Eccrine Porocarcinoma Treated by Mohs Micrographic Surgery: Over 6-Year Follow-up of 12 Cases and Literature Review Yaohui G. Xu, MD, PhD,* Juliet Aylward, MD,* B. Jack Longley, MD,* Molly A. Hinshaw, MD,* and Stephen N. Snow, MD†
BACKGROUND Eccrine porocarcinoma (EPC) has a poor prognosis after standard wide local excision (WLE), with 20% local recurrence, 20% regional and 12% distant metastatic rates. Mohs micrographic surgery (MMS) has been used as a promising treatment. OBJECTIVE
To review the use of MMS for EPC and assess treatment outcomes.
MATERIALS AND METHODS This was a retrospective case series of 12 EPC patients treated by MMS between 1984 and 2013 in the institution. Furthermore, a literature review revealed an additional 17 cases of EPC managed by MMS. RESULTS Of 29 cases of EPC treated by MMS, outcome was established in 27 cases. The patients had a significantly longer mean follow-up period of 6 years (range, 4–206 months), as compared with 19 months (range, 2–48 months) in reported cases. Two patients had regional lymph node metastasis after MMS. The regional metastatic rates to lymph nodes were 7% (2/27). There was no local recurrence, distant metastasis, or disease-specific death in the 27 cases studied. CONCLUSION To the best of the authors’ knowledge, this is the single largest case series of EPCs treated by MMS and the authors’ data demonstrated that MMS may be superior to the standard WLE. The authors have indicated no significant interest with commercial supporters.
E
ccrine porocarcinoma (EPC) is a rare malignant tumor of the intraepidermal (acrosyringeal) portion of the eccrine gland that represents only 0.005% to 0.01% of cutaneous neoplasms.1 Less than 300 cases of EPC have been reported since the original description by Pinkus in 1963 as “epidermotropic eccrine carcinoma.”2
Treatment of EPC remains challenging. Most patients are managed by wide local excision (WLE), and the reported local recurrence, and regional and distant metastatic rates are 20%, 20%, and 12%, respectively. Mortality rate is over 65% in patients with nodal metastases.3 Mohs micrographic surgery (MMS) has been shown effective in eradicating early EPC.3 The authors
present the clinical features and outcomes of 12 EPCs treated by MMS in the authors’ clinic, and 17 cases reported in the English literature.4–11 Methods This is a retrospective case series of 12 EPCs treated by MMS between January 1984 and September 2013 in the authors’ clinic. The study was approved by the University of Wisconsin Institutional Review Board Committee. Patient 1 was previously published by the senior author. Three patients died of other causes. The remaining 8 patients had routine follow-up skin examination by general dermatologists, and all were contacted by phone in September 2013 for the final confirmation of no recurrences or death. All 17 EPCs
*Department of Dermatology, University of Wisconsin, Madison, Wisconsin; †Department of Dermatology, Northwest Permanente, Portland, Oregon © 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved. ISSN: 1076-0512 Dermatol Surg 2015;41:685–692 DOI: 10.1097/DSS.0000000000000382
·
·
685
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
686
Pt
Age (y)/ Sex
Tumor Location
Tumor Duration
Mohs Stages
Defect (mm)
Local Rec
Regional Met
Distant Met
1* 2
FU (mo) Outcome
Author (Reference)
Year Published
36/WM
Chin
1.5 y
3
15 · 15
No
No
No
60
86/WM
Antihelix
NS
2
20 · 15
No
No
No
54
Alive
This study
1992
DOOC
This study
3
86/WM
Scalp
NS
2
18 · 20
No
No
No
16
DOOC
This study
2015 2015
4
73/WF
Chest
4y
3
24 · 56
No
Yes‡
No
206
Alive
This study
2015
5
71/WM
Chest
NS
1
31 · 24
No
No
No
127
Alive
This study
2015
6
71/WF
Chest
3 mo
1
90 · 65
No
No
No
4
DOOC
This study
2015
7
77/WM
Knee
12 y†
3
56 · 60
No
No
No
96
Alive
This study
2015
8
64/WM
Forehead
>1 mo
2
11 · 11
No
No
No
85
Alive
This study
2015
9
49/WF
Upper arm
NS
2
27 · 24
No
No
No
81
Alive
This study
2015
10
56/WM
Nose
1y
4
20 · 20
No
No
No
75
Alive
This study
2015
11
59/WF
Thigh
2.5 y
1
24 · 20
No
No
No
65
Alive
This study
2015
12
59/WM
Ankle
2y
3
45 · 45
No
No
No
37
Alive
This study
2015
13
71/AAF
Antihelix
20 y
1
NS
No
No
No
12
Alive
McMichael4
1999
14
79/WF
Foot
NS
2
NS
No
No
No
36
Alive
Wittenberg5
1999
15
74/WM
Scalp
NS
2
NS
No
No
No
48
Alive
Wittenberg5
1999
16
67/WF
Shin
NS
2
NS
No
No
No
12
Alive
Wittenberg5
1999
17
39/WF
Foot
NS
1
NS
No
No
No
5
Alive
Wittenberg5
1999
18
63/WF
Temple
NS
2
NS
No
No
No
12
Alive
Wittenberg5
1999
19
78/WF
Back
NS
1
NS
No
No
No
12
Alive
Johr6
2003
20
36/M
Chest
NS
2
NS
No
No
No
36
Alive
Wildemore7
2004
21
74/F
Shin
>10 y
2
NS
No
No
No
31
Alive
Wildemore7
2004
22
77/F
Thigh
NS
2
NS
No
No
No
29
DOOC
Wildemore7
2004
23
54/M
Ankle
NS
2
NS
No
No
No
15
Alive
Wildemore7
2004
24
71/M
Eyelid
6 wk
2
11 · 5
NS
NS
NS
NS
NS
D’Ambrosia8
2004
25
56/AAM
Back
1–2 y
3
NS
NS
NS
NS
NS
NS
Cowden9
2006
26
71/AAF
Heel
>1 y
NS
NS
No
No
No
2
Alive
Cowden9
2006
27
70/WM
Eyelid
3 mo
NS
9 · 14
No
No
No
6
Alive
Jain10
2008
28
63/M
Upper lip
4 mo
1
NS
No
No
No
12
Alive
Kwah11
2012
29
59/HF
Forearm
3y
NS
NS
No
Yesx
No
6
Alive
Vleugels12
2012
Case 1 to 12: from the authors’ series; cases 13 to 29: cumulative results from the literature. *Patient 1 was previously reported by Snow and Reizner3 in 1992. †Patient 4 developed regional lymph node metastasis 8 months after MMS. ‡Patient 7 had a “squamous cell carcinoma” excised from the left knee 12 years ago. xPatient 29 developed regional lymph node metastasis 2 years after MMS. AAF, African American female; AAM, African American male; DOOC, died of other causes; F, female; FU, follow-up; HF, Hispanic female; M, male; Met, metastasis; NS, not stated; Rec, recurrences; WF, white female; WM, white male.
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
DERMATOLOGIC SURGERY
TABLE 1. Clinical Characteristics and Outcomes of 29 EPC Cases Treated With MMS
25/27 (93%) 0/27 (0)
managed by MMS in the English literature were retrieved by searching the PubMed from 1963 to 2013. Treatment outcomes from MMS were compared with historical treatment efficacy from WLE.3
Results The Authors’ Series
F, female; FU, follow-up; M, male; Met, metastasis; NS, not stated; Rec, recurrences.
29 Total (%)
27
2 (1–4) NS 3.8 y (1 mo–20 y) 65 y (36–86) M 52%; F 48%
44 m (2–206)
0/27 (0)
2/27 (7%)
0/27 (0)
14/15 0/15 19 m (2–48) 17 Reported cases
15
1.8 (1–3) NS 4.6 y (6 wk–20 y) 65 y (36–79) M 41%; F 59%
0/15
1/15
0/15
11/12 0/12 0/12 1/12 0/12 76 m (4–206) 2.9 y (1 mo–12 y) 32 mm (11–78) 2.2 (1–4) 66 y (36–86) M 67%; F 33% 12 12 The authors’ series
Sex (%)
Mean Duration Mean Age (y) Total Cases Cases With (N) FU
TABLE 2. Summary of 29 EPC Cases Treated With MMS
Mean Mean Defect Size Mohs (mm) Stages
Mean FU (mo)
DiseaseLocal Regional Distant Specific Rec Met (%) Met Death
Mohs Success Rate (%)
XU ET AL
Twelve EPC cases treated by MMS were summarized in Tables 1 and 2. The mean age at diagnosis was 66 years (range, 36–86). Sixty-seven percent of patients were male, and all were white. Among the 12 lesions, 3 were located on the lower extremities (25%), 3 on the chest (25%), 1 on the upper extremity (8%), and 5 on the head and neck areas (42%), with 1 each on the chin, antihelix, scalp, forehead, and nasal bridge. The mean lesion size was 16 mm (range, 6–47). The mean duration of lesion presence was 2.9 years (range, 1 month to 12 years). Patient 7 had a histologically diagnosed “squamous cell carcinoma of left medial knee” excised 12 years before the gradual recurrence of a subcutaneous nodule that was further diagnosed as EPC (Table 1 and Figure 1). Before histologic diagnosis, Patients 8 and 11 had cryotherapy and Patient 10 had topical 5-fluorouracil. The lesions had various but nonspecific morphologies. Four patients (25%) were immunocompromised. Patient 6 had renal transplant, Patient 9 had renal and pancreatic transplant, and Patient 11 was HIV-positive. Clinical presentation was shown for Patients 7 and 12 (Figures 1 and 2). Before MMS, all cases had established diagnosis of EPC through a skin biopsy using routine hematoxylin and eosin–stained paraffin-embedded sections. Two patients were initially misdiagnosed as basal cell carcinoma (BCC)/squamous cell carcinoma (SCC) (Patient 6) and SCC (Patient 12). Eccrine porocarcinoma is notoriously known for being overdiagnosed or misdiagnosed; therefore, original biopsy slides and frozen section Mohs slides were retrieved for all 12 cases and reviewed by 2 board-certified dermatopathologists (B.J.L. and M.A.H.). The diagnosis of EPC was confirmed for all cases. All were invasive EPCs.
41:6:JUNE 2015
687
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
Figure 1. Eccrine porocarcinoma. Patient 7 presented with a slowly growing subcutaneous nodule on the left medial knee, where a “squamous cell carcinoma” was excised 12 years ago. (A) Before MMS. (B) Defect after the completion of MMS. (C) Defect repaired by primary linear closure.
Case 12 had adjacent poroma. Case 11 had concurrent in situ and invasive component. Three cases (25%, Patients 6, 9, and 12) had numerous mitoses greater than 14/high-power field (HPF). Seven lesions (58%, Patients 1, 3, 4, 5, 7, 11, and 12) had an infiltrative border, and the rest of 5 patients (42%) had a pushing border. Necrosis en masse was seen in half cases (50%, Patients 5, 6, 7, 9, 11, and 12). Four cases (33%, Patients 2, 4, 6, and 9) had moderate cytological atypia, and the remaining 8 patients (67%) had well-differentiated cytology. All contained mature ducts. Lymphovascular invasion, perinueral invasion, and Pagetoid spread were not identified in sections reviewed. Representative histology was shown for Patients 7, 9, and 12 (Figures 3, 4, and 5). Generally, perioperative imaging such as a computed tomography (CT) scan, as well as sentinel lymph node biopsy, was not routinely offered to the patients treated before 2010. Patient 12 did undergo CT scan of the abdomen and pelvis, neither of which showed meta-
static disease, and the patient declined sentinel lymph node biopsy. All patients had MMS as monotherapy (Tables 1 and 2). The mean Mohs stages required to achieve a tumorfree plane were 2.2 (range, 1–4). The depth of excision was at least deep subcutaneous fat or fascia. The mean postoperative size was 32 mm (range, 11–78). There had been no local recurrences, distant metastasis, or disease-specific death, with a mean follow-up of 6 years (range, 4–206 months). Excluding 3 patients who died from other causes (Patients 2, 3, and 6), the remaining 9 patients had a follow-up of 3 to 17 years (range, 37–206 months). One patient (Patient 4) who had EPC on the left lateral chest 4 years before presentation developed left axillary regional lymph node metastasis 8 months after MMS. Node dissection showed metastatic EPC involving 8 of 29 nodes, with extracapsular extension. No local recurrence or distant metastasis was noted. The patient underwent lymphadenectomy and electron beam radiation therapy without chemotherapy. There was no evidence of locoregional tumor recurrence 17 years postoperatively.
Figure 2. Eccrine porocarcinoma. Patient 12 presented with a dusky red, slightly hyperpigmented plaque on the ankle for 2 years (Photograph courtesy of Dr. Thomas Rozum). The central ulcer was resulted from the recent biopsy. (A) Before MMS. (B) Defect after the completion of MMS. (C) Defect repaired by a full-thickness skin graft.
688
DERMATOLOGIC SURGERY
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
XU ET AL
Figure 3. Frozen section evaluation of the lesion from Patient 9 before MMS showed characteristic histology of EPC. Within the epidermis and extending into the dermis, there was a lobular neoplasm composed of poromatous basaloid cells that were hyperchromatic, pleomorphic with ductal differentiation, numerous mitoses, scattered atypical large multinucleated nuclei, and focal necrosis. Hematoxylin and eosin, original magnifications ·20 (A); ·40 (B); ·100 (C); ·400 (D).
Cases Reported by Others Seventeen reported cases of EPC treated by MMS were summarized in Tables 1 and 2. The mean age at diagnosis was 65 years (range, 36–79). Forty-one percent of patients were male. Three patients were African American, one was Hispanic, and the rest
were White. Seven of the 17 lesions were located on the lower extremities (41%), 3 on the trunk (18%), 1 on the upper extremity (6%), and 6 (35%) on the head and neck areas, with 1 each on the antihelix, scalp, temple, and upper cutaneous lip and 2 on the eyelid. The size of defects was not reported in most cases. The mean duration of lesion presence was 4.6 years (range, 6 weeks to 20 years). Patients 21 and 26 had recurrence after previous excisions. Clinical presentations varied, and all cases were reportedly diagnosed as EPC histologically. The mean Mohs stages were 1.8 (range, 1–3). Fifteen patients for whom follow-up was available showed no local recurrences, distant metastasis, or diseasespecific death, with a mean follow-up of 19 months (range, 2–48). One patient died from other causes (Patient 22). Another (Patient 29) developed left axil-
Figure 4. Frozen section evaluation of the subcutaneous nodule from Patient 7 before MMS exhibited scar tissue overlying an infiltrative neoplasm that resembles squamous cell carcinoma. The presence of ducts lined by malignant cells indicates the correct diagnosis of EPC. Hematoxylin and eosin, original magnification ·20.
lary regional lymph node metastasis 2 years after MMS excision of an EPC on the left ventral forearm. No local recurrence or distant metastasis was noted. The patient underwent a radical left axillary lymphadenectomy and postoperative radiation and remained
41:6:JUNE 2015
689
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
controlled trial is feasible and the understanding of its clinical characteristics and treatment options is extrapolated from a few retrospective case series and case reports. Mohs micrographic surgery has been reported effective in excising early EPC in a few case reports. Presented here is the single largest case series of 29 EPC tumors treated by MMS, in an attempt to assess the patient and tumor characteristics and clinical outcomes. The epidemiology and tumor presentations of the 29 patients treated by MMS are compatible with what has been reported. The mean age of onset was 65 years, with no clear predilection of gender (Table 2). The duration from tumor onset until treatment averaged 3.8 years (range, 1 month to 20 years). Eccrine porocarcinoma most frequently arises from the lower extremity,3 could potentially arise from any cutaneous location and, in the authors’ case series, was the most common on the head and neck (42%).
Figure 5. (A) Permanent section evaluation of the original shave biopsy from Patient 12 and (B) frozen section evaluation of the residual plaque before MMS revealed an infiltrative basaloid tumor connected to the epidermis with a lobulated architecture, necrosis, and ductal differentiation. Hematoxylin and eosin, original magnification ·20 (A); ·40 (B).
disease-free clinically and radiographically at her 6-month follow-up. Summary of 29 Cases Twenty-nine patients were treated by MMS as monotherapy, and outcome was established in 27 cases. There were no local recurrences, distant metastasis, or disease-specific death over a mean follow-up of 6 years (the authors’ series) and 19 months (reported cases). The regional metastatic rates to lymph nodes were 7% (2/27). Success rate of MMS as a primary treatment for EPC was 93% (25/27).
Discussion Eccrine porocarcinoma, a rare malignant tumor of acrosyringium, poses a considerable management challenge to clinicians. Because of its rarity, no random
690
The delay in the diagnosis and treatment of EPC is multifactorial, for EPC can be mistaken for other cutaneous tumors clinically and histologically. Clinically, EPC often presents as a pink or flesh-colored papule with a nondescriptive morphology and therefore a broad differential diagnoses, including seborrheic keratosis, pyogenic granuloma, nonmelanoma skin cancer, amelanotic melanoma, extramammary Paget disease, and metastatic cancer.13 Eccrine porocarcinoma may develop de novo or result from malignant transformation of a preexisting poroma, often heralded by ulceration, bleeding, growth, or symptoms of pruritus or tenderness.3,13 The diagnosis of EPC is based on histological confirmation. Microscopically, EPC is a malignant tumor connected to the epidermis and consists of asymmetric cords and lobules of characteristic poromatous basaloid cells with ductal differentiation and cytological atypia.1,14 Acrosyringeal differentiation is present, recognized by the presence of multiple lumina lined with a cuticular border. Tumor cells are often hyperchromatic with nuclear polymorphism and periodic acid–Schiff-positive with diastase digestible granules in their cytoplasms. Necrosis and mitoses are often seen. Pagetoid infiltration of the overlying epidermis is occasionally seen. Approximately 10% to 15% of
DERMATOLOGIC SURGERY
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
XU ET AL
EPC tumors are in situ. Up to 18% of EPC are contiguous with histologic zones of benign poroma and are therefore presumed to have arisen from that preexisting poroma.1,15 Tumor cell variants include basaloid, squamous, mucinous, clear cells, pigmented, and spindle cells, and this varied morphology, particularly in context of a partial biopsy, may contribute to the relatively high rates of misdiagnosis as SCC, adenocarcinoma, sebaceous carcinoma, metastasis, Paget disease, melanoma, and sarcoma.15 Two studies reported that 37%16 and 50%17 of cases had been misdiagnosed in their initial histological evaluation. In the authors’ series, 2 cases were initially misinterpreted as BCC/SCC and SCC (17%). In the largest published case series evaluating clinicopathological correlation of EPC, Robson and colleagues1 examined 69 cases and noted that 3 histological features, tumor depth over 7 mm, mitoses greater than 14/HPF, and the presence of lymphovascular invasion, were independently predictive of death. Moreover, after adjusting for mitosis and depth, an infiltrative border, not a pushing border, was strongly predictive of local recurrence. The poor prognosis of an infiltrative border was confirmed by a second study, where 24 EPC cases were divided into pushing, infiltrative, and pagetoid subtypes.16 The local, regional, and distant recurrences for the aforementioned 3 subtypes were 0/7, 4/10, and 2/2 cases, respectively. The authors attempted to closely examine the histological features of EPCs treated by MMS. All 17 reported cases were confirmed by routine histology but some lacked histological details. Patient 29, the case with regional metastasis, did possess several adverse features including a depth of 10 mm, lobular infiltrative growth, and numerous mitoses. In the 12 cases, 3 (25%) had numerous mitoses greater than 14/HPF. Seven (58%) had an infiltrative border. Lymphovascular invasion was not identified and the depth of tumor was not assessable because most cases were transected at the base of partial biopsies. Although it is unfeasible to objectively compare all histological details of the cases to others, the authors had a fair collection of cases with infiltrative border (58%) and numerous mitoses (25%).
Immunohistochemical staining is not commonly done when a diagnosis can be rendered by routine hematoxylin and eosin staining, but it can be helpful to confirm acrosyringeal differentiation of EPC and help differentiate EPC from its mimickers. Common markers used with various degrees of expression include cytokeratin, epithelial membrane antigen, carcinoembryonic antigen, p53 protein, and p16.18 Management of EPC remains controversial because there is no consensus on treatment. Most patients undergo conventional local excision, often with wide margins. However, a review of 105 cases of EPCs treated by WLE showed 20% local recurrence, 20% regional and 12% distant metastatic rates, and a mortality rate over 65% in patients with nodal metastases.3 Similarly, in Robson’s series, outcome was established in 54 cases that yielded a local recurrence of 17%, regional metastasis of 19%, distant metastasis of 11%, and mortality of 7%.1 In contrast, in 27 EPC patients treated by MMS as monotherapy, there were no local recurrences, distant metastasis, or disease-specific death over a mean follow-up of 6 years (the cases) and 19 months (reported cases). The regional metastatic rates to lymph nodes were 7% (2/27). Success rate of MMS as primary treatment for EPC was 93% (25/27). The patient who had regional axillary lymph node metastasis survived up to 17 years after lymphadenectomy and electron beam radiation therapy. The reported case with regional metastasis had also been disease-free, albeit during a shorter follow-up period of 6 months. In the study of 24 EPCs with 3 histologic subtypes (pushing, infiltrative, and pagetoid), recurrences seemed to be associated with histologic subtypes but not with the surgical margin taken, specifically $2 cm (6 patients),
BACKGROUND Eccrine porocarcinoma (EPC) has a poor prognosis after standard wide local excision (WLE), with 20% local recurrence, 20% regional and 12% distant metastatic rates. Mohs micrographic surgery (MMS) has been used as a promising treatment. OBJECTIVE
To review the use of MMS for EPC and assess treatment outcomes.
MATERIALS AND METHODS This was a retrospective case series of 12 EPC patients treated by MMS between 1984 and 2013 in the institution. Furthermore, a literature review revealed an additional 17 cases of EPC managed by MMS. RESULTS Of 29 cases of EPC treated by MMS, outcome was established in 27 cases. The patients had a significantly longer mean follow-up period of 6 years (range, 4–206 months), as compared with 19 months (range, 2–48 months) in reported cases. Two patients had regional lymph node metastasis after MMS. The regional metastatic rates to lymph nodes were 7% (2/27). There was no local recurrence, distant metastasis, or disease-specific death in the 27 cases studied. CONCLUSION To the best of the authors’ knowledge, this is the single largest case series of EPCs treated by MMS and the authors’ data demonstrated that MMS may be superior to the standard WLE. The authors have indicated no significant interest with commercial supporters.
E
ccrine porocarcinoma (EPC) is a rare malignant tumor of the intraepidermal (acrosyringeal) portion of the eccrine gland that represents only 0.005% to 0.01% of cutaneous neoplasms.1 Less than 300 cases of EPC have been reported since the original description by Pinkus in 1963 as “epidermotropic eccrine carcinoma.”2
Treatment of EPC remains challenging. Most patients are managed by wide local excision (WLE), and the reported local recurrence, and regional and distant metastatic rates are 20%, 20%, and 12%, respectively. Mortality rate is over 65% in patients with nodal metastases.3 Mohs micrographic surgery (MMS) has been shown effective in eradicating early EPC.3 The authors
present the clinical features and outcomes of 12 EPCs treated by MMS in the authors’ clinic, and 17 cases reported in the English literature.4–11 Methods This is a retrospective case series of 12 EPCs treated by MMS between January 1984 and September 2013 in the authors’ clinic. The study was approved by the University of Wisconsin Institutional Review Board Committee. Patient 1 was previously published by the senior author. Three patients died of other causes. The remaining 8 patients had routine follow-up skin examination by general dermatologists, and all were contacted by phone in September 2013 for the final confirmation of no recurrences or death. All 17 EPCs
*Department of Dermatology, University of Wisconsin, Madison, Wisconsin; †Department of Dermatology, Northwest Permanente, Portland, Oregon © 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved. ISSN: 1076-0512 Dermatol Surg 2015;41:685–692 DOI: 10.1097/DSS.0000000000000382
·
·
685
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
686
Pt
Age (y)/ Sex
Tumor Location
Tumor Duration
Mohs Stages
Defect (mm)
Local Rec
Regional Met
Distant Met
1* 2
FU (mo) Outcome
Author (Reference)
Year Published
36/WM
Chin
1.5 y
3
15 · 15
No
No
No
60
86/WM
Antihelix
NS
2
20 · 15
No
No
No
54
Alive
This study
1992
DOOC
This study
3
86/WM
Scalp
NS
2
18 · 20
No
No
No
16
DOOC
This study
2015 2015
4
73/WF
Chest
4y
3
24 · 56
No
Yes‡
No
206
Alive
This study
2015
5
71/WM
Chest
NS
1
31 · 24
No
No
No
127
Alive
This study
2015
6
71/WF
Chest
3 mo
1
90 · 65
No
No
No
4
DOOC
This study
2015
7
77/WM
Knee
12 y†
3
56 · 60
No
No
No
96
Alive
This study
2015
8
64/WM
Forehead
>1 mo
2
11 · 11
No
No
No
85
Alive
This study
2015
9
49/WF
Upper arm
NS
2
27 · 24
No
No
No
81
Alive
This study
2015
10
56/WM
Nose
1y
4
20 · 20
No
No
No
75
Alive
This study
2015
11
59/WF
Thigh
2.5 y
1
24 · 20
No
No
No
65
Alive
This study
2015
12
59/WM
Ankle
2y
3
45 · 45
No
No
No
37
Alive
This study
2015
13
71/AAF
Antihelix
20 y
1
NS
No
No
No
12
Alive
McMichael4
1999
14
79/WF
Foot
NS
2
NS
No
No
No
36
Alive
Wittenberg5
1999
15
74/WM
Scalp
NS
2
NS
No
No
No
48
Alive
Wittenberg5
1999
16
67/WF
Shin
NS
2
NS
No
No
No
12
Alive
Wittenberg5
1999
17
39/WF
Foot
NS
1
NS
No
No
No
5
Alive
Wittenberg5
1999
18
63/WF
Temple
NS
2
NS
No
No
No
12
Alive
Wittenberg5
1999
19
78/WF
Back
NS
1
NS
No
No
No
12
Alive
Johr6
2003
20
36/M
Chest
NS
2
NS
No
No
No
36
Alive
Wildemore7
2004
21
74/F
Shin
>10 y
2
NS
No
No
No
31
Alive
Wildemore7
2004
22
77/F
Thigh
NS
2
NS
No
No
No
29
DOOC
Wildemore7
2004
23
54/M
Ankle
NS
2
NS
No
No
No
15
Alive
Wildemore7
2004
24
71/M
Eyelid
6 wk
2
11 · 5
NS
NS
NS
NS
NS
D’Ambrosia8
2004
25
56/AAM
Back
1–2 y
3
NS
NS
NS
NS
NS
NS
Cowden9
2006
26
71/AAF
Heel
>1 y
NS
NS
No
No
No
2
Alive
Cowden9
2006
27
70/WM
Eyelid
3 mo
NS
9 · 14
No
No
No
6
Alive
Jain10
2008
28
63/M
Upper lip
4 mo
1
NS
No
No
No
12
Alive
Kwah11
2012
29
59/HF
Forearm
3y
NS
NS
No
Yesx
No
6
Alive
Vleugels12
2012
Case 1 to 12: from the authors’ series; cases 13 to 29: cumulative results from the literature. *Patient 1 was previously reported by Snow and Reizner3 in 1992. †Patient 4 developed regional lymph node metastasis 8 months after MMS. ‡Patient 7 had a “squamous cell carcinoma” excised from the left knee 12 years ago. xPatient 29 developed regional lymph node metastasis 2 years after MMS. AAF, African American female; AAM, African American male; DOOC, died of other causes; F, female; FU, follow-up; HF, Hispanic female; M, male; Met, metastasis; NS, not stated; Rec, recurrences; WF, white female; WM, white male.
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
DERMATOLOGIC SURGERY
TABLE 1. Clinical Characteristics and Outcomes of 29 EPC Cases Treated With MMS
25/27 (93%) 0/27 (0)
managed by MMS in the English literature were retrieved by searching the PubMed from 1963 to 2013. Treatment outcomes from MMS were compared with historical treatment efficacy from WLE.3
Results The Authors’ Series
F, female; FU, follow-up; M, male; Met, metastasis; NS, not stated; Rec, recurrences.
29 Total (%)
27
2 (1–4) NS 3.8 y (1 mo–20 y) 65 y (36–86) M 52%; F 48%
44 m (2–206)
0/27 (0)
2/27 (7%)
0/27 (0)
14/15 0/15 19 m (2–48) 17 Reported cases
15
1.8 (1–3) NS 4.6 y (6 wk–20 y) 65 y (36–79) M 41%; F 59%
0/15
1/15
0/15
11/12 0/12 0/12 1/12 0/12 76 m (4–206) 2.9 y (1 mo–12 y) 32 mm (11–78) 2.2 (1–4) 66 y (36–86) M 67%; F 33% 12 12 The authors’ series
Sex (%)
Mean Duration Mean Age (y) Total Cases Cases With (N) FU
TABLE 2. Summary of 29 EPC Cases Treated With MMS
Mean Mean Defect Size Mohs (mm) Stages
Mean FU (mo)
DiseaseLocal Regional Distant Specific Rec Met (%) Met Death
Mohs Success Rate (%)
XU ET AL
Twelve EPC cases treated by MMS were summarized in Tables 1 and 2. The mean age at diagnosis was 66 years (range, 36–86). Sixty-seven percent of patients were male, and all were white. Among the 12 lesions, 3 were located on the lower extremities (25%), 3 on the chest (25%), 1 on the upper extremity (8%), and 5 on the head and neck areas (42%), with 1 each on the chin, antihelix, scalp, forehead, and nasal bridge. The mean lesion size was 16 mm (range, 6–47). The mean duration of lesion presence was 2.9 years (range, 1 month to 12 years). Patient 7 had a histologically diagnosed “squamous cell carcinoma of left medial knee” excised 12 years before the gradual recurrence of a subcutaneous nodule that was further diagnosed as EPC (Table 1 and Figure 1). Before histologic diagnosis, Patients 8 and 11 had cryotherapy and Patient 10 had topical 5-fluorouracil. The lesions had various but nonspecific morphologies. Four patients (25%) were immunocompromised. Patient 6 had renal transplant, Patient 9 had renal and pancreatic transplant, and Patient 11 was HIV-positive. Clinical presentation was shown for Patients 7 and 12 (Figures 1 and 2). Before MMS, all cases had established diagnosis of EPC through a skin biopsy using routine hematoxylin and eosin–stained paraffin-embedded sections. Two patients were initially misdiagnosed as basal cell carcinoma (BCC)/squamous cell carcinoma (SCC) (Patient 6) and SCC (Patient 12). Eccrine porocarcinoma is notoriously known for being overdiagnosed or misdiagnosed; therefore, original biopsy slides and frozen section Mohs slides were retrieved for all 12 cases and reviewed by 2 board-certified dermatopathologists (B.J.L. and M.A.H.). The diagnosis of EPC was confirmed for all cases. All were invasive EPCs.
41:6:JUNE 2015
687
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
Figure 1. Eccrine porocarcinoma. Patient 7 presented with a slowly growing subcutaneous nodule on the left medial knee, where a “squamous cell carcinoma” was excised 12 years ago. (A) Before MMS. (B) Defect after the completion of MMS. (C) Defect repaired by primary linear closure.
Case 12 had adjacent poroma. Case 11 had concurrent in situ and invasive component. Three cases (25%, Patients 6, 9, and 12) had numerous mitoses greater than 14/high-power field (HPF). Seven lesions (58%, Patients 1, 3, 4, 5, 7, 11, and 12) had an infiltrative border, and the rest of 5 patients (42%) had a pushing border. Necrosis en masse was seen in half cases (50%, Patients 5, 6, 7, 9, 11, and 12). Four cases (33%, Patients 2, 4, 6, and 9) had moderate cytological atypia, and the remaining 8 patients (67%) had well-differentiated cytology. All contained mature ducts. Lymphovascular invasion, perinueral invasion, and Pagetoid spread were not identified in sections reviewed. Representative histology was shown for Patients 7, 9, and 12 (Figures 3, 4, and 5). Generally, perioperative imaging such as a computed tomography (CT) scan, as well as sentinel lymph node biopsy, was not routinely offered to the patients treated before 2010. Patient 12 did undergo CT scan of the abdomen and pelvis, neither of which showed meta-
static disease, and the patient declined sentinel lymph node biopsy. All patients had MMS as monotherapy (Tables 1 and 2). The mean Mohs stages required to achieve a tumorfree plane were 2.2 (range, 1–4). The depth of excision was at least deep subcutaneous fat or fascia. The mean postoperative size was 32 mm (range, 11–78). There had been no local recurrences, distant metastasis, or disease-specific death, with a mean follow-up of 6 years (range, 4–206 months). Excluding 3 patients who died from other causes (Patients 2, 3, and 6), the remaining 9 patients had a follow-up of 3 to 17 years (range, 37–206 months). One patient (Patient 4) who had EPC on the left lateral chest 4 years before presentation developed left axillary regional lymph node metastasis 8 months after MMS. Node dissection showed metastatic EPC involving 8 of 29 nodes, with extracapsular extension. No local recurrence or distant metastasis was noted. The patient underwent lymphadenectomy and electron beam radiation therapy without chemotherapy. There was no evidence of locoregional tumor recurrence 17 years postoperatively.
Figure 2. Eccrine porocarcinoma. Patient 12 presented with a dusky red, slightly hyperpigmented plaque on the ankle for 2 years (Photograph courtesy of Dr. Thomas Rozum). The central ulcer was resulted from the recent biopsy. (A) Before MMS. (B) Defect after the completion of MMS. (C) Defect repaired by a full-thickness skin graft.
688
DERMATOLOGIC SURGERY
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
XU ET AL
Figure 3. Frozen section evaluation of the lesion from Patient 9 before MMS showed characteristic histology of EPC. Within the epidermis and extending into the dermis, there was a lobular neoplasm composed of poromatous basaloid cells that were hyperchromatic, pleomorphic with ductal differentiation, numerous mitoses, scattered atypical large multinucleated nuclei, and focal necrosis. Hematoxylin and eosin, original magnifications ·20 (A); ·40 (B); ·100 (C); ·400 (D).
Cases Reported by Others Seventeen reported cases of EPC treated by MMS were summarized in Tables 1 and 2. The mean age at diagnosis was 65 years (range, 36–79). Forty-one percent of patients were male. Three patients were African American, one was Hispanic, and the rest
were White. Seven of the 17 lesions were located on the lower extremities (41%), 3 on the trunk (18%), 1 on the upper extremity (6%), and 6 (35%) on the head and neck areas, with 1 each on the antihelix, scalp, temple, and upper cutaneous lip and 2 on the eyelid. The size of defects was not reported in most cases. The mean duration of lesion presence was 4.6 years (range, 6 weeks to 20 years). Patients 21 and 26 had recurrence after previous excisions. Clinical presentations varied, and all cases were reportedly diagnosed as EPC histologically. The mean Mohs stages were 1.8 (range, 1–3). Fifteen patients for whom follow-up was available showed no local recurrences, distant metastasis, or diseasespecific death, with a mean follow-up of 19 months (range, 2–48). One patient died from other causes (Patient 22). Another (Patient 29) developed left axil-
Figure 4. Frozen section evaluation of the subcutaneous nodule from Patient 7 before MMS exhibited scar tissue overlying an infiltrative neoplasm that resembles squamous cell carcinoma. The presence of ducts lined by malignant cells indicates the correct diagnosis of EPC. Hematoxylin and eosin, original magnification ·20.
lary regional lymph node metastasis 2 years after MMS excision of an EPC on the left ventral forearm. No local recurrence or distant metastasis was noted. The patient underwent a radical left axillary lymphadenectomy and postoperative radiation and remained
41:6:JUNE 2015
689
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
POROCARCINOMA TREATED BY MOHS
controlled trial is feasible and the understanding of its clinical characteristics and treatment options is extrapolated from a few retrospective case series and case reports. Mohs micrographic surgery has been reported effective in excising early EPC in a few case reports. Presented here is the single largest case series of 29 EPC tumors treated by MMS, in an attempt to assess the patient and tumor characteristics and clinical outcomes. The epidemiology and tumor presentations of the 29 patients treated by MMS are compatible with what has been reported. The mean age of onset was 65 years, with no clear predilection of gender (Table 2). The duration from tumor onset until treatment averaged 3.8 years (range, 1 month to 20 years). Eccrine porocarcinoma most frequently arises from the lower extremity,3 could potentially arise from any cutaneous location and, in the authors’ case series, was the most common on the head and neck (42%).
Figure 5. (A) Permanent section evaluation of the original shave biopsy from Patient 12 and (B) frozen section evaluation of the residual plaque before MMS revealed an infiltrative basaloid tumor connected to the epidermis with a lobulated architecture, necrosis, and ductal differentiation. Hematoxylin and eosin, original magnification ·20 (A); ·40 (B).
disease-free clinically and radiographically at her 6-month follow-up. Summary of 29 Cases Twenty-nine patients were treated by MMS as monotherapy, and outcome was established in 27 cases. There were no local recurrences, distant metastasis, or disease-specific death over a mean follow-up of 6 years (the authors’ series) and 19 months (reported cases). The regional metastatic rates to lymph nodes were 7% (2/27). Success rate of MMS as a primary treatment for EPC was 93% (25/27).
Discussion Eccrine porocarcinoma, a rare malignant tumor of acrosyringium, poses a considerable management challenge to clinicians. Because of its rarity, no random
690
The delay in the diagnosis and treatment of EPC is multifactorial, for EPC can be mistaken for other cutaneous tumors clinically and histologically. Clinically, EPC often presents as a pink or flesh-colored papule with a nondescriptive morphology and therefore a broad differential diagnoses, including seborrheic keratosis, pyogenic granuloma, nonmelanoma skin cancer, amelanotic melanoma, extramammary Paget disease, and metastatic cancer.13 Eccrine porocarcinoma may develop de novo or result from malignant transformation of a preexisting poroma, often heralded by ulceration, bleeding, growth, or symptoms of pruritus or tenderness.3,13 The diagnosis of EPC is based on histological confirmation. Microscopically, EPC is a malignant tumor connected to the epidermis and consists of asymmetric cords and lobules of characteristic poromatous basaloid cells with ductal differentiation and cytological atypia.1,14 Acrosyringeal differentiation is present, recognized by the presence of multiple lumina lined with a cuticular border. Tumor cells are often hyperchromatic with nuclear polymorphism and periodic acid–Schiff-positive with diastase digestible granules in their cytoplasms. Necrosis and mitoses are often seen. Pagetoid infiltration of the overlying epidermis is occasionally seen. Approximately 10% to 15% of
DERMATOLOGIC SURGERY
© 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
XU ET AL
EPC tumors are in situ. Up to 18% of EPC are contiguous with histologic zones of benign poroma and are therefore presumed to have arisen from that preexisting poroma.1,15 Tumor cell variants include basaloid, squamous, mucinous, clear cells, pigmented, and spindle cells, and this varied morphology, particularly in context of a partial biopsy, may contribute to the relatively high rates of misdiagnosis as SCC, adenocarcinoma, sebaceous carcinoma, metastasis, Paget disease, melanoma, and sarcoma.15 Two studies reported that 37%16 and 50%17 of cases had been misdiagnosed in their initial histological evaluation. In the authors’ series, 2 cases were initially misinterpreted as BCC/SCC and SCC (17%). In the largest published case series evaluating clinicopathological correlation of EPC, Robson and colleagues1 examined 69 cases and noted that 3 histological features, tumor depth over 7 mm, mitoses greater than 14/HPF, and the presence of lymphovascular invasion, were independently predictive of death. Moreover, after adjusting for mitosis and depth, an infiltrative border, not a pushing border, was strongly predictive of local recurrence. The poor prognosis of an infiltrative border was confirmed by a second study, where 24 EPC cases were divided into pushing, infiltrative, and pagetoid subtypes.16 The local, regional, and distant recurrences for the aforementioned 3 subtypes were 0/7, 4/10, and 2/2 cases, respectively. The authors attempted to closely examine the histological features of EPCs treated by MMS. All 17 reported cases were confirmed by routine histology but some lacked histological details. Patient 29, the case with regional metastasis, did possess several adverse features including a depth of 10 mm, lobular infiltrative growth, and numerous mitoses. In the 12 cases, 3 (25%) had numerous mitoses greater than 14/HPF. Seven (58%) had an infiltrative border. Lymphovascular invasion was not identified and the depth of tumor was not assessable because most cases were transected at the base of partial biopsies. Although it is unfeasible to objectively compare all histological details of the cases to others, the authors had a fair collection of cases with infiltrative border (58%) and numerous mitoses (25%).
Immunohistochemical staining is not commonly done when a diagnosis can be rendered by routine hematoxylin and eosin staining, but it can be helpful to confirm acrosyringeal differentiation of EPC and help differentiate EPC from its mimickers. Common markers used with various degrees of expression include cytokeratin, epithelial membrane antigen, carcinoembryonic antigen, p53 protein, and p16.18 Management of EPC remains controversial because there is no consensus on treatment. Most patients undergo conventional local excision, often with wide margins. However, a review of 105 cases of EPCs treated by WLE showed 20% local recurrence, 20% regional and 12% distant metastatic rates, and a mortality rate over 65% in patients with nodal metastases.3 Similarly, in Robson’s series, outcome was established in 54 cases that yielded a local recurrence of 17%, regional metastasis of 19%, distant metastasis of 11%, and mortality of 7%.1 In contrast, in 27 EPC patients treated by MMS as monotherapy, there were no local recurrences, distant metastasis, or disease-specific death over a mean follow-up of 6 years (the cases) and 19 months (reported cases). The regional metastatic rates to lymph nodes were 7% (2/27). Success rate of MMS as primary treatment for EPC was 93% (25/27). The patient who had regional axillary lymph node metastasis survived up to 17 years after lymphadenectomy and electron beam radiation therapy. The reported case with regional metastasis had also been disease-free, albeit during a shorter follow-up period of 6 months. In the study of 24 EPCs with 3 histologic subtypes (pushing, infiltrative, and pagetoid), recurrences seemed to be associated with histologic subtypes but not with the surgical margin taken, specifically $2 cm (6 patients),
