What every physician needs to know:
Melanoma: an introduction and its initial evaluation
Melanoma occurs in over 76,300 individuals in the US annually, with over 10,000 deaths. If diagnosed early, melanoma is easily cured. The AJCC staging provides valuable information on prognosis. In early stage T1a, stage IA the cure rate is well over 90%. In contrast, those patients with deep (>4mm) ulcerated T4b primary lesions have a survival of under 50% at 5 and 10 years.
The recognition of a melanoma early in its course is therefore of great importance and can markedly impact upon the patient’s survival. While the ABCD (asymmetry, border, color, diameter) of melanoma is well known, the E for evolution is not as appreciated as it should be. New or changing lesions can look very benign, but harbor deep invading melanoma.
Location: biology – therapy
The histology and location of melanomas have recently become linked to its biology. (Figure 1) Furthermore, the biology of melanoma has now become relevant to its systemic therapy (see below):
Lesions on the torso or other body areas where sun exposure is only intermittent associated with intense burns as a child or young adult are now recognized to display mutations in the BRAF gene at V600 in over 50% of cases. The importance is the fact that BRAF V600 mutations are a target for new therapy.
In contrast, lesions found on the hands (primarily palms) or feet (primarily soles) or under the nailbed (subungal) are acral by definition. In these cases mutations in the BRAF gene at V600 are observed in only 15-20% of cases while mutations of CKIT are reported in 15-20% of cases, which also may impact therapeutic options.
Melanoma originating in the head and neck region and occasionally on the extremities where there is chronic solar damage display a different genetic make-up. These lesions demonstrate BRAF V600 mutations in only about 10-20% of the time, while CKIT mutations may be seen in up to 5%.
Melanoma can originate from mucosal surfaces such as the rectum/anus, sinuses, oral cavity, vaginal/vulvar surfaces. These lesions originating from mucosal surfaces have always been viewed as extremely aggressive. Recent studies have demonstrated only very rare BRAF mutations and about 20% with CKIT mutations.
Melanomas can also originate from the uveal tract of the eye. These melanomas primarily originate from the choroid and demonstrate mutations in the G-binding proteins GNAQ and GNA11, in the majority of these cases. BRAF and CKIT mutations are extremely rare if they occur at all in the uveal tract melanomas. About 15-20% of melanomas of the cutaneous surfaces express mutations in the NRAS protein. These predominantly occur at codon 61 and less frequently at codon 12 and 13. These are activating mutations that are generally mutually exclusive of BRAF V600 mutations.
Later the significance of these genetic alterations in melanoma will be discussed in the context of treatment.
Are you sure your patient has melanoma? What should you expect to find?
Disease manifestations of melanoma
Primary melanoma can originate from any body surface including skin, mucosa including vaginal, anal, and oral/gastrointestinal tract, and even the uveal tract of the eye.
In addition, a relatively large number of melanomas are diagnosed at sites of metastases without an obvious primary site. Those cases with unknown primaries can present as an involved single lymph node basin, isolated skin or soft tissue mass, or even an isolated lesion in a major organ (lung and brain) mimicking a primary brain tumor or lung cancer.
In some cases, skin metastases are very difficult to discern from primary lesions.
Interestingly, patients presenting with an isolated lymph node site and no primary lesion appear to have a better prognosis than those with a known primary lesion.
In addition, many times disseminated disease is the first presentation of melanoma without a known primary lesion.
Frequently, melanoma can present as a changing lesion at the primary site which may bleed, be pruritic, fail to heal, or have discharge.
Lesions of the acral surfaces and other mucosal surfaces may be especially difficult to diagnose unless there is a high index of suspicion.
Lesions may occur in subungual sites and appear as traumatic hematomas under the nail.
Anorectal melanomas frequently masquerade as benign thrombosed hemorrhoids.
Distant metastases can involve a great diversity organs. This includes frequent skin, lymph node, and soft tissue involvement in nearly 40% of patients.
Small bowel is a site where melanoma is the most frequent metastatic tumor. Small bowel obstruction or a slow gastrointestinal bleed and anemia are two common scenarios with small bowel metastases.
Brain is a site which appears especially receptive to spread of melanoma. Most, if not all patients, with disseminated disease will develop brain metastases in the setting of uncontrolled metastatic disease.
In the past, bone involvement had been considered to be infrequent. More recently, with widespread use of PET scanning, it has been recognized more frequently. It typically features bone marrow involvement as opposed to lytic or sclerotic cortical lesions.
Additionally, on occasion, primary melanomas originating in the extremities can disseminate in the affected limb prior to involvement of the draining lymph nodes for long periods of time. This in-transit disease is most prevalent in the leg, but can also be observed in the upper extremities or in the head and neck region. These lesions are a result of disease trapped in lymphatics where it can become confluent and a significant local or regional problem. The lower extremity lesions can occasionally be treated through isolation of the limb and perfusion of the extremity with chemotherapy (melphalan), while heating the leg. This limits systemic exposure to the high doses of chemotherapy administered to the leg. While local control can be achieved, patients will frequently die of systemic disease. However, isolated limb disease can still be a prolonged symptomatic problem. Direct injection into superficial melanoma lesions with Talimogene laherparepvec (an oncolytic virus expressing GM-CSF often simply called “T-VEC”) is also effective and was approved for this indication in 2015.
Beware of other conditions that can mimic melanoma:
Diseases that may mimic melanoma
This is frequently an issue when patients present with a mass or disseminated disease without a history of an underlying malignancy. Melanoma is frequently poorly differentiated and a consideration whenever there is a malignancy of unknown primary. The melanoma cells demonstrate large nucleoli and occasionally evidence of pigment which makes the diagnosis easier if melanin is present.
Most helpful are special immunohistochemical stains, including the following:
These are by no means uniformly positive, but usually one of them demonstrates expression and directs the diagnosis in the direction of melanoma. Therefore melanoma is a consideration, when cases may present similar to breast cancer (axillary lymph node enlargement), lymphoma (with diffuse lymphadenopathy and B symptoms), or even as a primary brain tumor (glioblastoma multiforme).
At times, primary lesions can appear very unimpressive with none of the commonly described characteristics (ABCD). Frequently nodular melanoma is described to look like a “blood blister.” Other lesions may be new or changed but still have well defined borders, no color variation, and even small size (diameters less than 5mm). Some may be flesh colored without any pigmentation (amelanotic).
Histologically the differentiation between an invasive melanoma (Clark level II and III) that does not grow vertically and a melanoma-in-situ or dysplastic nevus can be difficult. Many times these may be treated as early stage melanoma with appropriate margins (1 cm or less). The lesion which can be most difficult to discern from melanoma is the Spitz nevus which occurs in children and young adults.
By gross and microscopic appearance, both lesions have many classic features of malignancy. The melanoma cells usually exhibit great variation, increase size under the microscope great atypia and “invasion into connective tissue.” Most frequently, a diagnostic problem is confronted in young adults. Mutational analysis has been used to help differentiate benign from malignant. In the end, many patients undergo wide excision and sentinel lymph node mapping both as a diagnostic and therapeutic procedure.
Which individuals are most at risk for developing melanoma:
Both acute intense ultraviolet radiation in the form of UVA and chronic sun exposure in the form of UVB have been associated with risk of melanoma.
Intense sun exposure with one or more severe burns at a younger age.
Chronic sun exposure over many years – generally older males.
Exposure to tanning booths among younger adults or teenagers.
Clinical setting associated with significant host immune suppression – especially with Cyclosprin A or tacrolimus.
Bone marrow transplantation – organ allograft.
Patient’s own history of melanoma.
Family history of melanoma, especially those with 3 or more first degree relatives with or without pancreatic cancer.
Dysplastic nevi syndrome.
Skin color, red hair, freckles, easy burnability – association with MC1R polymorphism and melanomas with the BRAFV600
Prevalence is gradually increasing in older males but for many years ratio had been 1:1 males versus females.
Median age now 57 years but frequency is higher in females in their 20s and males over 50 years of age.
Most prevalent in Caucasians especially with light skin and easy burnability – Australian, Scandinavian, North American.
Rare in more pigmented races/ethnicities – African, African-American, Hispanic, Asian. More frequently acral and mucosal in these populations.
What laboratory and imaging studies should you order to characterize this patient's tumor (ie, stage, grade, CT/MRI vs PET/CT, cellular and molecular markers, immunophenotyping, etc.) How should you interpret the results and use them to establish prognosis and plan initial therapy?
Primary lesion biopsy
The manner in which the cutaneous lesion is first biopsied in order to make a diagnosis is of great importance. Either an excisional biopsy or a punch biopsy through the depth of the deepest part of the lesion is greatly preferred over a shave biopsy. This will allow a much better appreciation for the initial stage of the primary.
The Breslow depth, Clark level, presence of ulceration, and number of mitoses (>1/mm2) are all used to stage primary lesions and to decide on further staging. Physicians need to know that it is rare that a patient with a primary melanoma needs systemic staging such as, positron emission tomography (PET) scans, computed tomography (CT) scans, or brain imaging.
Systemic staging is generally indicated only for very deep primary lesions (T4) with ulceration (T4b) or those with documented involved lymph nodes. At present there is even some debate whether systemic staging is required for microscopic lymph node involvement and early primary lesions (less than 2mm in depth).
Patients who undergo sentinel lymph node mapping based on the stage and depth of their primary have their lymph nodes carefully analyzed by the pathologist for even a minute collection of melanoma cells. This involves thin sectioning and staining with standard H&E as well as immunohistochemical staining for melanoma markers.
Sentinel lymph node involvement can be categorized according to tumor burden: diameters less than 0.1mm; 0.1-1.0mm; or greater than 1.0mm or even in some cases >10.0mm. These differences in tumor burden correlate with both tumor involvement in the non-sentinel nodes from the completion lymph node dissection (CLND) and patient outcome/survival.
See Table I. TNM staging of skin melanoma.
Due to the challenges of accessing archival tissue, it is reasonable to perform further molecular analysis at the time of diagnosis in cases that have a sizeable risk of recurrence requiring systemic therapy (Stage IIIA-C, IIB, IIC, and possibly earlier). These include:
BRAF: with BRAF and MEK inhibitors now available, knowledge of the presence of the BRAF V600 mutation is potentially critical. The combination of Trametinib (MEK inhibitor) and Dabrafenib (BRAF inhibitor) or Cobimetinib (MEK inhibitor and Vemurafenib (BRAF inhibitor) have become the standard of care for BRAFV600
mutant melanoma based on several phase III trials (see below). Adjuvant studies are underway. Metastatic studies demonstrate a strong survival advantage over single-agent BRAF targeted therapy and, by inference, over standard chemotherapy. This mutation will be found in approximately 50% of cutaneous melanomas originating from the torso or other body areas with intermittent sun exposure. Although less frequently, it is also seen in melanomas from chronic sun damage sites, and acral sites as well.
CKIT: CKIT mutations are observed in acral and mucosal melanomas primarily but also in regions with chronic sun damage (5-10%) and less frequently on intermittent sun exposed skin. KIT inhibitors have proven to be active in these melanomas when disseminated.
Other genetic alterations are likely to gain in importance over the next few years. Already studies are directed at NRAS mutated melanoma. Clinical activity has been observed with MEK inhibitors and now larger phase III trials are examining the true benefit of these agents in NRAS mutant melanoma. Early results have demonstrated a significant benefit for progression-free survival (PFS) but no difference in overall survival (OS) when the MEK inhibitor, Binmetinib was compared to chemotherapy. The ability to use next generation sequencing especially with a targeted capture approach has allowed one to explore for mutations in the entire sequence of hundreds of different cancer related genes.
Genetic mutation analysis may be delayed until treatment is indicated, but in those cases, tissue may not be available. Additionally, it may be worth considering genetic mutation analysis in patients with very high-risk disease but no overt metastatic disease.
What therapies should you initiate immediately i.e., emergently?
Emergency scenarios associated with melanoma
It is rare that a primary melanoma is associated with a clinical emergency, but it is critical that it is recognized as early as possible by the patient and physician as suspicious and a biopsy performed. Other situations where emergent surgery or radiation may be indicated include:
Surgery for large, painful and ulcerated lymph node metastases which are impinging on vessels or nerves.
Surgery for brain metastases that are isolated and symptomatic (or asymptomatic) when the extracranial disease is limited or controlled.
Surgery for small bowel metastases which are causing blood loss and anemia or bowel obstruction.
Radiation therapy in patients with controlled or limited systemic disease. Stereotactic radiation therapy for limited number of brain metastases (<5 lesions) whole brain radiation therapy (WBRT) may follow in some cases.
WBRT for multiple brain metastases in patients with poorly controlled systemic disease or poor performance status is still performed. However, the value of WBRT in melanoma is questionable and this modality is frequently avoided.
Other clinical emergencies associated with disseminated cancer seen with melanoma:
Spinal cord compression – radiation therapy or surgical debulking and then radiation therapy.
Pathologic bone fracture of long bone (weight bearing) – surgery and radiation therapy.
What should the initial definitive therapy for melanoma be?
Approach to Primary lesions
1. Adequate whole thickness punch biopsy or incisional biopsy.
2. If primary is narrow (<1 mm) and it is greater than 0.75 mm, displays Clark level IV (into the reticular dermis), ulceration, or excess mitoses (>1/mm2) then a sentinel lymphatic (SLN) staging is considered. Lesions greater than 1mm are generally mapped at the time of the wide excision.
3. Initially any laboratory testing or CT or PET scanning is not recommended.
4. A wide local excision recommended (when feasible) –
melanoma in situ- excision with clear borders
melanoma primary T1a/b: 1 cm margin
melanoma primary T2a/b: 1-2 cm margin
melanoma primary T3a-T4b: 2 cm margin
5. At time of sentinel lymph node staging – chest x-ray and LDH.
6. If deep primary lesion greater than 4 mm or positive lymph nodes (stage N1a-N3) – full body staging (PET preferred or CT of chest/abdomen/pelvis) and brain MRI or CT scan.
7. For patients with palpable lymph nodes, it may be reasonable to perform lymphatic mapping to determine if there are additional draining lymph node basins. In addition, some palpable lymph nodes may turn out to be benign or reactive. Nevertheless, if gross lymph nodes are found then a complete lymph node dissection should be performed. Optimally, this entails removal and analysis of greater than 15 lymph nodes from the head and neck, greater than 10 lymph nodes from the axilla, or greater than 5 lymph nodes from the groin.
Upon completion of primary and regional staging, the AJCC staging criteria provide the patient and physician with an excellent understanding of their prognosis effecting their need or acceptance of adjuvant therapy with IFN, their eligibility for clinical trials, and the intensity of clinical follow-up. In the inguinal lymph nodes some surgeons will include pelvic lymph node dissections if either suspicious on CT scan or a positive Cloquet’s node (node at entry to pelvis).
Sentinel lymph nodes
An important question to address is whether the sentinel lymph node needs to be mapped. This decision is based on the depth of the lesion (Breslow stage).
Lesions with the following characteristics should be considered for lymph node mapping:
greater than 0.75-1.0 mm in depth
Clark level IV (into the reticular dermis)
excess mitoses (>1/mm2)
Either of the latter two characteristics constitute a T1b stage. Lesions greater than 1mm are generally mapped at the time of the wide excision. At the time of sentinel lymph node biopsy it is reasonable to obtain a serum lactate dehydrogenase (LDH) and chest x-ray. The risk of lymph node disease increases with depth and ulceration of the primary.
Completion lymph node dissection
Standard of care remains completion lymph node dissection (CLND) for any positive sentinel lymph node, but the approach is evolving and there are ongoing studies to evaluate observation with periodic ultrasound versus CLND for positive sentinel lymph nodes. Some studies have suggested this group of patients may be followed with close observation. Some experts advocate against performing CLND on those with under 0.1 mm of tumor in their standard lymph node dissection (SLND).
Once staging of sentinel lymph node is completed, patients may generally proceed directly to a CLND. Systemic and brain imaging can be performed at completion of all surgery, when the entire T and N staging can accurately be defined and risk for distant outcomes is much better appreciated. This then completes the initial staging of melanoma and decisions concerning adjuvant therapy can be made.
There are some exceptions to this general rule including those patients with very deep ulcerated lesions (T4b) who may have systemic disease even in the absence of nodal disease. At the end of primary and regional staging, the AJCC staging criteria provides the patient and physician with an excellent understanding of their prognosis effecting their need or acceptance of adjuvant therapy with IFNα, pegylated IFNα, or Ipilimumab, their eligibility on clinical trials, and the intensity of clinical follow-up.
Finally, sentinel lymph node mapping and completion lymph node dissection ultimately may improve the regional control of melanoma, although there is no definite evidence that the use of sentinel lymph node staging improves overall survival. A recently accrued trial has examined the role of completion lymph node dissection versus periodic lymph node ultrasound in patients with involved sentinel lymph node(s). Some data shows that very small volume of disease in the SLN (<0.1 mm diameter for largest deposit) is associated with an excellent prognosis and does not require completion LN dissection.
Patients who present with palpable or symptomatic regional lymph node involvement with melanoma, whether synchronous with the primary melanoma, or asynchronous following the primary some time later, or without a known primary, should undergo complete staging of the body (PET or CT) and brain (MRI or CT).
If staging reveals no other disease, a CLND should be performed. In the inguinal lymph nodes some surgeons will include pelvic lymph node dissections if either suspicious on CT scan or a positive Cloquet’s node (node at entry to pelvis).
Adjuvant therapy for high-risk surgically resected disease
There is still a great deal of controversy toward the treatment of high-risk melanoma with 12 months of high-dose interferon (IFN) adjuvant therapy. While mortality from therapy is very rare, there is a major impact on quality of life with depression, myalgias, arthralgias, anorexia and a number of other chronic adverse events.
Several Eastern Cooperative Oncology Group (ECOG) trials demonstrate a significant survival advantage for high-risk melanoma patients with either stage T4a/b or nodal disease (IIB, IIC, IIIA/B/C). However, overall survival benefit has been inconsistent among the many trials and meta-analyses show only a 3-5% absolute and 10% relative increase in survival. The benefit has not been consistently seen in one subset of high-risk patients versus another subset and varies among different clinical trials.
Several retrospective analyses have suggested that IFN largely benefits patients with ulcerated primaries and microscopic lymph node disease with little benefit to those without ulcerated lesions or with large macroscopically involved lymph nodes. There has been no question that high-dose IFN can delay the recurrence of melanoma by a number of months, but due to its toxicity and duration of treatment this has been seen as only a minimal advance by many.
Evidence for adjuvant systemic therapy
E1684 – most patients at stage IIIB-C disease; small size of study (287 pts total)- demonstrated an overall survival advantage of approximately 9% from 37% to 46% at 5 years and significant DFS improvement.
E1690 – no statistical advantage to high dose IFN in terms of overall survival. This studied included many T4a/b cases without lymph node dissection. It has been hypothesized that patients in the observation arm who relapsed in regional LNs may have experienced IFN salvage.
E1694- IFN compared to GM2-KLH vaccine – both overall survival and DFS benefit.
The EORTC trials with PEG-Intron demonstrate no overall survival benefit but clear DFS improvement.
High-dose IFN regimen:
IFN alfa-2b 20 million units/m2 IV 5 days per week for 4 weeks; followed by
IFN alfa-2b 10 million units/m2 subcutaneously three days per week for 11 months.
Due to toxicities and prolonged duration of therapy, the identification of patient subsets most likely to benefit from high-dose interferon has received considerable focus. There is no data that suggests the benefit is greater in IIIA than IIIB disease (initially suggested by EORTC with pegylated IFN [sylatron]). An EORTC retrospective analysis demonstrates an apparent benefit in patients with N1a disease and ulcerated lymph nodes and this will be assessed prospectively.
PEG-IFN administered weekly for 2 years has just recently been approved by the FDA based on previous trials under the guidance of the EORTC. While the trials did not demonstrate any survival advantage it did demonstrate a clear improvement in disease-free survival and metastatic disease-free survival.
Early results from an EORTC trial (18071) have demonstrated superiority of ipilimumab at 10 mg/kg (not the approved dose for metastatic disease) over placebo in relapse-free survival (RFS) with a median RFS of 26 months versus 17 months for placebo (HR 0.75; 95% CI, 0.64-0.90; P=0.001). Overall survival data are not yet available. Trials comparing several doses (3 mg/kg and 10 mg/kg) of ipilimumab to high-dose IFN have completed accrual with results pending. The ipilimumab arm experienced significant autoimmune related toxicities with both diarrhea and colitis of all grades in up to 40% of patients and endocrinopathies of grade 3 and 4 in over 15% of patients. Five deaths were observed in the ipilimumab arm.
Currently, IFN, PEG-IFN, and ipilimumab (10 mg/kg IV) are approved adjuvant treatments for stage III melanoma. IFN is also approved as adjuvant therapy for high-risk stage II melanoma. Results of the E1609 trial comparing IFN and ipilimumab are widely anticipated. Other adjuvant treatment strategies under investigation include immune checkpoint inhibitors such as anti-PD-1 (nivolumab or pembrolizumab) or targeted treatment for BRAFV600
Radiation therapy may be considered in two scenarios:
Desmoplastic melanoma- found primarily on scalp demonstrating a strong stromal fibrotic reaction, primarily dermal infiltration, and neurotropic growth pattern, are associated with a high incidence of local and in-transit recurrence and possibly a lower rate of lymph node involvement. Several single arm studies suggest a benefit to local radiation of the primary especially when margins appear inadequate. It is also now known that these tumors have a high rate of single nucleotide variants (SNV) or mutations across the genome and may respond to anti-PD-1/PD-L1 therapies.
High-risk regional lymph nodes- these include 4 or more lymph nodes, extracapsular extension frequently with matted lymph nodes, palpable head and neck lymph nodes. Only recently post-operative radiation in this stage of patients has demonstrated a benefit from radiation, with a lower incidence of local and regional recurrence but a dismal overall survival from disseminated disease, and some increase in lymphedema in the treated arm and other long-term morbidities.
Treatment of systemic stage IV and unresectable advanced stage IIIC
The AJCC has divided staging of M1 disease into M1a (soft tissue, skin, and distant lymph nodes), M1b (lung), or M1c (visceral organs or elevated serum LDH above the upper limit of normal [ULN]). These stages have prognostic significance with median survival in the past, prior to effective standard treatment, at approximately 13-15 months for M1a, 10-12 months for M1b, and 6-8 months for M1c. Significant LDH elevation alone provides the most significant prognostic marker of poor outcome. In addition, as with most other cancers, performance status has prognostic significance. However, changes in management and the available agents have changed these results markedly now with some median OS of >24 months and possibly longer with checkpoint-based immunotherapy.
When a patient presents with isolated metastases that can be resected, there is a potential for a long disease-free period and even more lengthy overall survival. These patients likely have a very unique biology that limits their metastases in number at a time (oligoclonal). They can be managed over many years with surgery alone. They have a much better prognosis even if M1c (very unlikely high LDH).
Systemic treatment options according to molecular characterization of the melanoma are listed in Table II.
What should you tell the patient and the family about prognosis?
The prognosis of primary melanoma is well defined by a set of staging procedures and large database reviews by the AJCC with over 18,000 patients.
Based on T stage one can be assured of a very low risk of recurrence and death (albeit not zero) if disease is T1a.
Ulceration upstages each lesion.
Ulceration of the primary melanoma with regional lymph node involvement represents a very poor prognostic situation.
Presence of regional lymph node involvement is likely overall the most important prognostic factor for localized melanoma.
Those patients with N1b, N2b, or N3 disease have a much worse prognosis, with those with N3 disease having a risk of recurrence as high as 85-90% in some cases.
The disease can recur in-transit between the primary and the regional lymph nodes. This occurs primarily in those patients with a deep primary that is ulcerated or a desmoplastic melanoma on the scalp or elsewhere. Local recurrence or in-transit disease carries an extremely poor prognosis and will likely recur beyond the region in time, but can remain limited to the region for long periods of time (years) without distant metastases. Recurrence in regional lymph nodes has become less frequent due to the incidence of SLND, but still represents a site of potential involvement.
Metastatic disease is most frequent in soft tissue, skin, lymph nodes and lungs. Other visceral organs are also involved including liver, brain, adrenal glands, small intestines, and even bone. Metastatic disease carries a very poor prognosis but presently there are a number of factors which should be taken into consideration and mortality rates and duration of survival is influenced by a number of factors.
Systemic treatment – treatment of disseminated disease
Until recently, options for systemic treatment of unresectable metastatic disease were very limited and in general clinical trials was recommended as standard therapy for most patients.
Chemotherapy has demonstrated some activity, but with more rigorous disease monitoring response rates have dipped to below 10% and the overall survival had not been improved from a median of approximately 8 months. Single-agent therapy with Dacarbazine, Temozolomide, taxanes have all proven to have response rates in the 5-15% range and no major influence on survival.
Chemotherapy combinations with dacarbazine and platinums slightly improved response rates but at the cost of significantly more toxicity and no improvement in survival. Even the addition of biotherapy with Interleukin-2 and Interferon to combination chemotherapy ultimately was unable to improve overall survival. Therefore today, the role of chemotherapy is extremely limited and is rarely a good palliative approach.
With the advent of both new checkpoint-based immunotherapy and BRAF targeted therapy with combined BRAF and MEK inhibitors, the entire landscape and outlook for melanoma patients has changed.
Checkpoint Inhibitor Immunotherapy
Iplimumab is an cytotoxic T lymphocyte antigen-4 (CTLA-4) monoclonal antibody which acts to inhibit a checkpoint in the initial activation of the T lymphocyte. These checkpoints act to modulate the T cell activation and bring an activated T cell back to its resting state. The drug can induce autoimmune events due to the release of an activated immune system not only against the cancer but also normal tissues. These events include colitis with rash, diarrhea, hepatitis, endocrinopathies (pituitary, thyroid, adrenal gland), uveitis, and even neuropathies. Ipilimumab approval had been based on two randomized phase III clinical trials. First published in 2010 was a randomized double blinded trial where patients were randomized to Ipilimumab alone, ipilimumab with a peptide vaccine, or the peptide vaccine alone. Ipilimumab was administered at 3 mg/kg. The randomization was 1:3:1 and the peptide vaccine was derived from the gp100 protein and restricted for patients carrying the HLA-A0201.
Patients were all HLA-A0201 positive and had failed previous therapy. While the response rates were 11% or under for the Ipilimumab arms and the PFS increase was not impressive, there was a significant overall survival advantage to those on ipilimumab with or without peptide vaccine.
The median overall survival was 10.0 months among patients receiving ipilimumab plus peptide vaccine, as compared with 6.4 months among patients receiving peptide vaccine alone (hazard ratio for death, 0.68; P<0.001). The median overall survival with ipilimumab alone was 10.1 months (hazard ratio for death in the comparison with peptide vaccine alone, 0.66; P = 0.003).
In 2011, another trial examined the benefit of ipilimumab as initial therapy in metastatic melanoma. This trial compared Dacarbazine with or without Ipilimumab and explored a higher dose of ipilimumab. Ipilimumab was given at 10 mg/kg IV in this trial and at 3 mg/kg IV in the trial discussed above in previously treated patients.
Overall survival was significantly longer in the group receiving ipilimumab plus dacarbazine than in the group receiving dacarbazine plus placebo (11.2 months vs. 9.1 months, with higher survival rates in the ipilimumab–dacarbazine group at 1 year (47.3% vs. 36.3%), 2 years (28.5% vs. 17.9%), and 3 years (20.8% vs. 12.2%) (hazard ratio for death, 0.72; P<0.001).
Ipilimumab 3 mg/kg IV every 3 weeks x 4 doses. Follow-up maintenance therapy has not been proven to add efficacy. However, for patients receiving a benefit from a single course of Ipilimumab, a reinduction with another course of Ipilimumab can be effective at inducing another benefit.
Anti-Programmed Cell Death-1 (anti-PD-1) or anti-Programmed Cell Death-L1 (anti-PD-L1) antibodies.
There are numerous antibodies binding to and inhibiting these targets that are at various stages of development. For melanoma, the anti-PD1 antibodies pembrolizumab and nivolumab are currently approved.
Similar to ipilimumab, pembrolizumab is associated with multiple autoimmune toxicities. However, the adverse event profile of these agents is somewhat different from anti-CTLA-4 drugs, with lower rates of endocrinopathies and diarrhea, but potentially more pneumonitis. In clinical trials, response rates approximate 40% among patients who have not received prior CTLA4-targeted therapies, and over 30% among those previously treated with ipilimumab. Responses tend to be durable, with a median duration of almost 2 years. Among all patients, 2-year overall survival rates approach 50%. Pembrolizumab has been shown to be superior to Ipilimumab in terms of PFS and OS as first-line therapy for metastatic melanoma. In patients who have progressed after Ipilimumab, Pembrolizumab was found to be superior to chemotherapy of choice in PFS and OS. The benefit of first-line PD1-directed therapy is most apparent in patients whose tumors do not harbor BRAFV600
mutations. In BRAFV600
cases, the optimal sequence of immunotherapy and targeted therapy remains controversial and is being addressed in an ongoing trial (EA6134).
Pembrolizumab 2 mg/kg IV every 3 weeks until disease progression or unacceptable toxicity. (In the near future, the agent may be administered with a fixed dose not dependent on patient body weight.)
Another anti-PD1 monoclonal antibody, nivolumab, has a similar toxicity profile to pembrolizumab. Objective response rates of approximately 40% have been seen in both previously untreated patients, and just slightly lower in those who have progressed on prior anti-CTLA-4 therapy. Its efficacy is generally considered the same as pembrolizumab.
Initially, nivolumab in previously untreated patients with advanced melanoma was tested in a double-blind phase 3 study. Four hundred and eighteen previously untreated patients who had metastatic melanoma without a BRAF V600 mutation were randomized to receive nivolumab (at a dose of 3 mg/kg every 2 weeks) or dacarbazine chemotherapy every 3 weeks. The primary end point was overall survival. At 1 year, the overall rate of survival was 72.9% (95% confidence interval [CI], 65.5 to 78.9) in the nivolumab group, as compared with 42.1% (95% CI, 33.0 to 50.9) in the dacarbazine group (hazard ratio for death, 0.42; 99.79% CI, 0.25 to 0.73; P<0.001). The median progression-free survival was 5.1 months in the nivolumab group versus 2.2 months in the dacarbazine group (hazard ratio for death or progression of disease, 0.43; 95% CI, 0.34 to 0.56; P<0.001). The objective response rate was 40.0% (95% CI, 33.3 to 47.0) in the nivolumab group versus 13.9% (95% CI, 9.5 to 19.4) in the dacarbazine group (odds ratio, 4.06; P<0.001).
Four hundred and five melanoma patients all who had failed ipilimumab therapy were randomized 2:1 between either nivolumab 3 mg/kg versus chemotherapy of choice. With more than 6 months’ follow-up of 20 nivolumab-treated patients and 47 patients treated with chemotherapy, response rate was 32% versus 11%. In a subset analysis, PDL1+ tumors had a 44% RR versus 20% in PDL1 negative tumors.
Nivolumab has also been studied in combination with ipilimumab. The combination demonstrated a higher objective-response rate (over 50%) and progression-free survival (over 11 months) for the entire population compared with ipilimumab or nivolumab alone (not a primary endpoint). However, toxicities were increased. Only about 50% of patients are able to complete the initial 12 weeks of combined therapy due to toxicities. Interestingly those patients do equally well to those who can continue therapy in terms of response rate and outcome. The benefit appeared to be more pronounced in patients with melanoma that did not express PD-L1 by IHC. Again overall survival information is pending and is critical to make decisions about treatment selection.
Nivolumab 240 mg IV every 2 weeks until disease progression or unacceptable toxicity.
Combined nivolumab and ipilimumab regimen:
Ipilimumab 3 mg/kg IV and nivolumab 1 mg/kg IV given every 3 weeks for 4 cycles followed by nivolumab 240 mg IV every 2 weeks until disease progression or unacceptable toxicity. However, the optimal duration of therapy is not known. Many physicians should consider stopping therapy after 6 months if the disease is in a CR or a maximal response has been achieved. More recently, there are studies comparing this regimen to a regimen using the same schedule but with Ipilimumab at 1 mg/kg IV and Nivolumab at 3 mg/kg IV in hopes of maintaining the efficacy and decreasing toxicity.
An ongoing critical question is the selection and use of predictive biomarkers for checkpoint immunotherapy. In general, outside of the combination CTLA4 and PD1 therapy, PD-L1 expression on tumor cells (and in some cases on inflammatory cells) is associated with higher responses to anti-PD-1 therapy. However, it appears that up to 20% of melanoma patients with PD-L1-negative tumors could still respond to therapy. Other biomarkers under investigation include whole exome mutation burden (especially observed in patients who have an underlying DNA repair defect such as MSI, POLE, BRCA-2 etc.), the presence of a dense CD8+T cell infiltrate especially at the “invasive front,” the presence of a RNA expression signature consistent with IFN induced genes or other immune signatures, T cell clonal expansion based on next generation sequencing of the TCR, HLA class II expression on tumor cells, and possibly presence or absence of genetic alterations.
High-dose interleukin-2 (IL2)
In those patients deemed to be inappropriate for complete surgical resection, systemic treatments are the primary therapeutic approach. While the checkpoint inhibitors have become frontline therapy in most patients with advanced melanoma, high-dose interleukin-2 (IL2) has been approved since 1998 for the same indication. This is a very toxic therapy that should be administered to select patients with excellent organ function (cardiac, pulmonary, and no active brain metastases) and aged under 70 years, in a setting and situation with expertise and significant experience in the use of the agent. It only has a 15-20% response rate.
However, after 5-10 years, a small number of patients (5-8%) are disease-free in complete remission. Almost no patients have relapsed after 3 years in complete remission. Therefore, for a very select group of patients, IL2 does hold out a potential for long-term cure. This regimen is associated with a capillary leak syndrome that can include hypotension, fluid retention, renal and hepatic hypoperfusion, and pulmonary edema. Its efficacy following progression on checkpoint inhibitors is unknown.
High-dose IL2 regimen:
IL-2 600,000 or 720,000 IU/kg given in 15-minute infusions IV every 8 hours for a maximum of 14 doses on day 1 and day 15 schedule every 6 to 8 weeks.
Patients rarely receive all 28 doses, with most individuals receiving 18-22 doses.
Molecular biology of cutaneous melanoma
The Genetics of Melanoma
The Cancer Genome Atlas (TCGA) provides an ideal basis for a discussion of targeted therapy of melanomas. The TCGA is based on melanoma tissue that were all cutaneous in origin. The TCGA provides an overall framework for melanoma genomic classification based on presumed driver mutations. This approach has classified melanoma by the predominant driver, BRAF, NRAS, NF1 or “Triple Wild Type” (TWT). The major driver oncogenes and several of those present within the TWT population strongly support mitogen-activated protein kinase (MAPK) pathway inhibition as an important component of any targeted therapy, since over 90% of melanomas in the TCGA have driver genes activating this pathway.
As mentioned, recurrent mutations in the serine threonine kinase BRAF at the 600th codon offered a previously unrecognized approach in treating this disease. Most commonly, this mutation involved a valine to glutamine substitution (V600E), but also included other changes (most often valine to arginine – V600K). Other, non-V600 mutations were also identified in a smaller proportion of melanomas (roughly 5%), primarily in exons 11 and 15, and most commonly codons 466, 469, 597, and 601. Other alterations in BRAF have also been identified at lower frequencies, including fusions, gene amplifications, and kinase domain duplications. The discovery of recurrent NRAS mutations at codons 12, 13, and 61 actually predated the identification of recurrent BRAF mutations. These changes are present in 15-20% of melanomas, with codon 61 mutations overwhelmingly predominating, and less frequent codon 12 and 13 mutations. Less than 5% of the RAS mutated melanomas involve KRAS or HRAS. Loss of function and truncating mutations in the tumor suppressor gene NF1 were also identified in approximately 15% of melanomas (8-11). NF1 mutations often co-occur with other MAPK activating mutations, suggesting that although they promote MAPK signaling, additional alterations are required for oncogenic pathway activation. Other low frequency mutations that promote MAPK signaling were also identified, including those in KRAS, HRAS, MAP2K1, CRAF, and various receptor tyrosine kinase encoding genes. Notably, these “driver” type mutations are largely mutually exclusive (with the exception of NF1). Thus, MAPK signaling is dysregulated and promoted in nearly all melanomas, suggesting that targeting this pathway could be effective in multiple genetically-defined subtypes.
The other cohort of cutaneous melanomas has been identified as triple wild type (TWT) melanomas. Oncogenic drivers in this group are diverse, and include KIT, GNAQ and GNA11 (drivers of uveal melanoma), PDGFRα amplifications, CTNNB1, and EZH2. The overall frequency of a UV signature (C to T inversion or CC to TT inversion) is seen in less than 30% of TWT melanomas, while >90% of those from the other genetic classifications demonstrate a UV signature (15, 16). Finally, copy number alterations (CNA) and other structural events such as TERT, MDM2, MITF, and PDL1 amplifications, may also occur. Despite the somewhat distinct genetic makeup of this group, evidence of MAPK activation is still present in most TWT melanomas.
The stereotypic oncogenic “driver” mutations such as BRAF and NRAS were often found to co-exist with cooperating genetic alterations that promote an invasive phenotype. These include CDKN2A loss or mutations, TP53 mutations, TERT promoter mutations, PTEN loss, or other alterations in the PI3K-AKT pathway. These changes occur in a predictable, stepwise fashion. BRAF V600E mutations occur in benign nevi, whereas NRAS mutations and several other drivers occur in intermediate lesions. These intermediate lesions and melanoma in-situ also frequently harbor TERT promoter mutations. CDKN2A deletions, PTEN loss, and TP53 mutations occur in invasive melanomas. Furthermore, total mutational burden, the total number of somatic nucleotide variants identified in the tumor, accumulate with malignant progression. These co-occurring mutations may provide additional therapeutic targets.
Oncogenic driver mutations are also associated with unique clinical patterns. BRAF mutations occur more commonly in melanomas from skin with intermittent sun exposure, and are infrequent in melanomas arising from skin with chronic sun damage (CSD). If BRAF mutations are identified in skin with CSD, they frequently are V600K rather than V600E. NRAS mutations occur in a relatively predictable 15-20%, regardless of anatomic location (with the exception of uveal melanoma). These alterations are also correlated with an inferior overall prognosis and thicker primary tumors. KIT mutations are present in 15-20% of acral and mucosal melanomas and 2% of CSD melanomas, but rarely in other cutaneous melanomas. GNAQ and GNA11 mutations occur in nearly 90% of uveal melanomas, but rarely in other subtypes. NF1 mutations also commonly occur in skin with CSD and are associated with a high overall burden of somatic mutations.
Tyrosine kinase/signal transduction inhibitors
The 40-50% of cutaneous melanomas that harbor BRAFV600
mutations have activation of the MAP kinase pathway. Pre-clinical studies support the importance of this activated oncogene in the proliferation and survival of melanomas.
An inhibitor to BRAF V600mutant, called PLX4032/RG7204/vemurafenib (Zelboraf), has demonstrated an overall objective confirmed response rate of over 50%, a progression-free survival of nearly 7 months, and an overall survival of well over 13-16 months.
A phase III trial of treatment-naïve patients with BRAFV600
mutant melanoma were treated with vemurafenib versus dacarbazine. The improvement in overall survival and progression-free survival resulted in early termination of the trial at an interim monitoring point. A phase II trial of vemurafenib has demonstrated a median overall survival of nearly 16 months, an outcome far superior to those seen in other large trials.
Vemurafenib 960 mg orally twice daily.
Dose reduction may be needed due to rash, arthralgias or other systemic complaints. Cutaneous squamous cell tumors known primarily as keratoacanthomas are frequent in patients (found in up to 25% of patients) receiving BRAF inhibitors (appears to be more frequent with vemurafenib compared to dabrafenib). Treatment with surgery is adequate and in general patients do not need dose interruption of vemurafenib. This effect appears largely due to the ability of vemurafenib to accelerate the growth of pre-cancerous RAS mutated or MAP kinase activated skin lesions.
Another BRAF V600inhibitor, dabrafenib has also demonstrated response rates in the 50+% range with median progression-free survival of 5.3 months, far superior to chemotherapy. The trial comparing dabrafenib to chemotherapy allowed crossover to dabrafenib in the chemotherapy progressing patients. This made the overall survival endpoint difficult to meet. As time has passed the long-term results suggest an overall survival of 17-18 months.
Toxicities include the same hyperproliferative skin lesions as with vemurafenib but a lower rate of cuSCC. The photosensitivity induced by vemurafenib was not seen in patients receiving dabrafenib. However, the incidence of fevers and chills is considerably higher with dabrafenib than with vemurafenib. Arthralgias and rash also occur in a similar frequency to vemurafenib.
Dabrafenib 150 mg orally twice a day.
Single-agent therapy with trametinib, a MEK inhibitor, improved both PFS and OS compared to chemotherapy in a trial of patients with BRAF V600mutant melanoma also allowing crossover. The response rates were lower than in the BRAF inhibitor trials, but PFS was approximately 4.8 months (only slightly shorter than dabrafenib). Principal toxicities include acneiform rash, fatigue, visual symptoms, and rare cardiac dysfunction
Trametinib 2 mg orally once per day.
Combination dabrafenib + trametinib
The combination regimen has become the standard of care regimen for patients receiving BRAF directed therapy. The regimen has demonstrated an improvement in both PFS (against dabrafenib) and both PFS and OS (against vemurafenib). Recent follow-up shows that overall survival is improved with combination versus dabrafenib. Certain toxicities are considerably less including hyperproliferative skin lesions as cuSCC, while others such as fever and chills are significantly increased.
A phase III trial randomized patients in a blinded placebo design between the combination trametinib plus dabrafenib versus dabrafenib. Four hundred and twenty-three previously untreated patients with melanoma harbouring a BRAF V600E or V600K mutation received either combination of dabrafenib (150 mg orally twice daily) and trametinib (2 mg orally once daily) or dabrafenib and placebo. The median progression-free survival was 9.3 months in the dabrafenib-trametinib group and 8.8 months in the dabrafenib-only group (hazard ratio for progression or death in the dabrafenib-trametinib group, 0.75; 95% CI, 0.57-0.99; P=0.03). The overall response rate was 67% in the dabrafenib-trametinib group and 51% in the dabrafenib-only group (P=0.002). At 6 months, the interim overall survival rate was 93% with dabrafenib-trametinib and 85% with dabrafenib alone (hazard ratio for death, 0.63; 95% CI, 0.42 to 0.94; P=0.02). The rate of cutaneous squamous-cell carcinoma was lower in the dabrafenib-trametinib group than in the dabrafenib-only group (2% vs. 9%), whereas fever occurred in more patients (51% vs. 28%) and was more often severe (grade 3, 6% vs. 2%) in the dabrafenib-trametinib group. With longer follow-up overall survival was improved with the combination regimen in the range of 17 versus 24 months.
A phase III trial compared the same combination to vemurafenib alone. The trial randomly assigned 704 metastatic melanoma patients with a BRAF V600 mutation to receive either a combination of dabrafenib (150 mg twice daily) and trametinib (2 mg once daily) or vemurafenib (960 mg twice daily) orally as first-line therapy. At the pre-planned interim overall survival analysis, the overall survival rate at 12 months was 72% in the combination-therapy group and 65% in the vemurafenib group (hazard ratio for death in the combination-therapy group, 0.69; 95% CI, 0.53-0.89; P=0.005). Median progression-free survival was 11.4 months in the combination-therapy group and 7.3 months in the vemurafenib group (hazard ratio, 0.56; 95% CI, 0.46-0.69; P<0.001). The objective response rate was 64% in the combination-therapy group and 51% in the vemurafenib group (P<0.001). With further follow-up overall survival advantage was observed. Rates of severe adverse events and study-drug discontinuations were similar in the two groups. Cutaneous squamous-cell carcinoma and keratoacanthoma occurred in 1% of patients in the combination-therapy group and 18% of those in the vemurafenib group.
Dabrafenib 150 mg orally twice daily + trametinib 2 mg orally once per day.
Combination vemurafenib plus cobimetinib was evaluated in a phase III trial in which 495 patients with previously untreated advanced BRAFV600
mutant melanoma were randomly assigned to vemurafenib plus cobimetinib or vemurafenib plus placebo. With a median follow-up of 14.2 months, results included the following: PFS, the primary endpoint of the trial, was significantly increased with vemurafenib plus cobimetinib compared with vemurafenib plus placebo (median 12.3 versus 7.2 months, HR 0.58, 95% CI 0.46-0.72). Overall objective response rate was increased with vemurafenib plus cobimetinib (70 versus 50 percent), as was the complete response rate (16 versus 11 percent). Median overall survival was significantly longer with cobimetinib plus vemurafenib compared with placebo plus vemurafenib (22.3 versus 17.4 months, HR 0.70, 95% CI 0.55-0.90).
Cobimetinib was approved by the US Food and Drug Administration (FDA) for use in combination with vemurafenib for patients with metastatic melanoma and a V600 mutation in the BRAF .
Vemurafenib plus cobimetinib regimen:
Vemurafenib 960 mg PO daily over 28 days plus Cobimetinib 60 mg PO daily for 21 out of 28 days
For patients whose tumors lack BRAFV600
mutations, testing for CKIT mutations may be appropriate. There is clear clinical activity of c-kit inhibitors, but the consistency of response is much less than for BRAF inhibitors. Potentially up to 25% of patients will respond. Some of these responses can be durable. C-kit inhibitors include:
NRAS mutant melanoma (Q61)
MEK162 (Binimetinib), an ATP non-competitive inhibitor of MEK1 and MEK2, has been tested clinically in patients with metastatic melanoma harboring BRAF or NRAS mutations. Early encouraging results in NRAS mutated patients have been noted, with an objective response rate of 20%. The median PFS was 3.7 months with a median duration of response of 7.6 weeks. This is consistent with rapid development of acquired resistance, even for responding patients. This phase I/II study led to a phase III trial comparing binimentinib versus chemotherapy enrolling 402 NRAS Q61 mutant melanoma. Twenty percent had prior checkpoint inhibitors, including 13% with ipilimumab and only a few percent (5-6%) receiving prior anti-PD1. The study met its primary endpoint with an improvement of PFS (HR=0.62, p = 0.001). However, the increase in median PFS from 1.5 to 2.8 months was not clinically impressive. ORR favored binimetinib 15% versus 7%. Median overall survival for all patients enrolled was no different (11 versus 10.1 months, HR = 1.0). Results of this trial may lead to approval of the first targeted agent for NRAS mutant melanoma, but the results fall short of the hoped-for improvement.
Single-agent and combination chemotherapy regimens have shown modest activity in melanoma, particularly in individuals with good functional status and lung/non-visceral sites of disease. Response rates range between 5-15%, and there is no clear effect on overall survival.
Dacarbazine 800-1000 mg/m2; IV every 3 weeks
Temozolomide 150-200 mg/m2; orally for 5 days every 21-28 days
Carboplatin AUC 5-6 IV plus paclitaxel 175-225 mg/m2; IV every 21 days
What if scenarios in the treatment of melanoma.
What if a patient presents with isolated cutaneous lesion which is dermal and has no clear connection to the epidermis?
These are difficult cases where we do not know if this is a primary and where we simply do not see the characteristic atypical cellular activity at the epidermal/dermal junction or a dermal metastases from another melanoma.
In this situation it is most favorable to the patient to treat him/her as if this is a primary melanoma and perform a sufficient wide excision and sentinel lymph node mapping if indicated by the depth. In this case patients may have the best chance for long-term survival.
What if a patient has history of melanoma 3 years ago and now presents with an asymptomatic lesion found on chest x-ray?
In this case, imaging should determine the presence of other metastases (CT or PET and brain scan). If none are found, the pulmonary lesion should be resected. There is no need to attempt a biopsy, whether it is a metastatic lesion from melanoma or a primary lung cancer, surgery would be indicated.
Obviously this will be influenced by the actual appearance of the lesion. Infiltrative, cystic lesions or calcified lesions should lead to an evaluation for infection or chronic inflammatory disease. The extent of surgery can be determined intra-operatively when a frozen section if performed.
What if a patient presents with several nodules on the extremity proximal to his primary melanoma – what are the options?
This scenario is not infrequent, especially in association with lesions that are deep (>4 mm) and ulcerated. They can be managed by either:
Surgery – especially if there are few, their recurrences are infrequent and surgery has little morbidity. This can be a strategy for years in some cases.
Local injection of the lesions – this approach is taken when surgery is less practical, based on number or location of lesions. Injection has been with use of a number of agents including BCG, IFN, GM-CSF, or Interleukin-2. Application of imiquimod (Toll-like receptor [TLR] agonist) to the region could also be considered. Recently, another option is Talimogene Laheparepvec (T-Vec), an intratumorally administered, genetically modified clinical herpes simplex virus-1 strain. It appears quite effective in controlling regional disease (in-transit) even in tumor nodules that are not injected. Rarely systemic responses can occur. It has been FDA approved based on a large phase III trial versus systemic GM-CSF.
Limb perfusion or infusion – this is the most aggressive approach but can be very successful in achieving regional control of disease when lesions are too numerous and/or diffuse to consider local measures alone. Likely, this would be the case when there are many more than 5 lesions, lesions with wide distribution, and lesions that return after short periods of time (weeks to a few months). This approach attempts to isolate the extremity and then cannulate either the artery or vein and while infusing high doses of melphalan, heat the leg to high temperatures above 41°C. This does carry risks including arterial thrombosis and compartment syndrome, but can control regional disease for years in well over 50% of patients.
Systemic therapy is a consideration either alone or in combination with surgery. Patients with in-transit lesions will generally go on to develop systemic disease and provide an excellent opportunity to look at new treatments on clinical trials. They also are ideal clinical settings to obtain serial biopsies for defining biomarkers associated with treatment. Standard chemotherapy does have a higher response rate with these patients than those with visceral disease, but these responses are frequently transient.
What if a patient has rapidly growing symptomatic metastatic melanoma and you cannot get the patient’s tumor analyzed for BRAF mutation due to long delays in getting the outside tissue; would you just empirically administer the BRAF inhibitor?
While there is a 50% chance for some cutaneous lesions to express this mutation, it would not be a good idea to empirically treat patients without knowledge of their melanoma mutation status. If the melanoma does not express the BRAFV600
mutation, it very well could have its growth accelerated. Another option is combination checkpoint inhibitor therapy, which can induce rapid control of disease in some patients.
An increase in frequency of cutaneous squamous cell carcinoma (cuSCC) with BRAF inhibitors is likely due to their ability to enhance activation of the wild type (WT) RAF molecules, especially in the setting of upstream activation primarily through RAS. BRAF wild type melanomas might undergo accelerated growth through this same mechanism.
What if a patient has a BRAF V600 mutant melanoma and has limited asymptomatic disease in lung and lymph nodes for the past several months with little growth, would you administer anti-PD-1 alone or combination of anti-PD-1 with anti-CTLA-4, or combined BRAF and MEK inhibitor therapy?
In this case, all three options are reasonable. Over the next few years, how to use these agents in sequence or in combination will hopefully be clarified. Without definitive information at this time, most experts would now recommend the combination of checkpoint inhibitors with very close monitoring of toxicities. If a patient progresses on the combination, they can be treated with one of the other agents.
Follow-up surveillance and therapy/management of recurrences.
There are few long-term studies evaluating the role of surveillance in melanoma patients that provide established evidence based guidelines. In all cases imaging should be considered to evaluate new signs and symptoms.
Stage IA, IB, IIA
There is no role for routine imaging.
History and physical exam every 3 months x 2 years, then every 6 months x 3 years, then annually after 5 years.
Role of serum LDH and Chest X-ray is unproven.
Stage IIB, IIC, IIIA
History and physical exam every 3 months x 2 years, then every 6 months x 3 years, then annually after 5 years.
Role of routine imaging is unproven.
Consideration to annual imaging with PET scan for the initial 3 years, but not recommended.
Chest X-ray and LDH every 6 months for initial 3 years given consideration.
Stage IIIB, IIIC
History and physical exam every 3 months x 2 years, then every 6 months x 3 years, then annually after 5 years.
Role of routine imaging is unproven.
However, consider imaging (PET scan or C/A/P CT scan) at 6 month intervals for the initial 3 years due to the very high risk.
Chest X-ray and LDH at 6 month intervals for years 4-5.
Stage IV (previously resected)
In most cases, consideration to imaging (PET scan or C/A/P CT scan) every 3 months for the initial 2 years and then every 6 months for a minimum of 3 additional years. After 5 years, consider annual imaging.
The type of imaging will be influenced by resected site of recurrence. For skin only disease this may be performed less frequently.
Treatment of isolated metastatic disease
Surgery is a treatment that should be considered in a number of metastatic disease settings. It can provide a palliative role for patients with:
1. Lymph node or soft tissue disease where tissue ulceration and necrosis creates a local problem for the patient.
2. Pain control of soft tissue or nodal disease.
3. Brain metastasis – for symptomatic lesions, rapid relief can be achieved by resection.
4. Either bleeding or obstruction from small bowel metastases.
5. Tumor with uncontrolled bleeding.
In some cases, surgery may provide more definitive therapy for patients with asymptomatic isolated metastases. Retrospective studies, adjuvant vaccine trials, and even prospective trials demonstrate 5 year survivals of 30-50% in patients with resected metastatic disease. The outcome is best when:
1. A single metastasis versus greater than 2 metastases.
2. Single organ versus multiple organ involvement.
3. Soft tissue or skin metastases versus visceral metastases.
4. Prolonged interval since previous disease recurrence.
In some of these patients who have a short disease-free duration when their metastasis occurs, consideration could be given to a short 3-6 month course of systemic therapy (clinical trial) prior to surgery. This approach may not only lead to disease regression but also provide some assurance that disease would not recur very soon after surgery.
What other clinical manifestations may help me to diagnose melanoma?
There are several essential components of the history and physical exam of a patient with melanoma. History must include a very thorough family history for cancer including melanoma, pancreatic cancer, breast cancer, especially male breast cancer (BRCA2) and history of other family members with many atypical nevi (large and asymmetrical).
Some effort should also be made to determine the hair color and skin sensitivity to sun and presence of freckles in other family members.
A critical component to the physical exam is a thorough and complete skin exam, with all under garments off, including an exam of the genitalia, rectum, and oral mucosa. Other melanomas may be found, and the presence of other suspicious nevi that require close follow-up can be identified.
Lastly, the hair color, eye color and presence of very light skin and many freckles should also be considered in terms of the risk of other melanomas and relationship of sun burns to development of melanoma.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- What every physician needs to know:
- Are you sure your patient has melanoma? What should you expect to find?
- Beware of other conditions that can mimic melanoma:
- Which individuals are most at risk for developing melanoma:
- What laboratory and imaging studies should you order to characterize this patient's tumor (ie, stage, grade, CT/MRI vs PET/CT, cellular and molecular markers, immunophenotyping, etc.) How should you interpret the results and use them to establish prognosis and plan initial therapy?
- What therapies should you initiate immediately i.e., emergently?
- What should the initial definitive therapy for melanoma be?
- What should you tell the patient and the family about prognosis?
- What if scenarios in the treatment of melanoma.
- Follow-up surveillance and therapy/management of recurrences.
- What other clinical manifestations may help me to diagnose melanoma?