Paget disease of the breast (PDB)

INTRODUCTION — In 1874, Sir James Paget described 15 women with chronic nipple ulceration who all went on to develop cancer of the involved breast within two years [1]. The ulceration was described as an eczema-like eruption on the nipple and areola with a copious clear yellowish exudate. Ultimately, this would become known as Paget disease of the breast (PDB) or mammary Paget disease. While Paget believed that the nipple changes were themselves benign, it was subsequently discovered that the characteristic cells within the epidermis of the nipple (Paget cells) were in fact malignant [2].

This topic review will discuss the clinical presentation, pathogenesis, pathology, diagnosis, and management of mammary Paget disease.

EPIDEMIOLOGY — PDB is much less common than other presentations of breast cancer, accounting for only 1 to 3 percent of new cases of female breast cancer that are diagnosed annually in the United States [3-7]. PDB can occur in men; however, this is an extremely rare finding. Although cases have been described in patients ranging in age from 26 to 88, the peak incidence is between 50 and 60 [8-11].

Histologic (subclinical) evidence of PDB may be more frequent. In a series of 3000 consecutive mastectomy specimens, the incidence of clinical Paget disease was 0.7 percent, but histologic evidence of the disease was present in 4.9 percent of the mastectomies [12].

At least some epidemiologic data suggest that the incidence of PDB is decreasing over time [13]. In a report of 1738 cases of PDB reported to the Surveillance, Epidemiology, and End Results (SEER) database, the incidence of PDB decreased by 45 percent between 1988 and 2002, while the overall incidence of breast cancer increased over this same time period.

Although PDB is a less common presentation of breast cancer than a palpable mass or mammographic abnormality, it is an important consideration in the differential diagnosis of a chronic persistent abnormality of the nipple.

CLINICAL PRESENTATION — The hallmark of PDB is a scaly, raw, vesicular, or ulcerated lesion that begins on the nipple and then spreads to the areola (picture 1). Occasionally, a bloody discharge is present. PDB is usually unilateral, although bilateral cases have been described [14]. Nipple retraction is also a rare finding but may occur with more advanced disease.

Pain, burning, and/or pruritus are commonly associated with PDB and may be present before the development of clinically apparent disease. Thus, any patient complaining of these symptoms but without an obvious lesion should be followed closely if the symptoms persist. Despite the superficial nature of the findings, the median duration of signs and symptoms prior to histologic diagnosis is between six and eight months [8,9].

DIAGNOSTIC WORKUP — The diagnostic workup of suspected PDB focuses on both establishing the diagnosis and identifying an underlying breast cancer. An underlying breast cancer (in situ or invasive) is present in 85 to 88 percent of cases, although often without an associated breast mass or mammographic abnormality [13].

●A palpable breast mass is present in 50 percent of cases of PDB; the mass is often located more than 2 cm from the nipple-areolar complex [3,5].

●In 20 percent of cases, a mammographic abnormality is present without a palpable mass [15].

●Approximately 12 to 15 percent of cases are not associated with a palpable mass, mammographic abnormality, or parenchymal breast cancer [13].

●In 25 percent of cases, there is neither an underlying mass nor a mammographic abnormality, but an occult ductal carcinoma in situ (DCIS) is present.

Patients with either a palpable mass or a mammographic abnormality are more likely to have an underlying invasive cancer, while DCIS is found more often in the absence of a palpable mass or mammographic abnormality [16].

History and physical examination — A detailed history should document the length of time the lesion has been present and any associated symptoms, including pain, nipple discharge, bleeding, burning, and pruritus. The history should also cover the patient's individual risk profile for breast cancer, including any risk factors for hereditary breast cancer. (See "Factors that modify breast cancer risk in women" and "Overview of hereditary breast and ovarian cancer syndromes".)

Bilateral breast examination should be performed, focusing on associated breast abnormalities. An underlying breast malignancy is present in 85 to 88 percent of cases, and a palpable mass is present in approximately one-half. (See "Clinical manifestations, differential diagnosis, and clinical evaluation of a palpable breast mass", section on 'Physical examination'.)

Differential diagnosis — The differential diagnosis includes both benign etiologies (eczema, contact as well as radiation dermatitis, nipple adenoma) and malignant conditions (Bowen's disease [squamous carcinoma of the epidermis], basal cell carcinoma, superficial spreading malignant melanoma) [17-21].

Given the differential diagnosis, a short course of topical steroids is often considered if eczema or dermatitis is suspected. However, partial healing or transient improvement in the eczematous changes may occur even in the presence of PDB. In order to avoid undue delay in definitive diagnosis, any persistent nipple abnormality beyond four to six weeks should be considered an indication for biopsy and/or breast imaging.

Skin biopsy and histology — Nipple scrape cytology can accurately diagnose PDB [22,23], but the diagnosis is usually established by full-thickness punch or wedge biopsy of the nipple. Punch biopsy is more widely used. Wedge biopsy may contain lactiferous ducts, which may identify DCIS as the etiology of PDB. (See "Skin biopsy techniques".)

Paget cells — The histologic hallmark of PDB is the presence of malignant, intraepithelial adenocarcinoma cells (Paget cells) occurring singly or in small groups within the epidermis of the nipple (picture 2). The cells are often large but may be similar in size to keratinocytes. The cytoplasm is pale to clear, and nuclei are usually high-grade with prominent nucleoli. In many instances, the Paget cells retract from the surrounding keratinocytes and appear to be within a vacuole.

Immunohistochemistry — A useful immunohistochemistry (IHC) diagnostic panel for a nipple biopsy without underlying invasive cancer or DCIS may include CK7, carcinoembryonic antigen (CEA; polyclonal), p63, ER, HER2, and either SOX10, MART-1(MelanA), or HMB-45 if melanoma is also in the biopsy differential diagnosis [24].

Histologically, PDB can mimic malignant melanoma, particularly if the cells have incorporated melanin from the adjacent epidermis. The presence of cytoplasmic mucin vacuoles can aid in the diagnosis, as can special stains such as mucicarmine, which may highlight the vacuoles. IHC may also be helpful in diagnostically difficult cases (table 1) [25]. PDB can be distinguished from malignant melanoma by positive IHC for CEA (polyclonal) and negative IHC for S-100 protein; however, some cases of PDB are S-100 positive [26], and CEA may not always be expressed. MART-1, SOX10, or HMB-45 may be more useful than S-100 in confirming malignant melanoma.

Paget cells are positive for low molecular weight cytokeratins, a feature that can help distinguish PDB from squamous carcinoma of the epidermis (Bowen's disease), which typically expresses high molecular weight cytokeratins and p63 [27,28]. CK7 is probably the most useful marker, but caution must be exercised in interpreting CK7 stains in the absence of other typical histologic features of PDB. Although positive in PDB, CK7 may also be expressed in benign Toker cell hyperplasia [29,30], and some breast carcinomas may not express CK7 [31]. Toker cells in the nipple may be derived from or related to lactiferous duct epithelium, and malignant transformation of these cells may explain some cases of PDB without associated ductal carcinoma [32].

Hormone receptors — If present, estrogen and/or progesterone receptor positivity is extremely helpful. Unfortunately, approximately one-half of PDB cases do not express hormone receptors, a finding that is not surprising considering the frequent association with high-grade invasive ductal carcinomas that are also likely to be hormone receptor-negative and/or HER2 positive.

Between 84 and 91 percent of cases of PDB overexpress HER2. HER2 overexpression in PDB likely explains the worse prognosis reported for invasive breast cancer with Paget disease [33]. (See "HER2 and predicting response to therapy in breast cancer".)

Mammography and ultrasonography — Bilateral mammography is mandatory to identify an associated mass as well as exclude synchronous cancers or widespread calcifications that might preclude breast conservation therapy. Approximately one-half of cases of PDB have an associated mammographic abnormality [15,34]. The mammographic findings in 17 women in one study included suspicious microcalcifications, a mass, architectural distortion, nipple retraction, and asymmetric thickening of the nipple-areolar complex [15]. In seven cases, the mammographic abnormalities were located distant from the nipple-areolar complex.

As noted above, while women with a negative mammogram in the setting of PDB appear to have a low likelihood of underlying invasive breast cancer (particularly in the absence of a palpable mass [16,35]), an occult associated intraductal carcinoma may be present and quite extensive. In the above noted series, all of the 17 women with negative mammograms had DCIS, five with extensive multicentric disease [15]. In another series, mammography failed to identify 64 percent of cases of multifocal underlying breast cancer (both in situ and invasive cancer) in women with PDB and no palpable mass [36]. (See "Breast imaging for cancer screening: Mammography and ultrasonography".)

Ultrasound should be used to further evaluate and guide core biopsy of any palpable mass or mass-like mammographic abnormality. Whole breast ultrasound in addition to mammography has been examined in the context of Paget disease but does not appear to increase sensitivity over mammography alone [37-39].

Magnetic resonance imaging — Magnetic resonance imaging (MRI) is a sensitive imaging tool for invasive breast cancer, particularly for mammographically occult disease, and its sensitivity for detecting DCIS, particularly high-grade DCIS, is also high. MRI may be most useful for women at high risk for breast cancer. (See "Breast ductal carcinoma in situ: Epidemiology, clinical manifestations, and diagnosis", section on 'Magnetic resonance imaging' and "Clinical features, diagnosis, and staging of newly diagnosed breast cancer".)

There are few data addressing the utility of breast MRI in women with PDB and a negative mammogram [40-46]. The largest series included 34 women with PDB, 32 of whom had a histologically confirmed breast cancer (7 invasive, 25 DCIS) [40]. All patients underwent preoperative mammography, while 13 had an ipsilateral breast MRI. Among the 23 women with biopsy-proven PDB and negative mammography, eight underwent MRIs that detected otherwise occult disease in four women (three DCIS, one invasive cancer), three of whom required mastectomy because of extensive disease. However, MRI did not detect three cancers (two unifocal DCIS, one invasive cancer) in the other four patients.

Thus, MRI may disclose occult cancer in some women with PDB and no findings on mammography or physical examination, potentially allowing directed treatment of the ipsilateral breast. However, a negative preoperative study cannot reliably exclude an underlying cancer.

In addition, because MRI is highly sensitive but not highly specific, MRI may reveal abnormalities that might lead to the overuse of mastectomy rather than breast-conserving therapy. Thus, if MRI is to be performed, it should only be undertaken at an institution that has the capability of performing MRI-guided biopsies, and the patient should be counseled regarding the high false positive rate of breast MRI and the possible need for additional biopsies.

Biopsy of underlying abnormalities — In addition to nipple biopsy, any underlying masses or mammographic abnormalities must be biopsied to assist in decisions about local treatment of the breast and the need for evaluation of the axilla.

PATHOGENESIS — Two theories have been proposed to explain the pathogenesis of PDB: the epidermotropic theory, which is by far the more widely accepted, and the transformation theory.

Epidermotropic theory — According to the prevailing epidermotropic theory, the Paget cell arises from an underlying mammary adenocarcinoma, with the neoplastic ductal epithelial cells migrating through the ductal system of the breast into the epidermis of the nipple.

In many series, immunohistochemical (IHC) staining is concordant in Paget cells and the ductal epithelial cells but discordant in Paget cells and the epidermal keratinocytes of the surrounding nipple tissue (table 1) [9,47-52].

Similarly, expression of several molecular markers is concordant in PDB and underlying parenchymal breast tumors [47,53]. Over 80 percent of PDBs have overexpression or amplification of the gene for HER2 [17,18,53,54]. In one report, all 23 cases of PDB stained with an anti-HER2 monoclonal antibody; staining was also positive in all the underlying ductal carcinomas but never in normal nipple epidermis [8]. It is hypothesized that the spread of Paget cells to the nipple epidermis from the duct system may be mediated through a motility factor that exerts its effect via the HER2 receptor (figure 1) [55].

Taken together, these data lend support to a common genetic alteration and/or possibly a common progenitor cell for both the Paget cells and the underlying ductal carcinoma.

Transformation theory — In contrast, the transformation theory proposes that epidermal keratinocytes within the nipple transform into malignant Paget cells and that PDB in fact represents an epidermal carcinoma in situ that is independent of any underlying ductal carcinoma [19]. George Thin proposed this theory in 1881, believing that secretions from the breast ducts damaged the epithelium, transforming normal keratinocytes into cancer cells [2].

The transformation theory was bolstered by the following observations:

●In a small percentage of cases, no parenchymal cancer can be identified in association with PDB.

●When an underlying parenchymal carcinoma is present, it is often located peripheral to the nipple, suggesting two independent neoplastic processes. As an example, in one series of 80 women with breast cancer and PDB, 29 of the underlying breast tumors were located more than 2 cm from the areolar margin [10].

●Desmosomal attachments have been identified between Paget cells and adjacent epidermal keratinocytes [56]. These authors describe a "pre-Paget cell" with an appearance intermediate between that of a keratinocyte and a Paget cell, suggesting that epidermal cells can acquire the characteristics of ductal cells as they undergo malignant transformation. Others suggest the presence of a unique precursor, the Toker cell, to explain cases of PDB with no contiguous ductal carcinoma [32,57].

The transformation theory has fallen out of favor, in part because if enough tissue sections are studied, involvement of the large (lactiferous) ducts immediately beneath the nipple can usually be demonstrated.

STAGING — The presence of Paget disease of the breast does not change the stage of an underlying breast cancer. If an associated invasive breast cancer or ductal carcinoma in situ is not identified, Paget disease is classified as Tis (Paget) disease. (See "Tumor, node, metastasis (TNM) staging classification for breast cancer".)

TREATMENT — The most important aspect of presentation that influences the treatment decision is the presence or absence of an ipsilateral palpable breast mass or mammographic abnormality. The prognosis of PDB is based upon the underlying breast cancer, and treatment should be guided by the stage of the tumor and other prognostic and/or predictive factors.

Simple mastectomy has been the historic standard treatment for PDB. However, the widespread use of breast-conserving treatment (BCT) for invasive and in situ breast carcinoma has led to its use for PDB. In contrast to invasive breast cancer, treatment recommendations are limited by the lack of prospective randomized trials comparing mastectomy to BCT or comparing various BCT options in patients with PDB. Most reported series are small, and patient selection, treatment techniques, and median follow-up vary from study to study. (See "Breast-conserving therapy" and "Mastectomy" and "Ductal carcinoma in situ: Treatment and prognosis".)

PDB with a palpable mass or abnormal imaging — When PDB is present in association with a palpable mass or defined mammographic abnormality, the associated breast cancer tends to be a more advanced stage (ie, larger size, multifocal, more frequent presence of invasive disease and axillary node positivity) than if a mass is absent [10,16,36]. The incidence of axillary metastases and survival in women who present with PDB with and without a palpable mass is shown here (table 2).

Both the nipple-areolar complex and the underlying cancer must be excised. Many patients will require a mastectomy. However, if nipple-areolar resection and wide local excision of the palpable mass or area of mammographic abnormality can be performed with an acceptable cosmetic result and negative margins, breast-conserving surgery followed by whole breast radiotherapy (RT) is an appropriate local treatment option. If the breasts are large, resection of the palpable mass and nipple-areolar complex with nipple reconstruction and elective contralateral breast reduction can be combined with definitive breast RT to achieve an adequate oncologic outcome with acceptable breast contour and symmetry.

Women with multicentric cancer or diffuse calcifications should be treated by mastectomy with axillary evaluation, for example by sentinel lymph node biopsy (SLNB). Many women with PDB and a palpable mass are not amenable to BCT because of the distance between the primary tumor and the nipple-areolar complex [10]. If resection of both the nipple-areolar complex and the palpable mass with a negative margin would result in an unacceptable cosmetic result, women may elect for simple mastectomy with or without breast reconstruction. (See "Breast-conserving therapy", section on 'Patient selection for BCT' and "Mastectomy", section on 'Indications for mastectomy'.)

PDB with no palpable mass or imaging abnormality — Despite the absence of a palpable mass or mammographic abnormality, underlying carcinoma is present in most patients with PDB [36]. The majority will have DCIS, but invasive cancer is present in one-fourth to one-third of cases [3,5,58]. Standard treatment options include mastectomy or breast-conserving therapy (resection of the nipple-areolar complex followed by whole breast irradiation) (picture 3).

Simple mastectomy with or without breast reconstruction is an effective option [4,59]. In a series of 37 patients treated by simple mastectomy alone, the local recurrence rate was only 5 percent, with a median follow-up of 40 months [4].

Several reports describe the use of limited surgery or nipple-areolar resection followed by breast RT for treatment of PDB in women without a palpable breast mass (table 3) [8,35,60-63]. Rates of local recurrence, successful salvage, and distant disease-free survival are comparable to those seen with DCIS [64]. The following represents the range of findings:

●A study of 36 patients with PDB and no underlying palpable mass or mammographic abnormality who underwent complete or partial resection of the nipple-areolar complex followed by RT demonstrated a local recurrence rate of 11 percent at a median follow-up of 112 months [8]. Of the 22 patients undergoing complete nipple-areolar resection followed by whole breast RT with a boost (total dose 61.5 Gy), three (14 percent) developed a local recurrence. In contrast, two of six women (33 percent) treated with the same dose of RT but only a partial nipple-areolar resection recurred locally, one with distant metastases. All four women with an isolated local recurrence were treated successfully with salvage mastectomy and remained disease free at the time of last follow-up.

●The European Organization for Research and Treatment of Cancer (EORTC) conducted a prospective study of BCT for PDB [35]. This study excluded women with invasive cancer. Of the 61 women enrolled in the study with histologically proven PDB, 97 percent had no associated palpable mass, and 84 percent had a normal mammogram. Associated DCIS was present in 93 percent, and the remainder had no underlying parenchymal carcinoma. Treatment consisted of complete nipple-areolar resection followed by whole breast RT (median dose 50 Gy). There were four local recurrences (6.5 percent), with a median follow-up of 6.4 years. Three recurrences were invasive, and one was DCIS; one patient with an invasive recurrence subsequently died of metastatic disease.

●Several subsequent retrospective studies comparing BCT with mastectomy showed that the type of surgery did not demonstrate a difference in outcomes (eg, survival, recurrence) of patients with PDB [13,65-68]. The long-term outcomes in PDB are primarily determined by the associated invasive cancer or DCIS. (See 'Prognosis' below.)

Taken together, these data suggest that complete resection of the nipple-areolar complex with central lumpectomy, as clinically appropriate, followed by whole breast RT is a reasonable alternative to mastectomy for women with PDB and no palpable mass or mammographic abnormality provided good cosmetic outcome and negative margins can be achieved. A variety of techniques can be used to reconstruct the nipple-areolar complex, and cosmesis has been assessed as good to excellent in the majority of patients so treated. (See "Oncoplastic breast surgery".)

In some cases of PDB without a mass or mammographic abnormality, the pathologic evaluation of the nipple/areolar complex may detect secondary invasion into the subjacent dermis. In one series of six cases with secondary dermal invasion, there was no underlying breast carcinoma in one case, DCIS only in three cases, DCIS with microinvasion in one case, and invasive carcinoma in one case. Treatment was determined by the underlying disease. One patient died of unrelated causes, and the remainder were alive at last follow-up [69]. In a second series of seven patients, five patients had microinvasive (<1 mm) dermal foci, and the two remaining patients had 2 and 3 mm foci of dermal invasion. Two patients at high risk for breast cancer underwent mastectomy, and the average-risk patients were treated with breast-conserving therapy and RT; no recurrences were reported (median follow-up 20 months [range 4 to 66 months]) [70]. While there is limited information available on patients with secondary dermal invasion, this finding does not indicate more adverse outcomes and should be distinguished from locally advanced breast cancer with skin involvement (T4b).

Management of the axilla — The risk of axillary metastases in series of patients with PDB is higher in women with invasive cancer and a palpable mass (table 2) [9,10,58,59,71,72].

Evaluation and treatment of the axilla in PDB are the same as for any breast cancer and depend on the underlying cancer (see "Overview of management of the regional lymph nodes in breast cancer"):

●Patients with DCIS do not require axillary investigation unless the disease is extensive enough to merit mastectomy. If mastectomy is planned, sentinel lymph node (SLN) biopsy is often (and appropriately) performed preemptively in order to avoid complete axillary lymph node dissection in cases with an invasive component identified at final pathology. (See "Overview of sentinel lymph node biopsy in breast cancer".)

●If invasive disease has been identified, patients with clinically node-negative disease should undergo SLN biopsy at the time of wide excision. An SLN can be successfully identified in 97 percent of patients with PDB [72]. Management of a positive SLN is discussed in detail elsewhere. (See "Overview of sentinel lymph node biopsy in breast cancer".)

●Patients with a clinically positive or suspicious axilla should undergo an initial ultrasound-guided fine needle aspiration (FNA) or core needle biopsy of the palpable axillary nodes; if disease is confirmed histologically, full axillary lymph node dissection (ALND) is recommended. In contrast, if the FNA or core biopsy is negative, SLB biopsy should be performed. SLN biopsy after neoadjuvant chemotherapy may be an option for some highly selected patients with clinically positive axillary nodes. (See "General principles of neoadjuvant management of breast cancer", section on 'Positive axilla prior to treatment'.)

●For patients with pure PDB without an associated mass and a clinically negative axilla, there is some controversy as to the need for axillary evaluation when breast conservation is planned. Given the high incidence of DCIS alone in these patients, SLN biopsy may not be necessary and can be considered as a second operation if invasive disease is discovered [72]. In a series of 19 patients with PDB only, an invasive component was found in 27 percent and positive SLNs in 11 percent [58]. Based on these numbers, some consider the likelihood of invasive disease high enough to recommend an SLN biopsy in this situation routinely. (See "Overview of sentinel lymph node biopsy in breast cancer".)

Alternative approaches — Mastectomy and BCT (surgery plus whole breast RT) represent standard approaches to treatment of PDB without an underlying mass or mammographic abnormality. Less aggressive approaches have been examined, including ipsilateral breast RT following only a diagnostic nipple biopsy [60-63] and breast-conserving surgery without RT (table 4) [4,59,73,74]; however, only small numbers of patients have been studied, and reported results are variable. Thus, neither of these approaches can be considered a standard treatment option for most patients with PDB in the absence of an underlying mass.

Few areas in breast cancer treatment are as contentious as the obligate need for RT as a component of BCT in women with DCIS. Proponents of RT for all women with DCIS argue that local recurrence rates are substantially higher when RT is withheld and that one-half of all local recurrences are invasive. Others contend that RT represents overly aggressive therapy for many women with DCIS and that by weighing factors of prognostic importance (ie, grade, size, age, and distance to margins), a sizable subset can be identified who have excellent recurrence-free survival with excision alone. These issues are all discussed in detail elsewhere. (See "Ductal carcinoma in situ: Treatment and prognosis".)

Use of tamoxifen — There are no data addressing the efficacy of endocrine therapy, including tamoxifen, in reducing the risk for local recurrence in patients with PDB without an underlying invasive carcinoma or DCIS who are treated with breast-conserving therapy. Recommendations regarding endocrine therapy (tamoxifen or aromatase inhibitors) as well as other forms of adjuvant systemic therapy such as chemotherapy and trastuzumab should be based solely upon the characteristics of any associated invasive carcinoma or DCIS. (See "Adjuvant endocrine and targeted therapy for postmenopausal women with hormone receptor-positive breast cancer" and "Selection and administration of adjuvant chemotherapy for HER2-negative breast cancer" and "Ductal carcinoma in situ: Treatment and prognosis", section on 'Systemic treatment'.)

PROGNOSIS — The prognosis of PDB is dependent upon the presence of an underlying invasive ductal carcinoma or axillary node metastases. Tumor stage is a better indicator of prognosis than the presence of Paget disease per se.

As noted above, PDB presenting with a palpable mass is usually associated with more advanced disease than cases without a palpable mass; because of this, five-year survival rates are also lower (table 5) [9,10,59,71]. In general, for women with PDB associated with a palpable mass, five-year survival ranges from 20 to 60 percent, while for women without a palpable mass, survival ranges from 75 to 100 percent [10,19,59,71].

For patients with a palpable mass and invasive ductal carcinoma, five-year cause-specific survival ranges from 37 to 43 percent, while in patients without a mass and with ductal carcinoma in situ (DCIS) only, the five-year cause-specific survival is 90 to 100 percent [3-5,8].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Breast cancer".)

SUMMARY AND RECOMMENDATIONS

●Definition – Paget disease of the breast (PDB) is clinically characterized by a persistent scaling, eczematous, or ulcerated lesion involving the nipple-areolar complex. The pathologic hallmark and defining feature is the presence of malignant, intraepithelial adenocarcinoma cells (Paget cells) within the epidermis of the nipple. (See 'Introduction' above.)

●Clinical presentation – PDB is associated with an underlying breast cancer (in situ [ductal carcinoma in situ; DCIS] and/or invasive) in 85 to 88 percent of cases, often without an associated breast mass or mammographic abnormality. (See 'Clinical presentation' above.)

●Diagnosis – The diagnosis of PDB can usually be established by punch or full-thickness wedge biopsy of the nipple. Biopsy of any underlying mass or mammographic abnormality is required to assist in decisions about local treatment of the breast and the need for evaluation of the axilla. (See 'Diagnostic workup' above.)

●Diagnostic evaluation – The diagnostic workup of all patients with PDB should include careful physical examination of both breasts and mammography. Breast magnetic resonance imaging (MRI) may be considered in a patient with a negative physical examination and mammogram. All mammographically occult lesions that are identified by MRI do not represent cancer, and confirmatory biopsy is needed. Thus, MRI should only be obtained at institutions with the capability of performing MRI-guided biopsy. (See 'Diagnostic workup' above.)

●Pathogenesis

•Women who have a palpable mass in association with PDB are more likely to harbor invasive breast cancer than those with no palpable mass, and they are also more likely to have involved axillary lymph nodes. (See 'PDB with a palpable mass or abnormal imaging' above.)

•Women with PDB and no palpable underlying mass or radiographic abnormality have a lower likelihood of harboring invasive breast cancer; the majority have underlying DCIS. (See 'PDB with no palpable mass or imaging abnormality' above.)

●Treatment – The prognosis of PDB is based upon the underlying breast cancer, and treatment should be guided by the stage of the tumor and other prognostic and/or predictive factors.

•If an invasive breast cancer is identified, it should be treated like any other invasive breast cancer, with consideration of adjuvant systemic therapy. (See 'Treatment' above.)

•Evaluation and treatment of the axilla in PDB are the same as for any breast cancer and depend on the underlying cancer identified and the planned surgical procedure (breast-conserving surgery or total mastectomy). (See 'Management of the axilla' above.)

•Both mastectomy and breast-conserving surgery followed by whole breast radiotherapy (RT) are acceptable treatment options. Breast-conserving surgery and RT should only be performed when a nipple-areolar resection and wide local excision of the underlying cancer can be performed with an acceptable cosmetic result and negative margins. (See 'Treatment' above.)

•Recommendations regarding the use of endocrine therapy as well as other forms of adjuvant systemic therapy in patients with PDB should be based upon the characteristics of the underlying carcinoma. There are no data to support the use of endocrine therapy in women with PDB without an associated invasive cancer or DCIS. (See 'Use of tamoxifen' above.)