This Article Abstract Full Text (PDF) Erratum (v17,p3365) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jatoi, I. Articles by Osborne, C. K. Search for Related Content PubMed PubMed Citation Articles by Jatoi, I. Articles by Osborne, C. K.

Significance of Axillary Lymph Node Metastasis in Primary Breast Cancer

From the Department of Surgery, Brooke Army Medical Center, and Department of Medical Oncology, University of Texas Health Science Center at San Antonio, San Antonio, TX.

Address reprint requests to Ismail Jatoi, MD, PhD, Department of Surgery, Brooke Army Medical Center, 3851 Roger Brooke Dr, San Antonio, TX 78234-6200.

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: Axillary lymph node status is the single most important prognostic variable in the management of patients with primary breast cancer. Yet, it is not known whether metastasis to the axillary nodes is simply a time-dependent variable or also a marker for a more aggressive tumor phenotype. The purpose of this study was to determine whether nodal status at initial diagnosis predicts outcome after relapse and therefore also serves as a marker of breast cancer phenotype.

PATIENTS AND METHODS: Survival experience after first relapse in 1,696 primary breast cancer cases was analyzed using Cox proportional hazards regression. The following explanatory variables and their first-order interactions were considered: number of axillary lymph nodes involved (zero v one to three v four or more), hormone receptor status (any estrogen receptor [ER] negativity v ER negativity/progesterone receptor positivity v other ER positivity), primary tumor size (< 2 cm v 2 to 5 cm v > 5 cm), site of relapse (locoregional v distant), disease-free interval (< 1.5 years v 1.5 to 3 years v > 3 years), adjuvant endocrine therapy (none v any), adjuvant chemotherapy (none v any), and menopausal status (pre-, peri-, or postmenopausal).

RESULTS: Axillary lymph node status, site of relapse, and hormone receptor status were all highly significant as main effects in the model. After adjustment for other variables, disease-free interval alone was only modestly significant but interacted with nodal status. After disease-free interval, hormone receptor status, and site of relapse were accounted for, survival after relapse was poorer in node-positive cases, when compared with node-negative cases. The hazard ratios for patients with one to three and four or more involved nodes were 1.2 (95% confidence interval [CI], 0.8 to 1.9) and 2.5 (95% CI, 1.8 to 3.4), respectively.

CONCLUSION: Patients with four or more involved nodes at initial diagnosis have a significantly worse outcome after relapse than node-negative cases, regardless of the duration of the disease-free interval. We conclude that nodal metastasis is not only a marker of diagnosis at a later point in the natural history of breast cancer but also a marker of an aggressive phenotype.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
NODE-POSITIVE BREAST cancers have a worse prognosis than node-negative cases.¹ However, the significance of nodal metastasis is poorly understood. Late in the 19th century, Halsted² proposed that breast cancer spreads first to the axillary lymph nodes and then to distant sites. Thus, nodal metastasis was viewed as an indicator of tumor chronology. The better prognosis of node-negative tumors was attributed to timely resection, before distant metastasis via the axillary lymphatics had occurred.³

In more recent years, large randomized trials have shown that neither the extent of the mastectomy nor delay in the treatment of the axilla has any influence on the prognosis of patients with operable breast cancer.^4-6 In addition, long-term follow-up of node-negative patients reveals that 30% eventually die of metastatic breast cancer.⁷ Thus, the axilla does not seem to serve as a nidus for further spread of the cancer, as postulated by Halsted. Yet, nodal status is still considered an indicator of tumor chronology, and the better prognosis of node-negative patients is generally attributed to lead time bias.⁸ However, an alternative possibility may account for the difference in prognosis between node-negative and node-positive patients. Nodal status is perhaps also a marker of tumor biology, with node-positive tumors having a more aggressive phenotype.

In this study, we set out to better understand the significance of nodal metastasis through a multivariate analysis of the survival data from the San Antonio Tumor Bank (San Antonio, TX). To reduce the uncertainty associated with lead time bias, we correlated nodal status at initial diagnosis with outcome after relapse. If nodal status is, at least in part, a marker of tumor phenotype, then node-positive patients should have a shorter interval from first relapse to death than node-negative patients. Alternatively, if nodal status is simply a time-dependent variable, then it should not predict outcome after relapse.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient Population
The patients in this study were identified from a large database of patients who had hormone receptor assays or flow cytometry performed in our laboratory (University of Texas Health Science Center at San Antonio [UTHSCSA]). Eligible patients were initially diagnosed between 1970 and 1991 and presented with primary breast cancer treated with either radical or modified radical mastectomy, or lumpectomy and axillary node dissection with postoperative radiation therapy. These patients had no evidence of distant metastases at the time of diagnosis and were followed, as previously described,⁹ for relapse (defined as the first clinically recognized evidence of local or distant recurrence) and survival. Of 2,156 patients known to have relapsed locally, distantly, or both, we omitted 132 whose cancer recurred less than 6 months after the initial diagnosis and another 72 for whom we had no follow-up after relapse. Additional cases were omitted due to missing data (estrogen receptor [ER] results not available from the primary tumor, n = 178; menopausal status unknown, n = 61; and adjuvant therapy unknown, n = 17). The final data set, comprising 1,696 relapsed primary breast cancer cases, is summarized in Table 1.

Survival time is defined as time from relapse to last contact or death. As of June 1998, there have been 1,207 deaths, and median follow-up after relapse among living cases was 2.3 years. Survival of 35 cases still alive more than 10 years after relapse was truncated to 10 years for display in the survival curves.

Prognostic Variables
A number of potentially prognostic clinical and biologic variables were available. ER and progesterone receptor (PgR) status had been measured by ligand binding assay in our laboratory. Tumors with 3 fmol/mg protein of ER or 5 fmol/mg protein of PgR are considered hormone receptor–positive. Primary tumor size and number of positive axillary lymph nodes were abstracted from medical records. The disease-free interval (DFI), defined as the time from diagnosis to first relapse, was divided roughly at the tertiles to create DFI groups. About a third of relapses were locoregional (ie, chest wall, axillary, etc), while two thirds were distant (ie, liver, lung, bone, etc) or a combination of multiple sites.

Statistical Analysis
Univariate survival curves were calculated using the Kaplan-Meier method and compared using the log-rank test.^10,11 Multivariate analysis of survival was performed using Cox proportional hazards regression.¹² The purpose of the analysis was to examine the contribution of positive lymph nodes to survival after relapse, after accounting for the potential effects of other explanatory factors. Explanatory variables were treated as categorical variables and modeled using dummy variables, coded 0 or 1. The following main effects, and their first-order interactions, were considered: hormone receptor status (any ER- v ER+/PgR- v ER+/PgR+ or ER+/PgR unknown), number of positive axillary lymph nodes (zero v one to three v four or more), primary tumor size (< 2 cm v 2 to 5 cm v > 5 cm), DFI (< 1.5 years v 1.5 to 3 years v > 3 years), site of relapse (locoregional or any distant site), adjuvant endocrine therapy (none v any), adjuvant chemotherapy (none v any), menopausal status (premenopausal + perimenopausal v postmenopausal). The Cox model was pruned using hierarchical backward elimination, which requires removal of interaction terms before removal of associated main effects. Elimination stopped when all remaining terms eligible for removal from the model were significant at the alpha = 0.01 level. Initially, the data set was split in half randomly, with one subset being used for model pruning. The final pruned model was then fit to the second subset. Because the results were virtually identical, we show the final pruned model refit to the entire data set. Proportional hazards assumptions were tested using graphical methods^13,14 and were not violated. Model results were summarized by computing model estimates and approximate 95% confidence intervals (CIs) of 2-year survival after relapse for all possible combinations of the explanatory variables in the final model. Analyses were performed using SAS software version 6.12 (SAS, Inc, Cary, NC).

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Univariate Analysis
The characteristics of the 1,696 cases included in this study are summarized in Table 1. At diagnosis, they were predominantly postmenopausal (71.7%) with ER+ tumors (73.2%). Tumor sizes tended to be greater than 2 cm (70%), and the majority of cases were lymph node–positive (63.3%). The median DFI was 2.2 years, and 68.6% of relapses included a distant site. A majority of patients were diagnosed before 1984 (76.7%) and did not receive adjuvant chemo- or endocrine therapy (62.2% and 69.2%, respectively).

Two-year survival after first relapse was 51% (95% CI, 48% to 63%). In univariate survival analysis, only menopausal status was not at least marginally associated with survival after first relapse (Table 1). Survival curves for number of axillary lymph nodes and type of relapse are shown in Figs 1 and 2. Hormone receptor status, number of axillary lymph nodes, tumor size, adjuvant chemotherapy, DFI, and site of relapse were all highly significant, with the worst outcome being associated with positive nodes, large tumor size, adjuvant chemotherapy, short DFI, and distant relapse. Adjuvant chemotherapy may be associated with poorer postrelapse outcome because patients with biologically aggressive disease were selected for treatment, or possibly because tumors that relapsed after therapy were more resistant to further therapy. Endocrine adjuvant therapy (or selection of patients as candidates for therapy) was only modestly associated with a worse outcome. Similarly, patients diagnosed after 1984 (n = 395) did slightly worse after relapse (P = .034). This is likely due to the association between short DFI and poorer survival and to the association between chemotherapy and poorer survival. Patients with short DFIs are slightly overrepresented in the later cohort (P = .014 by ² test for independence). Similarly, more patients in the later cohort were treated with adjuvant chemotherapy (P = .001 by ² test for independence).

View larger version (17K): [in this window] [in a new window]   Fig 1. Survival after first relapse by number of positive axillary lymph nodes.

View larger version (16K): [in this window] [in a new window]   Fig 2. Survival after first relapse by site of relapse.

Multivariate Analysis
Cox regression was used to examine the independent prognostic value of each variable after adjusting for potential confounding effects of other variables. Variables comprising two groups, such as site of relapse, were represented by a single indicator variable taking on values of 0 or 1, whereas variables comprising three groups (eg, tumor size) were represented by two indicator variables, and therefore contribute two coefficients to the regression equation and two degrees of freedom. Interactions were represented by products of the corresponding main effects. The full range of possible interaction effects was therefore represented by combining main effect and interaction terms. The final pruned model (Table 2) was constructed by beginning with a full model containing terms for each prognostic variable and all of the corresponding two-factor interactions and progressively removing nonsignificant terms.

Site of relapse, number of axillary lymph nodes, and hormone receptor status were all highly significant as main effects in the model (Table 2). The risk of death after relapse was more than double in patients with a distant relapse compared with those with a local relapse (hazard ratio, 2.1) and similarly increased in patients who were diagnosed with four or more positive lymph nodes compared with those who were node-negative at diagnosis. Patients whose primary tumors were ER+/PgR+ had half the risk of those with ER- tumors. After adjustment for other variables, DFI alone was not significant as a main effect but interacted strongly with nodal status. A long DFI mitigates the otherwise deleterious effects of having four or more nodes.

There were no other significant interactions detected, and tumor size, menopausal status, adjuvant endocrine and chemotherapy, and all associated interactions were not significant and were eliminated from the model. Although the level of significance for retention was set ahead of time at alpha = 0.01, no other terms achieved even the 0.05 level. Several representations of hormone receptor status were considered, including ER status alone, with nearly identical results to those above.

Cox model estimates of 2-year survival after relapse and 95% CIs are summarized in Table 3 and Fig 3. Regardless of other factors, survival probability was lower for hormone receptor–negative cases and for those with distant relapse. Overall, DFI had relatively little effect, but the effect of the nodes x DFI interaction is easily seen as a difference in the shapes of the curves across the panels. For short DFI, the prognosis for those with four or more nodes was markedly worse than for those with fewer nodes at diagnosis (Fig 3A). For patients with a long DFI (Fig 3C), there was less differential due to nodes, and the number of nodes involved was not important. Patients with an intermediate DFI exhibited an intermediate pattern.

View larger version (57K): [in this window] [in a new window]   Fig 3. Estimated probability of 2-year survival after first relapse and approximate 95% CIs (shown as vertical bars) as a function of axillary lymph node status, hormone receptor status, site of relapse (—, local; – – –, any distant), and DFI of (A) less than 1.5 years, (B) 1.5 to 3 years, or (C) more than 3 years.

After DFI, hormone receptor status, and site of relapse were accounted for, the predicted survival after first relapse of node-negative cases was always better than that of cases with four or more nodes at diagnosis, suggesting that lymph node involvement is, at least in part, a marker of biologic behavior that continues to exert an effect, even after relapse.

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mueller¹⁵ analyzed the annual rates of death from the National Surgical Adjuvant Breast and Bowel Project 04 study and Connecticut Tumor Registry and found that they were consistently higher for node-positive patients than for node-negative patients. Therefore, he postulated that the two stages of breast cancer represented biologic variants of the same disease. However, his analysis did not account for the confounding effects of other prognostic factors on survival, and the possibility that nodal status was a surrogate for these other variables could not be excluded. Subsequently, Mittra and MacRae¹⁶ undertook a meta-analysis of published correlations between various prognostic factors in breast cancer and concluded that axillary lymph node status is simply a reflection of the chronologic age of the tumor. Yet, a meta-analysis has pitfalls, many of which were acknowledged by the authors. These include publication bias (journals tend to report only positive findings) and variability in laboratory analysis criteria (for instance, different laboratories adopt different criteria as to what constitutes ER positivity and ER negativity). In order to account for the shortcomings of the two previous studies, we undertook a multivariate analysis using a single, large database. All laboratory data were generated from the same institution, thereby accounting for the possible confounding effects of other prognostic factors on survival and yet eliminating the deficiencies of the meta-analysis.

The clinical history of breast cancer is often marked by four milestones: inception, diagnosis, distant relapse, and death.⁸ The time of inception is not known, and therefore the interval between inception and diagnosis is not known and probably highly variable between patients. If the survival advantage of node-negative patients is due to lead time bias, then these patients may seem to do better only because their cancers are diagnosed earlier than in the node-positive group, resulting in a longer follow-up time to death. To reduce the uncertainty associated with lead time bias, we correlated nodal status at initial diagnosis with the interval between two known points in the clinical history of breast cancer: relapse and death. A long interval would suggest a biologically indolent tumor, while a shorter interval would indicate a more aggressive tumor phenotype. If nodal status is simply a marker of delay in diagnosis, then it should not correlate with outcome after relapse.

The importance of nodal status in predicting outcome after relapse is controversial. Some investigators have reported that nodal status has prognostic importance after relapse.^17-19 However, others have found no correlation.^20-24 There were relatively few patients in most of these studies, and the relationship between the number of involved nodes and prognosis after relapse was generally not examined (patients were categorized simply as node-positive or node-negative). More recently, several authors have reviewed the literature on this subject and concluded that nodal status has no value in predicting outcome after relapse and is therefore simply an indicator of tumor chronology.^8,25,26 Several years ago, a report from our institution suggested that nodal status does have prognostic importance after relapse.¹⁹ In the present study, we evaluate an expanded cohort of patients from that database and correlate the extent of nodal involvement with outcome after relapse.

Patients with large, hormone receptor–negative or node-positive tumors are often selected to receive adjuvant systemic therapy. Thus, it is not surprising that patients selected for adjuvant systemic therapy have a worse outcome after relapse in the univariate analysis but not the multivariate model, where the effect of these other confounding variables is taken into account. Similarly, the size of the primary tumor correlates with outcome after relapse in the univariate analysis but not the multivariate model. Therefore, one might speculate that tumor size is a surrogate for nodal status. In the multivariate model, only nodal status, site of relapse, and hormone receptor status are independent predictors of outcome after relapse. After these main effects and the interaction between number of nodes involved and DFI are accounted for, nodal status remains a significant predictor of outcome after relapse. Indeed, when compared with node-negative patients, those with four or more involved nodes have a significantly worse outcome after relapse. However, for patients with only one to three involved nodes, the outcome is not significantly different from that of the node-negative patients. Thus, the number of involved nodes (rather than simply the absence or presence of nodal involvement) is a key determinant of prognosis after relapse.

The risk of axillary lymph node metastasis increases as tumor size increases, which suggests that nodal metastasis is indicative of tumor chronology.^27,28 Yet, our study suggests that nodal status has prognostic importance after relapse, indicating that it is also a marker for tumor phenotype. These findings are not necessarily inconsistent. One might speculate that there is a continuum from slow-growing tumors with late axillary involvement to more aggressive tumors with early metastasis to the axilla.²⁹ A 1-cm, node-positive tumor might be chronologically early but biologically more aggressive when compared with a 2-cm, node-negative tumor. Thus, all breast cancers may eventually metastasize to the axillary nodes, with the propensity for early or late metastasis having prognostic importance after relapse. Therefore, nodal status may indicate both tumor chronology and phenotype.

According to the theory of cancer screening, promulgated by Cole and Morrison,³⁰ the first round of any screening program (prevalent screen) should detect a greater fraction of indolent tumors, with a longer preclinical phase, while a greater percentage of aggressive tumors with a shorter preclinical phase should be detected in the screening rounds that follow (incident screens). In the Greater Manchester Breast Cancer Screening Unit, the prevalent screen detected a larger percentage of node-negative tumors, even though size of the average was greater than that of tumors detected at the incident screens.³¹ A similar trend was observed in the Edinburgh Breast Cancer Screening Trial.³² Taken together, the theory of cancer screening and the results from the screening centers underscore the importance of nodal metastasis as a marker of tumor phenotype and support the findings of our study.

Since the time of Halsted, clinicians have generally regarded nodal metastasis as a time-dependent variable. This view is too simplistic, and the relevance of nodal metastasis to tumor phenotype merits additional investigation. Nodal metastasis may indicate a highly malignant tumor or serve as a marker of host response. Indeed, one might speculate that a weakened host response results in early metastasis to the axillary lymph nodes and a poorer prognosis. Further studies may clarify these issues and provide important insights into cancer growth and control.

	ACKNOWLEDGMENTS

 
Supported by National Institutes of Health grants no. PO1 CA 30195 and P50 CA58183 and Cancer Center Support grant no. P30 CA54174.

	NOTES

 
The opinions or assertions contained herein are the private views of the authors and are not to be construed as reflecting the views of the Departments of the Army, Air Force, or Defense.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Dent DM: Axillary lymphadenectomy for breast cancer. Arch Surg131:1125-1127, 1996[Medline]

2. Halsted WS: A clinical and histological study of certain adenocarcinoma of the breast and a brief consideration of the supraclavicular operation and of the results of operation for cancer of the breast from 1889 to 1898 at the Johns Hopkins Hospital. Ann Surg28:557-576, 1898[Medline]

3. Eggers C: Cancer surgery: The value of radical operations for cancer after the lymphatic drainage area has become involved. Ann Surg106:668-689, 1937[Medline]

4. Wood WC: Progress from clinical trials on breast cancer. Cancer74:2606-2609, 1994 (suppl 9) [Medline]

5. Fisher B, Redmond C, Fisher ER, et al: Ten year results of a randomized clinical trial comparing radical mastectomy and total mastectomy with or without irradiation. N Engl J Med312:674-681, 1985[Abstract]

6. Cancer Research Campaign Working Party: Cancer Research Campaign (King's/Cambridge): Trial for Breast Cancer. Lancet2:55-60, 1980[Medline]

7. Bonadonna G: Evolving concepts in the systemic adjuvant treatment of breast cancer. Cancer Res52:2127-2137, 1992[Free Full Text]

8. Mittra I: Axillary lymph node metastasis in breast cancer: Prognostic indicator or lead-time bias? Eur J Cancer29:300-302, 1993

9. Clark GM, Dressler LG, Owens MA, et al: Prediction of relapse or survival in patients with node negative breast cancer by DNA flow cytometry. N Engl J Med320:627-633, 1989[Abstract]

10. Kaplan EL, Meier PL: Nonparametric estimation from incomplete observations. J Am Stat Assoc53:457-481, 1958

11. Cox DR, Oakes D: Analysis of survival data. New York, NY, Chapman and Hall, 1984

12. Cox DR: Regression models and life-tables (with discussion). J R Stat Soc B34:187-200, 1972

13. Grambsch PM, Therneau TM: Proportional hazards tests and diagnostics based on weighted residuals. Biometrika81:515-526, 1994[Abstract/Free Full Text]

14. Therneau TM, Grambsch PM, Fleming TR: Martingale-based residuals for survival models. Biometrika77:147-160, 1990[Abstract/Free Full Text]

15. Mueller CB: Stage II breast cancer is not simply a late stage I. Surgery104:631-638, 1988[Medline]

16. Mittra I, MacRae KD: A meta-analysis of reported correlations between prognostic factors in breast cancer: Does axillary lymph node metastasis represent biology or chronology? Eur J Cancer27:1574-1583, 1991

17. Shek LLM, Godolphin W, Spinelli JJ: Oestrogen receptors, nodes and stage as predictors of post-recurrence survival in 457 breast cancer patients. Br J Cancer56:825-829, 1987[Medline]

18. Lionetto R, Pronzato P, Bertelli GF, et al: Survival of patients with relapsing breast cancer: Analysis of 302 patients. Oncology43:278-282, 1986[Medline]

19. Clark GM, Sledge GW Jr Osborne CK, et al: Survival from first recurrence: Relative importance of prognostic factors in 1,015 breast cancer patients. J Clin Oncol5:55-61, 1987[Abstract]

20. Hahnel R, Woodings T, Vivian AB: Prognostic value of estrogen receptors in primary breast cancer. Cancer44:671-675, 1979[Medline]

21. Pater JL, Mores D, Loeb M: Survival after recurrence of breast cancer. Can Med Assoc J124:1591-1595, 1981[Abstract]

22. Howell A, Harland RNL, Bramwell VHC, et al: Steroid-hormone receptors and survival after first relapse in breast cancer. Lancet1:588-591, 1984[Medline]

23. Howat JMT, Harris M, Swindell R, et al: The effect of oestrogen and progesterone receptors on recurrence and survival in patients with carcinoma of the breast. Br J Cancer51:262-270, 1985

24. Williams MR, Todd JH, Nicholson RI, et al: Survival patterns in hormone treated advanced breast cancer. Br J Surg73:752-755, 1986[Medline]

25. Anonymous: Prognostic factors in breast cancer: Biology or chronology? Lancet340:517-518, 1992 (editorial) [Medline]

26. Tubiana-Hulin M, Hacene K, Martin PM, et al: Prognostic factor clustering in breast cancer: Biology or chronology? Eur J Cancer 31A:283-284, 1995

27. Hartveit F: Axillary metastasis in breast cancer: When, how, and why? Semin Surg Oncol5:126-136, 1989[Medline]

28. Tabar L, Gad A, Holmberg LH, et al: Reduction in mortality from breast cancer after mass screening with mammography. Lancet1:829-832, 1985[Medline]

29. Koscielny S, Le MG Tubiana M: The natural history of breast cancer: The relationship between involvement of axillary lymph nodes and the initiation of distant metastases. Br J Cancer59:775-782, 1989[Medline]

30. Cole P, Morrison AS: Basic issues in population screening for cancer. J Natl Cancer Inst64:1263-1272, 1980

31. Holland PA, Walls J, Boggis CRM, et al: A comparison of axillary node status between cancers detected at the prevalence and first incidence breast screening rounds. Br J Cancer74:1643-1646, 1996[Medline]

32. Anderson TJ, Alexander F, Chetty U, et al: Comparative pathology of prevalent and incident cancers detected by breast cancer screening. Lancet1:519-522, 1986[Medline]

Submitted August 11, 1998; accepted April 19, 1999.

This article has been cited by other articles:

				M. Schmidt, A. Victor, D. Bratzel, D. Boehm, C. Cotarelo, A. Lebrecht, W. Siggelkow, J. G. Hengstler, A. Elsasser, M. Gehrmann, et al. Long-term outcome prediction by clinicopathological risk classification algorithms in node-negative breast cancer--comparison between Adjuvant!, St Gallen, and a novel risk algorithm used in the prospective randomized Node-Negative-Breast Cancer-3 (NNBC-3) trial Ann. Onc., September 29, 2008; (2008) mdn590v1. [Abstract] [Full Text] [PDF]

				J. C. Harrell, W. W. Dye, D. M.E. Harvell, M. Pinto, P. Jedlicka, C. A. Sartorius, and K. B. Horwitz Estrogen Insensitivity in a Model of Estrogen Receptor Positive Breast Cancer Lymph Node Metastasis Cancer Res., November 1, 2007; 67(21): 10582 - 10591. [Abstract] [Full Text] [PDF]

				T. Nakagawa, S. R. Martinez, Y. Goto, K. Koyanagi, M. Kitago, T. Shingai, D. A. Elashoff, X. Ye, F. R. Singer, A. E. Giuliano, et al. Detection of Circulating Tumor Cells in Early-Stage Breast Cancer Metastasis to Axillary Lymph Nodes Clin. Cancer Res., July 15, 2007; 13(14): 4105 - 4110. [Abstract] [Full Text] [PDF]

				T. Nakagawa, S. K. Huang, S. R. Martinez, A. N. Tran, D. Elashoff, X. Ye, R. R. Turner, A. E. Giuliano, and D. S.B. Hoon Proteomic Profiling of Primary Breast Cancer Predicts Axillary Lymph Node Metastasis Cancer Res., December 15, 2006; 66(24): 11825 - 11830. [Abstract] [Full Text] [PDF]

				S.-C. Chen, H.-K. Chang, Y.-C. Lin, W.-M. Leung, C.-S. Tsai, Y.-C. Cheung, S. Hsueh, L.-C. See, and M.-F. Chen Prognosis of Breast Cancer After Supraclavicular Lymph Node Metastasis: Not a Distant Metastasis Ann. Surg. Oncol., November 1, 2006; 13(11): 1457 - 1465. [Abstract] [Full Text] [PDF]

				J. C. Harrell, W. W. Dye, D. C. Allred, P. Jedlicka, N. S. Spoelstra, C. A. Sartorius, and K. B. Horwitz Estrogen receptor positive breast cancer metastasis: altered hormonal sensitivity and tumor aggressiveness in lymphatic vessels and lymph nodes. Cancer Res., September 15, 2006; 66(18): 9308 - 9315. [Abstract] [Full Text] [PDF]

				M. Schaapveld, E. G. E. de Vries, W. T. A. van der Graaf, R. Otter, J. de Vries, and P. H. B. Willemse The Prognostic Effect of the Number of Histologically Examined Axillary Lymph Nodes in Breast Cancer: Stage Migration or Age Association? Ann. Surg. Oncol., April 1, 2006; 13(4): 465 - 474. [Abstract] [Full Text] [PDF]

				M. Shinozaki, D. S.B. Hoon, A. E. Giuliano, N. M. Hansen, H.-J. Wang, R. Turner, and B. Taback Distinct Hypermethylation Profile of Primary Breast Cancer Is Associated with Sentinel Lymph Node Metastasis Clin. Cancer Res., March 15, 2005; 11(6): 2156 - 2162. [Abstract] [Full Text] [PDF]

				J E A Somner, J M J Dixon, and J S J Thomas Node retrieval in axillary lymph node dissections: recommendations for minimum numbers to be confident about node negative status J. Clin. Pathol., August 1, 2004; 57(8): 845 - 848. [Abstract] [Full Text] [PDF]

				R. C. F. Leonard, M. Lind, C. Twelves, R. Coleman, S. van Belle, C. Wilson, J. Ledermann, I. Kennedy, P. Barrett-Lee, T. Perren, et al. Conventional Adjuvant Chemotherapy Versus Single-Cycle, Autograft-Supported, High-Dose, Late-Intensification Chemotherapy in High-Risk Breast Cancer Patients: A Randomized Trial J Natl Cancer Inst, July 21, 2004; 96(14): 1076 - 1083. [Abstract] [Full Text] [PDF]

				J. Pittman, E. Huang, H. Dressman, C.-F. Horng, S. H. Cheng, M.-H. Tsou, C.-M. Chen, A. Bild, E. S. Iversen, A. T. Huang, et al. Integrated modeling of clinical and gene expression information for personalized prediction of disease outcomes PNAS, June 1, 2004; 101(22): 8431 - 8436. [Abstract] [Full Text] [PDF]

				G. V. Glinsky, T. Higashiyama, and A. B. Glinskii Classification of Human Breast Cancer Using Gene Expression Profiling as a Component of the Survival Predictor Algorithm Clin. Cancer Res., April 1, 2004; 10(7): 2272 - 2283. [Abstract] [Full Text] [PDF]

				K. Suga, Y. Yuan, M. Okada, N. Matsunaga, A. Tangoku, S. Yamamoto, and M. Oka Breast Sentinel Lymph Node Mapping at CT Lymphography with Iopamidol: Preliminary Experience Radiology, February 1, 2004; 230(2): 543 - 552. [Abstract] [Full Text] [PDF]

				V. Cocquyt, K. Moeremans, L. Annemans, P. Clarys, and S. Van Belle Long-term medical costs of postmenopausal breast cancer therapy Ann. Onc., July 1, 2003; 14(7): 1057 - 1063. [Abstract] [Full Text] [PDF]

				M P G Leers, R H M G Schoffelen, J G M Hoop, P H M H Theunissen, J W A Oosterhuis, H v. Bijl, A Rahmy, W Tan, and M Nap Multiparameter flow cytometry as a tool for the detection of micrometastatic tumour cells in the sentinel lymph node procedure of patients with breast cancer J. Clin. Pathol., May 1, 2002; 55(5): 359 - 366. [Abstract] [Full Text] [PDF]

				Y. Nieto, S. Nawaz, R. B. Jones, E. J. Shpall, P. J. Cagnoni, P. A. McSweeney, A. Baron, C. Razook, S. Matthes, and S. I. Bearman Prognostic Model for Relapse After High-Dose Chemotherapy With Autologous Stem-Cell Transplantation for Stage IV Oligometastatic Breast Cancer J. Clin. Oncol., February 1, 2002; 20(3): 707 - 718. [Abstract] [Full Text] [PDF]

				P. J. Goodwin, M. Ennis, K. I. Pritchard, M. E. Trudeau, J. Koo, Y. Madarnas, W. Hartwick, B. Hoffman, and N. Hood Fasting Insulin and Outcome in Early-Stage Breast Cancer: Results of a Prospective Cohort Study J. Clin. Oncol., January 1, 2002; 20(1): 42 - 51. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Erratum (v17,p3365) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jatoi, I. Articles by Osborne, C. K. Search for Related Content PubMed PubMed Citation Articles by Jatoi, I. Articles by Osborne, C. K.