxidative stress is thought to play a major role in etiology of many cancers, including breast cancer. 8-hydroxy20 deoxyguanosine (8-OHdG) is the most abundant marker of oxidative DNA damage. The aim of this study was to assess the extent of oxidative DNA damage in different breast cancer molecular surrogate subtypes to investigate the prognostic relevance and role of oxidative base lesion (8-OHdG) in the etiology of breast cancer. Materials and Methods: 8-OHdG expression was immunohistochemicaly studied on tissue microarrays constructed from 152 patients with invasive breast cancer. Expression was correlated with other prognostic factors, as well as different breast cancer molecular surrogate subtypes such as luminal A, luminal B [human epidermal growth factor receptor 2 (HER2) negative], luminal B (HER2 positive), HER2-enriched ad triplenegative tumors. Results: Triple-negative breast carcinomas (TNBCs) were more frequently 8-OHdG negative compared with non-TNBCs (P = 0.036). There was no statistically significant difference between 8-OHdG expression and other breast cancer molecular subtypes. In univariate analysis, there was no significant difference between 8-OHdG expression and breast cancer–specific death, although in multivariate analysis 8-OHdG overexpression was associated with better breast cancer–specific survival (BCSS) (odds ratio = 0.04, 95% confidence interval, 0.002-0.62). In Cox regression analysis, patients with moderate and strong 8-OHdG expression had 0.9 times smaller breast cancer death hazard ratio than patients with negative 8-OHdG expression. Conclusions: Oxidative stress may have less impact in the pathogenesis of TNBCs compared with other surrogate breast cancer molecular subtypes. 8-OHdG may be a promising biomarker in the prediction of prognosis for breast cancer patients. Received for publication February 18, 2014; accepted July 24, 2014. From the *Department of Pathology, Clinical Hospital “Sv. Duh”; Departments of wMedical Oncology; and zPathology, University Hospital Center Zagreb, Zagreb, Croatia. D.V. is currently receiving a Grant (#108-1080058-0046) “Breast cancermolecular, genetic and clinical characteristics” from Ministry of Science and Technology, Zagreb, Croatia. The authors declare no conflict of interest. Reprints: Danica Jakovcevic, MD, Department of Pathology, Clinical Hospital “Sv. Duh,” University Hospital Center Zagreb, Kisˇ paticˇeva 12, Zagreb 10000, Croatia (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
Key Words: breast cancer molecular subtypes, oxidative stress, 8-OHdG (Appl Immunohistochem Mol Morphol 2015;23:696–703)
O
f the total incidence of cancer in women, breast cancer represents about 22.9% and accounts for 13.7% of the mortality.1 Mainly determination of tumor size, steroid receptor status, lymph node status, and human epidermal growth factor receptor 2 (HER2) status have guided prognostic predictions and adjuvant therapy recommendations for patients with early-stage breast cancer. As a result, some patients might be overtreated or undertreated. Therefore, numerous biomarkers are investigated to help determine the optimal therapy, and prediction of therapeutic response and prognosis. Molecular profiling based upon variations in gene expression has identified several distinct breast cancer subtypes that differ markedly in prognosis and therapeutic targets they express.2–4 However, gene expression profiling is often impractical because of the time and expenses requested. Therefore, some studies have proposed that using immunohistochemical (IHC) surrogates for molecular subtyping can provide much of the prognostic information obtained by gene expression profiling.5,6 Thus, readily available estrogen receptor (ER), progesterone receptor (PR), and HER2 status of tumor can be used to approximate breast cancer subtypes. It is known that development of cancer is a multistep process, and its underlining etiology is still largely unknown. During the past decade, free radicals induced DNA damage has been proposed to play important role in carcinogenesis.7,8 Oxygen radicals are constantly generated within mammalian cells as a consequence of normal cellular metabolism. Leakage from the mitochondrial electron transport chain is the main source of reactive oxygen species in physiological conditions.9,10 Oxidative stress occurs when the concentration of reactive oxygen species generated exceeds the antioxidant capability of that cell. Free radicals can cause strand breaks, alterations in pyrimidine and purine bases and sister chromatid exchanges.11 This may inactivate tumor suppressor genes within tumor cells, or further increase expression of protooncogenes.12,13 The hydroxyl radical (OH) plays a major role in DNA oxidation, and as a result many types of oxidized nucleoside have been reported.14,15 8-hydroxy-20 deoxyguanosine (8-OHdG)
696 | www.appliedimmunohist.com Appl Immunohistochem Mol Morphol Volume 23, Number 10, November/December 2015 Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
Appl Immunohistochem Mol Morphol
Volume 23, Number 10, November/December 2015
is one of the most commonly occurring of these DNA modifications and is frequently used as a specific marker of oxidative DNA damage.16 8-OHdG is thought to be promutagenic because it induces G:C to T:A transversion unless repaired before replication.17 Therefore, this lesion is a potential biomarker of carcinogenesis.18 Its creation is regulated by local antioxidant capacity and DNA repair enzymes, particularly DNA glycosilase (hOGG1), which removes 8-OHdG from damaged DNA strands.19 Furthermore, carcinoma cells are frequently under persistent oxidative stress, which is shown by higher levels of 8-OHdG in cancer cells.20,21 Several studies have shown significantly higher 8-OHdG levels in breast carcinoma cells compared with surrounding healthy tissue.22,23 More recently, Karihtala and colleagues found that expression of 8-OHdG is greatly reduced in invasive breast carcinomas compared with premalignant lesions and that the absence of 8-OHdG is an independent prognostic factor of poor prognosis in breast carcinomas.24,25 It is still unknown whether there is correlation between oxidative DNA damage and breast cancer molecular subtypes. Therefore, the aim of this study was to determine whether 8-OHdG expression is related to different breast cancer subtypes, prognosis, or clinicopathologic prognostic factors such as steroid receptor and HER2 status, proliferation rate, tumor size, nodal status, and metastatic spreading. These data may help evaluate the possible role of 8-OHdG as a prognostic marker in breast carcinoma.
MATERIALS AND METHODS Breast biopsies were obtained from 212 consecutive patients with primary breast cancer diagnosed and treated in Clinical Hospital Centre Zagreb from October 2002 to October 2003. Only 152 patients with invasive breast carcinomas and with complete clinical follow-up data were included. The mean age of patients with invasive breast carcinoma was 56 years (range, 49 to 66.8 y). The youngest participant was 30 years old, whereas the oldest one was 83 years old. Median follow-up time was 112 months (range, 87.3 to 116 mo). For all patients pathohistologic data (tumor size, histologic type, histologic and nuclear grade, steroid receptor and HER2 status, proliferative index, lymph node status, lymphovascular invasion, and distant metastasis) were obtained as well as treatment information. Patients were separated into 5 groups according to surrogate molecular subtype determined by immunohistochemistry as follows: luminal A (ER positive or PR positive and Ki-67 < 14%), luminal B “HER2 negative” (ER, PR positive, HER2 negative, and Ki-67Z14%), luminal B “HER2 positive” (ER, PR, and HER2 positive), HER2-enriched (ER, PR negative, HER2 positive), and triple-negative (ER, PR, and HER2 negative).26 Histologic tumor type was determined using the World Health Organization classification.27 Histologic grade was assessed using Elston-Ellis method.28 Immunohistochemistry for ER, PR, and HER2 was performed on formalin-fixed, paraffin-embedded tissue slides Copyright
r
2015 Wolters Kluwer Health, Inc. All rights reserved.
Breast Cancer Molecular Subtypes and Oxidative DNA
with standard avidin-biotin-immunoperoxidase staining method using an automatic stainer (Dako, Glostrup, Denmark). Positive staining for ER and PR was defined as nuclear staining in >1% of tumor nuclei as descried previously.29 HER2 was considered positive if scored as 3+ on >30% of tumor area according to Herceptest manufacturer instructions, whereas cases with 0 and 1+ were regarded as negative. For patients with 2+ HER2 expression, additional chromogenic in situ hybridization tests were performed for the evaluation of gene amplification using criteria that >6 signals per nucleus represent HER2 amplification.30 Proliferation index was estimated by IHC staining with Ki-67 antibody, the cut offs of 14% of positive nuclei on 500 counted tumor cells.5
Tissue Microarrays (TMAs) and IHC Determination of 8-OHdG 8-OHdG IHC analysis was carried out using TMAs. TMAs were constructed as described.31,32 Briefly, 3 representative areas from each tumor were punched from paraffin block with special needle (Sakura, Japan) and placed on a new recipient paraffin block at the given coordinates for simultaneous analysis of multiple tissue samples. Recipient paraffin block was then left overnight in the incubator at 451C to allow cylinders to connect and integrate into the recipient block. Tissue microchips where than cut into 4- to 5-mm-thick sections and stained with standard IHC avidinbiotin-immunoperoxidase method using 8-OHdG antibody (Abcam, Cambridge, MA). Staining of 8-OHdG was previously described25 as nuclear staining. In brief, slides were incubated with a 1:50 primary antibody dilution against 8-OHdG for 60 minutes at room temperature. Secondary antibody was from EnVision (K8000; Dako) and diaminobenzidine (Dako) was used as chromogen. Surgical pathologists scoring the TMA were blinded to the clinicopathologic characteristics and outcome of each patient. 8-OHdG immunostaining results were presented as IHC score.33 According to the percentage of positive tumor cells for each tumor triplet average points were awarded: 0 = negative, 1 = 1% to 10%, 2 = 10% to 50%, 3 = > 50% of stained cells. Multiplying the intensity of staining (Fig. 1) with the number of points assigned to the percentage of stained cells the score value was established from 0 to 9. In our study, 4.6% patients have shown negative IHC score, whereas 44.7% patients had weak, 35.5% moderate, and 15.1% strong IHC score.
Statistical Analysis All the analyses were carried out using SPSS 17.0 (SPSS Inc., Chicago, IL) statistical software package. The significance of associations was determined by using w2 test, Man-Whitney and Kruskal-Wallis test. Univariate and multivariate prediction of specific mortality for breast cancer were carried by means of logistic regression and odds ratios with 95% confidence intervals (CIs) for each variable. Overall survival (OS) and disease-free survival (DFS) were analyzed by means of Kaplan-Meier curves and Cox regression. Statistical significance was set at P < 0.05. www.appliedimmunohist.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
697
Appl Immunohistochem Mol Morphol
Jakovcevic et al
Volume 23, Number 10, November/December 2015
FIGURE 1. The intensity of 8-hydroxy-20 deoxyguanosine immunohistochemical staining. A, No staining (score 0); (B) weak staining (score 1); (C) moderate staining (score 2); (D) strong staining (score 3) (A and D: magnification: 400).
RESULTS The clinicopathologic data of 152 breast carcinoma patients are summarized in Table 1. The majority of patients were older than 50, had invasive ductal carcinoma, histologic and nuclear grade II, and tumor
Breast Cancer Molecular Subtypes and Oxidative DNA Damage Danica Jakovcevic, MD,* Natalija Dedic-Plavetic, MD, PhD,w Damir Vrbanec, MD, PhD,w Antonia Jakovcevic, MD,z and Jasminka Jakic-Razumovic, MD, PhDz
Background: Oxidative stress is thought to play a major role in etiology of many cancers, including breast cancer. 8-hydroxy20 deoxyguanosine (8-OHdG) is the most abundant marker of oxidative DNA damage. The aim of this study was to assess the extent of oxidative DNA damage in different breast cancer molecular surrogate subtypes to investigate the prognostic relevance and role of oxidative base lesion (8-OHdG) in the etiology of breast cancer. Materials and Methods: 8-OHdG expression was immunohistochemicaly studied on tissue microarrays constructed from 152 patients with invasive breast cancer. Expression was correlated with other prognostic factors, as well as different breast cancer molecular surrogate subtypes such as luminal A, luminal B [human epidermal growth factor receptor 2 (HER2) negative], luminal B (HER2 positive), HER2-enriched ad triplenegative tumors. Results: Triple-negative breast carcinomas (TNBCs) were more frequently 8-OHdG negative compared with non-TNBCs (P = 0.036). There was no statistically significant difference between 8-OHdG expression and other breast cancer molecular subtypes. In univariate analysis, there was no significant difference between 8-OHdG expression and breast cancer–specific death, although in multivariate analysis 8-OHdG overexpression was associated with better breast cancer–specific survival (BCSS) (odds ratio = 0.04, 95% confidence interval, 0.002-0.62). In Cox regression analysis, patients with moderate and strong 8-OHdG expression had 0.9 times smaller breast cancer death hazard ratio than patients with negative 8-OHdG expression. Conclusions: Oxidative stress may have less impact in the pathogenesis of TNBCs compared with other surrogate breast cancer molecular subtypes. 8-OHdG may be a promising biomarker in the prediction of prognosis for breast cancer patients. Received for publication February 18, 2014; accepted July 24, 2014. From the *Department of Pathology, Clinical Hospital “Sv. Duh”; Departments of wMedical Oncology; and zPathology, University Hospital Center Zagreb, Zagreb, Croatia. D.V. is currently receiving a Grant (#108-1080058-0046) “Breast cancermolecular, genetic and clinical characteristics” from Ministry of Science and Technology, Zagreb, Croatia. The authors declare no conflict of interest. Reprints: Danica Jakovcevic, MD, Department of Pathology, Clinical Hospital “Sv. Duh,” University Hospital Center Zagreb, Kisˇ paticˇeva 12, Zagreb 10000, Croatia (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
Key Words: breast cancer molecular subtypes, oxidative stress, 8-OHdG (Appl Immunohistochem Mol Morphol 2015;23:696–703)
O
f the total incidence of cancer in women, breast cancer represents about 22.9% and accounts for 13.7% of the mortality.1 Mainly determination of tumor size, steroid receptor status, lymph node status, and human epidermal growth factor receptor 2 (HER2) status have guided prognostic predictions and adjuvant therapy recommendations for patients with early-stage breast cancer. As a result, some patients might be overtreated or undertreated. Therefore, numerous biomarkers are investigated to help determine the optimal therapy, and prediction of therapeutic response and prognosis. Molecular profiling based upon variations in gene expression has identified several distinct breast cancer subtypes that differ markedly in prognosis and therapeutic targets they express.2–4 However, gene expression profiling is often impractical because of the time and expenses requested. Therefore, some studies have proposed that using immunohistochemical (IHC) surrogates for molecular subtyping can provide much of the prognostic information obtained by gene expression profiling.5,6 Thus, readily available estrogen receptor (ER), progesterone receptor (PR), and HER2 status of tumor can be used to approximate breast cancer subtypes. It is known that development of cancer is a multistep process, and its underlining etiology is still largely unknown. During the past decade, free radicals induced DNA damage has been proposed to play important role in carcinogenesis.7,8 Oxygen radicals are constantly generated within mammalian cells as a consequence of normal cellular metabolism. Leakage from the mitochondrial electron transport chain is the main source of reactive oxygen species in physiological conditions.9,10 Oxidative stress occurs when the concentration of reactive oxygen species generated exceeds the antioxidant capability of that cell. Free radicals can cause strand breaks, alterations in pyrimidine and purine bases and sister chromatid exchanges.11 This may inactivate tumor suppressor genes within tumor cells, or further increase expression of protooncogenes.12,13 The hydroxyl radical (OH) plays a major role in DNA oxidation, and as a result many types of oxidized nucleoside have been reported.14,15 8-hydroxy-20 deoxyguanosine (8-OHdG)
696 | www.appliedimmunohist.com Appl Immunohistochem Mol Morphol Volume 23, Number 10, November/December 2015 Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
Appl Immunohistochem Mol Morphol
Volume 23, Number 10, November/December 2015
is one of the most commonly occurring of these DNA modifications and is frequently used as a specific marker of oxidative DNA damage.16 8-OHdG is thought to be promutagenic because it induces G:C to T:A transversion unless repaired before replication.17 Therefore, this lesion is a potential biomarker of carcinogenesis.18 Its creation is regulated by local antioxidant capacity and DNA repair enzymes, particularly DNA glycosilase (hOGG1), which removes 8-OHdG from damaged DNA strands.19 Furthermore, carcinoma cells are frequently under persistent oxidative stress, which is shown by higher levels of 8-OHdG in cancer cells.20,21 Several studies have shown significantly higher 8-OHdG levels in breast carcinoma cells compared with surrounding healthy tissue.22,23 More recently, Karihtala and colleagues found that expression of 8-OHdG is greatly reduced in invasive breast carcinomas compared with premalignant lesions and that the absence of 8-OHdG is an independent prognostic factor of poor prognosis in breast carcinomas.24,25 It is still unknown whether there is correlation between oxidative DNA damage and breast cancer molecular subtypes. Therefore, the aim of this study was to determine whether 8-OHdG expression is related to different breast cancer subtypes, prognosis, or clinicopathologic prognostic factors such as steroid receptor and HER2 status, proliferation rate, tumor size, nodal status, and metastatic spreading. These data may help evaluate the possible role of 8-OHdG as a prognostic marker in breast carcinoma.
MATERIALS AND METHODS Breast biopsies were obtained from 212 consecutive patients with primary breast cancer diagnosed and treated in Clinical Hospital Centre Zagreb from October 2002 to October 2003. Only 152 patients with invasive breast carcinomas and with complete clinical follow-up data were included. The mean age of patients with invasive breast carcinoma was 56 years (range, 49 to 66.8 y). The youngest participant was 30 years old, whereas the oldest one was 83 years old. Median follow-up time was 112 months (range, 87.3 to 116 mo). For all patients pathohistologic data (tumor size, histologic type, histologic and nuclear grade, steroid receptor and HER2 status, proliferative index, lymph node status, lymphovascular invasion, and distant metastasis) were obtained as well as treatment information. Patients were separated into 5 groups according to surrogate molecular subtype determined by immunohistochemistry as follows: luminal A (ER positive or PR positive and Ki-67 < 14%), luminal B “HER2 negative” (ER, PR positive, HER2 negative, and Ki-67Z14%), luminal B “HER2 positive” (ER, PR, and HER2 positive), HER2-enriched (ER, PR negative, HER2 positive), and triple-negative (ER, PR, and HER2 negative).26 Histologic tumor type was determined using the World Health Organization classification.27 Histologic grade was assessed using Elston-Ellis method.28 Immunohistochemistry for ER, PR, and HER2 was performed on formalin-fixed, paraffin-embedded tissue slides Copyright
r
2015 Wolters Kluwer Health, Inc. All rights reserved.
Breast Cancer Molecular Subtypes and Oxidative DNA
with standard avidin-biotin-immunoperoxidase staining method using an automatic stainer (Dako, Glostrup, Denmark). Positive staining for ER and PR was defined as nuclear staining in >1% of tumor nuclei as descried previously.29 HER2 was considered positive if scored as 3+ on >30% of tumor area according to Herceptest manufacturer instructions, whereas cases with 0 and 1+ were regarded as negative. For patients with 2+ HER2 expression, additional chromogenic in situ hybridization tests were performed for the evaluation of gene amplification using criteria that >6 signals per nucleus represent HER2 amplification.30 Proliferation index was estimated by IHC staining with Ki-67 antibody, the cut offs of 14% of positive nuclei on 500 counted tumor cells.5
Tissue Microarrays (TMAs) and IHC Determination of 8-OHdG 8-OHdG IHC analysis was carried out using TMAs. TMAs were constructed as described.31,32 Briefly, 3 representative areas from each tumor were punched from paraffin block with special needle (Sakura, Japan) and placed on a new recipient paraffin block at the given coordinates for simultaneous analysis of multiple tissue samples. Recipient paraffin block was then left overnight in the incubator at 451C to allow cylinders to connect and integrate into the recipient block. Tissue microchips where than cut into 4- to 5-mm-thick sections and stained with standard IHC avidinbiotin-immunoperoxidase method using 8-OHdG antibody (Abcam, Cambridge, MA). Staining of 8-OHdG was previously described25 as nuclear staining. In brief, slides were incubated with a 1:50 primary antibody dilution against 8-OHdG for 60 minutes at room temperature. Secondary antibody was from EnVision (K8000; Dako) and diaminobenzidine (Dako) was used as chromogen. Surgical pathologists scoring the TMA were blinded to the clinicopathologic characteristics and outcome of each patient. 8-OHdG immunostaining results were presented as IHC score.33 According to the percentage of positive tumor cells for each tumor triplet average points were awarded: 0 = negative, 1 = 1% to 10%, 2 = 10% to 50%, 3 = > 50% of stained cells. Multiplying the intensity of staining (Fig. 1) with the number of points assigned to the percentage of stained cells the score value was established from 0 to 9. In our study, 4.6% patients have shown negative IHC score, whereas 44.7% patients had weak, 35.5% moderate, and 15.1% strong IHC score.
Statistical Analysis All the analyses were carried out using SPSS 17.0 (SPSS Inc., Chicago, IL) statistical software package. The significance of associations was determined by using w2 test, Man-Whitney and Kruskal-Wallis test. Univariate and multivariate prediction of specific mortality for breast cancer were carried by means of logistic regression and odds ratios with 95% confidence intervals (CIs) for each variable. Overall survival (OS) and disease-free survival (DFS) were analyzed by means of Kaplan-Meier curves and Cox regression. Statistical significance was set at P < 0.05. www.appliedimmunohist.com |
Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.
697
Appl Immunohistochem Mol Morphol
Jakovcevic et al
Volume 23, Number 10, November/December 2015
FIGURE 1. The intensity of 8-hydroxy-20 deoxyguanosine immunohistochemical staining. A, No staining (score 0); (B) weak staining (score 1); (C) moderate staining (score 2); (D) strong staining (score 3) (A and D: magnification: 400).
RESULTS The clinicopathologic data of 152 breast carcinoma patients are summarized in Table 1. The majority of patients were older than 50, had invasive ductal carcinoma, histologic and nuclear grade II, and tumor
