Portal de Periódicos da CAPES

Standards for diagnosis and management of ductal carcinoma in situ (DCIS) of the breast. American College of Radiology. American College of Surgeons. College of American Pathologists. Society of Surgical Oncology

1998; Wiley; Volume: 48; Issue: 2 Linguagem: Inglês

10.3322/canjclin.48.2.108

1542-4863

David P. Winchester, E. A. Strom,

Cancer Genomics and Diagnostics

Standards of care in the diagnosis and management of any disease should be based on the best available scientific information. Such information is derived from prospective, randomized clinical trials through the cooperative group or intergroup mechanism, single or multi-institutional prospective trials, prospective nonrandomized trials, retrospective studies, and personal experience. A collaborative effort of the American College of Radiology, the American College of Surgeons, the College of American Pathologists, and the Society of Surgical Oncology in 1992 culminated in a publication entitled “Standards for Breast-Conservation Treatment.”1 Both invasive and noninvasive diseases were considered in that document, although the focus was on invasive breast cancer. A task force of the same four national organizations decided that a sufficient body of knowledge had developed about ductal carcinoma in situ of the breast for a separate standard to be published. Before the widespread use of screening mammography, ductal carcinoma in situ (DCIS) was an infrequently encountered problem that was routinely treated by mastectomy. As a result, a limited amount of information exists about the natural history of DCIS to act as a basis for treatment decisions. In addition, most DCIS seen today is identified mammographically because of the presence of microcalcifications, and it is unclear whether the biologic potential of this subclinical DCIS is the same as that of clinically evident DCIS. Total mastectomy, excision and irradiation, and excision alone all have been advocated as management strategies for DCIS. The acceptance of breast-conserving therapy for invasive carcinoma has stimulated great interest in the use of this technique for the management of DCIS. However, data from randomized trials comparing mastectomy with lumpectomy and irradiation in patients with invasive carcinoma cannot be directly extrapolated to patients with DCIS. In the patient with invasive carcinoma, the risk of metastatic disease is present at the time of diagnosis and is not altered by local recurrence in the breast. In DCIS, the risk of metastases at the time of diagnosis is negligible, so an invasive local recurrence carries with it the risk of increased breast cancer mortality. The appropriateness of breast-conserving approaches in DCIS should be determined by the incidence of invasive recurrence in the breast and the results of salvage therapy. The evaluation of the results of different local therapies in DCIS is complicated by changes in the presentation of DCIS, differences in the extent of mammographic and pathologic evaluation over time, and the long natural history of the disease. Treatment selection for the individual patient with DCIS requires a clinical, mammographic, and pathologic evaluation. The term “ductal carcinoma in situ” encompasses a heterogeneous group of lesions. Before a patient's suitability for breast conservation with or without irradiation or the necessity of mastectomy can be determined, the extent and character of the patient's disease must be identified by a thorough evaluation. An adequate history and physical examination include a complete assessment of the patient's overall health status. Much of the information needed to determine a patient's suitability for breast-conservation therapy can be obtained from a directed history and physical examination. The elements of the history and physical examination specific for breast cancer, which are listed in Table 1 of this article and Table 8 of “Standards for Diagnosis and Management of Invasive Breast Carcinoma” (page 90), represent information that may influence the selection of local therapy. DCIS most commonly presents mammographically as microcalcifications. How-ever, about 10% of mammographically evident DCIS is an uncalcified mass. Occasionally, DCIS is diagnosed without mammographic findings. Recent mammographic evaluation (usually within 3 months) before biopsy or definitive surgery is needed to establish the appropriateness of breast-conservation treatment by defining the extent of the patient's disease. Because the contralateral breast should be evaluated, bilateral mammography is required. Preoperatively, the diagnosis of DCIS can be suggested by mammography, but a definitive diagnosis depends on pathologic evaluation of the specimen. Because imaging techniques cannot determine whether the basement membrane has been violated, mammography cannot determine whether a tumor is invasive. Additionally, because peritumoral inflammation or fibrosis (or both) can cause a mass to be present along with microcalcifications, the diagnosis of DCIS requires histopathologic examination. The subtypes of DCIS can be suggested on the basis of characteristic patterns of calcifications, but these patterns are not diagnostic; the definitive diagnosis depends on the analysis of tissue by the pathologist. The mammogram may underestimate the extent of DCIS. This may be true with increasing lesion size. However, in all cases an effort should be made to determine the extent of tumoral calcifications preoperatively. In addition to routine mediolateral oblique and craniocaudal views, magnification views and any other special views that may be required should be obtained in an attempt to identify areas of calcified tumor that otherwise might not be apparent. The entire breast should be carefully examined to determine if areas of tumor are present elsewhere in it, which would influence a decision about breast-conserving treatment. The role of other imaging modalities, especially magnetic resonance (MR) imaging, in staging the extent of tumor within the breast has yet to be established. When breast-conservation treatment is appropriate, the goals of any surgical procedure on the breast are total removal of the tissue suspected or known to be malignant and minimal cosmetic deformity. These goals apply to both diagnostic biopsy and definitive local excision. Failure to consider them at all stages may jeopardize conservation of the breast. DCIS can present as a palpable mass, but this presentation is unusual. The standards described for palpable invasive disease apply to palpable DCIS. Because DCIS most commonly presents as microcalcifications, image-directed procedures are necessary for diagnosis and treatment. Stereotactic core-needle biopsy of the breast— performed by qualified radiologists, surgeons, or other physicians—can be the initial approach for sampling suspicious nonpalpable mammographic abnormalities. Ultrasound-guided biopsy is useful for nonpalpable masses but usually cannot be relied upon for biopsy of microcalcifications. Not all patients with microcalcifications are ideal candidates for stereotactic biopsy. Some patients' breasts may be too small to be accommodated by the stereotactic system. The thickness of the breast must be adequate to allow the full throw of the automated biopsy device when such a device is used. Abnormalities just under the skin may pose technical problems in some cases. If calcifications are widely separated, generating useful stereotactic coordinates to guide needle biopsy may be difficult. When microcalcifications are not tightly clustered or when the sensitivity or resolution of the stereotactic imaging system is hampered so that individual microcalcifications are not well visualized, accurate localization and retrieval of microcalcifications within core biopsy specimens may be difficult. Finally, the difficulty of the procedure increases with an uncooperative patient. If any one or a combination of these adverse factors exists, image-directed open surgical biopsy is the preferred approach. For lesions amenable to stereotactic breast biopsy, multiple cores should be obtained, and the specimen should be radiographed to confirm an adequate sampling of the microcalcifications. Leaving some microcalcifications at the site is desirable because if DCIS is diagnosed, they can accurately image-direct the surgeon to definitive excision. If all microcalcifications or the entire mass has been removed, a marker can be left at the biopsy site so that the area can be localized. If a presurgical diagnosis of DCIS is made by percutaneous core-needle biopsy, physicians should be aware that areas of invasive carcinoma are found in about 20% of cases at the time of surgical excision. Nonpalpable, mammographically evident lesions that are excised surgically should be localized presurgically with a guide, such as a guide wire. Any suspicious lesion detected by mammography requires presurgical localization to ensure accurate removal of the abnormal area and to avoid excess sacrifice of breast tissue. The localization method can be needle-hook wire, dye injection, or a combination of both. Localization should be precise and may require positioning of more than one wire. Labeled craniocaudal and lateral films that show the hook wire should be sent to the operating room for the surgeon's orientation. The current diagnostic films may be of additional value to the surgeon. The surgeon should assess the exact location by triangulation (based on the position, depth of penetration, and angle of the wire) and place the incision closest to the tip of the wire to achieve the best cosmetic result (see Fig. 2 of “Standards for Diagnosis and Management of Invasive Breast Carcinoma,” page 97). Tunneling should be avoided, and the skin incision should be made as close to the lesion as possible. The incision should be long enough to permit the removal of the specimen in one piece. Removal of the lesion in numerous fragments should be avoided because doing so precludes margin assessment and size determination. Curvilinear skin incisions are preferable (see Fig. 1 of “Standards for Diagnosis and Management of Invasive Breast Carcinoma,” page 96). Periareolar incisions are not appropriate for lesions in the periphery of the breast. The procedure can be done readily with the patient under local anesthesia with or without intravenous sedation. Meticulous hemostasis is critically important. Hematoma formation produces changes that are difficult to interpret by physical examination. In addition, the evolving scar from a hematoma may make interpretation of mammography difficult. These changes may be long lasting and lead to unnecessary biopsy because of the difficulty in evaluation. A better cosmetic result can be achieved by allowing the biopsy cavity to fill with serum, although reapproximation of the biopsy cavity may be appropriate under some circumstances. Drains in the breast should be avoided. Skin incisions should be closed with a subcuticular technique. The specimen should be radiographed intraoperatively to determine that the mammographic lesion has been excised and to direct pathologic analysis to the site in question in the removed tissue. Magnification and compression of the specimen increase the resolution of the radiograph. The specimen film should be correlated with a preoperative mammogram and interpreted without delay. Absence of the mammographic abnormality on the specimen radiograph usually indicates that it has not been removed. If the diagnosis is DCIS, extension of calcification (or mass) to the margin of the specimen suggests that a residual tumor might be present in the breast and that further resection along that margin may be indicated. The specimen radiograph is not adequate to determine the completeness of excision. Histologically negative margins also do not guarantee complete removal of the lesion because DCIS may grow in a discontinuous fashion. A postoperative mammogram should be obtained to document complete removal of the mammographic abnormality. It can be done as soon as the patient can tolerate compression. Magnification views may be helpful. Margin status and the postoperative mammogram are complementary means of assessing the completeness of excision. If re-excision is done, another postexcision mammogram should be obtained to reassess the tumorectomy site. The previous biopsy site must be re-excised carefully to ensure negative margins of resection, avoid excess removal of breast tissue, and achieve good cosmesis. If the presence of microcalcifications is the indication for reexcision, needle localization should be considered. Proper orientation of the original biopsy specimen avoids removal of an already adequate margin. When the site of inadequate margins is not known, a rim of tissue must be removed around the previous biopsy site. Axillary node metastases are uncommon in DCIS. Axillary dissection is not necessary for the management of most patients with DCIS. Unsuspected invasive or microinvasive carcinoma occurs more frequently in association with extensive DCIS of high nuclear grade. Most of these patients need mastectomy to encompass the disease. Therefore, a low axillary sampling or level I dissection performed when the mastectomy is done avoids a second operative procedure if invasive carcinoma is found in the mastectomy specimen. If a clinically suspicious node is found during surgery, a frozen section should be done, and if the node is positive, a level I and II axillary node dissection should be performed. The excised tissue should be submitted for pathology examination with appropriate clinical history and anatomic site specifications, including laterality (right or left breast) and quadrant. For wide excisions or segmental breast resections, the surgeon should orient the specimen (e.g., superior, medial, lateral) for the pathologist with sutures or other markers. The specimen radiograph should be available for the pathologist to review while examining the specimen. The pathologist's gross examination should document the type of surgical specimen when this information is provided (e.g., excisional biopsy, quadrantectomy), the size of the specimen, and the proximity of the tumor (if visible) or biopsy site to the margins of excision. The presence or absence of tumor at the margins of excision is determined by marking the margins with India ink or using another suitable technique. The entire mammographic lesion, and as much of the remaining specimen as practical, should be submitted for histologic examination. Additionally, the margins of the specimen must be thoroughly evaluated, particularly those closest to the lesion. Frozen-section examination of image-guided needle biopsies of nonpalpable lesions or mammographically directed biopsy done for microcalcifications is strongly discouraged. Distinguishing between atypical ductal hyperplasia and DCIS may be impossible in frozen-section preparations, and small foci of microinvasion may be lost or rendered uninterpretable by freezing artifact. Frozen sections should be prepared only when enough tissue is present that the final diagnosis will not be compromised (i.e., grossly visible tumors larger than 1.0 cm) and when the information is needed for immediate therapeutic decisions. DCIS traditionally has been classified primarily by architectural pattern. In this system, DCIS is divided into comedo, cribriform, micropapillary, papillary, and solid subtypes. This classification was developed at a time when all patients with DCIS were treated by mastectomy, however, and the histologic subclassification of DCIS was largely an academic exercise. With the increasing use of breast-conserving therapy for DCIS, lesions more likely to recur or progress to invasive cancer must be identified. Several new classification systems have been proposed, based primarily on nuclear grade, necrosis, or both. Many studies have supported the clinical relevance of this approach, showing that high nuclear grade or necrosis is associated with a higher risk of early local recurrence after breast-conservation therapy. Although a single classification system for DCIS has not yet been uniformly accepted, the pathologist should clearly report the nuclear grade of the lesion and the presence or absence of necrosis. If a specific grading system for DCIS is used, this should be stated in the pathology report. The report also should include the architectural patterns present because they may have clinical relevance (e.g., the micropapillary pattern may be more prone to involvement of multiple quadrants, independent of nuclear grade). The extent (size) of DCIS is an important factor in treatment decisions. In contrast to most invasive cancers, however, DCIS is difficult to measure because it is usually nonpalpable and cannot be identified grossly. Although precise measurement may not be possible, the pathologist may be able to estimate the extent of DCIS, and this information should be included in the pathology report. Several methods for estimating the extent (size) of DCIS have been suggested, including (1) directly measuring the size of the lesion if it is confined to a single slide; (2) determining the size after submitting the entire specimen for microscopic examination in sequence and in sections of uniform thickness (2 to 3 mm); (3) estimating the percentage of breast tissue involved by DCIS in relation to the total specimen; and (4) reporting the total number of slides examined and the number with DCIS. The assessment of surgical margins is arguably the most important aspect of the pathologic evaluation of breast tumor excisions in patients being considered for breast conservation. Although the definitions of “positive” and “negative” margins vary among institutions, microscopic extension of DCIS to surgical margins usually results in further surgery. The pathologist should clearly specify in the pathology report whether DCIS is transected at the surgical margin, and if not, how close the lesion is to the nearest margin. Determinations of estrogen and progesterone receptors, DNA content (ploidy), S-phase, and oncogene amplification are not necessary for noninvasive breast carcinomas. How the specimen was received (e.g., number of pieces, fixative, orientation) The laterality and quadrant of the excised tissue and the type of procedure, as specified by the surgeon Size of the specimen in three dimensions Whether the entire specimen was submitted for histologic examination The histologic features of DCIS (e.g., nuclear grade, necrosis, architectural pattern) An estimate of the extent (size) of DCIS (if possible) The location of microcalcifications (e.g., in DCIS, in benign breast tissue, or both) The presence or absence of DCIS at the margins of excision. If possible, the distance of the lesion or biopsy site from the margin should be stated. Without mature data from clinical trials, it is the collective responsibility of the surgeon, pathologist, radiation oncologist, and radiologist to integrate all available data so that treatment options and recommendations can be articulated clearly to the patient. The surgeon must decide, based on imaging studies and the pathology consultation report, whether the patient is a candidate for a breast-conserving approach. If so, local recurrence must be further discussed. Local recurrence with total mastectomy is rare. Local recurrence is observed at a higher rate in patients treated with breast conservation, but the impact of these local recurrences on overall survival is probably small. Finally, patients need to understand the excellent prognosis for this disease with either surgical approach. Information on 39,010 patients with DCIS of the breast diagnosed between 1985 and 1993 was collected through the National Cancer Data Base. During the 8 years of analysis, the use of breast-conservation therapy increased from 31% to 54%. Tumors with favorable sizes and grades were associated with increased rates of breast preservation and lower rates of axillary dissection and use of radiotherapy. In the early years of the study, radiotherapy was administered to only 38% of patients postoperatively, but by 1993 it was used in 54%. The axilla was dissected in 49% of the patients initially and in 37% by the end of the study. The inappropriately high rate of axillary dissection and the low rate of postoperative radiation therapy should improve through dissemination of clinical trial results and professional education. No prospective randomized trials have compared the treatment of DCIS by mastectomy with treatment by breast conservation. Studies from single institutions, including patients with both clinically evident and mammographically detected DCIS, indicate that 1% to 2% of patients treated by mastectomy will relapse, either regionally or systemically (Table 2). Presumably, relapse is caused by unrecognized foci of invasive tumor present in the breast. Thus, although mastectomy results in cure rates approaching 100%, it may be overtreatment for many patients with DCIS, particularly those with small, mammographically detected lesions. In 1985, the National Surgical Adjuvant Breast and Bowel Project (NSABP) began protocol B-17, a prospective randomized study to evaluate the worth of postoperative radiation therapy after lumpectomy for patients with DCIS. The initial results, with an average follow-up of 43 months, were published in 1993.2 Additional details of the pathology were published in 1995.3 In 1997, the results were updated.4 For this analysis, 814 patients were eligible for evaluation, with a mean time in the study of 90 months (range, 67 to 130 months). All patients had been followed for more than 5 years, and 35% had been followed for more than 8 years. Thirty-eight percent of these patients had axillary dissections, all of which were negative. The total number of ipsilateral breast tumor recurrences was 151, and 70 (46.4%) recurrences were invasive. Most of the ipsilateral breast tumor recurrences were at or near the original lesion. Analysis by treatment arm showed 104 ipsilateral breast tumor recurrences among 403 patients treated with lumpectomy only (25.8%) (Table 3). The cumulative rate of ipsilateral breast tumor recurrence at 8 years was only slightly higher, 26.8%. (For comparison, the cumulative ipsilateral breast tumor recurrence rate at 5 years reported in 1993 was 20.9%.2) Fifty-one ipsilateral breast tumor recurrences were noninvasive (13.4%) and 53 were invasive (13.4%). The rate of ipsilateral breast tumor recurrence was markedly reduced for patients who received radiation therapy in addition to lumpectomy. Only 47 ipsilateral breast tumor recurrences occurred in 411 patients treated with lumpectomy plus radiation therapy (11.4%). The cumulative ipsilateral breast tumor recurrence rate at 8 years was only slightly higher, 12.1%. (Again for comparison, the cumulative ipsilateral breast tumor recurrence rate at 5 years for patients treated with lumpectomy and radiation therapy as reported in 1993 was 10.4%.2) Thirty ipsilateral breast tumor recurrences were noninvasive (8.2%), and 17 were invasive (3.9%) (Table 3). The P values were significantly different when patients who received radiation therapy were compared with patients who did not. The update confirmed the original conclusions of NSABP protocol B-17 that ipsilateral breast tumor recurrence of both invasive and noninvasive breast cancer is significantly reduced by postlumpectomy radiation therapy. In the original analysis, margin status and the presence of comedonecrosis were independent predictors of ipsilateral breast tumor recurrence. Other randomized trials that have been started in North America and Europe (Table 4) compare the results achieved using breast-conserving surgery alone with those of breast-conserving surgery plus radiation therapy. All of them (except the Swedish trial) require histologically “negative” margins as an entry criterion. (Attempts by national cooperative groups in Germany and Denmark to perform randomized trials comparing surgery alone with surgery and radiotherapy had to be abandoned because of poor accrual.) As yet, no data are available from these trials. The results of conservative surgery and radiation for DCIS from retrospective series are presented in Table 5. The crude incidence of breast tumor recurrence ranges from 4% to 18%. Deaths caused by breast cancer have been reported in up to 4% of patients treated in studies with a median follow-up of 10 years or fewer. The long-term results of conservative surgery and radiation for DCIS were reported by Solin et al.5, 6 This collaborative study of 10 institutions in the United States and Europe analyzed outcome in 268 patients. Seventy-eight percent of the tumors were detected by mammography alone. The 10-year actuarial risk of breast recurrence was 16%, and the 10-year actuarial cause-specific survival was 97%.5 The 15-year actuarial breast recurrence was 19%, and the 15-year actuarial cause-specific survival was 96%. Median follow-up was 10.3 years.6 Various clinical, pathologic, and treatment-related factors have been assessed for their ability to identify patients with a substantial risk of recurrence in the treated breast. For these patients, mastectomy may be recommended. One factor that appears to be associated with a high risk of recurrence is the presence of residual malignant-appearing calcifications on a postbiopsy mammogram. Failure to remove these calcifications before radiation has resulted in a 100% recurrence rate in the few patients reported.7, 8 DCIS presenting as a bloody nipple discharge was noted in earlier series to be associated with a higher risk of recurrence. In the collaborative study, however, no increased risk appeared to exist in this group of patients.6 The significance of young age (less than 40 years) is controversial. Three studies have observed an increased risk of breast tumor recurrence (approximately 25%) in young women who have DCIS treated with conservative surgery and radiation compared with older women (approximately 10%).9-11 However, four additional studies have found no correlation between young age and breast recurrence rates.6, 8, 12, 13 A similar controversy exists regarding a positive family history of breast cancer. Two series7, 12 have reported a higher breast tumor recurrence rate (approximately 40%) in women with a positive family history compared with those who do not have such a history (approximately 10%). However, a third series found no such association.10 The impact of young age and a positive family history of breast cancer on treatment options in women with DCIS requires further evaluation. The contribution of various pathologic factors (histologic subtype, nuclear grade, necrosis) to the risk of breast recurrence in patients treated with conservative surgery and radiation is controversial. It was suggested initially that high-grade or comedo DCIS was associated with a higher rate of breast tumor recurrence.5, 14 However, in the collaborative study, the 10-year actuarial breast recurrence rate was 18% for tumors with the combination of both comedo pattern and a high nuclear grade versus 15% for DCIS in which these factors were absent (P = 0.15).6 The median interval to recurrence for comedo DCIS was 3.1 years versus 6.5 years for non-comedo DCIS. Therefore, series with shorter follow-up tend to underestimate the number of recurrences in low-grade or non-comedo DCIS, and recurrences in high-grade or comedo DCIS predominate. The influence of necrosis on breast recurrence rates remains to be determined. Silverstein and colleagues15 have designed the Van Nuys Prognostic Index. This is a quantitative algorithm that uses tumor size, margin width, and a pathologic classification based on nuclear grade and comedo-type necrosis to predict the likelihood of local recurrence with breast-conservation treatment for patients with DCIS. Scores range from a low of 3 (best prognosis) to a high of 9 (worst prognosis). Their results suggest that patients who score 3 or 4 generally have small, well-excised, low-grade lesions that can be treated successfully with excision alone. The Van Nuys Prognostic Index must be confirmed by others before it is accepted for widespread use. Most breast tumor recurrences in patients undergoing conservative surgery and radiation for DCIS occur near the primary tumor, and approximately 50% are invasive cancers.6-8, 12, 14, 16-20 Invasive recurrences appear at later intervals (5 years) than do noninvasive recurrences (4 years) and may occur in a separate quadrant.9, 17 Nearly all patients who develop a noninvasive recurrence are salvaged with mastectomy, and approximately 75% of those with an invasive recurrence are salvaged.7, 8, 12, 16, 18-21 Over the last 10 years, the method of detection of DCIS has changed significantly. Approximately 85% of all DCIS is now detected solely as a mammographic finding, which is most often characterized by the presence of microcalcifications. Earlier reports of conservative surgery and radiation for DCIS do not accurately reflect outcome for mammographically detected DCIS because many included clinically evident DCIS (palpable mass or bloody nipple discharge), and detailed mammographic and pathologic correlation was frequently lacking. Unfortunately, the results of these earlier series were compared with those of conservative surgery alone for mammographically detected DCIS and often claimed to be equal. The results of conservative surgery and radiation for mammographically detected DCIS are shown in Table 6. The 10-year actuarial breast tumor recurrence rate ranges from 6% to 23%, with a 10-year cause-specific survival of 96% to 100%. The variation in the results reported reflects differences in patient selection, the extent of surgical resection, and the degree of mammographic and pathologic correlation. Increasing evidence exists that wide surgical excision8 and negative margins of resection diminish the risk of breast tumor recurrence in patients with mammographically detected DCIS treated with conservative surgery and radiation.8, 9, 22 In the collaborative study (which had a median follow-up of 9.3 years), the crude breas