Protocol for the Examination of Specimens From Patients With Ductal Carcinoma In Situ of the Breast

This protocol applies to ductal carcinoma in situ (DCIS) without invasive carcinoma or microinvasion. The TNM staging system for carcinoma of the breast of the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer (UICC) is recommended.

The following types of breast specimens and procedures may be reported with the checklist:

Removal of breast tissue without the intent of removing the entire breast. The nipple is only rarely removed with excisions. Excisions include specimens designated biopsies, partial mastectomies, lumpectomies, and quadrantectomies.

• Wire-guided localization excisions: If a nonpalpable lesion is detected by mammography, ultrasound, or magnetic resonance imaging (MRI), a wire is placed to identify the location of the lesion for the surgeon. Mammography or ultrasound should be used to document the presence of the targeted lesion in the excised tissue. The specimen radiograph (if performed) and the results of the radiologic evaluation should be available to the pathologist. Because MRI utilizes vascular uptake, it is not possible to image the specimen by using this technique to document the presence and location of the lesion.

• Excisions without wire localization: These excisions are generally performed for palpable masses or to excise major ducts behind the nipple to evaluate nipple discharge.

Removal of all breast tissue, including the nipple and areola.

• Simple mastectomy: This procedure consists of a total mastectomy without removal of axillary lymph nodes.

• Skin-sparing mastectomy: This is a total mastectomy with removal of the nipple and only a narrow surrounding rim of skin.

• Modified radical mastectomy: This procedure consists of a total mastectomy and an axillary dissection. In the checklist, the breast and lymph node specimens are documented separately.

• Radical mastectomy: This procedure consists of a total mastectomy with removal of the pectoralis muscles and currently is performed only rarely. This type of specimen and procedure can be indicated on the checklist as “Other.”

The following types of specimens should not be reported using this checklist:

• Very small incisional biopsies (including core needle biopsies).

• Excisions after a core needle biopsy showing invasive carcinoma or DCIS with microinvasion (invasion measuring ≤0.1 cm). This type of case should be reported by using the protocol for invasive carcinoma of the breast,1 and the report should incorporate information from the prior needle biopsy.

• Reexcision of a biopsy site or mastectomy after an excision, unless a carcinoma of a higher T category than that in prior specimens is present. If invasive carcinoma is found on a subsequent excision, the protocol for invasive carcinomas of the breast1 should be used.

Specimen sampling for specimens with DCIS has the following goals2–6:

• The clinical or radiologic lesion for which the surgery was performed must be examined microscopically. If the lesion is a nonpalpable imaging finding, the specimen radiograph and/or additional radiologic studies may be necessary to identify the lesion. When practical, the entire specimen should be submitted in a sequential fashion for histologic examination. If this is not possible, at least the entire region of the targeted lesion should be examined microscopically. If DCIS, lobular carcinoma in situ (LCIS), or atypical hyperplasia is identified, all fibrous tissue should be examined.2 

• All other gross lesions noted in the specimen must be sampled.

• The margins must be evaluated for involvement by DCIS. If the specimen received is sectioned or fragmented, this should be noted, as this will limit the ability to evaluate the size of lesions and the status of margins. If the specimen is an incisional biopsy, margins need not be evaluated.

For specimens with a known diagnosis of DCIS (eg, by prior core needle biopsy), it is highly recommended that the entire specimen be examined by using serial sequential sampling to exclude the possibility of invasion, to completely evaluate the margins, and to aid in determining extent.7–9 If an entire excisional specimen or grossly evident lesion is not examined microscopically, it is helpful to note the approximate percentage of the specimen or lesion that has been examined.

Carcinomas present in excisions removed for lesions seen best by MRI studies are generally not grossly evident and not seen on specimen radiography.10 Complete microscopic evaluation of all tissue is necessary to detect all cancers in these specimens.

Recording the specimen size is important, because the volume of tissue excised has been associated with the likelihood of recurrence.11 

Tissue may be taken for research studies or assays that do not involve the histologic examination of the tissue (eg, reverse transcription–polymerase chain reaction or RT-PCR) only when taken in such a way to be able to evaluate the tissue for small areas of invasion. For example, a thin slice of tissue taken for research studies should be matched with an adjacent slice of tissue that will be examined microscopically.

Patients with DCIS may have lymph nodes sampled in the following situations.

• Extensive DCIS: Patients with extensive DCIS are more likely to have areas of invasion, and it may be difficult or impractical to examine all involved areas of the breast microscopically.12–14 A lymph node with a macrometastasis would indicate an occult area of invasion.

• Pathologic findings raising concern for invasion or microinvasion (invasion measuring ≤0.1 cm) on a prior needle biopsy or excision: If invasion has been documented, the checklist for invasive carcinoma of the breast1 should be used.

• Imaging findings (eg, an irregular mass) or clinical findings (eg, a large palpable mass) that increase the likelihood that stromal invasion is present.13 

• Planned mastectomy: The additional sampling of low lymph nodes or a sentinel lymph node does not result in increased morbidity. If the node or nodes are negative, and invasive cancer is found, another surgical procedure for node sampling can be avoided.

Most tumor cells in lymph nodes of patients with DCIS would be classified as isolated tumor cells.15,16 If a larger nodal metastasis is found and the breast tissue has not been entirely submitted for microscopic examination, additional sampling should be considered to attempt to identify invasive carcinoma.12,14 

Grossly uninvolved nodes should be submitted in their entirety for histologic examination, whereas a representative section of a grossly positive node may be submitted. Small nodes (eg, 0.2 to 0.3 cm) may be submitted intact, but larger nodes should be thinly sectioned. If nodes are inked with different colors before slicing, an accurate count of positive nodes can be obtained when slices of multiple nodes are included in the same cassette. An accurate assessment of the number of positive lymph nodes is a critical prognostic indicator.

Sentinel lymph nodes are identified as such by the surgeon, generally by uptake of radiotracer or dye.

Although not required for pT classification or stage assignment, the size (extent) of DCIS is an important factor in patient management.5,17 Extent (as determined by a number of different methods) is correlated with the likelihood of residual disease after reexcision,18–21 close or positive margins,18,21 local recurrence,22–24 and the possibility of missed areas of invasion.13,14 Extent is not as important for predicting local recurrence when wide margins are obtained.22,23,25 

Extent is an estimation of the volume of breast tissue involved by DCIS. Mammographic assessment of DCIS, usually based on distribution of calcifications, frequently underestimates, and sometimes overestimates, the extent of DCIS. Precise measurement of extent is generally difficult or impossible for the following reasons26:

• Ductal carcinoma in situ involves the ductal system in a complex 3-dimensional branching pattern that is usually only apparent by microscopic examination. When gross findings are present (eg, areas of tissue thickening and/or punctate necrosis), they often do not correspond to the entire area of involvement.

• The ductal system and surrounding tissue is highly compressible (a fact that any woman who has undergone mammography can verify). Specimens may be distorted during surgery or specimen processing or compressed during specimen radiography.27,28 

• Diagnostic gaps in ductal involvement may be present (particularly for low-grade DCIS).

• Ductal carcinoma in situ is often not removed in one excision and may be present in multiple specimens from 1 surgical procedure or in multiple specimens from multiple procedures. This is more likely in cases of large areas of involvement.

The mean or median extent of DCIS is 1.4 to 2.7 cm8,9,18,21 but ranges from 0.1 cm to extensive involvement of all 4 quadrants of the breast. Although a precise measurement is often not possible, an estimate of the extent of DCIS is clinically important (Table 1).

There are multiple methods for estimating the extent of DCIS (see Figure):

• DCIS in 1 block: The area involved by DCIS can be measured from a single slide, if DCIS is present in only 1 block. If separate foci are present, the largest distance between foci should be reported. This method will underestimate the extent of DCIS when multiple blocks are involved and should not be used in such cases.8 

• Serial sequential sampling: The entire specimen is blocked out in such a way that the location of each block can be determined. The extent of the DCIS can be calculated by using a diagram of the specimen, the thickness of the slices, and the location of the involved blocks.7–9 This method is recommended for all excisions likely to harbor DCIS or with previously diagnosed DCIS (eg, by diagnosis on a prior core needle biopsy).

• Nonsequential sampling: The number of blocks involved by DCIS is correlated with the extent of DCIS of up to 4 cm.8 Multiplying the number of blocks involved by DCIS by the approximate width of a tissue section gives an estimate of the extent. Two studies have shown that multiplying by 0.3 cm underestimates the extent of DCIS, and multiplying by 0.5 cm may overestimate the extent.8,9 Therefore, multiplying by 0.4 cm is recommended, unless there is additional information suggesting that a different number would yield a more accurate result. This method may underestimate extent if not all areas of DCIS are sampled. Therefore, it is recommended that all tissue likely to be involved by DCIS be sampled (eg, all grossly abnormal tissue and all tissue with radiologically suspicious calcifications). When feasible, the entire specimen should be examined microscopically.

This method may result in a larger estimation of extent than with the serial sequential sampling method when DCIS is present in a large volume of tissue (in 3 dimensions) rather than in a predominantly linear distribution. The best estimate for correlation with outcomes (eg, residual disease or recurrence) will require further studies.

This method can be applied to any specimen and will give a better estimation of extent than measuring extent on a single slide when multiple blocks contain DCIS.

• Margins: If DCIS involves or is close to 2 opposing margins, the distance between the margins can be used as the extent of the DCIS within the specimen.

• Gross lesions: In some cases of high-grade DCIS, there may be a gross lesion that can be measured. Confirmation of the gross size must be confirmed by microscopic evaluation.

The largest estimate obtained by using any of these methods should be used to report the estimated size (extent) of the DCIS.

This protocol applies only to cases of DCIS. The protocol for invasive carcinoma of the breast1 applies if invasion or microinvasion (≤0.1 cm) is present. This protocol should not be used for classic/typical lobular carcinoma in situ (LCIS) (eg, extent and margins are not generally reported), but can be used for rare difficult-to-classify cases of carcinoma in situ with features of both DCIS and LCIS (eg, architectural patterns of both DCIS and LCIS or cytologic features of LCIS (without expression of E-cadherin, but with high-grade nuclei and/ or necrosis). In some cases, clinicians may choose to treat such cases as DCIS.

When DCIS involves nipple skin, the tumor cells may disrupt epithelial tight junctions, resulting in seepage of extracellular fluid and formation of the scale crust recognized clinically as Paget disease of the nipple. If there is no associated invasive carcinoma, the cancer is classified as DCIS (ie, Tis [Paget]). Most of these cases are strongly positive for HER-2.

In some cases, immunohistochemical studies for myoepithelial cells may be helpful to confirm a diagnosis of DCIS and to exclude invasion.6,29 

The architectural pattern has been reported traditionally for DCIS.3–5 However, nuclear grade and the presence of necrosis are more predictive of clinical outcome.

The nuclear grade of DCIS is determined using 6 morphologic features (Table 2).4,30 

The presence of necrosis is correlated with the finding of mammographic calcifications (ie, most areas of necrosis will calcify). DCIS that presents as mammographic calcifications often recurs as calcifications. Necrosis can be classified as follows.

• Central (“comedo”): The central portion of an involved ductal space is replaced by an area of expansive dirty necrosis that is easily detected at low magnification. Ghost cells and karyorrhectic debris are generally present. Although central necrosis is generally associated with high-grade nuclei (ie, comedo DCIS), it can also occur with DCIS of low or intermediate nuclear grade. This type of necrosis often correlates with a linear and/ or branching pattern of calcifications on mammography.

• Focal (punctate): Small foci, indistinct at low magnification, or single-cell necrosis.

Necrosis should be distinguished from secretory material, which can also be associated with calcifications, cytoplasmic blebs, and histiocytes, but does not include nuclear debris.

Whenever feasible, the specimen should be oriented to identify specific margins.

Margins may be identified by sutures or clips placed on the specimen surface or by other means of communication between surgeon and pathologist and should be documented in the pathology report. Margins can be identified microscopically in several ways, including by using multiple colored inks, by submitting the margins in specific cassettes, or by the surgeon submitting each margin as a separately excised specimen. Inks should be applied to the surface of the specimen, taking care to avoid penetration into the specimen.

If margins are sampled with perpendicular sections, the pathologist should report the distance from the DCIS to the closest margin, when possible. Because of the growth pattern of DCIS in the ductal system, a negative but close margin does not ensure the absence of DCIS in the adjacent tissue.

A positive margin requires ink on DCIS. If the specimen is oriented, the specific site(s) of involvement (eg, superior margin) should also be reported.

The deep margin may be at muscle fascia. If so, the likelihood of additional breast tissue beyond this margin (and therefore possible involvement by DCIS) is extremely small. A deep muscle fascial margin (eg, on a mastectomy specimen) is unlikely to have clinical significance.

A superficial (generally anterior) margin may be immediately below the skin, and there may not be additional breast tissue beyond this margin. However, some breast tissue can be left in skin flaps, and the likelihood of residual breast tissue is related to the thickness of the flap.31 

Specimen radiography is important to assess the adequacy of excision. Compression of the specimen should be minimized, as it can severely compromise the ability to assess the distance of the DCIS from the surgical margin.27 Mechanical compression devices should be used with caution and preferably reserved for nonpalpable lesions that require this technique for imaging (eg, microcalcifications).

If DCIS is present at the margin, the extent of margin involvement is associated with the likelihood of residual disease19,20:

• Focal (eg, DCIS at the margin in a <0.1 cm area in 1 block)

• Minimal/moderate (between focal and extensive)

• Extensive (eg, DCIS at the margin in an area ≥1.5 cm or in 5 or more low-power fields and/or in 8 or more blocks)

Patients may be treated with endocrine therapy, chemotherapy, or HER-2–targeted therapy before surgical excision, either as part of a protocol or during treatment of a contralateral carcinoma. After treatment of women with invasive carcinoma, it has been observed that the invasive carcinoma may respond to a greater extent than the accompanying DCIS. The histologic changes occurring in DCIS after treatment have not been well described and will probably vary with the specific agents used. Comparison to a pretreatment specimen is necessary to attribute histologic changes to the effects of treatment. The significance of histologic changes in DCIS is unknown.

If the patient had invasive carcinoma before treatment, but only DCIS after treatment, additional classification systems are available to evaluate residual carcinoma in the breast and lymph nodes.32 

Reporting Lymph Nodes.—The pathology report should state the total number of lymph nodes examined (including the number of sentinel nodes), the number of nodes with metastases, and the greatest dimension of the largest metastatic focus. If a patient has at least 1 macrometastasis, only nodes with micrometastases and macrometastases are included for the total number of involved nodes for N classification.33 Nodes with isolated tumor cells are not included in this count.

One section from grossly involved nodes may be examined. All other lymph nodes should be thinly sectioned and entirely submitted for microscopic evaluation. A single H&E section from each lymph node block is considered sufficient for routine evaluation. If additional studies are performed, these should be documented (ie, additional H&E levels or immunohistochemical studies). The presence of extranodal tumor extension should be included in the pathology report because it may be associated with a higher frequency of axillary recurrence.

If lymph node sampling has not been performed or if information about a prior lymph node sampling is unavailable, “X” is used rather than a number in the N designation. A pN classification requires removal of lymph nodes with pathologic examination.

The classification is based on axillary lymph node dissection with or without sentinel lymph node dissection. Classification based solely on sentinel lymph node dissection without subsequent axillary dissection is designated “(sn)” for sentinel node, eg, pN0(i+)(sn).

Isolated tumor cells (ITCs) are defined as single tumor cells or small cell clusters not greater than 0.2 mm.34–36 They may be detected by routine histologic examination or by immunohistochemical or molecular methods. ITCs do not usually show evidence of malignant activity (eg, proliferation or stromal reaction). Micrometastases may show histologic evidence of malignant activity (eg, proliferation or stromal reaction).

Almost all tumor cells present in lymph nodes of patients with DCIS are isolated tumor cells. Isolated tumor cells detected in cases of DCIS have not been shown to have prognostic importance.15,16 If a larger metastasis is found, additional tissue sampling and review of slides are helpful to determine if an area of invasion is present.14 

The TNM staging system for carcinoma of the breast of the American Joint Committee on Cancer (AJCC) and the International Union Against Cancer (UICC) is recommended.33,37 Although the pathologist provides information, based on examination of the surgical specimen, about the individual T and N categories, the treating physician has the responsibility for grouping the TNM categories into a stage of disease.

By AJCC/UICC convention, the designation “T” refers to a primary tumor that has not been previously treated and pM implies microscopic examination of distant lesions. Clinical classification (cTNM) is usually carried out by the referring physician before treatment during initial evaluation of the patient or when pathologic classification is not possible.

Pathologic staging is usually performed after surgical resection of the primary tumor. Pathologic staging depends on pathologic documentation of the anatomic extent of disease and whether or not the primary tumor has been completely removed. If a biopsied tumor is not resected for any reason (eg, when technically not feasible), but the highest T and N categories or the M1 category of the tumor can be confirmed microscopically, then the criteria for pathologic classification and staging have been satisfied without total removal of the primary cancer.

The symbol “p” refers to the pathologic classification of the TNM, as opposed to the clinical classification, and is based on gross and microscopic examination. pT entails a resection of the primary tumor or biopsy adequate to evaluate the highest pT category. For identification of special cases, the “y” and “r” prefixes are used. Although they do not change the stage grouping, the “y” and “r” prefixes indicate cases needing separate analysis.

ypTNM

The “y” prefix is used for those cases in which classification is performed during or following initial therapy before surgical removal of the tumor (ie, neoadjuvant chemotherapy, radiation therapy, endocrine therapy, or combinations of these treatments). The ypT categorizes the extent of tumor actually present at the time of the examination. The “y” categorization is not an estimate of tumor before therapy (ie, before initiation of neoadjuvant therapy).

rpTNM

The “r” prefix indicates a recurrent tumor when staged after a documented disease-free interval.

In general, the presence of distant metastases are assessed using clinical methods or by other types of biopsies (eg, liver biopsies, lung biopsies). Therefore, no M classification should be provided when only a breast specimen and regional nodes are evaluated.

In some cases, other pathologic findings are important for the clinical management of patients.

If the biopsy was performed for a benign lesion and the DCIS is an incidental finding, this should be documented. An example would be the finding of DCIS in an excision for a palpable fibroadenoma.

Peritumoral vascular invasion is a very rare finding in association with DCIS alone. Additional sampling should be considered to attempt to identify an area of invasion. If there has been prior surgery or needle biopsy, the possibility of artifactual displacement of epithelial cells into lymphatics should be considered. Lymph node biopsy may be performed in patients with DCIS and lymphovascular invasion.

If there has been a prior core needle biopsy or incisional biopsy, the biopsy site should be sampled and documented in the report. If the intention was to completely re-excise a prior surgical site, the report should document biopsy changes at the margin that could indicate an incomplete excision. This protocol should only be used for reexcisions that reveal the largest extent of DCIS.

The hormone receptor status of DCIS may be evaluated for multiple reasons. The primary use of this information is to determine if patients with DCIS would benefit from hormonal therapy.

Two studies have addressed outcomes for patients with DCIS who underwent hormonal therapy, and both studies showed that fewer women in the tamoxifen-treated group developed subsequent breast cancers: 18% versus 14% in the UK/ANZ study38 and 13.4% versus 8.2% in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B24 study.39 However, this result was only statistically significant in the NSABP study. It is possible that the younger age of the patients in this study could have influenced these results, as a smaller benefit was observed in women older than 50 years. There was no benefit for survival in either study.

A subsequent analysis of estrogen receptor (ER) status for DCIS in a subset of patients in the NSABP trial showed that the reduction in subsequent breast cancers was greatest for women with ER-positive DCIS treated with tamoxifen.40 Little benefit was found in women with ER-negative DCIS, but because of the small number of events, a small clinically significant benefit was not excluded.

The Update Committee of the American Society of Clinical Oncology concluded that current data are insufficient to make a general recommendation for the use of ER status of DCIS for making decisions about tamoxifen treatment.41 National Comprehensive Cancer Network practice guidelines include determination of ER status as part of the workup of DCIS.42 Although a progesterone receptor (PR) test is often ordered in conjunction with an ER test, there are almost no data on the association of PR status and DCIS. As a result, many institutions do not assess PR status for cases of DCIS. In addition to considerations for hormonal treatment, information about hormone receptor and HER-2 status in DCIS may be useful for other reasons in some settings. As in invasive carcinoma, these markers identify different types of DCIS, including “ER-positive,” “HER-2–positive,” and “triple-negative” cancers.43,44 For invasive carcinomas, these immunoprofiles identify groups with different prognoses and response to different types of treatments. The usefulness of these markers for determining outcome or response to treatment for DCIS is under investigation. Some ongoing treatment protocols require marker information in DCIS for eligibility. In addition, recurrent carcinomas, in general, have the same immunoprofile as the prior DCIS.45–47 Therefore, this information may be helpful for some patients and clinicians in making decisions about local treatment that could affect the likelihood of such a recurrence.

Because marker status in DCIS is used primarily for clinical purposes and not for diagnosis (except in rare cases to help distinguish Toker cells in nipple skin from the cells of an underlying DCIS resulting in Paget disease), the decision to perform tests for these markers should be made in conjunction with the clinicians who will use this information.

The results of hormone receptor stains performed on a prior core needle biopsy can be included in the checklist for an excisional specimen. If the result of the study on the core needle biopsy is negative, additional studies on a larger area of DCIS in the excisional biopsy should be considered.

Most (75% to 80%) cases of DCIS are ER positive, with strong immunoreactivity in most cells (Table 3). Almost all cases of ER-negative DCIS are of high nuclear grade, and many are associated with necrosis. In some cases, it may be difficult to distinguish rare positive tumor cells from intermingled normal epithelial cells.

In addition to the interpretation, each pathology report should specify the type of fixation and processing (eg, formalin-fixed, paraffin-embedded sections), the antibody clone, the general form of detection system used, and the scoring system used (see College of American Pathologists' Laboratory Accreditation Program, Anatomic Pathology Checklist [questions related to reporting of results only, ANP.22988]48). Any deviation from the laboratory's standard processing and antigen retrieval protocol should be included. Appropriate positive and negative controls should be used and documented. If normal breast epithelial cells are not immunoreactive for estrogen receptor, the test should be repeated and, if confirmed, this result should be noted. Complete absence of ER positivity for DCIS and normal breast epithelial cells may be due to failure of the assay (eg, failure to apply primary antibody, expired reagent) or loss of immunogenicity of the tissue.

Cancer found in biopsies performed for microcalcifications will almost always be at the site of the calcifications or in close proximity.2,3,5 The presence of the targeted calcifications in the specimen should be confirmed by specimen radiography. The pathologist must be satisfied that the specimen has been sampled in such a way that the lesion responsible for the calcifications has been examined microscopically. The relationship of the radiologic calcifications to the DCIS should be indicated.

If calcifications are present in the specimen radiograph but not in the initial histologic sections, deeper levels should be examined. If needed, radiographs of the paraffin block(s) may be obtained to detect calcifications remaining in the block(s). If microcalcifications cannot be confirmed by routine microscopic evaluation, polarized light may be helpful to identify calcium oxalate crystals that will polarize, but are unstained in H&E sections. On rare occasions, calcifications do not survive tissue processing or prolonged fixation in formalin. Foreign material can sometimes simulate calcifications (eg, metallic fragments after surgery or trauma).

It is a Joint Committee requirement that clinical information be provided for pathology specimens. Relevant clinical information is often necessary for the accurate evaluation of breast specimens and includes:

• Family history of breast or ovarian cancer and/or BRCA1 or BRCA2 mutation

• Current pregnancy or lactation

• Prior breast biopsy or surgery (including implants)

• Prior breast cancer diagnosis (type, location in breast, date of diagnosis)

• Prior treatment that could affect the breast:

  • Radiation

  • Chemotherapy

  • Hormonal therapy (eg, tamoxifen, aromatase inhibitors, or oral contraceptives)

  • Systemic diseases that may affect the breast (eg, collagen vascular disease, sarcoidosis, Wegener granulomatosis)

• Type of lesion sampled:

  • Palpable mass

  • Nipple discharge

  • Nipple lesion (eg, scaling crust)

  • Imaging finding

    • Mammographic or ultrasound mass: shape of mass (irregular, circumscribed, ill defined, cystic or solid)

    • Mammographic calcifications

    • Mammographic architectural distortion

    • Prior core needle biopsy site, with or without a clip, with or without residual radiologic or clinical lesion

    • MRI-detected lesion

• Type of specimen:

  • Excision without wire localization

  • Excision with wire localization—for these specimens, the specimen radiograph with an interpretative note should be made available to the pathologist

  • Nipple duct excision

  • Total mastectomy

  • Lymph node specimen (sentinel node, nonsentinel node, limited axillary dissection, complete axillary dissection)