Melanoma Indianapolis

Jeffrey D. Wagner, MD
Director, Cutaneous Oncology Program
Wagner and Associates Plastic And Reconstructive Surgery Consultants of Indiana

Current Therapy of Cutaneous Melanoma

Introduction

The worldwide incidence of cutaneous melanoma continues to rise. Melanoma rates are rising faster than any cancer in men and are second only to lung cancer in women. It is estimated that approximately 1 in 75 persons born in the United States in the year 2000 will develop melanoma during their lifetime; about 20% of these people will develop advanced disease and die within 5 years of diagnosis. Although an improvement in survival for early stage melanoma has been observed, the mortality rate from melanoma in the United States has remained stable over the past 9 years. By virtue of its metastatic potential, melanoma accounts for the vast majority of deaths from cutaneous malignancies.

Several recent clinical studies have helped to clarify the biology of the disease and have changed the patterns of care for patients with melanoma. Although melanoma remains primarily a surgical disease, trends have been toward less radical surgical treatment, fewer lymph node dissections, and a more aggressive diagnostic approach. The advent of lymphatic mapping of regional lymph nodes (sentinel node biopsy) and interferon as the first effective postsurgical adjuvant therapy have had a major impact on the care of melanoma in the United States and elsewhere.

Clinical Presentation and Diagnosis

Early detection and diagnosis is a critical factor which probably accounts for most of the increase in overall survival rates of melanoma noted over the past several decades. Clinical characteristics (markers) which identify the patient at higher risk for development of melanoma include: skin type I or II (Caucasian with light skin and sun sensitivity), presence of multiple atypical nevi (moles), large congenital nevi, family history of numerous atypical nevi or of melanoma, and a history of prior significant sun exposure, particularly blistering sunburns. Melanoma is predominantly a disease of Caucasians, and there is a slight male predominance. Primary lesions occur most commonly on the trunk in males, while most melanomas in females occur in the lower extremities.

The cell of origin of most cutaneous melanomas is the epidermal melanocyte. Although there is some disagreement, most melanomas are thought to arise from epidermal melanocytes without an obvious precursor nevus. However, many melanomas seem to arise from a pigmented precursor skin lesion, usually an acquired nevus. Early signs of melanoma include the ABCDE’s: Lesion Asymmetry or border irregularity, Bleeding or crusting, a history of recent Change, or variegation in Color (some lesions are nonpigmented), Diameter over 7 millimeters, and development of an Elevated area (or palpable nodule) in a previously flat nevus. About 1-2% of primary melanomas arise from melanocytes on mucous membranes. Approximately 5% of melanomas present with metastatic disease in the absence of an identifiable primary lesion. The most common presentation for unknown primary melanoma is a lymph node mass. A few patients present with visceral metastases in the absence of a known primary lesion.

Although several noninvasive diagnostic techniques for evaluation of skin lesions exist, biopsy is indicated for all suspicious pigmented lesions. Because tumor thickness is the most important histologic feature in determining prognosis and treatment, biopsy technique is critical. For most small and medium sized lesions the ideal biopsy technique is complete full thickness excision of the lesion with a 1-2 millimeter surrounding margin of normal skin. The orientation of the biopsy wound closure should take into consideration a possible subsequent wider excision to decrease the necessity for extensive reconstructive procedures. For example, extremity biopsy incisions should be oriented along the longitudinal axis of the extremity. Large excisions or reconstructive procedures should generally not be performed at the time of the initial biopsy. If melanoma is confirmed, these maneuvers may unnecessarily enlarge the extent of the wide local excision and may interfere with lymphatic mapping.

If the suspicious lesion is large enough or located where total excision would necessitate reconstructive surgery, an incisional biopsy is recommended. Full thickness removal of the most elevated, thickest, and/or most clinically suspect part of the lesion is the objective. Several areas may need to be sampled in large or morphologically variable lesions. Partial thickness or shave biopsies are not recommended for suspected melanoma because these techniques may prevent accurate determination of tumor thickness.

Accurate pathologic interpretation of the biopsy specimen is the cornerstone that determines treatment and prognosis for clinically localized melanoma. The pathology report should contain a description of the tumor site, Breslow’s thickness, Clark’s level of invasion, presence of ulceration, dimensions, and margins. In difficult cases, additional expert pathology consultation should be considered.

Several characteristic of primary melanoma tumors help predict the prognosis and risk of metastases. Tumor thickness (measured in millimeters, the so called Breslow depth) is the strongest predictive characteristic for recurrence in patients with primary cutaneous melanoma and the single most important determinant of patient management today. Tumor thickness is a continuous variable associated with increasingly poor prognosis, and forms the primary basis for tumor staging. Because thickness is so important, melanomas are commonly (and somewhat arbitrarily) referred to as thin (generally those less than 1.0 mm thick), intermediate depth (1.0-4.0 mm thick) and thick (greater than 4 mm thick).

Anatomic levels of invasion relative to the various dermal layers of the skin are known as Clark’s levels and also correlate with outcome. The Clark’s level is an independent prognostic factor for thin melanomas and for those that arise in thin skin such as the eyelid and ear. Lymphatic host inflammatory response, histogenetic melanoma subtype, and growth phase have also been correlated with prognosis. Similarly, the presence of angiolymphatic invasion, regression, microsatellitosis, neurotropism, and a high tumor mitotic rate have been reported to correlate with prognosis.

Clinical factors such as anatomic site, ulceration, and gender are also prognostic factors for survival. Extremity melanomas generally have a better prognosis than those on the head and neck or trunk. However, distal extremity lesions, (foot and hand) have prognoses similar to truncal primaries. Mucosal and mucocutaneous melanomas have an overall poor prognosis. Ulceration has been shown to be a strong independent prognostic factor in multiple studies and has recently been adopted into the AJCC staging system. This may be because tumor thickness may be underestimated in ulcerated lesions, lending to a worse prognosis than would be predicted by the measurable tumor thickness alone. Numerous studies have shown women fare somewhat better than men with melanoma. The reasons for gender-related differences in survival are unknown.

Staging

The most current standard national staging system recommended by the American Joint Commission on Cancer is shown below. The primary factors in assigning AJCC stage are tumor thickness, ulceration, nodal status, and distant metastases. Essentially, Stage I represents early local disease (thin melanomas without evidence of metastases), stage II is more advanced local disease (intermediate depth melanomas without metastases), stage III is comprised of regional lymph node and/or intransit metastases, and stage IV denotes the presence of distant metastatic disease. Survival correlates closely with stage groupings.

American Joint Committee On Cancer Staging System For Malignant Melanoma*

Tumor (T)

Tx		Primary tumor cannot be assessed
T0		No evidence of primary tumor
Tis		Melanoma in-situ
T1		Melanoma £ 1.0mm thick, with or without ulceration
	T1a	Melanoma £ 1.0mm thick no ulceration and level II/III
	T1b	Melanoma £ 1.0mm thick with ulceration or level IV/V
T2		Melanoma 1.01-2.0mm thick, with or without ulceration
	T2a	Melanoma 1.01-2.0mm thick no ulceration
	T2b	Melanoma 1.01-2.0mm thick no ulceration
T3		Melanoma 2.01-4.0mm thick, with or without ulceration
	T3a	Melanoma 2.01-4.0mm thick no ulceration
	T3b	Melanoma 2.01-4.0mm thick with ulceration
T4		Melanoma > 4.0mm thick, with or without ulceration
	T4a	Melanoma > 4.0mm thick no ulceration
	T4b	Melanoma > 4.0mm thick with ulceration

Regional Lymph Nodes (N)

NX		Regional lymph nodes cannot be assessed
N0		No regional lymph node metastasis
N1		Metastasis in one lymph node
	N1a	Clinically occult (microscopic disease) metastasis**
	N1b	Clinically apparent (macroscopic) metastasis***
N2		Metastasis in 2-3 regional nodes or intralymphatic regional metastasis, without nodal metastasis
	N2a	Clinically occult microscopic metastasis
	N2b	Clinically apparent macrosopic metastasis
	N2c	Satellite or in-transit metastasis, without nodal metastasis
N3		Metastasis in 4 or more regional nodes, or matted Metastatic nodes, or in-transit/satellite metastasis, with metastasis in regional nodes

Distant Metastasis (M)

MX		Distant metastasis cannot be assessed
M0		No distant metastasis
M1	M1a	Metastasis to skin, subcutaneous tissues, or distant lymph nodes
	M1b	Metastasis to lung
	M1c	Metastasis to all other visceral sites or distant metastasis at any site associated with an elevated serum lactic dehydrogenase (LDH)
N3		Metastasis in 4 or more regional nodes, or matted Metastatic nodes, or in-transit/satellite metastasis, with metastasis in regional nodes

Pathological Stage Grouping
	Stage 0	Tis	N0	M0
Stage IA	T1a	N0	M0
Stage IB	T1b	N0	M0
	T2b	N0	M0
Stage IIA	T2b	N0	M0
	T3a	N0	M0
Stage IIB	T3b	N0	M0
	T4a	N0	M0
Stage IIC	T4b	N0	M0
Stage IIIA	T1-4a	N1-2a	M0
IIIB	T1-4b	N1-2a	M0
	T1-4a	N1-2b	M0
	Any T	N2c	M0
IIIC	T1-4b	N1-2b	M0
	Any T	N3	M0
Stage IV	Any T	Any N	M1

*Adapted from Balch CM, Buzaid AC, Soong SJ, et al. Final Version of the American Joint Committee on Cancer Staging for Cutaneous Melanoma. J Clin Oncol 2001; 19:3635-3648

The initial clinical evaluation of a patient with melanoma consists of a history and physical examination. The history should focus on prior personal or family history of melanoma, and inquire about neurologic, musculoskeletal, gastrointestinal, pulmonary, and other abnormal symptoms. Examination should include inspection of the skin for suspicious pigmented lesion or dysplastic nevi and for second primary lesions, which may be present in 3-5% of patients. Palpation for satellite or intransit nodules, which may be subcutaneous and nonpigmented, is performed. Palpation of the major lymph node basins is done, with particular attention to the potential primary draining nodal basins. Enlarged lymph nodes should be considered potentially malignant. The diagnostic approach to enlarged lymph nodes in a patient with melanoma often includes fine needle biopsy. If negative or equivocal, open biopsy should be performed.

In addition to history and physical examination, the initial staging evaluation for patients with clinically localized cutaneous melanoma (no obvious or suggestive evidence of metastatic disease) usually includes a chest radiograph, and blood chemistries, such as serum alkaline phosphatase and lactate dehydrogenase. Abnormal findings on screening tests, physical examination or history should be further investigated with additional directed radiological examination(s), as indicated. In contrast, routine extensive laboratory and/or radiological staging workup of the asymptomatic patient with primary melanoma localized to the skin is low-yield and not generally recommended.

Treatment of Primary Lesion: Wide Excision

The surgeon plays a central role in the definitive management of melanoma. Wide local surgical excision is the mainstay of treatment in primary melanoma and should be offered to virtually all patients. Surgical treatment is aimed at removing all melanoma cells at the primary site in order to cure patients with low likelihood of occult metastasis and to achieve durable local disease control in high-risk patients who may subsequently suffer from non-local recurrences. Wide excision means removal of tumor and/or biopsy site with a predetermined (usually measured) margin of normal skin and subcutaneous tissue down to the level of muscle. In deep primary tumors or after deep surgical biopsies, the fascia or muscle is sometimes included to insure a tumor-free deep margin. The surgical margins should be examined for tumor cells to insure complete removal.

The extent of excision for invasive melanomas is an area of evolving controversy. Traditionally, very wide excision margins of up to 3-5 centimeters were routine. Five prospective, randomized studies have addressed excision margins for melanomas of varying tumor thicknesses and form the basis for current recommendations. Collectively, the melanoma literature does not demonstrate a survival disadvantage for patients undergoing narrower excisions (1-2 centimeters) compared to wider margins, but this may be due to inadequate power to detect small but important differences in survival. Furthermore, there are no convincing differences in local recurrence related to different excision margins. Despite this, there has been a reluctance to adopt narrower excision margins for higher risk melanomas because of perceived possible differences in outcome.

The currently recommended surgical margins for excision of melanomas are based primarily upon tumor thickness. In-situ or preinvasive melanomas must be completely excised and typical margins are 5 mm. Thin melanomas (<1 mm deep) are usually excised with 1 cm margins. Melanomas greater than 1 mm deep are usually excised with wider margins of 1-2 cm. Selected high-risk melanomas, such as those > 4 mm deep, those with satellite nodules, or neurotropic melanomas are associated with a higher risk of local recurrence and may be excised with wider margins of 2-3 cm or more.

Recommended surgical margins should not be compromised routinely. A reconstructive procedure such as a flap or graft is usually favored over possibly inadequate excision margins. Very narrow margins of excision, such as frozen-section guided margins or Moh’s micrographic surgical excision technique are controversial and have not been studied in prospective randomized trials. Very narrow margins have been considered contraindicated by many experts for invasive melanoma, but may have a role in the management of melanoma in-situ.

Management of Regional Lymph Nodes

The most common site of initial recurrence and metastatic disease in melanoma is the regional lymph nodes. Lymph node metastases, when present, are the single most important prognostic variable for survival in melanoma, taking priority over other tumor and host factors. For this reason, pathological nodal staging for all patients with significant-risk but apparently localized (AJCC stage I and II) melanomas is recommended to help determine optimal management. Several prognostic factors have been identified in patients with lymph node positive melanomas. The most important of these is the number of involved nodes. Microscopic versus macroscopic disease is a significant predictor of outcome. Primary tumor characteristics such as thickness, ulceration, and location retain some secondary prognostic significance even when lymph nodes are involved.

Surgical excision of cancerous regional lymph nodes is the therapy of choice and can be curative. Surgical excision of lymph nodes that are proven to contain tumor is traditionally referred to as a therapeutic lymphadenectomy. However, by the time regional nodal metastases are clinically obvious, 70-80% of patients also have obvious or occult distant metastases from which they will eventually suffer treatment failure. Lymphadenectomy should also be considered for select patients with bulky lymph node metastases who also have distant metastases because the potential morbidity from uncontrolled locoregional disease has no satisfactory alternative treatment.

Theoretically, excision of occult melanoma nodal metastases at an early stage with a lower tumor burden could prevent progression of disease to distant sites and improve survival. For decades, major controversy has raged over whether early elective or prophylactic lymphadenectomy in patients with clinically normal lymph nodes offers any therapeutic advantage over observation with delayed lymphadenectomy in patients later develop clinically obvious nodal metastases. The selection criteria for prophylactic lymphadenectomy have traditionally been based on tumor thickness and ulceration.

Four prospective randomized trials of lymphadenectomy in the management of localized cutaneous melanoma have been reported. The collective melanoma literature does not demonstrate a convincing survival advantage for patients treated with prophylactic lymphadenectomy, but subsets of patients may benefit. Despite these four trials, the lymphadenectomy controversy has not been resolved. In the absence of conclusive data the decision to perform a prophylactic lymphadenectomy must be made on a case-by-case basis. The modern advent of sentinel node biopsy (see below) has largely eclipsed the relevance and practice of prophylactic lymphadenectomy today.

Sentinel Lymph Node Biopsy

Morton et al introduced the concept of sentinel lymph node biopsy (SNB) in 1992. Studies in animals and humans have demonstrated there are well-defined pathways leading from each cutaneous territory to specific regional node, the so-called sentinel nodes. The sentinel node has been shown to be a sensitive indicator of metastatic melanoma in regional nodes. Importantly, absence of metastatic melanoma in sentinel nodes indicates absence of disease in the remaining nodal basin. Therefore, preliminary nodal staging can be achieved with SNB. SNB provides staging information essentially equivalent to that obtained by complete lymphadenectomy but with less morbidity. With a sensitivity of over 95% in experienced hands, SNB can virtually eliminate unnecessary lymph node dissections. Because of its high accuracy and low morbidity, SNB has essentially replaced prophylactic lymphadenectomy as the standard approach for staging regional lymph nodes in the US.

Initially, SNB using blue dye alone was successful in only about 80% of patients undergoing the procedure. In experienced hands, success rates now approach 100% due to improvements in preoperative lymphoscintigraphy and intraoperative mapping techniques, including use of the gamma probe. The independent evolution of SNB at various centers supports the use of both visual blue dye and radiocolloid for SN localization. The two techniques are complimentary, and most experienced centers use both.

In the selection of patients for nodal staging with SNB, tumor thickness is the primary criteria. The sentinel node positivity rate in various reports ranges between 12-36%. A positive correlation between sentinel node positivity with tumor thickness has generally been observed, with SNB positivity rates less than 5-10% for AJCC T1 melanomas and 40-50% for AJCC T4 lesions. The ideal selection criteria for SNB are not known. Regional practices vary and are somewhat arbitrary. Although early practice frequently limited SNB to patients with melanomas deeper than 1 mm, there has been a recent trend to expand the indications for SNB to selected patients with thin, lower risk melanomas. The search for the optimal selection criteria for SNB in thin melanomas continues. Typically, it is recommended that patients with melanomas greater than 1 mm deep, those with ulcerated tumors, deep Clark’s levels, and high mitotic index should be considered for SNB. The recent trend is to offer SNB to patients with thin lesions thought to be at higher risk for recurrence than would be predicted by tumor thickness alone on a case-by-case basis.

Lymphadenectomy performed after a positive SNB is termed selective lymphadenectomy (SLND). The positivity rate for nonsentinel nodes removed by SLND is 7% to 35%. These numbers probably underestimate the true SLND positivity rate. The relatively low incidence of positive nodes in SLND specimens has led to speculation that SLND may be unnecessary after a positive SNB for some patients, such as those with thin melanomas. This is an area of ongoing debate and controversy. At this time, however, most expert melanoma surgeons recommend SLND as the most appropriate therapy after a positive SNB. The question of subsequent therapy after a positive SNB is the subject of ongoing prospective clinical trials.

Technical and therapeutic controversies notwithstanding, the status of the sentinel node is clinically relevant for at least 3 major reasons. First, the SN status has been shown to be an important independent prognostic factor, with a positive result predictive of high risk for treatment failure. The predictive value of a negative SNB is worthwhile in that it identifies patients with significantly lower risk of recurrence than would be predicted by tumor characteristics alone. Second, SNB identifies with minimal morbidity high-risk patients who may benefit from additional therapy such as SLND or adjuvant interferon, or who may be eligible for participation in clinical trials. SNB allows the surgeon to perform SLND based on histological confirmation, essentially eliminating the problem of false negative lymphadenectomy. Lastly, the psychological benefit for the patient whose SNB result does not reveal metastases seems to be quite significant. The possible therapeutic benefits of the SNB procedure itself and SNB-based therapeutic decisions (including SNLD) are currently unknown but are the subject of ongoing clinical trials.

Adjuvant Therapy

In 1996, Kirkwood, et al reported the results of ECOG 1684 trial comparing a one year course of high dose adjuvant interferon alfa-2b (IFN) to observation in a randomized group of 287 patients rendered without evidence of disease by surgery. The study population was comprised of patients with melanomas > 4.0 mm thickness and patients with surgically resected regional nodal metastases. There were statistically significant improved 5-year relapse-free survival rates (37% versus 26%) and overall survival rates (46% versus 37%) in the high-dose IFN group compared to untreated patients. This study was the primary basis for Food and Drug Administration approval of interferon alfa-2b for adjuvant therapy of resected high-risk melanoma. Several other interferon clinical trials have been completed testing interferon regimens for adjuvant therapy in patients with melanoma and have yielded conflicting results. Interferon treatment is costly, lengthy (typically one year) and toxic, with a wide variety of side effects. To date, however, interferon is the only adjuvant therapy that has ever been shown in a randomized clinical trial to have a survival benefit in melanoma patients.

At this time, the role of IFN in the treatment of melanoma is not clear, and there does not seem to be a consensus among experts. While a prolongation of time to recurrence has been consistently observed in several trials, prolongation of survival is not certain. The cumulative observations from several trials contain no direct evidence for efficacy of any IFN regimen other than the high dose regimen used in ECOG trial 1684, and the strongest evidence for efficacy is in node positive patients. The clinical importance of accurate nodal staging and the attractiveness of SNB are highlighted by this controversy. Furthermore, because intransit metastases, satellite nodules, and local recurrences all represent various forms of lymphatic metastases with a similar prognosis to nodal metastases, it has been argued these patients should be offered adjuvant IFN therapy. Ongoing clinical trials testing IFN and other forms of immune modulation should clarify the appropriate treatment groups, and refine the optimal agents, dosage, and duration of this adjuvant therapy.

Management of Recurrent and Metastatic Disease

Although most patients with melanoma will be cured, approximately 25% of patients presenting with melanoma apparently localized to the skin will develop a recurrence after initial surgery. Among patients who experience recurrent disease, 55%-67% appear within the first 2 years and up to 70% by 3 years following treatment of the primary tumor. Thick tumors and ulcerated lesions tend to recur earlier and thin melanomas at later post-treatment intervals. Patterns of recurrence are similar for early and late recurrences. Recurrent melanoma may rarely present 10 years or later after apparently successful treatment of the primary lesion.

Regional Lymph Nodes

Nodal recurrence is the most common site of initial recurrence in patients treated without prior LND, and is noted as the site of first relapse in 70-80% of such patients. In contrast, when patients are treated with a prior lymph node dissection, nodal recurrences are less common, and visceral metastases are more often noted as the initial site(s) of treatment failure. Therapy of clinically obvious nodal recurrences is complete lymph node dissection, which is curative for 25-40 % of patients, depending on the volume of nodal disease. Even in the presence of distant metastases, lymphadenectomy should be considered for selected patients because this potentially morbid condition has no satisfactory alternative treatment. It is important to excise all tumor, and radical procedures are sometimes justified.

Recurrence in the lymph node basin after lymph node dissection, particularly in the head and neck region, has been reported as high as 50%. This indicates surgery alone frequently fails to control metastatic disease in lymph nodes, particularly in head and neck melanomas. Because secondary operation for local relapse is difficult and frequently morbid, measures to improve locoregional control in selected patients are often considered. Nonrandomized studies suggest the use of adjuvant radiotherapy may improve local control after lymphadenectomy in the head and neck. To date, the role and optimal protocol for use of postoperative adjuvant radiotherapy have not been established by randomized clinical trials.

As previously noted, patients with nodal metastases are at high risk for treatment failure due to distant metastases. Therefore, patients with regional nodal recurrence should be considered for adjuvant IFN therapy after surgical therapy.

Local Recurrence

Local recurrence is variously defined as recurrence within 1-2 cm of an excision scar. Local recurrence is not common, occurring in only 1-3% of patients who have had a standard wide excision. Primary tumor characteristics such as increasing tumor depth, ulceration and microsatellitosis are associated with a higher rate of local recurrences. Local recurrence is strongly associated with the subsequent appearance of intransit, nodal and distant metastases and portends a grave prognosis. However, patients with local recurrence related to true regrowth of a tumor not widely excised (i.e., after a after missed diagnosis) seem to have a somewhat better prognosis. The management of local recurrence with generous surgical resection (1-3 cm) is the simplest, most common, and most locally effective form of therapy.

In transit Recurrence

Intransit metastases appear in 2-38% of patients surgically treated for primary melanoma. In transit disease is defined as dermal or soft tissue nodules located at least 2 cm from the primary lesion. Local recurrences after prior wide excision of at least 2 cm can also be considered intransit metastases. Factors predisposing to this pattern of recurrence include increasing thickness of primary lesion, ulceration, lymph node metastases and locally recurrent disease. In transit disease is viewed as a continuum of intralymphatic disease and carries a similar prognosis to regional nodal metastases. Management of solitary intransit metastases is usually with generous wide excision. Achieving control of in transit disease is often difficult, and additional recurrences can be expected in the majority of patients.

When multiple or recurrent intransit metastases appear in an extremity a potentially morbid and sometimes limb-threatening situation may exist. Isolation limb perfusion (ILP) should be considered in these scenarios. While prospective studies have not shown adjuvant to be of benefit in primary melanoma, ILP seems to have some palliative benefit in the therapy of selected patients with recurrent disease localized to an extremity. A variety of agents and protocols have been employed, but melphalan perfused with mild hyperthermia has been the most commonly used.. Overall response rates of 60-100% have been reported with melphalan ILP, with about half of these responses being complete and about half of complete responses being durable. In selected patients, a successful ILP helps maintain control of disease, preserving a functional extremity. Substantial potential for toxicity exists with ILP, but morbidity in large series from experienced centers is acceptable. With the advent of ILP the indications for major limb amputation for melanoma are rare.

Distant Metastases

Remote soft tissue or nodal metastases are seen as the initial site of relapse in about 2-16% of patients. Visceral metastases are seen in about 20% of patients as the initial site of recurrence. Although visceral metastases may be seen in any organ, the most common sites are lung, liver, brain, bone, and gastrointestinal tract. Most patients with visceral or soft tissue metastases are symptomatic or have physical exam findings. The exception to this general rule is lung metastases, which are usually asymptomatic, and are frequently found by x-rays.

Stage IV melanoma is generally considered incurable by currently available therapies. Patients with skin, subcutaneous, nonregional lymph node and pulmonary metastases fare somewhat better than patients with bone, brain, and liver metastases. The most realistic goal for the physician treating stage IV melanoma patients is palliation and preservation of quality of life. Experienced physician judgment and careful selection of patients for therapy are important to prevent excessive morbidity while maximizing quality of remaining life.

A minority of patients with stage IV melanoma has what appears to be limited disease and can be deemed candidates for potentially curative surgical resection. In most cases surgical resection with curative intent is performed for solitary lesions of the skin and distant nodal sites. Resection of isolated pulmonary metastases can also result in long disease-free survival in carefully selected patients. Resection of brain and gastrointestinal metastases is usually performed with palliative intent. Randomized trials evaluating the impact of surgical resection in patients with stage IV melanoma patients have not been reported.

Systemic therapies for stage IV melanoma are generally unsatisfactory. Melanoma is a relatively chemotherapy-resistant disease. Responses to cytotoxic agents are relatively infrequent, incomplete, and not typically durable. Single agent response rates are generally about 10 to 20%, with the most active single agent being dacarabazine (DTIC). Randomized trials have not demonstrated a convincing improvement in overall survival for single or multiple agent chemotherapy in stage IV melanoma. Early detection and therapy of distant metastatic disease has not been clearly demonstrated to improve patient outcomes. Because of toxicity, poor response rates and lack of survival benefit, chemotherapy for melanoma is usually considered palliative. At present, no consensus seems to exist on the utility of chemotherapy for stage IV melanoma.

Interleukin-2 (IL-2) is a T-cell growth factor with antitumor and immunomodulatory activity. High dose IL-2 produces overall responses similar to chemotherapy, with complete responses in a minority of patients. IL-2 has been approved for use in stage IV melanoma and is most often employed as a first or second-line agent for treatment for stage IV disease.

Radiation Therapy

Melanoma has historically been considered a relatively radioresistant tumor, and radiation therapy is not generally used with curative intent. However, radiation can be useful for palliation in selected patients with metastatic disease. Indications for radiotherapy in selected stage IV melanoma patients include impending spinal cord compression (often in combination with surgery), bone metastases, unresectable symptomatic visceral metastases, and treatment of extensive cutaneous metastases not amenable to surgical resection or isolation limb perfusion. Radiosurgical treatment of brain metastases with the gamma knife is commonly employed for local control of brain metastases.

Investigational Therapies

Because standard therapies for stage IV melanoma are ineffective, experimental therapies are often considered in the first-line therapeutic approach. A variety of novel approaches for treatment of metastatic melanoma are currently in clinical trials. Combinations of cytotoxic chemotherapy and biological therapy have recently been studied. Biochemotherapy regimens demonstrate increased toxicity, but may have enhanced antitumor activity. A number of other biological response modifiers with potential implications for melanoma patients, such as cytokines, interleukins and antiangiogenic agents, are in various phases of clinical trials. A variety of vaccine therapy approaches continue to generate heightened interest as potential therapy, although to date, none have shown a survival benefit in a randomized trial. These agents carry substantially less toxicity, suggesting potential utility as therapy for both measurable disease and as postresection adjuvant therapy. Investigational agents are typically only available to patients who participate in clinical research trials. These clinical trials are important to define the role of these and other investigational approaches for the treatment of metastatic melanoma.

Follow-Up and Surveillance

The basic assumptions in follow-up of any cancer patient are that early detection of recurrence will lead to treatment, which will ultimately impact favorably on outcome. It is difficult to define relationship between intensity of follow-up for melanoma and outcome. There is currently no consensus on the frequency of follow-up or surveillance testing in patients with melanoma. It has become common to arrange follow-up schedules based on the risk, timing, and patterns of relapse, as well as the treatment options available if a recurrence is detected. For example, thick melanomas or patients with nodal metastases are commonly followed at more frequent intervals than patients with low risk melanomas. It is reasonable to offer more intensive follow-up to patients where more options are available for treatment if recurrence is detected (i.e., academic centers specializing in the multimodality management of advanced disease). Less rigorous strategies are probably appropriate in the setting of a local community, and for patients not interested in pursuing investigational therapies.

The goals of follow-up are twofold: detection of second primary lesions (seen in 3-5% of primary melanomas), and detection of recurrent disease. In the setting of careful surveillance, most second primary melanomas that are discovered are thin and have a good prognosis. A number of reviews have appeared in the literature describing current practices in the follow-up of patients after surgical treatment for melanoma. Most treatable recurrences are usually noted by the patient and are detectable by physical exam or are hinted at by a careful history. The exception is pulmonary metastases, which are usually found by plain chest radiographs. Routine surveillance radiography has not been shown to improve survival. Likewise, routine laboratory testing is of limited clinical utility in finding asymptomatic visceral metastases. Because most recurrences appear within the first 2-3 years, follow-up visits should be more frequent during this time. The optimal duration of follow-up is unknown. Late recurrences more than 10 years after initial treatment are well recognized and the risk of second primary melanomas is lifelong.

Summary

The melanoma epidemic continues to pose a serious public health problem in the United States, and it is expected to continue to increase. Care of the melanoma patient has evolved rapidly over the last decade, fueled by cooperative clinical trials. Surgery remains the mainstay of therapy for most melanoma patients. Adjuvant therapy is now available for the postsurgical patient at high risk for relapse. For these reasons, a multidisciplinary approach represents the optimal means of patient care delivery for melanoma today. Dissemination of new knowledge from each specialty in a coordinated and collaborative manner is necessary to ensure optimal patient care in this changing field. Enrollment of patients with melanoma into prospective randomized trials is important to advance our knowledge and care of the melanoma patient into the next century.