Melanoma and Cutaneous Malignancies

Epidemiology

Even though melanoma accounts for less than 2% of skin cancer cases, it is currently the fifth most common cancer in men and the sixth most common cancer among women in the United States. Melanoma also causes the majority of skin cancer–related deaths. The American Cancer Society estimated that there would be 91,270 new cases of melanoma diagnosed in the United States in 2018, with 9320 deaths. The incidence of melanoma has steadily increased in the United States and worldwide over the last four decades ( Fig. 31.1 ). Worldwide, the highest incidence rates of melanoma are in Australia, New Zealand, North America, and northern Europe. The steady rise in incidence has been attributed to lifestyle changes leading to increased sun exposure as well as to improved surveillance and detection of early lesions.

The degree of pigmentation in the skin is a relative protective factor against cutaneous melanoma; those with lighter skin tones are at increased risk. As a result, cutaneous melanoma is predominantly a disease of whites. In particular, patients with a fair complexion, blonde or red hair, and blue eyes are at increased risk, as are those who sunburn easily, have a tendency to develop freckles, or have an inability to tan. In the United States, the average annual age-adjusted melanoma incidence per 100,000 persons is 32.2 for non-Hispanic white men and 19.4 for non-Hispanic white women, compared with 4.8 for Hispanic men and 4.6 for Hispanic women, 4.2 for Native American men (including Alaska Natives) and 4.6 for Native American women, 1.6 for Asian/Pacific Islander men and 1.1 for Asian/Pacific Islander women, and 1.1 for Black men and 1.0 for Black women. As one can see from these numbers, melanoma is slightly more common in men across most races/ethnicities.

Melanoma is predominantly seen in middle-age adults, although it occurs across all ages ( Fig. 31.2 ). Importantly, approximately one in six new cases of melanoma occurs in patients younger than 45 years old; accordingly, melanoma-related deaths can result in a high number of potential years of life lost. Genetic risk factors for melanoma include high-risk skin types (Fitzpatrick types I and II), family history of melanoma, and xeroderma pigmentosum. Patients with a prior history of melanoma or other skin cancers, as well as those with a large number of melanocytic nevi, dysplastic nevi, or giant congenital nevi, are also at increased risk. Environmental risk factors include episodes of intense intermittent sun exposure associated with severe blistering sunburns, immunosuppression, and upper socioeconomic status.

There is a clear association between ultraviolet (UV) radiation exposure and the development of melanoma. Intermittent, intense UV radiation exposure appears to be a strong causative factor for melanoma, as opposed to chronic sun exposure and the associated risk of nonmelanoma skin cancers (NMSCs). UV light can be classified as either UVA or UVB; UVA has a longer wavelength and penetrates more deeply into the skin than UVB. Although UVA radiation has long been known to play a major role in skin aging and wrinkling, growing evidence has implicated UVA radiation as a cause of melanoma skin cancer and NMSC alike. UVA is the predominant wavelength in tanning beds. Emerging evidence has demonstrated an epidemiological link between tanning bed use (even without sunburn) and melanoma. Another major risk factor for melanoma is UVB exposure from natural sunlight, especially among those with fair skin. UVB damages the skin’s more superficial epidermal layers and is the chief cause of sunburns. A direct link between UVB exposure and specific melanoma oncogenes has been demonstrated. A UV signature consistent with mutations caused by UV damage was found in over 90% of three of the four melanoma subtypes identified in The Cancer Genome Atlas project.

From a public health perspective, melanoma is a cancer for which there are clear primary prevention strategies that can prevent its development and reduce preventable deaths. Recommendations for reducing the risk of melanoma include avoidance of sunbathing and tanning beds, use of sun-protective clothing, and use of sunscreens. Population-based case-control studies and some randomized clinical trials have suggested that regular use of sunscreen can reduce the development of melanoma. From a policy standpoint, one strategy that has been used on a state-by-state basis in the United States is the ban of indoor tanning bed use by minors (age <18 years). Some countries, including Brazil and Australia, have banned commercial indoor tanning salons outright.

Precursor Lesions

Melanomas frequently arise de novo in otherwise normal skin; however, up to 40% may arise within preexisting lesions, including dysplastic nevi, congenital nevi, and Spitz nevi. Upward of 5% to 10% of melanoma patients will have a family history of the disease. Variously termed dysplastic nevus syndrome, familial atypical multiple mole–melanoma syndrome, and B-K mole syndrome, these syndromes include patients with melanoma in one or more first- or second-degree relatives and large numbers of melanocytic nevi (often >100). These nevi will often be classified as atypical or dysplastic on close clinical and histologic examination. Other cancers may also be present in the family history, particularly pancreatic cancer. These patients require detailed dermatologic evaluation several times annually, with periodic biopsies of the most suspicious lesions.

In general, a dysplastic nevus is a 6- to 15-mm macular (flat) pigmented skin lesion with indistinct margins and variable color. The clinical distinction between a nevus with dysplasia and a nevus without dysplasia is often difficult; therefore, these lesions require careful monitoring over time to evaluate for progression. Most nevi are benign, but some may reflect early atypia associated with increased intracellular growth signaling and can progress to invasive disease with the accumulation of additional mutations. Although most dysplastic nevi do not progress to melanoma, suspicious lesions require biopsy. Dysplastic nevi are typically described as having mild, moderate, or severe dysplasia on histologic examination. Nevi with mild dysplasia usually do not require excision with negative margins but should be closely observed over time. It has been common practice to excise moderately or severely dysplastic nevi with negative margins, although wide local excision (WLE) is unnecessary. However, there is controversy about whether moderately dysplastic nevi require negative-margin excision. The risk of melanoma in those patients with congenital nevi is proportional to the size and number of nevi. Small- or medium-sized congenital nevi represent a low risk and therefore are not observed unless they change in appearance. Giant congenital nevi (>20 cm in diameter) are rare and are estimated to occur in anywhere from 1 in 20,000 to 1 in 500,000 newborns, but they carry an increased lifetime risk for the development of melanoma ( Fig. 31.3 ). These patients are also at increased risk for other tumors, particularly sarcomas. Complete excision should be considered when possible. At a minimum, these patients should undergo regular dermatologic evaluation.

Spitzoid melanocytic lesions constitute a wide range of histopathology from the typical benign Spitz nevus to spitzoid melanoma. A Spitz nevus is a rapidly growing, pink or brown, benign skin lesion with little or no risk for further progression to melanoma. Whereas benign Spitz nevi are most common in children, lesions in adults are more likely to have atypical features or to represent melanoma with spitzoid features. Atypical features include size larger than 10 mm, asymmetry, ulceration, and poor circumscription; these lesions can be difficult to distinguish histologically from melanoma. Consultation with an expert dermatopathologist is recommended; however, even the most experienced dermatopathologists may have difficulty in determining the malignant potential of spitzoid tumors. Although complete excision with negative margins is adequate for an unequivocal Spitz nevus, the diagnosis is often unclear. If there is any concern that the lesion may be melanoma, WLE with margins appropriate for melanoma is performed. Sentinel lymph node (SLN) biopsy is appropriate for invasive spitzoid melanomas and can be used as a prognostic measure in indeterminate cases. The routine use of SLN biopsy for atypical Spitz tumors is controversial because atypical cells are often seen in SLNs that may not have any prognostic significance.

Pathogenesis

In general, melanoma has the highest mutational burden of any malignancy that has been studied in humans. Our understanding of the molecular mechanisms that underlie melanoma progression has increased greatly over the last decade through collaborative efforts such as The Cancer Genome Atlas Network. In a landmark study in 2015, the collaborative published a study in which 331 cutaneous melanoma tumors underwent an integrative genomic analysis. Using DNA, RNA, and protein-based analyses, the group identified four distinct subtypes of cutaneous melanoma based on the most prevalent significantly altered genes: mutant BRAF , mutant RAS , mutant NF1 , and triple-wild-type.

The BRAF subtype is the most common genomic subtype in cutaneous melanoma, comprising of approximately one-half of all cutaneous melanoma. The most common mutation is the V600E, followed by the V600K mutation. BRAF subtype patients are relatively younger compared to the other subtypes. The second most common subtype was the RAS subtype, which described about 30% of cutaneous melanoma. The third most frequent subtype was the NF1 subtype, found in approximately 15% of samples; these patients were typically older than the other subtypes. Finally, the triple-wild-type was a heterogeneous subtype lacking any of the three major mutations previously mentioned. This subtype had the lowest rate of UV signature changes (30%). The prognostic and therapeutic implications of these genomic subtypes are the subject of intense ongoing research as we attempt to further target our therapies for melanoma in the age of personalized medicine.

The canonical pathway through which nearly all of the genomic subtypes exert melanogenic effect is the mitogen-activated protein kinase (MAPK) signaling pathway ( Fig. 31.4 ). Normally, signals generated by extracellular receptors initiate a signaling cascade that passes down the MAPK pathway to modulate gene expression in the nucleus. Gain-of-function mutations affecting any of the constituent steps along this cascade ( BRAF and RAS subtypes) or loss of inhibitory steps ( NF1 ) can result in unchecked cellular growth.

Gain-of-function mutations alone are likely not enough to generate melanoma; BRAF mutations are observed at a similar frequency in some benign nevi and malignant disease. Additional loss of key tumor suppressor genes is necessary for further neoplastic development. For instance, an inactivating mutation in the gene CDKN2A occurs in 25% to 40% of familial melanomas. A well-characterized gene with key roles in cell cycle control, CDKN2A codes for two separate tumor suppressor proteins, INK4A (p16 ^INK4A ) and ARF (p14 ^ARF ). In the event of DNA damage or activated oncogenes, INK4A prevents the cyclin-dependent kinase 4 from stimulating the cell to progress through the cell cycle. Through the regulation of p53 levels, ARF also acts as a tumor suppressor in the face of DNA damage or amplified growth signals. ARF prevents degradation of p53, allowing this key regulatory protein to accumulate and either to arrest the cell cycle or to initiate apoptosis. Loss of either ARF or INK4A therefore removes a checkpoint in the cell cycle, increasing the risk of uncontrolled replication.

The second major canonical pathway in melanoma pathogenesis is the PI3/AKT pathway ( Fig. 31.4 ). The PI3/AKT pathway is potentially altered in all four genomic subtypes. Phosphatidylinositol 3,4,5-trisphosphate (PI3) stimulates AKT to increase cellular proliferation through the mammalian target of rapamycin signaling molecule. Phosphatase and tensin homologue (PTEN), whose gene is located on chromosome 10, inhibits PI3/AKT signaling. Loss of PTEN occurs in approximately 25% to 50% of nonfamilial melanomas and is a common culprit pathway for the development of resistance to therapies targeting BRAF/MEK inhibition.

Initial Evaluation

Melanoma commonly presents as an irregular pigmented skin lesion that has grown or changed over time. The ABCDE criteria are used to guide diagnosis and the decision to perform a biopsy of suspicious cutaneous lesions ( Box 31.1 ). The first and most important step in the evaluation of a patient diagnosed with melanoma is a thorough history and physical examination. The history should elicit factors related to the primary melanoma, including duration, change over time, and symptoms such as itching and bleeding. Other risk factors, such as sun exposure, tanning bed use, immunosuppression, prior history of cancer, and family history should be sought. A detailed physical examination should specifically include a complete skin examination with inspection and palpation of the skin to detect any other suspicious skin lesions, including in-transit disease. Palpation of the cervical, axillary, and inguinal lymph nodes should always be performed with palpation of the epitrochlear or popliteal nodes as appropriate for distal upper or lower extremity melanomas. Although it is widely recognized that a skin examination should be part of the routine physical examination by primary care physicians and others, it is rarely adequately performed. A full skin examination requires only that the patient undress, and it may take only 1 minute to perform a complete survey. Many lives have been saved by early detection of melanomas by physicians who took the time to evaluate the skin.

Most melanomas occur de novo but some can arise within a congenital or acquired nevus. Even for experienced clinicians, distinguishing between a benign nevus and an early melanoma can be difficult. Benign pigmented lesions are so prevalent that it is challenging to detect an early melanoma among many benign lesions. The most common benign pigmented skin lesions are seborrheic keratoses ( Fig. 31.5 ). Known for their propensity to accumulate over time in elderly patients, typically these are scaly, waxy, raised lesions with a stuck-on look that makes them appear as if they could easily be scraped off with a fingernail. The characteristic appearance usually is completely diagnostic and these lesions do not need to be removed. However, even the most experienced dermatologists have been fooled by what appeared to be an irritated seborrheic keratosis that turned out to be melanoma.

Additional atypical presentations include amelanotic melanoma; these lesions are not pigmented and present as a raised pink or flesh-colored skin lesion. A high index of clinical suspicion is needed, and particular attention should be paid to any history of change in a lesion. If a patient presents with a skin lesion that has changed in size, color, or shape and/or is itching or bleeding, there should be a low threshold for biopsy. Telling a patient that “we should keep an eye on it” usually means that it will be ignored.

Given the increased awareness of this disease, it is uncommon for patients to have advanced regional or distant metastatic disease at the time of initial melanoma diagnosis. Nonetheless, roughly 10% of patients will present with regional disease, while up to 5% may present with metastases. Regional disease refers to the lymphatic spread of tumor to the regional nodal basin, which are the lymph nodes that receive the first drainage from the site of the primary tumor. In-transit melanoma is a form of regional lymphatic metastasis in which the tumor spreads within the draining lymphatic channels and becomes evident as cutaneous or subcutaneous nodules between the site of the primary tumor and regional lymph nodes. Distant metastasis refers to the hematogenous spread of melanoma to distant organs. Although uncommon at the time of initial diagnosis, it is important to elicit symptoms of metastatic disease, such as any masses, neurologic symptoms or headaches, anorexia, weight loss, bone pain, or respiratory symptoms.

Pathology

The biopsy report is the most important piece of information needed by the surgeon to evaluate a new skin lesion and to develop a treatment plan. Given the consequences of a missed diagnosis of melanoma, pathologists often have a low threshold to classify equivocal lesions as melanoma. It is now common for a pathology report to contain a long description essentially stating that the lesion may be anything from a severely dysplastic nevus to melanoma in situ to early invasive melanoma. In such cases, the prudent decision is to treat such lesions as an early invasive melanoma with a 1-cm margin WLE. Although melanoma in situ does not invade beyond the basement membrane into blood vessels and lymphatics, it may be considered a premalignant lesion given that there remains a significant likelihood of progression to invasive melanoma. For this reason, at least 5-mm margins are recommended for these lesions (see “Wide Local Excision”).

Histology

Histologically, invasive cutaneous melanoma is divided into four major types based on growth pattern and location. These include lentigo maligna melanoma, superficial spreading melanoma, acral lentiginous melanoma, and nodular melanoma. All melanomas initially proliferate in the basal layer of the skin. As they multiply, these cells expand radially in the epidermis and superficial dermal layer, termed the radial growth phase. With time, growth begins in a vertical direction and the skin lesion may become palpable, known as the vertical growth phase. The vertical growth phase allows invasion into the deeper layers of the skin, where the tumor may ultimately achieve metastatic potential by invasion of blood vessels and lymphatic channels. While the histologic subtype is not a major factor in prognosis, some histologic subtypes progress to the vertical growth phase earlier in tumor development and are therefore more likely to present at an advanced stage.

The most common histologic type is superficial spreading melanoma ( Fig. 31.6 ). It is not necessarily associated with sun-exposed skin and most commonly appears on the trunk and proximal extremities. As the name suggests, superficial spreading melanoma initially appears as a flat-pigmented lesion growing radially. These lesions are often asymmetric with irregular borders and can display a wide variety of pigments. Untreated, these melanomas will subsequently develop a vertical growth phase, invade more deeply into the skin, and possibly ulcerate.

Lentigo maligna melanoma occurs most commonly on the sun-exposed areas of older individuals and presents as a flat, dark, variably pigmented lesion, with irregular borders and a history of slow development ( Fig. 31.7 ). Lentigo maligna melanomas may become relatively large prior to diagnosis, as the slow progression can escape the patient’s notice. The prognosis of lentigo maligna melanoma is better than for the other subtypes given the superficial nature of these tumors. Nonetheless, these lesions can pose challenging management problems because of their propensity to develop in cosmetically challenging areas such as the face. The histologic extent of the lesion may extend well beyond the clinically apparent borders of the pigmented lesion, hampering efforts to achieve negative margins. Before proceeding with complex tissue flaps for closure, it is prudent to ensure negative margins. This may necessitate delaying the closure until the final pathology report indicates negative margins of excision.

Acral lentiginous melanoma is classified by its anatomic site of origin. These tumors develop in the subungual areas beneath fingernails and toenails and on the palms of the hands and soles of the feet ( Fig. 31.8 ). This is the most common type of melanoma in nonwhite patients. The histologic appearance of acral lentiginous melanomas is similar to melanomas arising on the mucous membranes. The diagnosis is often made at an advanced stage, which accounts for the general poor prognosis of these tumors. Subungual acral lentiginous melanomas are often mistaken for subungual hematomas, leading to a delay in diagnosis. The distinguishing feature of subungual melanomas is that they do not change in position underneath the nail, while a hematoma should migrate distally with growth of the nail. Biopsy of subungual melanomas can be accomplished by performing a digital block with local anesthesia and removing the nail or by performing a punch biopsy through the nail itself.

Nodular melanomas are raised papular lesions that can occur anywhere on the body and tend to develop a vertical growth pattern early in their course ( Fig. 31.9 ). These melanomas can have atypical presentations that do not always conform to ABCDE criteria, including a higher rate of amelanotic lesions compared to the other subtypes. Nodular melanomas often have a poor prognosis because of greater average tumor thickness and frequent ulceration at initial presentation.

A fifth type of melanoma that has a distinct histologic character is desmoplastic melanoma. These lesions are characterized by a combination of melanoma cells with a prominent stromal fibrosis. The diagnosis can be challenging and the presentation delayed because these lesions are often amelanotic. Desmoplastic melanomas are classified as either pure or mixed depending upon the degree of desmoplasia present. Desmoplastic melanomas have a greater propensity for local recurrence and often exhibit neurotropism. Pure desmoplastic melanomas have a low risk of lymph node metastases and some have advocated forgoing a SLN biopsy in these lesions. Mixed desmoplastic melanomas have lymph node metastatic rates that are similar to other histologic subtypes; therefore, SLN biopsy should be considered for the usual indications in these lesions.

Depth Assessment

Dr. Wallace Clark first described a classification system for melanoma that correlated with survival in 1969. Known as Clark level of invasion, this scheme was based on the extent of invasion into the anatomic layers of the skin. Shortly after Clark introduced his levels of invasion, Dr. Alexander Breslow described a simpler system based on a measurement of the vertical thickness of the melanoma. Now known as Breslow thickness, this is the distance from the top of the granular layer down to the lowest tumor cell. Over time, Breslow thickness has largely supplanted Clark level, as it has been shown to be a more accurate method of predicting prognosis. Melanomas are commonly referred to as thin (<1 mm Breslow thickness), intermediate thickness (1–4 mm), and thick (>4 mm). As the thickness of the melanoma increases, the prognosis worsens.

Staging

Treatment

Wide Local Excision

Historically, aggressive surgical resection was recommended for cutaneous melanoma. Excision margins of 5 cm were surgical dogma for much of the previous century. Beginning in the 1970s, studies began to report no adverse outcomes in patients who underwent excision with narrower margins. Since that time, multiple randomized controlled trials evaluating excision margins have established current guidelines, summarized in Table 31.2 . The principal determinant for appropriate excision margins is the thickness of the primary tumor.

Table 31.2

Recommended margins of wide local excision (WLE).

Thickness	Wle Margin ∗
In situ	0.5 cm
<1 mm	1 cm
1–2 mm	1–2 cm †
>2–4 mm	2 cm
>4 mm	2 cm ‡

∗ Lesser margins may be justified in specific cases in order to achieve better functional or cosmetic outcome.

† A 1-cm margin may be associated with a slightly greater risk of local recurrence in this Breslow thickness category.

‡ There is no evidence that margins >2 cm are beneficial; however, greater margins may be considered for advanced melanomas when local recurrence risk is high.

Based largely upon clinical experience and consensus guidelines, 5-mm excision margins are generally recommended for melanoma in situ. There are two potential issues with 5-mm margins for melanoma in situ. One is that negative margins may not be routinely achieved with gross 5-mm margins; a near 100% margin negative rate can be achieved with gross margins closer to 10 mm versus 5 mm. The second issue is that there may be some diagnostic uncertainty with melanoma in situ. In those situations in which an invasive component is found once the entire tumor is excised (as opposed to the biopsy specimen), a 5-mm gross margin would be inadequate and reexcision would be needed for an oncologically sound operation. With this in mind, it may be prudent to attempt 1-cm margins for some melanoma in situ lesions in anatomic areas that will allow for easy primary closure.

In general, the randomized trials that have focused on margins of excision have focused on intermediate thickness melanoma. The general consensus and practice for thin (<1.0 mm) melanomas is that 1-cm margins are sufficient. Several randomized controlled trials have evaluated the margins needed for intermediate thickness melanoma. As in most solid organ malignancies, no benefit in overall survival has been shown with wider excision margins for melanoma, although locoregional recurrence may be influenced by the margin of excision.

The British Collaborative Trial randomized 900 patients with melanomas at least 2 mm in thickness to a 1-cm versus 3-cm margin excision. Locoregional recurrences (defined as recurrence within 2 cm of excision or in-transit recurrence) were higher in the 1-cm margin group compared to the 3-cm margin group (hazard ratio [HR], 1.26; P = 0.05), with no difference in overall survival. Based on this study, 1-cm margins are generally considered inadequate for intermediate or thick melanomas. The Swedish Melanoma Study group trial randomized 936 patients in nine European centers with cutaneous melanoma at least 2 mm in thickness to 2-cm versus 4-cm margins of excision. No statistically significant difference in overall survival was found, and the local recurrence rates were not statistically different. Based on these trials and others, the general consensus is that 2-cm margins should be attempted for all melanoma at least 2.0 mm in thickness. For 1 to 2-mm thick lesions, 2-cm margins are recommended, although narrower (1-cm) margins are acceptable in areas in which wider margins are not anatomically feasible. These narrower margins come with the understanding that local recurrence may be higher with 1-cm margins.

Technique

WLE can be performed under local anesthesia in most cases, although general anesthesia is preferred for patients who will also undergo SLN biopsy or lymphadenectomy. The appropriate margins of excision are measured from the edge of the lesion or previous biopsy scar. This usually is achieved with a fusiform incision that encompasses the margins of excision to allow primary closure ( Fig. 31.10 ). The incision is carried down to the muscular fascia so all of the skin and underlying subcutaneous tissue are in the margin of excision. Excision of the fascia is not necessary in most cases but may be performed for patients with thick primary tumors and limited subcutaneous tissue in whom the deep margin is of concern. The specimen is submitted for permanent section pathology; frozen section analysis of margins is not performed. In most cases, the incision is closed by mobilizing the skin without the need for complex tissue rearrangement or skin grafting. Rotational tissue flaps or skin grafts are rarely necessary, except for melanomas of the head and neck and distal extremities. In these anatomic areas, grafting or tissue rearrangements are useful techniques to achieve skin closure. Tumors arising in proximity to structures such as the nose, eye, and ear may require a compromise of conventionally oncologically sound margins to avoid deformities or disabilities. Subungual melanomas are treated with amputation of the distal digit. For fingers, ray amputations are unnecessary because the melanoma commonly involves only the distal phalanx, and amputation at the distal interphalangeal joint is sufficient. In all cases, resection should achieve histologically negative margins. It should be noted that the recommended margins of excision are the grossly measured margins; it is unnecessary to reexcise the melanoma if the final pathology report indicates that the measured distance from the melanoma to the edge of the excised skin is less than the recommended margin, unless the margin is involved or almost involved by tumor.

Mohs micrographic surgery (MMS) involves the sequential tangential excision of skin cancers with immediate pathologic margin assessment. It is used most often for NMSCs such as SCCs and BCCs with good results. In melanoma, Mohs surgery is used primarily for in situ lesions, although some centers have begun to use MMS for invasive melanoma. MMS is preferred for cosmetically sensitive areas such as the face, where it may minimize the skin defect while still achieving negative margins of excision. Success can be highly operator dependent and requires full pathologic examination of the excised margins. Although there have been several single-institution reports indicating that MMS results in low local recurrence rates for melanoma, it remains controversial. In general, MMS is not considered oncologically acceptable for melanoma, except in the hands of experienced centers in highly selected cases.