10.1245/s10434-006-9109-6
Annals of Surgical Oncology 13:1450-1456 (2006)
© 2006 Society of Surgical Oncology
Upstaging and Improved Survival of Early Breast Cancer Patients after Implementation of Sentinel Node Biopsy for Axillary Staging
Kimberly A. Vanderveen, MD,
Philip D. Schneider, MD, PhD,
Vijay P. Khatri, MD,
James E. Goodnight, MD, PhD and
Richard J. Bold, MD
Department of Surgery, Division of Surgical Oncology, University of California Davis Medical Center, 4501 X Street, Sacramento, California 95817, USA
Correspondence: Address correspondence and reprint requests to: Richard J. Bold, MD; E-mail: richard.bold{at}ucdmc.ucdavis.edu
 |
ABSTRACT
|
|---|
Background: Sentinel lymph node biopsy (SLNB) has become a standard for axillary staging for early breast cancer patients. Prior studies suggest that SLNB may be more sensitive for the identification of lymph node disease than axillary lymph node dissection (ALND). We hypothesized that SLNB use increases the incidence of node-positivity in early breast cancer patients compared to ALND. Furthermore, survival improves due to more accurate staging (stage migration).
Methods: Registry data from an NCI-designated cancer center was reviewed for breast cancer patients with T1 and T2 tumors for two 5-year periods: before (1993–1997) and after (2000–2004) SLNB implementation (1998). TNM staging was updated to conform to American Joint Committee on Cancer (AJCC) 2003 guidelines.
Results: There were no differences in tumor size or stage groupings between the two time periods (n = 316 and 577). There was a non-significant increase in the proportion of patients with lymph node involvement (32 vs. 27%; P = .16) after SLNB implementation; though a trend of increased incidence of single-node positive patients was observed (13 vs. 8%; P = .07). This was significant in patients with T1A/T1B tumors (10 vs. 3%; P = .04), though not seen in T1C or T2 tumors. Stage II survival improved in the later time period (P = .02).
Conclusions: The increase in single-node positivity after SLNB implementation supports the theory that SLNB is more sensitive than ALND. Improvements in survival are likely due to the stage migration of patients who would have been node-negative by ALND (but were found to be node-positive by SLNB) in addition to improvements in adjuvant therapy.
Key Words: Sentinel lymph node biopsy • Breast Cancer • Stage migration
|
INTRODUCTION
|
|---|
As new technologies develop, accuracy in staging cancer improves and the extent of disease can be more accurately defined. Improved detection by new technologies can result in up-staging a subset of patients that were previously under-staged, with a resulting improvement in survival for each TNM stage grouping. Although the prognosis for the stage groupings appears to improve, the individual subjects’ outcomes do not change. This effect has been dubbed the "Will Rogers Phenomenon," and was described by Feinstein et al. in reference to a stage migration and apparent improvement in stage-specific survival of lung cancer patients after the implementation of more sensitive diagnostic testing [computer-assisted tomography (CAT) scan].1 Based on the statistics reported by the National Cancer Institute (SEER database), there appears to be an increased number of breast cancer patients diagnosed with regional (node-positive) disease between 1993 and 2000.2 However, the precise staging of these patients as well as reasons for the observed change in stage distribution are unclear.
Recently, the staging criteria for breast cancer were updated,3 with an anticipated stage migration and shift in survival curves.4 In the last several years, there has also been a change in how lymph node involvement is detected. Prior to the late 1990s, axillary lymph node dissection (ALND – the removal of nodal tissue from the axilla based on anatomic boundaries) was the standard surgical staging procedure for evaluation of axillary spread.5 In the mid-1990s, sentinel lymph node biopsy (SLNB) was first described for use in breast cancer staging.6,7 The success of the initial reports of SLNB prompted several large national trials (NSABP B-32, ACOSOG Z0010 and Z0011),8,9 and by 2000, SLNB emerged as a standard axillary staging procedure.10–13 SLNB is an attractive alternative to ALND as the node receiving initial drainage from the tumor (the "sentinel node") can be identified and removed with much less morbidity than ALND.14–16 Additionally, SLNB is theorized to be more sensitive for detecting early lymph node metastasis, as the limited nodes that are removed are more carefully scrutinized by the pathologist and amenable to advanced pathologic techniques, such as immunohistochemical staining.17,18 Furthermore, the use of immunohistochemistry increases the detection of lymph node disease by more than two-fold and provides additional prognostic information.19
If SLNB is indeed better in the detection of early lymph node disease than ALND, it should result in a stage migration separate from the anticipated effects of the new staging system. No prior trials have performed a "head-to-head" comparison of SLNB and ALND for staging. In all of the comparison trials, patients who were randomized to ALND also had a SLNB performed as the initial part of their axillary dissection. Therefore, all patients received the higher-scrutiny pathological evaluation (serial sectioning of lymph nodes, etc.) for the sentinel lymph node that is not possible for routine ALND specimens.8,9 This confounds the comparisons between SLNB and ALND for accuracy of staging and impact on survival. We therefore sought a novel approach to comparing these techniques by looking at the incidence patterns of nodal disease and survival trends before and after implementation of SLNB for routine staging. This therefore compares SLNB with serial sectioning with standard ALND (without serial sectioning of LNs) as they are used in standard practice. Although this approach may be confounded by other time-dependent variables (such as changes in trends in screening/detection and treatment), given the current practice patterns and standards of care, it would be impossible to perform a trial which randomized patients to SLNB vs. standard ALND (without SLNB/serial sectioning). Therefore, the impact of SLNB on breast cancer outcomes can only be determined by comparisons with historical series.
We hypothesized that SLNB is more sensitive for identifying early lymph node disease, and we proposed to investigate the issue by comparing the stage of disease and incidence of node-positivity between two distinct time periods – before (1993–1997) and after (2000–2004) our group implemented SLNB for the evaluation of lymph node involvement in early breast cancer (which occurred in 1998). We were especially interested in the specific details of those patients who were found to have only a single lymph node involved with metastatic disease.
|
METHODS
|
|---|
Subjects
This study was approved by our institutional review board. Since 1985, all healthcare facilities in California have been required by state law to maintain cancer registries to collect and report cancer statistics to the statewide cancer registry. Regional statistics are also reported to the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) Program. De-identified data from our institutional registry (an NCI-designated cancer center) was reviewed for breast cancer patients with T1 and T2 tumors (candidates for SLNB). Two 5-year periods were analyzed, before (1993–1997) and after (2000–2004) our group implemented SLNB for the evaluation of lymph node involvement in early breast cancer (which occurred in 1998). Patients were excluded from the analysis for the following reasons: Stage IV disease at presentation, absence of pathologic staging information for T and N stage, male gender, T3 or T4 tumors, or when no aspect of the breast cancer treatment was provided at the UC Davis Medical Center. The information collected included: age at diagnosis, TNM stage at diagnosis, tumor size, number of positive lymph nodes, pathology report summaries, follow-up times, and vital status at last follow-up. Staging information was reviewed for all subjects, and TNM status was revised to conform to American Joint Committee on Cancer (AJCC) 2003 staging guidelines (to eliminate the stage migration bias between the old and new guidelines). Patients with N0i+ disease were considered node-negative. We included all eligible patients in our registry, even if their surgeries were done elsewhere, as long as staging information was available.
Data Analysis
The TNM staging and rates of node positivity were compared using chi-square analysis. Survival time was calculated as the time (in months) from diagnosis until death from any cause or last follow-up (as of July 12, 2005). Kaplan-Meier curves were generated and compared with the log-rank test. A P value <.05 was considered to be statistically significant. All statistical analyses were performed with STATA 9.0 (StataCorp LP, College Station, Tex.).
|
RESULTS
|
|---|
A total of 893 patients were identified that met inclusion criteria, 316 in the pre-SLNB (1993–1997) and 577 in the post-SLNB (2000–2004) time period. The average ages of the subjects were 56 and 57 years, respectively, and there were no significant differences in tumor size (by T stage) or stage grouping between the two time periods (Table 1
). In the earlier cohort, two patients underwent SLNB (at an outside facility); in the later cohort, 442 patients underwent SLNB. Of those who underwent SLNB in the later cohort, 107 underwent ALND as well; 39 patients underwent immediate ALND (eight for non-identification of the sentinel lymph node) and 68 patients underwent delayed ALND for a positive sentinel lymph node on final histopathologic evaluation. Thirty-four patients underwent an unidentified axillary staging procedure (no record of procedure type available and a sentinel lymph node not clearly identified in the pathology report).
Comparing the two groups with respect to N stage, there was a 5% increased incidence of lymph node positivity (from 27 to 32% incidence) in the later time period (Table 1
; P = .10). When the specifics of the lymph node positivity was examined, it was observed that only 8% of patients in the first time period had just a single positive lymph node, but this increased to 13% in the later time period, accounting for the entire shift in N stage (Table 1; P = .07). We then examined whether the increased incidence of metastatic disease in the lymph nodes, specifically the single-positive lymph node subgroup, was associated with T stage. There was a significant increase in patients with T1A or T1B tumors who were found to have a single-positive lymph node in the later time period compared to the earlier cohort (10 vs. 3%; Table 2; P = .04). An increase in node positivity or single-positive lymph node was not observed in T1C or T2 tumors.
Patients in the earlier time period had higher grade tumors compared with the later period (Table 1, P = .02). For the subset analysis by tumor size, this was only significant for patients with T2 tumors where 59% of patients had high grade tumors in the later period, compared with 40% in the earlier time period (Table 3; P = .03). There was no significant difference in tumor grade between the two time periods for patients with T1A/B or T1C tumors (Table 3; P = .25 and .16, respectively). On analysis by stage grouping, Stage II patients had more high grade tumors in the earlier time period than in the later time period (37 vs. 26%; P = .002). There was no difference in tumor grade in Stage I or Stage III disease (data not shown; P = .28 and .31, respectively).
View this table:
[in this window]
[in a new window]
|
TABLE 3. Tumor grade by tumor size (T stage) during the two different time periods. Patients with unknown stage excluded from analysis
|
|
We then examined survival outcomes in the two time cohorts to determine whether there was any impact of this nodal upstaging. Mean and median follow-ups in the earlier time period were 84 and 90 months, respectively; mean and median follow-ups in the later time period were 27 and 26 months, respectively. The stage-specific survival for both time periods is shown in Fig. 1. As the overall stage distribution of the two groups was similar, we first compared overall survival for all patients between the two time periods and noted an improved outcome in the later period, despite limited follow-up (Fig. 2; P = .02). Stage-specific survival was also compared between the two time cohorts; Stage I survival (Fig. 3A; P = .21) and Stage III survival (Fig. 3C; P = .22) did not significantly improve. However, overall survival for patients with Stage II survival did significantly improve (Fig. 3B; P = .03).
View larger version (13K):
[in this window]
[in a new window]
|
FIG. 1. Kaplan-Meier curves for overall survival of patients during the earlier time cohort (A) and the later time cohort (B) by stage grouping. Note that the time of follow-up is shorter for the later time cohort.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
FIG. 2. Kaplan-Meier curves for overall survival of all patients with early breast cancer, before and after implementation of sentinel lymph node biopsy for routine axillary staging. (P = .02 by log rank test).
|
|
View larger version (10K):
[in this window]
[in a new window]
|
FIG. 3. Kaplan-Meier curves for overall survival of patients with early breast cancer, before and after implementation of sentinel lymph node biopsy for routine axillary staging. 3A, Stage I (log rank, P = .21); 3B–Stage II (log rank, P = .03); 3C, Stage III (log rank, P = .82).
|
|
|
DISCUSSION
|
|---|
Following the implementation of SLNB for early stage breast cancer, we noted an increase in single-node positive patients, with a 5% shift of patients from N0 to N1 disease. Although we hypothesized that all tumor sizes would be affected (and did see an increased incidence in single-node positivity for all T1 and T2 patients), this increase was significant only for T1A and T1B tumors, though not for T1C or T2 tumors. This suggests that SLNB is indeed more sensitive than axillary dissection for the detection of early nodal metastasis and that the patients most likely to benefit from the improved detection of nodal metastasis are those with tumors less than 1 cm in size. Although our dataset does not allow evaluation of the exact size of lymph node metastasis, a benefit of SLNB has been the identification of small meta-static foci on pathologic exam of sentinel nodes. Therefore, it seems reasonable that small tumors would have a greater proportion of patients harboring microscopic nodal disease and that these patients would benefit most from SLNB compared with those with larger tumors. It is also possible that our dataset was underpowered to detect increases in the T1C and T2 patients. A significant shift in stage groupings was not observed, despite the increased incidence of single-node-positive and N1 patients.
An improvement in survival after SLNB implementation was also observed, though only in those with Stage II disease. It is unclear why only Stage II survival appeared to be so dramatically altered, while Stage I and III survival did not significantly change; however, this may be a result of the higher tumor grade in the Stage II patients from the earlier time period. Given the small numbers of our dataset, it is also possible that that our study was underpowered to detect stage migration-associated differences in these two groups, particularly for Stage I patients who generally have a very good prognosis. Therefore, the migration of just 5% of the patients who were converted from N0 to N1 (single node positive) may have been sufficient to improve Stage II survival, but not Stage I survival. The improvement in Stage II survival is unlikely to be affected by the upstaging of a small number of patients to Stage III disease, as this requires N2 (> 3 metastatic lymph nodes), which is an infrequent outcome of SLNB.20 Although our data suggest that additional T2 patients were diagnosed with positive nodes after SLNB, this was not statistically significant, and survival for T2 tumors independent of nodal stage did not change (data not shown), despite the increased proportion of lower grade tumors in the later time period.
We revised the staging for all subjects to conform to the AJCC 2003 staging guidelines to control for the anticipated changes in stage groupings between the two time periods. Two studies have investigated the effects of these changes, which have reclassified "N" status primarily based on number of positive nodes.3 Woodard et al. reported improved stage-specific survival curves resulting from the upstaging of high-risk patients from N1 to N2/3 (Stage II to III).4 Conversely, McCready et al. have reported that the new guidelines may downstage approximately 7% of patients by reclassifying small micrometastases (< .2 mm) as N0 rather than N1; this may also decrease the sensitivity of SLNB.21 The shifts we observed primarily affected the detection of single-node disease (shifts from N0 to N1, and Stage I to II). Because we used the 2003 guidelines for both time periods in our study, the increased detection of N1 disease in the later time period strengthens the assertion that SLNB is more sensitive than ALND for detecting early nodal disease. Furthermore, the increased overall detection of early nodal disease with SLNB may offset the downstaging of small micrometastases (< .2 mm).
The outcomes of the single-node positive patients are of particular interest, as the behavior of this subset of disease may have important clinical implications for therapy. The 2003 AJCC guidelines distinguish between N0 and N1 disease by the size of a single node metastasis (relative to .2 mm), and between N1 and N2 disease by the number of nodes positive (less than or greater than 3).3 However, single-node positivity (even at the immunohistochemical level) may affect prognosis and alter adjuvant treatment.19 We therefore sought to compare the survival of: N0 patients, N1 patients who are single-node positive, and N1 patients with two or three nodes positive. Because the "single-node positive" patients are not equivalent between the earlier and later time periods (because of the difference in accuracy between SLNB and ALND), we intended to use data from the later group only. Unfortunately, our dataset contained too few node-positive patients in the later group to perform this analysis. An additional limitation of our dataset is the absence of clinical staging prior to surgical staging. Of the patients in the later time cohort, 20% did not undergo SLNB, presumably due to contraindications to SLNB, such as clinical evidence of metastatic lymph node involvement. While the T stage between the two time cohorts is similar, we cannot address the question of whether a similar fraction of patients in the earlier time cohort would have been appropriate for SLNB.
Improvements in survival between the two time periods may also have been impacted by the altered treatment of a previously under-staged subgroup of patients; however, the extent of this impact is impossible to isolate. Changes in adjuvant therapy between the two time periods include the addition of taxane-based polychemotherapy for early disease patients. This may account for some of the improved survival, particularly in the increased 5% of patients that were node-positive after the implementation of SLNB. Preliminary results of taxane-based polychemotherapy trials have demonstrated improvements in 5-year survival by 4–6% versus non-taxane-based regimens.22 We may be seeing some of these early "apparent" survival benefits in our patients; particularly since the subgroup most benefited in these trials was those with one to three positive nodes.22
In summary, the transition from ALND to SLNB has increased the detection of lymph node metastases, primarily with respect to the detection of single-node disease. This impact is that there has been a 5% shift of patients from N0 to N1 disease. This impact is most significant in smaller (T1A and T1B) tumors. Although this does not represent a pure "Will Roger’s Phenomenon," as overall survival did improve in the later time period, we believe that a stage migration effect may partially account for improvements in Stage II survival in addition to the advances in adjuvant therapy and the more aggressive treatment of a previously under-staged group of patients.
|
ACKNOWLEDGMENTS
|
|---|
We would like to thank Juanita Pratt of the UC Davis Cancer Registry for providing the data for this study.
Received for publication May 10, 2006.
Accepted for publication June 2, 2006.
|
REFERENCES
|
|---|
- Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon. Stage migration and new diagnostic techniques as a source of misleading statistics for survival in cancer. NEJM 1985; 312:1604–8.[Abstract]
- Ghafoor A, Jemal A, Ward E, Cokkinides V, Smith R, Thun M. Trends in breast cancer by race and ethnicity. CA Cancer J Clin 2003; 53:342–55.[Abstract/Free Full Text]
- Singletary SE, Allred C, Ashley P, et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002; 17:3628–36.
- Woodard WA, Strom EA, Tucker SL, et al. Changes in the 2003 American Joint Committee on Cancer staging for breast cancer dramatically affect stage-specific survival. J Clin Oncol 2003; 21:3244.[Abstract/Free Full Text]
- NIH consensus conference Treatment of early-stage breast cancer. JAMA 1991; 65:391–5.
- Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgical resection and radiolocalization of the sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993; 2:335–9.[CrossRef][Medline]
- Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391–8.[Medline]
- Ross MI. Sentinel node dissection in early-stage breast cancer: ongoing prospective randomized trials in the USA. Ann Surg Oncol 2001; 77S–81S.
- Krag DN, Julian TB, Harlow SP, et al. NSABP-32: Phase III, randomized trial comparing axillary resection with sentinel lymph node dissection: a description of the trial. Ann Surg Oncol 2004; 11:208S–10S.[Abstract/Free Full Text]
- Edge SB, Niland JC, Bookman MA, Theriault RL, Ottesen R, Lepisto E, Weeks JC. Emergence of sentinel node biopsy in breast cancer as standard-of-care in academic comprehensive cancer centers. J Natl Cancer Inst 2003; 95:514–21.
- Consensus statement on guidelines for performance of sentinel lymphadenectomy for breast cancer. American Society of Breast Surgeons (October 2003). Available at: http://www.breastsurgeons.org/officialstmts/sentinel.shtml. (accessed December 29, 2005).
- Schwartz GF, Giuliano AE, Veronesi U. Proceedings of the consensus conference on the role of sentinel lymph node biopsy in carcinoma of the breast, April 19–22, 2001, Philadelphia, Pennsylvania. Cancer 2003; 94:2542–51.[CrossRef]
- Singh-Ranger G, Mokbel K. The sentinel node biopsy is a new standard of care for patients with early breast cancer. Int J Fertil Womens Med 2004; 49:225–7.[Medline]
- Schrenk P, Rieger R, Shamiyeh A, Wayand W. Morbidity following sentinel lymph node biopsy versus axillary lymph node dissection for patients with breast carcinoma. Cancer 2002; 88:608–14.
- Peintinger F, Reitsamer R, Stranzl H, Ralph G. Comparison of quality of life and arm complaints after axillary lymph node dissection vs. sentinel lymph node biopsy in breast cancer patients. Br J Cancer 2003; 89:648–52.[CrossRef][Medline]
- Schijven MP, Vingerhoets AJ, Rutten HJ, Nieuwenhuijzen GA, Roumen RM, van Bussel ME, Voogd AC. Comparison of morbidity between axillary lymph node dissection and sentinel node biopsy. Eur J Surg Oncol 2003; 29:341–50.[CrossRef][Medline]
- Cody HS, Borgen PI, Tan LK. Redefining prognosis in node-negative breast cancer: can sentinel lymph node biopsy raise the threshold for systemic adjuvant therapy?. Ann Surg Oncol 2004; 11:227S–30S.[Abstract/Free Full Text]
- Dowlatshahi K, Fan M, Snider HC, Habib FA. Lymph node micrometastases from breast carcinoma: reviewing the dilemma. Cancer 1997; 80:1188–97.[CrossRef][Medline]
- Cote RJ, Peterson HF, Chaiwun B, et al. Role of immunohistochemical detection of lymph-node metastases in management of breast cancer. International Breast Cancer Study Group. Lancet 1999; 354:896–900.[CrossRef][Medline]
- Chu KU, Turner RR, Hansen NM, Brennan MB, Bilchik A, Giuliano AE. Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection?. Ann Surg 1999; 229:536–41.[CrossRef][Medline]
- McCready DR, Yong WS, Ng AK, et al. Influence of the new AJCC breast cancer staging system on sentinel lymph node positivity and false-negative rates. J Natl Cancer Inst 2004; 96:873–5.[Abstract/Free Full Text]
- Montero A, Fossella F, Hortobagyi G, Valero V. Docetaxel for the treatment of solid tumours: a systematic review of clinical data. Lancet Oncol 2005; 6:229–39.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
J. H. Lee, E. L. Rosen, and D. A. Mankoff
The Role of Radiotracer Imaging in the Diagnosis and Management of Patients with Breast Cancer: Part 1--Overview, Detection, and Staging
J. Nucl. Med.,
April 1, 2009;
50(4):
569 - 581.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Brenner, A. Gondos, and V. Arndt
Recent Major Progress in Long-Term Cancer Patient Survival Disclosed by Modeled Period Analysis
J. Clin. Oncol.,
August 1, 2007;
25(22):
3274 - 3280.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
