TITLE: Metastatic Melanoma of Unknown
Primary: A Unique Entity?
SOURCE: Grand Rounds Presentation, Department of Otolaryngology
The University of Texas
Medical Branch (UTMB Health) <=
/span>
DATE: November 28, 2012
RESIDENT PHYSICIAN: Benjamin Walton, MD
FACULTY PHYSICIAN: Susan McCammon, MD
SERIES EDITOR: Francis B. Quinn, Jr., MD
ARCHIVIST: Melinda Stoner Quinn, MSICS
"This
material was prepared by resident physicians in partial fulfillment of
educational requirements established for the Postgraduate Training Program =
of
the UTMB Department of Otolaryngology/Head and Neck Surgery and was not
intended for clinical use in its present form. It was prepared for the purp=
ose
of stimulating group discussion in a conference setting. No warranties, eit=
her
express or implied, are made with respect to its accuracy, completeness, or
timeliness. The material does not necessarily reflect the current or past
opinions of members of the UTMB faculty and should not be used for purposes=
of
diagnosis or treatment without consulting appropriate literature sources and
informed professional opinion."
Introduction
Malignant melanoma is a cancer arising from pigme= nt cells in the skin called melanocytes. Melanoma is a primary skin cancer whose incidence is rising at a fas= ter rate than any other malignancies, except for lung cancer. A recent study fr= om Europe has shown that while the mortality for melanoma stayed relatively stable, the incidence has increased exponentially. The incidence is highest= in the countries Australia and New Zealand, followed by North America and Nort= hern Europe. According to the US National Cancer Institute Surveillance Epidemio= logy and Results database, there is an estimation that 68,130 individuals will be diagnosed with melanoma worldwide in 2010 and that in the US alone, 8700 pe= ople will die from melanoma. There are many risk factors associated with the development of cutaneous melanoma. These are generally broken up into environmental and genetic risk factors. Environmental risk factors generally include: an inability to tan, fair complex,= blue/green eyes, blonde/red hair, freckling, history of peeling sun burns, immunosuppression, teenage outdoor summer jobs, and a history of tanning bo= oth exposure. Genetic and past medical causes include CDKN2A (p16) mutation, fa= mily history of melanoma, history of a prior melanoma, actinic keratoses, non-me= lanoma skin cancer, xeroderma pigmentosa, atypical nevus, and giant congenital melanocytic nevus. Melanocytes derive from neural crest cells. During embryo genesis, the precursors for melanocytes migrate from t= he neural crest to the skin, uvea, leptomeninges, and mucous membranes. These = are all sites where melanoma can occur. Generally, they remain undifferentiated until = they are stimulated where they begin to produce melanin and become melanocytes. Rece= nt studies have looked into the pathways involved in melanoma. Probably, the m= ost important and most researched intracellular signaling pathway is the RAS/RA= F/ MEK/MAPK pathway. In this pat= hway, the BRAF protein has been found to be mutated in up to 66% of all melanoma.= Also the N-RAS protein is found to be mutated in = 15% of melanoma.
There are several histological classifications in= melanoma. Superficial spreading melanoma is the most common form of melanoma and aris= es in a pre-existing nevus. Nodular melanoma is generally regarded as a second most common form of melanoma. Lentigo maligna is an in situ melanoma and is generally a precursor for invasive lentigo malignant melanoma. Desmoplastic melanoma is a somewhat atypical form of melanoma. 73% are amelanotic, and this often= leads to delay in diagnosis. Often this can lead to cranial nerve and skull base involvement. Mucosal melanoma is somewhat different in both its nature and incidence. It is most commonly in the head and neck which presents up to 40= -50% of all incidence. However, it is less than 2% of all melanomas. The peak incidence for mucosal melanoma is generally during the 6th and 7th decade. Within the head and neck, the nasal cavity is the most commonly involved si= te. Generally, the anterior septum is the most likely location followed by the middle and inferior turbinate. The second most common site is the oral cavity, wit= h the hard palate and the maxillary alveolar gingival being the most common subsites. Mucosal melanomas generally do not present with regional spread as only 18.= 7% will have regional spread at the time of diagnosis. Also ocular melanoma can occur and is usually subdivided into choroidal or conjunctival melanoma.
Staging
Due to the unique nature of melanoma, staging sys= tem differs to other head and neck cancers and cutaneous malignancies. Two important classification= s one should understand are the Clark and Breslow microstaging systems. The Bresl= ow microstaging system is based on the microscopic depth of invasion of the melanoma in millimeters. The = Clark level is based on a one through five level depth of invasion and based on t= he separate levels of the skin. A Clark level I is confined to the epidermis w= hile Clark level II is a melanoma in both the epidermis and through the basal lamina. Clark level III is a melanoma infiltrating through the papillary de= rmis while Clark level IV is a melanoma invading the reticular dermis. Clark lev= el V is when the melanoma has infiltrated through the subcutaneous fat. The AJCC revised their staging in 2009 after multi-variate analysis of 30,946 patien= ts with stage I, 2, and 3 melanoma and 7,9722 patients with stage IV melanoma.= The T staging is based on the thickness of the melanoma. Due to the multi-variate analysis s= howing prognostic differences, ulcer= ation and number of mitoses were added. A T1 tumor is less than 1.0 mm thick. A T2 tumor is 1.01-2.00 mm thick, and a= T3 tumor is 2.01-4.00 mm thick. A T4 tumor is greater than 4.00 mm thick. Sub-dividing these further, ulceration will upstage each of these classifications as this is been found to be a poorer prognostic factor. N staging is based on the number of metastatic nodes. N1 classification is one metastatic node, and N2 is 2-3 number of metastatic nodes. N3 disease is 4+ metas= tatic nodes or nodes that are matted or in transit metastases/solid metastases wi= th metastatic nodes. Microscopic examination is now being used as part of the staging system. M stagi= ng is classified based on number of metastases.&= nbsp; These are M0, M1a, M1b, or M1c.&nbs= p; LDH is used as well in= the staging system. An elevated LDH will increase the staging to M1c. The differences most notably from = the 6 addition to the seventh addition (2009) of the melanoma staging system incl= ude the addition of mitotic rate = in T1 melanomas. Also, the immunochemical detection of nodal metastases has not b= een included for nodal classification. Prior to the latest addition, there was a lower threshold for staging N-positive disease. Classification changes now<= span style=3D'mso-spacerun:yes'> include all nodal metastases with isolated tumor cells or tumor deposits less than 0.1 mm. This will meet criteria for histologic or immunohistochemical detection of melanoma as N- positive.
Treatment Options
Melanoma is primarily a surgical disease as radia= tion and chemotherapy have not been found to be sufficient in treatment. Standard approaches include wide local excision, wide local excision with sentinel l= ymph node biopsy, wide local excision with elective lymphadenectomy, radiation therapy, and/or systemic therapy. Sentinel lymph node biopsy has become more widely accepted as a standard care for patients with melanoma. Multiple studies have shown the ef= ficacy of sentinel lymph node biopsy if identifying occult metastases. This aids in the staging and progn= osis of patients.
Lymphatic Drainage
In understanding both malignant melanoma and mela= noma of unknown primary, one must understand the lymphatic drainage patterns of = the head and neck region and their unpredictability. A study out at M.D. Anderson by Ow = et el specifically looked at melanoma and its drainage pathways. They found highly variable drainage pathways = but found that the majority of the face, forehead, and cheek drained into the parotid lymph nodes. The paro= tid gland severs as a major drainage pathway from the face and scalp before draining into the upper jugular nodes.&nbs= p; Ow et al also looked at different nodal fields and a number of instances where melanoma drained into multiple lymph node fields. They found that 42% of head and neck melanomas drained into greater than one nodal field. This differed from non-head and = neck melanomas which generally drained into one nodal field. Another study out o= f Croatia sought to examine areas where there was discordance between the primary site and the predicted nodal fields involved. They found that the posterior scalp and upper neck were the most discordant with their projections on drainage.= Of the sites examined, the co= ronal scalp had the highest predictable drainage patterns. Understanding the lymphatic draina= ge pathways can help focus the search for a primary site in occult disease.
Melanoma of Unknown Primary
Melanoma of unknown primary (MUP) is unique in ma= ny accounts which will be discussed in further detail. It was first described = in 1952 by Pack et al. Dasgupta = first described diagnostic criteria for this unique entity in 1963. MUP generally accounts for 1-8% of all melanomas. It can be divided into 2 clinical g= roups. These are: metastatic involvement to lymph nod= es or non-lymph node disease. There are many theories on the origin of this unique entity. Some theorize it is an unrecognized completely regressed primary melanoma that metastasized prior to regression. Lee et al have speculated t= hat this is the most probable explanation. Other theories include a previous ex= cised undiagnosed melanoma, concurrent and undetected melanoma, a melanoma arising from benign nevus cells found in lymph nodes, or a de-novo malignant transformation. Anbari et al = set analyzed their clinical experience with melanoma of unknown primary site. T= hey analyzed 40 patients and found that 22.5% of their patients had a clinically dysplastic nevus prior to developing a MUP. 20% of their patients had a his= tory of regressed skin lesion that was never diagnosed.
Regression in melanoma is well documented with a postulated frequency of 3.7-8.7%. This is thought to be more common in men. Theories on this gender inequality include sun exposure and ignorance for pigmented lesions by men.= Some authors believe men will ignore a lesion for a long enough period to allow for total regression. Melanoma is one of the most common tumors to undergo regression and only 2 other tumors share this characteris= tic. These are neuroblastoma and hypernephroma.= Studies differ on whether regression is found to be a positive or negative prognostic factor. I= n fact some studies have shown that in primary cutaneous melanoma, regression can = be a worse prognosis. There is plenty of experimental data on immunologic studie= s which support endogenous anti-melanoma immune response. Mauer et al identified circulating factors potentiating lymphocytic cytotoxicity in regressed melanomas. There are also increased lymphocytic infiltrates found in many regressed melanomas. Some studies have shown there is a favorable prognosis associated with the presence of tumor infiltrating lymphocytes and melanoma= .
Since 1963, there have been changes to the origin= al diagnostic criteria is set by Dasgupta.&nb= sp; Current diagnostic criteria include a metastatic melanoma confirmed clinically, histologically, or immunohistochemically, an absence of a previous cutaneous = tumor, pigmentedor not, which has been destroyed or excised without histology. The= re is also an exclusion of unusual primary sites, including areas in the urogenital, otolaryngologic, or ophthalmologic sites. Diagnositc recommendations for evaluation for patient presenting with melanoma of unkn= own a primary are based on these criteria. Patients should have a review of the previous skin biopsies, a full skin evaluation, CT or MRI imaging of the br= ain, CT imaging of the chest, abdomen and pelvis, a full ENT examination, and ophthalmologic examination, and a Wood's UV lamp examination. The Woods UV = lamp reveals areas of depigmentation or regressed melanomas. This appears to be a fairly strenuous evaluation and has been questioned as being redundant. Tos= et al reviewed 103 patients from 1986-2006. 61% of their patients presented wi= th lymph node metastases. Each of their patients underwent a full examination includ= ing otoscopy, rhinoscopy, laryngoscopy, ophthalmologic exam, sigmoidoscopy or proctoscopy, and gynecological examination when relevant. Of the 103 patien= ts, only one was found with a possible primary on examination. This patient was= thought to have an ocular melanoma; however, this was never proven. The study recommended against multiple special examinations in the workup for metasta= tic melanoma with unknown primary, but suggested a focused search based on hist= ory and physical. PET imaging has also been recently added and the workup for metastatic melanoma with unknown primary. It unfortunately has not been very good at finding unknown primaries as in squamous cell carcinoma. Kole et al study 20 patients of which 8 were melanoma of unknown primary site. None of= hese patients had a primary site identified after PET imaging. O’neill et al also examined = 40 patients diagnosed with a melanoma of unknown primary site, all of which un= derwent PET imaging. None of these patients had an identification of a primary site. While PETimaging has not been shown to help evaluate for a primary site, it= has be useful in O’neill’s study at identifying further metastatic disease. PET imaging is especially limited in the brain and scalp due to hi= gh background activity of the brain.
Due to the unique nature of melanoma of unknown p= rimary site, the staging system is slightly different and only takes into account = the N and M staging. Melanoma of unknown primary site is classified as stage III = or stage IV disease. Any metastatic disease automatically upstages the patient= to stage IV. Although theories f= or the nature of this unique disease cannot be fully explained, it has been shown = that this disease has a similar, if not improved, outcome compared to cutaneous melanomas with lymph node metastases. Experts theorize that immune responses causing regression of the primary tumor incur improved regional and distant control of the disease. With the unusual nature of melanoma of unknown prim= ary site, Pfeil et al sought to determine whether the AJCC 2009 melanoma classification was suitable for melanoma of unknown primary site. They also sought to determine prognostic factors in this patient population. Of the 8= 897 patients, 172 patients had a melanoma of unknown primary site or a 1.9% incidence. The study, comparing patient based on staging found survival rate significantly different. The survival differences included staging as defin= ed by lymph node metastases, number of satellite or in-transit metastases, the size of satellite/in-transit metastases, the number of regional lymph node metastases, number of distant metastases, the total number of metastases , = the M classification at diagnosis, and LDH level at diagnosis. Independent significant prognostic factors in these patients included : age less than 60 years old, stage at diagnosis, the total number of metastases, and the LDH level. Patients with regional metastatic disease had a significantly better overall survival when compared to those with distant metastases. The number= of metastatic regional lymph nodes also proved to be a significant factor in m= ulti-variate analysis. An elevated LDH was= found to be strongest predictor of unfavorable survival in patients with stage IV disease.
Prognosis
The prognosis of melanoma of unknown primary site appears to be better, or similar, to patients with a cutaneous melanoma wit= h metastases. Katz et al reviewed 19 separate case series from 1952 to 2001. The overall percentage of melanoma of unknown primary site in each study was between 1.2-8.1%. There was also consistent ratio of male greater than female. Of the 19 case series, 7 studies s= howed a better prognosis. Nine stud= ies showed a similar or better prognosis. Only one study from 1952 showed a w= orse prognosis. A study from M.D. Anderson performed a Cox regression analysis looking for disease-free survival, disease-specific survival, and overall survival in patients comparing melanoma of unknown primary site with a metastatic melan= oma with a known site (MKP). Overall survival was significantly different betwe= en the patients with unknown primary site and with melanoma of known primary s= ite. There was a statistically significant difference between male and female patients as well. Also, overall survival was statistically significant for patients with N3 disease. In patients with N1b disease, overall and disease-free survival was better in MKP subjects. However, in each of the other stag= es patients with melanoma of unknown primary source had both improve disease-f= ree and overall survival. There are many theories on the cause of such improved prognosis. Many theories are based on immune responses to the primary lesion with a likely complete regression. There are also theories based on the melanoma originating from benign melanocytes in lymph nodes. W= ith regards to treating the disease, the surgeon is able to rid the occult prim= ary with a nodal dissection. Pathological studies of lymph nodes have found ben= ign melanocytes in lymph nodes.
Treatment
The treatment options for patients with melanoma = of unknown primary site include surgery, radiation, and/or systemic therapy. T= he mainstay surgery for these patients is a lymphadenectomy or lymph node dissection. Lee et al found improved survival in patients after lymphadenec= tomy from a nodal metastasis from an unknown primary melanoma. Their research sh= owed a 5 year survival of 55% in patients after lymph node dissection compared at 27% in patients without lymph node dissection. Their conclusions include most pati= ents with melanoma of unknown primary site can generally expect long-term surviv= al after an adequate lymphadenectomy. They matched patients based on tumor bur= den and showed improved overall survival in the patients with melanoma of unkno= wn primary site compared to those with cutaneous melanoma. Also, patients with stage IV disease were examined by Lee et al. They found the patients did not have an improved prognosis and M1a disease but otherwise had a significant improvement in overall survival in patients with an unknown primary site ve= rsus those with a known primary site.
There's controversy on the use of radiation in pa= tients with melanoma and with a melanoma of unknown primary site. Most believe the melanoma, in general, is not radio sensitive. Ballo et al at M.D. Anderson showed positive data for local and regional control after surgical treatmen= t in patients with melanoma. They found regional control rates of 89% at 5 and 10 years in patients with stage I or stage II melanoma. They concluded that radiation is a useful adjunct especially in patients in whom lymph node dissection or systemic therapies are not options. The suggested indications= for adjuvant radiotherapy include evidence of extracapsular spread, multiple ly= mph nodes, large lymph nodes greater than 3 cm, or recurrent disease. Extracaps= ular spread incurs a 5% higher regional recurrence rate in patients. Ballo et al found that adjuvant radiotherapy reduced regional recurrence. However, this= was not statistically significant. Bastiannet et al performed a systemic review and found that radiothe= rapy improved locoregional control without improving overall survival. Moncrief = et al have suggested that the most important factor preventing locoregional recurrence of melanoma is the primary surgical clearance of the nodal field. Extracapsular spread and large or multiple nodes indicate a higher risk of distant spread and overall mortality. Both appear to remain unaffected by t= he use of adjuvant radiotherapy. Adjuvant treatments including chemotherapy and systemic therapy have been looked on. Most of these have suggested no improvement in comparing patients with melanoma of unknown primary site with those of a known primary site. As studies continued to look at gene-targeted therapy and melanoma, new drugs = are being tested to help in a fight to cure melanoma. Most of these medications= are now in either phase I or phase II trials. There is hope then the next sever= al years there will be new adjuvant treatments available for patients with mel= anoma both of unknown and a known primary site.
Conclusion
Metastatic melanoma of unknown primary site conti= nues to be unique entity in head and neck cancer. Due to its unique presentation= and differences with its counterpart, melanoma of unknown primary site should be treated aggressively. A thorough workup is required in the diagnosis of this disease. Surgery is the main treatment option for these patients, and an aggressive surgical approach wi= ll yield the best results. Questions continue on whether there is a really improved prognosis compared to cutaneous melanoma with metastases. Most pre= sent studies would agree that a melanoma of unknown primary site does incur a be= tter prognosis. There continues to be limited data on adjuvant therapy on patien= ts with a melanoma of unknown primary site. Further studies are currently being done evaluating the role of different genetic markers and protein mutations= in melanoma. Melanoma requires a multidisciplinary management team led by the surgeon
Works Cited
--
Metastatic Melanoma of Unknown Primary: A Unique Entity? November 2012