- Email: [email protected]
Influence of radiation dose on positive surgical margins in women undergoing breast conservation therapy
Int. J. Radiation Oncology Biol. Phys., Vol. 53, No. 3, pp. 680 – 686, 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/02/$–see front matter
PII S0360-3016(02)02761-X
CLINICAL INVESTIGATION
Breast
INFLUENCE OF RADIATION DOSE ON POSITIVE SURGICAL MARGINS IN WOMEN UNDERGOING BREAST CONSERVATION THERAPY STEVEN J. DIBIASE, M.D.,* LYDIA T. KOMARNICKY, M.D.,† DWIGHT E. HERON, M.D.,‡ GORDON F. SCHWARTZ, M.D.,§ AND CARL M. MANSFIELD, M.D.* *Department of Radiation Oncology, University of Maryland, Baltimore, MD; Departments of †Radiation Oncology and §Surgery, Thomas Jefferson University Hospital, Philadelphia, PA; ‡Department of Radiation Oncology, University of Pittsburgh, Pittsburgh, PA Purpose: Positive surgical margins adversely influence local tumor control in breast conservation therapy (BCT). However, reports have conflicted regarding whether an increased radiation dose can overcome this poor prognostic factor. In this study, we evaluated the influence of an increased radiation dose on tumor control in women with positive surgical margins undergoing BCT. Methods and Materials: Between 1978 and 1994, 733 women with pathologic Stage I–II breast cancer and known surgical margin status were treated at Thomas Jefferson University Hospital with BCT. Of these 733 patients, 641 women had a minimal tumor bed dose of 60 Gy and had documentation of their margin status; 509 had negative surgical margins, and 132 had positive surgical margins before definitive radiotherapy. Complete gross excision of the tumor and axillary lymph node sampling was obtained in all patients. The median radiation dose to the primary site was 65.0 Gy (range 60 –76). Of the women with positive margins (n ⴝ 132), the influence of higher doses of radiotherapy was evaluated. The median follow-up time was 52 months. Results: The local tumor control rate for patients with negative margins at 5 and 10 years was 94% and 88%, respectively, compared with 85% and 67%, respectively, for those women with positive margins (p ⴝ 0.001). The disease-free survival rate for the negative margin group at 5 and 10 years was 91% and 82%, respectively, compared with 76% and 71%, respectively, for the positive margin group (p ⴝ 0.001). The overall survival rate of women with negative margins at 5 and 10 years was 95% and 90%, respectively. By comparison, for women with positive surgical margins, the overall survival rate at 5 and 10 years was 86% and 79%, respectively (p ⴝ 0.008). A comparison of the positive and negative margin groups revealed that an increased radiation dose (whether entered as a dichotomous or a continuous variable) >65.0 Gy did not improve local tumor control (p ⴝ 0.776). On Cox multivariate analysis, margin status and menopausal status had prognostic significance for local tumor control and DFS. Conclusion: Patients with positive surgical margins have a higher risk of local tumor recurrence and worse survival when undergoing BCT. Higher doses of radiation are unable to provide an adequate level of local control in patients with positive margins. © 2002 Elsevier Science Inc. Positive margins, Radiation dose, Breast cancer.
INTRODUCTION In recent years, breast conservation therapy (BCT) has had an increasing role in the management of early-stage breast cancer (1– 6). At least 6 randomized trials have shown the equivalence of breast-conserving surgery and radiotherapy (RT) to mastectomy in the management of early-stage breast cancer (7). As the recommendation for BCT has increased in recent years, selection criteria and prognostic factors have been developed to identify appropriate women for this therapy (4, 8, 9). Patient and treatment risk factors have been identified that influence the outcome in BCT. A previous report from our institution revealed that positive
surgical margins significantly influenced local tumor control in women undergoing BCT (10, 11). With increased emphasis on BCT, some physicians contend that adverse prognostic factors can be overcome with more aggressive therapy (10, 11). After initial lumpectomy, many women are found to have positive surgical margins of resection (12, 13). Typically, an additional surgical procedure is performed to remove any residual tumor in the resection cavity (14). The need for repeated excision has been advocated in numerous reports that have documented the inferior outcome in women who undergo BCT in this setting. However, some physicians do not aggressively recommend repeated surgical excision in the setting of micro-
Reprint requests to: Steven J. DiBiase, M.D., Department of Radiation Oncology, University of Maryland Medical Center, 22 S. Greene St., Baltimore, MD 21201. Tel: (410) 328-2318; Fax: (410) 328-5279; E-mail: [email protected]
Presented at the 82nd American Radium Society meeting, London, England, April 1, 2000. Received Jul 31, 2001, and in revised form Jan 14, 2002. Accepted for publication Jan 29, 2002. 680
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
681
Table 1. Patient and tumor characteristics Characteristic Stage I II Nodes (⫹) Negative NI Estrogen receptor (⫹) Positive Negative Histologic finding Invasive ductal Other Age (y) ⬎55 ⱕ55 Menopausal status Unknown Pre Post Perimenopausal Chemotherapy Yes No Boost technique 192 Ir implant Electrons Total dose (Gy) ⬎65.0 ⱕ65.0
Negative margins (%)
Positive margins (%)
p*
331 (65) 178 (35)
71 (54) 61 (46)
0.017
460 (93) 32 (7)
115 (91) 11 (9)
0.381
267 (71) 111 (29)
86 (79) 22 (21)
0.119
451 (88) 58 (12)
116 (88) 16 (12)
0.816
265 (52) 244 (48)
61 (46) 71 (54)
0.284
13 (3) 144 (28) 325 (64) 27 (5)
3 (2) 38 (29) 86 (66) 5 (5)
0.910
106 (21) 403 (79)
37 (28) 95 (72)
0.063
300 (59) 209 (41)
75 (58) 57 (42)
0.870
22 (4) 487 (96)
27 (20) 105 (80)
0.020
* p indicates the comparison between the negative and positive margin groups with regard to patient and tumor characteristics.
scopically positive or close surgical margins (15, 16). In such a setting, higher doses of radiation are frequently delivered in an attempt to sterilize the treated breast to overcome the increased risk of local tumor recurrence (15, 17). However, there is a paucity of evidence in the literature as to whether higher doses of radiation are effective in lowering the risk of local tumor recurrence in patients with positive surgical margins after BCT. To clarify this issue of radiation dose and margin status, we retrospectively evaluated our experience. METHODS AND MATERIALS Patient selection Between 1978 and 1994, 733 women with American Joint Committee on Cancer Stage I–II breast carcinoma were treated with BCT at Thomas Jefferson University Hospital. Of the 733 women, 641 women had a minimal tumor bed dose of 60 Gy and had documentation of their surgical margin status after their lumpectomy; they are the focus of this report. Patient and tumor characteristics are described in Table 1. The median follow-up time for the 641 evaluated women was 52 months (range 2–154). The median follow-up time for the women with negative surgical margins was 53 months (range 2–154); the median fol-
low-up time for the group with positive margins was 50 months (range 4 –139). Surgery The details of the breast-conserving surgery have been described previously (10, 18). All breast-conservation women underwent shaved margin sampling. Patients underwent an initial, limited incisional biopsy of the primary tumor to obtain a tissue diagnosis. No attempt was made to remove the entire tumor at that time. Once a diagnosis of malignancy was obtained, a second surgical procedure was performed on a subsequent day to remove the remaining primary tumor. After the surgeon had excised the primary tumor site sufficiently, a tumor cavity remained. Rather than ink the excised tissue to assess for tumor presence, arcs of additional tissue following the contour of the cavity were excised by shaving the edges of the cavity as though one were “peeling an onion from the inside out (18). These tumor cavity shavings were excised separately, labeled, and submitted, usually from the medial, lateral, superior, inferior, and base of the wound. The pathologist processed these 5 tissue samples separately. Any tumor involvement within the tissue submitted as a margin biopsy was considered to be a positive margin. At the time of repeated excisional lumpectomy, a level I–II axillary lymph node dissection
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 3, 2002
Table 2. 5- and 10-year local tumor control and disease-free and overall survival based on the number of positive margins Local control (%)
Disease-free time
Overall survival (%)
Margins
5y
10 y
5y
10 y
5y
10 y
Negative (n ⫽ 509) Positive (n ⫽ 132)
94 85
88 67
91 76
82 71
95 86
90 79
was performed in most women. The margin status was obtained in all 641 patients, of whom 630 patients underwent repeated excisional margin sampling. In the 11 women who underwent one surgical procedure, a negative surgical margin was defined as tumor no closer than 2 mm from the inked surface of the primary specimen. One hundred thirtytwo patients (20%) had microscopically positive margins, and the remainder had pathologically confirmed negative margins. Radiotherapy Most women received whole breast RT using opposed, tangential, wedged fields on a 6-MV linear accelerator to a dose of 45.0 Gy to the 90% isodose of the treatment plan in 1.8-Gy daily fractions. Treatment was administered 5 d/wk for 5 weeks. The primary site was boosted with either electron beam therapy (n ⫽ 213) from a linear accelerator or with an 192Ir implant (n ⫽ 428). The technique of electron boosting has been described previously by Regine et al. (19), and Mansfield et al. (11) previously described the implant technique. The median boost dose was 20.0 Gy (range 15–31). During the period of this study, the boost dose varied according to the margin status, with higher boost doses delivered in the setting of positive surgical margins. Table 3 depicts the margin status and dose distribution of the study population. Chemotherapy Chemotherapy treatment policies varied during the 16 years of the study with regard to sequencing, number of cycles, and agents administered. Chemotherapy was delivered to 25% of the patients (160 of the 641 evaluated patients). Most patients with positive lymph nodes received systemic therapy, usually consisting of cyclophosphamide, methotrexate, and 5-fluorouracil, for 6 – 8 cycles. Seventeen percent of the patients (n ⫽ 109) received tamoxifen. The use of adjuvant chemotherapy was not different between the margin-positive and margin-negative patients. Follow-up All patients were followed by a radiation oncologist or surgeon at 3-month intervals for the first 2 years, every 6 months for 3 years, and yearly thereafter. The first mammogram was usually obtained at 6 months after RT completion, and subsequent bilateral mammograms were obtained yearly. Blood work and a chest X-ray were obtained
yearly. Any suspicion of recurrent tumor from mammographic or clinical examination required histologic verification. A local recurrence was defined as a malignant lesion occurring in the same quadrant as the original primary tumor and consisting of the same histologic subtype. All other breast lesions were described as new primary breast cancer. Statistical analysis Survival curves were determined using the Kaplan–Meier method with the period beginning at the onset of definitive RT. The log–rank test was used to compare survival curves. The Cox proportional hazards model was used for univariate and multivariate analyses. Local tumor control was defined as no evidence of disease in the whole breast. For analysis of disease-free survival (DFS), patients were required to be alive and without evidence of disease at the last follow-up examination. In the multivariate analysis, we first used the score method to evaluate the possible models. Then, among the best 5 models (according to score criteria) with 5 factors, we kept a factor if it was significant at p ⬍0.05 or if it improved the precision of the other estimates. Risk ratios for the significant factors were calculated, and the 95% confidence interval of the risk ratio is given in the parentheses in Table 2. It is strongly recommended to use the risk ratio together with its confidence interval for reporting these results. RESULTS The influence of surgical margin status on local tumor control and DFS is shown in Table 2. Women with negative margins had a 5- and 10-year actuarial local tumor control rate of 94% and 88%, respectively. However, women with positive margins had a statistically inferior 5- and 10-year local tumor control rate of 85% and 67%, respectively (Fig. 1, p ⫽ 0.001). Women with negative surgical margins had a 5- and 10-year DFS rate of 91% and 82%, respectively. In contrast, the DFS rate of women with positive surgical margins at 5 and 10 years was 76% and 71%, respectively, statistically significant as shown in Fig. 2 (p ⫽ 0.001). Furthermore, the overall survival rate of women with negative margins at 5 and 10 years was 95% and 90%, respectively. By comparison, in women with positive surgical margins, the overall survival rate at 5 and 10 years was 86% and 79%, respectively (p ⫽ 0.008; Fig. 3).
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
683
Fig. 1. Time to local tumor recurrence by surgical margin status in the 641 women treated with BCT.
The influence of radiation dose was evaluated in terms of local tumor control (Table 3). Women were stratified into a low-dose group (n ⫽ 592), receiving a total tumor dose of 60 – 65 Gy, and a high-dose group (n ⫽ 49), receiving 65–76 Gy. In the 105 women who had positive surgical margins and received a low dose, 15 women had recurrence. Similarly, in the 27 women with positive surgical margins who received a high dose, 3 women had recurrence. In the 3 women who developed local tumor recurrence in this group, 1 received 66
Gy and 2 received 70 Gy. The dose distribution for all 27 women who received ⬎65 Gy with positive margins was as follows: 65.5– 66.0 Gy (n ⫽ 1), 66 – 67 Gy (n ⫽ 11), 70 –71 Gy (n ⫽ 14), and 76 Gy (n ⫽ 1). The 5-year local tumor control rate in women with negative surgical margins who received a low dose vs. a high dose was 94% and 100%, respectively (p ⫽ 0.343). For women with positive surgical margins, the 5-year local tumor control rate for the low-dose vs. high-dose groups was 86% and 87%, respectively (p ⫽
Fig. 2. DFS by surgical margin status in the 641 women treated with BCT.
684
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 3, 2002
Fig. 3. Overall survival by surgical margin status in the 641 women treated with BCT.
0.667). Figure 4 illustrates the influence of dose on local control in women with different margin status in this series. When evaluating the positive margin group (n ⫽ 132) compared with the negative margin group (n ⫽ 509), an increased radiation dose (whether entered as a dichotomous or a continuous variable) ⬎65.0 Gy did not improve local tumor control (p ⫽ 0.776). Table 4 shows the results of Cox multivariate analysis. In terms of the time to local tumor recurrence, univariate analysis (Kaplan–Meier curve and log–rank test) showed that margin status (p ⫽ 0.0003), premenopausal status (p ⫽ 0.008), and chemotherapy (p ⫽ 0.0001) were significant factors for recurrence. Multivariate analysis using the Cox proportional hazard model showed that only margin status (p ⫽ 0.0001) and menopausal status (p ⫽ 0.0079) remained significant, with a risk ratio equal to 6.21 and 0.303, respectively. As for DFS time, univariate analysis showed that stage (p ⫽ 0.0001), margin status (p ⫽ 0.0001), premenopausal status (p ⫽ 0.001), and chemotherapy (p ⫽ 0.0004) were significant. Multivariate analysis showed that stage (p ⫽ 0.0001) and margin status (p ⫽ 0.0092) were significant factors for recurrence, with a risk ratio of 3.668 and 1.934, respectively. Table 3. Margin status and dose: 5-year local tumor control rates Margin status Negative Positive
Low dose*
High dose†
94% (n ⫽ 487) 86% (n ⫽ 105)
100% (n ⫽ 22) 87% (n ⫽ 27)
* Low dose ⫽ 60 – 65 Gy. † High dose ⫽ ⬎65 Gy.
DISCUSSION The results of this report have demonstrated that higher radiation doses ⬎65 Gy are ineffective at overcoming the adverse influence of positive surgical margins on local tumor control and DFS in women with invasive breast cancer. We have previously shown that positive surgical margins were an independent risk factor for tumor control (10). Twenty percent of the women in this series presented with positive surgical margins to their radiation oncologist. During this time (before 1994), the significant influence of margin status was not well understood. Women were commonly offered a marginally higher boost dose of radiation in an effort to overcome the higher recurrence rate due to the residual disease. However, we could not show a RT dose–response effect in the positive margin group that received higher doses of RT, at least at the doses delivered in this series ⱕ76 Gy. In the early, randomized trials conducted evaluating BCT to mastectomy, whole breast RT was delivered without a boost treatment (4). Because higher doses of whole breast RT have been shown to adversely influence the cosmetic outcome, the incorporation of a breast boost allowed higher doses of RT to be delivered to the surgical site with acceptable morbidity (20 –22). Recently, completed randomized trials have demonstrated improved tumor control in women receiving a boost treatment after conservative surgery and negative surgical margins (23, 24). However, such trials did not address women with positive surgical margins after surgery. It is logical to assume that if an advantage in local tumor control were obtained in women with negative margins, this benefit would also exist in women with known residual disease. However, whether an escalated dose of RT over standard treatment could overcome a higher burden of disease is unknown.
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
685
Fig. 4. Time to local tumor recurrence by radiation dose in the 641 women treated with BCT.
Prospective trials evaluating the influence of radiation dose on positive surgical margins have not been completed to date. Of the limited retrospective information published, the results have offered conflicting views on this issue. In the series by Cowen et al. (25), although 64% of the patients with positive margins received a dose ⱖ70 Gy, no dose effect was observed on local tumor control. Similarly, our series did not show improved local tumor control in the 27 women with doses ⬎65 Gy. However, others have suggested that sufficient doses of radiation can control residual disease after conservative surgery. Smitt et al. (16) showed a dose–response effect for doses ⱖ66 Gy. However, their definition of non-negative margins included positive, as well as close and intermediate margins (16). Of note, although Wazer et al. (26) noted a low recurrence rate for higher doses of radiation in close and intermediate margins
cases, higher doses were ineffective in positive margin patients. Solin et al. (15) provide the strongest evidence in support of a dose–response effect in women with positive surgical margins. In that series, higher doses of RT revealed a 5-year local failure rate of 0% in women with positive margins who did not receive chemotherapy or hormonal therapy. However, in their series, the largest group (n ⫽ 346) had unknown margins and the worse DFS, suggesting a selection bias (15). Our series also showed that margin status significantly influenced both DFS and overall survival. Figure 2 illustrates this point by a clear difference in the DFS curves based on the surgical margin status, which directly influences local tumor control. Although only 9% of the margin positive group was node negative, the significant separation in the DFS curves (Fig. 2) highlights the adverse influence of local residual dis-
Table 4. Relationship of patient/tumor characteristics to local tumor control and disease-free survival Local control Characteristic
Univariate
Stage Margin status (⫹) Nodes (⫹) Estrogen receptor (⫹) Histologic finding Premenopausal Chemotherapy Implant boost Electron boost Total dose†
0.6112 0.0003 0.9306 0.1907 0.0883 0.0084 0.0001 0.8329 0.3145 0.6629
Multivariate* 0.0001 (RR ⫽ 6.210)
0.0079 (RR ⫽ 0.303)
* Multivariate model is built based on backward model selection. Total dose is sum of whole breast dose and boost dose.
†
Disease-free time Univariate
Multivariate*
0.0001 0.0001 0.0621 0.3882 0.5038 0.0001 0.0004 0.8954 0.1006 0.6881
0.0001 (RR ⫽ 3.668) 0.0092 (RR ⫽ 1.934)
686
I. J. Radiation Oncology
● Biology ● Physics
ease. Additionally, multivariate analysis revealed that margin status significantly influenced a women’s chance to remain disease-free overall. A recent series by Cowen et al. (25) noted that the hazard of relapsing from metastases was 2.5 times higher after local recurrence, supporting the hypothesis that local and distant recurrences are tightly connected. Similarly, our data seem to be convincing that the presence of a positive surgical margin has an adverse effect on the survival of our patients with breast cancer. As illustrated in Fig. 3, margin status significantly influenced the overall survival of women in this series.
Volume 53, Number 3, 2002
The current series provides additional evidence that higher doses of RT are unable to provide an adequate level of local tumor control in patients with positive margins. To better answer this question, a large, randomized, Phase III trial is underway in Europe that randomizes women with positive surgical margins to a higher dose of RT (76 Gy) vs. a lower dose (24). Until the results of that trial are available, we currently recommend that all women with positive surgical margins undergo a repeat lumpectomy to obtain free margins. Maximizing local tumor control remains a central theme in BCT for women with invasive breast cancer.
REFERENCES 1. Fisher B, Anderson S, Redmond CK, Wolmark N, Wickerham DL, Cronin WM. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995;333:1456 – 1461. 2. Blichert-Toft M, Rose C, Andersen JA, et al. Danish randomized trial comparing breast conservation therapy with mastectomy: Six years of life-table analysis. In: Consensus development conference on the treatment of early-stage breast cancer. Journal of the National Cancer Institute monographs. No. 11. Washington, DC: Government Printing Office; 1992:19 –25. (NIH publication no. 90-3187.) 3. Canadian Association of Radiation Oncologists. Mastectomy or lumpectomy? The choice of operation for clinical stages I and II breast cancer. The Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer. CMAJ 1998;158(Suppl. 3):S15–S21. 4. Fowble B. Radiotherapeutic considerations in the treatment of primary breast cancer. J Natl Cancer Inst Monogr 1992;11: 49 –58. 5. Veronesi U, Zurrida S. Optimal surgical treatment of breast cancer. Oncologist 1996;1:340 –346. 6. Whelan TJ, Lada BM, Laukkanen E, et al., for the Provincial Breast Disease Site Group. Breast irradiation in women with early stage invasive breast cancer following breast conservation surgery. Cancer Prev Control 1997;1:228 –240. 7. Fowble B. Ipsilateral breast tumor recurrence following breast-conserving surgery for early-stage invasive cancer. Acta Oncol 1999;38(Suppl. 13):9 –17. 8. Heimann R, Powers C, Halpem HJ, et al. Breast preservation in stage I and II carcinoma of the breast: The University of Chicago experience. Cancer 1996;78:1722–1730. 9. Osteen RT. Selection of patients for breast conserving surgery. Cancer 1994;74(Suppl.):366 –371. 10. DiBiase SJ, Komarnicky LT, Schwartz GF, et al. The number of positive margins influences the outcome of women treated with breast preservation for early stage breast carcinoma. Cancer 1998;82:2212–2220. 11. Mansfield CM, Komarnicky LT, Schwartz GF, et al. Ten-year results in 1070 patients with stages I and II breast cancer treated by conservative surgery and radiation therapy. Cancer 1995;75:2328 –2336. 12. Gage I, Schnitt SJ, Nixon AJ, et al. Pathologic margin involvement, and the risk of recurrence in patients treated with breast-conserving therapy. Cancer 1996;78:1921–1928. 13. Ghossein NA, Alpert S, Barba J, et al. Breast cancer: Importance of adequate surgical excision prior to radiotherapy in the
14. 15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
local control of breast cancer in patients treated conservatively. Arch Surg 1992;127:411– 415. Kearney TJ, Morrow M. Effect of reexcision on the success of breast-conserving surgery. Ann Surg Oncol 1995;2:303–307. Solin LJ, Fowble BL, Schultz DJ, et al. The significance of the pathology margins of the tumor excision on the outcome of patients treated with definitive irradiation for early stage breast cancer. Int J Radiat Oncol Biol Phys 1991;21:279 –287. Smitt MC, Nowels KW, Zdeblick MJ, et al. The importance of the lumpectomy surgical margin status in long-term results of breast conservation. Cancer 1995;76:259 –267. Schmidt-Ullrich R, Wazer DE, Tercilla O, et al. Tumor margin assessment as a guide to optimal conservation surgery and irradiation in early stage breast carcinoma. Int J Radiat Oncol Biol Phys 1989;17:733–738. Schwartz GF, Lagios MD, Carter D, et al. Consensus conference on the classification of ductal carcinoma in situ: Proceedings of the Consensus Conference on the Classification of Ductal Carcinoma In Situ. Cancer 1997;80:1798 –1802. Regine WF, Ayyangar KM, Komarnicky LT, et al. Computer-CT planning of the electron boost in definitive breast irradiation. Int J Radiat Oncol Biol Phys 1985;20:177–178. Ghossein NA, Vilcoq J, Stacey P, et al. Is it necessary to irradiate the breast after conservative surgery for localized cancer? Arch Surg 1987;122:913–917. Neff PT, Bear HD, Pierce CV, et al. Long-term results of breast conservation therapy for breast cancer. Ann Surg 1996; 223:709 –717. Noguchi M, Kinne DW, Miyazaki I. Breast-conserving treatment: Controversies and consensus. J Surg Oncol 1996;62: 228 –234. Romestaing P, Lehingue Y, Carrie C, et al. Role of a 10-Gy boost in the conservative treatment of early breast cancer: Results of a randomized clinical trial in Lyon, France. J Clin Oncol 1997;15:963–968. Bartelink H, Collette L, Fourquet A, et al. Impact of a boost dose of 16 Gy on the local control and cosmesis in patients with early breast cancer: The EORTC “boost versus no boost” trial. Int J Radiat Oncol Biol Phys 2000;48(Suppl.):111. Cowen D, Houvenaeghel G, Bardou V, et al. Local and distant failures after limited surgery with positive margins and radiotherapy for node-negative breast cancer. Int J Radiat Oncol Biol Phys 2000;47:305–312. Wazer DE, Schmidt-Ullrich RK, Ruthazer R, et al. Factors determining outcome for breast-conserving irradiation with margin-directed dose escalation to the tumor bed. Int J Radiat Oncol Biol Phys 1998;40:851– 858.
PII S0360-3016(02)02761-X
CLINICAL INVESTIGATION
Breast
INFLUENCE OF RADIATION DOSE ON POSITIVE SURGICAL MARGINS IN WOMEN UNDERGOING BREAST CONSERVATION THERAPY STEVEN J. DIBIASE, M.D.,* LYDIA T. KOMARNICKY, M.D.,† DWIGHT E. HERON, M.D.,‡ GORDON F. SCHWARTZ, M.D.,§ AND CARL M. MANSFIELD, M.D.* *Department of Radiation Oncology, University of Maryland, Baltimore, MD; Departments of †Radiation Oncology and §Surgery, Thomas Jefferson University Hospital, Philadelphia, PA; ‡Department of Radiation Oncology, University of Pittsburgh, Pittsburgh, PA Purpose: Positive surgical margins adversely influence local tumor control in breast conservation therapy (BCT). However, reports have conflicted regarding whether an increased radiation dose can overcome this poor prognostic factor. In this study, we evaluated the influence of an increased radiation dose on tumor control in women with positive surgical margins undergoing BCT. Methods and Materials: Between 1978 and 1994, 733 women with pathologic Stage I–II breast cancer and known surgical margin status were treated at Thomas Jefferson University Hospital with BCT. Of these 733 patients, 641 women had a minimal tumor bed dose of 60 Gy and had documentation of their margin status; 509 had negative surgical margins, and 132 had positive surgical margins before definitive radiotherapy. Complete gross excision of the tumor and axillary lymph node sampling was obtained in all patients. The median radiation dose to the primary site was 65.0 Gy (range 60 –76). Of the women with positive margins (n ⴝ 132), the influence of higher doses of radiotherapy was evaluated. The median follow-up time was 52 months. Results: The local tumor control rate for patients with negative margins at 5 and 10 years was 94% and 88%, respectively, compared with 85% and 67%, respectively, for those women with positive margins (p ⴝ 0.001). The disease-free survival rate for the negative margin group at 5 and 10 years was 91% and 82%, respectively, compared with 76% and 71%, respectively, for the positive margin group (p ⴝ 0.001). The overall survival rate of women with negative margins at 5 and 10 years was 95% and 90%, respectively. By comparison, for women with positive surgical margins, the overall survival rate at 5 and 10 years was 86% and 79%, respectively (p ⴝ 0.008). A comparison of the positive and negative margin groups revealed that an increased radiation dose (whether entered as a dichotomous or a continuous variable) >65.0 Gy did not improve local tumor control (p ⴝ 0.776). On Cox multivariate analysis, margin status and menopausal status had prognostic significance for local tumor control and DFS. Conclusion: Patients with positive surgical margins have a higher risk of local tumor recurrence and worse survival when undergoing BCT. Higher doses of radiation are unable to provide an adequate level of local control in patients with positive margins. © 2002 Elsevier Science Inc. Positive margins, Radiation dose, Breast cancer.
INTRODUCTION In recent years, breast conservation therapy (BCT) has had an increasing role in the management of early-stage breast cancer (1– 6). At least 6 randomized trials have shown the equivalence of breast-conserving surgery and radiotherapy (RT) to mastectomy in the management of early-stage breast cancer (7). As the recommendation for BCT has increased in recent years, selection criteria and prognostic factors have been developed to identify appropriate women for this therapy (4, 8, 9). Patient and treatment risk factors have been identified that influence the outcome in BCT. A previous report from our institution revealed that positive
surgical margins significantly influenced local tumor control in women undergoing BCT (10, 11). With increased emphasis on BCT, some physicians contend that adverse prognostic factors can be overcome with more aggressive therapy (10, 11). After initial lumpectomy, many women are found to have positive surgical margins of resection (12, 13). Typically, an additional surgical procedure is performed to remove any residual tumor in the resection cavity (14). The need for repeated excision has been advocated in numerous reports that have documented the inferior outcome in women who undergo BCT in this setting. However, some physicians do not aggressively recommend repeated surgical excision in the setting of micro-
Reprint requests to: Steven J. DiBiase, M.D., Department of Radiation Oncology, University of Maryland Medical Center, 22 S. Greene St., Baltimore, MD 21201. Tel: (410) 328-2318; Fax: (410) 328-5279; E-mail: [email protected]
Presented at the 82nd American Radium Society meeting, London, England, April 1, 2000. Received Jul 31, 2001, and in revised form Jan 14, 2002. Accepted for publication Jan 29, 2002. 680
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
681
Table 1. Patient and tumor characteristics Characteristic Stage I II Nodes (⫹) Negative NI Estrogen receptor (⫹) Positive Negative Histologic finding Invasive ductal Other Age (y) ⬎55 ⱕ55 Menopausal status Unknown Pre Post Perimenopausal Chemotherapy Yes No Boost technique 192 Ir implant Electrons Total dose (Gy) ⬎65.0 ⱕ65.0
Negative margins (%)
Positive margins (%)
p*
331 (65) 178 (35)
71 (54) 61 (46)
0.017
460 (93) 32 (7)
115 (91) 11 (9)
0.381
267 (71) 111 (29)
86 (79) 22 (21)
0.119
451 (88) 58 (12)
116 (88) 16 (12)
0.816
265 (52) 244 (48)
61 (46) 71 (54)
0.284
13 (3) 144 (28) 325 (64) 27 (5)
3 (2) 38 (29) 86 (66) 5 (5)
0.910
106 (21) 403 (79)
37 (28) 95 (72)
0.063
300 (59) 209 (41)
75 (58) 57 (42)
0.870
22 (4) 487 (96)
27 (20) 105 (80)
0.020
* p indicates the comparison between the negative and positive margin groups with regard to patient and tumor characteristics.
scopically positive or close surgical margins (15, 16). In such a setting, higher doses of radiation are frequently delivered in an attempt to sterilize the treated breast to overcome the increased risk of local tumor recurrence (15, 17). However, there is a paucity of evidence in the literature as to whether higher doses of radiation are effective in lowering the risk of local tumor recurrence in patients with positive surgical margins after BCT. To clarify this issue of radiation dose and margin status, we retrospectively evaluated our experience. METHODS AND MATERIALS Patient selection Between 1978 and 1994, 733 women with American Joint Committee on Cancer Stage I–II breast carcinoma were treated with BCT at Thomas Jefferson University Hospital. Of the 733 women, 641 women had a minimal tumor bed dose of 60 Gy and had documentation of their surgical margin status after their lumpectomy; they are the focus of this report. Patient and tumor characteristics are described in Table 1. The median follow-up time for the 641 evaluated women was 52 months (range 2–154). The median follow-up time for the women with negative surgical margins was 53 months (range 2–154); the median fol-
low-up time for the group with positive margins was 50 months (range 4 –139). Surgery The details of the breast-conserving surgery have been described previously (10, 18). All breast-conservation women underwent shaved margin sampling. Patients underwent an initial, limited incisional biopsy of the primary tumor to obtain a tissue diagnosis. No attempt was made to remove the entire tumor at that time. Once a diagnosis of malignancy was obtained, a second surgical procedure was performed on a subsequent day to remove the remaining primary tumor. After the surgeon had excised the primary tumor site sufficiently, a tumor cavity remained. Rather than ink the excised tissue to assess for tumor presence, arcs of additional tissue following the contour of the cavity were excised by shaving the edges of the cavity as though one were “peeling an onion from the inside out (18). These tumor cavity shavings were excised separately, labeled, and submitted, usually from the medial, lateral, superior, inferior, and base of the wound. The pathologist processed these 5 tissue samples separately. Any tumor involvement within the tissue submitted as a margin biopsy was considered to be a positive margin. At the time of repeated excisional lumpectomy, a level I–II axillary lymph node dissection
682
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 3, 2002
Table 2. 5- and 10-year local tumor control and disease-free and overall survival based on the number of positive margins Local control (%)
Disease-free time
Overall survival (%)
Margins
5y
10 y
5y
10 y
5y
10 y
Negative (n ⫽ 509) Positive (n ⫽ 132)
94 85
88 67
91 76
82 71
95 86
90 79
was performed in most women. The margin status was obtained in all 641 patients, of whom 630 patients underwent repeated excisional margin sampling. In the 11 women who underwent one surgical procedure, a negative surgical margin was defined as tumor no closer than 2 mm from the inked surface of the primary specimen. One hundred thirtytwo patients (20%) had microscopically positive margins, and the remainder had pathologically confirmed negative margins. Radiotherapy Most women received whole breast RT using opposed, tangential, wedged fields on a 6-MV linear accelerator to a dose of 45.0 Gy to the 90% isodose of the treatment plan in 1.8-Gy daily fractions. Treatment was administered 5 d/wk for 5 weeks. The primary site was boosted with either electron beam therapy (n ⫽ 213) from a linear accelerator or with an 192Ir implant (n ⫽ 428). The technique of electron boosting has been described previously by Regine et al. (19), and Mansfield et al. (11) previously described the implant technique. The median boost dose was 20.0 Gy (range 15–31). During the period of this study, the boost dose varied according to the margin status, with higher boost doses delivered in the setting of positive surgical margins. Table 3 depicts the margin status and dose distribution of the study population. Chemotherapy Chemotherapy treatment policies varied during the 16 years of the study with regard to sequencing, number of cycles, and agents administered. Chemotherapy was delivered to 25% of the patients (160 of the 641 evaluated patients). Most patients with positive lymph nodes received systemic therapy, usually consisting of cyclophosphamide, methotrexate, and 5-fluorouracil, for 6 – 8 cycles. Seventeen percent of the patients (n ⫽ 109) received tamoxifen. The use of adjuvant chemotherapy was not different between the margin-positive and margin-negative patients. Follow-up All patients were followed by a radiation oncologist or surgeon at 3-month intervals for the first 2 years, every 6 months for 3 years, and yearly thereafter. The first mammogram was usually obtained at 6 months after RT completion, and subsequent bilateral mammograms were obtained yearly. Blood work and a chest X-ray were obtained
yearly. Any suspicion of recurrent tumor from mammographic or clinical examination required histologic verification. A local recurrence was defined as a malignant lesion occurring in the same quadrant as the original primary tumor and consisting of the same histologic subtype. All other breast lesions were described as new primary breast cancer. Statistical analysis Survival curves were determined using the Kaplan–Meier method with the period beginning at the onset of definitive RT. The log–rank test was used to compare survival curves. The Cox proportional hazards model was used for univariate and multivariate analyses. Local tumor control was defined as no evidence of disease in the whole breast. For analysis of disease-free survival (DFS), patients were required to be alive and without evidence of disease at the last follow-up examination. In the multivariate analysis, we first used the score method to evaluate the possible models. Then, among the best 5 models (according to score criteria) with 5 factors, we kept a factor if it was significant at p ⬍0.05 or if it improved the precision of the other estimates. Risk ratios for the significant factors were calculated, and the 95% confidence interval of the risk ratio is given in the parentheses in Table 2. It is strongly recommended to use the risk ratio together with its confidence interval for reporting these results. RESULTS The influence of surgical margin status on local tumor control and DFS is shown in Table 2. Women with negative margins had a 5- and 10-year actuarial local tumor control rate of 94% and 88%, respectively. However, women with positive margins had a statistically inferior 5- and 10-year local tumor control rate of 85% and 67%, respectively (Fig. 1, p ⫽ 0.001). Women with negative surgical margins had a 5- and 10-year DFS rate of 91% and 82%, respectively. In contrast, the DFS rate of women with positive surgical margins at 5 and 10 years was 76% and 71%, respectively, statistically significant as shown in Fig. 2 (p ⫽ 0.001). Furthermore, the overall survival rate of women with negative margins at 5 and 10 years was 95% and 90%, respectively. By comparison, in women with positive surgical margins, the overall survival rate at 5 and 10 years was 86% and 79%, respectively (p ⫽ 0.008; Fig. 3).
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
683
Fig. 1. Time to local tumor recurrence by surgical margin status in the 641 women treated with BCT.
The influence of radiation dose was evaluated in terms of local tumor control (Table 3). Women were stratified into a low-dose group (n ⫽ 592), receiving a total tumor dose of 60 – 65 Gy, and a high-dose group (n ⫽ 49), receiving 65–76 Gy. In the 105 women who had positive surgical margins and received a low dose, 15 women had recurrence. Similarly, in the 27 women with positive surgical margins who received a high dose, 3 women had recurrence. In the 3 women who developed local tumor recurrence in this group, 1 received 66
Gy and 2 received 70 Gy. The dose distribution for all 27 women who received ⬎65 Gy with positive margins was as follows: 65.5– 66.0 Gy (n ⫽ 1), 66 – 67 Gy (n ⫽ 11), 70 –71 Gy (n ⫽ 14), and 76 Gy (n ⫽ 1). The 5-year local tumor control rate in women with negative surgical margins who received a low dose vs. a high dose was 94% and 100%, respectively (p ⫽ 0.343). For women with positive surgical margins, the 5-year local tumor control rate for the low-dose vs. high-dose groups was 86% and 87%, respectively (p ⫽
Fig. 2. DFS by surgical margin status in the 641 women treated with BCT.
684
I. J. Radiation Oncology
● Biology ● Physics
Volume 53, Number 3, 2002
Fig. 3. Overall survival by surgical margin status in the 641 women treated with BCT.
0.667). Figure 4 illustrates the influence of dose on local control in women with different margin status in this series. When evaluating the positive margin group (n ⫽ 132) compared with the negative margin group (n ⫽ 509), an increased radiation dose (whether entered as a dichotomous or a continuous variable) ⬎65.0 Gy did not improve local tumor control (p ⫽ 0.776). Table 4 shows the results of Cox multivariate analysis. In terms of the time to local tumor recurrence, univariate analysis (Kaplan–Meier curve and log–rank test) showed that margin status (p ⫽ 0.0003), premenopausal status (p ⫽ 0.008), and chemotherapy (p ⫽ 0.0001) were significant factors for recurrence. Multivariate analysis using the Cox proportional hazard model showed that only margin status (p ⫽ 0.0001) and menopausal status (p ⫽ 0.0079) remained significant, with a risk ratio equal to 6.21 and 0.303, respectively. As for DFS time, univariate analysis showed that stage (p ⫽ 0.0001), margin status (p ⫽ 0.0001), premenopausal status (p ⫽ 0.001), and chemotherapy (p ⫽ 0.0004) were significant. Multivariate analysis showed that stage (p ⫽ 0.0001) and margin status (p ⫽ 0.0092) were significant factors for recurrence, with a risk ratio of 3.668 and 1.934, respectively. Table 3. Margin status and dose: 5-year local tumor control rates Margin status Negative Positive
Low dose*
High dose†
94% (n ⫽ 487) 86% (n ⫽ 105)
100% (n ⫽ 22) 87% (n ⫽ 27)
* Low dose ⫽ 60 – 65 Gy. † High dose ⫽ ⬎65 Gy.
DISCUSSION The results of this report have demonstrated that higher radiation doses ⬎65 Gy are ineffective at overcoming the adverse influence of positive surgical margins on local tumor control and DFS in women with invasive breast cancer. We have previously shown that positive surgical margins were an independent risk factor for tumor control (10). Twenty percent of the women in this series presented with positive surgical margins to their radiation oncologist. During this time (before 1994), the significant influence of margin status was not well understood. Women were commonly offered a marginally higher boost dose of radiation in an effort to overcome the higher recurrence rate due to the residual disease. However, we could not show a RT dose–response effect in the positive margin group that received higher doses of RT, at least at the doses delivered in this series ⱕ76 Gy. In the early, randomized trials conducted evaluating BCT to mastectomy, whole breast RT was delivered without a boost treatment (4). Because higher doses of whole breast RT have been shown to adversely influence the cosmetic outcome, the incorporation of a breast boost allowed higher doses of RT to be delivered to the surgical site with acceptable morbidity (20 –22). Recently, completed randomized trials have demonstrated improved tumor control in women receiving a boost treatment after conservative surgery and negative surgical margins (23, 24). However, such trials did not address women with positive surgical margins after surgery. It is logical to assume that if an advantage in local tumor control were obtained in women with negative margins, this benefit would also exist in women with known residual disease. However, whether an escalated dose of RT over standard treatment could overcome a higher burden of disease is unknown.
Radiation dose and margins in BCT
● S. J. DIBIASE et al.
685
Fig. 4. Time to local tumor recurrence by radiation dose in the 641 women treated with BCT.
Prospective trials evaluating the influence of radiation dose on positive surgical margins have not been completed to date. Of the limited retrospective information published, the results have offered conflicting views on this issue. In the series by Cowen et al. (25), although 64% of the patients with positive margins received a dose ⱖ70 Gy, no dose effect was observed on local tumor control. Similarly, our series did not show improved local tumor control in the 27 women with doses ⬎65 Gy. However, others have suggested that sufficient doses of radiation can control residual disease after conservative surgery. Smitt et al. (16) showed a dose–response effect for doses ⱖ66 Gy. However, their definition of non-negative margins included positive, as well as close and intermediate margins (16). Of note, although Wazer et al. (26) noted a low recurrence rate for higher doses of radiation in close and intermediate margins
cases, higher doses were ineffective in positive margin patients. Solin et al. (15) provide the strongest evidence in support of a dose–response effect in women with positive surgical margins. In that series, higher doses of RT revealed a 5-year local failure rate of 0% in women with positive margins who did not receive chemotherapy or hormonal therapy. However, in their series, the largest group (n ⫽ 346) had unknown margins and the worse DFS, suggesting a selection bias (15). Our series also showed that margin status significantly influenced both DFS and overall survival. Figure 2 illustrates this point by a clear difference in the DFS curves based on the surgical margin status, which directly influences local tumor control. Although only 9% of the margin positive group was node negative, the significant separation in the DFS curves (Fig. 2) highlights the adverse influence of local residual dis-
Table 4. Relationship of patient/tumor characteristics to local tumor control and disease-free survival Local control Characteristic
Univariate
Stage Margin status (⫹) Nodes (⫹) Estrogen receptor (⫹) Histologic finding Premenopausal Chemotherapy Implant boost Electron boost Total dose†
0.6112 0.0003 0.9306 0.1907 0.0883 0.0084 0.0001 0.8329 0.3145 0.6629
Multivariate* 0.0001 (RR ⫽ 6.210)
0.0079 (RR ⫽ 0.303)
* Multivariate model is built based on backward model selection. Total dose is sum of whole breast dose and boost dose.
†
Disease-free time Univariate
Multivariate*
0.0001 0.0001 0.0621 0.3882 0.5038 0.0001 0.0004 0.8954 0.1006 0.6881
0.0001 (RR ⫽ 3.668) 0.0092 (RR ⫽ 1.934)
686
I. J. Radiation Oncology
● Biology ● Physics
ease. Additionally, multivariate analysis revealed that margin status significantly influenced a women’s chance to remain disease-free overall. A recent series by Cowen et al. (25) noted that the hazard of relapsing from metastases was 2.5 times higher after local recurrence, supporting the hypothesis that local and distant recurrences are tightly connected. Similarly, our data seem to be convincing that the presence of a positive surgical margin has an adverse effect on the survival of our patients with breast cancer. As illustrated in Fig. 3, margin status significantly influenced the overall survival of women in this series.
Volume 53, Number 3, 2002
The current series provides additional evidence that higher doses of RT are unable to provide an adequate level of local tumor control in patients with positive margins. To better answer this question, a large, randomized, Phase III trial is underway in Europe that randomizes women with positive surgical margins to a higher dose of RT (76 Gy) vs. a lower dose (24). Until the results of that trial are available, we currently recommend that all women with positive surgical margins undergo a repeat lumpectomy to obtain free margins. Maximizing local tumor control remains a central theme in BCT for women with invasive breast cancer.
REFERENCES 1. Fisher B, Anderson S, Redmond CK, Wolmark N, Wickerham DL, Cronin WM. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995;333:1456 – 1461. 2. Blichert-Toft M, Rose C, Andersen JA, et al. Danish randomized trial comparing breast conservation therapy with mastectomy: Six years of life-table analysis. In: Consensus development conference on the treatment of early-stage breast cancer. Journal of the National Cancer Institute monographs. No. 11. Washington, DC: Government Printing Office; 1992:19 –25. (NIH publication no. 90-3187.) 3. Canadian Association of Radiation Oncologists. Mastectomy or lumpectomy? The choice of operation for clinical stages I and II breast cancer. The Steering Committee on Clinical Practice Guidelines for the Care and Treatment of Breast Cancer. CMAJ 1998;158(Suppl. 3):S15–S21. 4. Fowble B. Radiotherapeutic considerations in the treatment of primary breast cancer. J Natl Cancer Inst Monogr 1992;11: 49 –58. 5. Veronesi U, Zurrida S. Optimal surgical treatment of breast cancer. Oncologist 1996;1:340 –346. 6. Whelan TJ, Lada BM, Laukkanen E, et al., for the Provincial Breast Disease Site Group. Breast irradiation in women with early stage invasive breast cancer following breast conservation surgery. Cancer Prev Control 1997;1:228 –240. 7. Fowble B. Ipsilateral breast tumor recurrence following breast-conserving surgery for early-stage invasive cancer. Acta Oncol 1999;38(Suppl. 13):9 –17. 8. Heimann R, Powers C, Halpem HJ, et al. Breast preservation in stage I and II carcinoma of the breast: The University of Chicago experience. Cancer 1996;78:1722–1730. 9. Osteen RT. Selection of patients for breast conserving surgery. Cancer 1994;74(Suppl.):366 –371. 10. DiBiase SJ, Komarnicky LT, Schwartz GF, et al. The number of positive margins influences the outcome of women treated with breast preservation for early stage breast carcinoma. Cancer 1998;82:2212–2220. 11. Mansfield CM, Komarnicky LT, Schwartz GF, et al. Ten-year results in 1070 patients with stages I and II breast cancer treated by conservative surgery and radiation therapy. Cancer 1995;75:2328 –2336. 12. Gage I, Schnitt SJ, Nixon AJ, et al. Pathologic margin involvement, and the risk of recurrence in patients treated with breast-conserving therapy. Cancer 1996;78:1921–1928. 13. Ghossein NA, Alpert S, Barba J, et al. Breast cancer: Importance of adequate surgical excision prior to radiotherapy in the
14. 15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
local control of breast cancer in patients treated conservatively. Arch Surg 1992;127:411– 415. Kearney TJ, Morrow M. Effect of reexcision on the success of breast-conserving surgery. Ann Surg Oncol 1995;2:303–307. Solin LJ, Fowble BL, Schultz DJ, et al. The significance of the pathology margins of the tumor excision on the outcome of patients treated with definitive irradiation for early stage breast cancer. Int J Radiat Oncol Biol Phys 1991;21:279 –287. Smitt MC, Nowels KW, Zdeblick MJ, et al. The importance of the lumpectomy surgical margin status in long-term results of breast conservation. Cancer 1995;76:259 –267. Schmidt-Ullrich R, Wazer DE, Tercilla O, et al. Tumor margin assessment as a guide to optimal conservation surgery and irradiation in early stage breast carcinoma. Int J Radiat Oncol Biol Phys 1989;17:733–738. Schwartz GF, Lagios MD, Carter D, et al. Consensus conference on the classification of ductal carcinoma in situ: Proceedings of the Consensus Conference on the Classification of Ductal Carcinoma In Situ. Cancer 1997;80:1798 –1802. Regine WF, Ayyangar KM, Komarnicky LT, et al. Computer-CT planning of the electron boost in definitive breast irradiation. Int J Radiat Oncol Biol Phys 1985;20:177–178. Ghossein NA, Vilcoq J, Stacey P, et al. Is it necessary to irradiate the breast after conservative surgery for localized cancer? Arch Surg 1987;122:913–917. Neff PT, Bear HD, Pierce CV, et al. Long-term results of breast conservation therapy for breast cancer. Ann Surg 1996; 223:709 –717. Noguchi M, Kinne DW, Miyazaki I. Breast-conserving treatment: Controversies and consensus. J Surg Oncol 1996;62: 228 –234. Romestaing P, Lehingue Y, Carrie C, et al. Role of a 10-Gy boost in the conservative treatment of early breast cancer: Results of a randomized clinical trial in Lyon, France. J Clin Oncol 1997;15:963–968. Bartelink H, Collette L, Fourquet A, et al. Impact of a boost dose of 16 Gy on the local control and cosmesis in patients with early breast cancer: The EORTC “boost versus no boost” trial. Int J Radiat Oncol Biol Phys 2000;48(Suppl.):111. Cowen D, Houvenaeghel G, Bardou V, et al. Local and distant failures after limited surgery with positive margins and radiotherapy for node-negative breast cancer. Int J Radiat Oncol Biol Phys 2000;47:305–312. Wazer DE, Schmidt-Ullrich RK, Ruthazer R, et al. Factors determining outcome for breast-conserving irradiation with margin-directed dose escalation to the tumor bed. Int J Radiat Oncol Biol Phys 1998;40:851– 858.