- Email: [email protected]
Comparison of immunohistochemical analysis with estrogen receptor SP1 and 1D5 monoclonal antibodies in breast cancer
Pathology – Research and Practice 208 (2012) 657–661
Contents lists available at SciVerse ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original Article
Comparison of immunohistochemical analysis with estrogen receptor SP1 and 1D5 monoclonal antibodies in breast cancer Klesia Pirola Madeira a , Renata Dalmaschio Daltoé a,b , Gabriela Modenesi Sirtoli a , Lucas Cunha Dias de Rezende a , Alex Assis de Carvalho c , Isabella dos Santos Guimarães a , Ian Victor Silva a,d , Leticia Batista Azevedo Rangel a,e,∗ a
Biotechnology Program/RENORBIO, Health Sciences Center, Federal University of Espirito Santo, Brazil Department of Pharmacy and Nutrition, Federal University of Espirito Santo, Brazil c Department of Pathology, Health Sciences Center, Federal University of Espirito Santo, Brazil d Department of Morphology, Health Sciences Center, Federal University of Espirito Santo, Brazil e Department of Pharmaceutical Sciences, Health Sciences Center, Federal University of Espirito Santo, Brazil b
a r t i c l e
i n f o
Article history: Received 9 April 2012 Received in revised form 19 July 2012 Accepted 30 July 2012 Keywords: Estrogen receptor alpha Monoclonal antibodies Breast cancer
s u m m a r y In the present study, we aimed to evaluate estrogen receptor ER-alpha status in 61 breast cancer cases using Sp1 and 1D5 monoclonal antibodies. Tissue array platforms were generated containing samples of breast cancer and positive controls that were assayed by immunohistochemistry applying monoclonal primary antibodies anti-ER alpha, SP1 and 1D5. We noted a high concordance rate (96.7%) between the referred antibodies. Moreover, we calculated the Kappa factor (0.921), indicating that 1D5 and SP1 provided overlapping ER␣ expression results. Indeed, we observed controversial results only in 2 samples studied, which were ER-negative when stained with 1D5 and ER-positive when assessed with SP1. Total concordance of PS was obtained (Pearson and intraclass CF, 0.7351 and 0.6193, respectively). However, concordance between the antibodies seems to be more accurate in higher PS values. An excellent IS correlation between antibodies was observed throughout the population (Spearman’s CF, = 0.9150). Following the Allred score, 17 out of 42 positive BC samples diverged, with 1D5 always pointing to weaker staining than SP1. When calculating Spearman’s CF of Total Score (TS) within the population, an excellent correlation between both the antibodies ( = 0.9238) was noted. Nonetheless, the results were less concordant among the BC-positive cases ( = 0.7743). Indeed, 20 samples were differentially classified using the antibodies (only 3 had higher TS with 1D5). Considering the mean TS of all samples or of invasive ductal carcinoma, SP1 provided higher scores than 1D5 (p < 0.05). We recommend the use of the anti-ER RMAb SP1 due to the high probability that the BC ER␣ status can be determined accurately as the reagent provides higher IS. Therefore, the A-score was higher than the MMAb 1D5. Ultimately, higher IS and A-score decrease the possibility of ER␣ status misinterpretation and, consequently, inappropriate BC treatment that would compromise the patient’s quality of life and overall survival. We recommend the use of anti-ER RMAb SP1 due to the high probability that the BC ER status can be determined accurately as the reagent provides higher IS, therefore higher A-score, than the MMAb 1D5. © 2012 Elsevier GmbH. All rights reserved.
Introduction An accurate pathological analysis of tissue biopsies is crucial for a conclusive and precise disease diagnosis, as well as for establishing its prognosis and therapeutic guidelines. In this context, the evaluation of cellular markers in breast cancer (BC) samples has proven imperative in the determination of disease diagnosis and
∗ Corresponding author at: Avenida Marechal Campos, 1468, Maruípe, Vitória, Espírito Santo, Brasil. Departamento de Ciências Farmacêuticas, 2◦ Andar, Sala 8, CEP: 29043-900, Brazil. Tel.: +55 27 33357539. E-mail address: [email protected] (L.B.A. Rangel). 0344-0338/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.prp.2012.07.010
prognosis, the patients’ eligibility for targeted therapies, or the use of endocrine and monoclonal antibody therapies for treatment [13]. The precise assessment of the expression levels of the estrogen (ER) and progesterone (PR) receptors by BC cells is an important procedure during BC diagnosis. Phenotypic analysis correlates with disease outcome and endocrine therapy responsiveness (mainly regarding postoperative adjuvant systemic therapy of patients with invasive BCs and BC recurrences) [7,9]. Positive tumors usually correlate with better disease prognosis than the negative ones because the former is entitled to the endocrine therapy with selective modulators of ER, such as tamoxifene, aromatase inhibitors, and anastrozole [8,13].
658
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
Materials and methods Samples We analyzed 61 BC samples, surgically resected in the two public oncology reference hospitals in the state of Espírito Santo (Brazil) Cassiano Antonio de Moraes Hospital (HUCAM) and Santa Rita de Cássia Hospital (HSRC). All cases were revised by a pathologist, and the BC diagnosis was confirmed. Cohort definition Primary BC cases registered at HUCAM from 1997 to 2007 and at HSRC from 2001 to 2007 were included in the present study. All living patients gave written informed consent before enrollment in the analysis. This work was approved by the Human Research Ethics Committee of the Federal University of Espírito Santo. TMA platforms and ER-alpha immunohistochemistry assays
Fig. 1. This figure presents one discordant case. (A) Invasive Ductal Carcinoma (anti ER-alpha 1D5 immunohistochemical stain, 400×). (B) Invasive Ductal Carcinoma (anti ER-alpha SP1 imunohistochemical stain, 400×).
Therefore, using immunohistochemistry (IHC) for the determination of BC hormone receptor status is clinically invaluable because information obtained by this method is considered specific, sensitive, and affordable [5,10]. In this respect, both anti-ER mouse (MMAbs) and rabbit (RMAbs) monoclonal antibodies have been used clinically. Whereas two different MMAbs, 6F11 and 1D5, seem to provide equivalent results in BC biopsies [14], several authors have investigated the efficacy of these antibodies in BC samples. Kaplan and colleagues (2005) showed that the two antiER MMAbs, 6F11 and 1D5, provided equivalent results with respect to immunochemistry assays [11]. On the other hand, some authors have also compared the accuracy of the anti-ER MMAb 1D5 with the anti-ER RMAbs SP1, leading to controversial data. In 2006, Cheang et al. published a pioneer study comparing the accuracy of the ER antibodies, 1D5 and SP1, by immunohistochemistry, revealing an eight-fold higher affinity of the latter to ER in comparison to the former [5]. Nonetheless, Brock et al. (2009) concluded that the antiER 1D5 and SP1 are equivalent in immunohistochemistry assays [3]. Despite these conflicting results, many authors have pointed to the benefits of using RMAbs due to the their high affinity against the epitope of interest, therefore providing precise immunohistochemistry analysis due to the intense positive staining and low background staining. In the present study, ER-alpha status was evaluated in 61 BC cases using Sp1 and 1D5 monoclonal antibodies. The ER staining results of immunohistochemistry were compared by assessing the percentage of cells stained within the analyzed sample, as well as their staining intensity by applying the Allred score system.
Hematoxylin and eosin stained sections from each breast sample were reviewed by a pathologist to confirm the primary BC diagnosis, and to select one representative 2 mm area of the tumor embedded in a paraffin block for immunohistochemical analysis. Each 2 mm sample was further transferred to a transient tissue array platform, generating customized BC tissue microarray (TMA) platforms [12,14–16]. The immunohistochemistry assay followed a protocol optimized by our group in 5 m sample sections of the generated breast cancer TMA platforms [14–16]. Briefly, after removal of paraffin, sections were immersed in preheated antigen-retrieval solution (Tris/EDTA 1X, pH = 9), incubated at 95–99 ◦ C for 30 min, and then allowed to cool down to room temperature for 30 min. Then, sections were incubated for 3 h with the primary antibodies of interest: MMAb anti-human ER-alpha, clone 1D5 (DakoCytomation, Carpinteria CA), dilution 1:100 and RMAb anti-human ER-alpha, clone SP1 (Spring Bioscience, Pleasanton, CA), dilution 1:100. Sections were incubated with biotinylated universal secondary antibody (Dako Cytomation LSAB+ System-HRP, Dako Cytomation, Carpinteria CA) for 30 min. Endogenous peroxidase activity was blocked by 5 min of incubation in 3% (v/v) hydrogen peroxide. Antigen–antibody complexes were detected by the avidin–biotin–peroxidase method, using 3,3-diaminobenzidine (DAB) as chromogenic substrate. The sections were incubated with streptavidin conjugated to peroxidase (LSAB+ System-HRP, Dako Cytomation, Carpinteria CA) for 30 min and then with DAB (Liquid DAB+ Substrat Chromogen System, Dako Cytomation, Dako Cytomation, Carpinteria CA) for 5 min. Slides were counterstained with hematoxylin and immersed in 5% (p/v) ammonium hydroxide. Positive controls were included in the BC TMA platform in opposite positions of the chip in order to guarantee an accurate analysis of protein expression in breast tissue. Negative control experiments were conducted in the absence of the primary antibodies listed above, in parallel of all assays. Cases with less than 1% nuclear staining in tumor cells were considered negative for the expression of the protein of interest. Subcellular localization was also noted. The scoring system proposed for the analysis of the 4 TMA platforms was the Allred score. This method categorizes the samples based on a proportion score (PS) that classifies the cases into 6 groups according to the percentage of stained cells, and on an intensity score (IS) that divides them into 4 classes according to staining intensity. The PS groups are as follows: Score 0: negative; Score 1: <1%; Score 2: 1–10%; Score 3: 10–33.3%; Score 4: 33.3–66.6%; and Score 5: >66.6%. The IS classes are similar: Score 0: negative; Score 1: weak; Score 2: intermediate; and Score 3: strong. Finally, the two scores are combined, and the total score (TS) is given, considering
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
659
an A-score of 0 or 2–8 [1]. The IHC reactions were interpreted by a pathologist with expertise in breast pathology. Statistical analysis Categorical data are given in percentage, whereas quantitative data are demonstrated as mean. Concordance and correlation among the tested anti-ER monoclonal antibodies were accessed by calculating Kappa factor, Pearson’s correlation coefficient, Spearman’s correlation coefficient, or intraclass correlation coefficient. Staining patterns of each antibody were compared by using two statistical tests: paired t-test, a parametric test, and Wilcoxon test, a rank-based non-parametric test. Results We herein investigated the ER expression profile in 61 paraffinembedded primary BC of the following histological types: invasive ductal carcinoma (n = 44, 72.13%), in situ ductal carcinoma (n = 8, 13.11%), invasive lobular carcinoma (n = 4, 6.56%), medullary carcinoma (n = 2, 3.29), microinvasive ductal carcinoma (n = 1, 1.64%), mucinous carcinoma (n = 1, 1.64%), and in situ papillary carcinoma (n = 1, 1.64%) (Table 1). ER expression in BC cells was analyzed by immunohistochemistry using the anti-ER MMAb 1D5 or the anti-ER RMAb SP1. Taken as a whole, we noted a high concordance rate of 96.7% between the referred antibodies. Moreover, we calculated the Kappa factor of 0.921, indicating that 1D5 and SP1 provided overlapping ER expression results. Indeed, results were controversial only in 2 samples studied, which were ER-negative when stained with 1D5 and ER-positive when assessed with SP1 (one sample with 13% and the other with 56% ER-positive cells). Fig. 1 presents one discordant case. Finally, considering only the ER-negative BC cases, we found a higher prevalence of them when treated with MMAb 1D5 (n = 19, 31.15%) compared to RMAb SP1 (n = 17, 27.87%). As it is important to determine the ER status during BC diagnosis, we further investigated the accuracy of ER expression evaluated with anti-ER MMAb 1D5 or anti-ER RMAb SP1 by stratifying the ER-positive BC according to the PS, IS, or TS/A-score. Our analysis revealed no statistically significant difference between the mean PS obtained with the two antibodies of interest, either by analyzing the entire study population (mean PS with 1D5 and SP1 of 68.35 and 69.42, respectively; p-value using t-Student test 0.2329, and pvalue applying Wilcoxon test 0.3750) or by classifying it according to the tumor histological type (invasive ductal carcinoma: mean PS with 1D5 and SP1 of 66.48 and 67.76, respectively; p-value using t-Student test 0.2479, and p-value applying Wilcoxon test 0.4217, and in situ ductal carcinoma: mean PS with 1D5 and SP1 of 73.60 and 74.20, respectively; p-value using t-Student test 0.8850, and pvalue applying Wilcoxon test 1.000) (Table 3). To better understand the correlation between the PS obtained with the anti-ER antibodies 1D5 and SP1, we plotted our experimental data and designed a theoretical curve representing the total concordance between the PS of the antibodies. The calculated Pearson correlation coefficient of 0.7351 and an intraclass correlation coefficient of 0.6193 point to a regular to good correlation between the results. Graphically, we Table 1 Histological classification of the primary BC included in the present study. Primary BC histological type
n (%)
Invasive ductal carcinoma In situ ductal carcinoma Invasive lobular carcinoma Medullary carcinoma Microinvasive ductal carcinoma Mucinous carcinoma In situ pappilary carcinoma
44 (72.13%) 8 (13.11%) 4 (6.56%) 2 (3.29%) 1 (1.64%) 1 (1.64%) 1 (1.64%)
Fig. 2. (A) Comparison between the PS obtained with the anti-ER RMAb SP1 and MMAb 1D5 toward a theoretical curve of total concordance. (B) Linear regression of the PS obtained with each antibody and the lack of concordance calculated as the difference between PS. Pearson correlation coefficient: 0.7351; intraclass correlation coefficient: 0.6193.
could observe that, for higher values of PS, our experimental data are closer to the theoretical curve than for lower values of PS. This means that samples with higher PS values present a better concordance between the ER status evaluated with 1D5 and SP1 than samples with lower PS values (Fig. 2A). Furthermore, examining the difference between the PS results obtained with 1D5 and SP1, which is a statistical approach to estimating the lack of concordance between the PS values, and the PS for these antibodies, we observed that the lack of concordance is most commonly observed among BC samples fitted within the lower PS range obtained with the MMAb 1D5 (Fig. 2B). Exploring our data with regard to the IS values of the ER-positive BC cases, we noted a higher proportion of strongly stained samples (score 3 of 40.98%) obtained by the analysis pursued with the anti-ER RMAb SP1 when compared to that of the anti-ER MMAb 1D5 (24.59%) (Table 2). As this is still unexplored information in the literature, we decided to calculate the Spearman’s correlation coefficient , aiming to establish the conformity between the antibodies. Considering the entire population studied, we observed good correlation between SP1 and 1D5 ( = 0.9150). Nonetheless, the mean IS values obtained with the anti-ER antibodies were statistically different within the BC population as a whole (mean IS with 1D5 and SP1 of 2.33 and 2.45, respectively; p-value using tStudent test <0.0001, and p-value applying Wilcoxon test 0.0003), and within the invasive ductal carcinoma cases (mean IS with 1D5 and SP1 of 2.03 and 2.36, respectively; p-value using t-Student test <0.0001, and p-value applying Wilcoxon test 0.0022). On the other hand, no statistical difference was noted between the mean IS values obtained with the two antibodies within the in situ ductal carcinoma cases (mean IS with 1D5 and SP1 of 2.20 and 2.80, respectively; p-value using t-Student test 0.0704, and p-value applying Wilcoxon test 0.1489) (Table 3).
660
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
Table 2 Evaluation of the ER status of BC cases following the Allred score. Primary BC cases (n = 61) Proportion score (PS) 0 2 3 4 5 Intensity score (IS) 0 1 2 3 Total (TS)/A-score (PS + IS) 0 2 3 4 5 6 7 8
RMAb anti-ER SP1 n (%)
MMAb anti-ER 1D5 n (%)
17 (27.87) 0 3 (4.92) 15 (24.59) 26 (42.62)
19 (31.15) 1 (1.64) 3 (4.92) 11 (18.03) 27 (44.26)
17 (27.87) 5 (8.20) 14 (22.95) 25 (40.98)
19 (31.15) 11 (18.03) 16 (26.23) 15 (24.59)
17 (27.87) 0 0 2 (3.28) 4 (6.56) 10 (16.39) 5 (8.20) 23 (37.70)
19 (31.15) 0 1 (1.64) 2 (3.28) 6 (9.84) 7 (11.48) 12 (19.67) 14 (22.95)
Fig. 3. IS obtained with the anti-ER RMAb SP1 and MMAb 1D5 among the 17 BC cases with divergent ER status assessment. Spearman’s correlation coefficient: 0.9150 (for the entire sample) and 0.7345 (for subjects which were positive for both antibodies).
PS: Score 0 (negative), 1 (<1%), 2 (1–10%), 3 (10–33.3%), 4 (33.3–66.6%), and 5 (>66.6%). PI: Score 0 (negative), 1 (weak), 2 (intermediate), 3 (strong). TS/A-score: 0 (negative), 2–8.
Last but not least, we analyzed the ER status exclusively within the ER-positive BC cases stained with both the anti-ER antibodies MMAb 1D5 and RMAb SP1 (n = 42). Initially, we calculated the Spearman’s correlation coefficient in a statistical strategy to establish the conformity between the patterns provided by the two antibodies. Intriguingly, the anti-ER antibodies of interest were less concordant ( = 0.7345) compared to the entire population studied ( = 0.9150). In fact, among the 42 ER-positive primary BC cases, conflicting data were observed in 17 samples (40.48%). Of remarkable interest, in all 17 cases, the anti-ER MMAb 1D5 provided weaker ER staining when compared to the anti-ER RMAb SP1 (Fig. 3). In addition, we calculated the TS/A-score in the same ER-positive BC population (n = 42). Taking into account only the ER-positive cases with both the studied antibodies which ranked as TS/A-score 8 (this is more than 66.6% of ER-positive cells within the tumor mass with high expression level of the protein), we observed that 54.76% (n = 23) and 33.33% (n = 14) of the cases fitted into this category following the use of the anti-ER RMAb SP1 and MMAb 1D5, respectively. We then concluded that 47.62% (n = 20) were differentially classified, considering the TS/A-score, following the analysis of the ER status with the antibodies. Out of these, only 7.14% (n = 3) of the cases were ranked into a higher TS/A-score with 1D5 than with SP1 antibody (Fig. 4). Finally, we noted that the mean
Fig. 4. TS/A-score obtained with the anti-ER RMAb SP1 and MMAb 1D5 among the 20 BC cases with divergent ER status assessment. Spearman’s correlation coefficient: 0.9238 (for the entire sample) and 0.7743 (for subjects which were positive for both antibodies).
TS/A-score difference between the antibodies was statistically significant for the entire population studied, as well as for the invasive ductal carcinomas but not for in situ ductal carcinomas. Discussion The determination of steroid hormone receptors status, ER and PR, in BC samples represents one of the most important predictive
Table 3 Mean parameters of the Allred score.
Primary BC cases (n = 61) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score Invasive ductal carcinoma (n = 44) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score In situ ductal carcinoma (n = 8) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score * **
p-Value: t-Student test. p-Value: Wilcoxon test.
Anti-ER MMAb 1D5
Anti-ER RMAb SP1
p-Value*
p-Value**
42 68.35 2.33 6.64
44 69.42 2.45 6.98
0.2329 <0.0001 0.0007
0.3750 0.0003 0.0025
31 66.48 2.03 6.52
33 67.76 2.36 6.82
0.2479 <0.0001 0.0060
0.4217 0.0022 0.0121
5 73.60 2.20 7.00
5 74.20 2.80 7.40
0.8850 0.0704 0.1777
1.0000 0.1489 0.3458
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
parameters during the disease diagnosis, and its clinical relevance is irrefutable. In the present study, we investigated the ER alpha status of primary BC cases using two distinct anti-ERs, RMAb SP1 and MMAb 1D5, both clinically used in BC diagnosis routine. As expected, we observed a good statistical concordance between the results obtained with the two antibodies (Kappa factor = 0.921). On the other hand, the anti-ER RMAb SP1 seemed to be more efficient in identifying ER alpha-positive BC cases than the anti-ER MMAb 1D5. This could be due, at least partly, to the higher affinity of RMAb toward the epitope against which it was generated when compared to MMAb [4]. In this respect, it was established that the affinity of the RMAb SP1 is 8-fold that of the MMAb 1D5 for each of their epitopes [3]. It is also interesting that the two antibodies were raised against distinct epitopes within the ER alpha sequence: whereas 1D5 was developed to recognize a sequence within N-terminal domain of ER alpha, SP1 recognizes the ER alpha C-terminal portion [6,17]. Our data is in agreement with the findings of Cheang et al. (2006), who have also documented higher efficiency of the RMAb SP1 in identifying ER-positive BC cases than the MMAb 1D5 [5]. It is worth emphasizing the clinical impact of the conflicting ER status obtained with the ER antibodies SP1 or 1D5 in two BC cases of the population studied (n = 2/61; 3.28%). In this context, women would be considered eligible for endocrine therapy if their biopsy was analyzed with the anti-ER RMAb SP1, but ineligible for the treatment if the ER status of the cancerous cells was assessed with the anti-ER MMAb 1D5. This is particularly important if the epidemiological scenario of BC is undertaken in view of the disease as the main cause of cancer-related deaths among women worldwide. Also of interest, according to the American Cancer Society, the estimated incidence of BC in the US for the years 2011 and 2012 was calculated to reach approximately 231,000 and 226,000 new cases, respectively, suggesting a decreased-inclination of around 5000 new cases this year. In contrast, the expected BC-related death in the US is calculated to affect approximately 39,000 patients in 2011 and 2012; pointing to a constant probability that patients die of BC despite the lower incidence of new cases [2]. Considering the advances in BC diagnosis techniques, it is reasonable to speculate that BC statistics reflects, at least partly, inefficient strategies for the treatment of the disease. Moreover, it is of remarkable importance to extrapolate our findings to the global US BC statistics: a misdiagnosed ER status rate of 3.28% among new BC cases would represent approximately 7577 and 7413 cases in 2011 and 2012, respectively. These would, ultimately, be excluded from the endocrine therapy benefits. As initially expected, both anti-ER provided equivalent mean PS values, which corroborates our findings related to the concordance of the overall results obtained with RMAb SP1 and MMAb 1D5. Moreover, our data is in agreement with the observations of Cheang et al. (2006), who reported similar percentage rates of ER-positive cells in BC samples analyzed with the two anti-ER antibodies [5]. Intriguingly, however, a more detailed appreciation of our results has revealed that there is no linearity of the equivalence of the PS values obtained with both the antibodies of interest throughout the population studied. In fact, when we pursued to correlate the individual PS values of each sample obtained with the different antibodies through the confrontation of our experimental data with a theoretical curve representing a complete concordance between the reagents, we noted good concordance among the cases scored within high PS values with the two antibodies, but there was lack of concordance among the BC cases scored within low PS range with the MMAb 1D5. Again, our data indicate a better accuracy of the investigation of ER status in BC specimens with RMAb SP1 than with MMAb 1D5. To confirm that it is important to analyze both the percentage of cancer cells expressing ER (PS), as well as the intensity of the protein
661
expression (IS) in order to accurately determine the ER status in BC samples, we studied the IS values of the ER-positive BC cases. We noted a higher proportion of strongly stained samples (40.98% of the cases with IS score 3) obtained with the analysis pursued with the anti-ER RMAb SP1 when compared to that of the anti-ER MMAb 1D5 (24.59% of the cases with IS score 3). Moreover, the mean IS values obtained with the anti-ER antibodies were statistically different within the BC population as a whole, and within the invasive ductal carcinoma cases (that correspond to 65–85% of all BC cases). On the other hand, no statistical difference was noted between the mean IS values obtained with the two antibodies within the in situ ductal carcinoma cases (that is the main non-invasive type of BC). Despite the fact that only two of the BC cases studied were excluded from endocrine therapy due to inaccurate determination of their ER status, and that all other ER-positive cases were treated with tamoxifen or anastrozole and remain disease-free, we recommend the use of the anti-ER RMAb SP1 due to high probability that the BC ER status is determined accurately as the reagent provides higher IS, and hence a higher A-score than MMAb 1D5. Ultimately, higher IS and A-score decrease the possibility of ER status misinterpretation and, consequently, inappropriate BC treatment that would compromise the patient’s quality of life and overall survival. References [1] D.C. Allred, et al., Prognostic and predictive factors in breast cancer by immunohistochemical analysis, Mod. Pathol. 11 (1998) 155–168. [2] American Society of Clinical Oncology, Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer, J. Clin. Oncol. 14 (1996) 2843–2877. [3] J.E. Brock, J.L. Hornick, A.L. Richardson, et al., A comparison of estrogen receptor SP1 and 1D5 monoclonal antibodies in routine clinical use reveals similar staining results, Am. J. Clin. Pathol. 132 (2009) 396–401. [4] J.C. Bystryn, J.S. Jacobsen, et al., Comparison of cell–surface human melanomaassociated antigens identified by rabbit and murine antibodies, Hybridoma 1 (4) (1982) 465–472. [5] M.C. Cheang, D.O. Treaba, C.H. Speers, et al., Immunohistochemical detection using the new rabbit mono-clonal antibody SP1 of estrogen receptor in breast cancer is superior to mouse monoclonal antibody 1D5 in predicting survival, J. Clin. Oncol. 24 (2006) 5626–5628. [6] E.R. Fisher, S. Anderson, S. Dean, et al., Solving the dilemma of the immunohistochemical and other methods used for scoring estrogen receptor and progesterone receptor in patients with invasive breast carcinoma, Cancer 103 (2005) 164–173. [7] A. Goldhirsch, J.H. Glick, R.D. Gelber, et al., Meeting highlights: international expert consensus on the primary therapy of early breast cancer, Ann. Oncol. 16 (2005) 1569–1583. ˜ [8] M.A. Guisado, A.S.A. Munoz, M.C.L. Garrido, et al., Initialization of adjuvant hormonal treatment for breast cancer, Adv. Ther. 28 (2011) 66–84. [9] M.E. Hammond, D.F. Hayes, A.C. Wolff, et al., American society of clinical oncology/college of american pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer, J. Oncol. Pract. 6 (2010) 195–197. [10] Z. Huang, W. Zhu, Y. Meng, et al., Development of new rabbit monoclonal antibody to progesterone receptor (clone SP2): no heat pretreatment but effective for paraffin section immunohistochemistry, Appl. Immuno-histochem. Mol. Morphol. 14 (2006) 229–233. [11] P.A. Kaplan, S.R. Frazier, T.S. Loy, et al., 1D5 and 6F11: an immunohistochemical comparison of two monoclonal antibodies for the evaluation of estrogen receptor status in primary breast carcinoma, Am. J. Clin. Pathol. 123 (2) (2005) 276–280. [12] J. Kononen, L. Bubendorf, A. Kallioniemi, et al., Tissue microarrays for highthroughput molecular profiling of tumor specimens, Nat. Med. 4 (1998) 844–847. [13] Y.W. Moon, S. Park, J.H. Sohn, et al., Clinical significance of progesterone receptor and HER2 status in estrogen receptor-positive, operable breast cancer with adjuvant tamoxifen, J. Cancer Res. Clin. Oncol. 137 (7) (2011) 1123–1130. [14] L.B. Rangel, R. Agarwal, T. D’Souza, et al., Tight junction proteins claudin-3 and claudin-4 are frequently overexpressed in ovarian cancer but not in ovarian cystadenomas, Clin. Cancer Res. 9 (2003) 2567–2575. [15] L.B. Rangel, C.A. Sherman-Baust, R.P. Wernyj, et al., Characterization of novel human ovarian cancer-specific transcripts (HOSTs) identified by serial analysis of gene expression, Oncogene 22 (46) (2003) 7225–7232. [16] L.B. Rangel, R. Agarwal, C.A. Sherman-Baust, et al., Anomalous expression of the HLA-DR alpha and beta chains in ovarian and other cancers, Cancer Biol. Ther. 3 (2004) 1021–1027. [17] S. Rossi, L. Laurino, A. Furlanetto, et al., A comparative study between a novel category of immunoreagents and the corresponding mouse monoclonal antibodies, Am. J. Clin. Pathol. 124 (2005) 295–302.
Contents lists available at SciVerse ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Original Article
Comparison of immunohistochemical analysis with estrogen receptor SP1 and 1D5 monoclonal antibodies in breast cancer Klesia Pirola Madeira a , Renata Dalmaschio Daltoé a,b , Gabriela Modenesi Sirtoli a , Lucas Cunha Dias de Rezende a , Alex Assis de Carvalho c , Isabella dos Santos Guimarães a , Ian Victor Silva a,d , Leticia Batista Azevedo Rangel a,e,∗ a
Biotechnology Program/RENORBIO, Health Sciences Center, Federal University of Espirito Santo, Brazil Department of Pharmacy and Nutrition, Federal University of Espirito Santo, Brazil c Department of Pathology, Health Sciences Center, Federal University of Espirito Santo, Brazil d Department of Morphology, Health Sciences Center, Federal University of Espirito Santo, Brazil e Department of Pharmaceutical Sciences, Health Sciences Center, Federal University of Espirito Santo, Brazil b
a r t i c l e
i n f o
Article history: Received 9 April 2012 Received in revised form 19 July 2012 Accepted 30 July 2012 Keywords: Estrogen receptor alpha Monoclonal antibodies Breast cancer
s u m m a r y In the present study, we aimed to evaluate estrogen receptor ER-alpha status in 61 breast cancer cases using Sp1 and 1D5 monoclonal antibodies. Tissue array platforms were generated containing samples of breast cancer and positive controls that were assayed by immunohistochemistry applying monoclonal primary antibodies anti-ER alpha, SP1 and 1D5. We noted a high concordance rate (96.7%) between the referred antibodies. Moreover, we calculated the Kappa factor (0.921), indicating that 1D5 and SP1 provided overlapping ER␣ expression results. Indeed, we observed controversial results only in 2 samples studied, which were ER-negative when stained with 1D5 and ER-positive when assessed with SP1. Total concordance of PS was obtained (Pearson and intraclass CF, 0.7351 and 0.6193, respectively). However, concordance between the antibodies seems to be more accurate in higher PS values. An excellent IS correlation between antibodies was observed throughout the population (Spearman’s CF, = 0.9150). Following the Allred score, 17 out of 42 positive BC samples diverged, with 1D5 always pointing to weaker staining than SP1. When calculating Spearman’s CF of Total Score (TS) within the population, an excellent correlation between both the antibodies ( = 0.9238) was noted. Nonetheless, the results were less concordant among the BC-positive cases ( = 0.7743). Indeed, 20 samples were differentially classified using the antibodies (only 3 had higher TS with 1D5). Considering the mean TS of all samples or of invasive ductal carcinoma, SP1 provided higher scores than 1D5 (p < 0.05). We recommend the use of the anti-ER RMAb SP1 due to the high probability that the BC ER␣ status can be determined accurately as the reagent provides higher IS. Therefore, the A-score was higher than the MMAb 1D5. Ultimately, higher IS and A-score decrease the possibility of ER␣ status misinterpretation and, consequently, inappropriate BC treatment that would compromise the patient’s quality of life and overall survival. We recommend the use of anti-ER RMAb SP1 due to the high probability that the BC ER status can be determined accurately as the reagent provides higher IS, therefore higher A-score, than the MMAb 1D5. © 2012 Elsevier GmbH. All rights reserved.
Introduction An accurate pathological analysis of tissue biopsies is crucial for a conclusive and precise disease diagnosis, as well as for establishing its prognosis and therapeutic guidelines. In this context, the evaluation of cellular markers in breast cancer (BC) samples has proven imperative in the determination of disease diagnosis and
∗ Corresponding author at: Avenida Marechal Campos, 1468, Maruípe, Vitória, Espírito Santo, Brasil. Departamento de Ciências Farmacêuticas, 2◦ Andar, Sala 8, CEP: 29043-900, Brazil. Tel.: +55 27 33357539. E-mail address: [email protected] (L.B.A. Rangel). 0344-0338/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.prp.2012.07.010
prognosis, the patients’ eligibility for targeted therapies, or the use of endocrine and monoclonal antibody therapies for treatment [13]. The precise assessment of the expression levels of the estrogen (ER) and progesterone (PR) receptors by BC cells is an important procedure during BC diagnosis. Phenotypic analysis correlates with disease outcome and endocrine therapy responsiveness (mainly regarding postoperative adjuvant systemic therapy of patients with invasive BCs and BC recurrences) [7,9]. Positive tumors usually correlate with better disease prognosis than the negative ones because the former is entitled to the endocrine therapy with selective modulators of ER, such as tamoxifene, aromatase inhibitors, and anastrozole [8,13].
658
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
Materials and methods Samples We analyzed 61 BC samples, surgically resected in the two public oncology reference hospitals in the state of Espírito Santo (Brazil) Cassiano Antonio de Moraes Hospital (HUCAM) and Santa Rita de Cássia Hospital (HSRC). All cases were revised by a pathologist, and the BC diagnosis was confirmed. Cohort definition Primary BC cases registered at HUCAM from 1997 to 2007 and at HSRC from 2001 to 2007 were included in the present study. All living patients gave written informed consent before enrollment in the analysis. This work was approved by the Human Research Ethics Committee of the Federal University of Espírito Santo. TMA platforms and ER-alpha immunohistochemistry assays
Fig. 1. This figure presents one discordant case. (A) Invasive Ductal Carcinoma (anti ER-alpha 1D5 immunohistochemical stain, 400×). (B) Invasive Ductal Carcinoma (anti ER-alpha SP1 imunohistochemical stain, 400×).
Therefore, using immunohistochemistry (IHC) for the determination of BC hormone receptor status is clinically invaluable because information obtained by this method is considered specific, sensitive, and affordable [5,10]. In this respect, both anti-ER mouse (MMAbs) and rabbit (RMAbs) monoclonal antibodies have been used clinically. Whereas two different MMAbs, 6F11 and 1D5, seem to provide equivalent results in BC biopsies [14], several authors have investigated the efficacy of these antibodies in BC samples. Kaplan and colleagues (2005) showed that the two antiER MMAbs, 6F11 and 1D5, provided equivalent results with respect to immunochemistry assays [11]. On the other hand, some authors have also compared the accuracy of the anti-ER MMAb 1D5 with the anti-ER RMAbs SP1, leading to controversial data. In 2006, Cheang et al. published a pioneer study comparing the accuracy of the ER antibodies, 1D5 and SP1, by immunohistochemistry, revealing an eight-fold higher affinity of the latter to ER in comparison to the former [5]. Nonetheless, Brock et al. (2009) concluded that the antiER 1D5 and SP1 are equivalent in immunohistochemistry assays [3]. Despite these conflicting results, many authors have pointed to the benefits of using RMAbs due to the their high affinity against the epitope of interest, therefore providing precise immunohistochemistry analysis due to the intense positive staining and low background staining. In the present study, ER-alpha status was evaluated in 61 BC cases using Sp1 and 1D5 monoclonal antibodies. The ER staining results of immunohistochemistry were compared by assessing the percentage of cells stained within the analyzed sample, as well as their staining intensity by applying the Allred score system.
Hematoxylin and eosin stained sections from each breast sample were reviewed by a pathologist to confirm the primary BC diagnosis, and to select one representative 2 mm area of the tumor embedded in a paraffin block for immunohistochemical analysis. Each 2 mm sample was further transferred to a transient tissue array platform, generating customized BC tissue microarray (TMA) platforms [12,14–16]. The immunohistochemistry assay followed a protocol optimized by our group in 5 m sample sections of the generated breast cancer TMA platforms [14–16]. Briefly, after removal of paraffin, sections were immersed in preheated antigen-retrieval solution (Tris/EDTA 1X, pH = 9), incubated at 95–99 ◦ C for 30 min, and then allowed to cool down to room temperature for 30 min. Then, sections were incubated for 3 h with the primary antibodies of interest: MMAb anti-human ER-alpha, clone 1D5 (DakoCytomation, Carpinteria CA), dilution 1:100 and RMAb anti-human ER-alpha, clone SP1 (Spring Bioscience, Pleasanton, CA), dilution 1:100. Sections were incubated with biotinylated universal secondary antibody (Dako Cytomation LSAB+ System-HRP, Dako Cytomation, Carpinteria CA) for 30 min. Endogenous peroxidase activity was blocked by 5 min of incubation in 3% (v/v) hydrogen peroxide. Antigen–antibody complexes were detected by the avidin–biotin–peroxidase method, using 3,3-diaminobenzidine (DAB) as chromogenic substrate. The sections were incubated with streptavidin conjugated to peroxidase (LSAB+ System-HRP, Dako Cytomation, Carpinteria CA) for 30 min and then with DAB (Liquid DAB+ Substrat Chromogen System, Dako Cytomation, Dako Cytomation, Carpinteria CA) for 5 min. Slides were counterstained with hematoxylin and immersed in 5% (p/v) ammonium hydroxide. Positive controls were included in the BC TMA platform in opposite positions of the chip in order to guarantee an accurate analysis of protein expression in breast tissue. Negative control experiments were conducted in the absence of the primary antibodies listed above, in parallel of all assays. Cases with less than 1% nuclear staining in tumor cells were considered negative for the expression of the protein of interest. Subcellular localization was also noted. The scoring system proposed for the analysis of the 4 TMA platforms was the Allred score. This method categorizes the samples based on a proportion score (PS) that classifies the cases into 6 groups according to the percentage of stained cells, and on an intensity score (IS) that divides them into 4 classes according to staining intensity. The PS groups are as follows: Score 0: negative; Score 1: <1%; Score 2: 1–10%; Score 3: 10–33.3%; Score 4: 33.3–66.6%; and Score 5: >66.6%. The IS classes are similar: Score 0: negative; Score 1: weak; Score 2: intermediate; and Score 3: strong. Finally, the two scores are combined, and the total score (TS) is given, considering
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
659
an A-score of 0 or 2–8 [1]. The IHC reactions were interpreted by a pathologist with expertise in breast pathology. Statistical analysis Categorical data are given in percentage, whereas quantitative data are demonstrated as mean. Concordance and correlation among the tested anti-ER monoclonal antibodies were accessed by calculating Kappa factor, Pearson’s correlation coefficient, Spearman’s correlation coefficient, or intraclass correlation coefficient. Staining patterns of each antibody were compared by using two statistical tests: paired t-test, a parametric test, and Wilcoxon test, a rank-based non-parametric test. Results We herein investigated the ER expression profile in 61 paraffinembedded primary BC of the following histological types: invasive ductal carcinoma (n = 44, 72.13%), in situ ductal carcinoma (n = 8, 13.11%), invasive lobular carcinoma (n = 4, 6.56%), medullary carcinoma (n = 2, 3.29), microinvasive ductal carcinoma (n = 1, 1.64%), mucinous carcinoma (n = 1, 1.64%), and in situ papillary carcinoma (n = 1, 1.64%) (Table 1). ER expression in BC cells was analyzed by immunohistochemistry using the anti-ER MMAb 1D5 or the anti-ER RMAb SP1. Taken as a whole, we noted a high concordance rate of 96.7% between the referred antibodies. Moreover, we calculated the Kappa factor of 0.921, indicating that 1D5 and SP1 provided overlapping ER expression results. Indeed, results were controversial only in 2 samples studied, which were ER-negative when stained with 1D5 and ER-positive when assessed with SP1 (one sample with 13% and the other with 56% ER-positive cells). Fig. 1 presents one discordant case. Finally, considering only the ER-negative BC cases, we found a higher prevalence of them when treated with MMAb 1D5 (n = 19, 31.15%) compared to RMAb SP1 (n = 17, 27.87%). As it is important to determine the ER status during BC diagnosis, we further investigated the accuracy of ER expression evaluated with anti-ER MMAb 1D5 or anti-ER RMAb SP1 by stratifying the ER-positive BC according to the PS, IS, or TS/A-score. Our analysis revealed no statistically significant difference between the mean PS obtained with the two antibodies of interest, either by analyzing the entire study population (mean PS with 1D5 and SP1 of 68.35 and 69.42, respectively; p-value using t-Student test 0.2329, and pvalue applying Wilcoxon test 0.3750) or by classifying it according to the tumor histological type (invasive ductal carcinoma: mean PS with 1D5 and SP1 of 66.48 and 67.76, respectively; p-value using t-Student test 0.2479, and p-value applying Wilcoxon test 0.4217, and in situ ductal carcinoma: mean PS with 1D5 and SP1 of 73.60 and 74.20, respectively; p-value using t-Student test 0.8850, and pvalue applying Wilcoxon test 1.000) (Table 3). To better understand the correlation between the PS obtained with the anti-ER antibodies 1D5 and SP1, we plotted our experimental data and designed a theoretical curve representing the total concordance between the PS of the antibodies. The calculated Pearson correlation coefficient of 0.7351 and an intraclass correlation coefficient of 0.6193 point to a regular to good correlation between the results. Graphically, we Table 1 Histological classification of the primary BC included in the present study. Primary BC histological type
n (%)
Invasive ductal carcinoma In situ ductal carcinoma Invasive lobular carcinoma Medullary carcinoma Microinvasive ductal carcinoma Mucinous carcinoma In situ pappilary carcinoma
44 (72.13%) 8 (13.11%) 4 (6.56%) 2 (3.29%) 1 (1.64%) 1 (1.64%) 1 (1.64%)
Fig. 2. (A) Comparison between the PS obtained with the anti-ER RMAb SP1 and MMAb 1D5 toward a theoretical curve of total concordance. (B) Linear regression of the PS obtained with each antibody and the lack of concordance calculated as the difference between PS. Pearson correlation coefficient: 0.7351; intraclass correlation coefficient: 0.6193.
could observe that, for higher values of PS, our experimental data are closer to the theoretical curve than for lower values of PS. This means that samples with higher PS values present a better concordance between the ER status evaluated with 1D5 and SP1 than samples with lower PS values (Fig. 2A). Furthermore, examining the difference between the PS results obtained with 1D5 and SP1, which is a statistical approach to estimating the lack of concordance between the PS values, and the PS for these antibodies, we observed that the lack of concordance is most commonly observed among BC samples fitted within the lower PS range obtained with the MMAb 1D5 (Fig. 2B). Exploring our data with regard to the IS values of the ER-positive BC cases, we noted a higher proportion of strongly stained samples (score 3 of 40.98%) obtained by the analysis pursued with the anti-ER RMAb SP1 when compared to that of the anti-ER MMAb 1D5 (24.59%) (Table 2). As this is still unexplored information in the literature, we decided to calculate the Spearman’s correlation coefficient , aiming to establish the conformity between the antibodies. Considering the entire population studied, we observed good correlation between SP1 and 1D5 ( = 0.9150). Nonetheless, the mean IS values obtained with the anti-ER antibodies were statistically different within the BC population as a whole (mean IS with 1D5 and SP1 of 2.33 and 2.45, respectively; p-value using tStudent test <0.0001, and p-value applying Wilcoxon test 0.0003), and within the invasive ductal carcinoma cases (mean IS with 1D5 and SP1 of 2.03 and 2.36, respectively; p-value using t-Student test <0.0001, and p-value applying Wilcoxon test 0.0022). On the other hand, no statistical difference was noted between the mean IS values obtained with the two antibodies within the in situ ductal carcinoma cases (mean IS with 1D5 and SP1 of 2.20 and 2.80, respectively; p-value using t-Student test 0.0704, and p-value applying Wilcoxon test 0.1489) (Table 3).
660
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
Table 2 Evaluation of the ER status of BC cases following the Allred score. Primary BC cases (n = 61) Proportion score (PS) 0 2 3 4 5 Intensity score (IS) 0 1 2 3 Total (TS)/A-score (PS + IS) 0 2 3 4 5 6 7 8
RMAb anti-ER SP1 n (%)
MMAb anti-ER 1D5 n (%)
17 (27.87) 0 3 (4.92) 15 (24.59) 26 (42.62)
19 (31.15) 1 (1.64) 3 (4.92) 11 (18.03) 27 (44.26)
17 (27.87) 5 (8.20) 14 (22.95) 25 (40.98)
19 (31.15) 11 (18.03) 16 (26.23) 15 (24.59)
17 (27.87) 0 0 2 (3.28) 4 (6.56) 10 (16.39) 5 (8.20) 23 (37.70)
19 (31.15) 0 1 (1.64) 2 (3.28) 6 (9.84) 7 (11.48) 12 (19.67) 14 (22.95)
Fig. 3. IS obtained with the anti-ER RMAb SP1 and MMAb 1D5 among the 17 BC cases with divergent ER status assessment. Spearman’s correlation coefficient: 0.9150 (for the entire sample) and 0.7345 (for subjects which were positive for both antibodies).
PS: Score 0 (negative), 1 (<1%), 2 (1–10%), 3 (10–33.3%), 4 (33.3–66.6%), and 5 (>66.6%). PI: Score 0 (negative), 1 (weak), 2 (intermediate), 3 (strong). TS/A-score: 0 (negative), 2–8.
Last but not least, we analyzed the ER status exclusively within the ER-positive BC cases stained with both the anti-ER antibodies MMAb 1D5 and RMAb SP1 (n = 42). Initially, we calculated the Spearman’s correlation coefficient in a statistical strategy to establish the conformity between the patterns provided by the two antibodies. Intriguingly, the anti-ER antibodies of interest were less concordant ( = 0.7345) compared to the entire population studied ( = 0.9150). In fact, among the 42 ER-positive primary BC cases, conflicting data were observed in 17 samples (40.48%). Of remarkable interest, in all 17 cases, the anti-ER MMAb 1D5 provided weaker ER staining when compared to the anti-ER RMAb SP1 (Fig. 3). In addition, we calculated the TS/A-score in the same ER-positive BC population (n = 42). Taking into account only the ER-positive cases with both the studied antibodies which ranked as TS/A-score 8 (this is more than 66.6% of ER-positive cells within the tumor mass with high expression level of the protein), we observed that 54.76% (n = 23) and 33.33% (n = 14) of the cases fitted into this category following the use of the anti-ER RMAb SP1 and MMAb 1D5, respectively. We then concluded that 47.62% (n = 20) were differentially classified, considering the TS/A-score, following the analysis of the ER status with the antibodies. Out of these, only 7.14% (n = 3) of the cases were ranked into a higher TS/A-score with 1D5 than with SP1 antibody (Fig. 4). Finally, we noted that the mean
Fig. 4. TS/A-score obtained with the anti-ER RMAb SP1 and MMAb 1D5 among the 20 BC cases with divergent ER status assessment. Spearman’s correlation coefficient: 0.9238 (for the entire sample) and 0.7743 (for subjects which were positive for both antibodies).
TS/A-score difference between the antibodies was statistically significant for the entire population studied, as well as for the invasive ductal carcinomas but not for in situ ductal carcinomas. Discussion The determination of steroid hormone receptors status, ER and PR, in BC samples represents one of the most important predictive
Table 3 Mean parameters of the Allred score.
Primary BC cases (n = 61) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score Invasive ductal carcinoma (n = 44) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score In situ ductal carcinoma (n = 8) ER-positive cases Mean percentage of positive cells Mean IS Mean TS/A-score * **
p-Value: t-Student test. p-Value: Wilcoxon test.
Anti-ER MMAb 1D5
Anti-ER RMAb SP1
p-Value*
p-Value**
42 68.35 2.33 6.64
44 69.42 2.45 6.98
0.2329 <0.0001 0.0007
0.3750 0.0003 0.0025
31 66.48 2.03 6.52
33 67.76 2.36 6.82
0.2479 <0.0001 0.0060
0.4217 0.0022 0.0121
5 73.60 2.20 7.00
5 74.20 2.80 7.40
0.8850 0.0704 0.1777
1.0000 0.1489 0.3458
K.P. Madeira et al. / Pathology – Research and Practice 208 (2012) 657–661
parameters during the disease diagnosis, and its clinical relevance is irrefutable. In the present study, we investigated the ER alpha status of primary BC cases using two distinct anti-ERs, RMAb SP1 and MMAb 1D5, both clinically used in BC diagnosis routine. As expected, we observed a good statistical concordance between the results obtained with the two antibodies (Kappa factor = 0.921). On the other hand, the anti-ER RMAb SP1 seemed to be more efficient in identifying ER alpha-positive BC cases than the anti-ER MMAb 1D5. This could be due, at least partly, to the higher affinity of RMAb toward the epitope against which it was generated when compared to MMAb [4]. In this respect, it was established that the affinity of the RMAb SP1 is 8-fold that of the MMAb 1D5 for each of their epitopes [3]. It is also interesting that the two antibodies were raised against distinct epitopes within the ER alpha sequence: whereas 1D5 was developed to recognize a sequence within N-terminal domain of ER alpha, SP1 recognizes the ER alpha C-terminal portion [6,17]. Our data is in agreement with the findings of Cheang et al. (2006), who have also documented higher efficiency of the RMAb SP1 in identifying ER-positive BC cases than the MMAb 1D5 [5]. It is worth emphasizing the clinical impact of the conflicting ER status obtained with the ER antibodies SP1 or 1D5 in two BC cases of the population studied (n = 2/61; 3.28%). In this context, women would be considered eligible for endocrine therapy if their biopsy was analyzed with the anti-ER RMAb SP1, but ineligible for the treatment if the ER status of the cancerous cells was assessed with the anti-ER MMAb 1D5. This is particularly important if the epidemiological scenario of BC is undertaken in view of the disease as the main cause of cancer-related deaths among women worldwide. Also of interest, according to the American Cancer Society, the estimated incidence of BC in the US for the years 2011 and 2012 was calculated to reach approximately 231,000 and 226,000 new cases, respectively, suggesting a decreased-inclination of around 5000 new cases this year. In contrast, the expected BC-related death in the US is calculated to affect approximately 39,000 patients in 2011 and 2012; pointing to a constant probability that patients die of BC despite the lower incidence of new cases [2]. Considering the advances in BC diagnosis techniques, it is reasonable to speculate that BC statistics reflects, at least partly, inefficient strategies for the treatment of the disease. Moreover, it is of remarkable importance to extrapolate our findings to the global US BC statistics: a misdiagnosed ER status rate of 3.28% among new BC cases would represent approximately 7577 and 7413 cases in 2011 and 2012, respectively. These would, ultimately, be excluded from the endocrine therapy benefits. As initially expected, both anti-ER provided equivalent mean PS values, which corroborates our findings related to the concordance of the overall results obtained with RMAb SP1 and MMAb 1D5. Moreover, our data is in agreement with the observations of Cheang et al. (2006), who reported similar percentage rates of ER-positive cells in BC samples analyzed with the two anti-ER antibodies [5]. Intriguingly, however, a more detailed appreciation of our results has revealed that there is no linearity of the equivalence of the PS values obtained with both the antibodies of interest throughout the population studied. In fact, when we pursued to correlate the individual PS values of each sample obtained with the different antibodies through the confrontation of our experimental data with a theoretical curve representing a complete concordance between the reagents, we noted good concordance among the cases scored within high PS values with the two antibodies, but there was lack of concordance among the BC cases scored within low PS range with the MMAb 1D5. Again, our data indicate a better accuracy of the investigation of ER status in BC specimens with RMAb SP1 than with MMAb 1D5. To confirm that it is important to analyze both the percentage of cancer cells expressing ER (PS), as well as the intensity of the protein
661
expression (IS) in order to accurately determine the ER status in BC samples, we studied the IS values of the ER-positive BC cases. We noted a higher proportion of strongly stained samples (40.98% of the cases with IS score 3) obtained with the analysis pursued with the anti-ER RMAb SP1 when compared to that of the anti-ER MMAb 1D5 (24.59% of the cases with IS score 3). Moreover, the mean IS values obtained with the anti-ER antibodies were statistically different within the BC population as a whole, and within the invasive ductal carcinoma cases (that correspond to 65–85% of all BC cases). On the other hand, no statistical difference was noted between the mean IS values obtained with the two antibodies within the in situ ductal carcinoma cases (that is the main non-invasive type of BC). Despite the fact that only two of the BC cases studied were excluded from endocrine therapy due to inaccurate determination of their ER status, and that all other ER-positive cases were treated with tamoxifen or anastrozole and remain disease-free, we recommend the use of the anti-ER RMAb SP1 due to high probability that the BC ER status is determined accurately as the reagent provides higher IS, and hence a higher A-score than MMAb 1D5. Ultimately, higher IS and A-score decrease the possibility of ER status misinterpretation and, consequently, inappropriate BC treatment that would compromise the patient’s quality of life and overall survival. References [1] D.C. Allred, et al., Prognostic and predictive factors in breast cancer by immunohistochemical analysis, Mod. Pathol. 11 (1998) 155–168. [2] American Society of Clinical Oncology, Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer, J. Clin. Oncol. 14 (1996) 2843–2877. [3] J.E. Brock, J.L. Hornick, A.L. Richardson, et al., A comparison of estrogen receptor SP1 and 1D5 monoclonal antibodies in routine clinical use reveals similar staining results, Am. J. Clin. Pathol. 132 (2009) 396–401. [4] J.C. Bystryn, J.S. Jacobsen, et al., Comparison of cell–surface human melanomaassociated antigens identified by rabbit and murine antibodies, Hybridoma 1 (4) (1982) 465–472. [5] M.C. Cheang, D.O. Treaba, C.H. Speers, et al., Immunohistochemical detection using the new rabbit mono-clonal antibody SP1 of estrogen receptor in breast cancer is superior to mouse monoclonal antibody 1D5 in predicting survival, J. Clin. Oncol. 24 (2006) 5626–5628. [6] E.R. Fisher, S. Anderson, S. Dean, et al., Solving the dilemma of the immunohistochemical and other methods used for scoring estrogen receptor and progesterone receptor in patients with invasive breast carcinoma, Cancer 103 (2005) 164–173. [7] A. Goldhirsch, J.H. Glick, R.D. Gelber, et al., Meeting highlights: international expert consensus on the primary therapy of early breast cancer, Ann. Oncol. 16 (2005) 1569–1583. ˜ [8] M.A. Guisado, A.S.A. Munoz, M.C.L. Garrido, et al., Initialization of adjuvant hormonal treatment for breast cancer, Adv. Ther. 28 (2011) 66–84. [9] M.E. Hammond, D.F. Hayes, A.C. Wolff, et al., American society of clinical oncology/college of american pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer, J. Oncol. Pract. 6 (2010) 195–197. [10] Z. Huang, W. Zhu, Y. Meng, et al., Development of new rabbit monoclonal antibody to progesterone receptor (clone SP2): no heat pretreatment but effective for paraffin section immunohistochemistry, Appl. Immuno-histochem. Mol. Morphol. 14 (2006) 229–233. [11] P.A. Kaplan, S.R. Frazier, T.S. Loy, et al., 1D5 and 6F11: an immunohistochemical comparison of two monoclonal antibodies for the evaluation of estrogen receptor status in primary breast carcinoma, Am. J. Clin. Pathol. 123 (2) (2005) 276–280. [12] J. Kononen, L. Bubendorf, A. Kallioniemi, et al., Tissue microarrays for highthroughput molecular profiling of tumor specimens, Nat. Med. 4 (1998) 844–847. [13] Y.W. Moon, S. Park, J.H. Sohn, et al., Clinical significance of progesterone receptor and HER2 status in estrogen receptor-positive, operable breast cancer with adjuvant tamoxifen, J. Cancer Res. Clin. Oncol. 137 (7) (2011) 1123–1130. [14] L.B. Rangel, R. Agarwal, T. D’Souza, et al., Tight junction proteins claudin-3 and claudin-4 are frequently overexpressed in ovarian cancer but not in ovarian cystadenomas, Clin. Cancer Res. 9 (2003) 2567–2575. [15] L.B. Rangel, C.A. Sherman-Baust, R.P. Wernyj, et al., Characterization of novel human ovarian cancer-specific transcripts (HOSTs) identified by serial analysis of gene expression, Oncogene 22 (46) (2003) 7225–7232. [16] L.B. Rangel, R. Agarwal, C.A. Sherman-Baust, et al., Anomalous expression of the HLA-DR alpha and beta chains in ovarian and other cancers, Cancer Biol. Ther. 3 (2004) 1021–1027. [17] S. Rossi, L. Laurino, A. Furlanetto, et al., A comparative study between a novel category of immunoreagents and the corresponding mouse monoclonal antibodies, Am. J. Clin. Pathol. 124 (2005) 295–302.