Annals of Oncology Advance Access published online on July 24, 2008
Annals of Oncology, doi:10.1093/annonc/mdn535
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Prognostic value of micrometastases in sentinel lymph nodes of patients with breast carcinoma: a cohort study
1 Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht
2 Department of Nuclear Medicine, Diakonessenhuis Utrecht
3 Department of Pathology, Diakonessenhuis Utrecht, The Netherlands
4 Department of Surgery, Diakonessenhuis Utrecht
* Correspondence to: Dr T. van Dalen, Department of Surgery, Diakonessenhuis Utrecht, Bosboomstraat 1, 3582 KE Utrecht, The Netherlands. Tel: +31-30-2566225; Fax: +31-30-2566210; E-mail: tvdalen{at}diakhuis.nl
 |
Abstract |
|---|
 Top Abstract introductionpatients and methodsresultsdiscussionReferences |
|---|
Background: The prognostic meaning and thus indication for adjuvant therapy of lymphogenic micrometastases in breast cancer patients is still under debate.
Patients and methods: From 1999 to 2007, 703 patients with cT1–2N0 breast cancer underwent surgery including sentinel lymph node biopsy. Examination of sentinel lymph nodes consisted of hematoxylin and eosin and immunohistochemistry staining following serial sectioning of the sentinel node. Patients were divided into four groups: pN0 (n = 423), pN1micro (n = 81), pN1a (n = 130) and pN
1b (n = 69). Median follow-up was 40 months.
Results: At the end of follow-up, 53 patients had died and 64 had recurrent disease. Compared with pN0 and following adjustment for possible confounders, including adjuvant systemic treatment, overall survival was not significantly different for pN1micro while significantly worse for pN1a and pN1b {hazard ratio (HR) [95% confidence interval (CI)]: 0.59 [0.14–2.58], 4.31 [1.85–10.01], 10.66 [4.04–28.14], respectively}. Likewise, disease-free survival was not significantly different for pN1micro and worse for pN1a and pN1b (HR [95% CI]: 1.43 [0.67–3.02], 2.79 [1.37–5.66], 7.13 [3.27–15.54], respectively). Distant metastases were more commonly observed in the pN1micro than in the pN0 group, but still not as common as in the pN1a or pN1b group (HR [95% CI]: 4.85 [1.79–13.18], 10.34 [3.82–28.00], 23.25 [7.88–68.56], respectively).
Conclusion: Although the risk of distant metastases was higher in patients in the pN1micro than in the pN0 group, no statistically significant differences were observed in overall or disease-free survival between pN0 and pN1micro. Micrometastatic lymph node involvement in itself should not be an indication for adjuvant chemotherapy in breast cancer patients.
breast cancer, micrometastases, prognosis, sentinel lymph node
|
introduction |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionReferences |
|---|
Axillary staging is a hallmark of breast cancer surgery as metastatic lymph node involvement is a strong prognosticator. The presence of lymphogenic metastases and number of involved lymph nodes contribute importantly to adjuvant systemic treatment decisions. In the era of the sentinel lymph node biopsy (SLNB), lymph nodes are assessed more thoroughly for tumor involvement than before. Consequently, the proportion of patients diagnosed with micrometastatic lymph node involvement (i.e. tumor deposits >0.2 and <2 mm) has increased [1–7], and micrometastases are observed in up to 23% of breast cancer patients [1, 6, 8].
These micrometastases pose a clinical dilemma with regard to adjuvant treatment decisions because their prognostic meaning is currently unclear. Most studies on this topic originate from the pre-SLNB era, are retrospective and yield inconsistent results. Some earlier reports indicated a prognosis of patients with TNM (tumor–node–metastasis) stage pN1micro disease comparable to patients without lymph node involvement (pN0) [9], while others observed a prognosis comparable to patients with TNM stage pN1a disease [10]. In this study from the SLNB era, we evaluated the association between pN1micro disease and clinical outcome compared with patients with pN0, pN1a and pN1b disease.
|
patients and methods |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionReferences |
|---|
From June 1999 to January 2007, 720 consecutive patients with clinically T1–2N0 primary invasive breast cancer underwent surgical treatment including an SLNB procedure at the Diakonessenhuis Utrecht (a large regional teaching hospital in The Netherlands). These women were all enrolled in the current study. Data regarding demography, preoperative lymphoscintigraphy, operative procedure, pathology results and adjuvant treatment administration were collected prospectively. At the time of the introduction of the sentinel node procedure, the ethical committee of the hospital approved the routine use of the SLNB as a staging procedure. All patients received written information regarding the SLN procedure.
Patients who presented with a synchronous contralateral breast cancer (n = 8) or with multifocal carcinoma (n = 9) were excluded to avoid difficulty of ascribing patient outcome to tumor-specific lymph node status. The cohort available for analysis consisted of 703 patients.
SLN procedure
On the day of the operation, all patients received a combination of peritumoral intraparenchymal and s.c. injections of an average of 77.7 MBq (53–150 MBq) 99mTc nanocolloid. Continuous visualization was done and imaging started as soon as lymphatic drainage was visualized on the persistence scope and at 2–3 h after injection in both the anterior and the lateral direction. The operative procedure was done in the afternoon of the same day. A 
-ray detection probe was intraoperatively used for SLN identification. During the operation, both axillary and internal mammary (IM) sentinel nodes were retrieved. Axillary SLNs were visualized on lymphoscintigraphy and collected by axillary exploration in 99% of the patients. IM SLNs were visualized in 22% of the patients, and in 78% of the patients with visualized IM SLNs they could be successfully retrieved through an intercostal exploration. Detailed information about this procedure was presented earlier by our study group [11].
pathologic examination of the SLN
Intraoperative frozen section analysis of the axillary SLNs was done to enable axillary dissection during the same operative procedure in case of overt lymph node metastases. In addition, both axillary and IM SLNs were formalin fixed and paraffin embedded and four cuts from both halves were taken at 250-µm intervals starting from the center. The sections were stained with both hematoxylin and eosin (H&E) and immunohistochemically with an antibody against keratin. When the axillary SLN contained metastases, a subsequent axillary dissection was carried out, either during the first operation when the frozen section analysis revealed metastases or as a second operation when the metastases were detected on the paraffin slices.
classification of lymph node stage
On the basis of the pathologic findings of the SLN and the axillary dissection specimens, lymph node status was classified according to the 6th edition of the International Union Against Cancer (UICC) TNM classification [12].
- pN0: no regional lymph node metastasis;
- pNitc: clusters of tumor cells in regional lymph nodes <0.2 mm;
- pN1micro: metastasis in axillary lymph nodes with a size between 0.2 and 2.0 mm;
- pN1a: 1–3 axillary macrometastasis (at least one with size >2.0 mm);
- pN1b: a single positive IM lymph node (diagnosed by SLNB, not clinically apparent);
- pN1c: a combination of pN1a and pN1b;
- pN2a: 4–9 ipsilateral axillary macrometastases;
- pN3a:
10 axillary macrometastases.
- pNitc: clusters of tumor cells in regional lymph nodes <0.2 mm;
According to the UICC TNM classification, pNitc was classified as pN0.
nonoperative treatment
On the basis of the axillary lymph node status and primary tumor characteristics, adjuvant systemic and/or radiotherapeutic treatment was given in accordance with the Dutch national guidelines [13]. The finding of macrometastatic disease in the axillary lymph nodes or IM lymph node metastases (TNM stage pN1a–c/pN2/pN3) were indications for adjuvant therapy. Furthermore, primary tumor size, hormonal receptor status [estrogen receptor (ER) and progesterone receptor (PR)], modified Bloom–Richardson (BR) malignancy grade (classified according to the Nottingham modification [14]) and patient age were taken into account.
In patients with pN1micro disease and favorable primary tumor characteristics, the guideline states that it is not clear whether the prognosis of TNM stage pN1micro justifies adjuvant systemic treatment [13, 15, 16]. The uncertain benefit of, and indication for, adjuvant systemic treatment was discussed with all patients by the medical oncologist and the choice to give hormonal therapy and/or chemotherapy was made by the physician and the patient.
follow-up
Follow-up started at the date of first operation. Patients were routinely seen twice yearly during outpatient visits, and a mammography was made annually. Dates of subsequent locoregional recurrence, contralateral breast cancer, osseal or visceral metastasis and death were recorded prospectively until August 2007. General practitioners were actively contacted for additional information when patients had not visited the hospital for 12 months.
analysis
Several patients had co-variables with missing values: tumor size (n = 4), modified BR grade (n = 5), mitotic activity index (n = 14), PR status (n = 1), human epidermal growth factor receptor 2 (HER2)/neu status (n = 120), adjuvant radiotherapy (n = 3), adjuvant hormonal therapy (n = 22) and adjuvant chemotherapy (n = 22). As it has been shown that the analysis of data after omitting patients with a missing value reduces statistical power and often lead to a biased result, we used an imputation method to replace missing values [17, 18]. This was done for all variables with missing data except for HER2/neu status, which was deemed to be missing in too many patients (17%, due to the relatively recent introduction of HER2/neu assessment in clinical practice). In total, 40 patients (6%) with missing data on one or more co-variables had these values imputed by an expectation–maximization method [Missing Value Analysis command, SPSS 14.0 (SPSS, Chicago, IL)]. For this, we used all co-variables including information on outcome and lymph node status. All analyses that we report are on the basis of this dataset with imputed values, but we also analyzed the data using a complete-case approach to evaluate the robustness of our findings.
On the basis of the metastatic involvement of regional lymph nodes, patients were categorized into four groups for analysis: pN0 (including pNitc), pN1micro, pN1a and pN1b. Then, pN0, pN1micro, pN1a and pN1b groups were compared for differences in age, tumor characteristics and adjuvant therapy administration by plotting their mean (with standard deviation) or median (with range) for continuous data when appropriate and percentages for categorical data within each group. Differences were tested for statistical significance by one-way analysis of variance, Kruskal–Wallis test or chi-square test when applicable.
We assessed the association between lymph node status, with special interest in the pN1micro group, and patient outcome in several ways. End points were defined as overall survival (OS), disease-free survival (DFS) and its individual components (i.e. locoregional recurrence, contralateral breast cancer and osseal or visceral metastasis).
First, incidence rates of these outcomes were calculated for the total study population. Then, crude and age-adjusted incidence rates for each group were calculated (for the latter using the indirect standardization method in which pN0 group served as the standard).
Cox proportional hazards analyses were used to assess the risk of adverse patient outcome for pN1micro, pN1a and pN1b compared with the pN0 group. For OS analyses, follow-up ended at the date of death (event) or either at the date of lost to follow-up or at the end of follow-up (by censoring), whichever occurred first. For DFS analyses, follow-up ended as event at the date of death, locoregional recurrence, contralateral breast cancer or metastasis, whichever occurred first (women remaining free of disease until lost to follow-up or until the end of follow-up were censored at that date). For analyses regarding risk of metastasis per se, osseal or visceral metastasis, follow-up ended as event at the first occurrence of the specific outcome (women remaining free of the specific outcome until death, lost to follow-up or end of study were censored at that date). No Cox proportional hazards analyses were carried out for the risk of locoregional recurrence and contralateral breast cancer as the number of these events was deemed too small.
Several models were made for each outcome of interest: a model without adjustment for co-variables (crude model); a model with adjustment for age (continuous), tumor size (continuous) and BR grade (adjusted model 1) and a model with additional adjustment for adjuvant treatment (adjusted model 2). We explored whether continuous variables (age and tumor size) were nonlinearly related with the different outcomes by adding quadratic terms to the adjusted model 2. This did not improve the fit of the models, so transformation of these variables was not deemed necessary. Categorical variables were entered into the models by making use of dummy variables.
Selection of co-variables to include in the multivariate models was on the basis of a statistical significant univariate association with lymph node status.
Survival and hazard plots visualizing the relation between lymph node status and the different outcomes were derived from the Cox proportional hazards analyses (adjusted model 2, plotted in strata of lymph node status at the mean of the co-variables).
The proportionality of the hazard assumption was checked by log minus log plots and was not violated for any of the variables in the different models. We also assessed the magnitude of correlations between all estimates within the different models and found that there was no threat of multicollinearity.
All analyses were carried out with SPSS 14.0 (SPSS), and all tests were two-sided with a cut-off for statistical significance of 5%.
|
results |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionReferences |
|---|
At the end of follow-up (1 August 2007), 636 women of the cohort of 703 women were still alive (90.5%), 53 had died (7.5%) and 14 were lost to follow-up (2.0%). A total of 27 760 months of follow-up were accrued, with a median time of follow-up of 40 months. The median age of the patients was 59.4 (range 24.2–92.0) years. There were 423 patients in the pN0 group (60.2%), among them there were 28 patients with pNitc. There were 81 in the pN1micro group (11.5%), 130 in the pN1a group (18.5%) and 69 in the pN
1b group (9.8%). This latter group contained four (5.8%) patients classified as pN1b, 14 (20.3%) as pN1c, 27 (39.1%) as pN2a and 24 (34.8%) as pN3a/b. Various baseline characteristics were not evenly distributed between the groups (Table 1). Higher nodal status was associated with younger age, larger primary tumor size and higher BR grade. The proportions of ER-positive, PR-positive and HER2/neu-positive tumors were not different for all N-categories.
|
adjuvant postoperative treatment
Adjuvant hormonal therapy and chemotherapy were given more frequently with increasing nodal status (Table 2). In the pN1micro group, 26% underwent chemotherapy while 65% received hormonal therapy. Compared with the situation where the pN1micro group would have been treated strictly as patients with pN0 disease (on the basis of primary tumor and patient characteristics), the proportion of pN1micro patients actually receiving chemotherapy was as high as expected, while hormonal therapy was given twice as often. Similarly, compared with a strict treatment regimen as applied to pN1a disease, pN1micro patients were less often treated with adjuvant hormonal therapy and chemotherapy (Table 3).
|
|
patient outcome
During the observation period, 53 patients died, and the overall annual death rate was 2.3% (Table 4). Thirty patients died with and 23 without breast cancer recurrence. Breast cancer recurred in 64 patients: distant metastasis (n = 44), locoregional relapse (n = 14) and contralateral breast cancer (n = 14). The annual rates were 0.6% for locoregional recurrence, 0.6% for contralateral cancer and 1.9% for metastases (visceral: 1.3%; bone: 1.5%).
|
After adjustment for differences in age, tumor size and BR grade, OS in the pN1micro group was not significantly different when compared with women with pN0 disease [hazard ratio (HR) 0.43; 95% confidence interval (CI) 0.10–1.84]. In comparison to patients with pN0 disease, OS was significantly and gradually worse for pN1a and pN
1b patients (HR 2.44, 95% CI 1.22–4.88, and HR 5.90, 95% CI 2.81–12.43, respectively; Table 4 and Figure 1a). The differences in OS between pN1a and pN1b patients compared with pN0 patients increased when the multivariate analysis was additionally adjusted for adjuvant therapy, while results for pN1micro patients did not change (Table 4).
|
DFS was not significantly different in the pN1micro or pN1a group compared with pN0 patients (age-, tumor size- and BR grade-adjusted HR 0.99; 95% CI 0.48–2.05, and HR 1.39; 95% CI 0.78–2.46, respectively), but was significantly worse for pN
1b patients (age-, tumor size- and BR grade-adjusted HR 3.30, 95% CI 1.82–5.99; Table 4 and Figure 1b). Additional adjustment for adjuvant therapy did not change the results for the pN1micro group, but again strengthened the associations for both the pN1a and the pN1b groups leading to a significant increased risk in the pN1a group (Table 4). There was a proportional increased risk of distant metastases in relation to metastatic lymph node involvement (Table 5). Women with pN1micro disease had an almost three times higher risk of any distant metastases than pN0 patients (age-, tumor size- and BR grade-adjusted HR 2.93; 95% CI 1.10–7.77). The risk of visceral metastasis was not significantly increased, but the risk of bone metastasis was borderline significantly increased when adjusted for age, tumor size and BR grade (HR 2.88; 95% CI 0.90–9.19) and reached statistical significance after additional adjustment for adjuvant treatment (Table 5).
|
Women with pN1a disease had about a five times higher risk of both visceral and osseal metastases compared with the pN0 group, whereas women with pN
1b disease had a 14 times increased risk of visceral and an eight times increased risk of osseal metastases, all when adjusted for age, tumor size and BR grade. Additional adjustment for adjuvant therapy strengthened these associations, retaining similarly increased risks of visceral and osseal metastases in pN1a patients, and a predominantly higher increased risk of visceral metastases compared with osseal metastases in the pN1b group (Table 5 and Figure 2a and b).
|
Analyzing the data without imputation of missing values yielded largely similar results. Such a complete-case analysis tended to result in somewhat stronger relations (on average 4% inflation of effect estimates; standard deviation 8%). Conclusions on the basis of statistical significance were similar between the two approaches (data not shown).
|
discussion |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionReferences |
|---|
In a prospective cohort of patients who underwent surgery for clinically T1–2N0 breast cancer and staging by SLNB, micrometastatic lymph node involvement was observed in 11.5% of the patients. After a limited follow-up and following adjustment for age, tumor size, modified BR malignancy grade and adjuvant systemic therapy, we found OS to be comparable to patients with pN0 disease and DFS closely resembling pN0 disease too. The risk of distant metastases was, however, gradually higher for every subsequent N-category.
This study has several strengths. First, SLNs were uniformly examined by serial sectioning at 250-µm intervals. As a confirmation of the thoroughness of the standardized pathology examination, the SLNB procedure resulted in the observed high frequency of patients with TNM stage pN1micro in this study group. This is in line with other ‘SLNB-series’ reporting frequencies of micrometastases up to 23% [1,6–8]. Another quality-enhancing consequence of the SLNB procedure was the exploration for IM SLNs. This led to the exclusion of patients with IM metastases (pN1b and pN1c) that would otherwise have been classified as pN1a. In addition, the present series is a substantial cohort of patients with almost no lost to follow-up.
The main weakness of the present study was the limited follow-up period resulting in a small number of events. Consequently, firm conclusions are difficult to make. However, as the SLNB procedure is only used since the late 1990s, more mature data, although urgently needed [19], are not at hand. Furthermore, adjuvant systemic treatment in the pN1micro group was a potential source of bias. Hormonal therapy was prescribed twice as often as one would have advised on the basis of the primary tumor characteristics and the age of the patient, reflecting a tendency to give hormonal therapy in patients with pN1micro. Chemotherapy was only given to the proportion of patients in whom age and primary tumor characteristics justified its use. Recognizing the possibility of the treatment bias we therefore presented the multivariate analyses both without and with adjustment for adjuvant treatment. Lastly, although SLNB has largely improved staging of breast cancer patients, there remains a small but realistic chance of ‘missing’ micrometastases when examining SLNs following serial sectioning lymph nodes at 250-µm intervals as proposed by our national pathology guidelines [20] and what appears to be common clinical practice [21].
The prognostic significance of lymphogenic micrometastatic disease is still under debate. ‘Older’ studies report equivocal results. Some observed a prognosis for pN1micro comparable to the prognosis of pN0 patients [9, 22, 23], while others found the prognosis of pN1micro to be similar to pN1a patients [10,24–26]. A number of recently conducted large population-based studies observed OS for pN1micro patients to be between pN0 and pN1a, with a modestly increased risk of death for pN1micro versus pN0 [27–30].
The aforementioned studies invariably report on patients treated in the ‘pre-SLN era’. Before the introduction of the SLNB as a staging procedure in breast cancer surgery, the routine pathological examination of lymph nodes from axillary dissection specimens was done by halving the lymph nodes and examining one cut from both halves by H&E only. Nowadays, lymph nodes are investigated by serial sectioning a node at very small intervals. In the present study, 10 slices per node were examined. In addition, not only the number of examined cuts of lymph nodes has changed but also the routine use of immunohistochemical staining with antibodies against keratin has facilitated the detection of very small clusters of metastatic tumor cells in lymph nodes.
While the population-based studies from the pre-SLN era reported a 2%–5% frequency of pN1micro, we observed 11.5% of patients with TNM stage pN1micro and others found similar or even higher rates of micrometastases since the introduction of the SLN procedure. In part, this higher frequency is the result of patient selection since overt axillary metastases preclude the use of the SLNB, and these patients are not included in ‘SLN cohorts’. On the other hand, and undoubtedly more importantly, better examination of lymph nodes has contributed to the observed higher frequency of micrometastases. Not only has the frequency of lymphogenic micrometastases increased, at the same time even smaller, so-called sub-micrometastases are detected in a substantial proportion of the patients [3, 7, 31], and within the pN1micro group the smaller metastases appear to prevail [31].
Apart from the stage-migrating effect of the SLN procedure, adjuvant systemic treatment has changed substantially in the same period. During the last 10 years, determination of endocrine responsiveness of primary breast cancers has improved largely, resulting in an increasing proportion of ER-positive patients and a similar rise of patients who will receive endocrine therapy. Furthermore, and also taking place in the last decade, more patients receive adjuvant systemic therapy because of unfavorable tumor characteristics dictated by more strict St Gallen criteria [32]. Lastly, again in the same time frame, more potent (anthracycline-containing) drugs are being used [15, 33].
The current group of pN1micro patients that consists of patients with lesser metastatic burden being more aggressively treated is different from the group of patients who were diagnosed as having pN1micro before the introduction of the SLN procedure. Consequently, we argue that the prognostic value of pN1micro cannot be distilled from studies on the basis of studies of patients who were staged by axillary lymph node dissections.
The most robust data from the SLN era come from a study by Colleoni et al. from the European Cancer Institute in Milan. After 4 years of follow-up, they observed disease-free, as well as metastasis-free survival for the pN1micro group to be worse than pN0, with a hazard rate of 1.5 for DFS. However, OS was exactly the same after 4 years of follow-up and significantly better than OS of patients with pN1a disease [31]. It is noteworthy that the cohort of patients was not exclusively staged by the SLN procedure. Almost half of the 1959 patients in this study underwent axillary dissection as a staging procedure. Furthermore, pN1micro was considered an indication for adjuvant systemic therapy in their study.
In agreement with Colleoni, we found OS for pN1micro patients to be comparable to pN0 patients, and distant metastases were significantly more common in the pN1micro group. In contrast with the Milan group, patients with pN1micro were not routinely prescribed chemotherapy in the present study, and only 25% received adjuvant chemotherapy.
One explanation for the paradoxical observation that a difference in DFS did not translate into a difference in OS may be that the follow-up period is too short. Another explanation is our finding that there was a difference in bone and visceral metastases between the pN1micro group and the pN1a group. Visceral metastases, associated with a gloomy short-term prognosis, were not more common in the pN1micro group than in the pN0 group, while osseal metastases showed a proportional increase in relation to all N-categories. Lastly, competing other causes of death may also be an explanation for the fact that a difference in metastasis-free survival does not translate into a difference in OS.
Furthermore, although we observed a significantly higher risk of distant metastases in the pN1micro group and a not significantly worse DFS, the hazard rates for these end points were at least twice as high for the pN1a group. This implies that for all end points, the outcome in patients with pN1micro disease is at least better than that for patients with pN1a disease.
The relevance of the prognostic meaning of pN1micro is the answer to the clinical problem as to whether pN1micro should be an indication for adjuvant systemic treatment. Although follow-up is very limited, and despite the fact that clinicians tend to give hormonal therapy on the basis of the finding of micrometastases, the present data do not support the use of adjuvant systemic treatment in patients only because they have lymphogenic micrometastases in an SLN. On the basis of the observations that there was no statistical difference in OS between the pN0 and pN1micro group and that DFS was not significantly worse for pN1micro patients, we conclude that pN1micro in itself should at least not be an indication for adjuvant chemotherapy.
Received for publication March 11, 2008. Revision received June 25, 2008. Accepted for publication July 1, 2008.
|
References |
|---|
|
TopAbstractintroductionpatients and methodsresultsdiscussionReferences |
|---|
1. Cserni G. Metastases in axillary sentinel lymph nodes in breast cancer as detected by intensive histopathological work up. J Clin Pathol (1999) 52:922–924.[Abstract]
2. Cserni G. Axillary staging of breast cancer and the sentinel node. J Clin Pathol (2000) 53:733–741.
3. Cserni G. Complete sectioning of axillary sentinel nodes in patients with breast cancer. Analysis of two different step sectioning and immunohistochemistry protocols in 246 patients. J Clin Pathol (2002) 55:926–931.
4. Cserni G, Amendoeira I, Apostolikas N, et al. Pathological work-up of sentinel lymph nodes in breast cancer. Review of current data to be considered for the formulation of guidelines. Eur J Cancer (2003) 39:1654–1667.[CrossRef][Web of Science][Medline]
5. Tjan-Heijnen VC, Bult P, de Widt-Evert LM, et al. Micro-metastases in axillary lymph nodes: an increasing classification and treatment dilemma in breast cancer due to the introduction of the sentinel lymph node procedure. Breast Cancer Res Treat (2001) 70:81–88.[CrossRef][Web of Science][Medline]
6. van der Heiden-van der Loo M, Bezemer PD, Hennipman A, et al. Introduction of sentinel node biopsy and stage migration of breast cancer. Eur J Surg Oncol (2006) 32:710–714.[CrossRef][Medline]
7. van Rijk MC, Peterse JL, Nieweg OE, et al. Additional axillary metastases and stage migration in breast cancer patients with micrometastases or submicrometastases in sentinel lymph nodes. Cancer (2006) 107:467–471.[CrossRef][Web of Science][Medline]
8. Giuliano AE, Dale PS, Turner RR, et al. Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg (1995) 222:394–399.[Web of Science][Medline]
9. Huvos AG, Hutter RV, Berg JW. Significance of axillary macrometastases and micrometastases in mammary cancer. Ann Surg (1971) 173:44–46.[Web of Science][Medline]
10. Fisher ER, Swamidoss S, Lee CH, et al. Detection and significance of occult axillary node metastases in patients with invasive breast cancer. Cancer (1978) 42:2025–2031.[CrossRef][Web of Science][Medline]
11. Madsen EVE, Gobardhan PD, Bongers V, et al. The impact on post-surgical treatment of sentinel lymph node biopsy of internal mammary lymph nodes in patients with breast cancer. Ann Surg Oncol (2007) 14:1486–1492.[CrossRef][Web of Science][Medline]
12. Wittekind C, Greene FL, Hutter RVP, et al. TNM Atlas, Illustrated Guide to the TNM/pTNM Classification of Malignant Tumours. pp. 207–223. Springer, New York: 2005.
13. Rutgers EJ, Nortier JW, Tuut MK, et al. Dutch Institute for Healthcare Improvement guideline, "Treatment of breast cancer". Ned Tijdschr Geneeskd (2002) 146:2144–2151.[Medline]
14. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology (1991) 19:403–410.[Web of Science][Medline]
15. Kwaliteitsinstituut voor de Gezondheidszorg CBO,VvIK. Richtlijn Behandeling van het mammacarcinoom. Van Zuiden Communications. Alphen aan den Rijn 2005.
16. Bontenbal M, Nortier JW, Beex LV, et al. [Adjuvant systemic therapy for patients with resectable breast cancer: guideline from the Dutch National Breast Cancer Platform and the Dutch Society for Medical Oncology]. Ned Tijdschr Geneeskd (2000) 144:984–989.[Medline]
17. Greenland S, Finkle WD. A critical look at methods for handling missing covariates in epidemiologic regression analyses. Am J Epidemiol (1995) 142:1255–1264.
18. Harrell FE Jr, Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med (1996) 15:361–387.[CrossRef][Web of Science][Medline]
19. Dowlatshahi K, Fan M, Snider HC, Habib FA. Lymph node micrometastases from breast carcinoma: reviewing the dilemma. Cancer (1997) 80:1188–1197.[CrossRef][Web of Science][Medline]
20. Madsen EVE, van Dalen J, van Gorp J, et al. Strategies for optimizing pathologic staging of sentinel lymph nodes in breast cancer patients. Virchows Arch (2008) 20. PMID: 18563440.
21. Cserni G, Amendoeira I, Apostolikas N, et al. Discrepancies in current practice of pathological evaluation of sentinel lymph nodes in breast cancer. Results of a questionnaire based survey by the European Working Group for Breast Screening Pathology. J Clin Pathol (2004) 57:695–701.
22. Pickren JW. Significance of occult metastases. A study of breast cancer. Cancer (1961) 14:1266–1271.[CrossRef][Web of Science][Medline]
23. Attiyeh FF, Jensen M, Huvos AG, Fracchia A. Axillary micrometastasis and macrometastasis in carcinoma of the breast. Surg Gynecol Obstet (1977) 144:839–842.[Web of Science][Medline]
24. Wilkinson EJ, Hause LL, Hoffman RG, et al. Occult axillary lymph node metastases in invasive breast carcinoma: characteristics of the primary tumor and significance of the metastases. Pathol Annu (1982) 17:67–91.[Web of Science][Medline]
25. Rosen PP, Saigo PE, Braun DW, et al. Axillary micro- and macrometastases in breast cancer: prognostic significance of tumor size. Ann Surg (1981) 194:585–591.[Web of Science][Medline]
26. Byrne J, Horgan PG, England S, et al. A preliminary report on the usefulness of monoclonal antibodies to CA 15-3 and MCA in the detection of micrometastases in axillary lymph nodes draining primary breast carcinoma. Eur J Cancer (1992) 28:658–660.[CrossRef][Medline]
27. Maibenco DC, Dombi GW, Kau TY, Severson RK. Significance of micrometastases on the survival of women with T1 breast cancer. Cancer (2006) 107:1234–1239.[Medline]
28. Kuijt GP, Voogd AC, van de Poll-Franse LV, et al. The prognostic significance of axillary lymph-node micrometastases in breast cancer patients. Eur J Surg Oncol (2005) 31:500–505.[CrossRef][Medline]
29. Grabau D, Jensen MB, Rank F, Blichert-Toft M. Axillary lymph node micrometastases in invasive breast cancer: national figures on incidence and overall survival. APMIS (2007) 115:828–837.[CrossRef][Web of Science][Medline]
30. Chen SL, Hoehne FM, Giuliano AE. The prognostic significance of micrometastases in breast cancer: a SEER population-based analysis. Ann Surg Oncol (2007) 14:3378–3384.[CrossRef][Web of Science][Medline]
31. Colleoni M, Rotmensz N, Peruzzotti G, et al. Size of breast cancer metastases in axillary lymph nodes: clinical relevance of minimal lymph node involvement. J Clin Oncol (2005) 23:1379–1389.
32. 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.
33. EBCTCG. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet (2005) 365:1687–1717.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. B. Julian If Sentinel Node Micrometastases Are the Question, Are Clinical Trials the Answer? J. Clin. Oncol., October 1, 2009; 27(28): 4640 - 4641. [Full Text] [PDF]
|
|
|
|
 |
|
 M. de Boer, C. H.M. van Deurzen, J. A.A.M. van Dijck, G. F. Borm, P. J. van Diest, E. M.M. Adang, J. W.R. Nortier, E. J.T. Rutgers, C. Seynaeve, M. B.E. Menke-Pluymers, et al. Micrometastases or Isolated Tumor Cells and the Outcome of Breast Cancer N. Engl. J. Med., August 13, 2009; 361(7): 653 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Burstein, I. Souter, H. A. D'Alessandro, and D. C. Sgroi Case 25-2009 -- A 36-Year-Old Woman with Hormone-Receptor-Positive Breast Cancer N. Engl. J. Med., August 13, 2009; 361(7): 699 - 707. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||