Dermatology: Practical and Conceptual Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 1 Clinical Review of Mucosal Melanoma: The 11-Year Experience of a Referral Center Marco Spadafora1,2, Giacomo Santandrea1,3, Michela Lai1,2, Stafania Borsari2, Shaniko Kaleci4, Chiara Banzi5, Vincenzo Dario Mandato6, Giovanni Pellacani7, Simonetta Piana3, Caterina Longo2,4 1 Clinical and Experimental Medicine PhD Program, University of Modena and Reggio Emilia, Modena, Italy 2 Centro Oncologico ad Alta Tecnologia Diagnostica, Azienda Unità Sanitaria Locale - IRCCS di Reggio Emilia, Reggio Emilia, Italy 3 Pathology Unit, Azienda Unità Sanitaria Locale - IRCCS di Reggio Emilia, Reggio Emilia, Italy 4 Department of Dermatology, University of Modena and Reggio Emilia, Modena, Italy 5 Medical Oncology Unit, Azienda Unità Sanitaria Locale - IRCCS di Reggio Emilia, Reggio Emilia, Italy 6 Unit of Obstetrics and Gynecology, Azienda Unità Sanitaria Locale - IRCCS, Reggio Emilia, Italy 7 Dermatology Clinic, Department of Clinical Internal, Anesthesiological and Cardiovascular Sciences, Sapienza University of Rome, Rome, Italy Key words: mucosal melanoma, diagnosis, treatments, rare disease Citation: Spadafora M, Santandrea G, Lai M, et al. Clinical Review Of Mucosal Melanoma: The 11-Year Experience Of A Referral Center. Dermatol Pract Concept. 2023;13(1):e2023057. DOI: https://doi.org/10.5826/dpc.1301a57 Accepted: June 29, 2022; Published: January 2023 Copyright: ©2023 Spadafora et al. This is an open-access article distributed under the terms of the Creative Commons Attribution- NonCommercial License (BY-NC-4.0), https://creativecommons.org/licenses/by-nc/4.0/, which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original authors and source are credited. Funding: None. Competing Interests: None. Authorship: All authors have contributed significantly to this publication. Corresponding Author: Caterina Longo, Department of Dermatology, University of Modena and Reggio Emilia, Italy. Azienda Unità Sanitaria Locale – IRCCS di Reggio Emilia, Centro Oncologico ad Alta Tecnologia Diagnostica-Dermatologia, Reggio Emilia, Italy. Phone: +390522295612 E-mail: longo.caterina@gmail.com Introduction: Mucosal melanoma is a rare neoplasm. Late diagnosis is caused by occult anatomic sites and scarcity of symptoms. Novel biological therapies have now become available. Demographic, therapeutical and survival records on mucosal melanoma are scarce. Objectives: To provide an 11-year retrospective clinical review of real-world data on mucosal mela- nomas managed in a tertiary referral center in Italy. Methods: We included patients with histopathological mucosal melanoma diagnoses from January 2011 to December 2021. Data were collected until the last known follow-up or death. Survival anal- ysis was performed. Results: Among 33 patients, we found 9 sinonasal, 13 anorectal and 11 urogenital mucosal mela- nomas (median age 82, females 66.7%). Eighteen cases (54.5%) presented with metastasis (p<0.05). ABSTRACT 2 Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 Introduction Melanoma is a malignant tumor arising from melano- cytes [1]. Although melanocytes are mostly localized in the skin, their precursors reach also endodermal and ectoder- mal mucosae migrating from the neural crest [2]. Primary mucosal melanoma arises from mucosal membranes lining the head and neck (i.e. nasal and oral cavities), anorectal, vulvovaginal, and urinary tract in order of frequency [3,4]. The occult locations in which mucosal melanoma occurs preclude sun exposure as a predisposing risk factor; the eti- ologic factors driving tumorigenesis in mucosal melanoma have not been discovered yet [5]. Mucosal melanoma rep- resents 0.03% of all cancer diagnoses and 0.8-3.7% of all melanomas [5,6], with a higher incidence in women than men; different gender incidence is mainly due to vulvovagi- nal neoplasia, which represents alone 18% of mucosal mel- anomas [1]. Most patients with mucosal melanoma are diagnosed in a metastatic stage because of the late occurrence of symp- toms and the occult location of the primary tumor [7]. The most common symptoms in nasal cavity melanomas are unilateral nasal obstruction, mass lesion, and epistaxis while in the oral cavity symptoms such as swelling, ulcer- ation, bleeding, pain, or tooth mobility can occur. Anorectal melanomas usually manifest with rectal bleeding, anorectal discomfort, or prolapse of the tumor mass. In vulvovaginal melanomas presenting symptoms are bleeding, vulvar mass, pruritus, pain or irritation, micturition discomfort, and dis- charge [4]. There is no universal staging system for mucosal melano- mas. Head and neck mucosal melanomas are usually staged according to the American Joint Committee on Cancer (AJCC) criteria for head and neck cancer; vulvar melanoma can be staged following the AJCC criteria for cutaneous mel- anoma, while no staging criteria have been established for mucosal melanoma arising in the urethra, vagina, rectum, and anus [8]. Surgical excision with negative margins, which is the treatment of choice in mucosal melanomas, is often un- feasible because of an anatomically complex site of ori- gin [7,9]. Patients with unresectable or metastatic mucosal melanomas can be treated with the same regimen proposed for cutaneous melanoma [10], although the frequency of common driver BRAF is low compared to the cutaneous counterpart (50% vs 3-5%), with reduced usefulness of targeted therapy [1]. Mutation of the KIT gene is detected in about 25% of mucosal melanomas [1]; to date, guide- lines suggest KIT testing only when BRAF and, eventually, NRAS, mutational status have been established; KIT tar- geted therapy is usually administered as a second line ther- apy [8]. In the last few years, immune checkpoint inhibitors (ICIs) have become a preferred first-line approach for pa- tients with advanced or metastatic cutaneous melanoma. A recent review [7] showed that anti-CTLA-4 antibody ip- ilimumab has less efficacy as monotherapy than monoclo- nal antibodies targeting the PD-1 and PD-L1, which have proven more effective in the treatment of MMs, with pro- longed survival and acceptable toxicity. A sub-analysis of mucosal melanomas performed in a five-year survival trial showed similar data on efficacy [8,11]. The treatment regi- men currently authorized by Agenzia Italiana del Farmaco for advanced mucosal melanoma includes ipilimumab and anti-PD1 antibodies, while Imatinib is approved for unre- sectable metastatic melanoma in progression after immu- notherapy [8]. As primary mucosal melanoma is an exceedingly rare neoplasm, demographic, histopathological, therapeutical, and survival records on this topic are scarce. The current study aims to provide an 11-year retrospective clinical re- view of the real-world data on mucosal melanomas managed in a tertiary referral center in Italy. In the urogenital subgroup, only 4 patients (36.4%) had metastasis at diagnosis, all in regional lymph nodes. Sinonasal melanomas were surgically managed with a debulking procedure (44.4%); every case of anorectal and urogenital melanomas underwent radical surgery (30.8% and 45.5%). Fifteen patients were treated with biological therapy (p<0.05). Radiation therapy was used in all melanomas of the sinonasal region (p<0.05). Overall survival was longer for urogenital melanomas (26 months). Univariate analysis showed an increased hazard ratio for death in patients with metastasis. A negative prognostic value of metastatic status was reported by the multivariate model, while administration of first-line immunotherapy demonstrated a protective role. Conclusions: At diagnosis, the absence of metastatic disease is the most relevant factor that influences the survival of mucosal melanomas. Moreover, the use of immunotherapy might prolong the survival of metastatic mucosal melanoma patients. Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 3 Methods This was a retrospective study performed between 01/01/2011 to 31/12/2021. This study was approved by the Institutional Review Board of Azienda Unità Sanitaria Locale – IRCCS di Reggio Emilia, Italy (protocol number #2011/02347213). We included consecutive patients with a histopatho- logically confirmed diagnosis of mucosal melanomas. We excluded recurrent tumors, unknown primary melanomas, and cases for which histological slides were not available for re-evaluation. Data from the first clinical or instrumental diagnosis to the date of each patient’s last known follow-up appointment or death was obtained from digital medical records. We re- corded patient age, gender, location of the lesion, presenting symptoms, site of metastasis at diagnosis if any (locoregional lymph node involvement, cerebral, visceral, or multiple me- tastases when more than one of the previous sites was in- volved) and histopathological and molecular features (cell morphology, Breslow thickness and ulceration when not compromised by fragmentation or orientation of biopsy specimen, and mutational status). Surgery was recorded as debulking procedure or radical treatment. Systemic treatments were categorized according to current recommendations [8,10] as first-line biological treat- ment (nivolumab, pembrolizumab, or imatinib), as first-line chemotherapy when the patient was administered with sys- temic therapy in the pre-biological era, and as second-line treatments when therapy was switched to a different ther- apy because of disease progression. We also recorded if the patient underwent radiation therapy on the primary tumor, which was always managed with a cytoreductive-palliative purpose in our series [12]. Statistics Statistical analysis was performed using STATA® software version 17 (StataCorp, 4905 Lakeway Drive, College Sta- tion, Texas 77845 USA). Descriptive statistics were pre- sented for baseline demographic clinical characteristics for the entire group, as well as for the groups of patients with different locations. Continuous variables were presented as the number of patients (N), mean, standard deviation (SD), minimum  (min), and maximum (max) and compared be- tween subgroups using Unpaired Student’s t-test; categorical variables were presented as frequency (N, percentage [%]) and compared using Pearson’s chi‐squared test. Survival analysis was performed using the Kaplan-Meier method and comparison between the survival curves was done us- ing log-rank test. Univariate and multivariate analyses were done using the Cox-regression hazard model. Data from the univariate and multivariate regression analyses were ex- pressed as Hazard ratio (HR) with it 95% confidence in- terval (CI). A p<0.05 was considered statistically significant. Results Demographic, Clinical, and Treatment Data Among 33 patients with primary mucosal melanomas who were included in our analyses, we found 9 melanomas of the sinonasal region, 13 anorectal melanomas and 11 urogen- ital melanomas (of which 10 vulvovaginal melanomas and 1 bladder melanoma), as reported in Table 1 and Figure 1. Median age at diagnosis was 82 years (75-83), with no sub- stantial differences by anatomical site. More women than men had mucosal melanoma (n= 22, 66.7%). Median fol- low-up period was 11.5 months. Presenting symptoms for which physicians were con- sulted by the patient were recorded in Table 2 (in our series dermatologists, otolaryngologists, gynecologists, and endos- copists); most patients were then managed by the Skin Can- cer Tumor Board. We observed that 18 cases of mucosal melanoma (54.5%) out of 33 presented with metastasis at diagnosis (p<0.05) and that 91% and 66% of patients with anorectal and sin- onasal melanomas had at least one metastatic site at diagno- sis. In urogenital melanomas, 4 out of 11 (36.4%) showed metastases which were all diagnosed in locoregional lymph nodes (inguinofemoral nodes) and no distant metastases. Surgery: Clinical records reported that 39.4% of patients underwent surgical treatment (p<0.05). All the sinonasal melanomas which were surgically managed (44.4%) under- went a debulking procedure, while every surgically managed anorectal and urogenital melanoma had a radical intent pro- cedure (30.8% and 45.5%). Most patients did not undergo surgery because of metastasis at diagnosis (n=14), detection of different neoplasia at primary staging (n=2), unresectable tumor (n=3), age of the patient (n=1), death of the patient before surgery (n=1). Systemic treatment: We identified 15 patients treated with systemic biological therapy as first-line treatment (p<0.05); 13 received nivolumab, 2 subjects were treated with pembrolizumab and 1 subject with imatinib. Only 1 patient received first-line chemotherapy (cyclophosphamide, vincristine, and dacarbazine combination treatment). Furthermore, 5 patients were switched to a second-line systemic therapy; 3 of them were switched to ipilimumab, 1 to pembrolizumab and 1 to monotherapy with dacarbazine. Radiation therapy (RT): in our analysis, RT on the primary tumor has always been proposed as a cytoreductive-palliative treatment; it was performed in all cases of sinonasal mela- noma (p<0.05). Anorectal melanomas were treated with RT 4 Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 Table 1. Comparison of demographic, clinical, and survival data in patients with mucosal melanomas and stratification by the site of origin. Mucosal melanomas (n=33, 100%) Sinunasal (n=9, 27.3) Anorectal (n=13, 39.4%) Urogenital (n=11, 33.3%) p-value Age at diagnosis, median (IQR) 82 (75-85) 82 (75-83) 80 (72-87) 83 (80-85) 0.977 Female 22 (66.7) 5 (55,6) 6 (46,2) 11 (100) 0.015 Metastasis at diagnosis 18 (54.5) 4 (44,4) 10 (76.9) 4 (36.4) 0.021 Regional lymph nodes 8 (24.2) 0 (0.0) 4 (30.8) 4 (36.4) Visceral 4 (12.1) 3 (33.3) 1 (7.7) 0 (0.0) Cerebral 2 (6.1) 0 (0.0) 2 (15.4) 0 (0.0) ≥2 sites 4 (12.1) 1 (11,1) 3 (23.1) 0 (0.0) Surgical treatment 13 (39.4) 4 (44.4) 4 (30.8) 5 (45.5) 0.015Debulking 4 (12.1) 4 (44.4) 0 (0.0) 0 (0.0) Radical excision 9 (27.3) 0 (0.0) 4 (30.8) 5 (45.5) First-line biological therapy 17 (51.5) 5 (55.5) 6 (46.2) 5 (45.5) 0.005 Nivolumab 13 (39.4) 4 (44.4) 4 (30.8) 5 (45.5) Pembrolizumab 2 (6.1) 1 (11.1) 1 (7.7) 0 (0.0) Imatinib 1 (3.0) 0 (0.0) 1 (7.7) 0 (0.0) First-line chemotherapy 1 (3.0) 1 (11.1) 0 (0.0) 0 (0.0) 0.253 Second-line systemic therapy 5 (15.1) 2 (22.2) 2 (15.4) 1 (9.1) 0.383 Ipilimumab 3 (9.1) 2 (22.2) 1 (7.7) 0 (0.0) Pembrolizumab 1 (3.0) 0 (0.0) 1 (7.7) 0 (0.0) Dacarbazine 1 (3.0) 0 (0.0) 0 (0.0) 1 (9.1) Radiation therapy on the primary tumor 17 (51.5) 9 (100) 6 (46.2) 2 (18.2) 0.004 Overall survival, median (IQR) months 11 (6-25) 14 (6-22) 6 (2-11) 26 (11-34) 0.021 Overall survival rate%, 12 months 71 55 54 100 Overall survival rate%, 12 months 54 37 18 100 Median follow-up period (months) 11.5 - - - IQR: interquartile range Mucosal Melanoma n=33 Sinonasal Melanoma n=9 (27.3%) Anorectal Melanoma n=13 (39.4%) Urogenital Melanoma n=11 (33.3%) Figure 1. Flow diagram showing mucosal melanoma subdivided by anatomical site of origin. Table 2. Presenting symptoms of mucosal melanoma by anatomical region. Presenting symptoms (n) Site of primary melanoma Sinonasal Epistaxis (2), nasal obstruction (2), eyelid ptosis (1) Anorectal Rectorrhagia (4), tumor mass prolapse (2), anemia (2) Urogenital Tumor mass (6), pigmentation of external genitalia (1), vaginal bleeding (1) n= number(s) Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 5 cavities, melanocytes can be found both in the epithelium and in the stroma, mainly in the dark-skinned population. Delayed site-related detection of these melanomas can explain their architecture, which is often nodular or polyp- oid, and, at the same time, their thickness, which is usually higher than most cutaneous melanomas. Most mucosal melanomas are often associated with an in situ lateral spread on the mucosal surface (Figure 3), which is a supportive feature that the tumor is primitive. Cytologically, mucosal melanomas are remarkably variable; neoplastic cells can be spindle, rhabdoid, epithelioid, small, or giant pleomorphic (often multinucleated), thus causing challenging diagnostic problems in differentiating them from lymphomas, carcinomas or sarcomas. Necrotic areas are frequent and mitotic activity is usually high (Figure 4). Due to their heterogeneous appearance, immunohisto- chemical staining (S100, MART-1, HMB45 or SOX10) is often required to demonstrate the melanocytic origin of the neoplasm. In our series, epithelioid features were prevalent (25 out of 33 cases). Two cases were composed of spindle hyperpig- mented cells, two cases had a lymphocytic-like appearance, one was rhabdoid and, in three cases, the cytological pattern of growth was mixed (Figure 5). Mean Breslow thickness in mucosal melanoma was 6.7 with no appreciable difference between different anatomical sites; no Breslow thickness was reported in sinonasal mela- nomas due to the sparse and fragmented nature of biopsy specimens. Tumor ulceration was noted in 11 samples. In one case of vulvar melanoma, a residual melanocytic nevus was found at the periphery of the tumor. No samples undergoing biomolecular analysis had BRAF or NRAS mutation, while 3 cases had c-KIT and 3 cases KRAS mutations. in 46.3% of cases while only 2 patients with urogenital mel- anomas were treated with RT. Survival data: Overall survival (OS), expressed as a me- dian value, was longer in urogenital melanomas (26 months), than sinonasal melanomas (14 months) and anorectal mela- nomas (6 months). As reported in the Kaplan-Meier estimate (Figure 2), si- nonasal and anorectal melanoma carried higher mortality in the first two years; anorectal melanoma had a high mortality rate over time while sinonasal melanoma reached a plateau. Urogenital melanoma showed longer survival in the first two years from diagnosis with delayed mortality. The 24-month overall survival rate among mucosal mel- anoma patients was 54%. The 24-month overall mortality for the location was 37% in sinonasal melanomas, 18% in anorectal melanoma and 100% in urogenital melanoma. The last observed exit from the estimate was 58 months (Figure 2). The univariate analysis showed that metastasis at diag- nosis was significantly associated with mortality (Table 2). In the multivariate analysis, the risk factor significantly as- sociated with mortality was the presence of metastasis at diagnosis, while first-line immunotherapy demonstrated a protective role (Table 4). Histopathologic and Molecular Data Mucosal melanomas can occur in all sites where muco- sal melanocytes are present. While perianal, genital, and perioral skin normally harbors melanocytes (and, conse- quently, these are sites of origin of benign and malignant melanocytic tumors, albeit rare), melanocytes are usually absent in the bladder mucosa. In the rare case of bladder melanoma, a spread of melanocytes from the urethra could explain its etiopathogenesis. In the nasal and paranasal Analysis time Number at risk 33 12 3 0 6040200 0. 00 0. 25 0. 50 0. 75 1. 00 Kaplan–Meier survival estimateA) Sinonasal melanoma Number at risk Sinonasal melanoma Anorectal melanoma Urogeital melanoma 9 13 11 3 2 7 1 0 2 0 604020 Analysis time Kaplan–Meier survival estimatesB) 0 0. 00 0. 25 0. 50 0. 75 1. 00 0 0 Anorectal melanoma Urogintal melanoma Figure 2. Kaplan Meier estimates for overall survival. (A) OS in mucosal melanomas. (B) Stratification of the cohort by the site of the primary tumor. 6 Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 Table 3. Univariable model of risk factors for mortality in mucosal melanomas. HR 95% CI p-value Age at diagnosis 1.02 (0.97-1.07) 0.292 Gender Female ref. Male 1.93 (0.67-5.45) 0.216 Location Sinonasal ref. 1.90 (0.57-6.29) 0.291 Urogenital 0.25 (0.05-1.09) 0.066 Metastasis at diagnosis No ref. Yes 8.75 (1.79-42.76) 0.007 Site of metastasis at diagnosis Skin/regional lymph node ref. Visceral 3.30 (0.34-31.96) 0.302 Cerebral 6.21 (0.53-72.00) 0.144 ≥2 sites 2.50 (0.22-27.95) 0.456 Surgery No ref. Debulking 0.19 (0.02-1.59) 0.129 Radical 0.33 (0.09-1.16) 0.085 First-line biological therapy No ref. Nivolumab 0.88 (0.26-2.95) 0.840 Pembrolizumab 1.16 (0.13-9.63) 0.890 Imatinib 2.71 (0.31-23.21) 0.362 First-line chemotherapy No ref. Yes 1.84 (0.23-14.66) 0.561 Second-line systemic therapy No ref. Ipilimumab 0.71 (0.09-5.68) 0.752 Pembrolizumab-- Dacarbazine 1.42 (0.17-11.30) 0.740 Cytoreductive-palliative radiation therapy No ref. Yes 1.65 (0.56-4.90) 0.360 HR: hazard ratio CI: confidence interval Detailed descriptive data from histopathologic and mo- lecular data are reported in Table 5. Discussion This 11-year retrospective study shows that mucosal melanoma is a tumor typically arising in the older population, with a me- dian age of 82, not significantly influenced by site of origin [13]. Demographic data also confirm that mucosal melanoma has a higher prevalence in females [9]; this result is largely driven by cases of vulvovaginal melanoma in the urogenital melanoma subgroup. Surprisingly, we have no records of oral cavity muco- sal melanomas, which alone have been described as the second most frequent location in the head and neck region [2,8,9]. Histologically, most cases were made by epithelioid cells, regardless of the site of origin. In many cases, pleomorphic Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 7 Table 4. Multivariable model of risk factors for mortality in mucosal melanomas. HR 95% CI p-value Age at diagnosis 1.02 (0.96-1.09) 0.574 Gender Female ref. Male 3.33 (0.98-11.25) 0.107 Metastasis at diagnosis No ref. Yes 26.55 (3.18-221.42) 0.001 First-line biological therapy No ref. Nivolumab 0.10 (0.01-0.75) 0.025 Pembrolizumab 0.04 (0.01-0.91) 0.044 Imatinib 0.09 (0.0-2.02) 0.133 CI: confidence interval Figure 3. Vulvar melanoma (A) Clinical overview of a nodular hyperpigmented lesion of the vulva. (B) Dermoscopy of the nodular part shows a blue-black structureless area with shiny white structures and negative pigment network at the implant base. (C) Dermoscopy of the flat part shows blue and black globules and blotches, a blue-white veil, and shiny white structures. (D) Histology: the neoplasm is largely ulcerated, shows discrete necrotic areas, and reaches 11 mm thickness according to Breslow. At a higher power view, vascular invasion of neoplastic cells is visible (arrow). 8 Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 Figure 4. Architectural features of mucosal melanoma in hematoxylin and eosin. (A) A sinonasal melanoma growing as a polypoid, edematous, and highly vascularized mass. Melanoma involves the entire polyp with multiple confluent nodules, reaching at least 1 mm Breslow thickness. The transitional respiratory epithelium is still recognizable all around. (B) A nod- ular melanoma arising at the interface between intestinal and squamous perianal epithelium. Breslow thickness is 14 mm (C) A vaginal nodular melanoma without any visible in situ component in the squamous vaginal epithelium. The tumor is superficially ulcerated, and Breslow thickness is 9 mm. areas with bizarre, giant cells were present. Only two cases of spindle cell melanomas were recorded, even if spindle cell areas occurred in three cases with a mixed pattern of growth. The epithelioid features and the presence of melanin, which in rare cases can be massive, simplified the diagnosis; vice versa, differential diagnosis resulted challenging in two cases of lymphoma-like melanoma (one anorectal and one genital) and an immunohistochemical panel of stains was necessary to rule out a lymphoproliferative process. More than half of mucosal melanomas (54.4%) mani- fested with metastasis at diagnosis. Patients who were dis- covered with metastasis have an increased HR for death as highlighted by univariate analysis (Table 3); in addition, the prognostic value of metastatic status at diagnosis was reported by the multivariate model. These findings corrob- orate the evidence that diagnosing advanced-stage tumors has a relevant influence on prognosis [14]. As previously dis- cussed [9], the site of origin of neoplasia, mostly anatomically occult, is a relevant cause of late diagnosis. Incidence of mu- cosal melanomas in older age could be an additional cause, as older people can delay seeking physician consultation. The multivariate model also showed that the administra- tion of immunotherapy as a first-line treatment was a protec- tive factor for death, with a statistical significance both for nivolumab and pembrolizumab. No significance has been obtained for imatinib, because of the small sample. It is interesting to highlight that, despite the small sam- ple size in the current study, our data support the utility of Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 9 also reported [18,19] that females diagnosed with mucosal melanomas have higher survival, unrelated to the site of or- igin; thus, the longer survival in this cohort could also be driven by the prevalence of vulvovaginal melanoma in the subgroup. Surgical treatment with radical intent, undergone by all patients with urogenital melanoma in our series, may be an additional factor influencing longer survival. No case of male urogenital melanomas was reported in the database of the Pathology Unit of our hospital confirm- ing that mucosal melanoma of the male genital tract is a very rare occurrence [4]. Urogenital melanomas in our database manifested as a vegetating mass, new pigmentation of the external genitalia, or urogenital bleeding (Table 2). Anorectal Melanoma Patients with anorectal melanoma showed a younger age at diagnosis (Table 1). immunotherapy to prolong survival in metastatic mucosal mel- anoma [7]. A recent international cohort study [15] remarked on the efficacy of the anti-PD1 agent in prolonging survival in metastatic patients, even though to a lesser extent than in cutaneous melanoma. In fact, mucosal melanoma has been re- ported to have a lower expression of PD-L1 than its cutaneous counterpart [16] and of being more tolerogenic [17] probably because of the absent mutagenic role of UV radiation that pre- vents a high tumor mutational load and immunogenicity. Urogenital Melanoma In our series, urogenital melanoma showed 4 cases of metas- tasis in locoregional lymph nodes and no distant metastasis at diagnosis. Moreover, urogenital melanoma is the sub- group with the longer OS, with full survival at 24 months and decreased survival only after 26 months (Figure 2). The absence at diagnosis of disseminated disease is probably the most relevant factor that influences cohort survival. It was Figure 5. Unusual cytological features of mucosal melanoma in hematoxylin and eosin. (A) Pleomorphic areas in an epithelioid vulvar melanoma. Melanocytes are irregularly shaped, and nuclei are bizarre. The arrow indicates an abnormal mitotic figure. (B) A hyperpigmented genital melanoma, composed of middle-sized, mitotically active spindle cells. (C) A rectal melanoma made up of small, monomorphous lymphocytic-like cells. Immunohistochemical staining is necessary to confirm the melanocytic origin of the tumor (in the inset, HMB45 diffusely stains the cells). (D) A rectal melanoma with a rhabdoid appearance. Neoplastic cells show a large, eosinophilic cytoplasm and peripherally located round nucleus. 10 Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 Additional evidence emerging from Kaplan-Meier analy- sis is that the mortality of sinonasal mucosal melanomas was higher in the first 24 months and then decreased, reaching a plateau (Figure 2). Factors that may influence this behav- ior, such as radiation therapy, administered to all cohort pa- tients, are still unclear and need to be further analyzed in a larger sample. Sinonasal melanomas in our series manifested with ep- istaxis, nasal obstruction, and one case of eyelid ptosis (Table 2). Conclusions In conclusion, different strategies have to be developed to avoid late diagnosis; so far, there are no targeting cam- paigns or specific advice that can help physicians from dif- ferent specialties (i.e. gynecologists) to get early diagnoses of this rare melanoma subsets. As promising results, the use of immunotherapy might impact the natural history of mucosal melanomas in improving overall or disease-free survival. Even though all patients surgically managed in this sub- group underwent a radical intent procedure, Kaplan-Meier analysis showed the worst survival. This result is consistent with previous reports [15,18]. It is reasonable to suppose that a more widespread metastatic disease at diagnosis in this subgroup is the reason for this difference (Table 1). It was demonstrated that anorectal melanoma has an intrinsically aggressive behavior because of the high propensity to develop distant and brain metastases [18]; in our series, this subgroup is the only one manifesting with 2 cases of cerebral metastasis and with 3 cases of multiple visceral metastases at diagnosis. Anorectal melanomas diagnosed in our clinics had pre- senting symptoms such as rectal bleeding, anemia, and pro- lapse of the tumor mass (Table 2). Sinonasal Melanoma In the sinonasal region, Breslow thickness is not easy to ap- ply because of biopsy sampling. In fact, all surgical treat- ments proposed to this cohort of patients were debulking procedures that usually prevent obtaining a well-oriented full-thickness specimen for histopathological analysis. Table 5. Mucosal melanoma pathological data and stratification by the site of origin. Mucosal melanomas (n=33, 100%) Sinunasal (n=9, 27.3%) Anorectal (n=13, 39.4%) Urogenital (n=11, 33.3%) Breslow, n, mean ± SD (range) 10, 6.7 ±5.2 (0.6-17) - 3, 7.1 ±5.3 (1.4-12) 7, 6.5 ±5.6 (0.6-17) Ulceration No 1 (3.0) 0 (0.0) 0 1 (9.1) Yes 11 (33.3) 0 (0.0) 5 (38.5) 6 (54.5) Cytology Epithelioid cells 25 (75.7) 9 (100) 10 (76.9) 6 (54.5) Spindle cells 2 (6.1) 0 (0.0) 1 (7.6) 1 (9.1) Small cells 2 (6.1) 0 (0.0) 1 (7.6) 1 (9.1) Rhabdoid cells 2 (6.1) 0 (0.0) 1 (7.6) 0 (0.0) Mixed 3 (9.1) 0 (0.0) 0 (0.0) 3 (27.2) LVI No 7 (21.2) 1 (11.1) 1 (7.7) 5 (45.5) Yes 5 (15.2) 1 (11.1) 2 (15.4) 2 (18.2) BRAF, wt 25 (75.8) 8 (88.9) 10 (76.9) 7 (63.6) c-KIT V560D 3 (9.1) 0 (0.0) 1 (7.7) 2 (18.2) wt 8 (24.2) 2 (22.2) 5 (38.5) 1 (9.1) NRAS, wt 9 (27.3) 3 (33.3) 4 (30.8) 2 (18.2) KRAS p.A146V 1 (3.0) 0 (0.0) 1 (7.7) 0 (0.0) p.G12A 1 (3.0) 1 (11.1) 0 (0.0) 0 (0.0) p.G12C 1 (3.0) 0 (0.0) 1 (7.7) 0 (0.0) wt 1 (3.0) 1 (11.1) 0 (0.0) 0 (0.0) LVI: lymphovascular invasion Original Article | Dermatol Pract Concept. 2023;13(1):e2023057 11 11. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Five-Year Sur- vival with Combined Nivolumab and Ipilimumab in Ad- vanced Melanoma. N Engl J Med. 2019;381(16). doi:10.1056 / NEJMoa1910836. PMID: 31562797. 12. NCCN Guidelines Version 3.2022 Melanoma: Cutaneous. Up- dated April 2022. Accessed July 2022. https://www.nccn.org /professionals/physician_gls/pdf/cutaneous_melanoma.pdf 13. Gutiérrez-Castañeda LD, Nova JA, Tovar-Parra JD. Frequency of mutations in BRAF, NRAS, and KIT in different populations and histological subtypes of melanoma: a systemic review. Melanoma Res. 2020;30(1):62-70. doi:10.1097/CMR.0000000000000628. PMID: 31274706. 14. Kuk D, Shoushtari AN, Barker CA, et al. Prognosis of Muco- sal, Uveal, Acral, Nonacral Cutaneous, and Unknown Primary Melanoma From the Time of First Metastasis. The Oncologist. 2016;21(7):848-854. doi:10.1634/theoncologist.2015-0522 15. Dimitriou F, Namikawa K, Reijers ILM, et al. Single-agent anti-PD-1 or combined with ipilimumab in patients with mucosal melanoma: an international, retrospective, cohort study. Ann Oncol Off J Eur Soc Med Oncol. Published online June 16, 2022:S0923-7534(22)01731-8. doi:10.1016/j.an- nonc.2022.06.004. PMID: 35716907. 16. Kaunitz GJ, Cottrell TR, Lilo M, et al. Melanoma subtypes demonstrate distinct PD-L1 expression profiles. Lab Investig J Tech Methods Pathol. 2017;97(9):1063-1071. doi:10.1038 / labinvest.2017.64. PMID: 28737763. 17. Goldemberg DC, Thuler LCS, de Melo AC. An Update on Mu- cosal Melanoma: Future Directions. Acta Dermatovenerol Croat ADC. 2019;27(1):11-15. doi:. 18. Heppt MV, Roesch A, Weide B, et al. Prognostic factors and treatment outcomes in 444 patients with mucosal melanoma. Eur J Cancer Oxf Engl 1990. 2017;81:36-44. doi:10.1016/j. ejca.2017.05.014. PMID: 28600969. 19. Cinotti E, Chevallier J, Labeille B, et al. Mucosal melanoma: clin- ical, histological and c-kit gene mutational profile of 86 French cases. J Eur Acad Dermatol Venereol JEADV. 2017;31(11): 1834-1840. doi:10.1111/jdv.14353 References 1. Broit N, Johansson PA, Rodgers CB. Meta-Analysis and System- atic Review of the Genomics of Mucosal Melanoma. Mol Cancer Res MCR. 2021;19(6):991-1004. doi:10.1158/1541-7786. MCR-20-0839. PMID: 33707307 2. López F, Rodrigo JP, Cardesa A. Update on primary head and neck mucosal melanoma. Head Neck. 2016;38(1):147-155. doi:10.1002/hed.23872. PMID: 25242350. 3. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer. 1998;83(8):1664-1678. doi:10.1002 /(sici)1097-0142(19981015)83:8<1664::aid-cncr23>3.0.co;2-g. PMID: 9781962 4. Mihajlovic M, Vlajkovic S, Jovanovic P, Stefanovic V. Primary mucosal melanomas: a comprehensive review. Int J Clin Exp Pathol. 2012;5(8):739-753. doi:. 5. Spencer KR, Mehnert JM. Mucosal Melanoma: Epidemiology, Biology and Treatment. Cancer Treat Res. 2016;167:295-320. doi:10.1007/978-3-319-22539-5_13 6. Yde SS, Sjoegren P, Heje M, Stolle LB. Mucosal Melanoma: a Literature Review. Curr Oncol Rep. 2018;20(3):28. doi:10.1007 /s11912-018-0675-0. PMID: 29569184. 7. Li J, Kan H, Zhao L, Sun Z, Bai C. Immune checkpoint inhibi- tors in advanced or metastatic mucosal melanoma: a systematic review. Ther Adv Med Oncol. 2020;12:1758835920922028. doi:10.1177/1758835920922028. PMID: 32489431. 8. Linee guida melanoma AIOM 2021. AIOM. Published Septem- ber 23, 2021. Updated January 2022. Accessed January 2022./ https://www.aiom.it/linee-guida-aiom-2021-melanoma/ 9. Patrick RJ, Fenske NA, Messina JL. Primary mucosal mela- noma. J Am Acad Dermatol. 2007;56(5):828-834. doi:10.1016/j .jaad.2006.06.017. PMID: 17349716. 10. Seth R, Messersmith H, Kaur V, et al. Systemic Therapy for Mel- anoma: ASCO Guideline. J Clin Oncol Off J Am Soc Clin Oncol. 2020;38(33):3947-3970. doi:10.1200/JCO.20.00198. PMID: 32228358.