Rare Tumors of Skin: Malignant Cutaneous Adnexal Tumors and Clinicopathological Features

Abstract

Background:

Cutaneous adnexal tumors (CATs) are a large group of neoplasms of the skin and the pathological diagnosis can be challenging.

Objective:

The aim of this study was to examine clinicopathological features of malignant cutaneous adnexal tumors (MCATs) and to identify therapeutic and prognostic implications.

Materials and Methods:

Twenty-six patients diagnosed with MCATs and operated in our center were included. Demographic and clinical characteristics of patients, tumor size, histological subtype, perineural and lymphovascular invasion, re-excision status, lymph node metastasis, distant metastasis, recurrence, follow-up, and treatment choice were recorded. All lesions were excised with 1 to 2-cm surgical margin. Surgical margin was determined according to tumor size.

Results:

Of 26 patients, 14 were males and 12 were females (median age: 66 years). Median follow-up was 52.08 months. Tumor location was the face in 69.2% (n = 18) patients, followed by scalp and neck region. Four patients had coexisting skin tumors. Majority of MCATs (69.2%, n = 18) originated from sweat glands. Median tumor diameter was 4 (range, 0.8 to 17) cm. There was a statistically significant correlation between presence of perineural invasion and advanced disease (P = 0.036). Majority of the patients had Stage 1-2 disease, while two patients had Stage 4 disease. In six patients (23.1%), re-excision was performed due to surgical margin positivity.

Conclusion:

Incidence of MCATs has an increasing trend with advanced age and these tumors are more commonly seen in men and in visible sites of the body such as head and neck and scalp. Wide resection according to tumor size is the first-line treatment.

Introduction

Cutaneous adnexal tumors (CATs) are a large group of neoplasms of the skin and the pathological diagnosis can be challenging. Some of benign tumors are of clinical importance; i.e., multiple trichilemmomas may be a precursor of malignant tumors of the visceral organs or, in Cowden syndrome, breast carcinomas may develop. According to the 2017 World Health Organization (WHO) classification, the CATs are mainly divided into three groups based on their clinical, histological, and genetic features: sweat gland (apocrine and eccrine), hair follicle, and sebaceous gland origin.[1] Malignant CATs (MCATs) are rarer than benign CATs and they can be misdiagnosed as metastatic diseases due to overlapping clinical and pathological features. According to the Surveillance, Epidemiology, and End Results (SEER) database, the incidence of MCATs has increased to 5.1 per million by increasing 150% over the last three decades, showing a male predominance.[2] In the literature, many case reports and case series regarding MCATs highlight the importance of MCATs and the United States (US) National Health data have provided more information about these rare tumors.[2,3,4]

Due to their rare occurrence, the etiological factors of MCATs have not been fully understood, yet. The majority of these tumors have been thought to develop de novo, while a minority of them result from malignant transformation of a benign precursor, and some of them originate from nevus sebaceous.[5,6] In addition, CATs have been reported in autosomal-dominant Brooke-Spiegler and Muir-Torre syndromes.[7,8] Although rare, MCATs can be heterotopic and may arise from the extracutaneous organs such as breast, salivary gland, gastrointestinal system, respiratory system, oral cavity, or tongue.[9]

Across the geographic areas including Turkey, there are no comprehensive studies investigating clinicopathological features of MCATs in the literature. In the present study, we, therefore, aimed to examine clinicopathological features of MCATs and to identify therapeutic and prognostic implications in the light of literature data.

Materials and Methods

This single-center, retrospective study was conducted at Department of Plastic, Reconstructive and Aesthetic Surgery of a tertiary care center between January 1, 2013, and January 31, 2020. The study protocol was approved by the Institutional Review Board (No: 09.2020.540). A written informed consent was obtained from each patient for all diagnostic and therapeutic procedures. The study was conducted in accordance with the principles of the Declaration of Helsinki.

A total of 26 patients aged >18 years who were diagnosed with MCATs and operated in our center were included. Immunosuppressive patients were excluded from the study. Data including demographic and clinical characteristics of the patients, tumor size, histological subtype, perineural and lymphovascular invasion, re-excision status, lymph node metastasis, distant metastasis, coexisting skin tumors, recurrence, follow-up duration, and treatment choice (i.e., surgical, radiation therapy or chemotherapy) were retrospectively reviewed. All diagnoses were revised according to the 2017 WHO classification[1] and clinical and pathological Tumor, Node, Metastasis staging was performed.[10]

All lesions were excised with 1 to 2-cm surgical margin. Surgical margin was determined according to the tumor size as follows: 1 cm for <1-cm tumors, 1.5 cm for 1 to 2-cm tumors, and 2 cm for 2 to 4-cm tumors and ≥4-cm tumors.

Statistical analysis was performed using the SPSS version for Windows 15.0 software (SPSS Inc., Chicago, IL, USA). Continuous variables were expressed in mean ± standard deviation (SD), median (min-max), while categorical variables were expressed in number and percentage. The unpaired t test was used to compare continuous random variables between the time periods, while the Chi-square test was used to compare discrete random variables. A P value of <0.05 was considered statistically significant.

Results

Of a total of 26 patients included in the study, 14 were males and 12 were females with a median age of 66 (range, 33 to 90) years. The tumor location was the face in 69.2% (n = 18) of the patients, followed by scalp and neck region [Figure 1]. Four patients had coexisting skin tumors (squamous cell carcinoma [SCC], n = 2; basal cell carcinoma [BCC], n = 1; basosquamous cell carcinoma, n = 1). None of the patients had accompanying syndromes such as Muir-Torre syndrome.

Figure 1.

Tumor localizations

The majority of MCATs (69.2%, n = 18) originated from the sweat glands [Figure 1]. The rate of MCATs arising from hair follicles [Figure 2a and b] or sebaceous gland was lower [Table 1]. The median tumor diameter was 4 (range, 0.8 to 17) cm.

Figure 2.

(a) Sweat gland origin (Adenocarcinoma) present at the parietal region. The tumor size is 2*1,8*0,6 cm. (b) Malignant pilar tumor present at the occipital region. This was the biggest tumor of the study (17*14*9,5 cm) posterior view, lateral view

Table 1.

Distribution of tumor types

Tumor type	n	%
Sweat gland origin	18	69,2
Adenocarcinoma (non-specified)	9	34,6
Microcystic adnexal carcinoma	1	3,8
Porocarcinoma	3	11,5
Hidradenocarcinoma	4	15,4
Adenoid cystic carcinoma	1	3,8
Hair follicle origin	4	15,4
Malignant trichoblastoma	1	3,8
Malignant pilar tumor	3	11,5
Sebaceous gland origin	4	15,4
Sebaceous carcinoma	4	15,4
Total	26	100

Data are given in number and percentage, unless otherwise stated

Demographic characteristics of the patients, tumor stage, lymph node metastasis, distant metastasis, perineural invasion, lymphovascular invasion, re-excision status, and recurrence rates are summarized in Table 2. At the time of diagnosis, seven patients (26.9%) had a regional lymph node metastasis, while only one patient (3.8%) had a distant metastasis (distant lymph node metastasis). The majority of the patients had Stage 1-2 disease, while two patients had Stage 4 disease.

Table 2.

Demographic characteristics of the patients and univariate analysis of the MCATs’

Variable	Mean (SD)
Age, Year	66 (±18,2)
Follow-Up duration (month)	43,57 (±26,75)
	Number	%
Sex
Male	14	53,8
Female	12	46,2
Primary site
Scalp	7	26,9
Face and neck	11	42,3
Trunk	2	7,7
Upper extremities	3	11,5
Lower extremities	3	11,5
T Stage
Tx	-
T0	-
Tis	-
T1	9	34,6
T2	11	42,3
T3	5	19,2
T4	1	3,8
N Stage
Nx	-
N0	19	73,1
N1	7	26,9
N2	-
N3
M Stage
M0	25	96,2
M1	1	3,8
Stage
Stage I	9	34,6
Stage Ii	6	23,1
Stage Iii	9	34,6
Stage Iva	1	3,8
Stage Ivb	1	3,8
Perineural invasion
Yes	8	30,8
No	18	69,2
Lymphovascular invasion
Yes	3	11,5
No	23	88,5
Re-excision
Yes	6	23,1
No	20	76,9
Recurrence
No	24	92,3
Local	2	7,7
Regional	-
Distant	-

SD=Standard deviation. Data are given in mean±SD or number and percentage, unless otherwise stated

In six patients (23.1%), re-excision was performed due to surgical margin positivity. None of the patients received postoperative radiation therapy or chemotherapy. The median follow-up was 52.08 (range, 5 to 96) months. Two patients (7.7%) had local recurrence. Both of them were females aged 70 years and the tumors were localized in the head and neck region. The diagnoses were Stage 2 and Stage 4 sweat gland carcinomas in each. The follow-up of these two patients was 75 and 95 months.

Eight of the patients died during follow-up; however, only one died from MCAT. This patient had Stage 3 disease with a follow-up period of 76 months at the time of diagnosis. Due to the low number of deceased patients, disease-free survival and overall survival were unable to be estimated.

There was a statistically significant correlation between the presence of perineural invasion and advanced disease (P = 0.036), while no significant correlation was observed between lymph node metastasis, distant metastasis, and recurrence (P = 0.635, P = 1.000, P = 0.529). Lymphovascular invasion was not correlated with any other variables including advanced disease, lymph node metastasis, distant metastasis, and recurrence (P = 0.458, P = 0.540, P = 1.000, P = 1.000).

All patients were evaluated for clinical and radiological presence of regional lymph node metastasis. In two patients suspected of having radiological signs, lymph node dissection was performed which revealed metastatic disease. After a median follow-up of 8.6 (range, 3 to 14) months, five patients were suspected of having a lymph node metastasis based on imaging studies and lymph node dissection was performed. Pathological examination also revealed a metastatic disease in these patients. In 57.1% of the patients with a lymph node metastasis, tumor type was reported as an eccrine tumor. In the remaining patients with a lymph node metastasis, histological subtypes of the tumor were reported as a malignant pilar tumor, porocarcinoma, and sebaceous carcinoma.

Discussion

The fact that MCATs have a broad range of histological subtypes and can be easily misdiagnosed as non-melanoma skin cancers complicate the clinical and histopathological diagnosis.[11,12] Due to their rare occurrence, there is no consensus regarding the tumor behavior (recurrence and lymph node and distant metastasis), and treatment and follow-up. Current evidence has been obtained from SEER database[2,4] and two large-scale European studies,[13,14] and case reports and case series, as well as a limited number of small-scale studies. Therefore, we believe that our study provides valuable information to the body of knowledge on this special group of tumors.

In a recent European study, one- and five-year overall survival rates were reported as 95% and 84, respectively.[13] According to the SEER database, the five-year overall survival rate was 96% for cutaneous appendageal carcinomas and 97% for apocrine-eccrine carcinomas.[2] Both aforementioned studies have shown an increasing trend for overall survival, despite the increasing number of MCAT cases in recent years. As in our study, this can be attributed to the sophisticated imaging modalities to detect small tumors and early diagnosis. In addition, in line with the growing evidence regarding the tumor behaviors, tumor complexity can be quantified and appropriate treatments can be applied without delay.

Although there are several studies showing that MCATs can affect both sexes equally,[3] many authors have reported a male predominance.[2,15] Blake et al.[2] found that the incidence of MCATs was similar between male and female patients up to 50 years of age; however, its incidence significantly increased in men after 50 years of age. Similarly, in our study, the incidence of MCATs was 53.2% in men with a slight increase among those aged 50 years.

The median age of onset of MCATs has been reported as ≥60 years, consistent with our study.[2,16] In their study, Oyasiji et al.[15] showed that the incidence of MCATs increased in ≥60 years of age and advanced age (≥60 years) was an independent predictor of poor overall and recurrence-free survival. In our study, 61.5% of our patients were >60 years and 50% were >70 years of age. Consistent with the previous findings, the incidence of MCATs shows an increasing trend with advanced age.[2]

Previous studies have demonstrated that MCATs mostly involve the head and neck region with a varying range from 52 to 65.2%.[2,3,4,15] In the present study, this rate was slightly higher (69.2%), and face was the most commonly affected site, followed by scalp and neck region. It was the least commonly seen in the trunk, consistent with the literature.[2] The fact that head and neck region is the most commonly affected body site with an increasing incidence in recent years can be explained by the increased exposure to ultraviolet (UV) radiation from the sun and that malignant melanoma and non-malignant melanoma skin cancers are possibly linked to the exposure to UV radiation.[13] In a retrospective, population-based study including patients with head and neck sweat gland adenocarcinomas, scalp, and neck tumors were the significant prognostic indicators of disease-specific survival.[17] In our study, only patient died from an eccrine-type MCAT localized in the head and neck region. However, due to the low sample size, we cannot draw a firm conclusion on disease-free survival and overall survival in this group of patients.

In the present study, the most common type of MCATs was sweat gland tumors, similar to the European data, accounting for about 60% of all MCATs.[13] According to the US data, these cases are comprised of about 50% of all MCATs.[2] These regional differences suggest that UV exposure and racial backgrounds may be implicated in the etiology of MCATs.

In our study, the vast majority of the tumors (76.9%) were smaller than 4 cm. Unlike our findings, previous studies have shown a significantly lower rate (about 39%).[3,4] However, these data are related to registry databases and the number of cases with an unknown tumor size is relatively high due to missing data (Tx = 56.6%). In a study, Oyasiji et al.[15] found that 26% of the tumors were Tx tumors (i.e., main tumor cannot be evaluated due to the lack of information) and 62% of the tumors were smaller than 4 cm, consistent with our study. This can be attributed to the majority of tumors being localized in the visible parts of the body (i.e., head-neck and scalp) and early hospital admission.

The rate of lymph node metastasis at the time of diagnosis was 7.7% in our study, consistent with the literature (2 to 12%).[4,15] Elective lymph node dissection and sentinel lymph node biopsy (SLNB) are still controversial in the treatment of MCATs. Ogata et al.[18] reported a high metastasis rate up to 89% after elective lymph node dissection in apocrine carcinomas. The authors recommended lymph node screening in these patients. However, Michl et al.[19] found no micro-metastasis in eight patients with MCATs via SLNB, while other authors reported micro-metastasis in three of five patients (60%) through SLNB, indicating the diagnostic value of SLNB. In our study, all patients were evaluated clinically and radiologically for a regional lymph node metastasis and lymph node dissection was performed in patients with radiologically confirmed lymph node positivity. The patients without clinical lymph node metastasis were followed and therapeutic lymph node metastasis was decided in five patients (19.2%) with a lymph node metastasis after a median follow-up of 8.6 (range, 3 to 14) months. As the lymph node positivity was detected in 26.7% at the time of diagnosis and during follow-up in our study and lymph node positivity was shown to be associated with overall survival, disease-free survival, and recurrence-free survival in previous studies,[4,15] we believe that lymph node status should be evaluated. In the clinical practice, there are two ways of clinical and radiological lymph node evaluation: elective lymph node dissection and SLNB. There is no consensus on the location of elective lymph node dissection in MCATs. In patients with malignant melanoma, elective lymph node dissection is associated with high morbidity rates and, therefore, SLNB is the most widely adopted dissection method.[20] This is also the case for MCATs. There is a limited number of data regarding the complementary lymph node dissection in SLNB-positive cases. However, as these tumors are usually refractory to chemotherapy and radiation therapy,[21] complementary lymph node dissection can be recommended, particularly in patients with poor prognostic factors such as advanced age and large tumor size.[15] In the present study, the majority of the metastatic tumors (57.1%) are located in the sweat gland, indicating that histopathological lymph node involvement should be evaluated in MCATs.

Previous studies have reported a distant metastasis rate of 0 to 8.5%.[2] In our study, distant metastasis was identified in only one patient (3.8%) at the time of diagnosis. As regional lymph node metastasis and distant metastasis affect the survival adversely,[13] all MCATs should be evaluated at the time of diagnosis and using imaging studies during follow-up.

Although MCATs may invade the perineural spaces,[22] its exact incidence is still unclear. In their study, Oyasiji et al.[15] found perineural invasion in two patients (4%). In our study, however, this rate is extremely high as 30.8%. Perineural invasion is a poor prognostic factor for cutaneous SCCs and can be detected in 2.5 to 14% of patients.[23,24] The local recurrence rate of SCCs with perineural invasion has been estimated as 16 to 45% in the literature.[24] However, in our study, this rate was relatively lower (12.5%). Also, previous studies have reported the lymph node metastasis rate in SCC with perineural invasion as 10 to 50%,[24] we found a higher rate in our patients (37.5%). Also, we found that the presence of perineural invasion was associated with advanced disease (P < 0.05). This finding suggests that perineural invasion may be a poor prognostic factor for MCATs, similar to cutaneous SCCs.

Radiotherapy is recommended in non-operable MCATs with perineural invasion.[25] However, some authors have reported that radiotherapy is associated with poor prognosis or has no contributory effect to survival.[12] In addition, chemotherapy is thought to have a limited effect in MCATs and even many tumors show chemoresistance. Although no specific chemotherapy regimen has been developed for MCATs yet, doxorubicin, 5-fluorouracil, vincristine, and anthracycline can be used as monotherapy or combination therapy in patients with distant metastases.[21]

Surgical resection is the most commonly implemented treatment modality in MCATs.[2,4,13] In particular, wide resection is associated with decreased recurrence rates.[26] In a review, Nazemi et al.[27] reported that wide resection was the most preferred method in the treatment of eccrine porocarcinomas; however, the Mohs micrographic surgery could yield more favorable results. In the present study, we performed surgical resection in all patients and re-resection in six patients (23.1%) with tumor continuation in the surgical margins or proximity to the adjacent tissues. It is well known that surgical margin is of utmost importance in all skin cancers, as in malignant melanomas. However, there is no consensus regarding the exact surgical margins for each tumor size. In our study, the surgical margin was determined according to the tumor size as follows: 1 cm for <1-cm tumors, 1.5 cm for 1 to 2-cm tumors, and 2 cm for 2 to 4-cm tumors, and ≥4-cm tumors. The local recurrence rate was 7.7% in our patients and was lower than previous studies (12%).[15] This finding indicates that a wide surgical resection according to the tumor size may be helpful to prevent local recurrence.

The main limitation of the present study is that no standardized approach for treatment and follow-up was able to be implemented, due to the rare occurrences and a broad range of histological subtypes and terminological diversity of MCATs. The retrospective design with a relatively small sample size is another limitation.

In conclusion, our study results suggest that the incidence of MCATs has an increasing trend with advanced age and these tumors are more commonly seen in men and in visible sites of the body such as head and neck and scalp. As coexisting tumors (BCCs and SCCs) are associated with UV exposure, UV exposure has been implicated in the etiology of MCATs. Wide resection according to the tumor size is the first-line treatment. Furthermore, as perineural invasion is associated with advanced disease, it should be evaluated, as in other skin tumors such as SCC, and its prognostic value should be examined. Finally, lymph node dissection is recommended in case of lymph node positivity based on clinical and radiological findings and SLNB should be performed in case of micro-metastasis, if lymph nodes are negative.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.