Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
ARTICLES Long-Term Results of a Randomized Trial Comparing Breast-Conserving Therapy With Mastectomy: European Organization for Research and Treatment of Cancer 10801 Trial Joop A. van Dongen, Adri C. Voogd, Ian S. Fentiman, Catherine Legrand, Richard J. Sylvester, David Tong, Emmanuel van der Schueren, Peter A. Helle, Kobus van Zijl, Harry Bartelink Breast-conserving therapy (BCT) is widely accepted as an effective treatment option for patients with early-stage carcinomas of the breast (i.e., tumors 艋2 cm). The largest increase in the use of BCT was in the 1980s, initiated by the publication of three randomized studies showing similar survival rates for the patients who received radical mastectomy and those who received BCT (1–5). The three trials differed substantially with respect to their size and the disease characteristics of the patients. One trial, at Institut Gustave-Roussy, Paris, France, was relatively small, containing fewer than 100 patients in each arm, all having a tumor 2 cm or smaller (1). The Milan trial (2,3) (containing approximately 350 patients in each arm) also included only patients with tumors 2 cm or smaller. The trial of the Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 National Surgical Adjuvant Breast and Bowel Project (NSABP B-06) (4,5) (consisting of approximately 600 patients in each arm) included patients with tumors up to 4 cm. In all three trials, a microscopically complete excision was required for the patients in the BCT arm. In 1992, the first results of three subsequent trials—those of the U.S. National Cancer Institute (consisting of approximately 120 patients in each arm) (6), the European Organization for Research and Treatment of Cancer (EORTC) (containing approximately 450 patients in each arm) (7), and the Danish Breast Cancer Cooperative Group (DBCG) (containing approximately 450 patients in each arm) (8)—were published. The results of these trials provided further evidence of the efficacy of BCT for patients with tumors larger than 2 cm. To be able to detect even a moderate difference in survival and risk of local recurrence between BCT and mastectomy, data from all six trials were analyzed in a systematic overview, or meta-analysis, by the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) (9). This meta-analysis, which was published in 1995, found no differences in overall survival at 10 years. However, it had several limitations. First, data for individual patients in the large NSABP B-06 trial were not included. Second, most of the patients in the EORTC and DBCG trials had been followed for fewer than 10 years. Consequently, the 10year estimates of the EBCTCG overview are based largely on patients with tumors 2 cm or smaller, and evidence for the efficacy of BCT in patients with larger tumors after long-term follow-up remained limited. Other limitations of the meta- Affiliations of authors: J. A. van Dongen (Department of Surgery), H. Bartelink (Department of Radiotherapy), The Netherlands Cancer Institute, Amsterdam; A. C. Voogd, Comprehensive Cancer Center South, Eindhoven, The Netherlands; I. S. Fentiman (Hedley Atkins Breast Unit), D. Tong (Guy’s and St. Thomas’ Cancer Center), Guy’s Hospital, London, U.K.; C. Legrand, R. J. Sylvester, European Organization for Research and Treatment of Cancer Data Center, Brussels, Belgium; E. van der Schueren (deceased March 1998), Department of Radiotherapy, University Hospital Gasthuisberg, Leuven, Belgium; P. A. Helle, Department of Radiotherapy, Dr. Daniël den Hoed Cancer Center, Rotterdam, The Netherlands; K. van Zijl, Breast Unit, University of Stellenbosch, Tijgersberg, South Africa. Correspondence to: Harry Bartelink, M.D., Ph.D., Department of Radiotherapy, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands (e-mail: [email protected]). See “Notes” following “References.” © Oxford University Press ARTICLES 1143 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 Background: Breast-conserving therapy (BCT) has been shown to be as effective as mastectomy in the treatment of tumors 2 cm or smaller. However, evidence of its efficacy, over the long term, in patients with tumors larger than 2 cm is limited. From May 1980 to May 1986, the European Organization for Research and Treatment of Cancer carried out a randomized, multicenter trial comparing BCT with modified radical mastectomy for patients with tumors up to 5 cm. In this analysis, we investigated whether the treatments resulted in different overall survival, time to distant metastasis, or time to locoregional recurrence. Methods: Of 868 eligible breast cancer patients randomly assigned to the BCT arm or to the modified radical mastectomy arm, 80% had a tumor of 2.1–5 cm. BCT comprised lumpectomy with an attempted margin of 1 cm of healthy tissue and complete axillary clearance, followed by radiotherapy to the breast and a supplementary dose to the tumor bed. The median follow-up was 13.4 years. All P values are two-sided. Results: At 10 years, there was no difference between the two groups in overall survival (66% for the mastectomy patients and 65% for the BCT patients; P = .11) or in their distant metastasis-free rates (66% for the mastectomy patients and 61% for the BCT patients; P = .24). The rate of locoregional recurrence (occurring before or at the same time as distant metastasis) at 10 years did show a statistically significant difference (12% of the mastectomy and 20% of the BCT patients; P = .01). Conclusions: BCT and mastectomy demonstrate similar survival rates in a trial in which the great majority of the patients had stage II breast cancer. [J Natl Cancer Inst 2000;92:1143–50] analysis related to the difficulty of controlling for the differences in treatment techniques between the trials and the variations in defining local recurrence. The same limitations apply to a more recent overview (10), which came to the same conclusions as the meta-analysis of the EBCTCG. A longer term follow-up would help to resolve the question of the relative efficacy of BCT and mastectomy for breast cancer patients with large tumors. The EORTC Breast Cancer Cooperative Group trial began in 1980. The purpose of the trial was to compare BCT with mastectomy, in particular in the treatment of clinical stage II breast cancer. Consequently, patients with tumors up to 5 cm (clinical measurement) were randomly assigned to one therapy or the other. The first results were published after a median follow-up of 6 years (7,11). In this article, we report findings based on a median follow-up of 13.4 years. PATIENTS AND METHODS The EORTC study (trial 10801) was open to patient accrual from May 1980 to May 1986. A total of 902 patients with a diagnosis of clinical stage I or II invasive carcinoma of the breast (12) were enrolled in the trial. Major exclusion criteria were as follows: patient age of 71 years or older, Karnofsky index below 80% (i.e., unable to carry on normal activity or to do active work), tumor fixed to the muscles or of questionable operability for some other reason, a multicentric tumor, or a large tumor in a small breast. Also excluded were patients with a history of other malignancies (apart from basal cell carcinoma of the skin and in situ carcinoma of the cervix uteri) as well as patients who were unable to conform to strict follow-up and patients who were judged to be emotionally or psychologically unfit to undergo BCT or (optional) positive axillary apex biopsy. Patients were stratified by participating center, carcinoma stage (I or II), and menopausal status. The participating centers were as follows: Guy’s Hospital, London, U.K. (n ⳱ 420); The Netherlands Cancer Institute, Amsterdam (n ⳱ 202); University Hospital Gasthuisberg, Leuven, Belgium (n ⳱ 108); Breast Unit, University of Stellenbosch, Tijgersberg, South Africa (n ⳱ 80); Dr. Daniël den Hoed Cancer Center, Rotterdam, The Netherlands (n ⳱ 50); and three other hospitals in The Netherlands and Belgium (n ⳱ 42). Randomization was performed centrally at the EORTC Data Center in Brussels, Belgium. In the first 2 years of the trial (i.e., 1980 and 1981), a 2 : 1 randomization was performed— thus explaining why somewhat more patients were randomly assigned to BCT (466 patients were assigned to the BCT arm, whereas 436 were assigned to the mastectomy arm). Fifteen patients decided to undergo BCT after having been randomly assigned to mastectomy, and 15 decided to be treated with mastectomy after having been randomly assigned to BCT. In the statistical analysis of the data, however, these patients remained in the original treatment arm to which they had been assigned, according to the principle of intent to treat. The study forms of all patients were reviewed by one of the study coordinators to evaluate their eligibility for the trial. After review, 34 patients (16 in the mastectomy arm and 18 in the BCT arm) were excluded from the analysis. Twenty-five patients were considered to be ineligible because their tumors were more advanced or the histopathologic diagnosis was incorrect, two because of poor physical condition, one because of an incomplete examination prior to randomization, and six for unspecified reasons. Thus, 868 patients remained in the study, of whom 420 were assigned to the mastectomy group and 448 to the BCT group (Fig. 1). Of these 868 patients, 696 (80%) had a tumor of 2.1–5 cm (see Table 1). The investigations were approved by the local institutional committee for medical ethics in each of the centers participating in the study. Informed consent was obtained from all of the patients entered in the EORTC 10801 trial according to the approved guidelines of the participating centers. Treatment Patients assigned to the mastectomy arm underwent a modified radical mastectomy. Patients assigned to the BCT arm underwent a lumpectomy with an attempted margin of 1 cm of healthy tissue. Margins were not routinely inked to assess the microscopic completeness of the lumpectomy, and there were no instructions in the protocol for re-excision. In general, re-excision took place only when macroscopic (i.e., palpable) disease was left behind, which was the 1144 ARTICLES case for only a few patients. Microscopic margin involvement was observed in 217 of the 448 patients in the BCT arm. Microscopically incomplete excision was not a reason for exclusion. Lumpectomy was followed by radiotherapy directed to the breast (50 Gy over a 5-week period), with an additional booster dose of 25 Gy directed to the lumpectomy site via an iridium-192 implant. In 22 patients, implants could not be used for technical reasons; such patients were given an equivalent booster dose with external irradiation. At Guy’s Hospital, the irradiation protocol after lumpectomy was slightly different: A flexible iridium192 implant (20 Gy midplane) was placed during surgery, and external-beam radiotherapy was given at a minimum dose of 46 Gy. Furthermore, Guy’s Hospital accepted only patients whose tumors were no more than 4 cm in diameter. For patients in both treatment arms, irradiation of the parasternal lymph node region was indicated for patients with a centrally or medially localized tumor and for patients with a lateral tumor and histologically proven axillary lymph node metastases. Postoperative irradiation of the chest wall after mastectomy was indicated after a microscopically incomplete operation and was given to only 12 patients. According to the protocol of the trial, six cycles of chemotherapy with cyclophosphamide, methotrexate, and 5-fluorouracil were indicated for all patients 55 years old or younger with histologically proven axillary lymph node metastases. The dosages were as follows: cyclophosphamide, 100 mg/m2, given orally on days 1–14; methotrexate, 40 mg/m2, given intravenously on days 1 and 8; and 5-fluorouracil, 600 mg/m2, given intravenously on days 1 and 8. There were no instructions in the protocol concerning the use of hormonal therapy, but any adjuvant treatment used at one of the participating centers had to be similar for both arms and consistent throughout the trial. Indications for adjuvant chemotherapy, hormonal therapy, and adjuvant parasternal radiotherapy were consistent during the trial period and similar for both treatment arms. Some minor institutional differences in treatment techniques were not considered relevant because the randomization of the patients was stratified by center. In preparation for the data analysis, a quality-control program was carried out. This program included site visits and a review of the medical files of the patients who were reported to have died and of those who were still alive but had distant metastases or locoregional recurrence. Clinical results from patients’ follow-up visits to the centers were reported annually to the EORTC Data Center. Statistical Analysis End points were survival, time to distant metastasis, and time to locoregional recurrence. All relapses within the treated area (breast, chest wall, or axilla) were considered to be locoregional recurrences, including new tumors and those recurrences appearing before, at the same time as, or after the manifestation of distant disease. Supraclavicular recurrence was considered to be distant metastasis. Duration of survival, time to distant metastasis, and time to locoregional recurrence were measured from the date of randomization, were censored at the last follow-up date, and were estimated with the use of the product-limit method of Kaplan and Meier (13). The log-rank test was used to compare the results of BCT and mastectomy. Treatment comparisons were adjusted for the major prog- Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 Patients Fig. 1. Flow diagram for evaluation of breast-conserving therapy (BCT) versus mastectomy in a randomized clinical trial of patients with TNM (i.e., tumor– node–metastasis) clinical stage I or II breast cancer (see “Patients and Methods” section for details of staging). Incidence of recurrence and death in the two groups is shown in Table 2. nostic factors (clinical tumor size, pathologic lymph node status, and age) with the use of a stratified log-rank test and a Cox proportional hazards model (14). Using pathologic tumor size instead of the clinical size gave similar results. However, because information on pathologic tumor size was missing for 22% of the patients, we used clinical tumor size in the analysis. Moreover, clinical (and mammographic) tumor size is the only information on which to base the decision between BCT and radical mastectomy. All P values are two-sided. Locoregional recurrence and distant metastasis are generally considered to be nonindependent events. Therefore, the data were also analyzed according to the approach for competing risks and multiple failures discussed by Gooley et al. (15), to take into account patients who died or who had distant metastases before experiencing locoregional progression. RESULTS The characteristics of the treatment groups with respect to age, menopausal status, tumor stage, and adjuvant chemotherapy were similar (Table 1). Table 2 shows the incidence of events and causes of death in each group. There was no difference between the two treatment groups in overall duration of survival (P ⳱ .11), which at 10 years was 66.1% (95% confidence interval [CI] ⳱ 61.4%–70.7%) for patients assigned to mastectomy and 65.2% (95% CI ⳱ 59.7%– 70.7%) for patients assigned to BCT (Fig. 2). Adjustment for clinical tumor size (艋2 cm versus >2 cm), pathologic axillary lymph node status (negative versus positive), and age (<50 years old versus 艌50 years old) in a Cox proportional hazards model did not reveal a difference between the two treatment groups (hazard ratio [HR] ⳱ 1.13; 95% CI ⳱ 0.92–1.39) (Table 3). Distant Metastases There was no difference in time to distant metastasis between the two groups (P ⳱ .24). Distant metastasis-free survival at 10 years was 66.3% (95% CI ⳱ 61.6%–70.9%) for patients assigned to mastectomy and 60.5% (95% CI ⳱ 55.8%–65.2%) for patients assigned to BCT (Fig. 3). No statistically significant differences in the time to distant metastasis were found between the two treatment arms after adjustment for clinical tumor size, pathologic axillary lymph node status, and age (HR ⳱ 1.09; 95% CI ⳱ 0.88–1.35) (Table 3). Locoregional Recurrence In the BCT arm, 92 patients developed locoregional recurrence; in 16 of these patients, the locoregional recurrence was diagnosed after they developed distant metastasis. In the mastectomy arm, 56 patients developed locoregional recurrence; 11 of these patients were diagnosed after developing distant metastasis. Locoregional recurrences occurring after distant metastasis are considered to be clinically less important than those developing earlier or at the same time as distant metastasis, and the risk of underreporting is higher. In addition, many of the locoregional recurrences occurring after distant metastasis had not been confirmed by histologic examination. For these reasons, we did not take into account the locoregional recurrences occurring after distant metastasis in our analyses. In the BCT arm, the locoregional recurrence rate was 11.8% (95% CI ⳱ 8.5%–15.1%) at 5 years and 19.7% (95% CI ⳱ 15.4%–24.0%) at 10 years (Fig. 4). For the mastectomy arm, these estimates were 9.8% (95% CI ⳱ 6.7%–13.0%) and 12.2% (95% CI ⳱ 8.7%–15.7%), respectively (Fig. 4). A log-rank test demonstrated that the locoregional recurrence rate was signifiJournal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 Time Interval Until Locoregional Recurrence Eighty percent of the locoregional recurrences in the mastectomy arm were diagnosed within 5 years after primary treatment, compared with 59.3% in the BCT arm (Table 5). Among patients who experienced a locoregional recurrence, the median time until the diagnosis of locoregional recurrence was 3.9 years after BCT and 2.4 years after mastectomy. The overall survival for the patients with locoregional recurrence as a first or only event was similar for both study arms: 33 (63.5%) of the 52 patients with locoregional recurrence in the BCT arm died, compared with 17 (61%) of the 28 patients in the mastectomy arm. Locoregional Recurrence According to Participating Center The proportion of patients with locoregional recurrence varied considerably among the centers participating in the study. The locoregional recurrence rate ranged from 10.5% to 36.0% after BCT and from 4.6% to 21.3% after mastectomy. In all participating centers, the risk of locoregional recurrence appeared to be uniformly higher for patients in the BCT arm than for patients in the mastectomy arm, except in one, where a higher incidence of locoregional recurrence was observed in the mastectomy arm. DISCUSSION The results from the EORTC 10801 trial of mostly stage II breast cancer, now with a median follow-up of 13.4 years, showed no substantial difference in overall survival between BCT and mastectomy. Together with the long-term data from the NSABP-B06 trial, with a median follow-up of 12 years (16), and the NCI trial, with a median follow-up of 10 years (17), these new results provide evidence that BCT is a feasible option for patients with tumors larger than 2 cm. However, there was a statistically significantly higher risk of ARTICLES 1145 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 Overall Survival cantly higher for the patients assigned to BCT than for the patients assigned to mastectomy (P ⳱ .0097). When competing risk methods were used to take into account the fact that some patients will die or will experience distant metastasis before experiencing locoregional recurrence, the locoregional recurrence rate was still statistically significantly higher for BCT than for mastectomy. After adjustment for clinical tumor size, pathologic lymph node status, and patient age in a Cox model, the locoregional recurrence risk for the patients assigned to BCT was 1.64 times that for the patients assigned to mastectomy (95% CI ⳱ 1.12– 2.38) (Table 3). A higher risk of locoregional recurrence was also found for patients with a positive lymph node status compared with those with a negative lymph node status (HR ⳱ 1.71; 95% CI ⳱ 1.19–2.45) (Table 3). When considering HRs in various subgroups based on age, clinical tumor size, and pathologic lymph node status, we found that locoregional recurrence rates appeared to be higher in the BCT arm than in the mastectomy arm for patients 50 years old or older, for patients with a tumor larger than 2 cm, and for patients with a negative lymph node status (Table 4). In the BCT arm, the 10-year locoregional recurrence rate for patients with a microscopically complete excision was 17.6% (95% CI ⳱ 11.2%–22.3%). For patients with microscopic margin involvement, it was 26.5.% (95% CI ⳱ 22.2%–33.0%). Table 1. Characteristics of eligible patients Treatment Mastectomy (n ⳱ 420) Characteristic Breast-conserving therapy (n ⳱ 448) % No. of patients % Age, y <50 艌50 Unknown 156 264 0 37.1 62.9 176 271 1 39.3 60.5 0.2 Menopausal status Premenopausal Postmenopausal Artificial menopause Unknown 171 224 25 0 40.7 53.3 6.0 183 246 17 2 40.8 54.9 3.8 0.4 Tumor stage, clinical* I II Unknown 69 351 0 16.4 83.6 66 381 1 14.7 85.0 0.2 Tumor size, cm, clinical† 艋2.0 2.1–5.0 Unknown 87 333 0 20.7 79.3 83 363 2 18.5 81.0 0.4 Tumor size, cm, pathologic‡ 艋2.0 2.1–5.0 Unknown 140 180 100 33.3 42.9 23.8 114 242 92 25.4 54.0 20.5 Clinical lymph node status§ N0 N1A N1B Unknown 312 33 75 0 74.3 7.9 17.9 316 38 87 7 70.5 8.5 19.4 1.6 Pathologic lymph node status㛳 pN− pN+ Unknown 251 163 6 59.8 38.8 1.4 251 194 3 56.0 43.3 0.7 Lymph nodes examined, No. 0–9 10–19 20–29 艌30 62 168 123 67 14.8 40.0 29.3 16.0 64 180 133 71 14.3 40.2 29.7 15.8 Positive lymph nodes, No. 0 1 2–3 4–9 艌10 Unknown 251 68 38 39 18 6 59.8 16.2 9.0 9.3 4.3 1.4 251 92 52 29 21 3 56.0 20.5 11.6 6.5 4.7 0.7 Parasternal irradiation Yes No 167 253 39.8 60.2 202 246 45.1 54.9 69 44 50 42.3 27.0 30.7 92 52 50 47.4 26.8 25.8 Adjuvant systemic treatment,¶ pN+ patients Yes No Unknown *Reference (12). †Size as measured at palpation; mammographic measurement was used for patients with an unknown measurement at palpation. ‡Postsurgical size, as measured by the pathologist. §Lymph node status as determined by palpation. N0 ⳱ no palpable homolateral axillary lymph nodes; N1A ⳱ movable homolateral axillary lymph nodes, not considered to contain growth; N1B ⳱ movable homolateral axillary lymph nodes, considered to contain growth. 㛳Lymph node status as determined by histologic examination. pN− ⳱ no positive lymph nodes found; pN+ ⳱ positive lymph nodes found. ¶Chemotherapy with cyclophosphamide, methotrexate, and 5-fluorouracil or hormonal therapy with tamoxifen. locoregional recurrence (occurring in the absence of distant metastasis or before or at the same time as distant metastasis) for patients in the BCT arm than for patients in the mastectomy arm. To refine the selection of patients for either treatment, therefore, risk factors for local recurrence should be determined. Patho1146 ARTICLES logic lymph node status and, to a lesser degree, clinical tumor size appeared to be risk factors for locoregional recurrence. Locoregional recurrence rates seemed to be higher after BCT than after mastectomy in certain subgroups based on age, pathologic lymph node status, and clinical tumor size. However, the small Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 No. of patients Table 2. Incidence of recurrence and death in the two treatment arms* Mastectomy (n ⳱ 420) BCT (n ⳱ 448) Total (n ⳱ 868) No. of patients %† No. of patients %† No. of patients %† LRR only 14 3.3 27 6.0 41 4.7 Both LRR and DM LRR first Both at the same time DM first 14 17 11 3.3 4.0 2.6 25 24 16 5.6 5.4 3.6 39 41 27 4.5 4.7 3.1 DM only 112 26.7 120 26.8 232 26.7 Death Death from malignant disease Death from cardiovascular disease Death from other or unknown causes 165 136 11 18 39.3 82.4 6.7 10.9 208 170 19 19 46.4 81.7 9.1 9.1 373 306 30 37 43.0 82.0 8.0 9.9 No recurrence and no death 221 52.6 209 46.7 430 49.5 Fig. 2. Kaplan–Meier curves for overall survival in patients with a diagnosis of clinical stage I or II invasive carcinoma of the breast after breast-conserving therapy (BCT) and mastectomy. Five-year survival rates: mastectomy arm, 84.6% (95% confidence interval [CI] ⳱ 81.1%–88.2%); BCT arm, 82.3% (95% CI ⳱ 78.7%– 85.9%). Ten-year survival rates: mastectomy arm, 66.1% (95% CI ⳱ 61.4%–70.7%); BCT arm, 65.2% (95% CI ⳱ 59.7%–70.7%). Thirteen-year survival rates: mastectomy arm, 60.0% (95% CI ⳱ 55.0%–65.0%); BCT arm, 54.9% (95% CI ⳱ 50.1%–59.8%). O ⳱ number of observed deaths; N ⳱ number of patients at risk. Table 3. Cox proportional hazards model for death, distant metastasis (DM), and locoregional recurrence (LRR) before or at the same time as DM in 857* patients† Death DM LRR Hazard ratio (95% CI) P Hazard ratio (95% CI) P Hazard ratio (95% CI) P Univariate Treatment (unadjusted): BCT versus mastectomy 1.18 (0.96–1.45) .112 1.14 (0.92–1.41) .238 1.62 (1.12–2.34) .0097 Multivariate Treatment (adjusted): BCT versus mastectomy Clinical tumor size, cm, 2.1–5.0 versus 艋2.0 Pathologic lymph node status, pN+ versus pN− Age, 艌50 y versus <50 y 1.13 (0.92–1.39) 1.77 (1.29–2.41) 2.13 (1.74–2.63) 1.28 (1.03–1.59) .246 .0003 .0001 .025 1.09 (0.88–1.35) 1.73 (1.26–2.39) 1.99 (1.60–2.48) 1.06 (0.85–1.32) .451 .0008 .0001 .639 1.64 (1.12–2.38) 1.59 (0.95–2.66) 1.71 (1.19–2.45) 0.92 (0.63–1.32) .01 .081 .004 .637 *Eleven patients were excluded from the model because of missing information. †CI ⳱ confidence interval, BCT ⳱ breast-conserving therapy, pN+ ⳱ positive lymph nodes, and pN− ⳱ negative lymph nodes. number of patients with locoregional recurrence did not allow a formal analysis of risk factors for such recurrence. Although most locoregional recurrences developed in the first 5 years after treatment, they also developed later, especially in the BCT group. It has been reported that late recurrences after Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 BCT have a better prognosis than early recurrences (i.e., those occurring in the first 2 or 3 years), possibly because an increasing portion of the recurrences will be new primary tumors (18– 20). These new tumors would have a small impact on the results of BCT that would be detected only in a meta-analysis of several ARTICLES 1147 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 *BCT ⳱ breast-conserving therapy, LRR ⳱ locoregional recurrence, and DM ⳱ distant metastasis. †Percentages sum to more than 100% because some of the women with recurrence ultimately died. Fig. 3. Kaplan–Meier curves for time to distant metastasis in patients with a diagnosis of clinical stage I or II invasive carcinoma of the breast after breastconserving therapy (BCT) and mastectomy. Five-year distant metastasis-free survival rates: mastectomy arm, 73.6% (95% confidence interval [CI] ⳱ 69.3%– 77.9%); BCT arm, 70.2% (95% CI ⳱ 65.9%–74.6%). Ten-year distant metastasis-free survival rates: mastectomy arm, 66.3% (95% CI ⳱ 61.6%–70.9%); BCT arm, 60.5% (95% CI ⳱ 55.8%–65.2%). Thirteen-year distant metastasis-free survival rates: mastectomy arm, 63.9% (95% CI ⳱ 59.1%–68.8%); BCT arm, 57.6% (95% CI ⳱ 52.8%–62.4%). O ⳱ number of patients developing distant metastases; N ⳱ number of patients at risk. Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 Fig. 4. Kaplan–Meier curves for time to locoregional recurrence (as only event or before or at the same time as distant metastases) after breast-conserving therapy (BCT) and mastectomy. Five-year locoregional recurrence-free survival rates: mastectomy arm, 90.2% (95% confidence interval [CI] ⳱ 87.0%–93.4%); BCT arm, 88.2% (95% CI ⳱ 84.9%–91.5%). Ten-year locoregional recurrence-free survival rates: mastectomy arm, 87.8% (95% CI ⳱ 84.3%–91.3%); BCT arm, 80.3% (95% CI ⳱ 76.0%–84.6%). Thirteen-year locoregional recurrence-free survival rates: mastectomy arm, 86.9% (95% CI ⳱ 83.2%–90.6%); BCT arm, 78.5% (95% CI ⳱ 74.0%–83.1%). O ⳱ number of patients developing locoregional recurrence; N ⳱ number of patients at risk. thousand patients in each treatment arm with a follow-up of at least 10–20 years. Several studies (18,21,22) have tried to determine whether locoregional recurrence is a source of distant spread of disease, especially after BCT. These studies have suggested that the majority of the early locoregional recurrences after BCT are associated with the same prognostic factors as distant metastases and are alternative expressions of the same biologic process. However, these findings do not exclude the possibility of local recurrence as a source of distant spread of disease and subsequent death, as has been illustrated in studies of the role of adjuvant radiotherapy in lymph node-positive premenopausal women (23,24). In an analysis of more than 1000 patients treated at The Netherlands Cancer Institute (22), it was also shown that local recurrence was an independent prognostic factor for survival. In the study reported here, the proportion of patients with locoregional recurrence as the first or only recurrence event was small 1148 ARTICLES Table 4. Hazard ratios (HRs) for locoregional recurrence after breast-conserving therapy (BCT) compared with mastectomy in subgroups based on patient age, clinical tumor size, and pathologic lymph node status HR (95% CI),* BCT versus mastectomy Age, y 艋50 >50 1.48 (0.84–2.62) 1.73 (1.06–2.80) Clinical tumor size, cm 艋2.0 2.1–5.0 1.30 (0.51–3.29) 1.68 (1.12–2.51) Pathologic lymph node status† pN− pN+ 1.95 (1.13–3.34) 1.39 (0.83–2.34) *CI ⳱ confidence interval. †pN+ ⳱ positive lymph nodes and pN− ⳱ negative lymph nodes. Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 Table 5. Interval between primary treatment and locoregional recurrence (LRR) according to treatment group and occurrence of distant metastasis (DM) (excluding LRR after DM) Mastectomy arm LRR only LRR before DM BCT arm* LRR and DM at the same time Total LRR only LRR before DM LRR and DM at the same time Total Interval, y No. % No. % No. % No. % No. % No. % No. % No. % <2 2–5 >5 3 6 5 14 21.4 42.9 35.7 31.1 8 4 2 14 57.1 28.6 14.3 31.1 9 6 2 17 52.9 35.3 11.8 37.8 20 16 9 45 44.4 35.6 20.0 100 1 8 18 27 3.7 29.6 66.7 35.5 11 9 5 25 44.0 36.0 20.0 32.9 10 6 8 24 41.7 25.0 33.3 31.6 22 23 31 76 29.0 30.3 40.8 100 Total *BCT ⳱ breast-conserving therapy. Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 REFERENCES (1) Sarrazin D, Le M, Rouesse J, Contesso J, Petit JY, Lacour J, et al. Conservative treatment versus mastectomy in breast cancer tumors with macroscopic diameter of 20 millimeters or less. The experience of the Institut Gustave-Roussy. Cancer 1984;53:1209–13. (2) Veronesi U, Saccozzi R, Del Vecchio M, Banfi A, Clemente C, De Lena M, et al. Comparing radical mastectomy with quadrantectomy, axillary dissection, and radiotherapy in patients with small cancers of the breast. N Engl J Med 1981;305:6–11. (3) Veronesi U, Banfi A, Salvadori B, Luini A, Saccozzi R, Zucali R, et al. Breast conservation is the treatment of choice in small breast cancer: longterm results of a randomized trial. Eur J Cancer 1990;26:668–70. (4) Fisher B, Bauer M, Margolese R, Poisson R, Pilch Y, Redmond C, et al. Five-year results of a randomized clinical trial comparing total mastectomy and segmental mastectomy with or without radiation in the treatment of breast cancer. N Engl J Med 1985;312:665–73. (5) Fisher B, Redmond C, Poisson R, Margolese R, Wolmark N, Wickerham L, et al. Eight-year results of a randomized clinical trial comparing total mastectomy and lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1989;320:822–8. (6) Straus K, Lichter A, Lippman M, Danforth D, Swain S, Cowan K, et al. Results of the National Cancer Institute early breast cancer trial. J Natl Cancer Inst Monogr 1992;11:27–32. (7) van Dongen JA, Bartelink H, Fentiman IS, Lerut T, Mignolet F, Olthuis G, et al. Randomized clinical trial to assess the value of breast-conserving therapy in stage I and II breast cancer, EORTC 10801 trial. J Natl Cancer Inst Monogr 1992;11:15–8. (8) Blichert-Toft M, Rose C, Andersen JA, Overgaard M, Axelsson CK, Andersen KW, et al. Danish randomized trial comparing breast conservation therapy with mastectomy: six years of life-table analysis. Danish Breast Cancer Cooperative Group. J Natl Cancer Inst Monogr 1992;11: 19–25. (9) Early Breast Cancer Trialists’ Collaborative Group. Effects of radiotherapy and surgery in early breast cancer. An overview of randomized trials. N Engl J Med 1995;333:1444–55. (10) Morris AD, Morris RD, Wilson JF, White J, Steinberg S, Okunieff P, et al. Breast-conserving therapy vs mastectomy in early-stage breast cancer: a meta-analysis of 10-year survival. Cancer J Sci Am 1997;3:6–12 (11) van Dongen JA, Bartelink H, Fentiman IS, Lerut T, Mignolet F, Olthuis G, et al. Factors influencing local relapse and survival and results of salvage treatment after breast-conserving therapy in operable breast cancer: EORTC trial 10801, breast conservation compared with mastectomy in TNM stage I and II breast cancer. Eur J Cancer 1992;28A:801–5. (12) Harmer MH, editor. TNM classification of malignant tumours. 3rd ed. Geneva (Switzerland): International Union Against Cancer; 1978. p. 47–54. (13) Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–81. (14) Cox DR. Regression models and life tables (with discussion). J R Stat Soc Series B 1972;34:187–220. (15) Gooley TA, Leisenring W, Crowley J, Storer BE. Estimation of failure probabilities in the presence of competing risks: new representations of old estimators. Stat Med 1999;18:695–706. (16) Fisher B, Anderson S, Redmond CK, Wolmark N, Wickerham DL, Cronin ARTICLES 1149 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 relative to the number of patients with distant metastasis, and the higher risk of locoregional recurrence after BCT does not seem to be of sufficient magnitude to cause a substantial difference in rates of distant metastasis and overall survival. The prognosis of patients with locoregional recurrence after mastectomy and BCT has been described in more detail in another report (25). That report concluded that the prognosis of patients with locoregional recurrence after BCT is similar to that of patients with locoregional recurrence after mastectomy. We found substantial differences in locoregional recurrence rates between the centers participating in this study, both when comparing the two treatment arms and when looking at each treatment arm separately. These differences may be due to variations in treatment techniques, in the skills of surgeons as well as of pathologists and radiation oncologists, in the selection of patients for the trial, or in the distribution of risk factors within the population from which the patients were selected, or they could be due just to chance. The difficulty of ensuring uniformity between the participating centers with respect to these factors illustrates the difficulties of interpreting the data from a multicenter trial and underlines the need for quality assurance of treatment techniques and patient selection. Our finding that BCT is an effective procedure for patients with breast tumors up to 5 cm needs to be considered in the context of the eligibility criteria and the treatment techniques used in the trial. Nowadays, more attention would be paid to microscopic completeness of the tumor excision in combination with generally established risk factors of locoregional recurrence, such as young age and an extensive intraductal component. In addition, the EORTC 10801 trial is characterized by a high radiation dose after BCT, especially the high booster dose, which impaired the cosmetic results because of increased fibrosis of the breast. Nevertheless, evaluation of the quality of life in the EORTC 10881 trial showed that BCT helped to maintain the patients’ body image, resulted in a higher satisfaction with treatment, and yielded no substantial difference from mastectomy with respect to fear of recurrence (26). Trial 22881 of the EORTC is now studying whether lowering the radiation dose to the breast by omitting the booster dose for patients with a microscopically complete excision will result in an equivalent local control and improve cosmetic results after BCT. In conclusion, after a median follow-up of 13.4 years, BCT and mastectomy still demonstrate similar survival rates for patients with tumors up to 5 cm. Even after this long median follow-up time, the number of locoregional recurrences in both treatment arms remains small relative to the number of distant recurrences. (17) (18) (19) (20) (21) (23) (24) 1150 ARTICLES al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997;337:949–55. (25) van Tienhoven G, Voogd AC, Peterse JL, Nielsen M, Andersen KW, Mignolet F, et al. Prognosis after treatment for loco-regional recurrence after mastectomy or breast conserving therapy in two randomised trials (EORTC 10801 and DBCG-82TM). EORTC Breast Cancer Cooperative Group and the Danish Breast Cancer Cooperative Group. Eur J Cancer 1999;35:32–8. (26) Curran D, van Dongen JA, Aaronson NK, Kiebert G, Fentiman IS, Mignolet F, et al. Quality of life of early-stage breast cancer patients treated with radical mastectomy or breast-conserving procedures: results of EORTC Trial 10801. The European Organization for Research and Treatment of Cancer (EORTC), Breast Cancer Co-operative Group (BCCG). Eur J Cancer 1998;34:307–14. NOTES We are indebted to Dr. G. van Tienhoven (Academic Medical Center, Amsterdam, The Netherlands) for his advice and to Dr. C. Gawlik (Kassenärztliche Bundesvereinigung, Cologne, Germany) who previously analyzed data of this study. We thank Dr. G. Groot (St. Lucas Hospital, Amsterdam) and Dr. G. de Keizer (Maria Hospital, Tilburg, The Netherlands) for recruitment of additional patients and Ms. F. Mignolet (EORTC Data Center, Brussels, Belgium), Ms. D. Ridley (Guy’s Hospital, London, U.K.), Ms. G. Vierstra (Dr. Daniël den Hoed Cancer Center, Rotterdam, The Netherlands), Ms. M. Mahn (The Netherlands Cancer Institute, Amsterdam), and Ms. P. Vermeiren (University Hospital Gasthuisberg, Leuven, Belgium) for data management. Manuscript received October 18, 1999; revised May 16, 2000; accepted May 17, 2000. Journal of the National Cancer Institute, Vol. 92, No. 14, July 19, 2000 Downloaded from http://jnci.oxfordjournals.org/ at Instituto Nacional de Cancerologia on April 26, 2014 (22) 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: 1456–61. Jacobson JA, Danforth DN, Cowan KH, d’Angelo T, Steinberg SM, Pierce L, et al. Ten-year results of a comparison of conservation with mastectomy in the treatment of stage I and II breast cancer. N Engl J Med 1995;332: 907–11. Fisher B, Anderson S, Fisher ER, Redmond C, Wickerham DL, Wolmark N, et al. Significance of ipsilateral breast tumour recurrence after lumpectomy. Lancet 1991;338:327–31. Fourquet A, Campana F, Zafrani B, Mosseri V, Vielh P, Durand JC, et al. Prognostic factors of breast recurrence in the conservative management of early breast cancer: a 25-year follow-up. Int J Radiat Oncol Biol Phys 1989;17:719–25. Kurtz JM, Amalric R, Brandone H, Ayme Y, Jacquemier J, Pietra JC, et al. Local recurrence after breast-conserving surgery and radiotherapy. Frequency, time course, and prognosis. Cancer 1989;63:1912–7. Veronesi U, Marubini E, Del Vecchio M, Manzari A, Andreola S, Greco M, et al. Local recurrences and distant metastases after conservative breast cancer treatments: partly independent events. J Natl Cancer Inst 1995;87: 19–27. Kemperman H, Borger J, Hart A, Peterse H, Bartelink H, van Dongen J. Prognostic factors for survival after breast conserving therapy for stage I and II breast cancer. The role of local recurrence. Eur J Cancer 1995;31A: 690–8. Ragaz J, Jackson SM, Le N, Plenderleith IH, Spinelli JJ, Basco VE, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 1997;337:956–62. Overgaard M, Hansen PS, Overgaard J, Rose C, Andersson M, Bach F, et