| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Michigan Health System, 1500 E Medical
Center Dr., Ann Arbor, MI 48109-0302.
2 Department of Radiology, William Beaumont Hospital, Royal Oak, MI.
3 Department of Radiation Oncology, University of Michigan Health System, Ann
Arbor, MI.
4 Present address: Oncology Department, Radiation Oncology Unit, Sheba Medical
Center, Ramat-Gan, Israel.
5 Department of Radiation Oncology, William Beaumont Hospital, Royal Oak,
MI.
6 Present address: Department of Radiology, Henry Ford Hospital, West
Bloomfield, MI.
Received September 27, 2006;
accepted after revision January 26, 2007.
Address correspondence to R. W. Pinsky
(rpinsky{at}umich.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Following institutional review board approval, records of women treated with breast-conserving surgery and radiation therapy for DCIS who developed an ipsilateral breast tumor recurrence from 1981 to 2003 were reviewed retrospectively. Multiinstitutional database records showed 513 women were treated, of whom 42 (8.2%) developed local recurrence. Study criteria were fulfilled and complete records were available for 32 women. Mean age at initial diagnosis was 49 years (range, 26-73 years).
RESULTS. Of the 32 patients included in our study, 31 (97%) recurrences were mammographically apparent. Twenty-nine (91%) of 32 were diagnosed exclusively by mammography. Mammographic findings at recurrence were calcifications in 24 (75%) of 32, mass in six (19%) of 32, and distortion in one (3%) of 32. The mean time to recurrence was 4.5 years. Twelve (40%) of 30 had the recurrence in a remote quadrant from the original cancer. Recurrences were DCIS in 17 (53%) of 32, DCIS with microinvasion in six (19%) of 32, invasive ductal cancer in three (9%) of 32, invasive lobular cancer in two (6%) of 32, and mixed DCIS and invasive cancer in four (13%) of 32. Six (67%) of nine patients with invasive cancer (excluding microinvasion) had tumors smaller than 1 cm. Ninety-one percent of recurrences were minimal cancers. All recurrences were stage 0 or 1.
CONCLUSION. Mammography successfully detected ipsilateral breast tumor recurrence, predominantly as calcifications or masses, after breast-conserving surgery with radiation therapy for DCIS in 97% of cases. The recurrences were located at variable distances from the lumpectomy site. Ninety-one percent of recurrences were minimal cancers and all were early stage, connoting excellent prognosis.
Keywords: breast • breast cancer • cancer recurrence • DCIS • ductal carcinoma in situ • mammography • women's imaging
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mastectomy for DCIS has been associated with a long-term local regional disease-free survival of 95% or greater [1-5]. Although breast conservation surgery with whole-breast radiation therapy has proven to be effective, women electing this treatment are faced with the prospect of a local regional recurrence rate of approximately 1% per year [6]. This decreases with the addition of tamoxifen in selected women [7, 8]. Although many consider mastectomy to be curative, a small percentage of patients (1-4%) will eventually develop chest wall recurrence [1-5]. These recurrences are usually invasive carcinomas that are detected once they are palpable. Because the rate of local recurrence with lumpectomy and breast radiation is higher than with mastectomy, it is important for a woman electing this treatment path to understand not only the risk of recurrence but also the implications of a recurrence. We therefore reviewed our data of women experiencing a local recurrence after breast-conserving surgery and radiation therapy. We were particularly interested to observe the radiographic appearance, method of detection, location of recurrence, and, most important, the size and stage of recurrence that would provide information regarding prognosis after recurrence. This information should be helpful in guiding radiologists, oncologists, and patients facing treatment decisions and planning surveillance methods when breast conservation is chosen.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Study Population
We performed a retrospective review of women treated for DCIS with
breast-conserving surgery and whole-breast radiation therapy who subsequently
developed an ipsilateral breast tumor recurrence. Breast imaging studies and
medical records, including demographic and pathology records, were analyzed.
Local recurrence was defined as the presence of cancer anywhere in the treated
breast. The study population was gathered from the radiation oncology
databases from two hospitals, one a large, academic, National Cancer
Institute-designated comprehensive cancer center and the other a large private
teaching hospital. The databases were searched for cases in which the initial
cancer and recurrence occurred between January 1986 and December 2003 at one
institution and between January 1981 and December 2003 at the other. All
initial cancers were primary unifocal DCIS, stage 0 according to the American
Joint Committee on Cancer (AJCC) staging system, version 5
[9]. No microinvasion was
present on histopathology.
A total of 513 women were treated with breast-conserving surgery and radiation therapy for DCIS at the two institutions during the study period. Of these women, 42 developed local recurrence (crude recurrence rate, 8.2%). Complete records, including mammograms (or detailed mammographic reports) and medical records, were available for the 32 patients who constitute our study population. Seven patients were excluded for lack of mammogram data or incomplete medical records. Three were excluded for initial presentation with multicentric DCIS.
In both centers, whole-breast radiation therapy was delivered using two opposed tangential fields. All patients were treated with megavoltage radiation, generally 6 MV, to a dose of 45-50 Gy in 1.8- to 2.0-Gy fractions over 4.5-5 weeks; a supplemental boost to the tumor bed, usually administered using electrons, was delivered in most cases. The median total tumor bed dose was approximately 60 Gy at both institutions. No regional lymphatic radiation was delivered. Specific treatment details for both centers are described elsewhere [10, 11].
Mammography and Imaging Assessment
The routine for imaging included physician-monitored diagnostic mammography
for all of the initial DCIS studies, including magnification views for
calcifications. For women with tumor-associated calcifications on the
diagnostic mammogram, a postlumpectomy mammogram was performed to identify
residual calcifications. Mammographic evaluation of the affected breast after
lumpectomy and radiation therapy at one of the institutions included
craniocaudal, mediolateral oblique, and lateral views and a magnification view
of the lumpectomy site at 6 months, a bilateral mammogram with a magnification
view of the lumpectomy site at 12 and 24 months, and then bilateral mammograms
annually. At the second institution, a similar protocol was followed, omitting
the lateral view at 6 months and adding a second magnification view of the
lumpectomy site at 12 and 24 months. MRI and whole-breast sonography were not
routinely performed at the time of diagnosis or as part of an annual
surveillance. Digital mammography was not used. Compliance was not analyzed in
this study. All cases of recurrence were presented at a multidisciplinary
tumor board conference that included breast radiologists, surgeons, medical
oncologists, radiation oncologists, nurses, and other support staff.
The mammograms and clinical information were reviewed by two Mammography Quality Standards Act-certified radiologists with 14 and 22 years of experience in mammographic interpretation. A recurrence was defined as the second occurrence of cancer in the treated breast. Original mammogram films were used when available. Alternatively, detailed mammography reports were used if they contained information regarding method of detection, location, size, and mammographic appearance of the recurrence. Mammographic findings on the initial and recurrence images were classified as calcifications, mass plus calcifications, mass alone, distortion, asymmetry, edema, and negative examination. Calcifications were categorized as amorphous and indistinct, coarse and heterogeneous, pleomorphic, or linear and branching. When it was available, the BI-RADS final assessment category was recorded [12]. The distance from the site of the primary cancer was measured or estimated from the lumpectomy site when original films were not available for review. The quadrant and distance from the nipple were identified for both the initial and recurrent carcinomas. The method of presentation (mammographic, palpable, nipple discharge, or other) of both the original and the recurrent cancers was recorded.
Pathologic Assessment
The pathologic results of the recurrent cancers were recorded from the
medical record. These diagnoses were categorized as DCIS, DCIS with
microinvasion, invasive ductal cancer, invasive lobular cancer, or other
invasive cancer. Microinvasion was defined as DCIS with an invasive component
of 0.1 cm or less [9]. Lobular
carcinoma in situ was not considered malignant. When available, the type and
grade of DCIS and the presence or absence of axillary lymph node metastases
and angiolymphatic invasion were recorded. The size of the recurrent cancer
was based on pathologic size in the medical record. When this was not
available, the mammographic size of the abnormality was used as the size of
the recurrence. A tumor recurrence was further characterized as minimal cancer
if it was noninvasive or, if invasive, as measuring 1 cm or less
[13]. The status of the
original surgical margins was described as positive, negative, or close. A
close margin was defined as DCIS less than 3 mm from the inked margin.
Clinical Assessment
The time to recurrence was calculated from the date of initial diagnosis to
date of recurrence. Time to recurrent DCIS versus invasive recurrence was
compared. Time to recurrence in the same quadrant and elsewhere in the breast
was analyzed as well. Tamoxifen use was not analyzed.
Statistics
Two-sample Student's t tests were used to determine the
significance between the interval to a noninvasive recurrence versus an
invasive recurrence and the interval to recurrence in the same quadrant versus
recurrence elsewhere in the breast. A difference with a p value of
less than 0.05 was considered significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Thirty-two of the 42 patients with recurrence had adequate medical and radiographic records and unifocal DCIS, and they constitute our study population. Thirty-one (97%) of 32 cancer recurrences were mammographically visible. Twenty-nine (91%) were exclusively diagnosed on mammography. Two (6%) were diagnosed by both palpation and mammography, and one (3%) was diagnosed by visual inspection with a normal mammogram (Paget's disease). The two recurrent cases that were palpable were invasive cancer.
Original mammogram films were available for review in 28 (88%) of 32 patients. The mammographic appearances of the initial and recurrent cancer are detailed in Table 1. Seventy-five percent of recurrences presented mammographically as microcalcifications, and 18% presented as masses on the mammogram. The calcifications of the primary DCIS were all pleomorphic or of the linear branching type. Morphologic analysis of the calcifications was possible for 22 (92%) of 24 calcification-containing recurrences. Twenty-one (95%) were either pleomorphic or of the linear branching type. One (5%) was in the amorphous indistinct category. Eighty percent of the patients who presented initially with microcalcifications had a recurrence manifested by microcalcifications. The appearance of the calcifications in both the initial and recurrent tumors was available in 18 patients. In 17 (94%) of 18 patients, the recurrent tumor calcifications had a morphology similar to the initial DCIS (Figs. 1A, 1B, and 1C).
|
|
|
|
Recurrences were pure DCIS in 17 (53%) of 32 (including one case of Paget's disease with DCIS), DCIS with microinvasion in six (19%) of 32, invasive ductal carcinoma in three (9%) of 32, invasive lobular carcinoma in two (6%) of 32, and DCIS with invasive cancer in four (13%) of 32. Seventeen of the recurrences (53%) were stage 0 and 15 (47%) were stage 1. Of the 15 patients with an invasive recurrence, 12 (80%) measured less than 1 cm. Overall, the recurrences were minimal cancers in 29 (91%) of 32 patients. Excluding cases of microinvasion, the median and mean sizes of the invasive cancers were 0.8 and 1.1 cm, respectively (range, 0.5-2.0 cm). None of 15 invasive carcinomas had angiolymphatic invasion. Seven patients with invasive cancer and two with pure DCIS had lymph node sampling at the time of the recurrence, and all were negative for metastasis.
The mean distance between the initial and the recurrent cancers was 3.8 cm (median, 3.5 cm; range, 0-9 cm) in 30 patients for whom these data were available. Eighteen (60%) of 30 patients had their recurrence in the same quadrant as the original DCIS (Figs. 1A, 1B, 1C, and 2).
|
The mean time to recurrence for all patients was 4.5 years. See Figure 3 for distribution of frequency of time to recurrence. No significant difference was seen in time to recurrence between noninvasive and invasive cancers (3.8 and 5.2 years, respectively) (p = 0.14). However, a trend was seen toward longer time to an invasive recurrence. No significant difference was seen in the mean time to recurrence in the same quadrant versus mean time to recurrence elsewhere in the breast (4.9 vs 4.1 years) (p = 0.42).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The crude recurrence rate of 8.2% confirms the low rate of tumor recurrence after breast-conserving surgery with radiation therapy. Several studies have addressed the rate of local recurrence after the various treatment options for DCIS [3, 4, 10, 14-17]. In the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-17 trial, local recurrence rates after breast-conserving surgery and radiation therapy were reported to be 7% at 5 years and 12% at 8 years [14, 17, 18]. The European Organization for Research and Treatment of Cancer (EORTC) trials confirmed reduced recurrence rates with radiation therapy, reporting 10-year local recurrence rates of 14.8% for the group treated with radiation and 26% for the group treated with lumpectomy alone (p < 0.0001) [19]. The United Kingdom Coordinating Committee on Cancer Research (UKCCCR) study showed 5-year ipsilateral recurrence rates of 6% for the radiation therapy group versus 14% for the group with lumpectomy alone (p < 0.0001) [20].
Mastectomy is an alternative treatment option for DCIS; however, it does not completely eliminate the risk of local recurrence. The incidence of locally recurrent cancer after mastectomy for DCIS has been variably reported as 1-4% for follow-up periods of 5-9 years [2-6]. These recurrences occur in the form of chest wall recurrences, and typically are invasive cancers detected by palpation because mammographic evaluation is not performed [1, 4].
As screening mammography has become standard practice, the incidence of DCIS has increased. It is estimated to represent 22% of all breast cancers diagnosed in the United States in 2005 [21]. The American Cancer Society estimates of the annual incidence of DCIS has increased from 30,000 in 1996 to more than 58,000 in 2005. Just as the initial diagnosis of DCIS is primarily made mammographically, recurrences are also primarily detected mammographically. Postsurgical and postirradiation changes may diminish mammographic sensitivity [22, 23]. Postoperative seroma, hematoma, scarring, and edema can alter the density and architecture of the breast. Progressively better mammography, regular use of magnification views of the lumpectomy site before radiation therapy, a baseline study after radiation therapy, routine follow-up mammograms, and better surgical and radiation therapy techniques resulting in less deformity [22] may enhance our ability to detect suspicious changes in the breast.
Most patients who presented with calcifications in the original DCIS had recurrences with calcifications. The finding of new suspicious calcifications on a follow-up mammogram should prompt a biopsy. Similarity in morphology between the primary and recurrent calcifications was seen in 94% of these patients. We consider this an aid for the radiologist in assessing follow-up studies, which underscores the value of having prior films for comparison.
The prognosis of breast cancer is closely tied to its stage at the time of diagnosis. In our study, all recurrences were diagnosed as an AJCC stage 0 or 1. The high percentage of recurrence as DCIS (53%) and as minimal breast cancer (91%) predicts an excellent prognosis [24]. Orel et al. [25] reported a statistically significant association between detection of a tumor recurrence on mammography alone and a lower tumor stage after treatment for earlystage invasive breast cancers. A trend was also noted toward noninvasive histology and longer disease-free survival.
In our study, only 60% of the recurrences occurred in the same quadrant and 40% recurred elsewhere in the breast. Tumor recurrence has previously been reported to occur with greatest frequency at or close to the lumpectomy site [2, 26]. Fisher et al. [17] reported early results of the NSABP B-17 trial, in which 84% of recurrences were in the same quadrant of the breast or at the border of the quadrant. In a study by Liberman et al. [26] of breast-conserving surgery for DCIS, 90% of the recurrences occurred in the same quadrant. We found no significant difference between the time to recurrence in the same quadrant and time to recurrence elsewhere in the breast. The former was 4.9 years and the latter was 4.1 years (p = 0.42).
In our study, all recurrences occurred after 1 year from the date of diagnosis. The mean time between the initial and recurrent cancers was 4.5 years (range, 1-12 years), with a clustering between 1 and 7 years. This time interval is longer than the mean time to recurrence of 2.8 years reported by Weng et al. [3] in a similar treatment group. In a multiinstitutional study reported by Solin et al. [27], mean time to recurrence was 5.7 years. These studies suggest that long-term mammographic surveillance is necessary after breast-conserving surgery and radiation therapy for women diagnosed with DCIS. However, Doubeni et al. [28] showed that the percentage of breast cancer survivors having regular mammograms declines with time after diagnosis.
Our study has several potential limitations. Because all patients had whole-breast radiation therapy, these results may not apply to patients who were treated with lumpectomy alone or with partial breast irradiation. None of the patients had clinical MRI or whole-breast sonography to locate other primary or residual disease. The impact of these tests on detecting future recurrence is unknown. The use of mammography reports when actual films were not available may have introduced a measurement error. It was necessary in some cases to use the mammographic size of a lesion, which is not considered as accurate as a pathologic measurement. Tamoxifen was not in widespread use during a large part of this study period. Studies have shown the expected recurrence rate is lower with the use of tamoxifen as adjuvant treatment after radiation therapy [8]. Postlumpectomy mammography was routinely used in this study to identify residual disease. Institutions that do not perform postlumpectomy mammography may experience different outcomes. Women at both centers were evaluated at comprehensive breast centers using multidisciplinary approaches to diagnosis and treatment, which may influence the course of treatment.
In conclusion, ipsilateral breast tumor recurrence after breast-conserving surgery and radiation therapy for DCIS was an infrequent event. Mammography was highly effective at detecting early tumor recurrence, with 97% presenting with an abnormal mammogram. Ninety-one percent of recurrences were minimal cancers at detection, suggesting an excellent prognosis even when tumors recur.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
