Fax: (011) 81 467411654
Treatment of breast carcinoma in patients with clinically negative axillary lymph nodes using radiotherapy versus axillary dissection†
Article first published online: 8 OCT 2004
Copyright © 2004 American Cancer Society
Volume 101, Issue 10, pages 2155–2163, 15 November 2004
How to Cite
Fujimoto, N., Amemiya, A., Kondo, M., Takeda, A. and Shigematsu, N. (2004), Treatment of breast carcinoma in patients with clinically negative axillary lymph nodes using radiotherapy versus axillary dissection. Cancer, 101: 2155–2163. doi: 10.1002/cncr.20650
Presented in part at the 38th annual meeting of the American Society of Clinical Oncology, Orlando, FL, May 18–21, 2002.
- Issue published online: 29 OCT 2004
- Article first published online: 8 OCT 2004
- Manuscript Revised: 3 AUG 2004
- Manuscript Accepted: 3 AUG 2004
- Manuscript Received: 1 MAR 2004
- breast-conservation therapy;
- clinically negative axilla;
- lymphovascular invasion;
- tangential-field irradiation;
- three-field irradiation
The role of axillary lymph node dissection (AxD) for patients with breast carcinoma who have clinically negative lymph nodes (cN0) and undergo breast-conserving therapy has been controversial. If patients do not undergo AxD, then it is uncertain whether specific lymph node irradiation should be given. The authors compared the results obtained from patients w ho underwent AxD with the results from patients who received axillary irradiation (AxR) using one of two radiotherapy techniques.
Patients with T1–T2cN0 breast carcinoma were treated from 1983 to 2002 with either AxD (80 patients) or AxR (1134 patients received tangential-field [2-field] irradiation, and 303 patients received 3-field irradiation). The median follow-up was 161 months for the AxD group and 66 months for the AxR group (55 months for patients who received tangential-field irradiation, and 122 months for patients who received 3-field irradiation).
One patient in the AxD group and 35 patients in the AxR group had axillary recurrences. The 10-year cumulative axillary recurrence rates were 1.3% and 4.6% for the AxD group and the AxR group, respectively (P = 0.21). For patients with T1 tumors, the 10-year overall survival rates for the two groups were 94.7% and 92.7%, respectively (P = 0.34); and, for patients with T2 tumors, the 10-year overall survival rates were 92.5% and 89.1%, respectively (P = 0.34). In the AxR group, the 5-year axillary recurrence rates were 2.5% for patients who received tangential-field irradiation and 1.7% for patients who received 3-field irradiation (P = 0.18), and the 5-year regional recurrence rates for the two groups were 4.8% and 2.4%, respectively (P = 0.048). On multivariate analysis, positive lymphovascular invasion, outer tumor location, and larger tumor size were significant risk factors for regional failure.
For patients with cN0 breast carcinoma, AxD and AxR yielded the same overall survival rates. Most patients can be treated safely with tangential-field irradiation alone. Patients who are at increased risk of regional failure may benefit from three-field irradiation. Cancer 2004. © 2004 American Cancer Society.
Axillary lymph node dissection (AxD) provides excellent axillary control for patients with early-stage breast carcinoma, and the pathologic information gained regarding lymph node status remains the best predictor of the risk of systemic and regional recurrence. However, the role of AxD for patients with clinically negative lymph nodes (no regional lymph node metastasis; cN0) has been questioned.1–3 A recent patterns-of-care study demonstrated that AxD could be avoided for selected patients in clinical practice.4 The reported frequency of axillary lymph node metastases in cN0 patients has ranged from 26% to 38%.4 Thus, approximately 60–70% of patients with cN0 breast carcinoma may receive excessive treatment if their axillary lymph nodes are dissected. Moreover, AxD is associated with significant morbidity, including postoperative lymphedema of the involved arm, which develops in 9–37% of patients who undergo AxD.5–7
Since 1983, breast-conservation therapy has been used with increasing frequency at our institutions. By applying liberal criteria for patient selection, > 90% of breasts were preserved. In earlier years, AxD was performed routinely, regardless of clinical axillary lymph node status. We soon realized that most patients wished to avoid the morbidity caused by AxD. Therefore, we rapidly shifted our treatment policy for patients with cN0 breast disease by substituting axillary irradiation (AxR) for AxD since 1988.
In this study, we evaluated whether AxR could control cN0 disease as effectively as AxD with equal survival rates. We also evaluated whether there was any advantage of irradiating the supraclavicular fossa by adding a third field to the tangential fields used to treat the breast. Various risk factors also were assessed for axillary and regional failure in relation to the type of treatment.
MATERIALS AND METHODS
For the current study, 1810 women with T1–T2cN0M0 invasive breast carcinomas (i.e., tumors that measured ≤ 5 cm in greatest dimension [T1–T2], no regional lymph node metastases [N0], and no distant metastasis [M0]) underwent breast-conservation therapy from April, 1983, to December, 2002, at our institutions. Patients who had synchronous bilateral breast carcinomas, ductal/lobular carcinoma in situ without invasion, or a follow-up < 6 months after surgery were excluded from the analysis. In 1995, we began to perform sentinel lymph node biopsies, and these 38 patients also were excluded. Therefore, 1517 patients were eligible for the current analysis. The possible advantages and disadvantages of AxD and AxR were explained. We obtained informed consent from all patients who received AxR, because AxD had been the standard of care for the axilla.
Details of Treatment
The breast tumor was excised with a rim of healthy breast tissue to obtain clear macroscopic surgical margins. In the early years, 80 patients underwent complete AxD (Levels I–III), and those patients constitute the AxD group. Of these, 23 patients had involved axillary lymph nodes. The remaining 1437 patients were treated without AxD, and those patients constitute the AxR group.
For both groups, the whole breast was irradiated with tangential, opposing fields using a 6-MeV linear accelerator up to a total dose of 50 grays (Gy) given in 25 fractions at a rate of 5 fractions per week. The radiation dose was calculated at a depth equal to two-thirds of the distance between the skin overlying the breast and the base of the tangential fields at the point of midseparation. Boost irradiation to the tumor bed was not given. The maximal dose did not exceed 53 Gy. The posterior edges of the tangential fields were made coplanar. The cranial border of the tangential field was placed at the caudal edge of the clavicular head and was set such that the humeral head was not included in either the lateral or the medial tangential field. The entrance of the medial field was placed at the midline, and the entrance of the lateral field was placed at the axillary midline. The caudal border of the tangential field extended inferiorly below the inframammary fold to include the entire breast.
In earlier years, 303 patients were treated with a separate anterior field that was used to treat the supraclavicular fossa and axilla to a total of 50 Gy in 25 fractions prescribed to a depth of 3 cm. The lateral border of the third field extended to the axilla; the humeral head was blocked. The corner-block matching technique was used,8 with blocks placed on the tangential fields to match the anterior field, and a half-beam block was used to match the caudal edge of the anterior field and the cranial edge of the tangential fields precisely in a vertical transverse plane.
Adjuvant, systemic chemotherapy was given to 1208 patients (79.6%), with the decision to give chemotherapy based on age, menopausal status, tumor size, hormone receptor status, and nuclear grade. Chemotherapy usually began the day after surgery. Nearly all patients (n = 1200 patients) received combined cyclophosphamide, methotrexate, and 5-fluorouracil chemotherapy (cyclophosphamide, 700 mg/m2; methotrexate, 40 mg/m2; and 5-fluorouracil, 600 mg/m2, each given intravenously on Days 1 and 8 every 4 weeks for 1–6 cycles). For postmenopausal women who had tumors that overexpressed estrogen receptor, tamoxifen was given either by itself or after chemotherapy.
Histologic evaluation was performed by one of two pathologists for the entire study period; a rereview of slides for the purposes of this investigation was not performed. In addition to histologic subtype, other tumor characteristics also were recorded. Tumor size was recorded both macroscopically and microscopically. Lymphovascular invasion (LVI) was defined as the presence of tumor emboli within small, thin-walled vascular spaces in the breast tissue. This was classified further by each pathologist's judgment as follows: 0 (negative), LVI was absent; 1 (intermediate), LVI was present minimally to moderately; and 2 (positive), LVI was present heavily.
Histologic grade was determined using the Scarff–Bloom–Richardson grading system, with grade scored on a scale from 1 to 3.9 Estrogen and progesterone receptor levels were evaluated by enzyme immunoassay in the early years and, more recently, by immunohistochemical methods.
Follow-up was measured from the date of breast surgery. The median follow-up as of June 30, 2002, was 161 months (range, 44–234 months) in the AxD group and 66 months (range, 6–172 months) in the AxR group. The length of follow-up for patients who were treated with the 2-field and 3-field techniques were 55 months (range, 6–131 months) and 122 months (range, 10–171 months), respectively. Among the AxR group, 2 patients developed distant metastases at 14 months and at 68 months and were then lost to follow-up. Nine patients were censored without an event at a mean interval of 61 months (range, 13–134 months). Among the AxD group, 4 patients were lost to follow-up, all of whom were alive without disease prior to that (mean follow-up, 110 months; range, 90–128 months).
Lymph node recurrences in the supraclavicular fossa and axilla were defined as regional failures. Lymph node recurrences that occurred simultaneously with distant failure, in-breast recurrence, or a new contralateral breast carcinoma were scored as an event, but lymph node failures that occurred after these were not scored as events.
Any adenopathy that as judged suspicious by palpation and/or ultrasound was recorded as a recurrence. Although not all lymph node recurrences were confirmed by biopsy or cytology, all patients who underwent salvage AxD were identified with lymph node involvement. Salvage operations were not performed for patients who presented with concurrent distant metastases.
We used the chi-square test and the Fisher exact test for comparing characteristics between groups. Cumulative lymph node recurrence rates and overall survival rates were calculated by the Kaplan–Meier method, and statistical significance was calculated with the log-rank test. A Cox proportional-hazard analysis was used to determine the relative contribution of different factors to outcome.10 Factors with P values ≤ 0.1 in the univariate analysis were included in the Cox regression model. All statistical analyses were performed using Statistical Software.
The median patient age was slightly younger in the AxD group, which contained more premenopausal women. Other patient and tumor characteristics were similar in the two groups (Table 1). Most of the AxD and 3-field group lacked LVI and grade information because of historic comparison.
|Characteristic||AxD (%) (n = 80)||AxR (%) (n = 1437)||P value: AxD vs. AxR (tangential vs. three-field)|
|Tangential (n = 1134)||Three-field (n = 303)|
|Median||43||48||46||< 0.001 (0.17)|
|Premenopausal||70 (87.5)||732 (64.6)||194 (64.0)||< 0.001 (0.87)|
|Postmenopausal||10 (12.5)||402 (35.4)||109 (36.0)||—|
|Tumor size (cm)|
|T1N0||38 (47.5)||459 (40.4)||117 (38.5)||0.19 (0.56)|
|T2N0||42 (52.5)||675 (59.6)||186 (61.5)||—|
|Yes||55 (68.8)||902 (79.5)||251 (82.8)||0.12 (0.6)|
|No||25 (31.2)||232 (20.5)||52 (17.2)||—|
|Estrogen receptor status|
|Positive||35 (43.8)||698 (61.5)||154 (50.8)||—|
|Negative||17 (21.3)||314 (27.7)||80 (26.4)||0.87 (0.35)|
|Unknown||28 (34.9)||122 (10.8)||69 (22.8)||—|
|Progesterone receptor status|
|Positive||25 (31.3)||704 (62.1)||120 (39.6)||—|
|Negative||20 (25.0)||304 (26.8)||107 (35.3)||0.12 (< 0.001)|
|Unknown||35 (43.7)||126 (11.1)||76 (25.1)||—|
|Negative||0 (0.0)||501 (44.2)||14 (4.6)||0.07 (0.01)|
|Intermediate||5 (6.3)||532 (46.9)||39 (12.9)||—|
|Positive||1 (1.3)||64 (5.6)||4 (1.3)||—|
|Unknown||74 (92.4)||37 (3.3)||246 (81.2)||—|
|1–2||0 (0.0)||568 (50.0)||4 (1.32)||(0.75)|
|3||0 (0.0)||376 (33.2)||2 (0.68)||—|
|Unknown||80 (100)||190 (16.8)||297 (98.0)||—|
One of 80 patients (1.3%) developed an axillary recurrence in the AxD group, compared with 35 of 1437 patients (2.4%) in the AxR group. The 10-year cumulative axillary recurrence rates were 1.3% and 4.6%, respectively; the difference was not statistically significant (P = 0.21). There were 3 regional failures (3.8%) and 58 regional failures (4.0%) regional failures in the AxD group and the AxR group, respectively, and the respective 10-year actuarial rates were 2.7% and 6.9% (P = 0.30). Because the median follow-up was much longer in the AxD group than in the AxR group (161 months vs. 66 months, respectively), we reanalyzed the data by censoring the AxD group at a point at which the AxD group had the same median follow-up as the AxR group. When this was done, the 5-year cumulative risk for regional recurrence in the AxD group was equivalent to that of the AxR group (4.7% vs. 5.9% [P = 0.92], respectively). There also was an imbalance in the median length of follow-up between patients who received 2-field irradiation and patients who received 3-field irradiation (55 months and 122 months, respectively); that imbalance had no significant impact on outcome when calculated in the same manner.
For patients with T1 tumors, the 10-year overall survival rates for the AxD group and the AxR group were 94.7% and 92.7%, respectively (P = 0.34). For patients with T2 tumors, the respective rates were 92.5% and 89.1% (P = 0.34) (Fig. 1A,B).
There were 19 patients (1.7%) who developed axillary recurrences among 1134 patients who were treated with a 2-field technique, and there were 5 patients (1.7%) who developed axillary recurrences among 303 patients who were treated with a 3-field technique. The 5-year cumulative risk of axillary failure for patients who were treated with the tangential technique and patients who were treated with the 3-field technique was 2.5% and 1.7%, respectively (P = 0.18). There were 39 regional failures in the tangential-field group (3.4%) and 7 regional failures in the 3-field group (2.3%), and the 5-year actuarial risk of regional failure was 4.8% and 2.4%, respectively (P = 0.048) (Fig. 2). The 5-year overall survival rates for the 2 groups were 96.9% and 94.7%, respectively (P = 0.27). Overall survival rates for patients with T1 tumors were 97.9% and 94.0% in the tangential-field and 3-field groups, respectively; whereas the overall survival rates for patients with T2 tumors were 96.2% and 95.1%, respectively, (P = 0.29).
None of these axillary recurrences was judged inoperable technically, although 12 patients did not undergo surgery because of their preference for systemic therapy alone over other drug therapy (hormone therapy for 6 patients and intraarterial chemotherapy for 3 patients) or refusal (3 patients). Among the 24 patients who underwent therapeutic dissection, 3 patients developed rerecurred disease in the axilla.
We analyzed possible risk factors for axillary or regional (axilla plus supraclavicular) recurrence for patients in the AxR group. For all patients combined, a univariate analysis showed that tumor multicentricity, LVI positivity, and location in the outer quadrants of the breast were statistically significant risk factors for axillary failure (P = 0.05, P = 0.08 and P = 0.09, respectively). Factors that were significant for regional failure were LVI positivity (P = 0.003), tangential irradiation (P = 0.05), tumor size > 3 cm (P = 0.07), multicentricity (P = 0.08), and premenopausal status (P = 0.09) (Table 2). Variables with univariate P values ≤ 0.1 were included in a Cox regression multivariate analysis. The only significant factor for axillary failure was outer tumor location (P = 0.025); for regional failure, the factors positive LVI (P = 0.001), outer tumor location (P = 0.016), and tumor size > 3 cm (P = 0.04) were significant (Table 3). For patients who had information available on all risk factors (1096 patients in the tangential-field group and 57 patients in the 3-field group, respectively), the crude regional recurrence rate was 11.6% (11 of 95 patients) in patients with all 3 risk factors (positive LVI, tumors in the outer quadrants, and tumor > 3 cm), whereas the rate for patients without any of the 3 factors was 0% (0 of 7 patients).
|Characteristic||Crude rate (%)||Actuarial rate (%)|
|P value||Tangential (n = 1134)||Three-field (n = 303)||Tangential (n = 1134)||Three-field (n = 303)||P value|
|Five yr risk of axillary recurrence|
|Premenopausal||0.94||13/732 (1.8)||4/194 (2.1)||2.3||2.2||0.69|
|Postmenopausal||7/402 (1.7)||1/109 (0.92)||2.6||0.93||0.19|
|Tumor size (cm)|
|< 3||0.54||13/820 (1.6)||5/212 (2.4)||2.2||2.5||0.71|
|≥ 3||7/314 (2.2)||0/91 (0.0)||2.8||0||0.16|
|Solitary||0.05||15/992 (1.5)||5/291 (1.7)||2.1||1.8||0.35|
|Multicentric||5/142 (3.5)||0/12 (0.0)||3.9||0||0.89|
|In||0.09||5/471 (1.1)||2/137 (1.5)||1.4||1.5||0.45|
|Out||15/663 (2.3)||3/166 (1.8)||3||1.9||0.09|
|Negative||0.08a||4/501 (0.80)||0/14 (0.0)||1.1||0||0.90|
|Intermediate||14/532 (2.6)||2/39 (5.1)||3.2||5.2||0.98|
|Positive||2/64 (3.1)||0/4 (0.0)||4.6||0||0.71|
|Unknown||1/37 (2.7)||6/246 (2.4)||—||—||—|
|Five yr risk of regional recurrence|
|Premenopausal||0.09||20/732 (2.7)||4/194 (2.1)||3.6||2.2||0.24|
|Postmenopausal||19/402 (4.7)||3/109 (2.8)||6.9||2.8||0.09|
|Tumor size (cm)|
|< 3||0.07||23/820 (2.8)||6/212 (2.8)||3.9||2.9||0.35|
|≥ 3||16/314 (5.1)||1/91 (1.1)||6.7||1.2||0.03|
|Solitary||0.08||32/992 (3.2)||7/291 (2.4)||4.6||2.4||0.08|
|Multicentric||7/142 (4.9)||0/12 (0.0)||5.5||0||0.70|
|In||0.1||12/471 (2.5)||4/137 (2.9)||3.1||2.9||0.72|
|Out||27/663 (4.1)||4/166 (2.4)||5.3||2.3||0.01|
|Negative||0.003a||6/501 (1.2)||0/14 (0.0)||1.8||0.0||0.89|
|Intermediate||26/532 (4.9)||3/39 (7.7)||5.9||7.3||0.86|
|Positive||7/64 (10.9)||0/4 (0.0)||15||0.0||0.48|
|Unknown||1/37 (2.7)||8/246 (3.3)||—||—||—|
|Characteristic||OR||95% CI||P value|
|LVI (negative vs. positive)||2.19||0.78–4.71||0.09|
|Multicentricity (solitary vs. multicentric)||1.86||0.92–3.44||0.178|
|Location (in vs. out)||2.81||1.15–7.07||0.025|
|Menopausal status (postmenopausal vs. premenopausal)||0.58||0.33–1.02||0.06|
|Tumor size (< 3 cm vs. ≥ 3 cm)||1.80||1.01–3.21||0.04|
|LVI (negative vs. positive)||6.03||2.17–16.7||0.001|
|Location (in vs. out)||2.17||1.15–4.07||0.016|
|Multicentricity (solitary vs. multicentric)||1.77||0.85–3.69||0.13|
The use of chemotherapy did not influence either overall survival (P = 0.57) or axillary and regional control (P = 0.35 and P = 0.42, respectively) when patients with and without postoperative chemotherapy were compared. Among patients who received adjuvant chemotherapy, leukopenia < 2000 counts/mm3 was observed in 2 patients in the AxD group and in 20 patients in the AxR group, with 10 patients who developed febrile neutropenia in the latter group. Radiation-induced pneumonitis was observed in 6 patients in the AxD group and in 26 patients in the AxR group. Most of those patients were asymptomatic, and none of them required the use of steroids. Neither systemic chemotherapy nor 3-field irradiation increased the occurrence of radiation pneumonitis (P = 0.16 and P = 0.92, respectively).
One patient who was treated with AxR (3-field technique) developed arm edema with repeated episodes of cellulitis, whereas arm edema was observed in 15 patients (including 2 patients with infectious cellulitis) in the AxD group (0.07% and 19%, respectively; P < 0.0001). Three patients had mild shoulder contraction in the AxR group. Although one patient in the AxD group experienced a brachial nerve stretch injury during surgery, the symptoms gradually decreased, and arm function completely recovered. No patients in either group experienced brachial plexopathy or pathologic fracture.
The Role of AxD
In addition to the therapeutic purpose of AxD, axillary lymph node status was used as the most important criteria for deciding whether to give a patient adjuvant systemic therapy. In the late 1980s, however, selective avoidance of AxD for patients with favorable prognostic factors did not appear to sacrifice long-term outcome.11–17 For example, the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-04 trial assigned patients randomly to undergo either AxD or no axillary treatment.1 That trial showed no significant survival benefit from adding AxD to total mastectomy. It is worth noting that the axillary recurrence rates and overall outcomes in that study were very similar among patients who underwent AxD and patients who received AxR. A prospective randomized trial reported in 1992 from Institute Curie, France, initially reported (in 1992) that patients who underwent breast-conserving surgery with AxD has a superior disease-free and overall survival rate than patients who received 3-field irradiation alone without AxD,18 although the long-term results from that trial showed equivalent survival rates between the AxD group and the AxR group.19
Currently, decisions for adjuvant systemic therapy are based not only on axillary status but also on factors such as patient age, menopausal status, expression of hormone receptor, tumor size, and histologic grade.20 Therefore, for many patients with cN0 status, AxD is not necessary in the decision of whether to use adjuvant systemic therapy.
Axillary Control by Tangential Irradiation
In patients with a low risk of axillary involvement, several other studies have demonstrated low axillary failure rates with tangential irradiation.12–17, 21, 22 However, the patients in this study had an average risk of axillary involvement, because AxD results revealed occult metastasis in 29% of patients with cN0 status. Nevertheless, tangential irradiation without a separate, third field in patients with cN0 status who did not undergo AxD achieved satisfactory axillary control (5-year and 10-year cumulative risk, 2.4% and 4.7%, respectively). Although two-portal tangential irradiation is not expected to cover the entire axillary region,23–28 these low failure rates suggests that tangential irradiation, as applied in the current study, may be sufficient to achieve axillary control in this population. Another possible explanation of the low failure rates is the use of adjuvant chemotherapy. In this study, high-risk patients tended to receive chemotherapy. Because adjuvant chemotherapy decreased in-breast recurrence rates compared with irradiation alone in the NSABP B-06 trial,29, 30 it was possible that systemic chemotherapy in the current study made some contribution to the control of axillary lymph node metastasis.
Sentinel lymph node biopsy (SNB) appears to identify patients with axillary lymph node involvement reliably and with low morbidity. Since 1995, we have performed SNB in 38 selected patients for whom (in the absence of information on pathologic axillary lymph node status) we may consider adding a third field or for whom the axillary status may alter the type of systemic therapy (for example, patients who are diagnosed by core-needle biopsy with ductal carcinoma in situ with foci suspicious for an invasive component or patients who wish to preserve their fertility by avoiding systemic therapy). Chung et al. reported that standard tangential breast irradiation covered the sentinel lymph nodes in all patients they studied.31 Therefore, standard tangential irradiation could be validated as a minimal essential therapy for postoperative radiation therapy in low-risk patients.
Regional Control with Three-Field Irradiation
Three-field irradiation also has been used instead of AxD for patients with cN0 status.19, 32 In our series, although 3-field irradiation decreased the risk of regional recurrence compared with tangential irradiation only, the difference was small (4.8% vs. 2.4% at 5 years; P = 0.048). These results imply that even the third port may not be necessary for most patients with cN0 status. Zurrida et al. demonstrated a low rate of axillary failure in patients who did not undergo AxD and were treated either with tangential irradiation or with 3-field irradiation and had 5 years of follow-up.33 There also have been several randomized trials that compared tangential irradiation with three-field irradiation,34–36 all of which have demonstrated the equivalence of two modalities to date (although their long-term results are awaited). Takeda et al. also proposed an approach using modified tangential irradiation as an alternative to the three-field technique.37
Factors that Predict the Risk of Axillary/Regional Metastasis
Another strategy for avoiding the use of a 3-field technique for all patients with cN0 status is to identify subgroups of patients who are at high risk of lymph node failure if they are treated with tangential fields only. Various authors recently reported the possibility of predicting the risk of axillary lymph node involvement on the basis of tumor characteristics, such as tumor size, LVI, or histologic grade.38–48 For example, Wong et al. suggested that the presence of LVI is predictive of having ≥ 4 positive axillary lymph nodes for patients with clinical T1–T2 breast carcinoma and that patients with LVI-negative, early-stage breast carcinoma may be reasonable candidates for treatment with tangential radiation fields in the absence of AxD.49 Our results are almost consistent with those reports (Table 3). Patients with LVI-positive tumors, patients with tumors located in the outer quadrants, and patients with large tumors may require more intensive regional therapy. For example, a multivariate analysis of the current series indicated that an outer tumor location increased the risk of both axillary and regional failure significantly. When such patients with tumors in the outer quadrants were compared with the axillary irradiation techniques, three-field irradiation showed equivalent axillary control and better regional control compared with tangential irradiation. However, for patients with an inner tumor location, there were no significant differences between the two radiotherapy techniques, suggesting that this may be due to both higher risk of lymph node involvement due to closer location and inappropriate coverage with irradiation fields for such patients (Table 2). Further study will be necessary to confirm whether a separate, third field can achieve significantly better outcomes, especially for those high-risk patients.
For patients with early-stage, operable breast carcinoma who have cN0 disease, tangential irradiation appears to result in satisfactory regional control without sacrificing survival, and morbidity is acceptably low. Further analysis will be necessary to identify subgroups (such as patients who have tumors with LVI) that may benefit from three-field irradiation. If the axilla is negative clinically and if information from AxD or SNB on pathologic lymph node status will not affect the decision between adjuvant chemotherapy or hormone therapy, then tangential radiotherapy (which has negligible morbidity) is a reasonable alternative to AxD (which entails a life-long risk of arm edema).
- 10Regression models and life-tables. J R Stat Soc B. 1972; 4: 187–220..
- 23The relationship between the upper border of the tangential radiation portal for breast conserving therapy and the irradiation level of axillary lymph nodes [abstract]. Int J Radiat Oncol Biol Phys. 2002; 54(Suppl 2): 231., , , et al.
- 28Analysis of axillary coverage during tangential radiation therapy to the breast. Int J Radiat Oncol Biol Phys. 2001; 51(Suppl 1): 1–7., , , et al.
- 34Randomized clinical trial on the role of axillary radiotherapy in breast conservative management without axillary dissection for lesions < 1 cm [abstract]. Int J Radiat Oncol Biol Phys. 1998; 42(Suppl 1): 350., , , et al.
- 35Study on the role of quadrantectomy alone [abstract]. Radiother Oncol. 1998; 48(Suppl 1): S119., , , et al.