This Article Abstract Full Text (PDF) eLetters: Submit a response to this article Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services E-mail this article link to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Paik, S. Search for Related Content PubMed PubMed Citation Articles by Paik, S.

Breast Cancer

Development and Clinical Utility of a 21-Gene Recurrence Score Prognostic Assay in Patients with Early Breast Cancer Treated with Tamoxifen

Division of Pathology, National Surgical Adjuvant Breast and Bowel Project, Pittsburgh, Pennsylvania, USA

Correspondence: Soonmyung Paik, M.D., Division of Pathology, National Surgical Adjuvant Breast and Bowel Project, Four Allegheny Center 5th floor, Pittsburgh, Pennsylvania 15206, USA. Telephone: 412-359-5013; Fax: 412-359-3239; e-mail: soon.paik{at}nsabp.org

Received March 9, 2006; accepted for publication March 15, 2007.

	Learning Objectives

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
After completing this course, the reader will be able to:

1. List the strengths and limitations of the recurrence score and discuss how the TAILORx trial is trying to further the utility of the recurrence score.
   
2. Describe the need for robust prognostic factors for estrogen receptor–positive and node-negative breast cancer treated with tamoxifen.
   
3. Discuss the basic statistical requirements in prognostic marker development.
   

	ABSTRACT

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
Although patients diagnosed with axillary node–negative estrogen receptor–positive breast cancer have an excellent prognosis, about 15% of them fail after 5 years of tamoxifen treatment. Clinical trials have provided evidence that there is a significant benefit from chemotherapy for these patients, but it would be significant overtreatment if all of them were treated with chemotherapy. Therefore, context-specific prognostic assays that can identify those who need chemotherapy in addition to tamoxifen, or those who are essentially cured by tamoxifen alone, and can be performed using routinely processed tumor biopsy tissue would be clinically useful. Using a stepwise approach of going through independent model-building and validation sets, a 21-gene recurrence score (RS), based on monitoring of mRNA expression levels of 16 cancer-related genes in relation to five reference genes, has been developed. The RS identified approximately 50% of the patients who had excellent prognosis after tamoxifen alone. Subsequent study suggested that high-risk patients identified with the RS preferentially benefit from chemotherapy. Ideally the RS should be used as a continuous variable. A prospective study—the Trial Assigning Individualized Options for Treatment (Rx) (TAILORx)—to examine whether chemotherapy is required for the intermediate-risk group defined by the RS is accruing in North America.

Disclosure of potential conflicts of interest is found at the end of this article.

	DEVELOPMENT AND VALIDATION OF RECURRENCE SCORE ASSAY

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
In order to develop a context-specific prognostic assay that addresses the question "Who needs more than 5 years of tamoxifen among women diagnosed with axillary node–negative and estrogen receptor–positive breast cancer?" National Surgical Adjuvant Breast and Bowel Project (NSABP) investigators collaborated with scientists at Genomic Health, Inc. (Redwood City, CA) who developed methods for high-throughput real-time quantitative reverse transcription polymerase chain reaction (QRT-PCR) using RNA extracted from formalin-fixed paraffin-embedded tumor tissue (FPET) blocks. First, individual QRT-PCR assays optimized for fragmented RNA substrates that can be isolated from FPET samples were developed for 250 candidate genes identified through a literature and database search. Many genes identified as prognostic genes from microarray gene-expression profiling studies were included in this set of 250 genes. Three cohorts, including the tamoxifen-treated arm of NSABP trial B-20, were examined for expression of 250 genes. Correlating clinical outcome with expression levels yielded many prognostic genes from these three studies and the 16 top performing genes were identified for final model building and validation [1]. Relative expression levels of the 16 genes are measured in relationship to average expression levels of five reference genes. While a majority of the genes comprising these 16 genes are estrogen receptor (ESR1, PGR, BCL2, SCUBE2) and proliferation (Ki67, STK15, Survivin, CCNB1, MYBL2) related, there are other genes (HER-2, GRB7, MMP11, CTSL2, GSTM1, CD68, BACG1). The unscaled recurrence score (RSu) was calculated with the use of coefficients that are defined on the basis of a regression analysis of gene expression. The RS was rescaled from the RSu as follows: RS = 0 if RSu <0, RS = 20 x (RSu – 6.7) if 0 RSu 100, and RS = 100 if RSu >100. Final validation of the RS was achieved by examination of its performance in a completely independent cohort from NSABP trial B-14, which was not used in the model building process. The validation study was conducted with a rigorous predefined statistical analysis plan with prespecified outcome endpoints and cutoffs for RS. The predefined low-risk group (RS <18) had a significantly better prognosis than the higher-risk groups. Compared with the National Comprehensive Cancer Network or St. Gallen criteria, which assigned <10% of the patients from B-14 into the low-risk group [2, 3], the RS was able to categorize 50% of the patients into a low-risk group that had similar 10-year distant disease–free survival rates as a low-risk group identified by these criteria (unpublished data).

When reporting on the B-14 study, Paik et al. [1] used three subgroups arbitrarily chosen prospectively for Kaplan–Meier plot comparisons. However, the RS was developed originally as a continuous variable and should be used as such. For example, the prognosis of a patient with an RS of 17 (who was categorized as low risk in the B-14 study) would not be very different from that of a patient with an RS of 19 (categorized as intermediate risk). On the other hand, the prognosis of a patient with an RS of 2 would be fairly different from that of the one with an RS of 17, although both are in the low-risk group.

	TECHNICAL BACKGROUND

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
It is informative to understand the technology behind the RS assay and how it is different from microarray-based gene-expression studies.

Recent advances in genomics have provided tools that allow us to examine expression levels of entire genes in tumor cells. However, these gene-expression tools usually require very high quality RNA as a starting material. The main hurdle in adopting this technical advance is the lack of fresh frozen tumor tissue from large phase III adjuvant trials. Therefore, methods that allow gene-expression profiling of FPET specimens, which are often fairly old, are in need.

RNA isolated from FPET samples is a poor starting material for gene-expression profiling. Masuda et al. [4] analyzed the RNA extracted from FPET samples for chemical modifications. In freshly made FPET samples that were fixed and processed in ideal conditions (fixed in 10% buffered formalin at 4°C), RNA was fairly well preserved. Although the extracted RNA did not show any sign of degradation compared with fresh samples, they were poor substrates for cDNA synthesis and subsequent PCR amplification, so that only PCR amplification of short targets was possible. The investigators found the addition of mono-methylol (-CH₂OH) groups to all four bases, to varying degrees, and some adenines dimerized through methylene bridging. In addition to the chemical modification, however, RNA in FPET samples continue to be degraded or fragmented over time during storage for reasons unclear. Cronin et al. [5] systematically examined the quality of RNA extracted from breast cancer FPET specimens taken at different times. RNA from FPET archived for 1 year was less fragmented than RNA archived for 6 or 17 years.

The initial step in any gene-expression profiling method including RT-PCR and microarray is synthesis of cDNA from mRNA species extracted from the tumor tissue [6]. In the usual method of gene-expression profiling, reverse transcription of the mRNA using an oligo-dT primer designed to bind poly-adenylated tail is used to generate DNA complementary to mRNA (cDNA). In FPET, chemical modification, especially of poly-adenylated sequences, and fragmentation significantly limit cDNA synthesis using RNA extracted from paraffin blocks [4, 5].

Fragmented RNA extracted from paraffin blocks, however, can be a reasonable substrate for real-time RT-PCR assay if gene-specific priming is used for cDNA synthesis for each gene target instead of oligo-dT primed reverse transcription [5]. One additional problem of FPET is that the absolute signal of RT-PCR from the same amount of starting RNA decreases significantly if the blocks have been stored for a long time, resulting in about a 100-fold reduction in signal if the block is 10 years old compared with a freshly made block [5]. However, careful normalization based on genes with minimal variation in expression level among different tumor samples can largely compensate for these differences in absolute signal [5]. The RS assay is the result of combining high-precision and high-throughput robotics, careful optimization of the normalization method, and miniaturization of QRT-PCR assays that allow monitoring of 16 cancer genes at once using RNA extracted from about 30-µm thick slices from FPET blocks. Currently, all steps of the clinical RS assay (oncotype DX^TM; Genomic Health, Inc., Redwood, CA) are automated except the RNA extraction step.

	COMPARISON WITH EXISTING CLINICAL PROGNOSTIC TOOLS

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
The high cost of the RS assay (estimated at $3,400) can be justified only if it provides additional information over and beyond readily available clinicopathological markers. Bryant and colleagues, in collaboration with Peter Ravdin, M.D., compared RS with the Adjuvant! Online (AOL) algorithm developed by Dr. Peter Ravdin [7] in the NSABP B-14 cohort in a presentation made at the 2005 St. Gallen Breast Cancer Symposium (personal communications with Drs. John Bryant and Peter Ravdin). The AOL program provides individualized estimates of recurrence risk based on information such as patient age, tumor size, nodal involvement, histologic grade, etc. Since AOL does not provide risk categories, the AOL output from B-14 was rank-ordered so that a similar proportion of cases would be categorized as low risk (50%) when compared with the RS. In a multivariate analysis, both assays were independently prognostic, although agreement between the two assays was only 48%. However, the RS was able to tease out patients with a poor prognosis from the AOL low-risk category, so that the 10-year distant relapse rate was 5.6% for both AOL and RS low-risk patients. However, AOL's low-risk patients and the RS's intermediate- and high-risk patients had a 12.9% distant relapse rate at 10 years in the tamoxifen-treated cohort (p = .04). For AOL's intermediate- or high-risk patients, the distant relapse rate was 8.9% when the RS was low, whereas the distant relapse rate was 30.7% when the RS risk was intermediate or high (p < .0001). These data confirm that the RS assay adds to existing prognostic markers. The other way to explain the data is that both clinical and gene expression–based markers contribute to prognosis independently and therefore have to be considered together. The current version of AOL incorporates the RS in its prognostic consideration.

	FURTHER VALIDATION OF THE RS IN THE INTENDED CLINICAL CONTEXT

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
Habel et al. [8] conducted a nested case–control study in breast cancer patients diagnosed in 1985–1994 at 14 Kaiser hospitals. Eligibility included negative nodes, age <75 years, and no chemotherapy. Cases died of breast cancer prior to 2002. Up to three controls were matched to each case on age, race, tamoxifen treatment, facility, diagnosis year, and follow-up time. Among 4,964 potentially eligible patients, 220 cases and 570 matched controls were identified. The median age was 59 years (range, 28–74); 30.9% were treated with tamoxifen (mostly after 1988). The median follow-up time was 4.9 years for cases (time to death) and 12.9 years for controls. The RS was significantly associated with breast cancer death in estrogen receptor (ER)-positive patients treated with or without tamoxifen (p = .002). This study provided validation of the RS assay used in a community setting.

	FURTHER EXAMINATION OF THE RS IN A DIFFERENT CLINICAL CONTEXT

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
Two published studies that examined the clinical utility of the RS outside the originally targeted cohort (ER-positive, node-negative patients treated with tamoxifen) have been published.

The first one, from Cobleigh et al. [9], is the description of one of the cohorts used in the discovery and model-building step in the development of the RS. Seventy-eight patients with more than nine positive axillary lymph nodes were studied. Seventy-seven percent of patients had distant recurrence or breast cancer death. Univariate Cox analysis identified 22 genes associated with distant recurrence–free survival (DRFS). Fourteen of the 16 cancer-related genes used in the RS algorithm correlated with breast cancer recurrence, nine of them at p < .05 and 14 at p < .10. Of the 78 evaluable patients in this study, 11 (14%) had an RS of <18 and had a rate of distant recurrence at 10 years of 29% (95% confidence interval [CI], 0%–53%), 19 (24%) had an RS of 18–31 and had a rate of distant recurrence at 10 years of 72% (95% CI, 38%–88%), and 48 (62%) had an RS 31 and had a rate of distant recurrence at 10 years of 80% (95% CI, 63%–89%). Considering the fact that many genes included in the RS are already known prognostic factors, it is not surprising that the RS was prognostic in this cohort, widely different from the originally intended use.

The second study, from Esteva et al. [10], examined oncotype DX^TM in 149 untreated node-negative patients. The median age at diagnosis was 59 years, the mean tumor diameter was 2 cm, and 69% of tumors were ER positive. The median follow-up time was 18 years. The 5-year disease free–survival rate for the group was 80%. The RS was not predictive of distant disease recurrence. However, a high concordance between RT-PCR and immunohistochemical assays for ER, progesterone receptor (PgR), and human epidermal growth factor receptor 2 (HER-2) status was noted. Because patients with a good nuclear grade had a poor clinical outcome, this cohort seems to have some unknown selection bias that makes interpretation of the clinical correlation data difficult. However, the study is still important in that it validated the technical aspect of the RS assay by correlating it with immunohistochemical assays for ER, PgR, and HER-2.

	THE RS AND RESPONSE TO CHEMOTHERAPY

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
The neoadjuvant study reported by Gianni et al. [11] is the most interesting because it suggests a correlation between RS and response to chemotherapy. Patients with locally advanced breast cancer were treated with doxorubicin (60 mg/m²) and paclitaxel (200 mg/m²) every 3 weeks x 3, followed by weekly paclitaxel (80 mg/m²) x 12 before surgery, and adjuvant i.v. cyclophosphamide, methotrexate, and fluorouracil (CMF) every 4 weeks x 4 thereafter.

Of 93 patients with an available biopsy enrolled between 1998 and 2002, 89 (97%) had sufficient RNA extracted for analysis. The mean age was 50 years and the median tumor size was 6 cm. Pathologic complete response (pCR) was observed in 11 patients (12.4%). Quantitative gene-expression results were obtained for 383 genes. Using univariate analysis, a significant correlation between gene expression and pCR was observed for 87 genes (p < .05), including five with p < .001 and 30 with p < .01. Thirty of the 87 genes that correlated with pCR clustered by both expression and function into three groups—an ER group (e.g., ER, PGR, SCUBE2), a proliferation group (e.g., MYBL2, E2F1, MCM6), and an immune group (e.g., CD3z, CD18, FASL). pCR was more likely with lower expression for the ER group and higher expression for the proliferation and immune groups. Multivariate analysis indicated that combinations of genes are more powerful predictors of response than single genes. Notably, the RS was positively associated with the likelihood of pCR (p = .005), suggesting that the patients who are at the greatest recurrence risk are more likely to have chemotherapy benefit.

This hypothesis was tested by examination of all available FPET blocks from NSABP trial B-20 [12]. There were 651 evaluable patients (227 randomized to tamoxifen and 434 randomized to tamoxifen plus either CMF or MF). Patients with tumors that had a high RS (RS >30) had a large absolute benefit of chemotherapy (with an absolute increase in DRFS at 10 years of 27.6% ± 8.0%). Patients with tumors that had low RS (RS <18) derived minimal, if any, benefit from chemotherapy (with an absolute increase in DRFS at 10 years of –1.1% ± 2.2%). The test for interaction between chemotherapy treatment and RS was statistically significant (p < .05). Of particular importance is the fact that there was a linear relationship between RS and the degree of benefit from chemotherapy when RS was examined as a continuous variable in a Cox model.

	DO ALL TUMORS DERIVE A SIMILAR DEGREE OF BENEFIT?

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
The ability to predict response to chemotherapy is not a unique characteristic of the RS. It is now clear from many studies that tumor response to chemotherapy is not uniform and there is a correlation between biological aggressiveness and response to chemotherapy. One example is a study of urokinase-type plasminogen activator (uPA)/plasminogen activator inhibitor type 1 (PAI-1) by Harbeck and colleagues [13]. They analyzed samples from more than 8,000 women with node-negative breast cancer. Antigen levels of tumor biomarkers, uPA and its inhibitor, PAI-1, were found to correlate directly with the risk for disease recurrence. Therefore, the investigators hypothesized that patients whose primary tumors expressed low levels of uPA and PAI-1 might safely avoid the need for adjuvant chemotherapy. Conversely, patients with elevated uPA/PAI-1 antigen levels were presumed to have a higher risk for disease recurrence. Accordingly, patients categorized as high risk, based on uPA/PA-1 levels, derived significant benefit from chemotherapy, whereas low-risk patients gained little from the addition of chemotherapy. Rouzier et al. [14] have reported a correlation between molecular subclasses defined by gene-expression patterns and response to neoadjuvant chemotherapy in 82 patients treated at MD Anderson Cancer Center. The basal-like and HER-2 subgroups, which have been shown to be associated with poor prognosis, were also associated with the highest rates of pCR—45% (95% CI, 24%–68%) and 45% (95% CI, 23%–68%), respectively—whereas luminal tumors had a pCR rate of 6% (95% CI, 1%–21%). No pCR was observed among the normal-like cancers (95% CI, 0%–31%). Together with the Gianni et al. [11] study described before, this study provides strong support for the validity of findings from the B-20 trial in which RS was associated with degree of benefit from chemotherapy.

	PROSPECTIVE STUDY TO EXAMINE THE WORTH OF CHEMOTHERAPY IN INTERMEDIATE-RISK PATIENTS DEFINED BY THE RS

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
In North America, a large prospective study, called the Trial Assigning Individualized Options for Treatment (Rx) (TAILORx), which tests the value of chemotherapy for intermediate RS patients is being conducted [15]. In that study, patients diagnosed with axillary node-negative and ER-positive breast cancer are upfront tested with oncotype DX^TM, and those with scores of 12–25 are randomized to hormone therapy or hormone therapy plus chemotherapy, while the low-risk group is followed to validate their excellent prognosis. Tumor and blood samples collected from this trial will provide excellent resources to further improve the RS and develop/validate other interesting markers.

	REAL-WORLD PERFORMANCE OF THE RS

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
At the 2006 San Antonio Breast Cancer Symposium, Genomic Health, Inc. reported its experience of performing the RS assay in >20,000 cases in a single reference laboratory [16]. The cases were divided into two groups based on chronology of the assay performed. The distribution of the marker data was remarkably similar among the approximately 10,000 case groups, suggesting very good reproducibility of the assay. More importantly, the data showed that, while cases can be roughly divided into luminal, HER-2, and triple-negative subgroups, as suggested by Perou et al. [17], there exists a continuum in expression levels of ER, PgR, and HER-2. These data again underscore the need for clinical oncologists to change their way of looking at breast cancer—not as distinct categories but as a continuum of biology.

	DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
The author indicates no potential conflicts of interest.

	ACKNOWLEDGMENTS

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 
This work was supported in part by grants from the National Cancer Institute.

	REFERENCES

Top Learning Objectives Abstract Development and Validation of... Technical Background Comparison with Existing... Further Validation of the... Further Examination of the... The RS and Response... Do All Tumors Derive... Prospective Study to Examine... Real-World Performance of the... Disclosure of Potential... Acknowledgments References

 

1. Paik S, Shak S, Tang G et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004;351:2817–2826.[Abstract/Free Full Text]
2. Carlson RW, Anderson BO, Bensinger W et al. NCCN Practice Guidelines for Breast Cancer. Oncology (Williston Park) 2000;14:33–49.
3. Goldhirsch A, Wood WC, Gelber RD et al. Meeting highlights: Updated international expert consensus on the primary therapy of early breast cancer. J Clin Oncol 2003;21:3357–3365.[Abstract/Free Full Text]
4. Masuda N, Ohnishi T, Kawamoto S et al. Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples. Nucleic Acids Res 1999;27:4436–4443.[Abstract/Free Full Text]
5. Cronin M, Pho M, Dutta D et al. Measurement of gene expression in archival paraffin-embedded tissues: Development and performance of a 92-gene reverse transcriptase-polymerase chain reaction assay. Am J Pathol 2004;164:35–42.[Abstract/Free Full Text]
6. Paik S, Kim CY, Song YK et al. Technology insight: Application of molecular techniques to formalin-fixed paraffin-embedded tissues from breast cancer. Nat Clin Pract Oncol 2005;2:246–254.[CrossRef][Medline]
7. Ravdin PM. A computer program to assist in making breast cancer adjuvant therapy decisions. Semin Oncol 1996;23(suppl 2):43–50.[Medline]
8. Habel LA, Shak S, Jacobs MK et al. A population-based study of tumor gene expression and risk of breast cancer death among lymph node-negative patients. Breast Cancer Res 2006;8:R25.[CrossRef][Medline]
9. Cobleigh MA, Tabesh B, Bitterman P et al. Tumor gene expression and prognosis in breast cancer patients with 10 or more positive lymph nodes. Clin Cancer Res 2005;11:8623–8631.[Abstract/Free Full Text]
10. Esteva FJ, Sahin AA, Cristofanilli M et al. Prognostic role of a multigene reverse transcriptase-PCR assay in patients with node-negative breast cancer not receiving adjuvant systemic therapy. Clin Cancer Res 2005;11:3315–3319.[Abstract/Free Full Text]
11. Gianni L, Zambetti M, Clark K et al. Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer. J Clin Oncol 2005;23:7265–7277.[Abstract/Free Full Text]
12. Paik S, Tang G, Shak S et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. J Clin Oncol 2006;24:3726–3734.[Abstract/Free Full Text]
13. Harbeck N, Kates RE, Schmitt M et al. Urokinase-type plasminogen activator and its inhibitor type 1 predict disease outcome and therapy response in primary breast cancer. Clin Breast Cancer 2004;5:348–352.[Medline]
14. Rouzier R, Perou CM, Symmans WF et al. Breast cancer molecular subtypes respond differently to preoperative chemotherapy. Clin Cancer Res 2005;11:5678–5685.[Abstract/Free Full Text]
15. Sparano JA. TAILORx: Trial Assigning Individualized Options for Treatment (Rx). Clin Breast Cancer 2006;7:347–350.[Medline]
16. Shak S, Baehner FL, Palmer G et al. Subtypes of breast cancer defined by standardized quantitative RT-PCR analysis of 10,618 tumors. Breast Cancer Res Treat 2006;100(suppl 1):S295; Abstract 6118.
17. Perou CM, Sorlie T, Eisen MB et al. Molecular portraits of human breast tumours. Nature 2000;406:747–752.[CrossRef][Medline]

This article has been cited by other articles:

				F. Montemurro and M. Aglietta Hormone receptor-positive early breast cancer: controversies in the use of adjuvant chemotherapy Endocr. Relat. Cancer, December 1, 2009; 16(4): 1091 - 1102. [Abstract] [Full Text] [PDF]

				F. Correa Geyer and J. S. Reis-Filho Microarray-based Gene Expression Profiling as a Clinical Tool for Breast Cancer Management: Are We There Yet? International Journal of Surgical Pathology, August 1, 2009; 17(4): 285 - 302. [Abstract] [PDF]

				C. H. Barrios, C. Sampaio, J. Vinholes, and R. Caponero What is the role of chemotherapy in estrogen receptor-positive, advanced breast cancer? Ann. Onc., July 1, 2009; 20(7): 1157 - 1162. [Abstract] [Full Text] [PDF]

				C. M. Sturgeon, M. J. Duffy, U.-H. Stenman, H. Lilja, N. Brunner, D. W. Chan, R. Babaian, R. C. Bast Jr., B. Dowell, F. J. Esteva, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for Use of Tumor Markers in Testicular, Prostate, Colorectal, Breast, and Ovarian Cancers Clin. Chem., December 1, 2008; 54(12): e11 - e79. [Abstract] [Full Text] [PDF]

				R. Simon The Use of Genomics in Clinical Trial Design Clin. Cancer Res., October 1, 2008; 14(19): 5984 - 5993. [Abstract] [Full Text] [PDF]

				D. C. Sullivan Imaging as a Quantitative Science Radiology, August 1, 2008; 248(2): 328 - 332. [Full Text] [PDF]

				J. S. Ross, C. Hatzis, W. F. Symmans, L. Pusztai, and G. N. Hortobagyi Commercialized Multigene Predictors of Clinical Outcome for Breast Cancer Oncologist, May 1, 2008; 13(5): 477 - 493. [Abstract] [Full Text] [PDF]

				L.-P. P. Tran and J. L. Grabinski Chemotherapy for Early-Stage Breast Cancer: A Paradigm in Flux Journal of Pharmacy Practice, February 1, 2008; 21(1): 46 - 56. [Abstract] [PDF]

This Article Abstract Full Text (PDF) eLetters: Submit a response to this article Purchase Article View Shopping Cart Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services E-mail this article link to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Paik, S. Search for Related Content PubMed PubMed Citation Articles by Paik, S.