CME/SAM. Abstract. Anatomic Pathology / Ki-67 and p53 Expression in Breast Cancer

Transcription

1 Anatomic Pathology / Ki-67 and p53 Expression in Breast Cancer CME/SAM Hormone Receptor Status Rather Than HER2 Status Is Significantly Associated With Increased Ki-67 and p53 Expression in Triple-Negative Breast Carcinomas, and High Expression of Ki-67 but Not p53 Is Significantly Associated With Axillary Nodal Metastasis in Triple- Negative and High-Grade Non Triple-Negative Breast Carcinomas Jeong S. Han, MD, PhD, 1 Dengfeng Cao, MD, PhD, 2 Kyle H. Molberg, MD, 1 Venetia R. Sarode, MD, 1 Roshni Rao, MD, 3 Lisa M. Sutton, MD, 1 and Yan Peng, MD, PhD 1 Key Words: Ki-67; p53; Triple-negative; Breast cancer; Axillary nodal metastasis DOI: 1.139/AJCP9DV3EVZUATFV Upon completion of this activity you will be able to: describe the status of prognostic markers Ki-67 and p53 in triplenegative breast carcinomas vs high-grade non triple-negative breast carcinomas. discuss the difference between hormone receptor status and HER2 status in association with Ki-67 and p53 expression in triple-negative breast cancer. describe association of estrogen receptor with Ki-67 and p53 expression in high-grade breast carcinomas. define the potential prognostic value of Ki-67 in predicting axillary nodal metastasis in patients with high-grade breast cancer. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 311. Exam is located at Abstract Triple-negative (TN) breast carcinoma is associated with a higher recurrence rate and shorter survival and lacks the benefit of specific therapy. TN tumors usually express high levels of Ki-67 and p53 that are considered prognostic markers for breast cancer. We compared Ki-67 and p53 expression between TN and high-grade non-tn invasive carcinomas in a total of 214 cases and investigated an association between their expression and axillary nodal metastasis in these tumors. Our findings demonstrate that TN tumors are associated with significantly higher expression of Ki-67 and p53 compared with non-tn tumors, which may contribute to the poorer prognosis in TN tumors. Hormone receptor negativity rather than HER2 negativity is associated with the significantly increased Ki-67 and p53 expression in TN tumors. Furthermore, a high expression level of Ki-67 but not p53 is more likely to be associated with axillary nodal metastasis in these cases. Human breast carcinomas represent a collection of diverse tumors that vary in their natural history and responsiveness to therapy. Previous studies have classified breast carcinomas into different subtypes mainly based on their molecular and immunohistochemical profiles of hormonal receptors and human epithelial growth factor receptor 2 (HER2) overexpression. 1-4 Triple-negative (TN) breast carcinoma is characterized by negativity for estrogen receptor (ER), progesterone receptor (PR), and HER2. TN breast carcinoma is a high-risk tumor that lacks the benefit of specific therapy that targets these proteins. TN breast cancer is associated with a higher histologic grade, shorter survival, and a higher recurrence rate. 5 TN tumors usually express high levels of Ki-67 and p53 that are currently considered prognostic markers for patients with breast cancer. 1,2 Ki-67, proliferation index, is a nonhistone nuclear protein that is closely linked to proliferating cells. High Ki-67 expression has been shown to be associated with a higher histologic grade, larger tumor size, the presence of axillary lymph nodal metastasis, and shorter disease-free and overall survival in patients with breast cancer. 2,3,6-8 In addition, a positive correlation between Ki-67 expression and pathologic tumor response to neoadjuvant chemotherapy has been reported, implicating its role in response to treatment. 9 Some researchers used meta-analysis to unify various results of previous studies; they identified 524 genes that were significantly 23 Am J Clin Pathol 211;135: Downloaded 23 from DOI: 1.139/AJCP9DV3EVZUATFV

2 Anatomic Pathology / Original Article associated with breast cancer survival and found that 7% of these genes were strongly correlated with proliferation. 1 This meta-analysis highlighted the important role of proliferation in breast cancer prognosis. The p53 gene encodes for a 53-kDa nuclear phosphoprotein, which has been implicated in controlling cell-cycle regulation, differentiation, DNA repair, and apoptosis Unlike normal p53, nonfunctional mutated p53 accumulates in the nucleus of tumor cells, and, therefore, it can be detected by immunohistochemical analysis. Multiple studies have shown that p53 overexpression in breast cancer is associated with a worse outcome. 11 Recent studies suggested that p53 status might have a different predictive value for the efficacy of anthracycline/alkylating agent based chemotherapy regimen between TN and non-tn breast cancers. 14 The aims of our study were to compare Ki-67 and p53 expression between TN (histologic grade II or grade III tumors) and high-grade (grade II or grade III) non-tn invasive breast carcinomas, respectively, and to investigate an association between their expression and axillary nodal metastasis with an attempt of exploring a cutoff value of these markers in predicting nodal metastasis in these cases. We also further analyzed differential expression of Ki-67 and p53 in subgroups of non-tn tumors (ER+/PR±/HER2, ER+/PR±/ HER2+, and ER /PR /HER2+) to determine which biomarker (hormone receptors vs HER2) was correlated with Ki-67 and p53 overexpression. Materials and Methods This study was approved by the institutional review board. A total of 214 cases of high-grade carcinoma of the breast from the surgical pathology files of the University of Texas Southwestern Medical Center, Dallas, and Parkland Memorial Hospital, Dallas, between 23 and 29 were included in this study. Clinical information was obtained from electronic medical records. Clinical characteristics and immunohistochemical tumor profiles were analyzed in 81 TN tumors and 133 non-tn tumors. Of the 133 non-tn tumors, 18 (13.5%) were ER /PR /HER2+, 18 (13.5%) were ER+/ PR±/HER2+, and 97 (72.9%) were ER+/PR±/HER2. The expression of Ki-67 and p53 was compared between the TN and non-tn groups. In addition, we compared the Ki-67 and p53 expression between the TN tumors with nodal metastasis (32/81 [4%]) and the non-tn tumors with nodal metastasis (63/133 [47.4%]). Subsequently, the TN and non-tn tumors were combined as a whole (n = 214) and then divided into 2 subgroups based on the percentage of Ki-67+ and p53+ tumor cells (high Ki-67 or high p53 group, >1%; low Ki-67 or low p53 group, 1%), and ER expression and the percentage of positive axillary lymph nodes were compared between the 2 subgroups as well. Immunohistochemical Assay and Image Quantitation Methods We used quantitative immunohistochemical analysis to determine ER, PR, HER2, Ki-67, and p53 expression in the breast tumors. Immunohistochemical staining was performed using a TechMate 1 automated immunostainer (Ventana Medical Systems, Tucson, AZ). A known positive control section was included in each run to ensure proper staining. Rabbit immunoglobulin fraction (normal) or nonspecific IgG1 monoclonal diluted with phosphate-buffered saline was used as a negative control. Monoclonal antibodies were used for ER (1D5, prediluted, Ventana Medical Systems), PR (PgR, prediluted, Ventana Medical Systems), HER2 (dilution 1:1,2; DakoCytomation, Carpinteria, CA), Ki-67 (MIB-1, prediluted, Ventana Medical Systems), and p53 (DO-7, 1:16,; DAKO). Quantitative information for staining was obtained using the automated microscopy method, Automated Cellular Imaging System (ACIS, Clarient, San Juan Capistrano, CA). The ACIS system consisted of an automated robotic bright-field microscope module, a computer, and a Windows NT based software interface. The robotic microscope module scanned the immunohistochemically stained slides, and the computer monitor displayed the digitalized tissue images. After viewing the high-magnification images on the ACIS computer, several subregions of the digitalized tissue images were selected for analysis by the ACIS. To assess the level of tissue ER and PR expression, the ACIS provided the percentage of positively stained cells for ER and PR in the selected subregions. We used the manufacture s guidelines for the ACIS to determine tissue ER, PR, and HER2 expression. When the percentage of cells staining positive for ER and PR was 5% or more, the tumor was considered positive for ER and PR expression. Ki-67 and p53 were considered as high expression when positive cells were more than 1% and as low expression when positive cells were 1% or less. To assess HER2 overexpression, the ACIS provided an average score for 5 selected subregions of the tissue with the highest staining intensity for HER2. Greater than 3% of tumor cells stained with an average intensity score of 2. or more were considered to have HER2 overexpression (3+). A score of more than 1.4 to less than 2. was considered a borderline result (2+, 1%-3% of tumor cells stained), and a score of less than 1.4, negative ( or 1+, <1% of tumor cells stained). All of the results were confirmed by manual review by a pathologist (Y.P.). The criteria for determining triple negativity were based on immunohistochemical analysis and image quantitation of ER, PR, and HER2. Negative results were based on less than 5% positive staining of ER or PR, and staining of or 1+ for HER2. All positive, borderline, or negative HER2 results by immunohistochemical analysis were confirmed by fluorescent in situ hybridization. Positivity of Ki-67 and p53 was determined Downloaded from Am J Clin Pathol 211;135: DOI: 1.139/AJCP9DV3EVZUATFV 231

3 Han et al / Ki-67 and p53 Expression in Breast Cancer by the presence of nuclear staining. Grading was based on the percentage of stained tumor cells. A favorable prognostic category for both biomarkers is considered to be less than or equal to 1% of tumor cells staining positively. HER2 Fluorescent In Situ Hybridization Two of the unstained slides were cut at a 4-μm thickness. Dual-color fluorescent in situ hybridization was performed with the HER2 probe labeled with spectrum red and chromosome 17 specific centromere (D17Z1) probe labeled with spectrum green on sections cut from the same block. Deparaffinization, in situ hybridization, and staining were performed using the PathVysion kit (Abbott-Vysis Laboratories, Abbott Park, IL) per the manufacturer s protocols. Fluorescent signals in at least 6 nonoverlapping interphase nuclei with intact morphology were scored with a 1 objective, using a triple band-pass filter that permits simultaneous blue, green, and red colors. Only tumor cells from the site designated on the H&E-stained slide by the pathologist were scored for the number of red (HER2) and green (chromosome 17) signals. A case was scored as amplified if the ratio of the number of fluorescent signals of HER2 to chromosome 17 was greater than 2.2. Statistical Analysis Tumor size, mean age, and the expression of Ki-67 and p53 were compared between the TN and non-tn groups by 1-way analysis of variance with the Tukey post hoc test. Ki-67 and p53 expression was individually compared with ER and HER2 expression using the Pearson correlation test. Numbers of lymph nodes with metastasis were compared between high Ki-67 and low Ki-67 groups and between the high p53 and low p53 groups using the Fisher exact test. Results Comparison of Ki-67 and p53 Expression Between TN Tumors and Non-TN Tumors Ki-67 expression in the TN group (mean ± SE, 63.7% ± 3.6%) was significantly higher than that in the non-tn group (3.3% ± 2.4%; P <.1; unpaired t test). Similarly, p53 expression in the TN group (5.7% ± 4.8%) was significantly higher than that in the non-tn group (19.3% ± 2.9%; P =.45; unpaired t test). In addition, the TN tumors were significantly larger than the non-tn tumors (3.18 ±.45 cm vs 2.31 ±.21 cm; P <.5; unpaired t test). Subsequently, the non-tn tumors were further classified into 3 subgroups based on their hormone receptor (ER and/or PR) and HER2 status as follows: ER+/PR±/HER2, ER /PR /HER2+, and ER+/PR±/HER2+. Mean age of the patients, tumor size, and Ki-67 and p53 expression in the TN tumor group and the 3 subgroups of non-tn tumors are summarized in Table 1. The TN (ER /PR /HER2 ) and the ER /PR /HER2+ groups seemed to have younger patient populations; however, the mean age difference among the groups did not reach statistical significance (P =.1). A significant difference was observed in tumor size among the groups (P =.35), and the ER+/PR±/HER2 group showed a smaller average tumor size compared with the other groups. Differential expression of Ki-67 and p53 in subgroups is shown as mean percentage ± SEM in Table 1 and in graphs in Figure 1. Statistical significance values between groups (1-way analysis of variance followed by the Tukey post hoc test) are indicated in Figures 1A and 1B. Figure 1A shows that Ki-67 expression in the TN tumors was significantly increased compared with that in the non-tn tumors that were ER+/ PR±/HER2 (P <.1) or ER+/PR±/HER2+ (P <.1). However, there was no significant difference in Ki-67 expression between the TN tumors (ER /PR /HER2 ) and the non- TN tumors that were ER /PR /HER2+. Figure 1B illustrates that p53 expression in the TN tumors was significantly higher compared with the non-tn tumors that were ER+/PR±/ HER2 (P <.1) or ER+/PR±/HER2+ (P <.1), while there was no significant difference in p53 expression between the TN tumors and the non-tn tumors that were ER /PR / HER2+. Moreover, there was no significant difference in Ki-67 and p53 expression between ER+/PR±/HER2 and ER+/PR±/HER2+ subgroups of the non-tn tumors. In addition, Ki-67 expression in the ER /PR /HER2+ subgroup was higher than that in the ER+/PR±/HER2+ subgroup (56.2 ± Table 1 Mean Age, Tumor Size, Ki-67 Activity, and p53 Expression in TN Tumors and Non-TN Tumors in Three Subgroups * Non-TN Tumors TN Tumors ER+/PR±/HER2 (n = 97) ER+/PR±/HER2+ (n = 18) ER /PR /HER2+ (n = 18) ER /PR /HER2 (n = 81) P Age (y) 58.3 ± ± ± ± Tumor size (cm) 1.99 ± ± ± ± Ki-67 (%) ± ± ± ± 3.5 <.1 p53 (%) 9.6 ± ± ± ± 4.83 <.1 ER, estrogen receptor; PR, progesterone receptor; TN, triple negative (for ER, PR, and HER2). * Data are shown as mean ± SEM; P values obtained by 1-way analysis of variance. 232 Am J Clin Pathol 211;135: Downloaded 232 from DOI: 1.139/AJCP9DV3EVZUATFV

4 Anatomic Pathology / Original Article A Ki-67 Expression (%) ER+/PR±/HER2 ER+/PR±/HER2+ ER /PR /HER2+ ER /PR /HER2 B p53 Expression (%) ER+/PR±/HER2 ER+/PR±/HER2+ * ER /PR /HER2+ ER /PR /HER2 Figure 1 Expression of Ki-67 (A) and p53 (B) in triple-negative (TN; negative for estrogen receptor [ER], progesterone receptor [PR], and HER2) tumors and 3 subgroups of non-tn tumors. Data are shown as mean ± SEM. Statistically significant differences are indicated. * P <.5. P <.1. P <.1. One-way analysis of variance followed by the Tukey post hoc test was performed to compare individual groups vs ± 5.24), although this did not reach statistical significance. Expression of p53 in the ER /PR /HER2+ subgroup was significantly higher than that in the ER+/PR±/ HER2+ group (68.2 ± 9.3 vs 25.7 ± 8.2; P <.5). Correlation of Hormone Receptor and HER2 With Ki-67 and p53 Expression The preceding results demonstrated that hormone receptor (ER and/or PR) status rather than HER2 status was significantly associated with increased Ki-67 and p53 expression in the TN tumors. The Pearson correlation test was performed to further investigate the association between hormone receptor and HER2 status and Ki-67 and p53 expression. ER expression showed a significant negative correlation with Ki-67 expression (r =.527; P <.1) and with p53 (r =.534; P <.1). HER2 overexpression showed no significant correlation with Ki-67 expression (r =.33; P =.63) or with p53 expression (r =.78; P =.25). These findings further support our hypothesis that hormone receptor status, but not HER2 status, is closely associated with Ki-67 and p53 expression in TN breast cancers. Comparison of Ki-67 and p53 Expression Between TN Tumors With Nodal Metastasis and Non-TN Tumors With Nodal Metastasis To determine if Ki-67 and p53 expression was associated with axillary nodal metastasis in these carcinomas, we compared the Ki-67 and p53 expression between 32 TN tumors with nodal metastasis and 63 non-tn tumors with nodal metastasis. We found that the Ki-67 and p53 expression in the TN tumors with nodal metastasis was significantly higher than in the non-tn tumors with nodal metastasis (64.% ± 6.% vs 31.9% ± 3.4%; P <.1; 49.% ± 8.% vs 21.% ± 4.2%; P <.1, respectively), results similar to those of the preceding overall comparison between TN tumors and non-tn tumors. Comparison of Percentage of Positive Axillary Nodes and ER Expression of Tumors Between the High and Low Ki-67/p53 Groups To determine a cutoff value of Ki-67 and p53 expression in predicting axillary nodal metastasis in these cases, we combined the TN and non-tn tumors as a whole (n = 214) and then divided them into 2 subgroups based on the percentage of Ki-67+ or p53+ tumor cells: high Ki-67 or p53 group, more than 1%; low Ki-67 or p53 group, 1% or less. The percentages of positive axillary lymph nodes when an axillary dissection was performed were compared between the 2 subgroups. In addition, comparison of ER expression between the 2 subgroups was also conducted to further investigate the relationship between ER and Ki-67/p53 expression. The 44 tumors with low Ki-67 expression level showed significantly higher ER expression compared with the 17 tumors in the high Ki-67 group (69.3% ± 5.6% vs 41.4% ± 3.6%; P =.3; unpaired t test) Figure 2A. The 127 tumors with a low p53 expression level showed significantly higher ER expression compared with the 87 tumors in the high p53 group (65.8% ± 3.7% vs 19.8% ± 4.%; P <.1; unpaired t test) Figure 2B. These results provide further evidence that ER expression is negatively correlated with high Ki-67 and p53 expression. Downloaded from Am J Clin Pathol 211;135: DOI: 1.139/AJCP9DV3EVZUATFV 233

5 Han et al / Ki-67 and p53 Expression in Breast Cancer A 8 * B ER (%) 4 ER (%) High Ki-67 Low Ki-67 High p53 Low p53 Figure 2 Estrogen receptor (ER) expression in tumors with low vs high Ki-67 expression and low vs high p53 expression in triple-negative (TN; negative for ER, progesterone receptor, and HER2) and non-tn tumors. Data are shown as mean ± SEM. A, Expression of ER is significantly higher in the low Ki-67 group (69.3% ± 5.62%; n = 44; unpaired t test; * P =.3) compared with the high Ki-67 group (41.4% ± 3.55%; n = 17). B, Expression of ER is significantly higher in the low p53 group (65.8% ± 3.71%; n = 127; unpaired t test; P <.1) compared with the high p53 group (19.8% ± 3.99%; n = 87). High and low Ki-67 and p53 were defined as follows: high Ki-67, >1%; low Ki-67, 1%; high p53, >1%; and low p53, 1%. The high Ki-67 group included 17 cases. Results of sentinel lymph node biopsy (SLNB) and/or axillary lymph node dissection (ALND) were available for review in 144 of these 17 cases. A total of 1,35 lymph nodes from the 144 cases were examined, and 386 of them were positive for macrometastasis (386/135 [29.6%]). The low Ki-67 group had 44 cases, and results of SLNB and/or ALND were available for review in 36 of them. A total of 32 lymph nodes from the 36 cases were examined, and 49 of them were positive for macrometastasis (49/32 [16.2%]). The percentage of positive lymph nodes in the high Ki-67 group (29.6%) was significantly higher than that in the low Ki-67 group (16.2%). The contingency data were analyzed by using the Fisher exact test (P <.1) Figure 3A. The high p53 expression group had 87 cases, and results of SLNB and/or ALND were available for review in 7 of them. A total of 694 lymph nodes from the 7 cases were examined, and 183 of them were positive for macrometastasis (183/694 [26.4%]). The low p53 group had 127 cases, and results of SLNB and/or ALND were available for review in 11. A total of 913 lymph nodes from the 11 cases were examined, and 252 of them were positive for macrometastasis (252/913 [27.6%]). There was no significant difference in the percentage of positive lymph nodes between the high p53 and low p53 groups (P =.61; Fisher exact test) Figure 3B. Discussion Our study demonstrated that Ki-67 and p53 expression was significantly higher in TN breast cancers compared with high-grade non-tn tumors. This finding suggests that Ki-67 and p53 expression may have a role in a worse prognosis of TN breast cancers. It is conceivable that tumors with a worse outcome may have a higher proliferation index, Ki-67, and higher probability of mutation in tumor suppressor genes, such as p53. Because we specifically excluded low-grade (grade I) invasive carcinomas in this study, the higher expression of Ki-67 and p53 in TN tumors compared with expression in the non-tn tumors was less likely owing to their high histologic grade, which is commonly seen in TN tumors, and was more likely associated with the TN status of these breast cancers. Although previous studies showed that TN tumors had higher Ki-67 and p53 expression than non-tn tumors, 15,16 the authors did not exclude low-grade breast carcinomas in their studies for the comparison. Subsequently, we investigated whether hormone receptor negativity or HER2 negativity was associated with significantly increased Ki-67 and p53 expression in TN tumors. Our results revealed that hormone receptor (ER and/or PR) negativity, but not HER2 negativity, was significantly associated with high Ki-67 and p53 expression in TN tumors. Our analysis of subgroups based on hormone receptor and HER2 status indicated that Ki-67 and p53 expression was negatively 234 Am J Clin Pathol 211;135: Downloaded 234 from DOI: 1.139/AJCP9DV3EVZUATFV

6 Anatomic Pathology / Original Article A 1, * B 8 No. of Lymph Nodes No. of Lymph Nodes High Ki-67 Low Ki-67 High p53 Low p53 High Ki-67 Low Ki-67 LN+ LN High p53 Low p53 LN+ LN Figure 3 Association of Ki-67 and p53 expression with axillary nodal metastasis in triple-negative (TN; negative for estrogen receptor, progesterone receptor, and HER2) and non-tn tumors. A, Percentage of positive lymph nodes (LNs) in the high Ki-67 group (386/1,35 [29.6%]) is significantly higher than that in the low Ki-67 group (49/32 [16.2%]). B, High p53 and low p53 groups do not demonstrate a significant difference in the percentage of positive LNs (26.4% vs 27.6%; P >.5). Black bars, number of LNs with metastasis (LN+); white bars, number of LNs without metastasis (LN ). The contingency data were analyzed by using the Fisher exact test ( * P <.1). correlated with ER expression, but not with HER2 overexpression. Previous studies done in all types of breast carcinomas, including low-grade carcinomas, revealed that Ki-67 expression had a negative correlation with ER expression that supports our findings, but they did not examine the relationship between p53 and ER expression and the relationship between Ki-67/p53 and HER2 overexpression Owing to their higher expression in TN tumors, Ki-67 and p53 may be used as predictors for response to chemotherapeutic agents. Ki-67 overexpression has been implicated in a better response to chemotherapy agents. 9 The posttreatment Ki-67 expression level is currently considered to have significant prognostic and predictive value. 6,2,21 Although no consensus has been established so far, a few specific chemotherapy agents have been shown to be particularly effective in tumors with high Ki-67 expression. 9,22 It has been shown that breast tumors with a p53 mutation were generally ER and were associated with decreased disease-free survival. 11,14 The underlying mechanism that may involve this association is unclear. A few molecules, such as ellipticine, PRIMA-1, RITA, and MIRA-1, have been used for the restoration of p53 function, and some of them demonstrated antitumor activity during in vitro or in vivo experiments. 11 Our findings support the currently proposed molecular classification of breast cancer by gene expression profile, 4,23 in which breast carcinomas are classified into ER+ (luminal A and B) and ER (HER2+ and TN) groups. Different types of tumors were shown to have different tumor biology, different prognoses, and different responses to therapy. ER status may have a more important role than HER2 status in predicting prognosis of TN and non-tn breast cancers. In the luminal subgroup that encompasses tumors that express ER and genes related to activation of the ER pathway, the luminal B subtype of breast cancer expresses lower ER positivity and higher Ki-67 level and has a worse prognosis compared with the luminal A subtype. HER2+ tumors were originally considered to be associated with a worse prognosis in hormone receptor negative tumors. However, recent studies have shown that treatment with trastuzumab appeared to eliminate the disparity in outcomes. 24 The lack of specific therapy in TN tumors highlights the importance of further biologic characterization of these tumors, which will lead to the development of effective treatment strategies. Our findings reinforce the emerging notion that breast cancer should be classified according to its gene expression profile, in order to make accurate predictions about the outcome of the disease and select the optimal treatment for patients with breast cancer. Downloaded from Am J Clin Pathol 211;135: DOI: 1.139/AJCP9DV3EVZUATFV 235

7 Han et al / Ki-67 and p53 Expression in Breast Cancer Our study also indicated that the Ki-67 and p53 expression levels in TN tumors with nodal metastasis were significantly higher than those in non-tn tumors with nodal metastasis. To determine a cutoff value of Ki-67 and p53 expression in predicting axillary nodal metastasis in these cases, we combined the TN and non-tn tumors as a whole and then divided them into 2 subgroups: high Ki-67 or p53 (>1%) and low Ki-67 or p53 ( 1%). We found that the percentage of lymph nodes with metastasis was significantly higher in tumors with a high Ki-67 expression level compared with those with low Ki-67 expression, while there was no significant difference in the frequency of lymph node metastasis between the high p53 and low p53 groups. Some previous studies revealed that a high Ki-67 expression level was associated with more frequent lymph node metastasis in patients with all types of breast cancer, but they did not investigate an association between p53 expression and nodal metastasis. 7,25,26 Our findings demonstrated that a high expression level of Ki-67 (>1%) but not p53, although both are considered prognostic markers for breast cancer, was significantly associated with axillary nodal metastasis in patients with TN and high-grade non-tn tumors, indicating that Ki-67 has a better prognostic value than p53 for these breast carcinomas. Previous studies have demonstrated that TN breast tumors are more frequently associated with lymph node metastasis compared with non- TN tumors. Our results suggest that high Ki-67 expression may contribute to the higher risk for nodal metastasis in TN tumors. The overall prognostic value of p53 appears to be weaker than Ki-67 because a variety of studies have found conflicting data. 8,27,28 The more consistent results regarding the prognostic value of Ki-67 over p53 may be due to its significant association with higher nodal metastasis. To our knowledge, this study is the first to demonstrate that ER negativity rather than HER2 negativity is significantly associated with increased Ki-67 and p53 expression in TN breast carcinomas, and a high expression level of Ki-67, but not of p53, is significantly associated with axillary lymph node metastasis in TN and high-grade non-tn breast carcinomas. Although there appears to be significant prognostic and predictive value to an immunohistochemical assessment of Ki-67 and p53, it is still not recommended for routine clinical practice. This resistance is primarily due to continued unresolved issues regarding these markers. First, the cutoff values for Ki-67 and p53 are variable in different studies, making it difficult to conduct a comparative analysis. Standardized staining methods and image analysis are also required to achieve consistent measurement in different laboratories. Another potential problem is that the p53 overexpression detected by immunohistochemical analysis may not necessarily correlate directly with p53 mutations. Mutations of p53 are heterogeneous, and some of them may not produce a stable protein that is detectable by immunohistochemical analysis. 11 Wild-type p53 may accumulate in some conditions by binding other cellular proteins, producing positive immunohistochemical results. 11 In fact, it has been shown that correlation between immunohistochemical analysis and sequencing of p53 in breast cancers is less than 75% and that the immunohistochemical method has a lower sensitivity and specificity compared with sequence-based analysis. 29 Numerous proteins have been shown to interact with p53 and modify its activity. The p53 signaling pathway is closely linked to other signal transduction pathways such as BRCA1. 3 Although the direct mechanism is not evident at this time, our results suggest potential association between the ER signaling pathway and Ki-67 and p53 regulation. TN breast carcinomas are associated with significantly higher expression of Ki-67 and p53 compared with non- TN tumors, which may contribute to the poorer prognosis in TN tumors. Hormone receptor (ER and/or PR) negativity rather than HER2 negativity is significantly associated with the increased Ki-67 and p53 expression in TN tumors. Furthermore, high expression level of Ki-67 (>1%), but not of p53, although both are considered prognostic markers for breast cancer, is significantly associated with axillary nodal metastasis in patients with TN and high-grade non-tn tumors, indicating that Ki-67 has a better prognostic value than p53 for these breast carcinomas. From the 1 Department of Pathology and 3 Division of Surgical Oncology, The University of Texas Southwestern Medical Center, Dallas; and 2 Department of Pathology and Immunology, Washington University School of Medicine, St Louis, MO. Presented in abstract form at the 98th Annual Meeting of the United States and Canadian Academy of Pathology; Boston, MA; March 7-13, 29; and the 99th Annual Meeting of the United States and Canadian Academy of Pathology; Washington, DC; March 2-26, 21. Address correspondence to Dr Peng: Dept of Pathology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd, Dallas, TX References 1. Hanby AM. Aspects of molecular phenotype and its correlations with breast cancer behaviour and taxonomy. Br J Cancer. 25;92: Nicolini A, Carpi A, Tarro G. Biomolecular markers of breast cancer. Front Biosci. 26;11: Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2;46: Sorlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A. 21;98: Kaplan HG, Malmgren JA. Impact of triple negative phenotype on breast cancer prognosis. Breast J. 28;14: Am J Clin Pathol 211;135: Downloaded 236 from DOI: 1.139/AJCP9DV3EVZUATFV

8 Anatomic Pathology / Original Article 6. Jones RL, Salter J, A Hern R, et al. The prognostic significance of Ki67 before and after neoadjuvant chemotherapy in breast cancer. Breast Cancer Res Treat. 29;116: Gonzalez-Vela MC, Garijo MF, Fernandez F, et al. MIB1 proliferation index in breast infiltrating carcinoma: comparison with other proliferative markers and association with new biological prognostic factors. Histol Histopathol. 21;16: Rakha EA, El-Sayed ME, Green AR, et al. Prognostic markers in triple-negative breast cancer. Cancer. 27;19: Petit T, Wilt M, Velten M, et al. Comparative value of tumour grade, hormonal receptors, Ki-67, HER-2 and topoisomerase II alpha status as predictive markers in breast cancer patients treated with neoadjuvant anthracycline-based chemotherapy. Eur J Cancer. 24;4: Wirapati P, Sotiriou C, Kunkel S, et al. Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Res. 28;1:R65. doi:1.1186/ bcr Lacroix M, Toillon RA, Leclercq G. p53 and breast cancer, an update. Endocr Relat Cancer. 26;13: Lane DP. Cancer: p53, guardian of the genome. Nature. 1992;358: Vogelstein B, Kinzler KW. p53 function and dysfunction. Cell. 1992;7: Bidard FC, Matthieu MC, Chollet P, et al. p53 status and efficacy of primary anthracyclines/alkylating agent based regimen according to breast cancer molecular classes. Ann Oncol. 28;19: Tan DS, Marchio C, Jones RL, et al. Triple negative breast cancer: molecular profiling and prognostic impact in adjuvant anthracycline-treated patients. Breast Cancer Res Treat. 28;111: Rhee J, Han SW, Oh DY, et al. The clinicopathologic characteristics and prognostic significance of triple-negativity in node-negative breast cancer. BMC Cancer. 28;8:37. doi:1.1186/ Molino A, Micciolo R, Turazza M, et al. Ki-67 immunostaining in 322 primary breast cancers: associations with clinical and pathological variables and prognosis. Int J Cancer. 1997;74: Moriki T, Takahashi T, Kataoka H, et al. Proliferation marker MIB-1 correlates well with proliferative activity evaluated by BrdU in breast cancer: an immunohistochemical study including correlation with PCNA, p53, c-erbb-2 and estrogen receptor status. Pathol Int. 1996;46: Spyratos F, Ferrero-Pous M, Trassard M, et al. Correlation between MIB-1 and other proliferation markers: clinical implications of the MIB-1 cutoff value. Cancer. 22;94: Dowsett M, Smith IE, Ebbs SR, et al. Prognostic value of Ki67 expression after short-term presurgical endocrine therapy for primary breast cancer. J Natl Cancer Inst. 27;99: Sharma S, Saboorian HM, Frawley WH, et al. MIB1 labeling index as an indicator of chemoresponse in carcinoma of the breast. Appl Immunohistochem Mol Morphol. 24;12: Tewari M, Krishnamurthy A, Shukla HS. Predictive markers of response to neoadjuvant chemotherapy in breast cancer. Surg Oncol. 28;17: Cianfrocca M, Gradishar W. New molecular classifications of breast cancer. CA Cancer J Clin. 29;59: Dawood S, Broglio K, Buzdar AU, et al. Prognosis of women with metastatic breast cancer by HER2 status and trastuzumab treatment: an institutional-based review [published online ahead of print November 23, 29]. J Clin Oncol. 21;28: Marrelli D, Pinto E, Neri A, et al. Mib-1 proliferation index is an independent predictor of lymph node metastasis in invasive breast cancer: a prospective study on 675 patients. Oncol Rep. 26;15: Susini T, Nori J, Olivieri S, et al. Predicting the status of axillary lymph nodes in breast cancer: a multiparameter approach including axillary ultrasound scanning. Breast. 29;18: Erdem O, Dursun A, Coskun U, et al. The prognostic value of p53 and c-erbb-2 expression, proliferative activity and angiogenesis in node-negative breast carcinoma. Tumori. 25;91: Bosari S, Lee AK, Viale G, et al. Abnormal p53 immunoreactivity and prognosis in node-negative breast carcinomas with long-term follow-up. Virchows Arch A Pathol Anat Histopathol. 1992;421: Norberg T, Lennerstrand J, Inganas M, et al. Comparison between p53 protein measurements using the luminometric immunoassay and immunohistochemistry with detection of p53 gene mutations using cdna sequencing in human breast tumors. Int J Cancer. 1998;79: Hohenstein P, Giles RH. BRCA1: a scaffold for p53 response? Trends Genet. 23;19: Downloaded from Am J Clin Pathol 211;135: DOI: 1.139/AJCP9DV3EVZUATFV 237