Clinical Cancer Research Versailles No Abst Metabolism
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Clinical Cancer Research 14, 4168, July 1, 2008. doi: 10.1158/1078-0432.CCR-07-4543
© 2008 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mulligan, A. M.
Right arrow Articles by Andrulis, I. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mulligan, A. M.
Right arrow Articles by Andrulis, I. L.

Imaging, Diagnosis, Prognosis

Prognostic Effect of Basal-Like Breast Cancers Is Time Dependent: Evidence from Tissue Microarray Studies on a Lymph Node–Negative Cohort

Anna Marie Mulligan1,2, Dushanthi Pinnaduwage5, Shelley B. Bull3,5, Frances P. O'Malley2,6 and Irene L. Andrulis2,4,5,6

Authors' Affiliations: 1 Department of Laboratory Medicine, St. Michael's Hospital; 2 Laboratory Medicine and Pathobiology, 3 Public Health Sciences, and 4 Department of Medical Genetics and Microbiology, University of Toronto; and 5 Samuel Lunenfeld Research Institute and 6 Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, Canada

Requests for reprints: Frances P. O'Malley, Department of Pathology and Laboratory Medicine, Mount Sinai Hospital, 600 University Avenue, Toronto, Ontario, Canada M5G 1X5. Phone: 416-586-4548; Fax: 416-586-8628; E-mail: fomalley{at}mtsinai.on.ca.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 
Purpose: To determine whether data obtained from tissue microarrays (TMA) of a prospectively accrued node-negative breast cancer cohort are prognostically informative, we compared data derived from TMA with previously determined molecular markers. Subsequent to this validation, we examined outcome in specific subgroups defined using TMA data.

Experimental Design: A consecutive series of 1,561 patients were followed for recurrence (median follow-up of 107 months). Estrogen receptor, progesterone receptor, p53, and HER2 expression, examined using TMA constructed from 887 tumors, was compared with status evaluated previously by biochemical and molecular methods. The associations with risk of recurrence were examined for biomarkers as well as for HER2, luminal, and basal subgroups defined by immunohistochemical expression.

Results: In line with earlier molecular studies, a significant risk of recurrence was found in patients with HER2 overexpression (relative risk = 2.30; P = 0.002) and p53-positive tumors (relative risk = 1.81; P = 0.005) in univariate Cox model analysis. Although complete concordance between methodologies was not observed for estrogen receptor and progesterone receptor, their associations with disease-free survival were consistent with established prognostic findings. Patients with basal-type tumors fared worse within 36 months of diagnosis but not thereafter.

Conclusions: This study shows the clinical validity of TMA in evaluating the importance of prognostic markers in this cohort. Furthermore, it shows a marked time-dependent effect in tumor subgroups, most notable within the basal subgroup. Our data suggest that patients with basal-like tumors may be broadly separable into two clinically distinctive groups: those likely to experience disease recurrence in the short term and those that will experience long-term survival.

Axillary lymph node status in breast cancer is the most important prognostic factor for clinical outcome. Nevertheless, 20% to 30% of patients with lymph node–negative disease (LNN) will experience disease recurrence and distant metastases. This limit in the prognostic ability of lymph node status has led to a search for markers with prognostic power to aid in determining which negative lymph node patients may benefit from systemic therapy. We previously assembled a multicenter prospective cohort to examine the prognostic value of molecular alterations in lymph node-negative breast cancer using flash-frozen tissue (1). Briefly, we defined the molecular markers as our primary variables, set up a mechanism to obtain all available specimens, followed all eligible patients from a well-defined population, and designed the study to have sufficient power to detect clinically important differences in disease-free survival (DFS) when several markers are being evaluated. Flash-frozen specimens were obtained and nearly 600 of these have been used for prognostic studies (1–3). We evaluated HER2 gene amplification and found it to be adversely related to outcome (1). Furthermore, we found that patients with both HER2 amplification and p53 mutation (2) had lower DFS. Lymphovascular invasion and histopathologic grade were found to be independent predictors of recurrence in multivariate analysis (1). In more recent studies, RNA from 135 of the frozen specimens has been used for gene expression profiling and we have identified gene profiles that distinguish good versus poor prognosis groups.7 The importance of the latter genes as well as other putative markers, however, has yet to be proven.

High-throughput molecular screenings, such as these, generate vast amounts of data on genes putatively involved in disease progression at a rate exceeding that at which clinical significance can be determined. However, to have any meaningful outcome translating these data into practical applications requires clinical validation on large numbers of histopathologic specimens linked to detailed clinical information. Thus, we retrieved tissue blocks from 887 subjects in our prospectively accrued LNN cohort and constructed tissue microarrays (TMA). Our aims are 2-fold: to validate the use of TMA technology for prognostic studies, which is the focus of this study, and, following successful validation, proceed with using it to validate our molecular findings.

That using small cores of tissue may not be representative of the tumor has been critically evaluated in studies that compared TMA protein expression with those of full-face sections. Excellent correlations have been found in a variety of tumor types (4–7). Other groups have found TMA to detect clinicopathologic correlations similar to those reported previously using standard sections, thus further validating the TMA as a method for analyzing protein expression in tumors (4, 8, 9). However, given that TMA are not designed to evaluate tumors on an individual basis but rather to assess protein expression in large collectives of tumors, a more important validation method would be to assess their prognostic power in a population previously well-described in terms of molecular profile and clinical outcome. This study, to our knowledge, is the first to validate TMA technology in such a manner.

Here, in the subcohort of patients within the TMA, we compare the protein expression of estrogen receptor (ER), progesterone receptor (PgR), HER2, and p53 with the previously determined molecular or biochemical status in frozen tumor samples. We correlate expression status with several clinicopathologic factors and evaluate the prognostic capabilities of biomarker expression in terms of the relative risk (RR) of outcome in DFS with a comparison with those determined previously using molecular/biochemical methods.

Gene expression studies have identified three major subtypes of breast cancer: basal-like, HER2 positive/ER negative, and luminal (10) that differ in terms of clinical outcome (11). More recent studies have shown a particular immunohistochemical profile for each of these subgroups (12–14). Based on these, we assigned tumors to either the luminal, HER2, or basal subgroup and evaluated each group for outcome. In future studies, TMA from this well-characterized cohort will be useful for the assessment of newly identified genes/proteins as well as candidate genes/proteins.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 Disclosure of Potential...
 References
 
Patient cohort and clinical follow-up. In this study, a prospectively ascertained consecutive series of women with node-negative breast cancer was enrolled from eight Toronto hospitals from September 1987 to October 1996 as described previously (1). Written consent was obtained from all of the patients included in this study.

We have continued to follow all eligible women for recurrence and death. Characteristics of the patients are listed in Table 1 . DFS was taken as the time between diagnosis and the confirmation of nonbreast recurrence. All patients were monitored for death whether or not they experienced disease recurrence. Using clinical follow-up data, patient status on January 10, 2002 determined DFS time and censoring status. Follow-up data were monitored for an additional 6 months (up to July 10, 2002) to confirm patient status at the termination date. We observed 206 (13.2%) recurrences in the entire cohort of 1,561 patients (median follow-up of 107 months) and 112 (12.6%) recurrences in the TMA cohort of 887 patients (median follow-up of 100 months). Further details about patient eligibility, clinical follow-up, and determination of survival times can be found in our recent report (3).


View this table:
[in this window]
[in a new window]

  		Table 1. Patient and tumor characteristics according to TMA data availability

 
Molecular analyses and histopathology. Details of the molecular analyses for HER2 gene amplification, p53 mutational status, and histopathology can be found in earlier publications (1–3). ER and PgR status was examined at the time of surgery using ligand-binding assays.

TMA construction and immunohistochemical staining. Formalin-fixed paraffin-embedded tumor blocks were available on 887 patients. Cores of tissue were not taken from tumors less than 4 mm to not compromise tissue. The most common reason, however, for lack of tissue for TMA construction was the inability to obtain tumor blocks from some centers that had participated in the initial studies.

Areas of invasive carcinoma were selected from a H&E-stained section of each tumor and two 0.6-mm cores of tissue were taken from the corresponding areas of the paraffin block. The selected donor cores were embedded in a paraffin block and 4-µm sections were cut from this recipient block and used in series for immunohistochemical staining for ER, PgR, p53, CK5, and HER2 under the conditions described in Table 2 . Microwave antigen retrieval was carried out in a Micromed T/T Mega Microwave Processing Lab Station (ESBE Scientific). Sections were developed with diaminobenzidine tetrahydrochloride and counterstained in Mayer's hematoxylin.


View this table:
[in this window]
[in a new window]

  		Table 2. Summary of antibodies and conditions of use

 
Each of the immunohistochemical stained sections was scored using the Allred scoring method (15). Nuclear staining was scored for ER, PgR, and p53. Strong complete membrane staining was assessed for HER2. Membranous and/or cytoplasmic staining was scored for CK5. The raw score data were processed using a TMA deconvoluter software program (16). As two cores from each tumor were assessed, the larger of the two values was chosen for use in statistical analysis. We chose the higher of the two values and not the average to minimize the effect of false negatives on the array.

Statistical analysis. Analyses were done for the TMA subgroup of 887. Generalizability of this subgroup was assessed by comparing the characteristics of this subgroup with the remainder of the entire cohort. As each biomarker was allocated a "raw score" using the Allred method, a cut point to define positive versus negative was based on previous validations for ER (>2; ref. 17), PgR (>2; ref. 18), and HER2 (>4; ref. 20). For p53, the cut point was chosen as >3 based on previous work (19), and for CK5, the cut point for positivity was arbitrarily specified as ≥4.

Descriptive baseline analyses compared frequency distributions of known prognostic factors among groups defined by biomarker status (positive versus negative). Pairwise correlations of all the biomarkers were assessed by Spearman's coefficient of rank correlation. TMA-based protein expression of tumors for ER, PgR, p53, and HER2 was compared with ER and PgR by ligand-binding assay and to p53 mutation status and HER2 gene amplification status as determined previously by molecular methods. Agreement was assessed at the biomarker measurement level using concordance rates and chance-corrected agreement ({kappa}) statistics (21) and at the clinical prognostic level using RR estimates from the Cox proportional hazards model.

Univariate survival analysis (DFS) according to biomarker status was by the log-rank test with Kaplan-Meier survival curves and by the Cox proportional hazards model. Multivariate analysis (DFS) to assess the contribution of each biomarker and traditional prognostic factors was by the Cox proportional hazards model. Prognostic factors included in the analysis were menopausal status, tumor size, ER and PgR status by ligand-binding assay, age at diagnosis, adjuvant therapy received, histologic grade, and lymphatic invasion. RR for each factor was estimated by the hazard ratio in the Cox proportional hazards model. Short-term effects were assessed by estimating a RR for the first z36 months of follow-up for DFS as in previous studies of this cohort (2, 3). Formal tests for time-dependent hazard ratios were conducted by adding an indicator term for interaction with follow-up time.

For 656 tumors with complete HER2, ER, and CK5 data, subgroups (HER2, basal, and luminal) were defined as described previously (12) as follows: all HER2-positive tumors were classified in the HER2 subgroup (n = 57); ER-negative, HER2 negative, and CK5-positive tumors were classified as basal (n = 78); and ER-positive, HER2-negative tumors were termed luminal (n = 521) regardless of CK5 status. Univariate survival analysis (DFS) according to subgroup status was by the log-rank test with Kaplan-Meier survival curves and by the Cox proportional hazards model. Short-term effects on subgroups were assessed as described above.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 Disclosure of Potential...
 References
 
Patient and tumor characteristics
All lymph node–negative patients who consented to participate in the study were observed regardless of whether a tumor specimen was available for TMA construction (n = 1,561). The clinicopathologic characteristics of the patients for whom tissue was available (n = 887) and those for whom tissue was not available (n = 674) for use in TMA construction as well as the characteristics of the whole cohort (n = 1,561) are presented in Table 1. Patients on whom tissue was available for study were significantly more likely to have had larger tumors (P = 0.0013) and lower-grade tumors (P = 0.0002) and to have received systemic adjuvant therapy (P < 0.0001). These patients were also more likely to have had ER and PgR status established at the time of the original surgery and to have PgR-positive tumors. Other patient and tumor characteristics did not differ significantly between the groups of patients with and without tissue for study (data not shown).

Correlation of TMA biomarker immunohistochemistry and previously determined molecular markers
We reported previously on HER2 gene amplification and p53 mutation status as well as ER and PgR in a subset of the cohort (1, 2). To determine whether protein expression determined by TMA immunohistochemical methods was correlated with the molecular markers, we compared results for the subset of tumors with both measurements. As shown in Table 3 , ER and PgR biomarker expression was not entirely concordant with ER and PgR status as determined at the time of surgery by ligand-binding assay of frozen tissue (chance-corrected agreement of 0.71 and 0.58, respectively). HER2 protein overexpression and gene amplification showed moderate agreement (chance-corrected agreement of 0.50) mainly due to a number (n = 29) of amplified tumors that were negative for HER2 overexpression. The status of p53 as determined by immunohistochemistry and mutation status on the sample of 227 tumors common to both cohorts exhibited only fair agreement (chance-corrected agreement of 0.40). However, agreement improved when type of p53 mutation was considered (chance-corrected agreement of 0.57). No mutations were detected in 173 of the tumors; of these, 149 (86.1%) were also negative by immunohistochemistry. When the mutation group of 54 was separated into missense (n = 28) versus truncating (n = 26) mutations, we found that missense mutations were predominantly immunohistochemistry positive (92.9%), and truncation mutations were predominantly immunohistochemistry negative (88.5%).


View this table:
[in this window]
[in a new window]

  		Table 3. Comparison between immunohistochemical and molecular/biochemical biomarker values

 
Biomarker expression and clinicopathologic and biologic parameters
ER and PgR positivity was each associated with smaller tumor size, low tumor grade, special tumor type, adjuvant hormonal treatment, and postmenopausal status (data not shown). In contrast, HER2 overexpression was associated with higher-grade tumors, lymphatic invasion, associated ductal carcinoma in situ, and larger tumor size. Patients with tumors that were CK5 positive and/or p53 positive were more likely to be premenopausal, to have larger, higher-grade tumors, and to have received chemotherapy and were less likely to have received hormonal therapy. ER and PgR positivity as determined by TMA immunohistochemistry showed a negative correlation with p53 staining, CK5 positivity, and HER2 overexpression. HER2 overexpression correlated positively with p53 expression (data not shown).

Prognostic relevance of biomarker expression
Univariate analysis by TMA immunohistochemistry. Using a log-rank test for DFS with the entire follow-up, HER2 overexpression and p53 positivity were significantly associated with reduced DFS (P = 0.0017 and 0.0047, respectively). When each of the biomarkers was considered alone in a Cox proportional hazards model for the entire follow-up period (Table 4 ), we found a 2.30-fold increase in the risk of recurrence associated with HER2 overexpression [95% confidence interval (95% CI), 1.35-3.93; P = 0.002] and a 1.81-fold increase in the risk of recurrence associated with p53 staining (95% CI, 1.19-2.75; P = 0.005). ER, PgR, and CK5 were not associated with risk of recurrence with long-term follow up at the 5% significance level. However, the risk of recurrence depended on the duration of follow-up for ER (P = 0.02), PR (P = 0.03), CK5 (P = 0.02), HER2 (P = 0.08), and p53 (P = 0.06). With short-term follow-up (36 months), ER and PgR negativity as well as HER2 overexpression and p53 positivity were significantly associated with reduced DFS (Table 4).


View this table:
[in this window]
[in a new window]

  		Table 4. Results of DFS analysis by Cox proportional hazards models with immunohistochemical markers in TMA

 
Multivariate analysis by TMA immunohistochemistry. When each of the biomarkers was assessed in a multivariate model for long-term follow up that included traditional prognostic factors, none of the biomarkers retained significance at the 5% level (Table 4). However, with short-term follow-up (36 months), ER and PgR negativity and p53 positivity retained significance at the 5% level in the multivariate model for DFS (Table 4).

Comparison of univariate DFS analysis by TMA immunohistochemistry and molecular analysis on frozen tissue
Previously, we reported significant associations of disease recurrence according to HER2 and p53 status (determined by molecular analyses) and ER and PgR status (as determined by ligand-binding assay) in a subset of this cohort (n = 543 subjects; refs. 1, 2). To assess the comparability of the associations obtained from the TMA/immunohistochemistry subcohort (Table 4) with those from the subcohort studied previously by molecular methods, we repeated univariate DFS analysis in the latter group using updated follow-up information. As presented in Table 5 for 36-month follow-up, the RR estimates for DFS were consistent between the two groups in terms of direction, magnitude, and level of statistical significance. Additional comparisons of RR estimates obtained when the analyses were limited to the subset of 252 patients in common to both subcohorts also yielded generally consistent results albeit with reduced statistical significance due to the smaller sample size (data not shown).


View this table:
[in this window]
[in a new window]

  		Table 5. Comparison between immunohistochemical and molecular/biochemical markers in DFS analysis by univariate Cox proportional hazards model for short follow-up of 36 mo

 
Outcome according to tumor subgroup
When tumors were assigned into one of three subgroups, as shown in Fig. 1 , patients with luminal tumors tended to have better DFS over the entire follow-up period than those with tumors overexpressing HER2. In univariate analysis of DFS, the RR comparing the HER2 subgroup with the luminal subgroup was 1.90 (95% CI, 1.02-3.52; P = 0.04), the RR of HER2 to basal was 2.39 (95% CI, 0.94-6.06; P = 0.07), and the RR of basal to luminal was 0.79 (95% CI, 0.36-1.74; P = 0.56). The risk of recurrence depended on the duration of follow-up for basal-like tumors compared with luminal-like tumors (P = 0.03). The basal group did poorly in the short term but relatively better with long-term follow-up. The RR associated with the basal group compared with the luminal group was 1.92 (95% CI, 0.78-4.76) within the first 36 months and decreased to 0.79 (95% CI, 0.36-1.74) with full follow-up. The RR associated with the HER2 group compared with the basal group went from 0.92 (95% CI, 0.26-3.25) at 36 months to 2.39 (95% CI, 0.94-6.06) at full follow-up (P = 0.16 for time dependence test).


Figure 1
View larger version (7K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Kaplan-Meier DFS curves stratified according to HER2, luminal, and basal subgroups. The annotation at the right-hand side provides the number at risk and the number of events that occurred during each 12-mo follow-up time period.

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
Disclosure of Potential...
 References
 
Gene expression microarray profiling has the potential to revolutionize patient management by identifying prognostic markers that identify patients most likely to benefit from adjuvant therapy as well as putative predictive markers, which may act as potential targets for novel therapies. However, clinical validation of such markers in terms of disease outcome is an essential step in translating findings into practical clinical application. TMA technology makes it feasible to study putative biomarkers on large numbers of breast cancers while achieving standardized laboratory and evaluation conditions, thereby achieving cost efficiency (22). Here, we have confirmed the prognostic capabilities of TMA analysis in a large, prospectively accrued multicenter cohort, which is well described in terms of clinical outcome (1–3), thus confirming TMA to be an informative and valuable tool in biomarker expression analysis.

Whereas several studies have shown excellent correlation between protein expression on TMA compared with full-face sections, we compared protein expression with previously determined molecular and biochemical markers, with concordance rates ranging from fair to substantial. One would not expect complete concordance in the assessment of certain biomarkers such as p53. Missense mutations, but not truncation mutations, stabilize the protein and hence their detection by immunohistochemical methods in the former only (23–26). Recently, we have shown in a subcohort of 543 patients that women whose tumors had missense p53 mutations were found to be at significantly higher risk of recurrence and death compared with those with wild-type p53, and they also tended to have worse prognosis compared with those with truncating mutations (3). Our results show that immunohistochemistry can detect p53 missense mutations; therefore, those patients are more likely to have a poor outcome.

HER2 protein overexpression showed moderate agreement with gene amplification status despite the low prevalence of HER2 protein positivity (8.3%) in the TMA cohort compared with a HER2 amplification rate of 20% (1) as assessed by quantitative PCR. Heterogeneity in staining may partly explain the low rate of positivity observed in the TMA cohort. Cases showing equivocal staining were considered negative; some of these may have been deemed positive if HER2 DNA amplification had been evaluated. In addition, low-grade tumors, which are less likely to be positive for HER2 overexpression, showed greater representation within the TMA cohort when compared with the overall cohort. Yet, the role of TMA is not to evaluate individual tumors in-depth but to assess a population of tumors collectively. Assessing the TMA population in terms of outcome showed that HER2 status predicted for a poorer outcome, similar to what was seen previously using molecular methods for HER2 status determination. In fact, the univariate Cox model RR estimates obtained using the two methods of biomarker measurement, TMA-based versus the original molecular/biochemical values, were remarkably similar not only for HER2 but also for p53, ER, and PgR (Table 5), and associations detected as significant in the original sample of 543 patients with molecular/biochemical biomarkers were also detected in the sample of 887 patients with TMA-based immunohistochemical biomarkers.

When we examined outcome in terms of subgroup (luminal, HER2, or basal), a marked time-dependent effect was seen with tumors defined as basal-like (Fig. 1). Whereas several earlier studies have reported uniformly poor prognosis for basal-like tumors (11, 27–29), one recent study (30) has reported a poor short-term survival for patients with basal-like phenotype but a significantly better survival than nonbasal grade III tumors on long-term follow-up. Our findings, which show a significantly smaller RR in basal-like cancers at full follow-up compared with that seen at 36 months, corroborate this. Furthermore, it suggests that there are as yet unknown factor(s) that broadly separate patients with basal-like tumors into two clinically distinctive groups: those likely to succumb to their disease at an early stage and those expected to show long-term survival.

In summary, we have clinically validated the use of a TMA constructed from a prospectively accrued cohort of women with lymph node negative breast cancer by evaluating its prognostic capabilities using established biomarkers and found that basal-like tumors may be composed of subgroups with different survival patterns. This TMA will be an invaluable platform for the evaluation of newly discovered putative predictive and prognostic markers in this cohort.


   Disclosure of Potential Conflicts of Interest
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
References
 
No potential conflicts of interest were disclosed.


   Acknowledgments
 
We thank the Toronto Breast Cancer Group for contributions to this work and S. Tjan (medical technologist at Mount Sinai Hospital) for excellent technical assistance with immunohistochemical staining and TMA construction.


   Footnotes
 
Grant support: National Cancer Institute of Canada with funds provided by the Terry Fox Run (I.L. Andrulis, S.B. Bull, and F.P. O'Malley) and Canadian Institutes of Health Research Senior Investigator Award (S.B. Bull).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Presented as "Validation of tissue microarray analysis in a large prospectively accrued patient cohort of lymph node negative breast cancer" at the 96th Annual Meeting, U.S. and Canadian Academy of Pathology, March 24-30, 2007, San Diego, California.

7 Andrulis et al. Manuscript in preparation. Back

Received 10/ 4/07; revised 2/14/08; accepted 4/ 3/08.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 

  1. Andrulis IL BS, Blackstein ME, et al. neu/erbB-2 amplification identifies a poor-prognosis group of women with node-negative breast cancer. Toronto Breast Cancer Study Group. J Clin Oncol 1998;16:1340–9.[Abstract/Free Full Text]
  2. Bull SB, Ozcelik H, Pinnaduwage D, et al. The combination of p53 mutation and neu/erbB-2 amplification is associated with poor survival in node-negative breast cancer. J Clin Oncol 2004;22:86–96.[Abstract/Free Full Text]
  3. Ozcelik H, Pinnaduwage D, Bull SB. TP53 mutation type and ERBB2 amplification affects survival in node-negative breast cancer. Breast Cancer Res Treat 2007;105:255–65.[CrossRef][Medline]
  4. Mucci NR, Akdas G, Manely S, Rubin MA. Neuroendocrine expression in metastatic prostate cancer: evaluation of high throughput tissue microarrays to detect heterogeneous protein expression. Hum Pathol 2000;31:406–14.[CrossRef][Medline]
  5. Nenutil R, Smardova J, Pavlova S, et al. Discriminating functional and non-functional p53 in human tumours by p53 and MDM2 immunohistochemistry. J Pathol 2005;207:251–9.[CrossRef][Medline]
  6. Nocito A, Kononen J, Kallioniemi OP, Sauter G. Tissue microarrays (TMAs) for high-throughput molecular pathology research. Int J Cancer 2001;94:1–5.[CrossRef][Medline]
  7. Zhang D, Salto-Tellez M, Do E, Putti TC, Koay ES. Evaluation of HER-2/neu oncogene status in breast tumors on tissue microarrays. Hum Pathol 2003;34:362–8.[CrossRef][Medline]
  8. Camp RL, Charette LA, Rimm DL. Validation of tissue microarray technology in breast carcinoma. Lab Invest 2000;80:1943–9.[Medline]
  9. Hoos A, Urist MJ, Stojadinovic A, et al. Validation of tissue microarrays for immunohistochemical profiling of cancer specimens using the example of human fibroblastic tumors. Am J Pathol 2001;158:1245–51.[Abstract/Free Full Text]
  10. Perou CM, Sorlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000;406:747–52.[CrossRef][Medline]
  11. 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 2001;98:10869–74.[Abstract/Free Full Text]
  12. Nielsen TO, Hsu FD, Jensen K, et al. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 2004;10:5367–74.[Abstract/Free Full Text]
  13. Livasy CA, Karaca G, Nanda R, et al. Phenotypic evaluation of the basal-like subtype of invasive breast carcinoma. Mod Pathol 2006;19:264–71.[CrossRef][Medline]
  14. Kreike B, van Kouwenhove M, Horlings H, et al. Gene expression profiling and histopathological characterization of triple-negative/basal-like breast carcinomas. Breast Cancer Res 2007;9:R65.[CrossRef][Medline]
  15. Allred DC, Harvey JM, Berardo M, Clark GM. Prognostic and predictive factors in breast cancer by immunohistochemical analysis. Mod Pathol 1998;11:155–68.[Medline]
  16. Liu CL, Prapong W, Natkunam Y, et al. Software tools for high-throughput analysis and archiving of immunohistochemistry staining data obtained with tissue microarrays. Am J Pathol 2002;161:1557–65.[Abstract/Free Full Text]
  17. Harvey JM, Clark GM, Osborne CK, Allred DC. Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer. J Clin Oncol 1999;17:1474–81.[Abstract/Free Full Text]
  18. Mohsin SK, Weiss H, Havighurst T, et al. Progesterone receptor by immunohistochemistry and clinical outcome in breast cancer: a validation study. Mod Pathol 2004;17:1545–54.[CrossRef][Medline]
  19. Done S, Arneson NC, Ozcelik H, et al. p53 protein accumulation in non-invasive lesions surrounding p53 mutation positive invasive breast cancer. Breast Cancer Res Treat 2001;65:111–8.[CrossRef][Medline]
  20. O'Malley FP, Parkes R, Latta E, et al. Comparison of HER2/neu status assessed by quantitative polymerase chain reaction and immunohistochemistry. Am J Clin Path 2001;115:504–11.[Abstract/Free Full Text]
  21. Fleiss JL. Statistical methods for rates and proportions. 2nd ed. New York: John Wiley & Sons, Inc.; 1981.
  22. Tzankov A, Went P, Zimpfer A, Dirnhofer S. Tissue microarray technology: principles, pitfalls and perspectives-lessons learned from hematological malignancies. Exp Gerontol 2005;40:737–44.[CrossRef][Medline]
  23. Bartek J, Bartkova J, Lukas J, Staskova Z, Vojtesek B, Lane DP. Immunohistochemical analysis of the p53 oncoprotein on paraffin sections using a series of novel monoclonal antibodies. J Path 1993;169:27–34.[CrossRef][Medline]
  24. Hashimoto T, Tokuchi Y, Hayashi M, et al. p53 null mutations undetected by immunohistochemical staining predict a poor outcome with early-stage non-small cell lung carcinomas. Cancer Res 1999;59:5572–7.[Abstract/Free Full Text]
  25. Kupryjanczyk J, Thor AD, Beauchamp R, et al. p53 gene mutations and protein accumulation in human ovarian cancer. Proc Natl Acad Sci U S A 1993;90:4961–5.[Abstract/Free Full Text]
  26. Sommer SS, Cunningham J, McGovern RM, et al. Pattern of p53 gene mutations in breast cancers of women of the midwestern United States. J Natl Cancer Inst 1992;84:246–52.[Abstract/Free Full Text]
  27. Abd El-Rehim DM, Ball G, Pinder SE, et al. High-throughput protein expression analysis using tissue microarray technology of a large well-characterised series identifies biologically distinct classes of breast cancer confirming recent cDNA expression analyses. Int J Cancer 2005;116:340–50.[CrossRef][Medline]
  28. Sotiriou C, Neo SY, McShane LM, et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci U S A 2003;100:10393–8.[Abstract/Free Full Text]
  29. van de Rijn M, Perou CM, Tibshirani R, et al. Expression of cytokeratins 17 and 5 identifies a group of breast carcinomas with poor clinical outcome. Am J Pathol 2002;161:1991–6.[Abstract/Free Full Text]
  30. Fulford LG, Reis-Filho JS, Ryder K, et al. Basal-like grade III invasive ductal carcinoma of the breast: patterns of metastasis and long-term survival. Breast Cancer Res 2007;9:R4.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Am J EpidemiolHome page
L. Natarajan, M. Pu, B. A. Parker, C. A. Thomson, B. J. Caan, S. W. Flatt, L. Madlensky, R. A. Hajek, W. K. Al-Delaimy, N. Saquib, et al.
Time-Varying Effects of Prognostic Factors Associated With Disease-Free Survival in Breast Cancer
Am. J. Epidemiol., June 15, 2009; 169(12): 1463 - 1470.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mulligan, A. M.
Right arrow Articles by Andrulis, I. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mulligan, A. M.
Right arrow Articles by Andrulis, I. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online