On the Cusp of Extraordinary Advances in Extramammary Paget Disease

By Warren R. Heymann, MD

I always include extramammary Paget disease (EMPD) and Bowen disease in my differential diagnosis of anogenital rashes for fear of missing them. It is good form to biopsy such lesions should they not respond to standard therapies for the more likely diagnoses in this region (tinea, Candida, seborrheic dermatitis, or psoriasis). Even with this prudent approach, I have diagnosed EMPD only a few times in four decades of practice.

EMPD commonly occurs on apocrine rich regions, usually affecting anogenital skin or the axilla, although the eyelid, ear canal, and umbilicus may rarely be affected. EMPD usually presents as an erythematous plaque on the genitals of patients aged 60 to 80 years. Compared to Paget disease of the breast, where 90% of cases are associated with an underlying breast cancer, only 7% to 40% of EMPD cases have an associated underlying internal malignancy. EMPD is considered primary when arising as an intraepidermal neoplasm with the potential for invasion or metastasis. Secondary EMPD represents intraepithelial spread of malignant adenocarcinoma cells from an underlying internal malignancy. (1)

According to Merritt et al, “In cases of EMPD with coexistent internal cancer, the location of EMPD is often correlated with the site of the underlying malignancy. Perianal EMPD is associated with lower gastrointestinal malignancy and penoscrotal EMPD is associated with urinary tract malignancy. Distant internal malignancies reported in association with EMPD include breast carcinoma, ovarian carcinoma, bile duct carcinoma, hepatocellular carcinoma, renal cell carcinoma, lung carcinoma, stomach carcinoma, and pancreatic carcinoma. Additionally, secondary EMPD has been described in the setting of contiguous pagetoid spread from an underlying adenocarcinoma. Rates of contiguous spread from adjacent malignancy of up to 23.9% have been reported in the literature. In particular, contiguous spread is more likely to occur from the vulva, urinary tract, and anorectal regions.” (2)

he concept of risk stratification for EMPD is evolving. It is essential to biopsy suspicious lesions — the H&E will be diagnostic, but immunohistochemistry will help determine primary from secondary EMPD, with both staining positive for cytokeratin-7, but only secondary EMPD staining positive for cytokeratin-20. (1) In a study of 113 patients with vulvar extramammary Paget disease (VPD), 77% had noninvasive EMPD. Of this group, only 8% developed invasion. There were no disease-specific deaths reported in the women with noninvasive EMPD. The 5-year disease-specific survival rate was greater than 98% in noninvasive and microinvasive EMPD, but significantly worse in invasive (50%). The authors concluded that most patients have noninvasive VPD, which does not affect survival, and should be considered a chronic disorder with limited invasive potential. For those with invasive disease, survival decreases significantly.(3)

In a study of 199 patients with vulvar EMPD, 76.9% were diagnosed with cutaneous noninvasive VPD, and this group has no increased risk for developing malignancies of the breast, intestine, or urological tract. The authors suggested that routine screening for these malignancies in patients with cutaneous noninvasive EMPD may not be necessary. (4) In an accompanying editorial, Cooper and Matin stated that in practice, “All women with suspected VPD must be biopsied and appropriate immunocytochemistry performed to help distinguish between primary and secondary VPD. Those with secondary VPD/EMPD or invasive VPD/EMPD must continue to be examined and investigated for underlying malignancy in the bowel and urinary tract as malignancy is strongly associated. Those with a profile suggesting primary cutaneous VPD/EMPD (1a) may not require routine screening unless clinical symptoms and signs suggest otherwise. Unnecessary, invasive and expensive tests may be avoided.” (5)

For those patients with invasive EMPD new diagnostic tests may help prognosticate and aid in measuring therapeutic responses. Cell-free DNA (cfDNA) has been an indicator in several cancers. Midjiddorj et al have demonstrated that serum cfDNA levels are significantly elevated in patients with EMPD with or without metastasis compared with healthy controls. (6) The combination of serum carcinoembryonic antigen (CEA) and cytokeratin 19 fragment 21-1 has been demonstrated to be positive in 95% of cases of metastatic EMPD. (7) Future studies of these markers will determine their practical clinical utility.

Therapeutically, surgical excision is the cornerstone of treatment for EMPD, with Mohs micrographic surgery (MMS) considered the treatment of choice. Alternative treatments include topical 5-fluorouracil, imiquimod, photodynamic therapy, laser vaporization, chemotherapy, and radiation therapy, however data are limited in allowing for specific recommendations. (2)

Ultimately, the key to treating EMPD will be based on comprehending its pathogenesis. Although the current theory points to EMPD arising from Toker cells, future therapies will focus on modifying molecular pathways. An exome analysis from three patients with EMPD revealed mutations in several genes including new somatic mutations in the TP53 gene (8), which is a tumor suppressor gene — targeted therapies are on the horizon.

Point to Remember: EMPD must be assessed to determine if it is primary or secondary. Advances in diagnostic and molecular techniques will alter how these patients are assessed and treated.