Are You Confident of the Diagnosis?

What you should be alert for in the history

Mammary Paget’s disease (MPD) and extramammary Paget’s disease (EMPD) have insidious onsets and mimic numerous common inflammatory and infectious eruptions, such as eczema, psoriasis, intertrigo, and dermatophyte infection, among others. Patient history plays a valuable role in elucidating the diagnoses. History of a slowly growing plaque unresponsive to a variety of treatment attempts is often a clue. Patients ought to be questioned about any past history of a clinically similar eruption, as this would favor conditions that can come and go, such as a contact dermatitis and psoriasis.

Moreover, patients ought to be questioned about conditions associated with diagnoses in the differential, such as contact allergens (suggesting contact eczema), nail dystrophy and arthritis (suggesting psoriasis), or athlete’s foot (suggesting dermatophyte). For EMPD, pruritus is substantial and proves in many instances to be the reason for the patient’s visit.

For MPD and EMPD, a detailed review of symptoms is a critical part of the history because of MPD and EMPD’s associations with other malignancies. Any suggestion of an associated malignancy should of course be investigated, but also elevates the possibility of MPD and EMPD in one’s diagnosis of a clinically appropriate lesion.

Continue Reading

Characteristic findings on physical examination

MPD and EMPD are typically solitary, slowly-expanding and well-demarcated erythematous plaques, sometimes with scale, crust, and ulceration.

MPD always involves the nipple/areola complex and sometimes its surrounding skin. As most cases represent cutaneous spread of tumor cells from a ductal breast carcinoma, a palpable nodule may be appreciable on breast examination. A palpable nodule suggests invasive ductal carcinoma as opposed to an in situ tumor.

EMPD typically affects areas of apocrine gland distribution. The most commonly reported site for EMPD is a woman’s vulva and represents about 2% of primary vulvar tumors. Other commonly reported areas for EMPD are the penis, scrotum, anal, perianal, axillary, and umbilical skin (Figure 1). Still, there are exceptions to the rules, including multifocal eruptions and unusual locations, including the buttock, thigh, eyelid, and even external auditory canal.

Figure 1.

A 59-year-old Asian man with biopsy-proven EMPD. (Source: Anolik R, Liang C, Wang N, Rosenman K, Pomeranz M, Joe E. Extramammary Paget disease. Dermatol Online J. 2008;14[10]:15.)

Palpation of and around the EMPD lesion can be helpful in identifying possible associated malignancies, such as those of adnexal structures and visceral organs of the urogenital and gastrointestinal systems.

Expected results of diagnostic studies

The histology of MPD and EMPD is particularly helpful in differentiating these lesions from clinically similar skin eruptions. A characteristic feature of these tumors on histology is larger basophilic or amphophilic cells, known as Paget’s cells, scattered singly, in clusters, or in glandular structures throughout the epidermis (Figure 2, Figure 3). This appearance is so characteristic that when other pathologies share the appearance, it is described as having a “pagetoid spread.” Paget’s cells have a finely granular cytoplasm with enlarged, pleomorphic nuclei. Mucin is found in about 40% of MPD tumor cells, but in the vast majority, about 90%, of EMPD tumor cells.

Figure 2.

A histologic example of EMPD at 20x magnification (Courtesy of Shane A. Meehan, MD, Dermatopathology Section, Department of Dermatology, New York University School of Medicine)

Figure 3.

A histologic example of EMPD at 40x magnification (Courtesy of Shane A. Meehan, MD, Dermatopathology Section, Department of Dermatology, New York University School of Medicine)

Imaging studies and sometimes serologic tests are warranted in cases of MPD and EMPD. Notably, these tests are intended to screen for and diagnose possibleassociated malignancies, rather than clarify the skin diagnosis. See systemic implications and complications below for details on imaging and bloodwork.

Diagnosis confirmation

Several clinical and histologic diagnoses make up the differential diagnosis for MPD and EMPD.

Clinically, the differential diagnosis for MPD and EMPD is broad, including conditions that can form solitary erythematous plaques with scale. This includes eczema, psoriasis, seborrheic dermatitis, candidiasis, dermatophytosis, and Bowen’s disease/squamous cell carcinoma. For EMPD, which usually arise in areas of apocrine distribution, intertrigo also fits in the differential.

These diagnoses can be ruled out by history (ie, history of psoriasis, known contact allergens, etc), physical examination for associated features (ie, similar plaques on the elbows and knees for psoriasis), KOH (ie, to rule out fungal processes), empiric treatment for short period of time, and biopsy when necessary. Other less common pathologies also exist in the differential, such as necrolytic migratory erythema and zinc deficiency, but these are usually present as part of a largely evident composite of skin and mucosal changes instead of just a solitary lesion.

Histologically, any lesion that demonstrates the so-called “pagetoid spread” in which single and clustered cells are scattered throughout the epidermis can mimic MPD and EMPD. These conditions include superficial spreading malignant melanoma, squamous cell carcinoma, mycosis fungoides, Langerhans cell histiocytosis, and Spitz nevus. When histologic morphology alone will not differentiate the diagnoses, immunologic markers are very helpful.

Immunologic markers for cytokeratins will usually differentiate Paget’s cells from surrounding epithelial cells. Cam5.2, for example, are antibodies that bind to low-molecular-weight cytokeratins of MPD and EMPD. Adjacent epidermal cells are not highlighted given their expression of higher weight cytokeratins. Specific cytokeratins stains, such as for cytokeratins 7 and 20, offer additional diagnostic details.

Cytokeratin 7 is found on sweat glands, Merkel cells, Toker cells, and importantly Paget’s cells, but not the epidermis or hair follicle epithelium. It is therefore sensitive, but not specific. Cytokeratin 20 staining, on the other hand, is often found in EMPD, especially when there is an underlying carcinoma.

Several other immunostains also exist and are summarized in great detail in some reports. Gross cystic disease fluid protein (GCDFP-15 or BR-2) is a marker of apocrine epithelium and strongly suggests primary vulval and perianal EMPD. Epithelial membrane antigen (EMA)/Human milk fat globule membrane antigen (HMFG) is a marker for most cases of MPD and EMPD, but not squamous epithelium. Carcinoembryonic antigen (CEA) is a marker of adenocarcinoma, and stain most cases of primary EMPD.

Estrogen and progesterone receptors generally stain positively in MPD, as opposed to sweat gland tumors of the skin or EMPD. c-erb-B2 is an oncogene that produces a protein found in most cases of MPD, but only occasionally in EMPD. S100 is a calcium binding protein that, along with the monoclonal antibody HMB-45 and extracts of activated melanocytes and melanoma, is helpful in differentiating MPD and EMPD from melanocytic lesions.

Who is at Risk for Developing this Disease?

MPD is observed in the range of 0.7% to 4.3% of breast cancers. Because of its relationship with breast cancer, it is generally seen in postmenopausal women. EMPD is less common than MPD, although its true incidence is not known. It is more often found in adult Caucasian women between 50 and 80 years old, although in Japan there is a male predominance.

What is the Cause of the Disease?

MPD and EMPD are neoplastic processes. The causes and cells of origin are controversial.


MPD is largely believed to originate from underlying ductal carcinoma of the breast. Some, however, question whether Toker cells beget MPD, especially in the uncommon instances of MPD arising without underlying breast carcinoma. Toker cells are found intraepidermally in 10% of normal nipples. Their abundant cytoplasm and ability to appear singly or in clusters can mirror the appearance of MPD on pathology. Even immunostains of Toker cells and Paget’s cells of MPD can overlap, leaving the pathologist to rely more heavily on the pleomorphism and atypical nuclei of Paget’s cells to make the correct diagnosis.

In contrast, EMPD is thought to arise from one of two pathways, known as primary and secondary forms. Primary EMPD is believed to arise from intraepidermal ducts of apocrine glands or pluripotent keratinocyte stem cells. The secondary form, observed in roughly 25% of reported EMPD cases, results from contiguous spread from an underlying neoplastic dermal adnexal gland or nearby organ, especially one from the urogenital and gastrointestinal tract.

Systemic Implications and Complications

Since MPD is nearly always associated with an underlying ductal carcinoma of the breast, testing should be employed to understand the extent of the disease. Typical options include mammography, ultrasound, magnetic resonance imaging, sometimes thermography, possible tissue examination via fine needle aspiration, and always tissue examination of any surgically excised specimen.

Although EMPD is far less associated with underlying malignancy, imaging and sometimes other tests should still be employed, and modified based on location of the cutaneous eruption. Physicians should consider colonoscopy, cystoscopy, mammography, intravenous urography, abdominal and pelvic ultrasound or computed tomography based on proximity of the EMPD lesion to these areas. Some recommend serum prostate specific antigen testing.

Regardless of location, it is this author’s recommendation to perform age-appropriate cancer screening (eg, breast examination, colonoscopy, prostate examination, pap smear) whenever a case of EMPD is diagnosed.

Treatment Options

Table 1. Treatment options for MPD and EMPD

Table 1.
Medical Referral to an oncologistAdjuvant chemotherapy/ pharmacologic therapy (eg, tamoxifen) Medical Topical chemotherapy (eg, 5-fluoruracil, bleomycin, imiquimod)
Surgical Modified radical mastectomy with lymph node excisionBreast-conserving surgery (such as wide conic excision of the nipple/areola) Surgical Mohs micrographic surgeryRadical excision (eg, radical vulvectomy)
Physical modalities Radiotherapy Physical modalities RadiotherapyPhotodynamic therapyLaser ablation

Optimal Therapeutic Approach for this Disease

The management of MPD is based on prospective and retrospective series and has been reviewed in the literature. Referral to an oncologist and surgical oncologist is needed. The first-line therapy for MPD is surgical. Modified radical mastectomy has been the standard of care, but breast-conserving surgery is employed in certain circumstances. Conservative excision, such as conic excision, which involves the removal of the nipple-areola complex and the subareolar breast tissue, is reasonable in some cases when the nipple-areola changes are found without evidence of a palpable mass in the breast or mammographic abnormalities.

When the axilla demonstrates palpable nodes or the underlying ductal carcinoma is invasive, an axillary dissection is warranted in order to guide adjuvant therapy strategies. Sentinal lymph node biopsy has replaced typical dissection at some institutions with experience with these procedures. Adjuvant chemotherapy and pharmacologic therapy is then integrated based on the underlying ductal carcinoma’s characteristics, such as degree of invasion and hormone receptors.

Radiotherapy typically serves as an adjuvant therapy in settings of breast conservation. However, radiotherapy as a primary strategy for MPD results in recurrence rates higher than conservative surgical intervention such as lumpectomy with radiotherapy.

Surgery is the standard of care for EMPD. That said, any excision attempt comes with high local recurrence rates because of the typical lesion’s irregular margins, multicentricity, and frequent involvement of otherwise normal appearing skin.

Several studies now endorse Mohs micrographic surgery (Mohs surgery) as the best level of care for the patient, as it delivers fewer recurrences than wide local excision and spares normal tissue. One study showed a recurrence rate of 8% with Mohs surgery and 22% with wide local excision. Radical excision, such as radical vulvectomy, has also demonstrated low rates of recurrence but comes with substantial disfigurement such that this treatment course is not often pursued.

Nonsurgical options do exist for patients with EMPD, and these have been detailed in the literature, but these are generally reserved for poor surgical candidates or those refusing surgical intervention. Radiotherapy has reportedly been successful, although there are no randomized controlled trials endorsing this approach. Topical chemotherapy (eg, 5-fluorouracil, bleomycin, imiquimod) has reportedly been beneficial in some reports, although with questionable efficacy, and often it comes with many weeks of irritation and pain.

Systemic chemotherapy has been utilized but no widely accepted treatment regimens have been defined. Photodynamic therapy has been effective in some reports, wherein the clinicians have attempted topical and intralesional aminolevulinic acid. Finally, efforts for laser ablation are reported in the literature, but recurrences have been substantial.

Patient Management

Clinical follow-up is recommended for all patients with MPD and EMPD because of risks for recurrence and because of risk for associated internal malignancies. It is the author’s belief that oncologists should participate in the follow-up care, as well as site-specific specialists (ie, for cases of EMPD near the perianal skin, gastroenterologists should be part of the follow-up care).

There is no rule as to how and when to follow up. For MPD, follow-up depends in part on the nature of the underlying ductal carcinoma. However, all patients should have many years of regular breast examinations and mammography with or without other imaging. For non-invasive EMPD, some recommend biannual examinations for 3 years, then annually for 10. For invasive EMPD or those forms having an associated malignancy, follow-up is recommended more frequently, usually three to four times per year, especially in the first several years. Some clinicians coordinate follow-up based on EMPD location. For example, for perianal EMPD, some perform colonoscopies every 2 to 3 years.

Unusual Clinical Scenarios to Consider in Patient Management

As with most dermatologic conditions, there can be exceptions to the rule. Although MPD is almost always seen with an underlying ductal carcinoma, there are unusual situations in which this is not the case. In these cases, conservative surgical intervention can be selected. It is theorized the tumor arises from apocrine ducts in the skin much like primary forms of EMPD. Multifocal EMPD can exist, as well as EMPD arising in areas outside the apocrine gland distribution. You cannot rule out EMPD simply by anatomic location.

For EMPD lesions particularly suspicious for having clinically undetectable extension, a number of strategies can be employed to better define the tumor before excision. A patient can be treated preoperatively with topical 5-fluorouracil to better define the affected areas. One would expect a visible inflammatory reaction where 5-fluorouracil interacted with the tumor-containing skin. Other attempts at border definition include fluorescein staining as well as intraoperative cytokeratin 7 staining.

What is the Evidence?

Heymann, WR. “Extramammary Paget's disease”. Clin Dermatol. vol. 11. 1993. pp. 83-7. (A succinct and comprehensive review of EMPD.)

Lloyd, J, Flanagan, AM. “Mammary and extramammary Paget's disease”. J Clin Pathol. vol. 53. 2000. pp. 742-9. (A succinct and comprehensive review of MPD and EMPD.)

Yim, JH, Wick, MR, Philpott, GW, Norton, JA, Doherty, GM. “Underlying pathology in mammary Paget's disease”. Ann Surg Oncol. vol. 4. 1997. pp. 287-92. (Retrospective review of clinical, mammographic and pathologic data from 38 women with mammary Paget's disease treated between 1979 and 1995. This review correlates associated breast findings with disease stage and outcome to help guide evaluation and treatment.)

Smith, KJ, Tuur, S, Corvette, D, Lupton, GP, Skelton, HG. “Cytokeratin 7 staining in mammary and extramammary Paget's disease”. Mod Pathol. vol. 10. 1997. pp. 1069-74. (An assessment of which immunohistochemical stains are the best for the diagnosis of MPD and EMPD, as well as the best stains for identifying small foci of tumors in evaluating tumor margins.)

Lundquist, K, Kohler, S, Rouse, RV. “Intraepidermal cytokeratin 7 expression is not restricted to Paget cells but is also seen in Toker cells and Merkel cells”. Am J Surg Pathol. vol. 23. 1999. pp. 212-9. (A study showing that anti-CK7 is an effective but not 100% sensitive marker for Paget cells, staining 21 of 22 cases of MPD and 19 of 22 cases of EMPD, whereas CK20 stained 0 of 17 cases of MPD and 6 of 19 cases of EMPD. It also highlights the fact that anti-CK7 staining can be found in Toker cells and Merkel cells.)

Goldblum, JR, Hart, WR. “Perianal Paget's disease: a histologic and immunohistochemical study of 11 cases with and without associated rectal adenocarcinoma”. Am J Surg Pathol. vol. 22. 1998. pp. 170-9. (An analysis of the histologic and immunohistochemical features of perianal EMPD in 11 patients to determine the frequency and relationship of associated regional internal carcinoma and to gain insight into its histogenesis.)

Kanitakis, J. “Mammary and extramammary Paget's disease”. J Eur Acad Dermatol Venereol. vol. 21. 2007. pp. 581-90. (A review of MPD and EMPD.)

Marcus, E. “The management of Paget's disease of the breast”. Curr Treat Options Oncol. vol. 5. 2004. pp. 153-60. (A review of the management of MPD.)

O’Connor, WJ, Lim, KK, Zalla, MJ. “Comparison of Mohs micrographic surgery and wide excision for extramammary Paget's disease”. Dermatol Surg.. vol. 29. Jul 2003. pp. 723-727. (A comparison of Mohs micrographic surgery versus wide local excision for the treatment of EMPD.)

Shepherd, V, Davidson, EJ, Davies-Humphreys, J. “Extramammary Paget's disease”. BJOG. vol. 112. 2005. pp. 273-9. (A review of EMPD)