|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 21 (July 20), 2008: pp. 3530-3535 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.15.5630 Impact of Immediate Versus Delayed Axillary Node Dissection on Surgical Outcomes in Breast Cancer Patients With Positive Sentinel Nodes: Results From American College of Surgeons Oncology Group Trials Z0010 and Z0011
From the Department of Surgery, Duke University Medical Center, Durham, NC; University of Texas Southwestern Medical Center; Dallas Surgical Group, Dallas; The University of Texas M.D. Anderson Cancer Center, Houston, TX; Nashville Breast Center, Nashville, TN; Lakeland Regional Cancer Center, Lakeland; Surgical Associates of West Florida, Clearwater, FL; Advanced Surgical Associates, Flint, MI; and the John Wayne Cancer Institute, Santa Monica, CA Corresponding author: John A. Olson Jr, MD, PhD, Box 3873, Duke University Medical Center, Durham, NC 27710; e-mail: jaomd{at}duke.edu
Purpose Patients with breast cancer metastasis to the sentinel lymph nodes (SLNs) generally undergo completion axillary lymph node dissection (cALND), either concurrently with SLN biopsy or at a second procedure. The impact of the timing of cALND on pathologic results and complications in these patients has not been examined. Patients and Methods We examined outcomes from SLN-positive patients in American College of Surgeons Oncology Group (ACOSOG) trials Z0010 and Z0011. Pathologic data examined included primary tumor characteristics, total number of SLNs recovered, positive SLN(s) and non-SLN(s) identified. Complications assessed included axillary seroma, paresthesia, arm morbidity and range of motion, and lymphedema. Results A total of 1,003 assessable patients with SLN metastasis had immediate (n = 425) or delayed (n = 578) cALND. The median number of SLNs and axillary LNs removed were the same between groups. Patients who had immediate cALND more often had larger tumors, SLN metastasis identified intraoperatively, two or more positive SLNs, and higher pathologic N stage. Axillary paresthesia, seroma, and impaired extremity range of motion were more common in the immediate group during the early postoperative period, but not at later time points. There was no difference in lymphedema at any time point. Conclusion In ACOSOG trials Z0010 and Z0011, LN recovery and long-term complications were similar after either delayed or immediate cALND for patients with metastasis to SLNs. Patients who undergo immediate cALND experience more short-term morbidity. With respect to staging and complications, there is no clear detriment for patients with a positive SLN who undergo a second procedure for cALND.
Sentinel lymph node biopsy (SLNB) is an accepted alternative to axillary lymph node dissection (ALND) for staging the axilla in patients with early-stage breast cancer.1 This technique reliably identifies patients who have axillary lymph node metastasis.2,3 Patients with metastasis in the SLN(s) generally undergo completion ALND (cALND) for accurate staging and for prevention of axillary recurrence.4-6 Morbidity of ALND including lymphedema, arm paresthesia, chronic pain, and immobility occurs in 5% to 50% of patients.7-12 In contrast, SLNB has a 4% to 15% frequency of complications.2,12-14 For patients with positive SLNs, complications after the combination of SLNB and cALND include seroma (14%), paresthesia (39%), and lymphedema (11%). These events occur in up to 75% of patients undergoing cALND versus 25% of those undergoing SLNB alone.15 Increasingly, patients with positive SLNs are offered multiple alternatives to managing the axilla including cALND, radiotherapy, or observation. If outcomes after delayed and immediate cALND are similar, patients might choose a stepwise approach to decision making rather than consenting to both procedures at once. On the other hand, if a delayed cALND were associated with greater risk of complications and/or inferior staging, then this would be unacceptable. The impact of timing of cALND in patients with positive SLNs has not been investigated. Examination of the effects of immediate cALND (one surgery) compared with a delayed cALND (two surgeries) has practical relevance to breast cancer care. We hypothesized that a second surgery to perform cALND may result in greater morbidity because of technical difficulty imparted by scarring at the surgical site. This could also result in recovery of fewer lymph nodes and inferior staging. The aim of this study was to compare pathologic results and short-term complications between patients undergoing immediate versus delayed cALND after a positive SLNB. We examined patients enrolled onto the American College of Surgeons Oncology Group (ACOSOG) Z0010 and Z0011 SLN trials who underwent either immediate or delayed cALND, and found that total LN recovery and surgical complications are similar after delayed and immediate cALND.
Study Design The ACOSOG Z0010 trial was conducted to evaluate the incidence and impact of SLN and bone marrow micrometastases on relapse-free survival (RFS) and overall survival (OS) in patients with early-stage carcinoma of the breast treated with breast-conserving surgery (BCS) and radiation therapy. The ACOSOG Z0011 trial evaluated the impact of cALND on the outcomes of patients with carcinoma of the breast and one or two positive SLNs. Both studies examined complications after these procedures. Details regarding these institutional review board–approved studies are available at www.acosog.org. We analyzed pathologic outcomes and surgical complications for patients in the ACOSOG trials Z0010 and Z0011 who had metastasis to SLNs and underwent cALND in either an immediate or delayed fashion. The objectives of the study were to ascertain whether delayed cALND was associated with inferior axillary staging and increased morbidity compared with immediate cALND. Although both ACOSOG SLN trials documented the complications after these procedures, a specific aim to examine the impact of immediate versus delayed cALND was not originally planned.
Frequency distributions were used to summarize patient characteristics and adverse effects.
Patients and Treatment For patients with SLN metastasis undergoing cALND, timing of cALND was not dictated in either protocol and was at the discretion of patient and surgeon. For patients enrolled on the Z0011 trial either alone or in conjunction with the Z0010 trial, the decision for immediate cALND was based on whether the patient was enrolled on the trial before or after the SLN dissection. Surgical treatment of eligible patients in both trials consisted of BCS with SLN dissection. Patients requiring mastectomy were ineligible. The SLN procedure was performed using isosulfan blue dye, 99mTc-labeled sulfur colloid, or both. Conduct of the SLN procedure was not specified in the protocol. Postoperative whole-breast irradiation (45 to 50 Gy) without a supraclavicular field was prescribed in these trials. One hundred eleven participating surgeons were credentialed according to an ACOSOG skills-verification procedure.16 This process included documentation of outcomes of 20 SLN procedures confirmed with cALND or completion of a training program with adequate SLNB training. Skills verification was not required for ALND.
Outcomes Measured Pathologic data collected included primary tumor size, histologic subtype, Bloom-Richardson score, lymphovascular invasion (LVI), and the presence or absence of estrogen receptor (ER) and progesterone receptor. Lymph node pathologic data included total number of SLNs removed, number of SLNs with metastasis, total number of ALNs removed, and the number of ALNs with metastasis. Collection of complication data was mandatory for all Z0010 and Z0011 trial participants and was submitted prospectively to the ACOSOG statistical center. Complication data recorded in each study included axillary paresthesia, range of motion, lymphedema, wound infection, and axillary seroma. The presence or absence of paresthesia was determined subjectively by patient symptoms and objectively by physical examination. Data were analyzed as a binary variable without severity. Lymphedema was defined as an increase in arm circumference of 2 cm or more compared with baseline compared with the contralateral arm.12 Shoulder range of motion (ROM) was scored as degrees of abduction ranging from 45 to 180 degrees. Impairment was defined as a change in measured ROM of 25 degrees or greater. Surgical-complication data analyzed were collected at 30 days, 6 months, and 1 year. Extended time periods had greater than 25% missing data, making these time points unsuitable for the present analysis.
Of the 5,539 patients enrolled on the Z0010 protocol, 5,239 were assessable and 1,255 (24%) had metastasis to the SLN(s). From these 1,255 SLN-positive patients, 614 had cALND and 406 were registered onto the Z0011 protocol. The remaining 235 underwent cALND without enrollment to Z0011 or did not undergo cALND at all. A total of 891 SLN-positive patients were enrolled in the Z0011 protocol, of whom 821 were assessable and 389 were randomly assigned to and received cALND. A total of 209 patients were enrolled in both studies and had cALND. Data regarding timing of cALND, clinicopathologic variables, and surgical complications were available for a total of 1,003 patients in these trials (Fig 1).
Of the 1,003 patients, 425 patients (42%) had cALND at the same surgery (immediate) and 578 patients (58%) had cALND at a second surgery (delayed). Median time between SLNB and cALND in the delayed group was 19 days (range, 1 to 93 days). Study participation was a major factor in the approach to timing of cALND. Patients in the Z0010 trial had immediate cALND significantly more often than did those in the Z0011 trial (50% v 31%; P < .001; Fig A1, online only). The proportion of patients in each study who had immediate versus delayed cALND was constant throughout the years of study accrual (data not shown). Clinicopathologic characteristics of the patients who had immediate or delayed cALND are shown in Appendix Table A1 (online only). Patients in the two groups were comparable with respect to age, race, and BMI. Primary tumors were larger (median, 2.0 v 1.8 cm; P = .04) and more often ER negative (21% v 17%; P = .05) in the immediate group than in the delayed group. There was no difference in other primary tumor features including T stage, histologic subtype, Bloom-Richardson score, presence of LVI, or progesterone receptor status between groups. A median of 2.0 SLNs (range, 1 to 17 SLNs) were removed at the time of SLN dissection with no difference in the number of total nodes or total positive SLNs removed between immediate and delayed groups (Table 1). However, there was a significant difference in the frequency of finding two or more positive SLNs in the immediate group (35% v 28%; P = .02).
The results of cALND in both groups are listed in Table 2. Patients in both groups had the same number of non-SLNs recovered on cALND (median, 14 non-SLNs; range, 0 to 52 non-SLNs). Patients with positive SLNs who underwent immediate cALND had additional non-SLN metastasis identified 42% of the time (177 of 425), including five or more additional positive nodes in 56 patients (13%). Patients in the delayed cALND group had additional non-SLN metastasis identified 27% of the time (153 of 578), significantly less often than those having immediate cALND (P < .0001). Patients who had immediate cALND had significantly more total positive LNs (SLN and non-SLN) than those patients having delayed cALND (median, 2.0 v 1.0 LNs; P < .0001). This translated to a higher pathologic N stage in the immediate group (P < .0001).
Because of differences in the use of intraoperative detection methods (frozen section and/or touch prep) between the immediate and delayed groups, data were analyzed by method of SLN metastasis detection. Data regarding the method of SLN metastasis detection were available for almost 90% of patients in each group (Table 1). Seventy-six percent of patients (325 of 425) who had immediate cALND versus 21% of patients (119 of 578) who had delayed cALND had intraoperative detection of the SLN metastasis using frozen section, touch prep, or both (P < .0001). Examining only those patients who had the diagnosis of SLN metastasis made by frozen section, there was a significantly greater number of positive nodes recovered in the immediate than the delayed group (median, 2.0 v 1.0; P = .01). There was no difference in the number of recovered or positive SLNs or in the total number of ALNs recovered at the cALND procedure between groups. We performed multivariate analysis to identify factors associated with increasing numbers of positive lymph nodes identified on cALND (Table 3). Factors examined included tumor size, LVI, number of positive SLNs, SLN detection using frozen section or touch prep, as well as timing of cALND. Consistent with published results, tumor size, presence of LVI, and more than one positive SLN were significant predictors of additional non-SLN metastasis.21 In addition, immediate cALND was associated with an increasing number of positive non-SLNs.
Complications were compared according to the timing of cALND (Table 4). At 30 days, patients who underwent immediate cALND had more reported axillary paresthesia (51% v 35%; P < .0001) and impaired ROM (49% v 36%; P < .0001) compared with those who underwent delayed cALND. The increase in paresthesia observed after immediate versus delayed cALND persisted at 6 months of follow-up (50% v 42%; P = .03), but not at 1 year. Impaired range of motion observed at 30 days improved significantly at 6 months and 1 year, with no difference noted between groups. Patients who underwent delayed cALND tended to have more lymphedema at 6 months (13% v 10%; P = not significant) but this difference did not remain at 1 year. Postoperative axillary seromas were more frequent after immediate cALND than after delayed cALND. There was no difference in the resolution of surgical complications between the immediate and delayed groups. There was no operative mortality in these trials.
We performed multivariate analysis to identify factors associated with 30-day complications of paresthesia, impaired ROM, and seromas after cALND, as well as paresthesia, impaired ROM, and lymphedema at 1 year (Table 5; Appendix Table A2, online only). Factors examined included patient age, BMI, number of ALNs removed, and timing of cALND. At 30 days, removal of more than 10 ALNs and immediate cALND were significant predictors of paresthesia and impaired ROM. In addition, age older than 60 years and immediate cALND were predictors of seroma at 30 days. At 1 year, immediate dissection and removal of more than 10 ALNs removed retained significance for paresthesia and impaired ROM, respectively (Table A2).
The standard for patients with metastasis to the SLN(s) is cALND. Ideally, SLN metastases are identified at the time of SLNB, and cALND can proceed at the same surgery.17-19 Occasionally, a delayed approach to cALND is taken not only because of failure of intraoperative detection of SLN metastasis, but also to allow for review of the pathology and to obtain patient input in axillary management decisions. The uncertain survival benefit of cALND and its morbidity have led many patients and physicians to seek alternatives to cALND, including observation or axillary radiation therapy.20 We conducted this analysis of ACOSOG SLN data sets to determine whether a delayed approach to cALND influences final pathologic staging or complications after the cALND procedure. In this large multicenter study, we found equivalent recovery of ALNs in patients who had immediate or delayed cALND. A greater degree of LN involvement was observed in patients who had immediate cALND; however, this group of patients had larger tumors and a greater number of positive SLNs. The most plausible explanation for the observed lower frequency of non-SLN and total LN metastasis in the delayed group is selection bias on the part of surgeons offering Z0011 trial participation to patients judged to be at lower risk of having additional nodal disease before registration. The greater overall representation of larger, ER-negative tumors in the immediate group supports this notion. From the data available, we cannot exclude the possibility that other factors may explain the difference in node positivity observed in the immediate and delayed groups. It is possible that the completeness of the cALND is less in the delayed group, or that there is a difference in pathologic accuracy of identifying metastases in LNs collected during reoperative cALND. Although both could result in a lower number of metastases observed in the delayed group, these explanations are speculative and seem less likely because there were equal numbers of LNs recovered with both the delayed and immediate cALND. This retrospective analysis of the data may only raise such hypotheses. Analysis of other SLN data sets may clarify these findings given that it is unlikely that a randomized study of timing of cALND will be feasible. Several studies have documented morbidity after SLNB and after ALND.11-15 However, none have investigated the impact of timing of cALND on complications of these axillary procedures. We anticipated that a delayed cALND could be associated with greater morbidity because of the complexity of reoperative axillary surgery. Our results were contrary to this expectation. Immediate cALND was associated with a greater frequency of postoperative complications including axillary paresthesia, impaired ROM, and seromas. Lymphedema at 6 months was the only complication that tended to occur more frequently in the delayed group. This difference was not statistically significant and was not apparent at 1 year, reflecting either resolution of the process or an inadequate number of patients with sufficient follow-up. Our findings showing resolution of these short-term complications of ALND with longer follow-up is consistent with other reports.2,13 These findings apply to patients undergoing BCS. Because patients undergoing mastectomy were excluded from analysis, we cannot extrapolate the findings from this study to those patients. It is possible that SLN-positive patients undergoing mastectomy, especially with reconstruction, might experience greater morbidity with a second surgery, reinforcing the importance of immediate cALND in this group. Patients in the immediate cALND had a higher burden of axillary metastasis, suggesting that when metastases to SLNs are sufficiently large to be discovered during the SLN procedure, non-SLNs are more likely to harbor metastasis. This observation is in keeping with the importance of method of detection in models for predicting the likelihood of non-SLN metastasis for patients with positive SLNs.21 This finding has relevance to intraoperative decision making for patients with positive SLNs that are detected intraoperatively, suggesting that cALND should be performed in this setting. For such patients, strategies of either observation or radiation seem less compelling. In summary, pathologic assessment and morbidity after a delayed cALND were not significantly different from when the cALND was performed at the same time as the SLN procedure. In this respect, there is no disadvantage to delayed dissection. However, there is considerable added cost and emotional stress to the patient associated with additional surgical procedures, so we emphasize that it is best to use intraoperative SLN assessment and perform cALND when SLN metastases are found intraoperatively. For the few patients who wish to defer a decision regarding cALND (until all pathologic data have been retrieved) to assess risk of additional nodal metastases by inputting those data into a nomogram, our study shows that no increase in complications or compromise in pathologic assessment occurs after a delayed cALND. These data may be most useful when the SLN metastasis is discovered postoperatively. In this circumstance, there is often considerable multidisciplinary debate over whether there is greater morbidity to a delayed cALND, often favoring a decision to radiate the axilla in lieu of cALND. Our data indicate that concern of increased complications in a delayed ALND (compared with immediate cALND) should not be used to justify nonoperative treatment. The European Organisation for Research and Treatment of Cancer 10981 trial, which randomly assigned SLN-positive patients to cALND versus axillary radiation, may clarify the benefits and complications of these approaches in a comparative fashion. Until then, results of this study will inform decisions for axillary management in breast cancer patients undergoing BCS with metastasis to SLNs.
The author(s) indicated no potential conflicts of interest.
Conception and design: John A. Olson Jr, Armando E. Giuliano Administrative support: Kelly K. Hunt Provision of study materials or patients: Peter Beitsch, Pat W. Whitworth, Douglas S. Reintgen, Peter W. Blumencranz, A. Marilyn Leitch, Sukamal Saha, Kelly K. Hunt, Armando E. Giuliano Collection and assembly of data: Linda McCall Data analysis and interpretation: John A. Olson Jr, Linda McCall, A. Marilyn Leitch, Kelly K. Hunt Manuscript writing: John A. Olson Jr, Kelly K. Hunt Final approval of manuscript: John A. Olson Jr, A. Marilyn Leitch, Kelly K. Hunt
Supported by Grant No. U10-CA76001-11 from the National Cancer Institute, Bethesda, MD. Presented at the 60th Annual Cancer Symposium of the Society of Surgical Oncology, March 15-18, 2007, Washington, DC. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article. Clinical Trials repository link available on www.JCO.org.
1. Lyman GH, Giuliano AE, Somerfield MR, et al: American Society of Clinical Oncology guideline recommendations for sentinel lymph node biopsy in early-stage breast cancer. J Clin Oncol 23:7703-7720, 2005 2. Mansel RE, Fallowfield L, Kissin M, et al: Randomized multicenter trial of sentinel node biopsy versus standard axillary treatment in operable breast cancer: The ALMANAC Trial. J Natl Cancer Inst 98:599-609, 2006 3. Veronesi U, Paganelli G, Viale G, et al: A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer. N Engl J Med 349:546-553, 2003 4. Orr RK: The impact of prophylactic axillary node dissection on breast cancer survival: A Bayesian meta-analysis. Ann Surg Oncol 6:109-116, 1999[CrossRef][Medline] 5. Bland KI, Scott-Conner CE, Menck H, et al: Axillary dissection in breast-conserving surgery for stage I and II breast cancer: A National Cancer Data Base study of patterns of omission and implications for survival. J Am Coll Surg 188:586-595, 1999; discussion 595-596[CrossRef][Medline] 6. Harris JR, Osteen RT: Patients with early breast cancer benefit from effective axillary treatment. Breast Cancer Res Treat 5:17-21, 1985[CrossRef][Medline] 7. Liljegren G, Holmberg L: Arm morbidity after sector resection and axillary dissection with or without postoperative radiotherapy in breast cancer stage I: Results from a randomised trial—Uppsala-Orebro Breast Cancer Study Group. Eur J Cancer 33:193-199, 1997[CrossRef][Medline] 8. Warmuth MA, Bowen G, Prosnitz LR, et al: Complications of axillary lymph node dissection for carcinoma of the breast: A report based on a patient survey. Cancer 83:1362-1368, 1998[CrossRef][Medline] 9. Kuehn T, Klauss W, Darsow M, et al: Long-term morbidity following axillary dissection in breast cancer patients: Clinical assessment, significance for life quality and the impact of demographic, oncologic and therapeutic factors. Breast Cancer Res Treat 64:275-286, 2000[CrossRef][Medline] 10. Petrek JA, Senie RT, Peters M, et al: Lymphedema in a cohort of breast carcinoma survivors 20 years after diagnosis. Cancer 92:1368-1377, 2001[CrossRef][Medline] 11. Silberman AW, McVay C, Cohen JS, et al: Comparative morbidity of axillary lymph node dissection and the sentinel lymph node technique: Implications for patients with breast cancer. Ann Surg 240:1-6, 2004[CrossRef][Medline] 12. Wilke LG, McCall LM, Posther KE, et al: Surgical complications associated with sentinel lymph node biopsy: Results from a prospective international cooperative group trial. Ann Surg Oncol 13:491-500, 2006[CrossRef][Medline] 13. Fleissig A, Fallowfield LJ, Langridge CI, et al: Post-operative arm morbidity and quality of life: Results of the ALMANAC randomised trial comparing sentinel node biopsy with standard axillary treatment in the management of patients with early breast cancer. Breast Cancer Res Treat 95:279-293, 2006[CrossRef][Medline] 14. Temple LK, Baron R, Cody HS, III, et al: Sensory morbidity after sentinel lymph node biopsy and axillary dissection: A prospective study of 233 women. Ann Surg Oncol 9:654-662, 2002[CrossRef][Medline] 15. Lucci A, McCall LM, Beitsch PD, et al: Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J Clin Oncol 25:3657-3663, 2007 16. Posther KE, McCall LM, Blumencranz PW, et al: Sentinel node skills verification and surgeon performance: Data from a multicenter clinical trial for early-stage breast cancer. Ann Surg 242:593-599, 2005; discussion 599-602[Medline] 17. Brogi E, Torres-Matundan E, Tan LK, et al: The results of frozen section, touch preparation, and cytological smear are comparable for intraoperative examination of sentinel lymph nodes: A study in 133 breast cancer patients. Ann Surg Oncol 12:173-180, 2005[CrossRef][Medline] 18. Weiser MR, Montgomery LL, Susnik B, et al: Is routine intraoperative frozen-section examination of sentinel lymph nodes in breast cancer worthwhile? Ann Surg Oncol 7:651-655, 2000[CrossRef][Medline] 19. Rubio IT, Korourian S, Cowan C, et al: Use of touch preps for intraoperative diagnosis of sentinel lymph node metastases in breast cancer. Ann Surg Oncol 5:689-694, 1998[CrossRef][Medline] 20. Mansel RE, Goyal A: European studies on breast lymphatic mapping. Semin Oncol 31:304-310, 2004[CrossRef][Medline] 21. Van Zee KJ, Manasseh DM, Bevilacqua JL, et al: A nomogram for predicting the likelihood of additional nodal metastases in breast cancer patients with a positive sentinel node biopsy. Ann Surg Oncol 10:1140-1151, 2003[CrossRef][Medline] Submitted December 3, 2007; accepted March 14, 2008.
Related Editorial
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 tests were used to examine the relationship between both adverse effects (including axillary seroma, wound infection, paresthesia, and lymphedema) and clinical characteristics, including age, stage, body mass index (BMI), and the number of total and positive SLNs and ALNs removed, and timing of ALND. All analyses were performed with SAS statistical analysis software, version 8 (SAS Institute, Cary, NC). Values of P < .05 were considered significant. 
