Genitourinary melanoma: An overview for the clinician

Abstract

Genitourinary (GU) melanoma is a rare presentation of melanoma accounting for approximately 0.5% of all melanomas. GU melanomas include primary melanomas of the vulva, vagina, uterine cervix, ovary, penis, scrotum, urethra, bladder, ureter, and kidney. These melanomas are often diagnosed in advanced stages and stigma is thought to contribute to delays in presentation. As the likely diagnosing provider, it is imperative that dermatologists, urologists, and gynecologists are aware of these uncommon sites of presentation. While there have been major advances in the treatment of melanomas as a whole in the last 10 years, their applications to GU melanomas have often been overlooked. GU melanomas have not been included in many of the major phase III clinical trials which brought contemporary advanced treatments to market and the prognoses for GU melanomas remain poor. Due to the rarity of GU melanomas, much of the literature provides generalized recommendations across multiple different organs affected by GU melanomas or omits certain topics, making it difficult to appreciate the fundamentals of the individual presentations. This review aimed to provide background information on the pathogenesis and epidemiology of the different sites of GU melanomas and categorize data specific to the presentation, staging, treatment, and prognosis of each type of GU melanoma to guide the clinician. It was also meant to encourage a multidisciplinary approach to the management of these patients as it spans the expertise of surgical oncologists, medical oncologists, radiation oncologist, dermatologists, urologists, and gynecologists.

1. Introduction

Primary melanoma of the genitourinary (GU) tract is a type of melanoma that usually arises on mucosal surfaces of the female genital, male genital, and urinary tract, but can also arise on epidermal skin bearing surfaces [1,2]. All melanomas are thought to derive from neural crest cells; however, cutaneous melanomas are believed to arise from epidermal or dermal melanocytes, and mucosal melanomas are theorized to arise from melanoblasts [3,4]. Mucosal melanomas vary from cutaneous melanomas on a molecular level as well [5,6]. In cutaneous melanomas, BRAF somatic mutations are present in approximately 52%–66%, neuroblastoma RAS (NRAS) in 28%, and neurofibromin 1 (NF1) in 14%, while only 14% are wild-type for these oncogenes [7,8]. Comparatively, mucosal melanomas most frequently have mutations in KIT (up to 39%), NRAS (up to 18%), and NF1 (up to 16%), with BRAF mutations in only 0%–16% [5,6,9].

The presence of BRAF and KIT mutations in GU melanomas has been evaluated by multiple studies. In female GU melanomas, Hou et al. [10] found BRAF mutations in 26% of 51 cases of vaginal and vulvar melanomas, while they were observed in only 8.3% of 105 cases of non-GU mucosal melanomas which differs from prior findings of BRAF mutations in 4.3% of 416 pooled cases of vaginal and vulvar melanomas. Additionally, they found KIT mutations in 22% of vaginal and vulvar melanomas, compared to 8.8% in non-GU mucosal melanomas and 3% in non-GU cutaneous melanomas. NRAS mutations were found in 4% of vaginal and vulvar melanomas compared to 26% of all non-GU melanomas. Wylomanski et al. [11] also evaluated vulvovaginal melanomas and found BRAF mutations in 33% of the 15 cases of vulvar melanomas and none of seven cases of vaginal melanomas while KIT were present in 7% and 14%, respectively. No NRAS mutations were observed in vulvar melanomas, but they were present in 43% of primary vaginal melanomas. Saglam et al. [12] evaluated 13 female GU melanomas for mutations; BRAF mutations were observed in 23% (all vulvar primaries); KIT mutations were observed in 31% (equally divided between vulvar and vaginal primaries); and NRAS mutations were observed in 15% (all vulvar primaries). Both male and female mucosal melanomas were evaluated by Omholt et al. [13]; BRAF mutations were found in 9% of 23 cases of vulvar melanomas, none of seven cases of vaginal melanomas, and 20% of five cases of penile mucosal melanomas. KIT mutations were found in 35% of 23 cases of vulvar melanomas, none of seven cases of vaginal melanomas, and 20% of five cases of penile mucosal melanomas. NRAS mutations were found in 0%, 43%, and 20%, respectively. Penile melanomas were further evaluated by Oxley et al. [14] who did not find any BRAF or KIT mutations in 12 cases.

More recently, the expression of programmed death-ligand 1 (PD-L1) has been evaluated in melanomas. While it has been observed to be expressed in up to 76% of cutaneous melanomas, the same might not hold true for all GU melanomas [15,16]. Kaunitz et al. [17] evaluated PD-L1 expression in melanoma subtypes and found that it was expressed in 44% of 36 cases of mucosal melanomas. Hou et al. [10] specifically examined 51 vaginal and vulvar melanomas and found programmed death-1 expressed in 75% of the cases and PD-L1 expressed in 56% of the cases.

1.1. Epidemiology: distribution and determinants

Mucosal melanomas account for only approximately 1%–4% of new melanoma diagnoses, but GU melanomas represent 43%–45% of all mucosal melanomas [2,18,19]. Review of the Survival Epidemiology and End Results (SEER) database by Vyas et al. [1] revealed 817 cases of GU melanomas between 1992 and 2012. Eighty-nine percent were of the female genital tract, 6.6% were of the male genital tract, and 4.3% were of the urinary tract. Sanchez et al. [20] also reviewed the SEER database from 1973 to 2010 and found 1586 primary GU melanoma cases. In females, 75% were vulvar melanomas and 25% were vaginal melanomas. In males, 69% were penile melanomas and 31% were scrotal melanomas. Review of the North American Association of Central Cancer Registries by McLaughlin et al. [2] found 776 cases of genital tract mucosal melanomas from 1996 to 2000. The majority of the cases (723 cases) were in females; 77% were vulvar melanomas; 20% were vaginal melanomas; and 2.2% were cervical melanomas. The remaining 53 cases were in males; 66% were penile melanomas and 32% were scrotal melanomas.

GU melanomas are thought to represent 3%–7% of all cases of melanomas in females compared to less than 1% of cases in males [21,22]. McLaughlin et al. [2] found an overall age-adjusted incidence rate of 1.0 per million person-years, while it was only 0.2 per million person-years in males compared to 1.6 per million person-years in females. Similarly, Vyas et al. [1] found that both genital and urinary tract melanomas were over 10 times more common in females with an overall GU melanoma incidence rate of 1.74 per million person-years in females and 0.17 per million person-years in males. Likewise, Sanchez et al. [20] found an age-adjusted incidence of 1.80 per million person-years in females compared to 0.19 per million person-years in males. They also specifically examined urothelial melanomas and observed that the incidence was again higher in females compared to males, 0.06 versus 0.02 per million person-years, respectively.

Like cutaneous melanomas, non-Hispanic White people appear to be at highest risk for GU melanomas [1,2,20]. However, in available literature, the racial distribution of GU melanomas appears to be less skewed towards White people than in cutaneous melanomas [23,24]. A review of the SEER database from 1992 to 2005 found that the overall incidence ratio of mucosal melanomas in White to Black people was 2.30:1 and more specifically, in vulvar and vaginal melanomas, the incidence in White to Black people is 3.14:1 and 1.02:1, respectively [23]. A more expanded SEER database review from 1975 to 2016 found that in vulvar melanomas, 85.3% were White, 3.5% were African American, 7.3% were Hispanic, 2.9% were Asian or Pacific Islander, and 0.3% were American Indian or Alaska Native [24]. In vaginal melanomas, they found 71.7% were White, 9.5% were African American, 7.1% were Hispanic, 11.4% were Asian or Pacific Islander, and 0.2% were American Indian or Alaska Native.

Infection with human papillomavirus (HPV) may also contribute to the development of GU melanomas; however, the role is less clearly defined than in other GU malignancies. In a small series of six cases of vulvar melanomas and three cases of vaginal melanomas, HPV DNA was identified in 67% [25]. HPV-3, cutaneous, was identified in four of nine cases, and HPV-38, epidermodysplasia verruciformis-associated, was identified in four of nine cases; none of the cases had genital-mucosal HPV. In contrast, only one of 10 normal vulvar tissue controls tested positive for HPV DNA. This reflected similar trends seen in cutaneous melanomas in which HPV infection, especially cutaneous HPV, is associated with a higher rate of developing melanomas [26].

Age is also a risk factor with the highest incidence rate seen in those aged greater than 80 years regardless of sex [1,2]. The median ages of presentation are 66–68 years for female genital, 62 years for male genital, and 75 years for urothelial melanomas [20,21]. Unlike cutaneous melanomas, ultraviolet radiation exposure does not appear to be a risk factor for the development of mucosal melanomas [19,27]. Additionally, they appear to develop from preexisting nevi less frequently [28].

While the risk factors for GU melanomas are not entirely understood, the incidence of the disease has remained stable over several decades [20].

2. Site-specific presentation, staging, surgical treatment, and prognosis

2.1. Female genital melanomas

2.1.1. Vulvar melanomas

2.1.1.1. Presentation

Vulvar melanomas account for up to 10% of all vulvar malignancies [29]. Up to 12% of females have pigmented lesions of the vulva; however, only 4% of dermatologists check the vulva on annual examination [30,31]. Vulvar melanomas may present as papules, macules, or nodules with asymmetric borders, irregular coloration, and diameter greater than 7 mm (Fig. 1A) [32]. Up to 27% can present as amelanotic red colored papules [28]. Aside from a visible lesion, vulvar melanomas may present with a mass, pain, bleeding, pruritis, dysuria, discharge, irritation, or ulceration [33,34]. In order of frequency, it may arise on the labia majora, clitoral hood, and labia minora [34]. The majority develops on glabrous skin followed by the mucocutaneous border.

Figure 1.

Melanoma of the left labium minus with clinically and radiographically node-negative disease undergoing modified radical left vulvectomy with lymphoscintigraphy-guided sentinel lymph node biopsy. (A) Vulvar melanoma visible only after retraction of the labia majora; (B) Margins of resection; (C) Specimen for pathology; (D) Closure of modified radical left vulvectomy (images courtesy of Mitchel S. Hoffman, M.D. and Jonathan S. Zager, M.D.).

Vulvar melanomas may appear similar to cutaneous melanomas on dermoscopy, with structureless zones, atypical vascular supply, irregular globules, and with blue, gray, or white coloration or veil [35,36]. Similarly, reflectance-mode confocal microscopy may be used to identify vulvar melanomas with identifying features including loss of normal architecture of the chorion papillae and presence of atypical cells in the epithelium [37].

2.1.1.2. Evaluation and staging

Upon diagnosis of vulvar melanomas, clinical work-up should include computed tomography (CT) scan of the chest, abdomen, and pelvis (C/A/P) or whole-body positron emission tomography (PET)/CT scan with fluorodeoxyglucose (FDG) radiotracer, with or without magnetic resonance imaging (MRI) brain for all suspected Stage IIIA or higher as with non-GU cutaneous melanomas [[38], [39], [40]]. Fine needle aspiration (FNA) should be considered to assess any suspicious lymph nodes preoperatively [41]. Historically, vulvar melanomas were staged based on the International Federation of Gynecology and Obstetrics (FIGO) staging system for squamous cell carcinoma of the vulva, but the American Joint Committee on Cancer (AJCC) tumor-node-metastasis (TNM) staging system for cutaneous melanomas has been found to more accurately predict recurrence- and progression-free survival [33,42,43]. Nagarajan et al. [44] proposed a modified T-category which classifies vulvar melanomas into two categories based upon tumor thickness and mitotic rate to improve prognostication of overall survival (OS) and disease-specific survival (DSS) (Table 1A). Stage T1 is classified as tumor thickness less than or equal to 2.0 mm and a mitotic rate less than 2 per mm², while tumors greater than 2.0 mm and/or with a mitotic rate greater than or equal to 2 per mm² are classified as Stage T2.

Table 1.

Alternative staging systems applied to genitourinary melanoma.

Alternative staging system	Description
(A) Staging system proposed by Nagarajan et al. [44] (adjunct to TNM staging in vulvar melanoma)
Stage T1	≤2.0 with mitotic rate <2 per mm2
Stage T2	>2.0 and/or mitotic rate ≥2 per mm2
(B) 2018 FIGO staging for carcinoma of the cervix uteri [81] (applied to cervical melanoma)
Stage I	Carcinoma is strictly confined to the cervix uteri (extension to the corpus should be disregarded)
IA	Invasive carcinoma that can be diagnosed only by microscopy with maximum depth of invasion <5 mm
IA1	Measured stromal invasion <3 mm in depth
IA2	Measured stromal invasion ≥3 mm and <5 mm in depth
IB	Invasive carcinoma with deepest measured invasion ≥5 mm, lesion limited to the cervix uteri
IB1	Invasive carcinoma ≥5 mm in depth of stromal invasion and <2 cm in greatest dimension
IB2	Invasive carcinoma ≥2 cm and <4 cm in greatest dimension
IB3	Invasive carcinoma ≥4 cm in greatest dimension
Stage II	Carcinoma invades beyond the uterus, but has not extended to the lower 1/3 of the vagina or to the pelvic wall
IIA	Involvement limited to the upper 2/3 of the vagina without parametrial involvement
IIA1	Invasive carcinoma <4 cm in greatest dimension
IIA2	Invasive carcinoma ≥4 cm in greatest dimension
IIB	Parametrial involvement not up to the pelvic wall
Stage III	Carcinoma involves the lower 1/3 of the vagina and/or extends to the pelvic wall and/or causes hydronephrosis or non-functioning kidney and/or involves pelvic and/or involves paraaortic lymph nodes
IIIA	Carcinoma involves the lower 1/3 of the vagina with no extension to the pelvic wall
IIIB	Extension to the pelvic wall and/or hydronephrosis or non-functioning kidney (unless known to from another cause)
IIIC	Involvement of pelvic and paraaortic lymph nodes, irrespective of tumor size and extent
IIIC1	Pelvic lymph node metastasis only
IIIC2	Paraaortic lymph node metastasis
Stage IV	Carcinoma has extended beyond the true pelvis or has biopsy-proven involvement of the bladder or rectal mucosa.
IVA	Spread of the growth to adjacent organs
IVB	Spread to distant organs
(C) 2014 FIGO ovarian, fallopian tube, and peritoneal cancer staging system [90] (applied to ovarian melanoma)
Stage I	Tumor confined to ovaries or fallopian tubes
IA	Tumor limited to one ovary (capsule intact) or fallopian tube; no tumor on ovarian or fallopian tube surface; no malignant cells in ascites or peritoneal washings
IB	Tumor limited to both ovaries (capsules intact) or fallopian tubes; no tumor on ovarian or fallopian tube surface; no malignant cells in the ascites or peritoneal washings
IC	Tumor limited to one or both ovaries or fallopian tubes, with any of the following:
IC1	Surgical spill
IC2	Capsule ruptured before surgery or tumor on ovarian or fallopian tube surface
IC3	Malignant cells in the ascites or peritoneal washings
Stage II	Tumor involves one or both ovaries or fallopian tubes with pelvic extension (below pelvic brim) or primary peritoneal cancer
IIA	Extension and/or implants on uterus and/or fallopian tubes and/or ovaries
IIB	Extension to other pelvic intraperitoneal tissue
Stage III	Tumor involves one or both ovaries or fallopian tubes, or primary peritoneal cancer, with cytologically or histologically confirmed spread to the peritoneum outside the pelvis and/or metastasis to the retroperitoneal lymph nodes
IIIA
IIIA1	Positive retroperitoneal lymph nodes only (cytologically or histologically proven):
IIIA1 (i)	Metastasis up to 10 mm in greatest dimension
IIIA1 (ii)	Metastasis more than 10 mm in greatest dimension
IIIA2	Microscopic extrapelvic (above the pelvic brim) peritoneal involvement with or without positive retroperitoneal lymph nodes
IIIB	Macroscopic peritoneal metastasis beyond the pelvis up to 2 cm in greatest dimension, with or without metastasis to the retroperitoneal lymph nodes
IIIC	Macroscopic peritoneal metastasis beyond the pelvis more than 2 cm in greatest dimension, with or without metastasis to the retroperitoneal lymph nodes (including extension of tumor to capsule of liver and spleen without parenchymal involvement)
Stage IV	Distant metastasis excluding peritoneal metastases
IVA	Pleural effusion with positive cytology
IVB	Parenchymal metastases and metastases to extra-abdominal organs (including inguinal lymph nodes and lymph nodes outside of the abdominal cavity)
(D) Staging system as proposed by Bracken and Diokno [101] (applied to penile melanoma)
Stage I	Melanoma confined to the penis
Stage II	Melanoma with regional lymph node metastases
Stage III	Melanoma with distant metastases
(E) Staging system as proposed by Levine [121] (applied to urethral melanoma)
Stage A	Tumor confined to submucosa
Stage B	Tumor infiltrating periurethral muscle in females and corpus spongiosum in males
Stage C	Periurethral invasion including vagina, bladder, labia or clitoris in females and extending beyond the corpus spongiosum in males
Stage D	Metastasis to lymph nodes

FIGO, International Federation of Gynecology and Obstetrics; TNM, tumor-node-metastasis.

2.1.1.3. Surgical treatment

Vulvar melanomas are managed similarly to non-GU cutaneous melanomas with wide excision and sentinel lymph node biopsy (SLNB) as the standard of care, as shown in Fig. 1B–D [45]. More aggressive surgical management has been investigated including radical vulvectomy with bilateral inguinofemoral lymphadenectomy; however, it has not been demonstrated to have a survival benefit and has significant morbidity [24,[46], [47], [48]]. Standard surgical margins applied to cutaneous melanomas have been compared to more extensive surgical resection and no difference in survival has been observed [42,48]. Therefore, the recommended resection margins are at least 1 cm deep extending to the muscular fascia and 1 cm skin margins for melanomas ≤1.00 mm, 1–2 cm skin margins for melanomas 1.01–2.00 mm thick, and 2 cm for melanomas >2.00 mm thick [24,49].

Elective lymph node dissection does not appear to have a survival benefit in vulvar melanomas compared to lymph node dissection performed only with clinical evidence of nodal involvement [46]. SLNB has identified the sentinel node in 98% of documented cases and should be considered the standard of care, absent of any radiographic or clinical nodal-involvement [39]. Trifirò et al. [50] found that in 12 patients who underwent SLNB, the biopsy accurately predicted the status of remaining lymph nodes, which was confirmed on lymph node dissection in 10 of the patients. While there is still limited evidence, SLNB is recommended for evaluation of regional disease and omission of lymphadenectomy if negative [50,51]. Given the variable drainage of the pelvis, SLNB may be facilitated by single photon emission CT integrated with CT (SPECT-CT) to localize lymph nodes compared to planar lymphoscintigraphy [52].

2.1.1.4. Radiotherapy

Radiotherapy has roles in the neoadjuvant and adjuvant setting for operative candidates as well as a primary or palliative treatment in patients who are not surgical candidates [48]. In the neoadjuvant setting, the primary benefit is to facilitate more conservative surgery by reducing tumor size [45,48]. Adjuvant radiotherapy may be considered in patients with nodal involvement. It is important to note that the role of radiotherapy in vulvar melanomas is largely extrapolated from the application of radiotherapy in other mucosal melanomas [39].

2.1.1.5. Systemic therapy

Historically, chemotherapy may have had a role in the neoadjuvant setting; Janco et al. [53] reported that carboplatin and paclitaxel with bevacizumab shrunk a vulvar melanoma preoperatively preventing the need for skin grafting. A survival benefit has not been demonstrated with the use of adjuvant chemotherapy in vulvar melanomas [39]. In advanced disease, the role of biochemotherapy (cisplatin, vinblastine, dacarbazine, and/or tamoxifen plus interferon-alpha and/or interleukin-2) has been evaluated by Harting and Kim [54] and found to have a 10-month median survival with 36% partial response. Anti-PD-1/PD-L1 antibodies should be considered if appropriate, in a neoadjuvant setting, given the high rate of PD-1/PD-L1 mutations observed in vulvovaginal melanomas [10]. The role of immune checkpoint inhibitors specifically in vulvar and vaginal melanomas was examined by Quéreux et al. [55] and a 50% partial response was observed. Wohlmuth et al. [56] found that in vulvovaginal melanomas, ipilimumab had an overall response rate (ORR) of 13% while PD-1 inhibitors alone or in combination with cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) inhibitors had an ORR of 33%. Shoushtari et al. [57] found an ORR of 23% and median OS of 12 months to anti-PD-1 therapy in 35 patients with mucosal melanomas; 40% were vulvovaginal melanomas. While KIT mutations appear to be present in more than 20% of tumors, the role of targeted therapies has not been evaluated specifically in the treatment vulvar melanomas [39,58].

2.1.1.6. Prognosis

The 5-year survival rates for all vulvar melanomas ranged from 47% to 58% [18,20,59]. For localized, regional, and distant disease, they were 76%, 39%, and 22%, respectively [29]. Tumor thickness, specifically Breslow depth is the most significant predictor of recurrence in early-stage disease [42]. Independent factors that have been shown to significantly impact survival include age at diagnosis, lymph node status, mitotic rate, ulceration, and macroscopic amelanosis [24,28,39,58]. Lymph node status in particular appears to be the most important predictor of survival [24,58]. Both lymph node status and number of nodes involved are significantly associated with survival with rates of 65%–68%, 20%–29%, and 0%–20% for 0, 1, and 2 positive lymph nodes, respectively [29,49,60].

2.1.2. Vaginal melanomas

2.1.2.1. Presentation

Vaginal melanomas account for up to 5% of all vaginal malignancies [61]. It may arise anywhere within the vaginal canal; however, it most commonly presents in the lower third and on the anterior wall of the vagina [62,63]. Vaginal melanomas frequently present with vaginal bleeding, less frequently with vaginal discharge, mass, and pain [64,65]. Vaginal melanomas are often polypoid and ulcerated in appearance and while they are usually pigmented, up to 10% are amelanotic [63,65]. At the time of presentation, up to 50% of patients have positive regional lymph nodes and 20% have distant metastases [48,62]. The most common sites of metastases are to the lungs, liver, bones, and brain [65].

2.1.2.2. Evaluation and staging

Since vaginal melanomas are often advanced at presentation, CT C/A/P or whole-body FDG-PET/CT and brain MRI should be considered to detect metastatic disease [38,40,48,66]. Suspicious lymph nodes may be assessed with FNA [41]. The AJCC TNM staging system has been found to be superior to FIGO staging in vaginal melanomas [48,67]. AJCC Stage 0–II vaginal melanomas have been shown to have a significantly improved DSS compared to Stage III [67]. However, tumor size less than 3 cm versus greater than or equal to 3 cm appears to be the only significant predictor of survival in early-stage disease [[68], [69], [70]]. Therefore, Piura [48] has proposed a modification to the AJCC TNM staging system in which primary tumor size less than 3 cm versus greater than or equal to 3 cm is included in the T staging.

2.1.2.3. Surgical treatment

Surgical resection with pelvic exenteration or wide excision has been found to improve OS compared to nonsurgical management [71]. A significant difference in survival between radical resection with pelvic exenteration and vaginectomy versus conservative resection with wide excision has not been demonstrated; therefore, wide excision is considered an acceptable treatment with decreased morbidity [24,48,71,72]. Recommended wide excision margins are the same as vulvar melanomas with a depth of 1 cm tumor-free margin and width of 1.00 cm for melanomas ≤1.00 mm, 1.00–2.00 cm for melanomas 1.01–2.00 mm thick, and 2.00 cm for melanomas greater than 2.00 mm thick [24,41,48]. Vaginectomy and pelvic exenteration are acceptable as well if wide excision is not anatomically possible [48,71]. It has also been suggested for any vaginal melanomas greater than 3.00 mm in depth [73].

There are limited and mixed data regarding the role of lymphadenectomy and SLNB in clinically node negative vaginal melanomas; however, it appears that SLNB may be a reasonable approach to the evaluation of regional disease and should be considered standard of care [15,48,74]. The vaginal mucosa has variable drainage but, in general, the upper third drains to the external iliac nodes, the middle third drains to the common and external iliac nodes, and the lower third drains to the superficial inguinal and perirectal nodes [66]. As a result, SPECT-CT should be considered to aid sentinel lymph node localization [51,52].

2.1.2.4. Radiotherapy

Radiotherapy may be used in the neoadjuvant setting to reduce tumor size and facilitate more conservative surgery [48]. In the adjuvant setting—particularly in the presence of positive margins, tumors greater than 3 cm, or nodal disease—radiotherapy has been shown to reduce the risk of local recurrence [48,61,68,71]. Radiotherapy has also been shown to provide local control in patients with surgically unresectable disease [72]. The lymph node basins targeted by radiotherapy will vary based upon the location of the primary tumor within the vagina [48,66].

2.1.2.5. Systemic therapy

There are very limited data, but there may be a role for chemotherapy in the neoadjuvant and adjuvant setting, as well as in those with recurrent disease or those who are not operative candidates [45,53,71,75]. The use of adjuvant immunotherapy (interleukin-2, interferon-alpha, bacillus Calmette-Guerin, dendritic cells, lymphokine-activated killer cells, or measles vaccine) demonstrated a survival benefit compared to surgery alone, chemotherapy alone, or chemotherapy combined with immunotherapy [76]. Given the high rate of PD-1/PD-L1 mutations observed in vulvovaginal melanomas, treatment with these inhibitors should be considered [10]. As previously discussed, Wohlmuth et al. [56] found that in vulvovaginal melanomas, ipilimumab had an ORR of 13% while PD-1 inhibitors alone or in combination with CTLA-4 inhibitors had an ORR of 33%. Shoushtari et al. [57] found an ORR of 23% and median OS of 12 months to anti-PD-1 therapy in 35 patients with mucosal melanomas; 40% were vulvovaginal melanomas. The role of targeted therapies in vaginal melanomas is not well understood, but should be considered in those with targetable mutations [45].

2.1.2.6. Prognosis

The 5-year survival rates for vaginal melanomas are very poor and estimated at 5%–27% [18,20,71]. As previously mentioned, tumor size (< or ≥3 cm) is the greatest predictor of survival in early stage disease [[68], [69], [70]]. Nodal involvement is significantly associated with decreased survival [71]. Tumor thickness, ulceration status, and pathologic clearance of margins have been shown to predict DSS [67]. In addition, mitotic rate was found to predict DFS.

2.1.3. Uterine cervix melanomas

2.1.3.1. Presentation

Cervical melanomas may account for up to 9% of all malignancies of the uterine cervix [77].

Diagnostic criteria for primary cervical melanomas include presence of melanin in the cervical epithelium, absence of melanomas outside the cervix, junctional activity in the cervical epithelium adjacent to the lesion, and if present, metastatic spread consistent with the pattern of cervical malignancy [48,78,79]. It most commonly presents with abnormal bleeding including postcoital spotting and increased menstrual bleeding; however, 13% of diagnoses are made on routine gynecologic examination [78]. The lesion is often exophytic, polypoid, and varies in color with up to 45% of cervical melanomas presenting as amelanotic lesions [45,78].

2.1.3.2. Evaluation and staging

Once a patient is diagnosed with cervical melanoma using the above criteria, full body imaging is recommended to evaluate the extent of the primary tumor and if there is any metastatic spread [40,80,81]. The FIGO system for cervical carcinoma (Table 1B) is applied to cervical melanomas due similarities in presentation and pattern of spread [48,81]. In both FIGO and AJCC TNM systems, staging is affected by tumor depth, lymph node involvement, and distant metastases; however, FIGO is also based largely upon local anatomic involvement as well.

2.1.3.3. Surgical treatment

At this time, radical surgery, often radical hysterectomy with partial vaginectomy, is favored for the treatment of early-stage cervical melanomas [45,48,78,80]. Pelvic lymphadenectomy is advised for prognostication and planning of adjuvant treatment; however, the survival benefit of lymphadenectomy and the role of SLNB have not been evaluated [48,78].

2.1.3.4. Radiotherapy

Pelvic radiotherapy is recommended for palliation in late-stage cervical melanomas when surgical resection is not possible [45,48,78]. There may also be a role for adjuvant radiotherapy if the tumor is greater than 4 cm, and/or if there is regional nodal disease, positive surgical margins, and/or involvement of the corpus uteri or parametrium.

2.1.3.5. Systemic therapy

Single-agent dacarbazine is the most commonly used and only proven effective agent in recurrent or advanced cervical melanomas with up to 20% response rate [78]. The benefit of immunotherapy in cervical melanomas has not been demonstrated [82]. The role of targeted therapy specifically in cervical melanomas is not yet understood; however, it should be considered in patients with targetable mutations [83].

2.1.3.6. Prognosis

The 5-year survival for cervical melanomas is 10% with up to 88% passing within 3 years of diagnosis [45]. The most significant predictor of survival is FIGO stage [45,82,84]. However, even though most patients are diagnosed with early-stage disease, the prognosis remains poor [41,82]. The 5-year survival rates by stage are 19%, 11%, and 0% for Stages I, II, and III–IV respectively [82]. Tumor thickness, nodal status, lymphovascular invasion, and neovascularization have also been found to have prognostic value in cervical melanomas [45,84].

2.1.4. Ovarian melanomas

2.1.4.1. Presentation

Primary ovarian melanomas account for less than 1% of all female GU melanomas [1]. It occurs only in the presence of mature cystic teratomas as it is a malignant transformation of melanocytes in the ectoderm [48]. It presents at a younger age than other GU melanomas, with a median age of 47 years [85]. Since it is so obscure, the diagnosis of primary melanomas of the ovary is dependent upon criteria established by Cronje and Woodruff [86]. There must not be an extraovarian primary melanoma; there must be a unilateral ovarian tumor with teratoid element; and the patient's age and symptoms must correlate with other reported cases. The fourth criterion is that the tumor must have melanocytic junctional activity; however, this is only present in 50% of reported cases and not necessary for diagnosis [86,87]. Symptoms include lower abdominal pain, abdominal distension, palpable lower abdominal mass, dysuria, and dyschezia [87,88]. Diagnosis of ovarian melanomas is often made at the time of surgery as it is rarely suspected [87,89]. Ovarian melanomas spread in a nature similar to epithelial ovarian cancer with the addition of lymphatic and hematogenous routes, commonly involving the lymph nodes, lung, liver, and bone [88].

2.1.4.2. Evaluation and staging

The FIGO staging system for ovarian epithelial carcinoma (Table 1C) is commonly applied to ovarian melanomas [48,90]. Stage I is confined to the ovaries; Stage II extends below the pelvic brim; Stage III has cytologically or histologically confirmed spread to the peritoneum outside the pelvis and/or metastasis to the retroperitoneal lymph nodes; and Stage IV has distant metastases beyond peritoneal metastases [90]. However, some have suggested that the use of a depth-based staging system as is used in cutaneous melanomas may better estimate prognosis as it has been argued that FIGO staging does not provide valuable prognostication [87,89,91].

2.1.4.3. Surgical treatment

Total abdominal hysterectomy, bilateral salpingo-oophorectomy, and omentectomy is the procedure of choice in those who do not wish to preserve fertility [48,87,88]. In those with Stage IA to unilateral IC disease who wish to preserve fertility, unilateral salpingo-oophorectomy is a less radical option [48]. Pelvic and para-aortic lymphadenectomy is typically performed in the surgical management of ovarian melanomas, though there is no evidence supporting its benefit [48,88].

2.1.4.4. Radiotherapy

There is no evidence that radiotherapy improves prognosis in ovarian melanomas [87].

2.1.4.5. Systemic therapy

Cisplatin- and dacarbazine-based combination regimens have most commonly been used as adjuvant therapy for ovarian melanomas [48,88]. In a series of seven patients who received postoperative adjuvant chemotherapy (primarily cisplatin-based combination regimens), four had no evidence of disease at 12 months and three had passed within 9 months [85]. A single case of metastatic ovarian melanoma was successfully treated with adjuvant dacarbazine, vinblastine, and cisplatin after pneumonectomy with no evidence of disease at 18 months [88]. An important consideration for those that underwent ovary-sparing surgery is that chemotherapy should also be selected to preserve ovarian function; cisplatin-based regimens are often employed [48]. Immunotherapy with adjuvant interferon-alpha was attempted in a single case; however, the patient developed metastatic disease 12 months after surgery and was subsequently treated with dacarbazine, cisplatin, and vinblastine and passed away 17 months after initial diagnosis [89]. To our knowledge, the role of checkpoint inhibitors and targeted therapies has not been evaluated in ovarian melanomas, but there is at least one case of KIT mutations identified in an ovarian melanoma making it a potential therapeutic target [92].

2.1.4.6. Prognosis

In a series of 25 patients, 44% were alive without evidence of disease at median 18 months while the other 56% passed away at a median 8 months from diagnosis [85]. Of 25 patients in another series, 60% presented as Stage IA, 12% as Stage IC, 8% as Stage IIB, 16% as Stage III, and one had an unknown stage [93]. While most of these patients presented at early stages, only 21% were alive at 2 years. Similarly, Hyun and Mun [89] found that of 36 cases, 15 were deceased within 2 years and eight of those who passed away were Stage IA at diagnosis. This may be in part due to the aggressive nature of the disease, but also due to the poor prognostic value of the FIGO staging system for ovarian melanomas [87].

2.2. Male genital melanomas

2.2.1. Penile melanomas

2.2.1.1. Presentation

Penile melanomas account for less than 2% of all primary penile malignancies and less than 0.2% of all melanomas [94,95]. It most often presents as a nodular brown or blue-black lesion (Fig. 2A and B) that is frequently ulcerated [95]. It may be red and amelanotic in rare instances [96]. Patients are typically asymptomatic at presentation, but dysuria, hematuria, urethral discharge, obstructive symptoms, and even urinary fistula may be present in advanced disease [97]. At the time of presentation, over half of patients with penile melanomas had palpable inguinal lymphadenopathy in one series [20]. In order of frequency, penile melanomas may present at the glans penis, prepuce, penile shaft, or urethral meatus [98]. Penile melanomas are less likely to involve mucosa when compared to vulvar melanomas as there is theorized to be a lower concentration of melanocytes at the mucocutaneous border [99].

Figure 2.

Melanoma of the glans penis with biopsy-proven right inguinal nodal involvement undergoing partial penectomy with ventral phalloplasty and right inguinal complete lymph node dissection. (A and B) Penile melanoma presentation after retraction of the foreskin; (C and D) margins of resection; (E) specimen for pathology; (F) closure of partial penectomy with ventral phalloplasty and right inguinal lymphadenectomy (images courtesy of Philippe E. Spiess, M.D. and Jonathan S. Zager, M.D.).

2.2.1.2. Evaluation and staging

Since disease is often advanced at the time of presentation, completion of CT C/A/Por whole body FDG-PET/CT with or without brain MRI is recommended [20,38,40,100]. FNA should be considered to evaluate clinically positive lymphadenopathy [41]. A simple staging system developed by Bracken and Diokno [101] is often used in penile melanomas (Table 1D). Stage I disease is localized to the penis; Stage II involves regional lymph nodes; and Stage III is disseminated disease. AJCC TNM staging for cutaneous melanomas may also be applied to penile melanomas, though its prognostic ability is not well defined [22,38,40].

2.2.1.3. Surgical treatment

Aggressive surgical resection with total penectomy, perineal urethrostomy, and radical inguinal, iliac, and obturator lymphadenectomy was the preferred treatment in the past [102]. Recent recommendations favor more conservative management with wide excision or partial penectomy when possible to achieve standard margins applied to cutaneous melanomas, as seen in Fig. 2C–F [22,95,96,[103], [104], [105]]. There is a high recurrence rate regardless of surgical procedure; when comparing organ-sparing to amputative surgery in cutaneous penile melanomas, Bittar et al. [106] found recurrence rates of 19% and 13%, respectively. Cutaneous guidelines for lymphadenectomy and SLNB have also been extrapolated to penile melanomas [22,100]. Lymphoscintigraphy has been found to facilitate mapping of lymph node drainage in penile carcinomas and is recommended in SLNB for penile melanomas [96,107].

2.2.1.4. Radiotherapy

Adjuvant or palliative radiotherapy can be used but does not appear to have significant benefit in the treatment of penile melanomas [22,100,105].

2.2.1.5. Systemic therapy

Systemic therapy may be applied in the adjuvant or palliative setting [105]. At this time, there are not enough data to draw conclusions specifically for penile melanomas; however, there are single cases of success with adjuvant combined thymosin, bleomycin, vincristine, and cisplatin as well as with interferon [96,108]. Evaluation for targetable mutations is recommended since KIT mutations are present in up to 20% of penile melanomas and targeted therapies have shown promise in anal mucosal melanomas [13,100]. Despite lack of robust data, we feel that consideration should be given to use newer immunotherapy agents such as anti-PD-1 as a monotherapy or in combination with anti-CTLA-4 in a neoadjuvant setting when penile melanomas are advanced and/or associated with regional nodal disease.

2.2.1.6. Prognosis

The 5-year survival rates for all penile melanomas range from 20% to 69% [14,20,109]. When examining penile melanomas of only mucosal origin, van Geel et al. [109] found a 5-year survival of 31%; however, this was comparable to 5-year survival for cutaneous melanomas of similar depth. All patients with mucosal penile melanomas with regional nodal involvement or distant metastases passed away within 2 years of diagnosis even with surgical intervention. Poor prognostic factors include Breslow depth greater than or equal to 3.5 mm, tumor diameter greater than 15 mm, presence of ulceration, and microsatellitosis.

2.2.2. Scrotal melanomas

2.2.2.1. Presentation

Scrotal melanomas are exceedingly rare with less than 50 documented cases [20,110]. Like other cutaneous melanomas, they most commonly present as a pigmented macule or papule; however, there are rare cases of it presenting as an exophytic, ulcerated mass [22,111]. In one series, 19% of scrotal melanomas presented as AJCC TNM Stage I or II; 56% presented as Stage III; and 25% presented as Stage IV [110]. In another series, at presentation, tumor thickness was greater than 2 mm in 76% and 44% had regional or distant disease [106].

2.2.2.2. Evaluation and staging

Given that scrotal melanomas often present at late stages, imaging with CT C/A/P or whole-body FDG-PET/CT with or without MRI brain should be considered [38,106,110]. Palpable lymphadenopathy may benefit from sampling with FNA [41]. Scrotal melanomas are staged using the AJCC TNM staging system [22,111].

2.2.2.3. Surgical treatment

Bittar et al. [106] observed that local recurrence rate was 18% in 18 patients who underwent organ-sparing surgery while the two patients that underwent amputation did not recur. Sanchez-Ortiz et al. [22] found that all six patients in their series who underwent wide local excision had 0% local recurrence and 33% DSS at median follow-up of 36 months. It appears that the greater risk is for regional recurrence, which patients remain high-risk for even if they undergo scrotectomy [22,110]. Therefore, conservative management with wide local excision is recommended for local control [22]. Historically, elective lymphadenectomy was recommended in cases with Breslow depth greater than 1 mm, ulceration, or Clark Level IV or V involvement [22]. While the role of SLNB has not been described in scrotal melanomas to our knowledge, it is the standard of care in cutaneous melanomas and should be considered standard of care in scrotal melanomas, not elective lymph node dissection [38].

2.2.2.4. Radiotherapy

Radiotherapy in combination with systemic therapy has been applied in the metastatic setting; however, it is not a well described treatment of scrotal melanomas [112].

2.2.2.5. Systemic therapy

Adjuvant chemotherapy and immunotherapy in combination with radiotherapy as well as adjuvant chemotherapy alone in regional and distant metastatic disease have been described [22,112,113]. However, there does not appear to be a clear benefit. There is one case report of BRAF mutation targeted therapy in scrotal melanomas that is ongoing [114].

2.2.2.6. Prognosis

The 5-year survival for scrotal melanomas is up to 69% [20]. Stage of disease is the most significant predictor of outcome. Presence of distant metastases appears uniformly fatal [110].

2.3. Urinary tract melanomas

2.3.1. Urethral melanomas

2.3.1.1. Presentation

The urethra is the most common site for primary melanomas of the urinary tract; however, urethral melanomas comprise only 4% of all urethral malignancies [48]. Urethral melanomas are more likely to present in females (60%) and at an older age in females compared to males (67 years vs. 62 years) [115]. Time from onset of symptoms to diagnosis is nearly double in males (28 weeks vs. 15 weeks). Urethral cancer may present with a mass, dysuria, bleeding, hematuria, decreased urinary flow, incontinence, discharge, pain, weight loss, or may be asymptomatic [[115], [116], [117]]. In order of frequency, tumors present at the urethral meatus, distal urethra, fossa navicularis, and proximal urethra [115]. Urethral melanomas can be amelanotic in up to 20% of cases and have been mistaken for urethral caruncles [116,118]. The vast majority of patients present with late stage disease; El-Safadi et al. [115] found that 11% cases present as AJCC Stage I; 8% as Stage II; 37% as Stage III; and 44% as Stage IV.

2.3.1.2. Evaluation and staging

Evaluation with urethrocystoscopy and MRI pelvis can assist with surgical planning and evaluate the extent of soft tissue invasion [119]. CT C/A/P or whole-body FDG-PET/CT is recommended to evaluate for spread of disease [100,120]. The Levine staging system for urethral carcinoma (Table 1E) has been applied to urethral melanomas [121]. Tumor is confined to the submucosa in Stage A, infiltrates the periurethral muscle in females or corpus spongiosum in males in Stage B, invades beyond the periurethral muscle in females or beyond the corpus spongiosum in males in Stage C, and is metastatic to lymph nodes in Stage D. AJCC TNM staging for cutaneous melanomas has also been applied to urethral melanomas [22,38] and El-Safadi et al. [115] found that the T stage, a marker of depth of invasion, correlates significantly with survival.

2.3.1.3. Surgical treatment

Surgical treatment varies based upon the location of the urethral melanomas as well as between males and females [115,119]. In females, surgical resection may include total urethrectomy, partial urethrectomy, cystectomy, and vulvectomy [115]. In males, surgical treatment may include total penectomy, partial penectomy, prostatectomy, and cystectomy. Complete urethrectomy in early-stage disease may reduce the risk of local recurrence [63,117,120]. Margins of at least 2 cm are recommended [120]. A more radical surgical approach is recommended if there is evidence of local invasion [119]. Pelvic exenteration has been suggested for women with urethral melanomas greater than 3 mm in depth [73]. Lymph node dissection has not been shown to improve recurrence or DSS [117]. SLNB may be considered in patients without evidence of regional or distant disease [100,120].

2.3.1.4. Radiotherapy

Treatment of recurrence (using radiotherapy, chemotherapy, or immunotherapy) has been shown to improve survival, though the role of radiotherapy has not been examined specifically [115]. However, a benefit to using radiotherapy in particular has not been observed [100,120].

2.3.1.5. Systemic therapy

Adjuvant therapy with dacarbazine and interferon-beta (nimustine and vincristine held after one cycle) has achieved complete response in a patient with metastatic disease to the liver [118]. However, in another case, DAV-Feron (adjuvant dacarbazine, nimustine, vincristine, and interferon-beta) after radical surgery and lymphadenectomy in a patient with local disease at presentation did not appear to improve survival [122]. Adjuvant interferon-alpha may improve prognosis and should be combined with chemotherapy for synergistic effect [100,116,123]. Pembrolizumab has been used in the metastatic setting, but a survival benefit was not observed [119]. In a systematic review, the role of chemotherapy and immunotherapy was not examined individually, but treatment of recurrence with radiotherapy, chemotherapy, or immunotherapy was found to improve survival [115]. Targeted therapy, especially in those who have KIT mutations, should be considered [120].

2.3.1.6. Prognosis

The median survival of urethral melanomas is 26 months, without significant difference between males and females [115]. In a series of females, an OS of 27% was seen at 3 years [117]. Papeš and Altarac [120] reported a median survival of 16 months and 5-year survival of 10% in female urethral melanomas. Papeš et al. [100] also reported a 5-year survival of 10% in male urethral melanomas. Depth of invasion on T-stage, recurrent disease, pulmonary metastases, and nodal involvement have been associated with survival while tumor diameter has not been associated with prognosis [20,115]. In a smaller study, Oliva et al. [116] found that mucosal location and presence of nodular growth were more predictive of outcome than depth of invasion or tumor stage. The recurrence rate is approximately 60% in 1 year or up to 71% at a median 13 months [115,117]. Sites of recurrence include local recurrence (55%), inguinal lymph nodes (28%), distant metastases primarily to the lungs (38%), central nervous system (10%), and bones (6%) [115]. Treatment of recurrence with radiotherapy, chemotherapy, or immunotherapy has been shown to improve survival.

2.3.2. Bladder melanomas

2.3.2.1. Presentation

Bladder melanomas are very rare with roughly 30 cases reported and is slightly more common in males [124]. Presenting symptoms include hematuria, dysuria, pelvic pain, and pelvic mass; however, these are usually signs of locally advanced disease [125,126]. Since melanomas of the bladder much more commonly present as metastatic disease (up to 18% of metastatic melanomas metastasize to the bladder), a series of criteria have been established for diagnosis of primary bladder melanomas [48,125]. Patients must not have a history of prior cutaneous melanomas, must not have evidence of regressed cutaneous melanomas, must not have evidence of other visceral primary melanomas, must have a pattern of recurrence consistent with the primary visceral melanomas, and must have atypical melanocytes at the margins of the bladder lesion consistent with those seen in primary mucosal melanomas.

2.3.2.2. Evaluation and staging

Cystoscopy with transurethral biopsy is performed at diagnosis [48,126]. Additionally, intravenous pyelography and MRI pelvis are recommended to assess local involvement. CT C/A/P or whole-body FDG-PET/CT should be performed in addition to comprehensive dermatological, ophthalmological, and otorhinolaryngological exam to assess for regional or distant disease and rule out alternative primary melanomas. A staging system has not been applied to bladder melanomas in the literature; however, a distinction is made between tumors confined to the epithelium and those extending beyond the epithelium [126,127].

2.3.2.3. Surgical treatment

Transurethral resection of bladder tumor (TURBT) is recommended for bladder melanomas confined to the epithelium [126]. Bacillus Calmette-Guerin may be used as an adjunct to TURBT [124]. More aggressive surgery with partial cystectomy, or radical cystectomy is favored for localized disease extending beyond the epithelium [40,128]. The role of pelvic lymphadenectomy and SLNB in bladder cancer is not defined in the literature [48].

2.3.2.4. Radiotherapy

Radiotherapy has been used, but there are few data to support its value [40,48,129]. However, a single case of remission after combined radiotherapy and interferon-alpha in a patient who was not a surgical candidate has been reported [130].

2.3.2.5. Systemic therapy

The use of platinum-based chemotherapy has been proposed, but its effect is not documented [124]. Immunotherapy with interferon-alpha in combination with radiotherapy induced remission in single patient who was not an operative candidate [130]. However, a case of adjuvant interferon with dacarbazine was not as promising [131]. Adjuvant nivolumab following cystectomy in a case of metastatic bladder melanoma yielded clinical remission at 15 months, suggesting that checkpoint inhibitors may be beneficial especially in late-stage disease [124]. Tumors with BRAF, NRAS, or KIT mutations should be considered for targeted therapies [124].

2.3.2.6. Prognosis

In a systematic review, Barillaro et al. [124] found median survival of 21 months. In patients who underwent radical cystectomy, survival was 60% at median follow-up of 16 months while in patients who underwent conservative management, including TURBT, survival was 29% at median follow-up of 14 months. Size, depth of invasion, and presence of metastases are associated with a worse prognosis [127]. Survival beyond 3 years has not been observed in locally advanced disease [40,128].

2.3.3. Upper urinary tract melanomas

To our knowledge, only three cases of primary melanomas of the ureter have been reported [132]. The first two cases presented in the 1960s [[132], [133], [134]]. Recently, a single case of melanoma of the ureter presented in a patient who underwent nephrectomy for papillary renal cell carcinoma 7 years prior to presentation and was found on follow-up imaging to have a new, contralateral mass of the proximal ureter consistent with primary melanoma on biopsy [132]. The patient had metastatic disease to the lungs and scapular bone and received immunotherapy with progression of disease and ultimately passed away 16 months from diagnosis.

Primary renal melanomas are more commonly of metastatic origin, similar to bladder melanomas [135]. Thus, it has been proposed that criteria established by Ainsworth et al. [125] for primary bladder melanomas be applied to the diagnosis of primary renal melanomas [135]. To meet these criteria, the patient must not have history of prior cutaneous melanomas or evidence of regressed cutaneous melanomas, must not have evidence of other visceral melanomas, must have a pattern of recurrence consistent with the primary lesions, and atypical melanocytes or in situ melanomas should be present in the urothelium in the vicinity of the melanomas. There are only five reported cases of primary renal melanomas in the literature that meet these criteria [135]. Presenting symptoms included flank and lumbar pain, hematuria, and polyuria [135]. All patients underwent nephrectomy—three radical and two simple—as well as ureterectomy in one case, paracaval and interaortocaval lymph node dissection in one case, and liver biopsy in another case. All patients received adjuvant treatment, three received interferon-alpha—one with the addition of fotemustine, one received Bacillus Calmette-Guerin with a melanoma cell vaccine, and one received chemotherapy. Of the four with known follow-up, two were alive with no evidence of disease at 22 months and 27 months and two developed distant metastatic disease within 1 year of diagnosis.

Due to the limited number of cases, there are no guidelines on the management of upper urinary tract melanomas; however, Gakis et al. [136] have extrapolated data from the treatment of other advanced melanomas to establish recommendations. When resectable, treatment should include primary nephroureterectomy with regional lymphadenectomy. Adjuvant dacarbazine-based chemotherapy regimen is recommended when the primary tumor depth is greater than 1.5 mm, there are positive surgical margins, or there are positive lymph nodes. Unresectable disease should be treated with local radiation and dacarbazine-based chemotherapy. Palliative stenting of the upper urinary tract may be performed, but they advise that tumor manipulation with ureteroscopy or double-J stent placement may lead to distribution of melanoma cells throughout the ureter. Palliation with endoscopic resection of ureteral masses can also be considered [137]. Of note, these recommendations are for the management of melanomas metastatic to the upper urinary tract but may be applied to primary ureteral and renal melanomas in lieu of more relevant guidelines.

3. Conclusion

The presentation, diagnosis, staging, treatment, and prognosis are variable with regards to GU melanomas. This variability in addition to the rare nature of the diseases makes them difficult to study and establish definitive staging and treatment guidelines. There have been significant advances in the treatment of melanomas in the last 10 years, particularly in systemic therapy for late-stage disease and even consideration of neoadjuvant strategies [138]. While many of the trials that led to the approval of these drugs included cutaneous and mucosal GU melanomas, the generalizability of these data to GU melanomas is not well defined [139]. Unfortunately, even studies attempting to specifically examine these novel therapies in mucosal melanomas do not discuss the primary location or have very few GU-specific melanomas, the majority of which are vulvar or vaginal primaries [57,[140], [141], [142], [143]]. Additionally, these publications lag years behind updates to standard of care. To our knowledge, there is only one active clinical trial dedicated to the treatment of GU melanomas: a single-center, open-label study evaluating the safety and efficacy of anti-PD-1 antibody, camrelizumab, in the treatment of malignant melanomas of the female genital tract (NCT04593485). This paucity of treatment information in addition to the tendency for GU melanomas to present in advanced stages likely contributes to the poor prognosis associated with the disease [144]. Outcomes may be improved by increasing clinician awareness of the current literature and may benefit from future trials focusing specifically on the management of GU melanomas, particularly in advanced stages.

Author contributions

Study concept and design: Philippe E. Spiess, Jonathan S. Zager.

Data acquisition: Danielle K. DePalo, Kelly M. Elleson, Michael J. Carr, Philippe E. Spiess, Jonathan S. Zager.

Data analysis: Danielle K. DePalo, Kelly M. Elleson, Michael J. Carr, Philippe E. Spiess, Jonathan S. Zager.

Drafting of manuscript: Danielle K. DePalo.

Critical revision of manuscript: Kelly M. Elleson, Michael J. Carr, Philippe E. Spiess, Jonathan S. Zager.

Conflicts of interest

Jonathan S. Zager has advisory board relationships with Merck, Novartis, Philogen and Castle Biosciences, speaker's bureau for Castle Biosciences, Pfizer, and Sun Pharma. He also receives research funding from Amgen, Delcath Systems, Philogen, Provectus, and Novartis. He serves on the medical advisory board for Delcath Systems. Philippe E. Speiss. serves as the vice chair for the National Comprehensive Cancer Network Panel for bladder and penile cancer. He also serves as an author for UpToDate.com. The remaining authors have no relevant conflicts of interest to disclose.