Background
Pagetoid spread (PS) is defined as the proliferation of individual cells in the epithelium characterized by erythema and inflammation. It can manifest as mammary or extramammary Paget’s disease (EMPD) and is described as an apocrine gland tumor that can be benign or malignant with metastatic potential. EMPD has been reported at several extramammary sites, including the axilla, thigh, groin, perineum, scrotum, vulva, and perianal area [1, 2]. When EMPD affects the perianal region, it is called perianal Paget’s disease (PPD). PPD is usually associated with an underlying malignancy, such as anal, rectal, cervical, or urinary bladder adenocarcinoma, with an appearance similar to that of simple PPD without an underlying malignancy [3, 4]. The prognosis of simple PPD is relatively favorable, with overall and disease-free survival of approximately 60% at 5 years [5‐7]. However, PPD associated with an underlying malignancy has been reported to have a poor prognosis [5‐9].
Immunohistochemical analysis of the skin lesion is useful for differentiating whether or not PPD is associated with underlying anorectal adenocarcinoma or carcinoma. Immunohistochemical markers that are frequently used include cytokeratin 7 (CK7), CK20, gross cystic disease fluid protein 15 (GCDFP-15), and CDX2. Although CK7 is a sensitive marker for almost all pagetoid neoplasms, it is not practical for differentiating PPD because some rectal adenocarcinomas also express CK7 [10, 11]. CK20 might be expressed in colorectal carcinoma but not in simple PPD [12]. GCDFP-15, which is regarded as a specific marker for apocrine epithelium tissue, is usually positive in simple PPD [13‐15]. As mentioned above, PPD associated with anorectal carcinoma often has the CK20+/GCDFP-15− immunophenotype. In contrast, simple PPD shows a CK20−/GCDFP-15+ pattern [16]. CDX2, which has a high positivity rate in gastrointestinal cancers, including rectal cancer, has been used [17]. However, about 17% of colorectal cancers are negative for CK20 [18]. Therefore, it is necessary to evaluate the underlying malignancy using a combination of colonoscopy and radiology [19].
Anzeige
Staging classification and treatment outcome
A staging classification for PPD that includes appropriate treatment for each stage has been proposed (Table 1) [9, 20]. The prognosis is good for stage I (Paget’s cells found in the perianal epidermis and adnexa without primary carcinoma) but worsens for stage II (invasive cutaneous disease penetrating the basement membrane and entering the underlying stroma and/or synchronous localized malignancies, i.e., IIa adnexal malignancy and IIb visceral malignancy), stage III (regional metastatic disease), and stage IV (distant metastatic disease).
Table 1
Perianal Paget’s disease classification and accompanying suggested therapy
Stage | Description | Management |
|---|---|---|
I | Paget’s cells in the perianal epidermis and adnexa without primary carcinoma | Wide local excision |
IIA | Cutaneous Paget’s disease with associated adnexal carcinoma | Wide local excision |
IIB | Cutaneous Paget’s disease with associated anorectal carcinoma | Abdominoperineal resection |
III | Paget’s disease in which associated carcinoma has spread to regional nodes | Inguinal lymph nodes dissection and abdominoperineal resection/wide local excision |
IV | Paget’s disease with distant metastases of associated carcinoma | Chemotherapy, radiotherapy, local palliative management |
The treatment options for PPD depend on local (extent and depth of invasion) and regional (lymph node involvement) factors and the extent of systemic disease. Local excision (with macroscopic clearance of surgical margins) has been performed as a treatment for non-invasive PPD but was associated with a high local recurrence rate (40%) [19]. Wide local excision (WLE, >1 cm microscopic clearance of surgical margins) with a sphincter-saving technique was later proposed as the treatment of choice for PPD due to the higher survival rate in patients treated with WLE than in those treated with local excision and due to the better chance of cure and normal survival. The standard treatment for PPD associated with the anorectal canal is abdominoperineal resection (APR) [9, 20]. However, it is recommended that adenocarcinoma of the anal canal be managed in the same way as rectal cancer [21], and transanal local excision is appropriate for selected early-stage rectal cancers [22]. Therefore, even in cases of PPD associated with anal canal cancer, combined transanal local excision and WLE may be an alternative to APR.
The distinction between PPD with and without underlying anorectal cancer is essential [3, 23‐25] because of the differences in treatment methods and the prognosis [4]. Due to the rarity of PPD, few cases have been reported [24‐27], and long-term outcomes of treatment for PS are unknown. This study aims to analyze the clinical and pathological features of perianal Paget’s disease (PPD) and to explore rational treatment options and follow-up for this disease.
Methods
A search of the National Cancer Center Hospital database identified 14 cases of PPD diagnosed between 2006 and 2021. Cases of adenocarcinoma of the rectum or anal canal with pagetoid spread, including suspected and recurrent cases, were retrospectively reviewed for age, sex, tumor size, presence or absence of preoperative endoscopic findings, presence or absence of mapping biopsy, treatment methods, and long-term prognosis.
Anzeige
The study was approved by the National Cancer Center Hospital Institutional Review Board (code: 2017–437). The requirement for written informed consent was waived in view of the retrospective nature of the research and the anonymity of the study data.
Results
All 14 patients met the inclusion criteria. Eight patients (57.1%) were male, and 7 (50.0%) were female (Table 2). The median age was 74 (range, 55–84) years. Twelve patients (85.7%) were suspected of having PS before initial treatment. The most commonly used immunostaining for differentiation was CK7+/CK20+/GCDFP-15−/CDX2+. These 12 patients were clinically and radiologically evaluated for underlying anorectal carcinoma, which was identified in 6 cases. The median size of the primary tumor was 52 (9–110) mm. No primary lesions were detected in the other 6 cases. There were no distant metastases at the initial diagnosis; however, 1 patient had metastasis to the lateral lymph nodes, and another had metastasis to the inguinal lymph nodes.
Table 2
Clinicopathologic characteristics
Characteristic | Value | (n=14) |
|---|---|---|
Age, median (IQR), years | 74 | (67.5–78.5) |
Gender | ||
Male, No. (%) | 7 | (50%) |
Female, No. (%) | 7 | (50%) |
Preoperative diagnosis of pagetoid spread, No. (%) | ||
Diagnosed | 2 | (14%) |
Not diagnosed | 12 | (86%) |
Identification of underlying anorectal cancer, No. (%) | ||
Not found | 6 | (43%) |
Syncronous | 7 | (50%) |
Metachronous | 1 | (7%) |
Primary tumor site, No. (%) | ||
Anal canal | 7 | (50%) |
Lower rectum | 1 | (7%) |
Unknown | 6 | (43%) |
Immunohistochemistry, No. (%) | ||
Not confirmed | 2 | (14%) |
CK20+/CK7+ | 7 | (50%) |
CK20+/CK7− | 4 | (29%) |
CK20−/CK7+ | 0 | (0%) |
CK20−/CK7− | 1 | (7%) |
GCDFP15−a | 8 | (100%) |
CDX2+ | 11 | (92%) |
Histologya, No. (%) | ||
Adenocarcinoma (grade unknown) | 7 | (50%) |
Tub1/2 | 1 | (7%) |
Por/sig | 4 | (933%) |
Muc | 3 | (600%) |
Tumor size, median (IQR), cm | 53.5 | (26.3–87.5) |
Depth of invasion, No. (%) | ||
M | 3 | (21%) |
SM | 1 | (7%) |
MP | 2 | (14%) |
A | 1 | (7%) |
AI | 1 | (7%) |
Invasive (WLE) | 1 | (7%) |
Non-invasive (WLE) | 2 | (14%) |
Lymph node metastases at initial treatment, No. (%) | ||
Negative | 4 | (29%) |
Lateral lymph node | 1 | (7%) |
Inguinal lymph node | 2 | (14%) |
Unknown (clinically negative) | 7 | (50%) |
Initial treatment, No. (%) | ||
WLE (±transanal local excision | 7 | (50%) |
APR (±WLE) | 5 | (36%) |
Radiation | 1 | (7%) |
Unknownb | 1 | (7%) |
Mapping biopsyc, No. (%) | ||
Done | 5 | (42%) |
None | 7 | (58%) |
Resection statusc, No. (%) | ||
R0 | (0%) | |
R1 | (0%) | |
R2 | (0%) | |
Follow-up time, median (IQR), mo | 4.5 | (158.7–7.4) |
Recurrence, No (%) | 7 | (50%) |
Time to recurrence, median (IQR), mo | 34.6 | (18.2–81.7) |
5-year recurrence free survival | 55.9% | |
Mortality, No. (%) | 2 | (14%) |
5-year overall survival | 90.0% | |
The median follow-up period after diagnosis was 4.5 (range, 0.1–13.0) years. At the last follow-up, 7 patients were alive with no evidence of disease, 4 were alive with recurrent disease, 1 had transferred to another hospital, and 2 had died of disease. A summary of the 13 cases and their clinical course is presented in Table 3.
Table 3
A summary of cases attended
Case | Gender | Initial treatment | Immunohistological findings | Primary | Histology* | LN mets | Mapping biopsy | 1st relapse | Outcome | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Age | Lymph node dissection | CK7 | CK20 | GCDFP15 | CDX2 | Maximum diameter | Depth of invasion | Location of LN | Resection status | Interval | Treatment | Follow-up time (primary) | |
No detectable primary lesion | |||||||||||||
1 | M | WLE | − | + | NA | + | Unknown | Adenocarcinoma | Negative | Done | ILN | ILND | Dead of disease |
76 | None | 55mm | Invasive | R1 | 61 months | 6 years | |||||||
2 | M | WLE | + | + | − | + | Unknown | Adenocarcinoma | NA | Done | ILN | ILND | Alive with no evidence of disease |
55 | None | 30mm | Non-invasive | R1 | 34 months | 13 years | |||||||
3 | F | RT | − | + | − | − | Unknown | Adenocarcinoma | NA | None | Local | Medication | Alive with 2nd relapse |
78 | None | NA | NA | 87 months | 12 years | ||||||||
4 | F | APR with WLE | + | + | − | + | Unknown | Adenocarcinoma | Negative | None | NA | NA | Alive with no evidence of disease |
73 | D2 | 74mm | MP | R0 | 5 years | ||||||||
5 | F | WLE | − | − | NA | − | Unknown | Muc | Positive | Done | NA | NA | Alive with no evidence of disease |
63 | ILND | 80mm | Invasive | Inguinal | R0 | 2 years | |||||||
6 | M | WLE | + | + | − | − | Unknown | Adenocarcinoma | Negative | None | Local | WLE | Alive with 2nd relapse |
80 | None | 50mm | Invasive | R0 | 25 months | 5 years | |||||||
Detectable underlying anorectal carcinoma | |||||||||||||
7 | M | APR with WLE | − | + | NA | + | P | Muc/sig | Positive | None | ILN | ILND | Dead of disease |
68 | D3,LLND | 110mm | AI | Lateral pelvic | R1 | 7 months | 2 years | ||||||
8 | M | Transanal LE with WLE | − | + | − | + | P | Tub1 | NA | None | NA | Alive with no evidence of disease | |
72 | None | 9mm | SM1 | R0 | 5 years | ||||||||
9 | F | APR with WLE | + | + | − | + | P | Tub1/por/sig | Negative | Done | LD/ILN | LLND | Alive with no evidence of disease |
80 | D2 | 90mm | MP | R1 | 13 months | ILND | 2 years | ||||||
10 | M | APR with WLE | + | + | NA | + | P | Adenocarcinoma | Negative | None | NA | Alive with no evidence of disease | |
66 | D2 | 106mm | M | R0 | 3 years | ||||||||
11 | F | NA | + | + | − | + | Rb | NA | NA | NA | NA | Unknown | |
69 | None | Unknown | 3 years | ||||||||||
12 | F | Transanal LE with WLE | + | + | − | + | P | Por2/tub2/muc | NA | Done | Local | WLE | Alive with no evidence of disease |
84 | None | 52mm | M | R1 | 36 months | 4 years | |||||||
No preoperative diagnosis of pagetoid spread | |||||||||||||
13 | M | Trans anal LE* | NA | NA | NA | NA | P | Tub2/por | NA | NA | Local | WLE | Alive with 2nd relapse |
74 | None | 10mm | SM | 159 months | 13 years | ||||||||
14 | F | Chemoterapy | NA | NA | NA | + | P | Por/sig | Positive | None | NA | Alive with | |
74 | D3, LLND, ILND | 25mm | A | Inguinal | Positive | 1 year | |||||||
Seven patients (50%) developed recurrence, with a median time to recurrence of 34.6 (range 19.2–81.7) months. Three patients had local recurrence, and 4 (28.6%) had a recurrence in the inguinal lymph nodes. One patient had distant metastases. The time to the first relapse was 3 months, and that to the second relapse was 6 months. The 5-year overall survival rate was 90.0%.
The initial treatment in the 6 patients with no detectable primary lesion was WLE alone (n=4, 66.7%), APR + WLE (n=1, 16.7%), and radiotherapy (n=1, 16.7%). The outcomes in the 4 patients who underwent WLE alone were as follows: alive with no evidence of disease (n=2), alive with lung metastasis (n=1), and died after inguinal lymph node metastases at first relapse and liver and bone metastases at the second relapse (n=1).
The initial treatment of the 6 patients in whom anorectal carcinoma was detected was APR + WLE (n=3, 50%), transanal local excision + WLE (n=2, 33.3%), and referral to another hospital (n=1, 16.7%). The outcomes in the group that underwent APR + WLE were as follows: alive with no evidence of disease (n=2) and dead (n=1). The outcomes in the group that underwent transanal local excision + WLE were as follows: alive with no evidence of disease (n=1) and alive with recurrent disease (n=1).
In terms of initial treatment, there were 4 recurrences after WLE (with or without transanal local excision), 2 after APR, and 1 after radiotherapy. All 6 cases of recurrence after surgical resection had positive resection margins at the initial surgery. Mapping biopsy was performed in 5 (45.4%) of the 11 patients who underwent surgical resection, and 4 (80.0%) had positive margins. In the 6 cases without mapping biopsy, 2 (33.3%) had positive margins.
Discussion
PPD is usually associated with an underlying malignant anorectal tumor and has a relatively poor prognosis [23] with a high risk of local recurrence [28]. The rate of malignancy associated with PPD ranges from 33 to 86% [29]. In the present study, of the 11 cases of PPD associated with anal canal cancer who presented with CK20+/GCDFP-15− (GCDFP was not examined in some cases), anal cancer could be noted in 6 patients by endoscopic or radiographic evaluation. Immunohistological examination alone is not sufficient to identify underlying malignancy, and it is critical to apply PD staging (Table 1) in conjunction with endoscopic and radiologic evaluation.
Anzeige
In the present study, primary anorectal carcinoma could not be identified preoperatively in half of the patients, and APR was performed in only 1 case. There was 1 death in the WLE group. However, only half of the patients with anorectal cancer underwent APR + WLE to preserve the anus, although there were cases of recurrence-free survival of more than 5 years after WLE. Further investigations are needed to identify cases in which APR should be pursued aggressively and those in which WLE (with or without transanal local excision) can be considered.
The 2 deaths occurred in a patient without an identified primary tumor who underwent WLE alone as the initial treatment (case 1) and in a patient with a primary tumor identified preoperatively who underwent APR + WLE (case 7).
Of the 6 cases in which anal cancer was detected, APR + WLE was performed in 3 patients and Transanal Local Excision was performed in 2 cases. In the latter two cases, the depth of the primary tumor was M/SM1, and recurrence-free survival of 2 to 5 years was achieved. Further investigation is considered to be necessary.
On the other hand, both these fatal cases had inguinal lymph node metastasis as the first relapse, and distant metastasis several months later.
Anzeige
The time to first recurrence varies from 7 months to about 5 years, and it is difficult to predict prognosis in general from the course of the disease, but long-term and regular follow-up for more than 5 years is considered necessary to check for recurrence and distant metastasis.
Intraoperative frozen section analysis of the resection margin has been proposed to reduce the possibility of borderline invasion and minimize the local recurrence rate [19]. However, frozen section analysis of surgical margins in PPD can be misleading and dangerous because it may appear negative intraoperatively but become permanently positive on subsequent histological analysis. It is believed that permanent margin status is not a predictor of local recurrence and that a minimally invasive carcinoma measuring <1 mm probably does not have an adverse prognosis, whereas a deeply invasive carcinoma behaves as a fully malignant adenocarcinoma [30]. Of the five patients who underwent mapping biopsy in this study, four were positive for transection, but only one case resulted in local recurrence. This is consistent with previous studies that have shown that edge evaluation is not a predictor of local recurrence.
There are several potential limitations to this study. First, there is the possibility of selection bias due to the retrospective study design. Second, the sample size is very small (14 cases) and cannot shed light on treatment possibilities, especially the optimal approach and prognosis. Despite these limitations, we believe that our findings warrant more extensive investigation in patients with PPD.
Conclusions
Although skin biopsy and immunohistological diagnosis are useful for distinguishing underlying malignancy in patients with PPD, endoscopic and radiologic evaluation is mandatory. The time to first recurrence varies widely, and long-term and regular follow-up for more than 5 years is considered necessary for local recurrence and distant metastasis.
Anzeige
Acknowledgements
The authors thank Sigeki Sekine, all of whom served as staff members at the National Cancer Center Hospital.
Declarations
Ethics approval and consent to participate
Ethics approval was obtained from our hospital’s review board (NCC2017437).
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.