A Young Woman With a Small ER-Positive Breast Cancer, a Micrometastatic Axillary Lymph Node, and an Intermediate Oncotype DX Recurrence Score

SECOND OPINION

Multidisciplinary Consultations on Challenging Cases

The University of Colorado Health Sciences Center holds weekly second opinion conferences focusing on cancer cases that represent most major cancer sites. Patients seen for second opinions are evaluated by an oncologist. Their history, pathology, and radiographs are reviewed during the multidisciplinary conference, and then specific recommendations are made. These cases are usually challenging, and these conferences provide an outstanding educational opportunity for staff, fellows, and residents in training.

The second opinion conferences include actual cases from genitourinary, lung, melanoma, breast, neurosurgery, and medical oncology. On an occasional basis, ONCOLOGY will publish the more interesting case discussions and the resultant recommendations. We would appreciate your feedback; please contact us at second.opinion@uchsc.edu.

E. David Crawford, MD
Al Barqawi, MD
Guest Editors

University of Colorado Health Sciences Center
Univeristy of Colorado Cancer Center Denver, Colorado

The patient is a 39-year-old premenopausal woman who presents with a new diagnosis of breast cancer to our multidisciplinary second opinion clinic.

History

The patient originally noticed a lump at the 3 o'clock position in her left breast approximately 1 ½ years ago. A mammogram and ultrasound were negative at that time, and no further studies were pursued. Several months before the woman presented to us, however, the lump began to change. Another mammogram was obtained and followed by an ultrasound-guided core biopsy demonstrating adenocarcinoma of the breast. She did not have any symptoms or laboratories findings suggestive of metastatic disease.

The patient underwent lumpectomy and sentinel node biopsy 1 month later. Histology revealed a 0.8-cm grade 2 invasive ductal adenocarcinoma with negative margins. The lesion was strongly positive for estrogen and progesterone receptors (ER/PR) and negative for HER2. One of two sentinel lymph nodes was positive for microscopic involvement measuring 1.1 mm, signifying a diagnosis of T1b, pN1mi, M0 stage IIA breast cancer. The Oncotype DX recurrence score-obtained prior to knowledge about the nodal micrometastasis-was 21. With this information, the patient presented 1 month after her lumpectomy to the University of Colorado Breast Center for further management.

The patient has no other significant medical history. Her status is G2P2 with menarche at age 11, and she has ongoing, regular menses. She bore her first child at age 33 years and does not smoke or drink alcohol. Her maternal grandmother developed breast cancer at age 53.

Discussion

Dr. Daniel Bowles: This case brings up the role of micrometastatic nodal involvement and Oncotype DX in the prognosis and treatment of breast cancer, two areas of significant controversy. Let's start by discussing microscopic nodal involvement.

Micrometastatic Lymph Node Disease

Dr. Bowles: What is the exact definition of micrometastatic lymph node disease in breast cancer?

Dr. Meenakshi Singh: According to the 6th edition of the American Joint Committee on Cancer (AJCC) staging manual, micrometastases are defined as tumor deposits within the lymph node greater than 0.2 mm but not larger than 2.0 mm. There must also be a stromal reaction or other evidence of metastatic activity to identify a micrometastatic lymph node to avoid confusion with ectopic glands or artifact from biopsies.[1]

Dr. Anthony Elias: The reason we care about micrometastatic lymph node disease is because its presence changes disease stage, prognosis, and treatment choices, including further surgery. For this woman, is an axillary node dissection necessary?

Dr. Christina Finlayson: Currently, American Society of Clinical Oncology (ASCO) guidelines recommend an axillary dissection when the sentinel lymph node contains micro- or macrometastatic disease.[2] This recommendation is based on the 20% to 35% incidence of additional nodal disease found when an axillary dissection is performed for the finding of micrometastatic disease. Although axillary dissection improves the quality of staging, there is little or no evidence regarding the impact on local recurrence or disease survival.

Predictive models are being developed to further identify patients at highest risk for additional nodal disease. The Van Zee nomogram has been widely discussed as a possible mechanism for further defining the risk of nonsentinel lymph node metastases and balancing that risk against the risks of surgery.[3] Using the Van Zee nomogram, this patient has a 10% risk of further lymph node involvement.

Dr. Bowles: The Van Zee nomogram may not address microscopic nodal disease specifically. A recent abstract found that compared to an area under the receiver-operator curve (ROC) for macrometastatic lymph node disease of 0.724, the area under the ROC was 0.538 for microscopic nodal disease.[4] This suggests that the nomogram is less predictive of microscopic nodal disease. However, several other studies suggest that microscopic nodal disease increases the likelihood of disease within other axillary lymph nodes.

Two studies found that 13% to 18% of women with micrometastatic lymph node disease had additional positive lymph nodes on axillary dissection, compared to 51% with macrometastatic lymph node disease. The size of the primary tumor, the size of the micrometastasis, and the presence of lymphovascular invasion increased the risk.[5,6] These studies contrast with an earlier study suggesting that only 6% of patients have positive nonsentinel lymph nodes.[7]

While these data are somewhat disparate, a recent meta-analysis of women with micrometastatic sentinel lymph nodes found a pooled proportion of positive nonsentinel lymph nodes of 10% to 15%, depending on the detection method.[8] It's unclear to me whether axillary lymph node dissection is necessary if several studies suggest roughly a 10% to 20% chance of having additional positive nodes.

H&E Staining vs IHC

Dr. Finlayson: It is also important to know whether the disease deposit within the sentinel node was identified on hematoxylin and eosin (H&E) stain or by cytokeratin immunohistochemistry (IHC). Deposits identified by IHC only are staged as N0(i+). The most recent ASCO guidelines do not provide specific recommendations for or against axillary dissection in this situation. However, there has been some debate about whether to screen sentinel lymph nodes with IHC instead of H&E in order to increase sensitivity. It is unclear whether increased sensitivity will change prognosis and treatment in a meaningful way.

Dr. Bowles: That is an area where the data are somewhat mixed. One recent investigation evaluated sentinel lymph nodes using H&E staining and immunostaining for cytokeratin. They found that 26% of patients have lymph node disease by H&E staining, whereas IHC identified another 4% of patients, half of which had micrometastatic disease that was retrospectively seen on H&E staining, as well.[9] Other studies have corroborated that IHC increases the sentinel node involvement rate over H&E staining alone.[10,11]

The real issue is whether micrometastatic disease detected only by IHC portends a worse prognosis than node-negative disease or disease identified on H&E. Several small studies suggest that micrometastases detected only by IHC, and staged as pN0(i+) disease, are of little clinical significance, particularly in the context of adjuvant therapies.[12-14] These results were corroborated by a larger study of almost 700 women, which found that occult metastases identified by IHC were not associated with a worse prognosis.[15] On the contrary, the authors of another 700-patient study concluded that the risk of positive nonsentinel nodes was increased when sentinel node micometastases were detected only by IHC, and that this justifies completion axillary node dissection.[16] Clearly this is still an area of significant debate, although the majority of data seems to suggest that disease identified on H&E confers a worse prognosis than disease identified solely by IHC.

What is the University of Colorado Hospital Breast Center's protocol for identifying cancer within the lymph nodes?

Dr. Singh: Nodes that appear benign after touch/imprint cytology undergo permanent fixation and are stained with H&E. For ductal carcinomas, we do not perform IHC if the permanent section of the lymph node is negative by H&E. It is more difficult to distinguish between lobular carcinoma and the native lymphocytes in the sentinel lymph nodes. Therefore, lobular cancers with negative sentinel nodes by H&E will undergo cytokeratin immunostaining. If the sentinel lymph node is grossly abnormal, we perform an H&E stain on fresh frozen samples taken during surgery and submit the remainder of the node for permanent fixation. We feel this maximizes our ability to diagnose clinically important disease in the axilla and is in line with current guidelines.[2]

Axillary Dissection and Radiotherapy

Dr. Rachel Rabinovitch: I agree that standard therapy would be to perform a completion axillary dissection in order to (1) achieve tumor control in the axilla and (2) improve our understanding of the patient's axillary disease burden, staging, and risk of occult systemic disease. Knowing which patients have significant axillary disease and may benefit from an axillary dissection is important because completion axillary node dissection can result in the most chronic and life-affecting toxicities of breast cancer therapy. Another option for this patient would be axillary irradiation. If identification of further axillary disease would not affect systemic therapy recommendations, it's difficult to argue against offering axillary radiotherapy as an option.

In general, no axillary radiotherapy would be needed after a level I/II axillary dissection. We have reasonably good evidence that radiation to the level I and II nodes in a clinically negative axilla is an effective modality to prevent progression in the axillary tissue, even with a positive pathologic sentinel node.[17,18] Two studies have demonstrated equivalent survival in patients randomized to axillary dissection vs axillary radiotherapy, with only a 3% axillary failure rate at 15 years in the radiotherapy group.[18] Unfortunately, there are no data addressing our specific patient's scenario-the efficacy of axillary radiotherapy when a micrometastatic sentinel node has been identified. I would postulate that the likelihood of residual disease is the same or even smaller in our patient than in a patient whose sentinel node status is unknown, and would therefore feel that radiotherapy is certainly a reasonable treatment option.

Prognosis and Additional Treatment Options

Dr. Alexander Urquhart: I think microscopic nodal disease does provide some important prognostic information. It tells me that there is a population of cells within the tumor that has the ability to metastasize locally and spread distantly. The degree of risk for distant metastasis is still being determined and was a driving force to include N1mi and N0(i+) in the staging criteria. Retrospective analyses suggest that prognosis is a continuum, with microscopic nodal disease leading to a risk of distant recurrence somewhere in between node-negative and macroscopic nodal involvement.[19,20]

Dr. Virginia Borges: The American College of Surgeons Oncology Group (ACOSOG) Z0011 trial attempted to definitively answer the role of microscopic nodal disease and completion axillary dissection after sentinel node biopsy confirmation. Unfortunately, it closed without meeting its target accrual. Further axillary dissection would not likely alter my treatment recommendations, as I lean toward treating a situation like this with a regimen typical for node-positive breast cancer, especially given her young age and excellent health.

Dr. Elias: When counseling patients on the importance of further axillary surgery, an important consideration is whether it would make a difference in other therapies such as radiation therapy or the choice of chemotherapy.

For example, radiation therapy might be changed if she had chosen a mastectomy. There is controversy about postmastectomy radiation when one to three axillary lymph nodes are positive. Two randomized control trials support its use, although neither is from the United States.[21,22] These studies have been discounted by some investigators, in part because of the limited amount of surgery performed. On the other hand, less axillary surgery is performed now and those studies are closer to real life. However, I do not believe that there is consensus to irradiate the chest wall postmastectomy for N1mi disease. If the patient received a mastectomy and had a 2.5-cm tumor with lymphovascular invasion, would the presence of N1mi push you to recommend postmastectomy radiation therapy?

Dr. Rabinovitch: This is a controversial area with not enough available data. Randomized controlled trials of limited nodal disease suggest that even one positive lymph node increases the risk of locoregional recurrence-and that the risk of locoregional failure increases with increasing nodal disease. Prevention of locoregional recurrence translates into improvement in survival. Unfortunately, there's no breakdown in the published trials about the subgroup of patients with only one positive lymph node, let alone micrometastatic involvment in a single lymph node.

I would definitely consider offering postmastectomy radiation therapy to a premenopausal woman with lymphovascular invasion and a solitary micrometastatic lymph node, being frank that the benefit is likely to be small and the evidence supporting this approach is not available. We need to remember that the survival of women with chest wall recurrence is ultimately dismal.

What about the 0.8-cm tumor?

Oncotype DX

Dr. Borges: This woman had an Oncotype DX gene profile ordered by her local surgeon. The assay result is a raw number generated from a mathematical computation that takes the results of 16 genes into account and applies various weights for favorable or unfavorable prognosis associated with each gene. The raw score (recurrence score) is translated into a risk of recurrence over 10 years persisting after the presumed benefit of tamoxifen, as all women in the dataset that generated the assay had been treated with tamoxifen. The data demonstrated that the Oncotype DX recurrence score predicts distant recurrence in women with estrogen-positive, lymph node-negative cancer who were treated with tamoxifen.[23]

Dr. Bowles: In that original study, low risk was defined as a recurrence score less than 18, high risk was 31 or greater, and intermediate was everything in between. The result of this study has been corroborated by a large population-based study from a health management organization.[24] Furthermore, a recent follow-up study found that the Oncotype DX recurrence score might predict the magnitude of benefit from chemotherapy in patients with ER-positive, node-negative disease who were treated with tamoxifen.[25]

Oncotype DX is well validated in this population of patients. Beyond this group of patients, two of the studies used to design the assay suggested that tumor gene expression can help predict distant recurrence in women with 10 or more positive nodes[26] and response to chemotherapy in women with locally advanced breast cancer.[27] Recurrence score did not predict response to chemotherapy in a small study of women with stage II or III breast cancer, however.[28]

The interest in Oncotype DX in node-positive and node-negative breast cancer is expanding. One question that remains to be answered is how we interpret the Oncotype DX recurrence score in a patient with microscopic nodal or node-positive disease? In particular, how could it contribute to assessing the risk of local or distance recurrence and the need for systemic treatment?

Dr. Elias: The Oncotype DX has not been well studied in patients with node-positive or ER-negative disease, so I don't believe we have much real data to address this question. I think it's fair to believe that the Oncotype DX reflects biology. After all, pathologists already used very similar markers with IHC in their assessments of tumors. Oncotype DX probably has better accuracy than some of those tests. Of course, it's also expensive (approximately $3,500).

The interesting issue is how Oncotype DX might be able to risk-stratify people in the "gray zones" of breast cancer. For example, a tumor with a low Oncotype DX score is likely to have a less aggressive biology than one with a high score, yet some of these tumors can metastasize to the lymph node and beyond. Can Oncotype DX and nodal status in women who are N1mi or N0(i+) help determine whether they need aggressive chemotherapy? If the Oncotype DX score is high in N1mi or N0(i+), aggressive chemotherapy is likely to be indicated. Of course, this woman had an intermediate score of 21. Subset analysis of National Surgical Adjuvant Breast and Bowel Project (NSABP) B-20 in N0 patients suggested that the intermediate recurrence score had limited benefit from chemotherapy.

Individual Oncotype DX gene analysis would suggest that high mRNA levels of ER would predict a greater benefit from antiestrogenic therapy and low proliferative index might predict a lesser benefit from chemotherapy. For this patient, it is therefore possible that chemotherapy might not provide much benefit. Of major interest is the recent abstract from the Eastern Cooperative Oncology Group (ECOG) measuring Oncotype DX in women getting chemotherapy for breast cancer with 0 to 3 positive lymph nodes.[29] Recurrence score was clearly associated with prognosis, and women with a low recurrence score had an excellent prognosis despite lymph node involvement. The role of Oncotype DX or other gene-profiling techniques remains uncertain in cases of nodal involvement and needs further prospective study.

Oncotype DX and Locoregional Recurrence

Dr. Elias: What about Oncotype DX and locoregional recurrence?

Dr. Rabinovitch: Oncotype DX was initially designed to assess the risk of distant metastasis, not local recurrence. However, an interesting study by Mamounas and colleagues evaluated the potential association of Oncotype DX recurrence score with risk of locoregional recurrence.[30] These authors found that women in the NSABP B-14 and B-20 trials with node-negative, ER-positive breast cancer had significantly increased rates of locoregional recurrence with increased Oncotype DX recurrence scores. The patients in these trials were treated locally with either a modified radical mastectomy or lumpectomy, axillary dissection, and breast radiotherapy. Chemotherapy consisted of CMF (cyclophosphamide, methotrexate, fluorouracil [5-FU]) or MF (methotrexate, 5-FU) in B-20. This association between recurrence score and locoregional failure was identified regardless of systemic therapy given-placebo, tamoxifen, or tamoxifen plus chemotherapy.

Assuming this patient receives chemotherapy, breast and axillary irradiation, and tamoxifen, her risk of locoregional recurrence would be 2.7% according to this abstract. Of course, all the women in this study who received breast-conserving surgery were also treated with radiation, so it does not address the question as to who will benefit most from breast or axillary irradiation. I do not know of any studies using Oncotype DX to randomize patients to radiation or no radiation, but I think it would be a very interesting study.

Dr. Bowles: Are there studies using Oncotype DX to randomize patients to different systemic therapy regimens?

Dr. Borges: To my knowledge, there are no ongoing prospective clinical trials for node-positive or ER-negative breast cancer. TAILORx is a prospective randomized clinical trial using Oncotype DX to help determine the benefit of chemotherapy in women diagnosed with ER-positive, node-negative breast cancer. Women with such a diagnosis will undergo conventional surgical treatment and radiation as clinically indicated. If their Oncotype DX score places them in the predefined low-risk category, they will receive antiendocrine treatment alone. If they are considered high risk, they will receive a standard chemotherapy regimen as determined by their treating physician. If they fall into the intermediate category, they will be selected for randomization to antiendocrine treatment alone or antiendocrine and chemotherapy combination therapy.[31] I look forward to the results.

Dr. Bowles: Are there other gene-profiling systems available for breast cancer at this time?

Dr. Singh: Many profiles have been written about in the literature, but Oncotype DX is the most widely used breast cancer gene-profiling system available in the United States. One advantage of Oncotype DX is that it can be done on paraffin-embedded tissue, whereas many of the other systems need to be performed on fresh frozen tissue.

Dr. Elias: Of course, one of the reasons that ER-assays that required fresh frozen tissue fell out of favor was because the amount of tissue needed to do the assays was now larger than some of the tumors we were removing. It is unclear to me how much fresh frozen tissue will be needed for the new gene-profile assays. One commercialized assay using fresh frozen tissue is the MammaPrint, a test developed in Europe.[32,33] It's now being tested in the Microarray in Node Negative Disease may Avoid ChemoTherapy (MINDACT) trial. This investigation is categorizing women with early-stage breast cancer as high-risk or low-risk, either by clinical analysis or gene-expression analysis, and randomizing them depending on their risk groups. It will be an interesting trial.

Treatment Recommendations

Dr. Bowles: We have discussed the potential prognostic and therapeutic ramifications of this patient's micrometastatic nodal disease and intermediate Oncotype DX recurrence score. I was wondering if we could offer some treatment recommendations for women in this position.

Dr. Finlayson: For further staging purposes, current standard practice would be to offer this woman a completion axillary dissection. However, this may not improve her prognosis and could provide significant morbidity.

Dr. Rabinovitch: I agree that a completion axillary dissection would be standard care. I do think, however, that treating this woman with axillary radiation would be a very reasonable alternative for controlling her axillary disease if further pathologic information does not affect anyone's decision about systemic therapy options.

Dr. Urquhart: Regarding reducing the risk of distant recurrence, I would recommend adjuvant chemotherapy and I would use a taxane-containing regimen. I would treat her as a woman with node-positive breast cancer.

Dr. Borges: This patient would be eligible for inclusion in many node-positive chemotherapy trials, such as NSABP B-38 or Southwest Oncology Group (SWOG) S0221. She would also be eligible for inclusion in one of the studies looking at the role of ovarian ablation with antiendocrine therapy, such as the Tamoxifen and Exemestane Trial (TEXT), which is open here at the University of Colorado. She is, of course, eligible for most clinical trials only if she has a completion node dissection.

Outside of a clinical trial, I would treat her with a "third-generation" chemotherapy regimen that includes an anthracycline and taxane, such as docetaxel, cyclophosphamide, and doxorubicin or dose-dense doxorubicin and cyclophosphamide followed by paclitaxel. I would also treat her with tamoxifen for 5 years, given her young age. If her menses did not cease with chemotherapy, then I would discuss with her the possibility of adding ovarian ablation at that time, as previous data have suggested that women who maintain menstrual function may have a slightly higher risk of disease recurrence than those who do not.[34] I would need to incorporate her personal desires for future childbearing potential and quality-of-life concerns before embarking on ovarian ablation.

Follow-up

The pros and cons of aggressive therapy were discussed. Arguing against chemotherapy were minimal nodal disease and a small, strongly ER/PR-positive tumor. Arguing for aggressive therapy were her young age, high intermediate Oncotype DX recurrence score, and N1mi status. Ultimately, her age and generally more aggressive tumor persuaded the patient to pursue an aggressive course of treatment.

She was enrolled in NSABP B-38, which did not require a completion axillary dissection, and randomized to dose-dense doxorubicin and cyclophosphamide followed by paclitaxel and an antiestrogen therapy. This will be followed by radiation to the breast and axilla. Should her menses continue after chemotherapy and antiestrogen therapy are initiated, she will undergo ovarian suppression or ablation.

Financial Disclosure:The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.

References:

1. Greene FL, Page DL, Fleming ID, et al (eds): AJCC Cancer Staging Manual, 6th ed. New York, Springer, 2002.

2. Lyman GH, Giuliano AE, Somerfield MR, et al: American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 23:7703-7720, 2005.

3. Van Zee KJ, Manasseh DM, Bevilacqua JL, et al: A nomogram for predicting the likelihood of additional nodal metastases in breast cancer patients with a positive sentinel node biopsy. Ann Surg Oncol 10:1140-1151, 2003.

4. Alran S, De Rycke Y, Fourchotte V, et al: Validation and limitations of use of a breast cancer nomogram predicting the likelihood of non-sentinel node involvement after positive sentinel node biopsy (abstract 2002). Breast Cancer Res Treat 100(suppl 1), 2006.

5. Schrenk P, Konstantiniuk P, Wolfl S, et al: Prediction of non-sentinel lymph node status in breast cancer with a micrometastatic sentinel node. Br J Surg 92:707-713, 2005.

6. Masci G, Di Tommaso L, Del Prato I, et al: Micrometastasis in sentinel (SLN) and non-sentinel lymph nodes of breast cancer: An update including clinico-pathologic impact and survival (abstract 10602). J Clin Oncol 24(18S):576s, 2006.

7. Fournier K, Schiller A, Perry RR, et al: Micrometastasis in the sentinel lymph node of breast cancer does not mandate completion axillary dissection. Ann Surg 239:859-863, 2004.

8. Cserni G, Gregori D, Merletti F, et al: Meta-analysis of non-sentinel node metastases associated with micrometastatic sentinel nodes in breast cancer. Br J Surg 91:1245-1252, 2004.

9. Klevesath1 MB, Bobrow LG, Pinder SE, et al: The value of immunohistochemistry in sentinel lymph node histopathology in breast cancer. Br J Cancer 92: 2201-2205, 2005.

10. Dowlatshahi K, Fan M, Snider HC, et al: Lymph node micrometastases from breast carcinoma: Reviewing the dilemma. Cancer 80:1188-1197, 1997.

11. Cserni C, Amendoeira I, Apostolikas N, et al: Pathologic work-up of sentinel lymph nodes in breast cancer: Review of current data to be considered for the formulation of guidelines. Eur J Cancer 39:1654-1667, 2003.

12. Chung MA, Steinhoff MM, Cady B: Clinical axillary recurrence in breast cancer patients after a negative sentinel node biopsy. Am J Surg 184:310-314, 2002.

13. Calhoun KE, Hansen NM, Turner RR, et al: Nonsentinel node metastases in breast cancer patients with isolated tumor cells in the sentinel node: Implications for completion axillary node dissection. Am J Surg 190:588-591, 2005.

14. Chagpar A, Middleton LP, Sahin AA, et al: Clinical outcome of patients with lymph node-negative breast carcinoma who have sentinel lymph node micrometastases detected by immunohistochemistry. Cancer 103:1581-1586, 2005.

15. Hansen NM, Grube BJ, Te W, et al: Clinical significance of axillary micrometastases in breast cancer: How small is too small (abstract 91)? Proc Am Soc Clin Oncol 20:24a, 2001.

16. Houvenaeghel G, Nos C, Mignotte H, et al: Micrometastases in sentinel lymph node in a multicentric study: Predictive factors of nonsentinel lymph node involvement-Groupe des Chirurgiens de la Federation des Centres de Lutte Contre le Cancer. J Clin Oncol 24:1814-1822, 2006.

17. Galper S, Recht A, Silver B, et al: Is radiation alone adequate treatment to the axilla for patients with limited axillary surgery? Implications for treatment after a positive sentinel node biopsy. Int J Radiat Oncol Biol Phys 48:125-132, 2000.

18. Louis-Sylvestre C, Clough K, Asselain B, et al: Axillary treatment in conservative management of operable breast cancer: Dissection or radiotherapy? Results of a randomized study with 15 years of follow-up. J Clin Oncol 22:97-101, 2004.

19. Noguchi M: Therapeutic relevance of breast cancer micrometastases in sentinel lymph nodes. Br J Surg 89:1505-1515, 2002.

20. Colleoni M, Rotmensz N, Peruzzotti G, et al: Size of breast cancer metastases in the axillary lymph nodes: Clinical relevance of minimal lymph node involvement. J Clin Oncol 23:1379-1389, 2005.

21. Overgaard M, Jensen MB, Overgaard J, et al: Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet 353:1641-1648,1999.

22. Ragaz J, Olivotto IA, Spinelli JJ, et al: Locoregional radiation therapy in patients with high-risk breast cancer receiving adjuvant chemotherapy: 20-year results of the British Columbia randomized trial. J Natl Cancer Inst 97:116-126, 2005.

23. 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 351:2817-2826, 2004.

24. 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 8:R25, 2006.

25. 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 24: 3726-3734, 2006.

26. 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 11:8623-8631, 2005.

27. 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. 23:7265-7277, 2005.

28. Mina L, Soule SE, Badve S, et al: Predicting response to primary chemotherapy: Gene expression profiling of paraffin-embedded core biopsy tissue. Breast Cancer Res Treat 103:197-208, 2007.

29. Goldstein LJ, Gray R, Childs BH, et al: Prognostic utility of the 21-gene assay in hormone receptor (HR) positive operable breast cancer and 0-3 positive axillary nodes treated with adjuvant chemohormonal therapy (CHT): An analysis of Intergroup trial E2197 (abstract 526). J Clin Oncol 25(18S):9s, 2007.

30. Mamounas E, Tang G, Bryant J, et al: Association between the 21-gene recurrence score assay (RS) and risk of locoregional failure in node-negative, ER-positive breast cancer: Results from NSABP B-14 and NSABP B-20 (abstract 29). Breast Cancer Res Treat 94(suppl 1):S16, 2005.

31. National Cancer Institute: The TAILORx breast cancer trial. Available at www.cancer.gov/clinicaltrials/digestpage/TAILORx. Accessed August 10, 2007.

32. van't Veer LJ, Dai H, van de Vijver MJ, et al: Gene expression profiling predicts clinical outcome of breast cancer. Nature 415:530-536, 2002.

33. van de Vijver MJ, He YD, van't Veer LJ, et al: A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 347:1999-2009, 2002.

34. Pagani O, O'Neill A, Castiglione M, et al: Prognostic impact of amenorrhoea after adjuvant chemotherapy in premenopausal breast cancer patients with axillary node involvement: Results of the International Breast Cancer Study Group (IBCSG) Trial VI. Eur J Cancer 34:632-640, 1998.