Pathology – Research and Practice xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice
jou rn al hom epage: www.elsev ier .com/ locate /prp
riginal Article
xpression of cancer stem cell markers in basal and penta-negativereast carcinomas – A study of a series of triple-negative tumors
iego de Mendonc a Uchôaa,c,∗, Marcia Silveira Graudenza,b,c,idia Maria Callegari-Jacquesd,e, Carolina Rigatti Hartmanna, Bruna Pellini Ferreirab,ariana Fitarelli-Kiehle, Maria Isabel Edelweissa,b,c
Department of Pathology, Hospital de Clínicas de Porto Alegre (HCPA), Porto Alegre, BrazilSchool of Medicine, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, BrazilPost-Graduate Program in Medicine: Medical Sciences, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, BrazilStatistics Department, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, BrazilPost-Graduate Program in Genetics and Molecular Biology, Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, Brazil
r t i c l e i n f o
rticle history:eceived 27 September 2013eceived in revised form 15 January 2014ccepted 10 March 2014
eywords:reast cancerasalriple-negative
a b s t r a c t
Purpose: Breast cancer is a heterogeneous disease. Immunohistochemistry has given rise to triple-negative carcinoma (TNC). Concomitantly, biological origins of neoplasia and its heterogeneity has beenstrongly debated in cancer stem cells (CSC) theme. This study investigates the prevalence of basal (BCC)and penta-negative carcinomas (5NC) in TNC and establishes associations with CSC (CD44CD24).Materials and methods: 94 TNC were tested for CK5/6, HER1, CD44 and CD24, evaluated by a simpleimmunohistochemistry score and correlated with clinicopathological and survival data.Results: BCC had higher tumor grades than 5NC (p = 0.004). CD44 negativity (p = 0.007) and CD44−CD24+
phenotype (p = 0.013) were associated with less vascular invasion amongst TNC. CD44 expression was− −/low
D44D24
associated with BCC (p = 0.007). CD44 CD24 phenotype was associated with 5NC. None of the vari-ables were associated with clinical outcome.Conclusion: BCC and 5NC are closely related tumor subtypes. CD44−CD24−/low phenotype was associ-ated with 5NC and CD44−CD24+ phenotype was associated with vascular invasion. These results requirehistogenetic confirmation in larger studies.
Breast cancer is a heterogeneous disease. Its clinical presenta-ion, biological potential, response to treatment and prognosis caniffer significantly [1–6]. The diagnostic and prognostic factors tra-itionally used in breast cancer have proven insufficient in someases and there is a clear need for more refined diagnostic criteria.
Depending on immunohistochemistry (IH), triple-negativereast cancer (TNC) emerged in the literature due to its defined
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
oor prognosis and lack of important therapeutic modalities, suchs anti-estrogen and anti-HER2 therapies [7–17]. Likewise, in theistogenetic reclassification of breast tumors, basal cell carcinoma
∗ Corresponding author at: Department of Pathology, Hospital de Clínicas de Portolegre, Rua Ramiro Barcelos, 2350, Bom Fim, CEP 90035-903, Porto Alegre, Brazil.el.: +55 51 33598313; fax: +55 51 33598334.
(BCC) also exhibits aggressive biological potential [18–21]. In recentyears, there has been much discussion regarding the need to sepa-rate these two tumor subtypes, although approximately 70–80% ofthem are the same [22–31]. Despite a considerable overlap betweenthese two entities, there is a large component of both tumor groupsthat is unrelated [22,24,25,31,32]. Recent studies in TNC, on asimplified IH panel, have indicated that BCC exhibit a worse prog-nosis than “non-basal TNC”, i.e., penta-negative carcinomas (5NC)[26,31,33–35].
The biological origin of neoplasia through the theory of breastcancer stem cells (CSC) has been highlighted in the litera-ture [36–41]. The prospective identification of CSC through theCD44+CD24−/low phenotype was demonstrated in breast cancer[42,43] and, because it was also considered a marker of poor
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
prognosis and BCC, this CSC phenotype rapidly gained notoriety[38,44–46]. However, in a few studies, TNC and BCC subtypes havebeen compared for the different immunohistochemical expressionof CSC [47–50].
Condition at the end of the studyAlive 46 48.9Deceased 31 33.0NA 17 18.1
Type of material analyzedTMA (surgery and nodulectomy) 71 75.5
FH
ARTICLERP-51180; No. of Pages 8
D. de Mendonc a Uchôa et al. / Patholog
This is an IH study, designed to evaluate clinicopathological dif-erences between BCC and 5NC and to assess CSC (CD44 and CD24)xpression in BCC and 5NC in a series of TNC.
aterials and methods
All TNC were retrieved from the Hospital de Clínicas de Portolegre (HCPA) pathology department database between 2001 and006. The sample included 119 patients, 14 of whom were excludedecause the paraffin blocks were no longer available, nine due to
ittle or no residual tumor in the blocks (core-biopsies) and twoue to inappropriate neoplasia for testing (in situ carcinoma andone marrow metastasis). The final sample consisted of 94 patients.linical, anatomopathological and follow-up data were collected byearching HCPA database or from private medical records and byelephone. This study was approved by Research Ethics Committeef HCPA.
Tissue microarray (TMA) was performed in all cases wherehere was sufficient tumor for extraction with the removal of
3-mm diameter cylinder from each tumor. The IH techniqueas initiated on a Dako automated device (PT Link model) and
hen was processed on an automated Dako Autostainer Link8 (Dako, Carpinteria, CA, USA). CK5/6 Dako antibody (D5/16B4lone, ready to use, retrieval in a high-pH buffer for 20 min at8 ◦C), Zymed HER1 antibody (EGFR) (31g7 clone, ready to use,ith Pepsin Digest-all 3, retrieval for 5 min at 37 ◦C), Cell Mar-
ue CD44 antibody (MRQ-13 clone, 1:200 dilution, retrieval in high-pH buffer for 20 min at 98 ◦C) and Neomarquers CD24ntibody (SN3b clone, 1:50 dilution, retrieval in a low-pH bufferor 20 min at 98 ◦C) were used. The slides were assessed bywo experienced breast pathologists based on a consensus ofiagnostic observations and without prior knowledge of the clin-
copathological characteristics of the studied tumors. All casesere evaluated with respect to the quantity of positive tumor
ells, coding as 0 (0%), 1 (0.1–1%), 2 (1.1–10%), 3 (10.1–33%), 433–66%) and 5 (>66%). TNC were considered BCC when thereas expression of CK5/6 and/or HER1; 5NC were those that wereegative for both markers. For CK5/6, cytoplasmic staining wasaken into account; for HER1, membrane staining was consid-red. Expressions with score ≥1 were considered positive. ForD44 and CD24, membrane and cytoplasmic staining of neoplas-ic cells was considered. CD44 expression was considered positivef score ≥1. For CD24 scores of 4 and 5 were considered posi-ive and scores of 0, 1, 2 and 3 were considered negative or low−/low).
Data is presented as the means and standard deviations
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
r absolute numbers and percentages. The categorical vari-bles were compared using a Chi-squared test, and an exact-value was obtained for the test. Whenever necessary, annalysis of residuals was also performed. The analyses were
performed using SPSS (PASW Statistics 18.0 [2009]) available atwww.spss.com.hk/statistics and WinPEPI [51]. We consideredp-values <0.05 to be statistically significant.
Results
Clinicopathological characteristics are shown in Table 1. Eighty-five percent of TNC were ductal not otherwise specified (NOS). Themajority of the tumors (46%) were grade 3. Tumor size greater than2.0 cm was identified in 50 cases (53%), of which 15 (16%) were
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
larger than 5.0 cm. Some cases (17/94) were received for secondopinion and immunohistochemical examination, but without dataregarding the age of patients. The mean age was 55.4 years, rangingfrom 25 to 81. The mean follow-up time was 71.5 months (2–353).
Among the 94 TNC, 64% (60/94) were identified as BCC and 36%34/94) were identified as 5NC. In the BCC group, 52% (31) wereoncomitantly positive for CK5/6 and HER1. Individually, HER1 washe most common basal marker, with 48 positive cases (80%) versus3 (72%) CK5/6-positive cases (Fig. 1).
asal cell and penta-negative carcinomas and clinicopathologicalariables
There were no significant differences between BCC and 5NC forge (cut-off point on 40 years), type of carcinoma, tumor size andymph-vascular invasion (LVI) or axillary lymph node involvement.owever, tumor grade 3 was significantly more frequent amongCC than 5NC (p = 0.004; Table 2).
urvival analysis for basal and triple-negative carcinoma patients
Fifty percent of the 75 patients (48 BCC and 27 5NC) who coulde evaluated for overall survival were alive at 96 months. BCC andNC survival curves did not differ (p = 0.642; Fig. 2).
Tumor recurrence could be analyzed in 63 patients; 41 belongedo BCC and 22 to 5NC. Tumor recurrence curve was not differentetween these groups (p = 0.712; Fig. 3).
xpression of CD44 and CD24 and clinicopathological variables
We observed CD44 expression in 34/94 TNC (36%) and CD24 in0/94 cases (32%). CD44 staining identified in TNC was predomi-
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
antly membranous and CD24 was expressed in both membranend cytoplasm (Fig. 1). CD44 and CD24 expression and their asso-iation with clinicopathological variables are shown in Table 3. LVIas significantly associated with no CD44 expression (p = 0.007).
PRESSearch and Practice xxx (2014) xxx–xxx 3
CD24 expression did not show any association with the clinico-pathological variables examined in this study.
CD44CD24 phenotype and clinicopathological variables
We observed a difference (Table 4) between CD44CD24 pheno-types in relation to LVI (p = 0.013). A complementary analysis of theChi-squared residuals revealed that CD44−CD24+ was associatedwith LVI (p = 0.011).
Survival analysis according to CD44 and CD24 expression
There was no significant differences in survival either betweenpatients with or without CD44 expression (p = 0.341) as wellas between CD24+ and CD24−/low groups (p = 0.706) and amongCD44CD24 phenotypes (p = 0.401).
Regarding tumor recurrence (63 patients with available data),there was no significant difference between the groups for CD44(p = 0.531) and CD24 expression (p = 0.618) and neither betweenCD44CD24 immunophenotypes (p = 0.684).
CD44 and CD24 and the basal and penta-negative groups
The global expression of CD44 for TNC was 36% (34/94), ofwhich 28 cases were in the BCC group and only six were in the5NC group (p = 0.007). Regarding CD24, positivity was observedin 32% (30/94) and there was no significant difference betweenBCC and 5NC (p = 0.492). The frequencies of CD44CD24 combinedphenotypes differed between BCC and 5NC cases (p = 0.040); thisdifference was mainly due to a higher number of cases negativefor both markers (CD44−CD24−/low phenotype) in the 5NC group(residual p = 0.019). There are also some difference between groupsfor the CD44+CD24−/low class, the p value being near statistical sig-nificance (p = 0.064; Table 5).
Discussion
Basal and penta-negative breast carcinomas
The cut-off point for the basal markers used in our study waslow (>0%), very close or identical to that used in other studies[5,33,35,52–54]. The average frequency for BCC amongst TNC in theabove mentioned studies was 62.4%, very close to that observed inthe present investigation.
There are few published papers in which BCC and 5NC tumorswere discriminated by IH and the respective tumor grades com-pared. In the studies by Sutton and Nofech-Mozes et al., whichused CK5/6 and/or HER1 to discriminate BCC from a group of TNCcases, there was no difference in tumor grade [28,55]. Rakha et al.also failed to detect a difference using CK5/6, CK14, CK17 and/orHER1 to define BCC [26]. In contrast, Carey, Conforti and Cheanggroups, using CK5/6 and/or HER1 to define BCC, demonstrated thatBCC were more frequently classified as histological grade 3 whencompared with 5NC [33,52,53]. Other studies, with different IHpanels also reported difference in tumor grade between BCC and“non-basal TNC” patients [34,56]. We also observed a significantdifference between BCC and 5NC in relation to tumor grade 3, con-firming the distinction between these groups.
Some studies indicated that there is no age difference betweenBCC and 5NC subtypes [26,48,52,53]. However, Cheang and Blowsgroups, in larger samples, observed that BCC occurred more fre-quently than 5NC in patients younger than 40 years [33,35]. In our
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
study, with a smaller sample, we were unable to demonstrate thiseffect.
Regarding axillary lymph node involvement, we observedno difference between BCC and 5NC, like others studies
4 D. de Mendonc a Uchôa et al. / Pathology – Research and Practice xxx (2014) xxx–xxx
amon
[wtidt
p
Fig. 2. Kaplan–Meier survival curve
28,33,48,52,53,55]. However, in some publications, BCC presentedith axillary involvement less frequently than 5NC [26,35]. On
he other hand, Thike et al. observed higher axillary lymph nodenvolvement associated with BCC [56]. We did not observe anyifference between 5NC and BCC groups in regard to histologic type,
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
umor size and LVI which is similar to others [26,28,33,35,48,53,56].Worse overall and disease-free survival for BCC subtype com-
ared to other types of breast carcinomas has been suggested
Fig. 3. Kaplan–Meier disease-free survival curve
g basal and penta-negative tumors.
by several studies. A few of these, using tumor type selectionmethods based on IH, reported significant differences in survivalanalysis between BCC and 5NC. These are large samples stud-ies, with N ranging from 232 to 1645 TNC [26,33,35,56]. In aneven smaller sample (52 BCC and 60 “non-basal TNC”) followed
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
up for slightly over six years, a difference in survival analy-sis between these two groups was also observed [34]. However,Nofech-Mozes followed 132 TNC (116 BCC and 16 5NC) up for 10
ears and did not find a significant difference in survival [28]. Inur study (94 TNC followed up for approximately 6 years), no dif-erence was noted in survival analysis between BCC and 5NC. It isikely that a real difference in survival analysis between BCC andnon-basal TNC” exist, but it is probably small, because statisti-al significance was only obtained in studies with large samples
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
r with a higher proportion of cases in the “5NC arm”. Amonghe clinical variables examined in our study, we did not includedjuvant treatment, which may have influenced our survivalnalysis.
able 4linicopathological parameters according to CD44CD24 phenotypes.
n CD44+CD24−/low CD44
Age (years)
> 40 69 13 13
≤40 8 3 1
Carcinoma type
Ductal 80 17 14
Lobular 2 0 1
Medullarya 1 1 0
Metaplastic 1 0 0
Mixedb 1 0 0
Tumor size
≤2.0 cm 18 6 1
2.0–4.9 cm 38 8 6
≥5.0 cm 15 4 4
Tumor grade
Grade 1 3 1 1
Grade 2 26 2 2
Grade 3 43 13 9
Lymphovascular invasion
Present 21 3 8
Absent 26 10 2
Lymph nodes
Negative 34 9 4
1–3 LNs+ 18 5 3
≥4 LNs+ 22 2 4
a Atypical.b Ducal + lobular.
Breast cancer stem cell markers
The expression of cancer stem cell markers (CD44 and CD24) hasbeen previously reported in TNC and BCC patients [42,45–50,57,58].However, very few studies compare BCC and 5NC groups byimmunohistochemistry CSC phenotype. Honeth et al., identified a
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
significant difference between “basal TNC” and “non-basal TNC”such that CD44 expression and CD44+CD24− phenotype (withoutCD24low and cut-off point >0%) were more frequent in BCC. [47].In another sample of BCC and “non-basal TNC” tumors, CD44+ and
a Global heterogeneity test between tumor types regarding CD44CD24 phenotypb Residual analysis.
D44+CD24−/low phenotype (cut-off point >10%) were both asso-iated with BCC [48]. We observed CD44 expression in 36.1% ofNC (34/94), which is less than the prevalence in the aforemen-ioned studies. Despite using similar cut-off point (greater than%), the percentage of tumors expressing CD44 in the present sam-le was about half that obtained by Honeth et al. (36.1% versus4.7%). This difference may be due to immunohistochemistry anti-odies for the selection of the samples, as well as the use of differentD44 clone and dilutions for the assessment of CSC, which makeshe comparison of results difficult. Regardless of all the variablesnvolved, we observed that BCC exhibited a significantly higherifference in CD44 expression, supporting the suggestion of otheruthors that the hyaluronic acid receptor is indeed a character-stic IH marker of BCC, even when compared with a very similarubtype, such as the 5NC (“non-basal TNC”). Additionally, we iden-ified CD44−CD24−/low profile in 45.7% (43/94) of our TNC seriesnd observed that CD44−CD24−/low immunophenotype was asso-iated with 5NC. Giatromanolaki et al. indicated that CD44−CD24−
henotype was an independent variable of worse prognosis, asell as triple-negative state, in a series of non-selected breast
arcinomas [45]. CD44+CD24−/low profile is the most frequentlybserved CSC immunophenotype both in BCC and in TNC. In ourtudy, CD44+CD24−/low phenotype was more frequent in BCC, buthe difference failed to reach statistical significance. We believehat the difference between our results and those is due to a higherumber of CD44 negative cases in our study. In a recent study, noifferences between BCC and “non-basal TNC” in CD44 and CD24xpression were found and the best prognosis was observed in theD44+CD24−/low phenotype, which was not associated with TNC.
n addition, the authors observed that CD44−CD24+ phenotype washe most frequent among BCC [50]. As well as we have noted, mostf the studies do not show significant differences between BCC andnon-basal TNC” for isolated CD24 expression.
Lymphovascular invasion (LVI) is a poorly investigated variablemongst the clinicopathological parameters of studies involvingSC. In our series of TNC, CD44− tumors exhibited a significantlyaised rate of LVI. Lizak et al. suggested that CD44 expression wasignificantly lower in early carcinomas with positive axillary lymphodes. The authors did not provide TNC data [59]. These findings are
n agreement with the observation that a cell first needs to lose itsdhesion capacity before invading extracellular matrix to reach itsetastatic potential. In the study by Ahmed et al., however, CD44as not related with LVI. However, the authors did not evaluate
possible CD44 phenotype difference between “basal TNC” and
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
non-basal TNC” [50]. Regarding isolated CD24 expression and clin-copathological variables, there are conflicting results. Ahmed et al.dentified lower CD24 expression in carcinomas of higher histo-ogical tumor grades [50]. Ricardo et al. observed no associations
between CD24 and any of the clinicopathological variables [48]. Inour study, there were also no differences in clinical or pathologicalvariables or in the survival analysis.
In a combined analysis of CD44 and CD24 profiles with theclinical variables in our study, the frequency of LVI was also asso-ciated with CD44−CD24+ phenotype. Aigner et al. reported thatCD24 is capable of replacing PSGL-1 in its capacity for plateletand endothelial adhesion, suggesting that CD24+ carcinomas havemore metastatic capacity [60]. Kristiansen et al. reported a CD24+
association with regional lymph node metastases [61]. In TNC,CD44−CD24+ phenotype may be associated with factors involved inthe loss of adhesion capacity to extracellular matrix (CD44−), whilethere may be a rise in metastatic capacity (CD24+).
Regarding survival analysis and breast CSC, non-selected breastcarcinomas have exhibited discrepancies in the few availablestudies using IH as the main research method. CD44+CD24−/low
phenotype was associated with a worse prognosis for TNC in onestudy [49]. In another, CD44−CD24− phenotype was associatedwith a worse prognosis and this finding was also true for TNC [45].Ahmed et al. observed that CD44−CD24+ phenotype was the onlyone associated with poor prognosis. In other studies, CD44CD24profiles were not shown to be different when compared withinbreast cancer subtypes [46,48,57]. In most of these studies, the sur-vival analyses involved breast carcinoma series with no subdivisionof TNC (into “basal TNC” and “non-basal TNC”) for comparison withour results. We observed no differences in disease-free and over-all survival between BCC and 5NC in relation to CD44, CD24 andCD44CD24 phenotype, in an average follow-up time of 5.9 years.There is a lack of uniformity in the prognostic results of CD44CD24phenotypes among the aforementioned studies, suggesting thata mixture of several tumor subtypes (luminal, HER2, BCC, 5NC),defined by heterogeneous diagnostic criteria and in different sam-ples, may be confounding the results. In addition, CD44+CD24−/low
CSC phenotype is more common in a subtype (BCC or TNC) of lowerprevalence among breast cancers. Considering this, there are alsono experimental studies demonstrating a specific CSC immunophe-notype for luminal or HER2 carcinoma breast tumors.
BCC and 5NC are closely related tumors. Despite this, our resultsrevealed differences between these tumor types, as defined bya simple IH panel. The tumor grade 3, which is a characteristicfeature of carcinomas with poor prognoses, was associated withBCC. In our study, LVI was associated with CD44 negativity andCD44−CD24+ immunophenotype, which merits investigation byother studies with larger samples. Although not statistically signif-
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
icant, BCC in our sample also showed less LVI than 5NC. In our caseseries, we identified higher CD44 expression in BCC that, perhaps, isthe most important basal CSC marker. CD44−CD24−/low phenotypewas associated with the 5NC. Despite being unable to demonstrate
difference between BCC and 5NC in CD44+CD24−/low phenotype,t may indeed exist and thus further studies with larger samples
ay prove useful to enlighten this question.
onflict of interest
The authors declare no competing interest.
thical approval
This study was approved by Research Ethics Committee of Hos-ital de Clínicas de Porto Alegre (HCPA), under protocol # 08-446.
cknowledgements
Financial support for this research and publications by UFRGS-PGCM and FIPE/HCPA.
eferences
[1] C.M. Perou, T. Sorlie, M.B. Eisen, et al., Molecular portraits of human breasttumours, Nature 406 (2000) 747–752.
[2] T. Sorlie, C.M. Perou, R. Tibshirani, et al., Gene expression patterns of breastcarcinomas distinguish tumor subclasses with clinical implications, Proc. Natl.Acad. Sci. U. S. A. 98 (2001) 10869–10874.
[3] T. Sorlie, R. Tibshirani, J. Parker, et al., Repeated observation of breast tumorsubtypes in independent gene expression data sets, Proc. Natl. Acad. Sci. U. S.A. 100 (2003) 8418–8423.
[4] G. Viale, Histopathology of primary breast cancer, Breast 12 (2003) 391–396.[5] T.O. Nielsen, F.D. Hsu, K. Jensen, et al., Immunohistochemical and clinical char-
acterization of the basal-like subtype of invasive breast carcinoma, Clin. CancerRes. 10 (2004) 5367–5374.
[6] C.M. Perou, S.S. Jeffrey, M. van de Rijn, et al., Distinctive gene expression pat-terns in human mammary epithelial cells and breast cancers, Proc. Natl. Acad.Sci. U. S. A. 96 (1999) 9212–9217.
[7] E.A. Rakha, M.E. El-Sayed, A.R. Green, et al., Prognostic markers in triple-negative breast cancer, Cancer 109 (2007) 25–32.
[8] S. Cleator, W. Heller, R.C. Coombes, Triple-negative breast cancer: therapeuticoptions, Lancet Oncol. 8 (2007) 235–244.
10] W.J. Irvin Jr., L.A. Carey, What is triple-negative breast cancer? Eur. J. Cancer 44(2008) 2799–2805.
11] S.J. Dawson, E. Provenzano, C. Caldas, Triple negative breast cancers: clinicaland prognostic implications, Eur. J. Cancer 45 (Suppl. 1) (2009) 27–40.
12] W.D. Foulkes, I.E. Smith, J.S. Reis-Filho, Triple-negative breast cancer, N. Engl.J. Med. 363 (2010) 1938–1948.
13] R. Dent, M. Trudeau, K.I. Pritchard, et al., Triple-negative breast cancer:clinical features and patterns of recurrence, Clin. Cancer Res. 13 (2007)4429–4434.
14] H. Mersin, E. Yildirim, U. Berberoglu, et al., The prognostic importance of triplenegative breast carcinoma, Breast 17 (2008) 341–346.
15] G. Viale, L. Bottiglieri, Pathological definition of triple negative breast cancer,Eur. J. Cancer 45 (Suppl. 1) (2009) 5–10.
16] Y. Sasaki, H. Tsuda, Clinicopathological characteristics of triple-negative breastcancers, Breast Cancer 16 (2009) 254–259.
17] J.A. Billar, A.C. Dueck, C.C. Stucky, et al., Triple-negative breast cancers: uniqueclinical presentations and outcomes, Ann. Surg. Oncol. 17 (Suppl. 3) (2010)384–390.
18] E.A. Rakha, D.A. El-Rehim, C. Paish, et al., Basal phenotype identifies a poorprognostic subgroup of breast cancer of clinical importance, Eur. J. Cancer 42(2006) 3149–3156.
19] A.J. Evans, E.A. Rakha, S.E. Pinder, et al., Basal phenotype: a powerful prognosticfactor in small screen-detected invasive breast cancer with long-term follow-up, J. Med. Screen. 14 (2007) 210–214.
20] E.A. Rakha, J.S. Reis-Filho, I.O. Ellis, Basal-like breast cancer: a critical review, J.Clin. Oncol. 26 (2008) 2568–2581.
21] A.A. Luck, A.J. Evans, A.R. Green, et al., The influence of basal phenotype onthe metastatic pattern of breast cancer, Clin. Oncol. (R. Coll. Radiol.) 20 (2008)40–45.
22] E.A. Rakha, D.S. Tan, W.D. Foulkes, et al., Are triple-negative tumours and basal-like breast cancer synonymous? Breast Cancer Res. 9 (2007) 404, author reply405.
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
23] B. Kreike, M. van Kouwenhove, H. Horlings, et al., Gene expression profiling andhistopathological characterization of triple-negative/basal-like breast carcino-mas, Breast Cancer Res. 9 (2007) R65.
24] F. Bertucci, P. Finetti, N. Cervera, et al., How basal are triple-negative breastcancers? Int. J. Cancer 123 (2008) 236–240.
26] E.A. Rakha, S.E. Elsheikh, M.A. Aleskandarany, et al., Triple-negative breast can-cer: distinguishing between basal and nonbasal subtypes, Clin. Cancer Res. 15(2009) 2302–2310.
27] E. Lerma, A. Barnadas, J. Prat, Triple negative breast carcinomas: similarities anddifferences with basal like carcinomas, Appl. Immunohistochem. Mol. Morphol.17 (2009) 483–494.
28] S. Nofech-Mozes, M. Trudeau, H.K. Kahn, et al., Patterns of recurrence in thebasal and non-basal subtypes of triple-negative breast cancers, Breast CancerRes. Treat. 118 (2009) 131–137.
29] N. Kuroda, M. Ohara, K. Inoue, et al., The majority of triple-negative breast can-cer may correspond to basal-like carcinoma, but triple-negative breast canceris not identical to basal-like carcinoma, Med. Mol. Morphol. 42 (2009) 128–131.
30] A.A. Thike, P.Y. Cheok, A.R. Jara-Lazaro, et al., Triple-negative breast cancer: clin-icopathological characteristics and relationship with basal-like breast cancer,Mod. Pathol. 23 (2010) 123–133.
31] S. Badve, D.J. Dabbs, S.J. Schnitt, et al., Basal-like and triple-negative breastcancers: a critical review with an emphasis on the implications for pathologistsand oncologists, Mod. Pathol. 24 (2011) 157–167.
32] E.A. Rakha, M.E. El-Sayed, J. Reis-Filho, et al., Patho-biological aspectsof basal-like breast cancer, Breast Cancer Res. Treat. 113 (2009) 411–422.
33] M.C. Cheang, D. Voduc, C. Bajdik, et al., Basal-like breast cancer defined by fivebiomarkers has superior prognostic value than triple-negative phenotype, Clin.Cancer Res. 14 (2008) 1368–1376.
34] Z.B. Liu, J. Wu, B. Ping, et al., Basal cytokeratin expression in relation to immuno-histochemical and clinical characterization in breast cancer patients with triplenegative phenotype, Tumor 95 (2009) 53–62.
35] F.M. Blows, K.E. Driver, M.K. Schmidt, et al., Subtyping of breast cancer byimmunohistochemistry to investigate a relationship between subtype andshort and long term survival: a collaborative analysis of data for 10, 159 casesfrom 12 studies, PLoS Med. 7 (2010) e1000279.
36] D. Ponti, N. Zaffaroni, C. Capelli, et al., Breast cancer stem cells: an overview,Eur. J. Cancer 42 (2006) 1219–1224.
37] C.T. Jordan, M.L. Guzman, M. Noble, Cancer stem cells, N. Engl. J. Med. 355 (2006)1253–1261.
38] R. Liu, X. Wang, G.Y. Chen, et al., The prognostic role of a gene signa-ture from tumorigenic breast-cancer cells, N. Engl. J. Med. 356 (2007) 217–226.
39] M. Cariati, A.D. Purushotham, Stem cells and breast cancer, Histopathology 52(2008) 99–107.
40] B.J. Morrison, C.W. Schmidt, S.R. Lakhani, et al., Breast cancer stem cells: impli-cations for therapy of breast cancer, Breast Cancer Res. 10 (2008) 210.
41] K. Kai, Y. Arima, T. Kamiya, et al., Breast cancer stem cells, Breast Cancer 17(2009) 80–85.
42] M. Al-Hajj, M.S. Wicha, A. Benito-Hernandez, et al., Prospective identificationof tumorigenic breast cancer cells, Proc. Natl. Acad. Sci. U. S. A. 100 (2003)3983–3988.
43] C. Sheridan, H. Kishimoto, R.K. Fuchs, et al., CD44+/CD24− breast cancer cellsexhibit enhanced invasive properties: an early step necessary for metastasis,Breast Cancer Res. 8 (2006) R59.
44] E. Mylona, I. Giannopoulou, E. Fasomytakis, et al., The clinicopathologic andprognostic significance of CD44+/CD24(−/low) and CD44−/CD24+ tumor cells ininvasive breast carcinomas, Hum. Pathol. 39 (2008) 1096–1102.
45] A. Giatromanolaki, E. Sivridis, A. Fiska, et al., The CD44+/CD24− phenotyperelates to ‘triple-negative’ state and unfavorable prognosis in breast cancerpatients, Med. Oncol. 28 (2011) 745–752.
46] H.J. Kim, M.J. Kim, S.H. Ahn, et al., Different prognostic significance of CD24 andCD44 expression in breast cancer according to hormone receptor status, Breast20 (2011) 78–85.
47] G. Honeth, P.O. Bendahl, M. Ringner, et al., The CD44+/CD24− phenotype isenriched in basal-like breast tumors, Breast Cancer Res. 10 (2008) R53.
48] S. Ricardo, A.F. Vieira, R. Gerhard, et al., Breast cancer stem cell markers CD44,CD24 and ALDH1: expression distribution within intrinsic molecular subtype,J. Clin. Pathol. 64 (2011) 937–946.
49] M.O. Idowu, M. Kmieciak, C. Dumur, et al., CD44(+)/CD24(−/low) cancerstem/progenitor cells are more abundant in triple-negative invasive breast car-cinoma phenotype and are associated with poor outcome, Hum. Pathol. 43(2012) 364–373.
50] M.A. Ahmed, M.A. Aleskandarany, E.A. Rakha, et al., A CD44(−)/CD24(+) pheno-type is a poor prognostic marker in early invasive breast cancer, Breast CancerRes. Treat. 133 (2012) 979–995.
52] L.A. Carey, C.M. Perou, C.A. Livasy, et al., Race, breast cancer subtypes,and survival in the Carolina Breast Cancer Study, JAMA 295 (2006)2492–2502.
53] R. Conforti, T. Boulet, G. Tomasic, et al., Breast cancer molecular subclas-sification and estrogen receptor expression to predict efficacy of adjuvant
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
anthracyclines-based chemotherapy: a biomarker study from two randomizedtrials, Ann. Oncol. 18 (2007) 1477–1483.
54] M. Tischkowitz, J.S. Brunet, L.R. Begin, et al., Use of immunohistochemical mark-ers can refine prognosis in triple negative breast cancer, BMC Cancer 7 (2007)134.
55] L.M. Sutton, J.S. Han, K.H. Molberg, et al., Intratumoral expression level of epi-dermal growth factor receptor and cytokeratin 5/6 is significantly associatedwith nodal and distant metastases in patients with basal-like triple-negativebreast carcinoma, Am. J. Clin. Pathol. 134 (2010) 782–787.
56] A.A. Thike, J. Iqbal, P.Y. Cheok, et al., Triple negative breast cancer: outcome
Please cite this article in press as: D. de Mendonc a Uchôa, et al., Expbreast carcinomas – A study of a series of triple-negative tumors, Patho
correlation with immunohistochemical detection of basal markers, Am. J. Surg.Pathol. 34 (2010) 956–964.
57] M.A. Bernardi, A.F. Logullo, F.S. Pasini, et al., Prognostic significance of CD24and claudin-7 immunoexpression in ductal invasive breast cancer, Oncol. Rep.27 (2012) 28–38.
[
[
PRESSearch and Practice xxx (2014) xxx–xxx
58] S.Y. Park, H.E. Lee, H. Li, et al., Heterogeneity for stem cell-related markersaccording to tumor subtype and histologic stage in breast cancer, Clin. CancerRes. 16 (2010) 876–887.
59] J.S. Lyzak, M.L. Yaremko, W. Recant, et al., Role of CD44 in nonpalpable T1a andT1b breast cancer, Hum. Pathol. 28 (1997) 772–778.
ression of cancer stem cell markers in basal and penta-negativel. – Res. Pract (2014), http://dx.doi.org/10.1016/j.prp.2014.03.005
60] S. Aigner, C.L. Ramos, A. Hafezi-Moghadam, et al., CD24 mediates rolling ofbreast carcinoma cells on P-selectin, FASEB J. 12 (1998) 1241–1251.
61] G. Kristiansen, E. Machado, N. Bretz, et al., Molecular and clinical dissectionof CD24 antibody specificity by a comprehensive comparative analysis, Lab.Invest. 90 (2010) 1102–1116.
![Volvo Penta 7744726[1]](https://static.documents.pub/doc/80x56/540ce3167bef0aff298b465e/volvo-penta-77447261.jpg)
