- Email: [email protected]
Does Lymphovascular Invasion Predict Regional Nodal Failure in Breast Cancer Patients With Zero to Three Positive Lymph Nodes Treated With Conserving Surgery and Radiotherapy? Implications for Regional Radiation
Int. J. Radiation Oncology Biol. Phys., Vol. 78, No. 3, pp. 793–798, 2010 Copyright Ó 2010 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$–see front matter
doi:10.1016/j.ijrobp.2009.08.049
CLINICAL INVESTIGATION
Breast
DOES LYMPHOVASCULAR INVASION PREDICT REGIONAL NODAL FAILURE IN BREAST CANCER PATIENTS WITH ZERO TO THREE POSITIVE LYMPH NODES TREATED WITH CONSERVING SURGERY AND RADIOTHERAPY? IMPLICATIONS FOR REGIONAL RADIATION RIMOUN BOUTRUS, M.D.,* RITA ABI-RAAD, M.D.,* ANDRZEJ NIEMIERKO, PH.D.,* ELENA F. BRACHTEL, M.D.,y LEVI RIZK, B.S.,* ALEXANDRA KELADA, B.S.,* AND ALPHONSE G. TAGHIAN, M.D., PH.D.* Departments of *Radiation Oncology and yPathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA Purpose: To examine the relationship between lymphovascular invasion (LVI) and regional nodal failure (RNF) in breast cancer patients with zero to three positive nodes treated with breast-conservation therapy (BCT). Methods and Materials: The records of 1,257 breast cancer patients with zero to three positive lymph nodes were reviewed. All patients were treated with BCT at Massachusetts General Hospital from 1980 to December 2003. Lymphovascular invasion was diagnosed by hematoxylin and eosin–stained sections and in some cases supported by immunohistochemical stains. Regional nodal failure was defined as recurrence in the ipsilateral supraclavicular, axillary, or internal mammary lymph nodes. Regional nodal failure was diagnosed by clinical and/or radiologic examination. Results: The median follow-up was 8 years (range, 0.1–21 years). Lymphovascular invasion was present in 211 patients (17%). In univariate analysis, patients with LVI had a higher rate of RNF (3.32% vs. 1.15%; p = 0.02). In multivariate analysis, only tumor size, grade, and local failure were significant predictors of RNF (p = 0.049, 0.013, and 0.0001, respectively), whereas LVI did not show a significant relationship with RNF (hazard ratio = 2.07; 95% CI, 0.8–5.5; p = 0.143). The presence of LVI in the T2/3 population did not increase the risk of RNF over that for those with no LVI (p = 0.15). In addition, patients with Grade 3 tumors and positive LVI did not have a higher risk of RNF than those without LVI (p = 0.96). Conclusion: These results suggest that LVI can not be used as a sole indicator for regional nodal irradiation in breast cancer patients with zero to three positive lymph nodes treated with BCT. Ó 2010 Elsevier Inc. Breast cancer, Breast-conserving therapy, Regional nodal failure, Lymphovascular invasion, Regional irradiation.
lar invasion (LVI) remained a strong independent predictor for regional nodal involvement in most studies (7–9); however, its relationship to RNF and its ability to identify patients with zero to three positive nodes who are at a higher risk for developing RNF remain uncertain. The aim of this study was to examine the relationship between the presence of LVI in the primary tumor and the risk of RNF in breast cancer patients with zero to three positive lymph nodes treated with breast-conservation therapy.
INTRODUCTION Breast-conservative surgery plus radiotherapy has been established as an alternative for mastectomy in early-stage breast cancer (1, 2). For early-stage breast cancer the rate of nodal failure after breast-conservation therapy has remained consistently low in most studies, ranging from 0.5% to 3% in N0 patients and 2% to 6% in N1 patients (3–6); however, identifying the group of patients at higher risk for regional nodal failure (RNF) who would benefit from regional nodal irradiation (RNI) remains unclear. Although the presence of four or more pathologically proven axillary lymph nodes (N2 and N3) has been a standard indication for RNI, there is still no consensus regarding RNI indications in patients with negative or one to three positive lymph nodes. Lymphovascu-
METHODS AND MATERIALS After institutional review board approval with waived informed consent, the medical records of 1695 breast cancer patients treated with breast-conservation therapy at Massachusetts General Hospital
Reprint requests to: Alphonse G. Taghian, M.D., Ph.D., Department of Radiation Oncology, Massachusetts General Hospital, 100 Blossom Street, Cox Building 302, Boston, MA 02114. Tel: (617) 726-6050; Fax: (617) 726-3606; E-mail: [email protected] Presented at the 2009 Annual Meeting of the American Society of Therapeutic Radiology and Oncology (November 2, 2009, Chicago, IL).
Conflict of interest: none. Acknowledgment—Supported in part by the Jane Mailloux Fund, the Blanche Montesi Fund and the Tim Levy Fund for Breast Cancer Research (AGT). Received March 20, 2009, and in revised form Aug 19, 2009. Accepted for publication Aug 20, 2009. 793
794
I. J. Radiation Oncology d Biology d Physics
during the period January 1, 1980 to December 31, 2003 were retrospectively analyzed. Because the aim of the study was to evaluate the significance of LVI as a predictor of nodal failure, patients with unknown LVI status were excluded (n = 251). Patients with incomplete pathologic information regarding tumor size and nodal status (n = 133) or incomplete treatment information (n = 54) were also excluded. Nonmetastatic patients with invasive ductal or lobular carcinomas, pT1–3, pN0–1, and with complete pathologic information composed the study cohort for this analysis (n = 1257). All patients underwent conservative breast surgery, which included removal of the primary tumor with the intention of achieving free surgical margin, and surgical axillary staging in the form of axillary dissection or sentinel lymph node biopsy. Tumors were diagnosed in accordance with pathologic practice (10, 11). Lymphovascular invasion was defined as the presence of tumor in endothelial-lined spaces, located in uninvolved breast tissue in the vicinity or away from the tumor mass. Lymphovascular invasion was mostly diagnosed by conventional hematoxylin and eosin–stained sections, in some cases supported by immunohistochemical stains for lymphatic vessel markers that recently became available (12). Patients were considered to have suspicious LVI status when their tumors demonstrated some, but not all, of the criteria essential for the diagnosis of LVI, and those patients were considered as LVI negative in the present analysis. Other tumor factors analyzed included histologic grade, tumor size, nodal status, estrogen receptor status, and surgical margin status. Surgical margins were defined as negative with a surgical clearance of at least 0.2 cm, or close (<0.2 cm). Assessment of histologic grade was based on nuclear pleomorphism, tubule formation, and mitotic activity (13). The histopathologic evaluation of lymph nodes was routinely evaluated by (hematoxylin and eosin) stained sections; sentinel lymph nodes mostly included additional tissue sections stained with cytokeratin (immunohistochemistry). Postoperative radiotherapy was given to all patients through 4–6-MV linear accelerators delivering a dose of 45–50.4 Gy, followed by a 10–16-Gy boost. Regional lymph nodes received radiation treatment using a supraclavicular field in 10% of patients (n = 123), axillary field in 7.5% (n = 95), and/or internal mammary field in 5% (n = 64). Patients who received radiation through high tangential fields were excluded from the study. Regional nodal failure was defined as any recurrence of tumor as first event in the ipsilateral supraclavicular, axillary, or internal mammary lymph nodes, with or without local recurrence and/or distant metastases. Regional nodal failure was diagnosed either by clinical or radiologic examination. Time to nodal failure and overall survival (OS) were analyzed using univariate (Kaplan-Meier) and multivariate (Cox proportional hazards) methods. The equivalence of survival and cumulative incidence curves was tested using the log–rank test. In our 1257 cases with 19 events, assuming 0.8 power and a of 0.05, we were able to detect a fourfold increase in the hazard ratio (HR) using a two-sided test. We used the method of Freedman (14) and Schoenfeld (15). The strength of association between various risk/prognostic factors and the number of events of interest was evaluated using Fisher’s exact test. A p value of <0.05 was considered statistically significant. All tests of significance were two sided.
RESULTS Patient and tumor characteristics The median follow-up time was 8 years (range, 0.1–21 years). Median age at diagnosis was 55 years (range, 25–85
Volume 78, Number 3, 2010
Table 1. Patient, tumor, and treatment characteristics Age (y) Median, range <55 >55 Tumor size (cm), median, range pT stage T1 T2 T3 Tx Nodal stage N0 N1 Grade 1 2 3 Not assessed Lymphovascular invasion status Positive Negative Suspicious Final surgical margin Negative Positive or close Unknown Estrogen receptor status Positive Negative Unknown Radiotherapy Breast only Breast and supraclavicular Systemic therapy Chemotherapy only Hormonal therapy only Chemo- plus hormonal therapy No systemic therapy
55, 25–85 597 (47.5) 660 (52.5) 1.2 (0.05–6) 1038 (82.6) 212 (16.9) 4 (0.3) 3 (0.2) 961 (76.3) 296 (23.7) 267 (21.2) 553 (44.1) 347 (27.7) 90 (7) 211 (16.7) 979 (78) 67 (5.3) 1032 (82.1) 213 (17) 12 (0.9) 939 (74.7) 210 (16.7) 108 (8.6) 1099 (87.4) 158 (12.6) 155 (12.3) 499 (39.7) 305 (24.3) 298 (23.7)
Values are number (percentage) unless otherwise noted.
years). Lymphovascular invasion was detected in 17% of the cases (n = 211). Clinical and pathologic characteristics of the study cohort are illustrated in Tables 1 and 2. Because of the small number of patients with T3 tumors (n = 4), they were combined with patients with T2 tumors in one group for the sake of the analysis. No nodal failures were observed among those with T3 tumors. All patients were treated with conservative breast surgery and axillary evaluation (axillary dissection in 85.9% of the patients or sentinel node sampling in 14.1% of the patients), followed by postoperative radiotherapy. Patients received radiation treatment either to breast only (n = 1099, 87.4%) or to the breast and regional lymph nodes (n = 158, 12.6%). Adjuvant systemic therapy (hormonal or chemotherapy) was given to 76.8% of cases (n = 959). Nineteen patients (1.5%) developed RNF in our cohort; the distribution of failures is summarized in Table 3. Univariate analysis Univariate analysis of the effect of clinicopathologic variables on RNF is presented in Table 4. Patients with
LVI as a predictor of regional nodal failure d R. BOUTRUS et al.
Table 2. Patient, tumor, and treatment characteristics stratified by LVI status Characteristic Age (y), median, range pT stage T1 T2 T3 Tx pN stage N0 N1 Grade 1 2 3 Not assessed Estrogen receptor status Positive Negative Unknown Radiotherapy Breast only Breast and supraclavicular Systemic treatment Chemotherapy Hormonal therapy Chemo- plus hormonal therapy None
Table 4. Regional nodal failure in relation to patient, tumor, and treatment characteristics: Univariate analysis
LVI positive (n = 211)
LVI negative (n = 1046)
52 (28–83)
52 (26–77)
153 (72.5) 54 (25.5) 1 (0.5) 3 (1.5)
885 (84.7) 158 (15) 3 (0.3) 0
111 (53) 100 (47)
850 (81) 196 (19)
21 (10) 82 (39) 103 (49) 5 (2)
246 (24) 471 (45) 244 (23) 85 (8)
164 (78) 35 (16) 12 (6) 151 (72) 60 (28)
795
775 (74) 175 (17) 96 (9) 948 (91) 98 (9)
40 (19) 60 (28) 99 (47)
115 (11) 439 (42) 206 (20)
12 (6)
286 (27)
Abbreviation: LVI = lymphovascular invasion. Values are number (percentage) unless otherwise noted.
LVI-positive tumors had a higher rate of nodal recurrences than those with LVI-negative tumors (3.32% vs. 1.05%), and the difference was statistically significant (p = 0.02). Other significant factors associated with nodal failures were tumor grade (p = 0.008; Fig. 1), tumor size (p = 0.002; Fig. 2), estrogen receptor status (p = 0.008), and previous local failure (p < 0.001). In the LVI-positive group (n = 211), 60 patients received RNI and 3 of them (5%) developed RNF; 151 did not receive RNI and 4 of them (2.5%) developed RNF. The difference in RNFs between those who received and those who did not receive RNI was not statistically significant (p = 0.31).
Table 3. Sites of regional nodal failure Site of nodal failure
n (%)
SC Axilla IMC Axilla, SC Axilla, IMC SC, IMC Axilla, SC, IMC
7 (37) 6 (32) 1 (5) 2 (11) 1 (5) 1 (5) 1 (5)
Abbreviations: SC = supraclavicular region; IMC = internal mammary chain.
Characteristic Age (y) <40 40–60 >60 Tumor size (cm) <1.2* $1.2 pT Stage T1 T2/3 pN Stage N0 N1 Margin status Close or positive Negative Unknown LVI Status Negative Positive Grade 1 2 3 Not assessed ER status Negative Positive Unknown Radiotherapy Breast only Breast and supraclavicular Systemic treatment Yes No
Total no. Nodal failures, of patients n (%)
p 0.14
110 677 470
4 (3.64) 8 (1.18) 7 (1.49)
571 686
4 (0.7) 15 (2.19)
1038 216
11 (1.06) 8 (3.7)
961 296
13 (1.35) 6 (2.03)
213 1032 12
4 (1.88) 15 (1.45) 0
1046 211
11 (1.05) 7 (3.32)
267 553 347 90
0 6 (1.08) 12 (3.45) 1 (1.12)
210 939 108
8 (3.81) 9 (0.96) 2 (1.85)
1099 158
14 (1.27) 5 (3.16)
959 298
14 (1.46) 5 (1.68)
0.028 0.002 0.425 0.603
0.022 0.001
0.005
0.018 0.740
Abbreviations: LVI = lymphovascular invasion; ER = estrogen receptor. * Median tumor size.
The significance of the ratio of the number of positive nodes to the total number of the nodes removed from the axilla was analyzed but was not significantly associated with RNF (p = 0.161). Multivariate analysis On considering tumor size, grade, LVI status, and local failure in a Cox regression model, tumor size, grade, and local failure remained significantly associated with RNF. For tumor size (T1 vs. T2/3) the HR was 1.49 (95% CI, 1.00–2.22; p = 0.049); for tumor grade (1 and 2 vs. 3), the HR was 3.64 (95% CI, 1.30–10.15; p = 0.013); for local failure the HR was 8.6 (95% CI, 3.01–24.65; p = 0.0001). However, LVI status was not significant: HR 2.07 (95% CI, 0.8–5.52; p = 0.143). Further subgroup analysis by combining LVI and various risk factors (Table 5) revealed that the presence of LVI in the T2/3 population did not increase the risk of RNF over that for patients with no LVI (p = 0.15); additionally, patients with
796
I. J. Radiation Oncology d Biology d Physics
Volume 78, Number 3, 2010
Table 5. Regional nodal failure in patients with T2/3, Grade 3, and N1 disease, stratified according to presence or absence of LVI LVI positive Parameter
LVI negative
Total Nodal Total Nodal no failures, n (%) no. failures, n (%)
Grade 3 103 T2/3 55 N1 100 Grade 3 and T2/3 34
4 (3.8) 4 (7.27) 4 (4) 3 (8.82)
244 161 196 72
8 (3.27) 4 (2.48) 2 (1.02) 3 (4.17)
p 0.96 0.15 0.08 0.456
Abbreviation: LVI = lymphovascular invasion.
Fig. 1. Cumulative incidence of regional nodal failure in relation to tumor grade.
Grade 3 tumors and positive LVI status did not have a higher risk of RNF than those without. Although there was a trend toward a higher incidence of RNF among patients with N1 and positive LVI vs. negative LVI status, this difference was not statistically significant (p = 0.08). When the patients with both Grade 3 and T2/3 tumors were analyzed compared to those with Grade 1/2 and T1 tumors (Fig. 3), there was clear evidence of higher nodal failure rate among the former: 5.6% vs. 0.5% (HR 11.5; p = 0.0001). No difference in the rate of RNF was observed when those patients were further stratified according to the presence or absence of LVI (p = 0.456). Analysis of OS In the present analysis there was no statistically significant difference in OS between patients with LVI positive tumors and those without (Figs. 4 and 5). On multivariate analysis the only factors that were shown to be independently associated with significant OS difference were age (HR 1.04; p < 0.001; 95% CI, 1.02–1.06), tumor Grade 1 and 2 vs. 3 (HR 2.50; p < 0.001; 95% CI, 1.68–3.74), tumor size (HR 1.49; p < 0.0001; 95% CI 1.22–1.81), margin status negative vs. close or positive (HR 1.57; p = 0.040; 95% CI 1.02–2.42), and systemic therapy administration (HR
Fig. 2. Cumulative incidence of regional nodal failure in relation to tumor size.
0.5; p = 0.002; 95% CI 0.32–0.77; HR <1 denoting reduction in the probability of mortality with systemic therapy administration). DISCUSSION In the present study we tried to assess whether the presence of LVI can be used as a prognostic factor imposing a higher risk for nodal failure in women with Stage I/II breast cancer treated with conservative breast therapy. We found that there was a trend towards more nodal failures in the group of patients having LVI in their primary tumors (3.32% vs. 1.15% for LVI-negative tumors); this difference was statistically significant on univariate analysis (p = 0.02) but failed to maintain this significance in the multivariate analysis, in which HR was 2.07 (95% CI, 0.78–5.52; p = 0.143). Tumor size, grade, and local failure were the only variables proved to be independent prognostic factors for nodal recurrence on multivariate analysis: HR 1.49 for tumor size (95% CI, 1.00–2.22; p = 0.049), HR 3.64 for tumor grade (95% CI, 1.30–10.15; p = 0.013), and HR 8.6 for local failure (95% CI, 3.01–24.65; p = 0.0001). The crude rate of nodal failure at 10 years was higher among those with Grade 3 vs. Grade 1/2 (3.45% vs. 1.08%) and those with T2/3 vs. T1 (3.7% vs. 1.06%) tumors, yet the clinical significance could be seen when both adverse factors were present; RNF rate was 5.65% in patients with Grade 3 and T2/3 vs.
Fig. 3. Cumulative incidence of regional nodal failure in patients with Grade 3 and T2/3 tumors vs. Grade 1/2 and T1.
LVI as a predictor of regional nodal failure d R. BOUTRUS et al.
Fig. 4. Comparison of overall survival for node-negative patients with and without lymphovascular invasion (LVI). Log–rank, p = 0.3.
0.5% in patients with Grade 1/2 and T1, driving the HR up to 11.5 (p = 0.0001). When it comes to decision making, RNI for patients with zero to three positive lymph nodes remains a real challenge. In a survey of 1137 radiation oncologists from Europe and North America conducted by Ceilley et al. (16), 40.7% of the North Americans and 36.1% of the Europeans responded they would treat the supraclavicular area with a third field in patients with one to three positive nodes treated with lumpectomy. Faced with the low incidence of RNFs in early-stage breast cancer (1.5% in our study and 0.5–6% in most of the reported series [3–6]), clinicians always had to weigh the benefits from adding a supraclavicular field covering the supraclavicular area and the axillary apex vs. the potential harmful effects of radiation, such as lymphedema, pneumonitis, and brachial plexopathy (17, 18). Attempts have been made by many investigators to identify patient and tumor characteristics that would help guide the management of such patients. Lymphovascular invasion has been evaluated in many studies in an attempt to determine its prognostic significance in different breast cancer clinical settings. Galper et al. (19) retrospectively analyzed 691 patients with clinical Stage I and II breast cancer treated with gross total resection and tangential radiotherapy alone; the rate of regional nodal recurrence among patients with LVI was 6%, compared with 3% in patients without LVI, and this difference was not statistically significant (p = 0.18). In a series of 1500 patients with Stage I and II breast cancer, Grills et al. (20) reported the RNF incidence at 10 years to be 2.8%; lymphovascular invasion showed a statistically significant association with RNF on univariate analysis (p = 0.002), but when considered in a multivariate model with RNI, age, size of lymph node metastases, and the percentage of positive lymph nodes, RNI was the only significant independent factor. Strom et al. (21) analyzed a cohort of 1031 patients treated with mastectomy including axillary dissection and without postmastectomy radiotherapy. Lymphovascular invasion was not shown to be a significant predictor for nodal failures
797
Fig. 5. Comparison of overall survival for node-positive patients with and without lymphovascular invasion (LVI). Log–rank, p = 0.2.
in low–mid axilla, either on univariate or multivariate analysis. For high axillary and supraclavicular failures, LVI was found to be a significant predictor of failure on univariate (p = 0.0008) as well as multivariate analysis (HR 1.89; p = 0.007). However, when the analysis was restricted to patients with T1/T2 disease and one to three positive lymph nodes, the presence of LVI was not a significant predictor for high axillary failures (p = 0.65). The investigators concluded that RNI is not warranted for patient with fewer than four positive lymph nodes. In a large cohort of patients (n = 5688), Truong et al. (22) found an association between the presence of LVI and locoregional relapse on multivariate analysis (HR 1.45; p = 0.002; 95% CI, 1.17–2.02); however, when regional relapse was analyzed separately from local relapse, the 10-year KaplanMeier analysis did not show a significant relationship between LVI and regional relapse (p = 0.71). Lymphovascular invasion did not affect OS in our patient population; on multivariate analysis the only factors shown to be independently significant for OS difference were age, tumor grade, tumor size, margin status, and systemic therapy administration. In a recent study from the Danish Breast Cancer Cooperative Group (23) with 15,659 patients included, Ejlertsen et al. reported no effect of LVI on invasive-disease–free survival or OS in early-stage breast cancer patients (HR for invasive disease 0.84 [95% CI, 0.40–1.77; p = 0.65]; HR for death 0.44 [95% CI, 0.14–1.36; p = 0.15]). The investigators concluded that their results do not support that LVI has sufficient independent prognostic influence to move patients from a low-risk group to a high-risk group. The retrospective nature of our study carries the risk of selection and treatment bias. In addition, because the number of regional nodal relapses in our series was small, we were unable to conduct more subgroup analyses. Our data showed a higher rate of RNF among patients who received RNI (3.16%) than in those who did not receive RNI (1.27%). These data were reviewed in detail by two coauthors independently (R.B. and R.A.R.); we are unable to find any plausible
798
I. J. Radiation Oncology d Biology d Physics
explanation. However, we believe that the retrospective nature of the study and the relatively small number of events in this group would make such observation not amenable to generalization. Another important limitation is the diversity in clinical practice regarding surgical management of the axilla over time, which may have influenced treatment outcomes.
Volume 78, Number 3, 2010
CONCLUSION The results of our study suggest that LVI should not be used as a sole or the main indicator for RNI in breast cancer patients with zero to three positive lymph nodes treated with conservative breast surgery. However, large tumor size and high grade could potentially be considered when evaluating the risk of nodal failure.
REFERENCES 1. Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year followup of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002;347:1227–1232. 2. Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002;347:1233–1241. 3. Recht A, Pierce S, Abner A, et al. Regional nodal failure after conservative surgery and radiotherapy for early-stage breast carcinoma. J Clin Oncol 1991;9:988–996. 4. Halverson KJ, Taylor ME, Perez CA, et al. Regional nodal management and patterns of failure following conservative surgery and radiation therapy for stage I and II breast cancer. Int J Radiat Oncol Biol Phys 1993;26:593–599. 5. Renolleau C, Merviel P, Clough KB, et al. Isolated axillary recurrences after conservative treatment of breast cancer. Eur J Cancer 1996;32A:617–621. 6. Fowble B, Solin L, Schultz D, et al. Frequency, sites of relapse, and outcome of regional node failures following conservative surgery and radiation for early breast cancer. Int J Radiat Oncol Biol Phys 1989;17:703–710. 7. Davis B, Gelber R, Goldhirsch A, et al. Prognostic significance of peritumoral vessel invasion in clinical trials of adjuvant therapy for breast cancer with axillary node metastasis. Hum Pathol 1985;16:1212–1218. 8. Ravdin P, De Laurentiis M, Vendely T, et al. Prediction of axillary lymph node status in breast cancer patients by use of prognostic indicators. J Natl Cancer Inst 1994;86:1771–1775. 9. Wong J, O’Neill A, Recht A, et al. The relationship between lymphatic vessel invasion, tumor size, and pathologic nodal status: Can we predict who can avoid a third field in the absence of axillary dissection. Int J Radiat Oncol Biol Phys 2000;48: 133–137. 10. Rosen PP. Rosen’s breast pathology. 3rd ed. Philadelphia: Lippincott, Williams & Wilkins; 2008. 11. Tavassoli FA, Devilee P. Pathology and genetics of tumours of the breast and female genital organs. Lyon, France: IARC Press; 2003. 221–232. 12. Mohammed R, Martin S, Gill M, et al. Improved methods of detection of lymphovascular invasion demonstrate that it is the predominant method of vascular invasion in breast cancer
13. 14. 15. 16. 17.
18. 19.
20.
21.
22.
23.
and has important clinical consequences. Am J Surg Pathol 2007;31:1825–1833. Robbins P, Pinder S, de Klerk N, et al. Histological grading of breast carcinomas: A study of interobserver agreement. Hum Pathol 1995;26:873–879. Freedman LS. Tables of the number of patients required in clinical trials using the logrank test. Stat Med 1982;1:121–129. Schoenfeld D. The asymptotic properties of nonparametric tests for comparing survival distributions. Biometrika 1981;68: 316–319. Ceilley E, Jagsi R, Goldberg S, et al. Radiotherapy for invasive breast cancer in North America and Europe: Results of a survey. Int J Radiat Oncol Biol Phys 2005;61:365–373. Pierce S, Recht A, Lingos T, et al. Long-term radiation complications following conservative surgery (CS) and radiation therapy (RT) in patients with early stage breast cancer. Int J Radiat Oncol Biol Phys 1992;23:915–923. Lingos T, Recht A, Vicini F, et al. Radiation pneumonitis in breast cancer patients treated with conservative surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1991;21:355–360. Galper S, Recht A, Silver B, et al. Factors associated with regional nodal failure in patients with early stage breast cancer with 0-3 positive axillary nodes following irradiation alone. Int J Radiat Oncol Biol Phys 1999;45:1157–1166. Grills I, Kestin L, Goldstein N, et al. Risk factors for regional nodal failure after breast conserving therapy: Regional nodal irradiation reduces rate of axillary failure in patients with four or more positive lymph nodes. Int J Radiat Oncol Biol Phys 2003;56:658–670. Strom E, Woodward W, Katz A, et al. Clinical investigation: Regional nodal failure patterns in breast cancer patients treated with mastectomy without radiotherapy. Int J Radiat Oncol Biol Phys 2005;63:1508–1513. Truong P, Jones S, Kader H, et al. Patients with T1 to T2 breast cancer with one to three positive nodes have higher local and regional recurrence risks compared with node-negative patients after breast-conserving surgery and whole-breast radiotherapy. Int J Radiat Oncol Biol Phys 2009;73:357–364. Ejlertsen B, Jensen M, Rank F, et al. Population-based study of peritumoral lymphovascular invasion and outcome among patients with operable breast cancer. J Natl Cancer Inst 2009; 101:729–735.
doi:10.1016/j.ijrobp.2009.08.049
CLINICAL INVESTIGATION
Breast
DOES LYMPHOVASCULAR INVASION PREDICT REGIONAL NODAL FAILURE IN BREAST CANCER PATIENTS WITH ZERO TO THREE POSITIVE LYMPH NODES TREATED WITH CONSERVING SURGERY AND RADIOTHERAPY? IMPLICATIONS FOR REGIONAL RADIATION RIMOUN BOUTRUS, M.D.,* RITA ABI-RAAD, M.D.,* ANDRZEJ NIEMIERKO, PH.D.,* ELENA F. BRACHTEL, M.D.,y LEVI RIZK, B.S.,* ALEXANDRA KELADA, B.S.,* AND ALPHONSE G. TAGHIAN, M.D., PH.D.* Departments of *Radiation Oncology and yPathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA Purpose: To examine the relationship between lymphovascular invasion (LVI) and regional nodal failure (RNF) in breast cancer patients with zero to three positive nodes treated with breast-conservation therapy (BCT). Methods and Materials: The records of 1,257 breast cancer patients with zero to three positive lymph nodes were reviewed. All patients were treated with BCT at Massachusetts General Hospital from 1980 to December 2003. Lymphovascular invasion was diagnosed by hematoxylin and eosin–stained sections and in some cases supported by immunohistochemical stains. Regional nodal failure was defined as recurrence in the ipsilateral supraclavicular, axillary, or internal mammary lymph nodes. Regional nodal failure was diagnosed by clinical and/or radiologic examination. Results: The median follow-up was 8 years (range, 0.1–21 years). Lymphovascular invasion was present in 211 patients (17%). In univariate analysis, patients with LVI had a higher rate of RNF (3.32% vs. 1.15%; p = 0.02). In multivariate analysis, only tumor size, grade, and local failure were significant predictors of RNF (p = 0.049, 0.013, and 0.0001, respectively), whereas LVI did not show a significant relationship with RNF (hazard ratio = 2.07; 95% CI, 0.8–5.5; p = 0.143). The presence of LVI in the T2/3 population did not increase the risk of RNF over that for those with no LVI (p = 0.15). In addition, patients with Grade 3 tumors and positive LVI did not have a higher risk of RNF than those without LVI (p = 0.96). Conclusion: These results suggest that LVI can not be used as a sole indicator for regional nodal irradiation in breast cancer patients with zero to three positive lymph nodes treated with BCT. Ó 2010 Elsevier Inc. Breast cancer, Breast-conserving therapy, Regional nodal failure, Lymphovascular invasion, Regional irradiation.
lar invasion (LVI) remained a strong independent predictor for regional nodal involvement in most studies (7–9); however, its relationship to RNF and its ability to identify patients with zero to three positive nodes who are at a higher risk for developing RNF remain uncertain. The aim of this study was to examine the relationship between the presence of LVI in the primary tumor and the risk of RNF in breast cancer patients with zero to three positive lymph nodes treated with breast-conservation therapy.
INTRODUCTION Breast-conservative surgery plus radiotherapy has been established as an alternative for mastectomy in early-stage breast cancer (1, 2). For early-stage breast cancer the rate of nodal failure after breast-conservation therapy has remained consistently low in most studies, ranging from 0.5% to 3% in N0 patients and 2% to 6% in N1 patients (3–6); however, identifying the group of patients at higher risk for regional nodal failure (RNF) who would benefit from regional nodal irradiation (RNI) remains unclear. Although the presence of four or more pathologically proven axillary lymph nodes (N2 and N3) has been a standard indication for RNI, there is still no consensus regarding RNI indications in patients with negative or one to three positive lymph nodes. Lymphovascu-
METHODS AND MATERIALS After institutional review board approval with waived informed consent, the medical records of 1695 breast cancer patients treated with breast-conservation therapy at Massachusetts General Hospital
Reprint requests to: Alphonse G. Taghian, M.D., Ph.D., Department of Radiation Oncology, Massachusetts General Hospital, 100 Blossom Street, Cox Building 302, Boston, MA 02114. Tel: (617) 726-6050; Fax: (617) 726-3606; E-mail: [email protected] Presented at the 2009 Annual Meeting of the American Society of Therapeutic Radiology and Oncology (November 2, 2009, Chicago, IL).
Conflict of interest: none. Acknowledgment—Supported in part by the Jane Mailloux Fund, the Blanche Montesi Fund and the Tim Levy Fund for Breast Cancer Research (AGT). Received March 20, 2009, and in revised form Aug 19, 2009. Accepted for publication Aug 20, 2009. 793
794
I. J. Radiation Oncology d Biology d Physics
during the period January 1, 1980 to December 31, 2003 were retrospectively analyzed. Because the aim of the study was to evaluate the significance of LVI as a predictor of nodal failure, patients with unknown LVI status were excluded (n = 251). Patients with incomplete pathologic information regarding tumor size and nodal status (n = 133) or incomplete treatment information (n = 54) were also excluded. Nonmetastatic patients with invasive ductal or lobular carcinomas, pT1–3, pN0–1, and with complete pathologic information composed the study cohort for this analysis (n = 1257). All patients underwent conservative breast surgery, which included removal of the primary tumor with the intention of achieving free surgical margin, and surgical axillary staging in the form of axillary dissection or sentinel lymph node biopsy. Tumors were diagnosed in accordance with pathologic practice (10, 11). Lymphovascular invasion was defined as the presence of tumor in endothelial-lined spaces, located in uninvolved breast tissue in the vicinity or away from the tumor mass. Lymphovascular invasion was mostly diagnosed by conventional hematoxylin and eosin–stained sections, in some cases supported by immunohistochemical stains for lymphatic vessel markers that recently became available (12). Patients were considered to have suspicious LVI status when their tumors demonstrated some, but not all, of the criteria essential for the diagnosis of LVI, and those patients were considered as LVI negative in the present analysis. Other tumor factors analyzed included histologic grade, tumor size, nodal status, estrogen receptor status, and surgical margin status. Surgical margins were defined as negative with a surgical clearance of at least 0.2 cm, or close (<0.2 cm). Assessment of histologic grade was based on nuclear pleomorphism, tubule formation, and mitotic activity (13). The histopathologic evaluation of lymph nodes was routinely evaluated by (hematoxylin and eosin) stained sections; sentinel lymph nodes mostly included additional tissue sections stained with cytokeratin (immunohistochemistry). Postoperative radiotherapy was given to all patients through 4–6-MV linear accelerators delivering a dose of 45–50.4 Gy, followed by a 10–16-Gy boost. Regional lymph nodes received radiation treatment using a supraclavicular field in 10% of patients (n = 123), axillary field in 7.5% (n = 95), and/or internal mammary field in 5% (n = 64). Patients who received radiation through high tangential fields were excluded from the study. Regional nodal failure was defined as any recurrence of tumor as first event in the ipsilateral supraclavicular, axillary, or internal mammary lymph nodes, with or without local recurrence and/or distant metastases. Regional nodal failure was diagnosed either by clinical or radiologic examination. Time to nodal failure and overall survival (OS) were analyzed using univariate (Kaplan-Meier) and multivariate (Cox proportional hazards) methods. The equivalence of survival and cumulative incidence curves was tested using the log–rank test. In our 1257 cases with 19 events, assuming 0.8 power and a of 0.05, we were able to detect a fourfold increase in the hazard ratio (HR) using a two-sided test. We used the method of Freedman (14) and Schoenfeld (15). The strength of association between various risk/prognostic factors and the number of events of interest was evaluated using Fisher’s exact test. A p value of <0.05 was considered statistically significant. All tests of significance were two sided.
RESULTS Patient and tumor characteristics The median follow-up time was 8 years (range, 0.1–21 years). Median age at diagnosis was 55 years (range, 25–85
Volume 78, Number 3, 2010
Table 1. Patient, tumor, and treatment characteristics Age (y) Median, range <55 >55 Tumor size (cm), median, range pT stage T1 T2 T3 Tx Nodal stage N0 N1 Grade 1 2 3 Not assessed Lymphovascular invasion status Positive Negative Suspicious Final surgical margin Negative Positive or close Unknown Estrogen receptor status Positive Negative Unknown Radiotherapy Breast only Breast and supraclavicular Systemic therapy Chemotherapy only Hormonal therapy only Chemo- plus hormonal therapy No systemic therapy
55, 25–85 597 (47.5) 660 (52.5) 1.2 (0.05–6) 1038 (82.6) 212 (16.9) 4 (0.3) 3 (0.2) 961 (76.3) 296 (23.7) 267 (21.2) 553 (44.1) 347 (27.7) 90 (7) 211 (16.7) 979 (78) 67 (5.3) 1032 (82.1) 213 (17) 12 (0.9) 939 (74.7) 210 (16.7) 108 (8.6) 1099 (87.4) 158 (12.6) 155 (12.3) 499 (39.7) 305 (24.3) 298 (23.7)
Values are number (percentage) unless otherwise noted.
years). Lymphovascular invasion was detected in 17% of the cases (n = 211). Clinical and pathologic characteristics of the study cohort are illustrated in Tables 1 and 2. Because of the small number of patients with T3 tumors (n = 4), they were combined with patients with T2 tumors in one group for the sake of the analysis. No nodal failures were observed among those with T3 tumors. All patients were treated with conservative breast surgery and axillary evaluation (axillary dissection in 85.9% of the patients or sentinel node sampling in 14.1% of the patients), followed by postoperative radiotherapy. Patients received radiation treatment either to breast only (n = 1099, 87.4%) or to the breast and regional lymph nodes (n = 158, 12.6%). Adjuvant systemic therapy (hormonal or chemotherapy) was given to 76.8% of cases (n = 959). Nineteen patients (1.5%) developed RNF in our cohort; the distribution of failures is summarized in Table 3. Univariate analysis Univariate analysis of the effect of clinicopathologic variables on RNF is presented in Table 4. Patients with
LVI as a predictor of regional nodal failure d R. BOUTRUS et al.
Table 2. Patient, tumor, and treatment characteristics stratified by LVI status Characteristic Age (y), median, range pT stage T1 T2 T3 Tx pN stage N0 N1 Grade 1 2 3 Not assessed Estrogen receptor status Positive Negative Unknown Radiotherapy Breast only Breast and supraclavicular Systemic treatment Chemotherapy Hormonal therapy Chemo- plus hormonal therapy None
Table 4. Regional nodal failure in relation to patient, tumor, and treatment characteristics: Univariate analysis
LVI positive (n = 211)
LVI negative (n = 1046)
52 (28–83)
52 (26–77)
153 (72.5) 54 (25.5) 1 (0.5) 3 (1.5)
885 (84.7) 158 (15) 3 (0.3) 0
111 (53) 100 (47)
850 (81) 196 (19)
21 (10) 82 (39) 103 (49) 5 (2)
246 (24) 471 (45) 244 (23) 85 (8)
164 (78) 35 (16) 12 (6) 151 (72) 60 (28)
795
775 (74) 175 (17) 96 (9) 948 (91) 98 (9)
40 (19) 60 (28) 99 (47)
115 (11) 439 (42) 206 (20)
12 (6)
286 (27)
Abbreviation: LVI = lymphovascular invasion. Values are number (percentage) unless otherwise noted.
LVI-positive tumors had a higher rate of nodal recurrences than those with LVI-negative tumors (3.32% vs. 1.05%), and the difference was statistically significant (p = 0.02). Other significant factors associated with nodal failures were tumor grade (p = 0.008; Fig. 1), tumor size (p = 0.002; Fig. 2), estrogen receptor status (p = 0.008), and previous local failure (p < 0.001). In the LVI-positive group (n = 211), 60 patients received RNI and 3 of them (5%) developed RNF; 151 did not receive RNI and 4 of them (2.5%) developed RNF. The difference in RNFs between those who received and those who did not receive RNI was not statistically significant (p = 0.31).
Table 3. Sites of regional nodal failure Site of nodal failure
n (%)
SC Axilla IMC Axilla, SC Axilla, IMC SC, IMC Axilla, SC, IMC
7 (37) 6 (32) 1 (5) 2 (11) 1 (5) 1 (5) 1 (5)
Abbreviations: SC = supraclavicular region; IMC = internal mammary chain.
Characteristic Age (y) <40 40–60 >60 Tumor size (cm) <1.2* $1.2 pT Stage T1 T2/3 pN Stage N0 N1 Margin status Close or positive Negative Unknown LVI Status Negative Positive Grade 1 2 3 Not assessed ER status Negative Positive Unknown Radiotherapy Breast only Breast and supraclavicular Systemic treatment Yes No
Total no. Nodal failures, of patients n (%)
p 0.14
110 677 470
4 (3.64) 8 (1.18) 7 (1.49)
571 686
4 (0.7) 15 (2.19)
1038 216
11 (1.06) 8 (3.7)
961 296
13 (1.35) 6 (2.03)
213 1032 12
4 (1.88) 15 (1.45) 0
1046 211
11 (1.05) 7 (3.32)
267 553 347 90
0 6 (1.08) 12 (3.45) 1 (1.12)
210 939 108
8 (3.81) 9 (0.96) 2 (1.85)
1099 158
14 (1.27) 5 (3.16)
959 298
14 (1.46) 5 (1.68)
0.028 0.002 0.425 0.603
0.022 0.001
0.005
0.018 0.740
Abbreviations: LVI = lymphovascular invasion; ER = estrogen receptor. * Median tumor size.
The significance of the ratio of the number of positive nodes to the total number of the nodes removed from the axilla was analyzed but was not significantly associated with RNF (p = 0.161). Multivariate analysis On considering tumor size, grade, LVI status, and local failure in a Cox regression model, tumor size, grade, and local failure remained significantly associated with RNF. For tumor size (T1 vs. T2/3) the HR was 1.49 (95% CI, 1.00–2.22; p = 0.049); for tumor grade (1 and 2 vs. 3), the HR was 3.64 (95% CI, 1.30–10.15; p = 0.013); for local failure the HR was 8.6 (95% CI, 3.01–24.65; p = 0.0001). However, LVI status was not significant: HR 2.07 (95% CI, 0.8–5.52; p = 0.143). Further subgroup analysis by combining LVI and various risk factors (Table 5) revealed that the presence of LVI in the T2/3 population did not increase the risk of RNF over that for patients with no LVI (p = 0.15); additionally, patients with
796
I. J. Radiation Oncology d Biology d Physics
Volume 78, Number 3, 2010
Table 5. Regional nodal failure in patients with T2/3, Grade 3, and N1 disease, stratified according to presence or absence of LVI LVI positive Parameter
LVI negative
Total Nodal Total Nodal no failures, n (%) no. failures, n (%)
Grade 3 103 T2/3 55 N1 100 Grade 3 and T2/3 34
4 (3.8) 4 (7.27) 4 (4) 3 (8.82)
244 161 196 72
8 (3.27) 4 (2.48) 2 (1.02) 3 (4.17)
p 0.96 0.15 0.08 0.456
Abbreviation: LVI = lymphovascular invasion.
Fig. 1. Cumulative incidence of regional nodal failure in relation to tumor grade.
Grade 3 tumors and positive LVI status did not have a higher risk of RNF than those without. Although there was a trend toward a higher incidence of RNF among patients with N1 and positive LVI vs. negative LVI status, this difference was not statistically significant (p = 0.08). When the patients with both Grade 3 and T2/3 tumors were analyzed compared to those with Grade 1/2 and T1 tumors (Fig. 3), there was clear evidence of higher nodal failure rate among the former: 5.6% vs. 0.5% (HR 11.5; p = 0.0001). No difference in the rate of RNF was observed when those patients were further stratified according to the presence or absence of LVI (p = 0.456). Analysis of OS In the present analysis there was no statistically significant difference in OS between patients with LVI positive tumors and those without (Figs. 4 and 5). On multivariate analysis the only factors that were shown to be independently associated with significant OS difference were age (HR 1.04; p < 0.001; 95% CI, 1.02–1.06), tumor Grade 1 and 2 vs. 3 (HR 2.50; p < 0.001; 95% CI, 1.68–3.74), tumor size (HR 1.49; p < 0.0001; 95% CI 1.22–1.81), margin status negative vs. close or positive (HR 1.57; p = 0.040; 95% CI 1.02–2.42), and systemic therapy administration (HR
Fig. 2. Cumulative incidence of regional nodal failure in relation to tumor size.
0.5; p = 0.002; 95% CI 0.32–0.77; HR <1 denoting reduction in the probability of mortality with systemic therapy administration). DISCUSSION In the present study we tried to assess whether the presence of LVI can be used as a prognostic factor imposing a higher risk for nodal failure in women with Stage I/II breast cancer treated with conservative breast therapy. We found that there was a trend towards more nodal failures in the group of patients having LVI in their primary tumors (3.32% vs. 1.15% for LVI-negative tumors); this difference was statistically significant on univariate analysis (p = 0.02) but failed to maintain this significance in the multivariate analysis, in which HR was 2.07 (95% CI, 0.78–5.52; p = 0.143). Tumor size, grade, and local failure were the only variables proved to be independent prognostic factors for nodal recurrence on multivariate analysis: HR 1.49 for tumor size (95% CI, 1.00–2.22; p = 0.049), HR 3.64 for tumor grade (95% CI, 1.30–10.15; p = 0.013), and HR 8.6 for local failure (95% CI, 3.01–24.65; p = 0.0001). The crude rate of nodal failure at 10 years was higher among those with Grade 3 vs. Grade 1/2 (3.45% vs. 1.08%) and those with T2/3 vs. T1 (3.7% vs. 1.06%) tumors, yet the clinical significance could be seen when both adverse factors were present; RNF rate was 5.65% in patients with Grade 3 and T2/3 vs.
Fig. 3. Cumulative incidence of regional nodal failure in patients with Grade 3 and T2/3 tumors vs. Grade 1/2 and T1.
LVI as a predictor of regional nodal failure d R. BOUTRUS et al.
Fig. 4. Comparison of overall survival for node-negative patients with and without lymphovascular invasion (LVI). Log–rank, p = 0.3.
0.5% in patients with Grade 1/2 and T1, driving the HR up to 11.5 (p = 0.0001). When it comes to decision making, RNI for patients with zero to three positive lymph nodes remains a real challenge. In a survey of 1137 radiation oncologists from Europe and North America conducted by Ceilley et al. (16), 40.7% of the North Americans and 36.1% of the Europeans responded they would treat the supraclavicular area with a third field in patients with one to three positive nodes treated with lumpectomy. Faced with the low incidence of RNFs in early-stage breast cancer (1.5% in our study and 0.5–6% in most of the reported series [3–6]), clinicians always had to weigh the benefits from adding a supraclavicular field covering the supraclavicular area and the axillary apex vs. the potential harmful effects of radiation, such as lymphedema, pneumonitis, and brachial plexopathy (17, 18). Attempts have been made by many investigators to identify patient and tumor characteristics that would help guide the management of such patients. Lymphovascular invasion has been evaluated in many studies in an attempt to determine its prognostic significance in different breast cancer clinical settings. Galper et al. (19) retrospectively analyzed 691 patients with clinical Stage I and II breast cancer treated with gross total resection and tangential radiotherapy alone; the rate of regional nodal recurrence among patients with LVI was 6%, compared with 3% in patients without LVI, and this difference was not statistically significant (p = 0.18). In a series of 1500 patients with Stage I and II breast cancer, Grills et al. (20) reported the RNF incidence at 10 years to be 2.8%; lymphovascular invasion showed a statistically significant association with RNF on univariate analysis (p = 0.002), but when considered in a multivariate model with RNI, age, size of lymph node metastases, and the percentage of positive lymph nodes, RNI was the only significant independent factor. Strom et al. (21) analyzed a cohort of 1031 patients treated with mastectomy including axillary dissection and without postmastectomy radiotherapy. Lymphovascular invasion was not shown to be a significant predictor for nodal failures
797
Fig. 5. Comparison of overall survival for node-positive patients with and without lymphovascular invasion (LVI). Log–rank, p = 0.2.
in low–mid axilla, either on univariate or multivariate analysis. For high axillary and supraclavicular failures, LVI was found to be a significant predictor of failure on univariate (p = 0.0008) as well as multivariate analysis (HR 1.89; p = 0.007). However, when the analysis was restricted to patients with T1/T2 disease and one to three positive lymph nodes, the presence of LVI was not a significant predictor for high axillary failures (p = 0.65). The investigators concluded that RNI is not warranted for patient with fewer than four positive lymph nodes. In a large cohort of patients (n = 5688), Truong et al. (22) found an association between the presence of LVI and locoregional relapse on multivariate analysis (HR 1.45; p = 0.002; 95% CI, 1.17–2.02); however, when regional relapse was analyzed separately from local relapse, the 10-year KaplanMeier analysis did not show a significant relationship between LVI and regional relapse (p = 0.71). Lymphovascular invasion did not affect OS in our patient population; on multivariate analysis the only factors shown to be independently significant for OS difference were age, tumor grade, tumor size, margin status, and systemic therapy administration. In a recent study from the Danish Breast Cancer Cooperative Group (23) with 15,659 patients included, Ejlertsen et al. reported no effect of LVI on invasive-disease–free survival or OS in early-stage breast cancer patients (HR for invasive disease 0.84 [95% CI, 0.40–1.77; p = 0.65]; HR for death 0.44 [95% CI, 0.14–1.36; p = 0.15]). The investigators concluded that their results do not support that LVI has sufficient independent prognostic influence to move patients from a low-risk group to a high-risk group. The retrospective nature of our study carries the risk of selection and treatment bias. In addition, because the number of regional nodal relapses in our series was small, we were unable to conduct more subgroup analyses. Our data showed a higher rate of RNF among patients who received RNI (3.16%) than in those who did not receive RNI (1.27%). These data were reviewed in detail by two coauthors independently (R.B. and R.A.R.); we are unable to find any plausible
798
I. J. Radiation Oncology d Biology d Physics
explanation. However, we believe that the retrospective nature of the study and the relatively small number of events in this group would make such observation not amenable to generalization. Another important limitation is the diversity in clinical practice regarding surgical management of the axilla over time, which may have influenced treatment outcomes.
Volume 78, Number 3, 2010
CONCLUSION The results of our study suggest that LVI should not be used as a sole or the main indicator for RNI in breast cancer patients with zero to three positive lymph nodes treated with conservative breast surgery. However, large tumor size and high grade could potentially be considered when evaluating the risk of nodal failure.
REFERENCES 1. Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year followup of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med 2002;347:1227–1232. 2. Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002;347:1233–1241. 3. Recht A, Pierce S, Abner A, et al. Regional nodal failure after conservative surgery and radiotherapy for early-stage breast carcinoma. J Clin Oncol 1991;9:988–996. 4. Halverson KJ, Taylor ME, Perez CA, et al. Regional nodal management and patterns of failure following conservative surgery and radiation therapy for stage I and II breast cancer. Int J Radiat Oncol Biol Phys 1993;26:593–599. 5. Renolleau C, Merviel P, Clough KB, et al. Isolated axillary recurrences after conservative treatment of breast cancer. Eur J Cancer 1996;32A:617–621. 6. Fowble B, Solin L, Schultz D, et al. Frequency, sites of relapse, and outcome of regional node failures following conservative surgery and radiation for early breast cancer. Int J Radiat Oncol Biol Phys 1989;17:703–710. 7. Davis B, Gelber R, Goldhirsch A, et al. Prognostic significance of peritumoral vessel invasion in clinical trials of adjuvant therapy for breast cancer with axillary node metastasis. Hum Pathol 1985;16:1212–1218. 8. Ravdin P, De Laurentiis M, Vendely T, et al. Prediction of axillary lymph node status in breast cancer patients by use of prognostic indicators. J Natl Cancer Inst 1994;86:1771–1775. 9. Wong J, O’Neill A, Recht A, et al. The relationship between lymphatic vessel invasion, tumor size, and pathologic nodal status: Can we predict who can avoid a third field in the absence of axillary dissection. Int J Radiat Oncol Biol Phys 2000;48: 133–137. 10. Rosen PP. Rosen’s breast pathology. 3rd ed. Philadelphia: Lippincott, Williams & Wilkins; 2008. 11. Tavassoli FA, Devilee P. Pathology and genetics of tumours of the breast and female genital organs. Lyon, France: IARC Press; 2003. 221–232. 12. Mohammed R, Martin S, Gill M, et al. Improved methods of detection of lymphovascular invasion demonstrate that it is the predominant method of vascular invasion in breast cancer
13. 14. 15. 16. 17.
18. 19.
20.
21.
22.
23.
and has important clinical consequences. Am J Surg Pathol 2007;31:1825–1833. Robbins P, Pinder S, de Klerk N, et al. Histological grading of breast carcinomas: A study of interobserver agreement. Hum Pathol 1995;26:873–879. Freedman LS. Tables of the number of patients required in clinical trials using the logrank test. Stat Med 1982;1:121–129. Schoenfeld D. The asymptotic properties of nonparametric tests for comparing survival distributions. Biometrika 1981;68: 316–319. Ceilley E, Jagsi R, Goldberg S, et al. Radiotherapy for invasive breast cancer in North America and Europe: Results of a survey. Int J Radiat Oncol Biol Phys 2005;61:365–373. Pierce S, Recht A, Lingos T, et al. Long-term radiation complications following conservative surgery (CS) and radiation therapy (RT) in patients with early stage breast cancer. Int J Radiat Oncol Biol Phys 1992;23:915–923. Lingos T, Recht A, Vicini F, et al. Radiation pneumonitis in breast cancer patients treated with conservative surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1991;21:355–360. Galper S, Recht A, Silver B, et al. Factors associated with regional nodal failure in patients with early stage breast cancer with 0-3 positive axillary nodes following irradiation alone. Int J Radiat Oncol Biol Phys 1999;45:1157–1166. Grills I, Kestin L, Goldstein N, et al. Risk factors for regional nodal failure after breast conserving therapy: Regional nodal irradiation reduces rate of axillary failure in patients with four or more positive lymph nodes. Int J Radiat Oncol Biol Phys 2003;56:658–670. Strom E, Woodward W, Katz A, et al. Clinical investigation: Regional nodal failure patterns in breast cancer patients treated with mastectomy without radiotherapy. Int J Radiat Oncol Biol Phys 2005;63:1508–1513. Truong P, Jones S, Kader H, et al. Patients with T1 to T2 breast cancer with one to three positive nodes have higher local and regional recurrence risks compared with node-negative patients after breast-conserving surgery and whole-breast radiotherapy. Int J Radiat Oncol Biol Phys 2009;73:357–364. Ejlertsen B, Jensen M, Rank F, et al. Population-based study of peritumoral lymphovascular invasion and outcome among patients with operable breast cancer. J Natl Cancer Inst 2009; 101:729–735.