| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
1 Department of Surgery, University Medical Centre Utrecht, P.O. Box 85500, , 3508 GA, Utrecht, The Netherlands
2 Julius Centre for Health Sciences and Primary Care, University Medical Centre Utrecht, P.O. Box 85500, 3508 GA, Utrecht, The Netherlands
3 Comprehensive Cancer Centre Middle Netherlands, P.O. Box 19079, 3501 DB, Utrecht, The Netherlands
Correspondence: Address correspondence and reprint requests to: Adriaan Hennipman, MD, PhD; E-mail: a.hennipman{at}chir.azu.nl.
| ABSTRACT |
|---|
|
|
|---|
Methods: A total of 499 female patients with 502 DCIS lesions treated in the period 1989 to 2002 were retrospectively reviewed. Survival rates were calculated by the Kaplan-Meier method, and differences were tested by using the log-rank test. The association of variables with local recurrence was analyzed by using the
2 test.
Results: Treatment constituted of lumpectomy in 329 patients (65%). Thirty-eight patients (8%) had disease-positive margins, and for 41 patients (8%) the margin status was not known. Eighty tumors recurred, for a local recurrence rate of 13% after 4 years compared with 17% for patients treated with breast-conserving surgery only. Risk factors for ipsilateral recurrences were younger age (<50 years), treatment with breast-conserving surgery only, and presence of disease-involved surgical margins.
Conclusions: Conservative treatment of DCIS results in high recurrences rates, and outcomes can be improved by performing more radical surgery. Because radiotherapy has been shown to be effective in preventing recurrent disease, and, to date, no subgroups have been identified in which radiation can be omitted, its use is recommended, especially in younger patients.
Key Words: Ductal carcinoma-in-situ Breast Surgical treatment Local recurrences Risk factors
| INTRODUCTION |
|---|
|
|
|---|
An important treatment-related risk factor for local recurrences is the adequacy of surgical excision.11,12 This is demonstrated by the fact that most recurrences after breast-conserving surgery occur at the same site of and are clonally related to their primary lesions.1315
Recently, two large-scale randomized studies revealed that radiotherapy after breast-conserving surgery for DCIS decreased the incidence of local recurrences.1621 However, it has also been shown that radiotherapy cannot compensate for an inadequate excision in reducing the risk of local recurrence.2224 Therefore, a microscopically complete excision remains the cornerstone in treatment of DCIS. Whether all patients with DCIS should receive radiotherapy after breast-conserving surgery remains a topic of active debate.11,21 The aim of this population-based study was to assess the outcome of conservative surgical treatment of patients with DCIS without radiotherapy and to identify subgroups of patients at increased risk for local recurrence.
| PATIENTS AND METHODS |
|---|
|
|
|---|
Through the Cancer Registry from the Comprehensive Cancer Centre of the Middle Netherlands, all patients with a first diagnosis of DCIS in the period between January 1989 and December 2002 were identified. Patients with a simultaneously occurring invasive breast lesion and/or other malignancies (except for nonmelanoma skin cancer) were excluded. A total of 604 patients were eligible, and their medical records were reviewed. For 36 patients, no medical record was available. Another 69 patients were excluded because of a microinvasive component of the initial DCIS lesion (12 patients), DCIS with lobular carcinoma-in-situ (8 patients), invasive breast carcinoma as an initial diagnosis (9 patients), simultaneous contralateral invasive breast carcinoma (3 patients), a history of breast cancer (6 patients), no further information or follow-up data (20 patients), or other criteria (no diagnosis of DCIS or an unknown malignancy elsewhere; 11 patients).
From the remaining 499 patients with 502 DCIS lesions, all available clinical, pathologic, and follow up data were collected. The following data were noted: detection method (clinical diagnosis or mammographic detection), mammographic appearance (calcifications, architecture distortion, or both), age at diagnosis, family history of breast cancer (both first and second degree), location of primary DCIS (inner quadrant, outer quadrant, or central), size of DCIS lesion (according to mammographic or pathology reports, if specified), surgical treatment (including the performance of re-excision and the total number of surgical procedures), use of adjuvant radiotherapy, pathologic classification of the DCIS lesion, and final margin status (positive or negative, and margin width, if specified).
Follow-up started at the date of the first surgical procedure and ended at the last visit at the outpatient clinic. A local recurrence was defined as a pathology-proven carcinoma, both noninvasive and invasive, anywhere in the treated breast. Contralateral "recurrences" were not considered treatment failures.
Survival rates for ipsilateral recurrences were calculated by using the Kaplan-Meier method. Statistical significance of differences between survival curves of subgroups was tested by using the log-rank test. The association of clinical, pathologic, and treatment-related variables with recurrent disease was analyzed by using the
2 test. For continuous variables, Students t-test was used. For all statistical analyses, a P value <.05 was considered statistically significant.
| RESULTS |
|---|
|
|
|---|
|
After 1 or more surgical procedures (mean, 1.54; median, 1.00), 329 patients (65%) were treated by lumpectomy, and 173 patients (35%) were treated by mastectomy. Twenty patients (4%) received postoperative radiotherapy as a result of inclusion in the European Organization for Research and Treatment of Cancer (EORTC) 10853 trial. At the start of follow-up, 38 patients (8%) had disease-involved margins (including 10 patients with disease-involved margins after mastectomy), and in 41 patients (8%), margin status was not specified.
From 214 DCIS lesions (43%), a histopathologic classification could be obtained from pathology reports: 29 lesions (19%) were classified as grade 1, or well differentiated; 45 lesions (29%) were classified as grade 2, or moderately differentiated; and 79 lesions (52%) were classified as grade 3, or poorly differentiated. A total of 52 DCIS tumors were classified as comedo type, whereas 9 were classified as noncomedo type DCIS. Pathology reports noted "ductal carcinoma-in-situ" only in 288 patients (57%).
During follow-up, five nonbreast cancerrelated deaths were recorded, for an overall survival rate of 98.7%. Ninety-eight patients (20%) developed a breast cancerrelated failure in either the treated or contralateral breast. One patient developed an ipsilateral recurrence and contralateral breast tumor simultaneously.
Eighty tumors (16%) recurred in the treated breast and thus were considered treatment failures. The median interval from surgery to ipsilateral recurrence was 30 months (range, 1130 months). Approximately half of these recurrent tumors (45%) were invasive breast carcinomas.
Table 2
presents the 4-year local recurrencefree survival rates calculated according to the Kaplan-Meier method. The cumulative incidence of all local recurrences after 4 years was 13%, whereas the cumulative incidence for invasive local recurrences after 4 years was 6%.
|
|
At 4 years of follow-up, younger patients (<50 years) had worse local recurrencefree survival rates compared with older patients (Fig. 1
; 4-year local recurrencefree survival, 81% vs. 88%, respectively; P = .02). The mean follow-up time was comparable for both age groups (P = .56). The mean age in the ipsilateral recurrence group was 54 years, compared with 58 years in the nonrecurrent or contralateral recurrence group (P = .01). Although younger patients underwent more surgical procedures (mean, 1.70 vs. 1.49 in older patients; t-test; P < .01), the results on margin status did not differ between age groups (negative margins vs. positive or unknown margin status; P = .76). Furthermore, younger patients were treated with a mastectomy less frequently compared with older patients (48% vs. 30%; P < .01). In patients for whom the histopathologic classification was known, high-grade DCIS lesions were found in 83% of younger patients compared with 71% in older patients, but this difference reached no statistical significance (P = .25).
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Preventing local recurrences is the main goal in treatment of patients with DCIS. Two large-scale randomized studies have been performed to compare conservative treatment of DCIS by breast-conserving surgery only with excision followed by radiotherapy in reducing local recurrences.1621 The National Surgical Adjuvant Breast and Bowel Project (NSABP) B17 trial, published in 1998, revealed after 90 months of follow-up that 27% of patients treated with breast-conserving surgery alone developed local recurrence, compared with 12% when surgery was followed by radiotherapy.17
The EORTC 10853 trial, which included patients with DCIS lesions up to 5 cm and excluded patients
70 years, revealed a 4-year local recurrence rate of 16% in the surgery-only group, compared with 9% in patients treated by postoperative radiotherapy.19 However, a comparison of a subgroup of patients treated with lumpectomy followed by radiotherapy outside and inside the trial showed significantly worse 4-year local recurrencefree rates: 83% outside the trial versus 98% inside the trial (P = .03).21 For eligible patients treated with breast-conserving surgery only outside the trial, the 4-year recurrence-free rate was 93%, compared with 89% for randomized patients (P = .45).21 Neither the NSABP nor EORTC trial revealed a beneficial effect of radiotherapy on either distant recurrences or breast cancerrelated mortality.
Besides the use of radiotherapy, other risk factors have been identified that are associated with an increased risk of local recurrences. Histopathologic factors associated with increased recurrence rates are a high nuclear grade of DCIS and the presence of necrosis.3,18,20,2529 In our analysis, no differences in recurrence rates for nonhigh-grade and high-grade lesions were found. This can be mainly attributed to the small number of lesions for which a histopathologic classification was known and to the use of different histopathologic classification systems in the various hospitals over time.
As has been outlined, a microscopically complete removal of DCIS is an important factor in preventing recurrent disease. However, in daily practice, it can be difficult to obtain surgical margins free of DCIS. Excision of a screening-detected DCIS lesion has to be performed with a localization procedure without the guidance of a palpable mass. Furthermore, mammographic estimates of DCIS size frequently do not correlate with the histological DCIS size.30 It has been shown that 43% of patients with initially clear margins, defined as a rim of normal breast tissue of
1 mm, had residual DCIS at re-excision or mastectomy.22
In our study population, 35% of patients were ultimately treated with mastectomy because of tumor-involved surgical margins after initial excision. This percentage of mastectomies in our population is comparable to results published from a large population-based study performed in the United States (34% mastectomy rate).1 Furthermore, at the start of follow-up, 38 patients (8%) had positive margins, and in 41 patients (8%) margin status could not be specified. It is noteworthy that 6% of patients treated with mastectomy had disease-involved margins. From both the NSABP and EORTC trials, results on margin status were reported, and these are comparable to our data. An analysis of 77% of patients from the NSABP cohort showed that 16% of patients had uncertain or involved surgical margins. A comparable analysis of 85% of patients from the EORTC cohort revealed that from 12% of patients no information on margins was available and that 9% had close (margin width
1 mm) or involved margins.1621
The difficulty of achieving a complete excision of DCIS can be explained by three-dimensional studies of growth patterns of DCIS. These studies showed that most lesions involve a single duct system and that gaps between DCIS rarely exceeded 10 mm.3032 According to this growth pattern, complete eradication of the lesion could be achieved in approximately 90% of cases if the lesion were excised with a rim of at least 10 mm of healthy breast tissue. Silverstein et al.23 showed that excision of DCIS with a margin of
10 mm obtained excellent local control with or without radiotherapy (3 [2.3%] local recurrences out of 133 patients after a median follow-up of 72 months). Furthermore, this study showed that if DCIS was excised with a margin of
1 mm, adjuvant radiotherapy indeed decreased the rate of local recurrence from 58% to 30% in 8 years, but this still is an unacceptably high rate of local failure.23 Clearly, radiotherapy does not compensate for surgical margins not free of disease.22,23
Although a margin width of 10 mm has been recommended, obtaining such a margin is difficult (only two of our patients had a margin width of
10 mm). Furthermore, widespread recommendation of this margin width could give rise to the performance of more mastectomies, which could be considered as overtreatment of noninvasive DCIS.
Another important risk factor for local recurrence in this study is age: patients <50 years of age had an increased risk for local recurrence. For invasive breast carcinoma, younger age has found to correlate with local recurrence.33 The EORTC trial revealed patient age <40 years as a risk factor for recurrence.16 In retrospective series, younger age also has been identified as a risk factor for ipsilateral recurrence.3438 Our data confirm these findings: younger patients (<50 years) were more likely to develop a local recurrence. This age difference could not be attributed to different treatment strategies: no difference in mean age was found in the mastectomy and breast-conserving surgery groups. It can be hypothesized that DCIS lesions in younger patients have a worse biological behavior. Indeed, in our population, younger patients had lesions of higher grade, but this difference did not reach statistical significance. Otherwise, surgeons treating younger patients could be more likely to perform less radical excisions to obtain a better cosmetic outcome.39 However, the volume of excised tissue is inversely related to cosmetic outcome but directly related to residual DCIS and, therefore, the risk of local recurrence.39,40 Indeed, we found more surgical procedures in younger patients, although no difference in final margin status was found.
The previously mentioned risk factors for local recurrences (histopathologic grading, margin width of excision, tumor size, and age) have been summarized by Silverstein11 in the University of Southern California/Van Nuys Prognostic Index scoring system. The goal of this classification system is to identify certain subgroups of patients with DCIS who have an increased risk of local recurrences, thereby guiding the treatment strategy. For example, some patients probably require no adjuvant treatment after complete local excision (older patients with small, low-grade lesions excised with a wide margin width), whereas others are at such an increased risk for local recurrence that mastectomy could be a reasonable treatment option. Still, the validity of this classification system has to be prospectively confirmed in large groups of patients.
In conclusion, we showed that treatment of DCIS with conservative surgical treatment without radiotherapy results in relatively high recurrence rates. Clearly, this treatment option for all patients with DCIS is not sufficient. Because the cornerstone in the treatment of DCIS remains radical excision, more e orts should be made to perform radical surgery, because 8% of our patients had involved margins at the start of follow-up. On the basis of randomized trials, one could expect lower recurrence rates if patients were treated with radiotherapy after breast-conserving surgery. However, it is not clear whether all patients with DCIS will benefit from radiotherapy. Prospectively validated data to identify subgroups of patients at such a low risk for local recurrence that radiation can be omitted safely are not available yet. However, because younger patients (<50 years) have an increased risk for local recurrence, the adjuvant use of radiotherapy in this group of patients is recommended strongly.
| APPENDIX 1. Members of the Comprehensive Cancer Centre Middle Netherlands Surgical Oncology Group |
|---|
|
|
|---|
Received for publication June 17, 2005. Accepted for publication December 22, 2005.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |