Primary anorectal melanoma: an update.

The anorectum is a rare anatomic location for primary melanoma. Mucosal melanoma is a distinct biological and clinical entity from the more common cutaneous melanoma. It portrays worse prognosis than cutaneous melanoma, with distant metastases being the overwhelming cause of morbidity and mortality. Surgery is the treatment of choice, but significant controversy exists over the extent of surgical resection. We present an update on the state of the art of anorectal mucosal melanoma. To illustrate the multimodality approach to anorectal melanoma, we present a typical patient.

Introduction

Malignant melanoma accounts for under 5% of all cases of cutaneous malignancies in the United States, with an annual incidence of approximately 70,000 1. However, melanoma is the most fatal skin malignancy, with an annual mortality approaching 10,000. The vast majority of cases of melanoma arise in the skin, while primary melanoma arising from mucosal surfaces occurs in under 2% of cases 2. Primary mucosal melanoma (MM) is slightly more predominant in females, largely due vulvovaginal melanoma 3. The distribution of MM involves the head and neck in 55.4% of cases, the anorectum in 23.8%, the female genital tract in 18.0%, and the urinary tract in 2.9% 2. The diagnosis of anorectal mucosal melanoma (ARMM) portends a particular poor prognosis 4, and a standardized evidence based treatment approach is not well defined due to the rarity of this disease 5. Herein, we use a typical patient along with relevant literature to illustrate the multimodality approach to ARMM, and discuss important factors associated with the treatment and prognosis of ARMM.

Typical presentation of anorectal mucosal melanoma

To illustrate a typical presentation of ARMM and its accompanying multimodality approach, we present a 76-year old female presented with a 2-month history of intermittent blood per rectum and tenesmus. On digital rectal examination, a palpable mass at 4 o'clock was noted approximately 2 cm proximal to the anal verge. Rigid proctoscopy demonstrated a 3 cm ulcerated lesion along the dentate line. Multiple biopsies were taken, and pathologic results were consistent with malignant melanoma. Endoscopic ultrasound demonstrated a 2.5 X 3.5 X 3cm mass with no pathologic lymph nodes visualized. A staging positron emission tomography (PET) scan with computed tomography (CT) was performed, with no evidence of distant metastases.

The patient underwent wide local excision of the primary lesion, preceded by a sentinel lymph node (SLN) mapping with a submucosal injection of 99mTC-Nanocoll. A single pre-sacral SLN was found and biopsied.

The final pathology from the resection specimen demonstrated spindle-shaped, melanotic cells invading into the muscularis propria. The cells demonstrated a high mitotic index, with evidence of vascular invasion. Immunohistochemical (IHC) staining showed strongly positive staining with HMB-45, MART-1, and S100. Additionally, the SLN was found to harbor micrometastases.

Subsequently, the patient received adjuvant chemo-radiation therapy with three cycles of radiation to the surgical site and three cycles of temozolomide. During routine follow-up imaging six months post-procedure, the patient was found to have metastatic deposits in the mesorectal lymph nodes, along with the liver and lung. Fourteen months later, the patient developed rectal bleeding and obstructive symptoms, and was found to have local recurrence in the rectum. A repeat wide local excision was performed, along with a diverting colostomy.

Discussion

Malignant melanoma is the fifth leading cause of new cancer diagnosis in males in the United States, and the sixth in females 6. This incidence is rising by approximately 3% per year 7, especially among young adults 8. These increases may be attributed to increased screening and codification of pathologic standards for diagnosis 9. MM is a rare type of malignant melanoma, and widely considered to be a distinct clinical entity from CM based on its poor prognosis 10. Though mucosal melanoma arising from various head and neck mucosal surfaces comprises over 50% of MM, the anorectum is a common anatomical site of MM, and is the third most common location for malignant melanoma after cutaneous melanoma (CM) and ocular melanoma 2, 3, 11. ARMM is also a distinctly rare tumor of the anal canal, comprising less than 4% of malignancies in that region 12.

Considerable dispute exists regarding the cell of origin for ARMM. While the presence of melanocytes has been relatively well described in the mucosa of the head and neck 13, 14 and the esophagus 15, 16, its presence in the intestinal mucosa from the stomach to distal rectum is controversial 17. Historically, melanocytes have been found within the transition zone beneath the dentate line and increases in number distally toward the anoderm 18, 19. This has led to the presumption that anorectal melanoma arises from normal melanocytes distal to the dentate line that extend proximally into the rectum 20. Staining for melanoma markers using various techniques for HMB-45 and S100 have demonstrated that melanocytes present rarely in the mucosal epithelium above the dentate line in normal patients 21. Additionally, a proliferation of normal melanocytes within the colorectal epithelium proximal to the dentate line has been seen in patients with melanoma arising within the proximal anal canal 21, 22. This has led to the conclusion that ARMM can arise directly from melanocytes located in the intestinal epithelium of the proximal anus or distant rectum hence, primary anorectal melanoma 23.

Embryologically, cells migrating from the neural crest that enter the dorsolateral pathway differentiate into melanocytes that eventually populate their eventual sites of colonization, while cells entering the ventral pathway are the neurogenic precursors of the peripheral and enteric nervous system 24. Neural crest cells are typified by a characteristic set of transcription factors including Snail2 (Slug), Sox10, FoxD3, and Sox9 25. Melanocyte migration and differentiation involves a complex interplay of cell signaling pathways 26. Specific mutations in the c-Kit/stem cell factor (SCF) pathway, the endothelin receptor type B/endothelin pathway, and the Sox10 transcription factor pathway being associated with a variety of related pigment and enteric nervous system disorders including piebaldism, Waardenburg syndrome, and Hirschprung's disease 27-32.

Exposure to ultraviolet (UV) rays, especially UV-B rays at a wavelength of 290 to 320 nm, is clearest risk factor for the development of cutaneous malignancies, including melanoma 33, 34. While cumulative exposure increases the risk of the more common basal cell (BCC) and squamous cell carcinomas (SCC) of the skin, CM is associated with intense, intermittent exposure to UV rays 35. Two well described nucleic acid lesions caused by UV-B include 6-4 photoproducts (6-4PP) and cyclobutane pyramidine dimers (CPD) 36. Accumulation of these signature mutations are seen in BCC and SCC, including p16/INK4A and p53 37. Mutations in BRAF is the most common driver mutation in malignant melanoma, found in upwards of 60% of all CMs and 80% of melanocytic nevi, the vast majority being the V600E amino acid substitution at exon 15 38-40. While the signature UV-associated 6-4PP and CPD nucleic acid mutations are not found to be the etiology of BRAF mutations, there is evidence that faulty repair of UV-A associated oxidative damage and other UV-B associated mutations may be the etiology of BRAF mutations in CM 41.

The role of photocarcinogenesis in CM is highlighted by the different patterns of mutations associated with MM. Edwards et al evaluated primary MM tumors from varying anatomical locations and found no evidence of BRAF mutation, in stark contrast to the prevalence of this mutation in CM 42. Along with a paucity of BRAF mutations 43, similar studies have shown decreased NRAS mutations in MM compared to CM 44, 45, while KIT driver mutations are increased in MM 46, 47. Other mechanisms of pathogenesis for MM have been investigated, including viral infections 48 and the use of tobacco 49, but no clear etiology of MM has been elucidated.

Due to its relative rarity, the treatment of ARMM is controversial. While it is clear that surgical resection is favored, the extent of surgery has been called into question as upwards of 25% of patients with ARMM present with inoperable tumors, either because of distant metastases or aggressive locoregional disease 2, 3. 60% of patients present with local lymphatic spread 2. This aggressive local disease led to the suggestion that an abdominoperineal resection (APR) be the treatment of choice in order to address local lymph nodes 50-53. Brady et al reported on 71 patients with ARMM treated with either abdominoperineal resection (APR), wide local excision (WLE), or biopsy and/or fulgration only 54. While no significant difference in survival was found regardless of operative approach, the local recurrence rate was 8% for APR versus 20% for local therapies.

Given the morbidity of APR and the clear quality of life advantages of WLE 55, there has been persistent concerns regarding an aggressive surgical approach with no clearly demonstrated survival advantage 5, 56. Iddings et al reported on patterns of treatment and outcomes in patients with ARMM from the SEER database and found no significant different in five-year survival between patients undergoing APR versus WLE (17% and 19%, respectively) 57. Nilsson et al found that margins of resection significantly predicted long-term advantage, with 5-year survivals of 19% for patients receiving an R0 resection, compared to 6% for patients with R+ resections, regardless of the type of surgery 58. Of note, patients undergoing APR were significantly more likely to receive an R0 resection (76%) compared to those undergoing WLE (26%). Our patient did undergo WLE, but notably had an R1 resection, with microscopic involvement of the margins of resection. The role of reoperation for margins after an initial WLE has not been addressed in the literature.

As was the case in our patient, distant metastases areoverwhelmingly the cause of mortality in patients with ARMM, leading to the suggestion that WLE be the initial procedure of choice 59. However, as demonstrated in our case, patients with local recurrence often require repeat excisions, occasionally including salvage APR 60. With reports of local recurrence rates as high as 65% with WLE alone 61, investigations into improved local control with the addition of radiation have been undertaken. Kelly et al reported on 54 patients undergoing WLE followed by adjuvant radiation, and showed a local recurrence rate of 18% at 5 years 62.

A lingering question in the treatment of ARMM involves the approach to nodal disease. The use of sentinel lymph node (SLN) mapping has been reported, with subclinical nodal disease found in the inguinal and pelvic nodal basin SLNs 62-66. No series large enough has been reported to draw conclusions on the effect of SLN biopsy on recurrence or survival, leading some to conclude that the procedure should not be performed 67. The presence of regional lymph node metastases has not been shown to affect recurrence patterns 4, lending further support to the avoidance of lymphadenectomy. In our patient, with a positive mesorectal SLN, we opted to avoid pelvic exploration as there is no evidence supporting an improved outcome with extended lymphadenectomy.

The preponderance of data suggests that, while extended surgical procedures for ARMM afford no survival benefit, there appears to be significant benefit to achieving an R0 resection. We recommend, based on a review of the literature, the use of intraoperative frozen sections to insure that WLE is performed to negative margins. Reoperative excision of margins may be necessary to render patients disease free and thereby increase the likelihood of cure.

Conclusion

ARMM is a rare malignancy that seems to lack the risk factors associated with cutaneous malignancies. The diagnosis of ARMM portends a poor prognosis, with distant metastases being a common and fatal development. There is a clear advantage to complete R0 surgical excision. While neither surgery nor radiation has had a clear impact on survival, adjuvant radiotherapy does appear to provide greater loco-regional disease control. Due to the rarity of this disease process, no prospective, randomized trials can definitively elucidate the ideal multimodality therapy for ARMM. Despite this, our retrospective review of the literature leads us to recommend a complete R0 resection for patients with ARMM.

Contributing Author Declaration

We certify that all individuals who qualify as authors have been listed; each has participated in one or more of the following areas: conception and design of this work, the acquisition and/or analysis of data, the writing, and/or critical revision of the document, and supervision of this cooperative research effort. All contributing authors approve of the submission of this version of the manuscript and assert that the document represents valid work. If information derived from another source was used in this manuscript, we obtained all necessary approvals to use it and made appropriate acknowledgements in the document. All contributing authors take public responsibility for this work.

Funding

All authors declare no source of funding.

Disclaimer

The views expressed in this manuscript are those of the authors and do not reflect the official policy of Department of the Army, the Department of Defense or the United States Government.

Copyright protection

Some of the contributing authors are military service members or employees of the U.S. Government, and this work was prepared as part of their official duties. Title 17 U.S.C. 105 provides the “Copyright protection under this title is not available for any work of the United States Government.” Title 17 U.S.C. 101 defines a U.S. Government work as a work prepared by a military service member or employee of the U.S. Government as part of that person's official duties.