Plastic surgery

PART II

SKIN AND SOFT TISSUE

CHAPTER 14  DERMATOLOGY FOR PLASTIC SURGEONS II CUTANEOUS MALIGNANCIES

DANIEL J. CERADINI AND KEITH M. BLECHMAN

INTRODUCTION

Skin cancer is the most common form of cancer diagnosed in the United States, exceeding the combined incidence of breast, prostate, lung, and colon cancer cases annually. It is estimated that one in five Americans will be diagnosed with skin cancer during their lifetime.1 The incidence of cutaneous malignancies has increased dramatically over the past several decades. The most recent analysis of the incidence of non-melanoma skin cancer in the US population suggests that nearly 3.5 million new cases are diagnosed annually in 2 million patients.2 This figure has tripled from the estimated annual incidence of cutaneous malignancy from 1994 and signifies a major public health problem that accounts for over 2 billion dollars in healthcare spending annually. The treatment of skin cancer comprises over 75% of all procedures performed by plastic surgeons in the United States.3

BASAL CELL CARCINOMA

Epidemiology and Risk Factors

The vast majority of skin cancer cases diagnosed in the United States are either basal cell carcinoma (BCC) or squamous cell carcinoma. BCC is by far the most common cutaneous malignancy worldwide and accounts for nearly 80% of all skin cancers. There is marked worldwide geographic variability, most often affecting light-skinned populations in locations with the highest UV exposure. Although mortality is rare, locally aggressive BCC can result in significant patient morbidity.

The predominant risk factor for BCC is intense sunlight and UV exposure. Other risk factors include Fitzpatrick skin types I–II, a family history of skin cancer, male sex, smoking, human papilloma virus (HPV), exposure to arsenic or hydrocarbons, previous radiation, and immunodeficiency resulting from either acquired immunodeficiency syndrome or systemic drugs required for transplant recipients. While most arise sporadically, BCC is also associated with several clinical syndromes, including Bazex syndrome, Gorlin syndrome (basal cell nevus syndrome), and xeroderma pigmentosum. Constitutive activation of the sonic hedgehog signaling pathway is thought to play a significant role in BCC pathogenesis.

Diagnosis and Staging

Most BCCs arise in the head and neck and sun-exposed extremities and can be categorized into one of several different tumor types that exhibit distinct growth patterns. Tumor type is a key prognostic factor and guides selection of treatment options. Nodular BCC is the most common tumor type that classically presents as a dome-shaped nodular papule with a pearly surface, scattered telangiectasias, and rolled borders. As it enlarges, it typically ulcerates centrally, giving it the classic rodent ulcer appearance (Figure 14.1). Cystic BCC is a variation of the nodular type characterized by a cystic mucin-filled central core that retains the clinical appearance of the nodular tumors. Pigmented BCC is another variation of a nodular tumor that has brown or black macules, which can be easily confused with seborrheic keratosis or nodular malignant melanoma. Superficial BCC is the second most common tumor type. It appears as a demarcated multi-centric erythematous patch, frequently occurring on the trunk and extremities. The surface of this lesion is often scaly and ulcerated and can be misdiagnosed as a cutaneous fungal infection, discoid eczema, actinic keratosis, or psoriasis. Although there is frequently “normal” appearing skin between the tumor foci suggesting that each arises separately, this subtype exhibits a significant radial growth pattern and each focus is actually connected, likely arising from a single primary focus. Morpheaform or sclerosing BCC is the most aggressive tumor type, usually found in the head and neck. It is the most difficult type to diagnose and manage due to its insidious onset and infiltrative growth characteristics. It frequently appears as a poorly defined flat indurated plaque that resembles a scar without a history of trauma. Histologically, this tumor exhibits numerous thin linear extensions that can reach into the deep dermis, making surgical resection difficult and recurrence frequent.

Tissue biopsy definitively establishes the diagnosis and characterizes the histologic subtype. Additional workup is generally not required due to the very low rate of metastasis (<0.05%) and should be reserved for patients where a genetic disorder or clinical syndrome is suspected. The overall cure rate for BCC can exceed 90%; however, recurrence is frequent and more difficult to manage. The likelihood of curative treatment is determined by several prognostic factors, which allows clinicians to identify high-risk lesions that have an increased likelihood for recurrence of a more aggressive course (Table 14.1). Risk stratification guides selection of an appropriate treatment modality for tumor eradication. Staging of BCC is rarely performed due to the very low frequency of spread to lymph nodes or distant sites.

FIGURE 14.1. Nodular basal cell carcinoma. Central ulceration gives rise to the classic rodent ulcer appearance. (Right photo courtesy of Christine Liang, MD.)

Surgical Management

The options for primary tumor eradication of BCC can be divided into destructive or surgical/excisional modalities. Destructive treatment options are generally reserved for low-risk basal cell tumors and use a variety of methods to destroy neoplastic tissue including electrosurgery, cryosurgery, topical 5-fluorouracil, topical imiquimod, intralesional interferon, radiation, and photodynamic therapy. Notably, these modalities do not definitively ensure a margin clear of neoplasm. However, in selected low-risk cases, the overall success rate can be excellent.

Surgical or excisional treatment of BCC can be used in low-risk and high-risk cases. Simple excisional biopsy of low-risk lesions in anatomically simple areas (trunk and extremities) can result in success rates over 95%.4 In order to achieve histologically negative margins, guidelines exist to assist the surgeon: for tumors <1 cm a clinical margin of 4 to 5 mm, and for tumors >1 cm a clinical margin of 5 to 10 mm is recommended.5 In high-risk cases, especially on the face where obtaining adequate margins may result in significant deformity, direct excisional biopsy allows for histologic evaluation of the surgical margins to ensure that they are free of tumor to maximize the aesthetics of reconstruction. Mohs’ micrographic surgery is the most definitive modality and treatment of choice in high-risk BCC of anatomically complex areas on the face. High-risk tumors are serially excised and the entire margin of resection is examined histologically by the Mohs’ surgeon. Presence of tumor at the margins is mapped, and further excision of affected areas is serially performed and examined until the margins are clear. Mohs’ surgery achieves over 98% cure rate in primary tumors and over 95% in recurrent cases, although it is more expensive and time consuming to perform. The primary advantage of Mohs’ surgery is that it spares normal tissue and anatomic structures in complex areas while ensuring negative margins and excellent cure rates. Based on a recent randomized clinical trial comparing Mohs’ surgery to direct excision of BCC, Mohs’ surgery resulted in better outcomes for treatment of recurrent basal cell tumors, while there were no significant statistical differences in the treatment of primary BCC.6

BCC generally carries a good prognosis as the tumors tend to grow slowly and metastasize very rarely, but can result in significant morbidity due to local invasion. One-third of recurrences occur in the first year following treatment, half in the second year, and two-thirds in the third year regardless of the treatment modality.7 Patients should be monitored every 6 months for the first year following treatment and annually thereafter.

SQUAMOUS CELL CARCINOMA

Epidemiology and Risk Factors

Cutaneous squamous cell carcinoma (cSCC) is the second most common skin cancer, accounting for 15% to 20% of all cases of skin cancer in the United States. Similar to BCC, there is marked geographic variability in the incidence of cSCC, with more patients affected in areas with increased sun exposure. Although primary tumors can be locally invasive, it is frequently diagnosed in the early stages when it is a highly curable disease. Approximately 3,000 patients die from cSCC annually in the United States,8 and the incidence of more aggressive or advanced tumors is increasing.

Chronic cumulative sun exposure is the prevalent risk factor for cSCC, and both UVA and UVB are implicated in tumor pathogenesis. This is significant because the sun protection factor in sunscreens only measures protection against UVB. The incidence of cSCC increases significantly with age, likely reflecting an increased cumulative exposure to sunlight. Other environmental risk factors for cSCC include a history of radiation, chronic inflammation (as in Marjolin’s ulcer), and exposure to arsenic and hydrocarbons. Chronic immunosuppression secondary to organ transplantation markedly increases the risk of cSCC up to 250 times the general population and is closely correlated to the type of transplant, immunosuppressive drug burden, and time since transplantation.9,10 Host risk factors for cSCC include Fitzpatrick I–II skin types, fair hair, previous history of non-melanoma skin cancer, and infection with HPV. Additionally, certain inherited disorders such as xeroderma pigmentosum, epidermolysis bullosa, and albinism confer a genetic susceptibility to developing cSCC. UV-induced mutations in the p53 tumor suppressor gene are thought to be the molecular mechanism of malignant transformation of keratinocytes.

Diagnosis and Staging

The majority of cSCC is diagnosed on sun-exposed skin of the head and neck, dorsum of hands, lower arms, and legs. Unlike BCC, however, cSCC can arise from a premalignant actinic keratosis, identified as an area of erythematous, rough, scaly plaque that exhibits dysplastic growth and malignant potential. Up to 80% of cSCC tumors arise in association with a preexisting actinic keratosis, although overall <1% of all actinic keratoses undergo malignant transformation annually.11-13 Features of actinic keratosis that are associated with malignant transformation include inflammation, diameter >1 cm, rapid growth, ulceration, bleeding, and erythema.14 A cutaneous horn is a clinical variant of actinic keratosis that presents as a hyperkeratotic protuberance shaped like a cone extending above the plane of the skin. Approximately 15% of cutaneous horns actually contain cSCC,15 and excision is indicated.

cSCC in situ, also referred to as Bowen’s disease, frequently presents as a slowly growing, erythematous, scaly patch. It is most frequently diagnosed in older patients (>60 years) and can occur anywhere on the body including the mucosal surfaces. When cSCC in situ occurs on the mucocutaneous epithelium of the glans of the penis or labia majora, it is referred to as erythroplasia of Queyrat. It occurs most often in uncircumcised men and is thought to be associated with chronic irritation, infection with HPV, and immunosuppression. It classically appears as a velvety red plaque on the glans of the penis. Progression to invasive cSCC occurs in up to 33% of cases over variable periods of time.16 When cSCC in situ occurs in the oral or genital mucosa, it presents as adherent white patches clinically referred to as leukoplakia. Notably, this must be differentiated from other causes of leukoplakia such as chronic irritation (usually from smoking), candidal infection, and HPV infection. This often requires a biopsy of suspicious lesions. Squamous cell carcinoma develops in 10% to 20% of all patients with leukoplakia.

Keratoacanthoma is a rapidly growing nodule (over weeks to months) with a central ulceration or keratin plug that is found mainly in sun-exposed skin (Figure 14.2). Left untreated, it may spontaneously involute. Keratoacanthoma is felt to be a low-grade variant of cSCC, but is clinically difficult to distinguish from high-grade invasive cSCC. Shave biopsies are not helpful in making this distinction; therefore, surgical excision is recommended.

Invasive cSCC penetrates the basement membrane to reach the dermis and either arises de novo or is associated with actinic keratosis (Figure 14.3). Characteristic lesions are firm, raised, pink- or flesh-colored papules with frequent keratinization, scaling, ulceration, or crusting on the surface. These most often represent well-differentiated tumor types. Poorly differentiated lesions are typically soft, granulomatous nodules with areas of hemorrhage, necrosis, and ulceration and lacking in keratinization. Invasive cSCC associated with actinic keratosis in sun-exposed areas has a low metastatic risk and a favorable prognosis. De novo invasive cSCC, however, is a high-risk variant typically occurring in immunocompromised hosts or in areas of chronic irritation (such as burns) and has a metastatic rate as high as 14%.17

FIGURE 14.2. Keratoacanthoma with characteristic umbilicated center and keratin plug. If left untreated, it typically undergoes a period of rapid growth followed by spontaneous regression over several months; however, it can progress to squamous cell carcinoma with metastases.

FIGURE 14.3. Invasive cutaneous squamous cell carcinoma.

Diagnosis of cSCC is made by tissue biopsy to distinguish it from other neoplasms or cutaneous inflammatory conditions. In addition to definitive tumor diagnosis, patients with cSCC should undergo clinical examination of the appropriate draining lymph node basins. Palpable nodes should be biopsied by fine needle aspiration. Routine imaging studies for cSCC is not indicated, but should be obtained in patients that exhibit specific neurological symptoms or regional lymphadenopathy.

For the first time, the American Joint Committee on Cancer (AJCC) has introduced a completely separate staging system for cSCC, which was formerly incorporated into the “Carcinoma of the Skin” comprised of 80 different non-melanoma skin cancers. This new TNM staging system utilizes a multidisciplinary evidence-based experience to more accurately describe the history and prognostic outcomes of cSCC (Tables 14.2 and 14.3).18 Due to the fact that most cSCC occurs in the head and neck, this system is meant to be consistent with the AJCC Head and Neck Staging system. The new cSCC staging system has several notable changes. The T staging (Tumor Characteristics) of the TNM system has been modified to eliminate the 5 cm size criteria and invasion of extradermal structures criteria to define a T4 lesion. Instead, a new list of “high-risk” features has been added, which impacts the overall T staging. Of these features, tumor grade now also contributes to the overall stage groups. The N component (Regional Lymph Nodes) of the TNM system has been totally revised to incorporate data indicating that overall survival decreases with increased node size and number involved.

While the majority of cSCC is diagnosed and cured in the early stages, the reported rate of regional metastasis ranges from 0.5% to 10%. While there is no consistent definition or stratification of what features of a primary tumor are considered “high risk” for regional spread, there are a number of tumor- and patient-specific characteristics that can be used to guide management. Tumor-specific features that are considered “high risk” include tumors located on the ears, lips, or within chronic wounds or scars, horizontal size >2 cm, thickness of 2 to 6 mm (low risk) or >6 mm (high risk), poorly differentiated cell types, perineural invasion, and rapidly growing or recurrent lesions.19 Patients who are organ transplant recipients or who are diagnosed with chronic lymphocytic leukemia, small lymphocytic lymphoma, epidermolysis bullosa, or HIV/AIDS are more likely to exhibit more aggressive tumor types and disease progression (Table 14.1).

Surgical Treatment

The surgical treatment options for cSCC are similar to those of BCC and are based on assessing the risk of local regional recurrence or distant metastasis. In selected low-risk cases, destructive treatment modalities can be used with excellent results. Direct surgical excision can be used for both low-risk and high-risk lesions. In order to increase the chance of achieving histologically negative margins, the recommended surgical margin for low-risk lesions is 4 mm, and for high-risk lesions it is 6 to 10 mm. An increasing number of high-risk features of the primary tumor may require a larger margin of resection. In anatomically complex areas of the face or in particularly high-risk cSCC tumors, Mohs’ micrographic surgery is the treatment of choice. Since cSCC tends to metastasize to the lymph nodes preferentially, there is some interest and initial success in using sentinel lymph node biopsy to diagnose subclinical lymph node metastasis and stage high-risk tumors. However, more controlled prospective randomized trials are required to determine whether detection of subclinical nodal metastasis will result in better clinical outcomes.20 There is currently no role for adjuvant therapy in patients who are at risk for recurrence. Patients with distant metastasis or advanced local disease not amenable to surgery or other treatment modalities require systemic chemotherapy.

MALIGNANT MELANOMA

Epidemiology and Risk Factors

Malignant melanoma is the most deadly form of skin cancer, diagnosed in 114,900 patients (46,770 noninvasive and 68,130 invasive) and resulting in 8,700 deaths in the United States in 2010.1 The incidence has been steadily increasing over the past 50 years (currently 20 in 100,000 people), with notable growth in young white women 15 to 39 years (3% annual increase over the past 15 years) and white men over 65 years (5.1% annual increase since 1975). Overall, it is the fifth most common malignancy in men and the seventh in women, and 1 in 52 people will be diagnosed with melanoma in their lifetime. Although melanoma accounts for less than 5% of all skin cancer cases, it results in greater than 75% of skin cancer deaths—approximately one person dies of melanoma every hour. Indeed, melanoma is a significant public healthcare problem—the incidence is increasing, there is a predominant modifiable risk factor (sun exposure), and it is curable in the very early stages but not in advanced disease.

Assessment of risk factors for melanoma can be divided into host factors and environmental factors. Host factors that confer a higher risk of developing melanoma include physical attributes such as fair features, Fitzpatrick I–II skin types, and blue/green eyes. Additionally, numerous common congenital nevi, atypical nevi, and giant nevi are all associated with increased risk. A personal history of melanoma is thought to confer a 5% lifetime risk of developing a second melanoma. Finally, familial melanoma accounts for approximately 10% of all cases and is associated with mutations within the cyclin-dependent kinase inhibitor 2A locus (CDKN2A), cyclin-dependent kinase 4 (CDK4), and melanocortin 1 receptor (MC1R).

Perhaps the most significant and modifiable environmental risk factor for melanoma is sun exposure, particularly intermittent and intense exposure. UVA and UVB exposure are both strongly associated with melanoma. One or more blistering sunburns early in life or greater than five sunburns at any age increases the lifetime risk of developing melanoma twofold. Regular broad spectrum sunscreen use may reduce the risk of developing invasive melanoma.21 It is important to note that while intense sun damage is very strongly associated with melanoma, it is not necessarily required for malignant transformation, as a significant number of lesions arise in relatively sun-protected areas (soles of feet, anus, and vagina). This underscores the complex multifactorial role of host and environmental factors in melanoma pathogenesis.

Diagnosis and Staging

Since early detection of melanoma is critical to improve public healthcare outcomes, the simple yet effective ABCDE diagnostic tool was developed at the New York University Langone Medical Center to educate the public and general healthcare practitioners, simplifying the decision to biopsy a suspicious lesion (Figure 14.4).22 This tool utilizes five simple criteria for identifying pigmented lesions that are suspicious for melanoma: Asymmetry, Border Irregularity, Color Variegation, Diameter > 6 mm, and Evolution or change in the appearance of lesion over time. Using these criteria, the vast majority of melanomas are detected by clinical examination. It is important to note, however, that a minority of lesions are atypical and can be nonpigmented (5%), resemble other types of cutaneous malignancies (basal or squamous cell carcinoma), or be smaller than 6 mm in size.

FIGURE 14.4. Malignant melanoma. Lesions suspicious for melanoma demonstrate asymmetry, border irregularity, color variegation, diameter >6 mm, and evolution or change in the appearance. (Photos courtesy of Richard L. Shapiro, MD.)

Melanoma can be classified into five clinical and histologic growth patterns, each of which has unique clinical characteristics: superficial spreading, nodular, lentigo maligna, acral lentiginous, and desmoplastic melanoma. With the exception of nodular melanoma, the remaining subtypes originate from an in situ radial growth phase that does not have metastatic capability. The prognostic significance of growth pattern remains controversial, although there are some genetic factors identified within the subtypes that may have prognostic value. Superficial spreading melanoma presents as flat or slightly elevated lesion with variegate pigmentation, most commonly occurring on the trunk in men and the legs in women, in patients aged 30 to 50 years. As the name would imply, the growth pattern is typically superficial and radial with scattered atypical melanocytes in the epidermis. It is the most common subtype in the Caucasian population and likely contributes significantly to the increasing incidence of melanoma over the past 30 years. Nodular melanoma is the second most common growth pattern, which often lacks the classic features commonly identified by the ABCDE melanoma screening tool. These lesions commonly present as a smooth, single colored (black or brown) elevated nodule or an ulcerated mass on examination, frequently affecting the legs or trunk. They are typically thicker and more advanced at the time of diagnosis, largely due to a relatively short or lack of radial growth phase. Overall, nodular melanomas account for most thick melanomas; however, the survival rate and prognosis is similar to that of the other clinical types when thickness and ulceration are taken into account. Both nodular and superficial spreading melanomas are associated with increased sun exposure in fair-skinned individuals. Lentigo maligna melanoma (Figure 14.5) is a slow growing lesion with radial spreading that typically arises in long-standing pigmented lesion on chronically sun-damaged anatomic sites (head and arms) in older patients. Hypopigmented lesions are also possible within this subtype. It occurs most often in fair-skinned older individuals, with an average age of diagnosis at 65 years, and is associated with solar elastosis of the surrounding skin. Acral lentiginous melanoma affects only 2% to 8% of Caucasians, but accounts for up to 36% of melanoma diagnosed in African Americans, making it the most common subtype within this demographic.23 It commonly occurs on the palms of the hand, sole of the feet, or beneath the nail plate (subungual) and presents at a more advanced stage with an aggressive course compared with the other subtypes. Subungual variants commonly present as a longitudinal line of pigment extending the length of the nail plate, with the hallmark spread of the pigment to the proximal nail fold referred to as Hutchinson sign. Finally, desmoplastic melanoma is a relatively uncommon subtype that presents as an unremarkable plaque or nodule and can easily be misdiagnosed at an early stage. It affects older patients (although not as old as lentigo maligna melanoma) most commonly in the head and neck and occurs in men twice as often as in women. Desmoplastic melanoma is frequently associated with nerve invasion and spread along fascial planes and tends to be thicker at the time of diagnosis. They are locally aggressive with a higher rate of local recurrence, but exhibit a low incidence of lymph node involvement. These clinical features more closely resemble a soft tissue sarcoma, suggesting that the underlying biology of desmoplastic melanoma may be unique compared with the other subtypes.

Definitive biopsy remains a critical factor in both establishing the diagnosis and providing valuable information about staging and prognosis, as histologic characteristics of melanoma are powerful independent predictors of 5- and 10-year survival rates. A full thickness excisional biopsy with a 1 to 2 mm margin of normal tissue is the method of choice for suspicious lesions. In larger lesions located in areas where a complete excision may be technically difficult or result in significant deformity (i.e., areas of the face), it may be necessary to perform an incisional biopsy or multiple punch biopsies. This should include the most raised area of the lesion. Shave biopsies should never be performed on lesions with a high clinical suspicion of melanoma as there is an unacceptably high rate of positive deep margins, which precludes accurate staging and treatment. All patients diagnosed with cutaneous melanoma undergo a thorough skin assessment and clinical evaluation of the relevant nodal basins. Further screening workup for newly diagnosed patients with invasive melanoma includes chest X-ray, complete blood count, liver function tests, and serum lactate dehydrogenase (LDH). Abnormal findings in the review of systems or these screening modalities should prompt further imaging studies such as computed tomography (CT) scanning or positron emission tomography (PET) scan. Patients with head and neck primary tumors are likely to benefit from CT or PET imaging to identify suspected nodal involvement. Further metastatic workup includes serum alkaline phosphatase, serum creatinine, body CT imaging, MRI of the brain, and bone scan.

FIGURE 14.5. Lentigo maligna melanoma is a slowly progressive melanoma in situ that consists of malignant cells without invasive growth. (Photo courtesy of Richard L. Shapiro, MD.)

The AJCC recently updated the staging system for melanoma from the previous system described in 2001 (Tables 14.4 and 14.5).18 The widespread use of sentinel lymph node biopsy as well as a large collaborative effort among clinicians has allowed the AJCC to update the staging guidelines using evidence-based data collected over the past decade. Using a population of 38,918 melanoma patients, the 2009 AJCC Staging and Classification guidelines identified several new findings. Tumor thickness, mitotic rate, and ulceration were the most significant prognostic factors associated with localized tumors. Furthermore, since mitotic rate is the second most powerful predictor of survival after tumor thickness, it was recommended that the mitotic rate replace Clark’s level as a criterion for defining T1b lesions. Clark’s level is no longer recommended for use in melanoma staging. The N component (Regional Lymph Nodes) of the TNM classification system was modified to include the number of metastatic nodes, overall tumor burden, and ulceration of the primary lesion. The M component (Distant Metastasis) of the TNM classification system continues to be defined by the site of distant metastasis and elevation of serum LDH, which has been identified as an independent predictor of survival outcome in stage IV patients. Finally, all patients with microscopic nodal metastasis diagnosed by sentinel lymph node biopsy (including those identified by immunohistochemical staining) regardless of the overall tumor burden are classified as stage III, which gives a more accurate survival estimate for stage IIB–IIIA disease.

The observed 5- and 10-year survival rates by stage of patients diagnosed with melanoma are summarized in Table 14.6. However, prognostication in melanoma is a complex process and continues to evolve based on observations in clinical trials. Thus, factors not accounted for by the AJCC staging system, which are known to have an impact on prognosis (age, sex, anatomic location of the primary lesions, and number of distant metastases), contribute to the individual patient’s long-term outlook. In an effort to develop a statistical model to predict survival outcome in localized melanoma using individual patient characteristics, the AJCC Melanoma Task Force released an Internet-based prognosis tool based on clinical data from >50,000 melanoma patients (http://www.melanomaprognosis.org).24 This tool was successfully validated on an independent data set and represents a valuable tool for clinicians to use to provide patients with real-time evidenced-based prognosis data based on their individual case characteristics.

Surgical Management

Wide and Deep Excision. Surgical excision not only is critical for establishing the diagnosis but is also the definitive management of malignant melanoma. Historically, 5 cm margins were advocated for local tumor excision based on observations that melanoma had a propensity to recur adjacent to the primary site. Over the past few decades, however, the guidelines for surgical margins have been redefined by several randomized prospective clinical trials25-31 and are largely based on the thickness of the primary lesion (Table 14.7). In many cases, the primary tumor can be managed with a full thickness elliptical excision, down to the level of deep muscular fascia, with primary closure. Challenging anatomic sites include the ear, face, hands, and feet. Melanoma of the ear is generally treated by full thickness wedge excision and primary closure due to the proximity of the underlying cartilage to the thin overlying skin. Primary lesions of the face can be particularly challenging, and every effort should be made to excise the primary lesion with recommended margins. However, narrower margins in anatomically complex areas for intermediate thickness lesions (1 to 4 mm) may be considered—there is a higher rate of local recurrence, but no significant impact on long-term survival.25 Invasive melanoma of the fingers and toes often requires amputation through the mid-phalanx proximal to the primary lesion (Figure 14.6). In subungual melanoma of the index, middle, ring, or little fingers, this requires amputation through the mid-portion of the middle phalanx; for the thumb, through the proximal phalanx. Likewise, for the great toe (most common site of digital melanoma) and remaining toes, amputation through the mid-proximal phalanx is recommended. Palmar or plantar melanoma requires excision down to the palmar/plantar fascia with primary closure or local tissue rearrangement. Dorsal lesions on the hands/feet or web-space lesions require soft tissue resection down to the tendon or bone with skin grafting or local flap coverage.

Sentinel Lymph Node Biopsy. The sentinel lymph node is the first lymph node in the drainage basin to receive afferent lymphatic communication from the primary tumor site, prior to spread to the other nodes in this region. Based on observations that this functionally defined node was nearly universally involved when lymph node spread of melanoma was detected in lymphadenectomy specimens, it was postulated that selective sampling of this important “marker” could serve as an accurate predictor of involvement of the rest of the nodal basin. Supported by numerous prospective randomized clinical trials, the feasibility and accuracy of sentinel lymph node biopsy has been definitively established. Due to the growing experience and success of sentinel lymph node biopsy, it is standard of care for patients at high risk for nodal metastases.

Most practitioners advocate sentinel lymph node biopsy in clinical stage I/II melanoma with tumor thickness from 1.00 to 4.00 mm and clinically negative node basins (Table 14.7). Additionally, consideration can be given to patients with tumors between 0.76 and 1.00 mm with features such as ulceration, lymphovascular invasion, age <40 years, significant vertical growth phase, and increased mitotic rate. Finally, patients with >4.00 mm tumors and clinically negative nodes benefit from the prognostic information obtained from sentinel node sampling. Technical limitations to sentinel lymph node biopsy include a previous wide and deep excision with extensive reconstruction and local tissue rearrangement, anatomic sites where there are more than one drainage basin (i.e., scalp), or anatomic sites where the primary is very close to the sentinel node (i.e., overlying the parotid) and γ-detection is difficult. In patients with a previous wide and deep excision who underwent simple closure, sentinel lymph node biopsy is a viable option and should be performed where indicated.

The process of sentinel lymph node biopsy involves mapping the sentinel lymph node by two complementary techniques: preoperative lymphoscintigraphy and direct intraoperative visualization of draining lymphatic patterns using a blue dye. Typically on the morning of surgery, patients receive an injection of γ-emitting radioactive colloid (commonly technetium-99m) around the primary tumor site, followed by serial images of the γ-emission pattern. As the colloid enters the lymphatic channels surrounding the lesion, it travels to the first lymph node where it collects, forming a “hot spot” on the emission imaging (Figure 14.7). This identifies the anatomic location of the sentinel node, but gives no information as to whether it contains metastatic melanoma. This area is marked and the patient is sent to the operating suite with the images to guide the surgeon in identification of the sentinel node. Following induction of general anesthesia, a second lymphatic mapping technique is used for intraoperative identification of the sentinel node. Blue dye (isosulfan blue or methylene blue) is injected intradermally around the primary melanoma and massaged for approximately 5 minutes to augment dye flow in the lymphatic channels. Based on the post-emission imaging and marking, a limited incision is made over the site of the proposed sentinel node, and dissection into the lymphatic basin is guided by the use of a handheld γ-probe and the presence of blue dye in lymphatic channels or nodes. The ideal sentinel node identification would be a single blue lymph node with significant γ-emission; however frequently, the blue dye is absent and identification can be reliably made by radioactivity alone. The sentinel node is then excised, taking care not to disturb the channels of the surrounding lymph node basin, and sent to pathology in formalin for permanent fixation, sectioning, and immunohistochemistry. Frozen sections are generally discouraged, although advances in immunostaining may make this technique more widely acceptable. Survey of the remaining lymphatic tissue should reveal <10% of the highest sentinel node emission. Greater than 10% suggests the presence of additional “sentinel” nodes, which should be excised and processed in a similar fashion. Following the sentinel node biopsy, the primary lesion is excised and closure of both sites proceeds.

FIGURE 14.6. Typical level of amputation for a subungual melanoma.

FIGURE 14.7. Lymphoscintigraphy demonstrates localization of radioactive colloid to the sentinel lymph node. (Photo courtesy of Russell S. Berman, MD.)

Routine histologic examination with H&E staining has traditionally been the gold standard for sentinel node examination and identifies 15% to 20% of micrometastatic disease in sentinel lymph node specimens. However, it is vulnerable to miss up to 12% of true-positive nodes. Immunohistochemical staining has enhanced the sensitivity of the micrometastasis detection by up to 40%.32 Due to its utility, immunohistochemical identification of micrometastatic nodes has been incorporated into the 2009 AJCC staging system. S-100 is the most sensitive marker for the detection of melanoma (detected in >90% melanomas) but is also expressed in a broad range of different cell types and neoplasms, limiting its specificity as a marker. HMB-45 and MART-1 are more specific cellular markers detected in melanoma, but each lacks sensitivity. Therefore, a combination of these tumor markers is used in the pathologic examination. Recently, reverse transcriptase polymerase chain reaction has been used as a molecular staging tool, and ongoing trials are evaluating its utility.

Lymphadenectomy. Complete surgical lymphadenectomy is indicated in patients with clinically involved nodes diagnosed by examination, fine needle aspiration, and/or sentinel lymph node biopsy.Nodal status is the most important prognostic factor in staging malignant melanoma. The revised AJCC staging system for melanoma accounts for the number of positive nodes as well as the overall tumor burden. Patients with only one positive node have a better prognosis than patient with multiple nodes. Some controversy exists, however, on the overall benefit of elective lymph node dissection in patients with clinically uninvolved nodes. While multiple clinical trials have failed to show any benefit overall of non-selective elective lymph node dissection, there may be some data to suggest that it may offer a survival benefit in select cases.

While potentially therapeutic, a complete lymphadenectomy carries a significant risk for substantial morbidity. Postoperative lymphedema is a major source of physical and psychological distress in patients already coping with a diagnosis of melanoma. The rate of lymphedema following axillary and inguinal lymphadenectomy can be as high as 30% and 60%, respectively, compared with the incidence of lymphedema in patients who have only had a sentinel node biopsy, which ranges from 3% to 7%. It is important to counsel patients that while the use of sentinel node biopsy significantly reduces the risk of lymphedema, it does not eliminate it completely.

Advanced Melanoma. The prognosis for stage IV melanoma is poor, with only 10% to 15% of patients living past 5 years. A number of modalities have been studied in clinical trials including immunotherapy (high-dose interleukin 2, interferon alpha, combination therapy, adoptive immunotherapy, and vaccines), systemic chemotherapy (dacarbazine and fotemustine), isolated limb perfusion, and radiotherapy, each with limited success in highly selected cases. However, there are no provocative data indicating that any of these treatments reliably prolongs survival in stage IV melanoma. In the absence of effective medical therapy, surgical management of metastatic melanoma may offer a survival benefit to selected patients with isolated or limited metastatic burden.33

MERKEL CELL CARCINOMA

Merkel cell carcinoma is a relatively rare (approximately 1,500 cases annually) but aggressive cutaneous malignancy, typically affecting fair-skinned patients with a history of sun exposure who are over 50 years of age. It has been associated with UV exposure, a history of non-melanoma skin cancer, radiation exposure, immunosuppression, organ transplantation, B-cell neoplasia, and ectodermal dysplasia. Merkel cell carcinoma typically presents as a solitary, painless, erythematous nodule on sun-exposed skin, with about half of cases arising in the head and neck and approximately 10% in the periocular region (Figure 14.8). The most common site of metastasis is the lymph nodes and 25% to 30% of patients initially present with lymph node involvement. Clinically, Merkel cell carcinoma resembles melanoma in its aggressive behavior and lymphatic spread. Diagnosis is established by biopsy, which features dense nodules of small basophilic cells in the dermal space often with lymphovascular invasion that must be differentiated from other small cell neoplasms, particularly small cell carcinoma of the lung. Further workup should include chest imaging to help exclude this possibility.

FIGURE 14.8. Merkel cell carcinoma of the scalp. (Photo courtesy of Joseph F. Merola, MD.)

Surgical management of the primary tumor is wide excisional biopsy, often with 2 to 3 cm margins. There is a high rate of local recurrence, which ranges from 40% to 45%.34 In anatomically complex areas, Mohs’ surgery should be considered. Due to the propensity of Merkel cell carcinoma to spread to the lymph node basin first prior to metastasizing to distant sites, sentinel lymph node biopsy or elective lymph node dissection is recommended. The most powerful prognostic factor in determining survival is the presence of lymph node metastasis. Radiation is considered an adjuvant therapy for tumors >2 cm and can be used following surgical management of both the primary site and lymph node basin. Chemotherapy currently has a very limited role in treatment.

Several different staging systems of Merkel cell carcinoma exist, making prognostication difficult when comparing outcomes data from disparate studies. Five-year survival rates for localized disease range from 44% to 68% and for regional or metastatic spread 23% to 42%.35,36 Recently, the AJCC developed a more standardized evidence-based staging system exclusively for Merkel cell carcinoma, no longer including it with all other non-melanoma skin cancers (Tables 14.8 and 14.9).18

DERMATOFIBROSARCOMA PROTUBERANS

Dermatofibrosarcoma protuberans (DFSP) is a soft tissue sarcoma of dermal fibroblastic origin, the pathogenesis of which is poorly understood. It occurs rarely, with an annual incidence of only 4.2 per million. DFSP often presents on the trunk and extremities of adults as a slowly growing, painless, violaceous or erythematous nodular plaque. However, its clinical appearance can vary, and development of the tumor has been associated with trauma or scar formation (Figure 14.9). Lesions suspicious for DFSP should undergo needle, incisional, or excisional biopsy to confirm a diagnosis. Most of these tumors are histologically classified as “low grade” and are associated with a favorable prognosis following surgical resection. However, the fibrosarcomatous “high-grade” variant (approximately 10% of cases) represents a much more aggressive tumor with metastatic potential. If signs and symptoms of metastatic spread are present, further evaluation with CT or PET imaging may be warranted. Despite a high recurrence rate, metastatic disease only occurs in 1% to 2% of cases. Most patients with DFSP have an excellent outcome, and mortality is rare.

Wide local excision with 3 cm margins is recommended, although subclinical extensions of tumors that appear as normal collagen may frequently lead to inadequate resection. This failure to achieve negative margins despite wide local excision is thought to be responsible for the high rates of local recurrence found with DFSP. In a study of 159 patients from Memorial Sloan Kettering Cancer Center, only 53% were found to have negative margins despite wide local excision, and 21% developed a local recurrence.37 Some authors advocate Mohs’ micrographic surgery for DFSP, particularly in cosmetically sensitive areas, but larger studies have not been published. Adjuvant radiation therapy has traditionally been used to reduce the risk of local recurrence when residual disease remains after surgery. In addition, the advent of targeted molecular therapies, such as the selective tyrosine kinase inhibitor imatinib mesylate, has provided new effective and safe options for adjuvant treatment of DFSP.38

FIGURE 14.9. Dermatofibrosarcoma protuberans arising from a prior scar. (Photo courtesy of Richard L. Shapiro, MD.)

SEBACEOUS CARCINOMA

Sebaceous carcinoma (also referred to as sebaceous gland carcinoma, sebaceous cell carcinoma, or meibomian gland carcinoma) is a rare, aggressive neoplasm that originates from cells within the sebaceous glands. A majority of sebaceous carcinomas occur in the periocular region, commonly in the eyelid, and usually arise from the meibomian glands of the tarsus. These tumors may arise de novo, but some have been shown to originate from preexisting sebaceous lesions such as sebaceous nevus. Muir-Torre syndrome is an autosomal dominant skin condition characterized by sebaceous skin tumors (including sebaceous carcinoma) associated with internal malignancies such as gastrointestinal, gynecologic, or urologic tumors. The most common clinical presentation of sebaceous carcinoma is a painless, round, subcutaneous nodule. However, due to their varied clinical and histologic appearance, sebaceous carcinomas commonly mimic benign or less aggressive lesions, which may lead to a delay in diagnosis or inappropriate treatment. Patients presenting with atypical or recurrent chalazion, eyelid thickening, or persistent blepharitis should prompt evaluation for sebaceous carcinoma. Older age and female sex are important risk factors for developing these tumors.

Orbital imaging may be performed if there is extensive periocular involvement. Histologic diagnosis is confirmed by either incisional or excisional biopsy, which should include full thickness skin, tarsus, and palpebral conjunctiva. The mainstay of treatment is surgical, using wide local excision with 5 to 6 mm margins or by Mohs’ micrographic surgery. Tumor extension posteriorly may require orbital exenteration. Recurrence occurs in up to a third of cases and metastases arise in about 25% of patients. Radiation therapy is reserved for treatment of metastatic disease or for orbital involvement in patients who do not wish to undergo exenteration.

References

1.  American Cancer Society. Cancer Facts & Figures 2010. Atlanta, GA: American Cancer Society; 2010.

2.  Rogers HW, Weinstock MA, Harris AR, et al. Incidence estimate of nonmelanoma skin cancer in the United States, 2006. Arch Dermatol. 2010;146:283-287.

3.  American Society of Plastic Surgeons. 2008 Reconstructive Surgery Procedures. Arlington Heights, IL: American Society of Plastic Surgeons; 2008.

4.  Silverman MK, Kopf AW, Bart RS, Grin CM, Levenstein MS. Recurrence rates of treated basal cell carcinomas. Part 3: surgical excision. J Dermatol Surg Oncol. 1992;18:471-476.

5.  Wolf DJ, Zitelli JA. Surgical margins for basal cell carcinoma. Arch Dermatol. 1987;123:340-344.

6.  Mosterd K, Krekels GA, Nieman FH, et al. Surgical excision versus Mohs’ micrographic surgery for primary and recurrent basal-cell carcinoma of the face: a prospective randomised controlled trial with 5-years’ follow-up. Lancet Oncol. 2008;9:1149-1156.

7.  Rowe DE, Carroll RJ, Day CL Jr. Long-term recurrence rates in previously untreated (primary) basal cell carcinoma: implications for patient follow-up. J Dermatol Surg Oncol. 1989;15:315-328.

8.  Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60:277-300.

9.  Jensen P, Hansen S, Møller B, et al. Skin cancer in kidney and heart transplant recipients and different long-term immunosuppressive therapy regimens. J Am Acad Dermatol. 1999;40:177-186.

10.  Hartevelt MM, Bavinck JN, Kootte AM, Vermeer BJ, Vandenbroucke JP. Incidence of skin cancer after renal transplantation in The Netherlands. Transplantation. 1990;49:506-509.

11.  Glogau RG. The risk of progression to invasive disease. J Am Acad Dermatol. 2000;42:23-24.

12.  Mittelbronn MA, Mullins DL, Ramos-Caro FA, Flowers FP. Frequency of pre-existing actinic keratosis in cutaneous squamous cell carcinoma. Int J Dermatol. 1998;37:677-681.

13.  Criscione VD, Weinstock MA, Naylor MF, et al. Actinic keratoses: natural history and risk of malignant transformation in the Veterans Affairs Topical Tretinoin Chemoprevention Trial. Cancer. 2009;115:2523-2530.

14.  Quaedvlieg PJ, Tirsi E, Thissen MR, Krekels GA. Actinic keratosis: how to differentiate the good from the bad ones? Eur J Dermatol. 2006;16:335-339.

15.  Yu RC, Pryce DW, Macfarlane AW, Stewart TW. A histopathological study of 643 cutaneous horns. Br J Dermatol. 1991;124:449-452.

16.  Micali G, Innocenzi D, Nasca MR, Musumeci ML, Ferraú F, Greco M. Squamous cell carcinoma of the penis. J Am Acad Dermatol. 1996;35:432-451.

17.  Cassarino DS, Derienzo DP, Barr RJ. Cutaneous squamous cell carcinoma: a comprehensive clinicopathologic classification—part two. J Cutan Pathol. 2006;33:261-279.

18.  Edge SB, Byrd DR, Compton CC, et al., eds. AJCC Cancer Staging Manual. New York, NY: Springer; 2010.

19.  Brantsch KD, Meisner C, Schönfisch B, et al. Analysis of risk factors determining prognosis of cutaneous squamous-cell carcinoma: a prospective study. Lancet Oncol. 2008;9:713-720.

20.  Ross AS, Schmults CD. Sentinel lymph node biopsy in cutaneous squamous cell carcinoma: a systematic review of the English literature. Dermatol Surg. 2006;32:1309-1321.

21.  Green AC, Williams GM, Logan V, Strutton GM. Reduced melanoma after regular sunscreen use: randomized trial follow-up. J Clin Oncol. 2010;29:257-263.

22.  Abbasi NR, Shaw HM, Rigel DS, et al. Early diagnosis of cutaneous melanoma. JAMA. 2004;292:2771-2776.

23.  Bradford PT, Goldstein AM, McMaster ML, Tucker MA. Acral lentiginous melanoma: incidence and survival patterns in the United States, 1986–2005. Arch Dermatol. 2009;145:427-434.

24.  Soong SJ, Ding S, Coit D, et al. Predicting survival outcome of localized melanoma: an electronic prediction tool based on the AJCC Melanoma Database. Ann Surg Oncol. 2010;17:2006-2014.

25.  Thomas JM, Newton-Bishop J, A’Hern R, et al. Excision margins in high-risk malignant melanoma. N Engl J Med. 2004;350:757-766.

26.  Veronesi U, Cascinelli N. Narrow excision (1-cm margin). A safe procedure for thin cutaneous melanoma. Arch Surg. 1991;126:438-441.

27.  Veronesi U, Cascinelli N, Adamus J, et al. Thin stage I primary cutaneous malignant melanoma. Comparison of excision with margins of 1 or 3 cm. N Engl J Med. 1988;318:1159-1162.

28.  Khayat D, Rixe O, Martin G, et al. Surgical margins in cutaneous melanoma (2 cm versus 5 cm for lesions measuring less than 2.1-mm thick). Cancer. 2003;97:1941-1946.

29.  Cohn-Cedermark G, Rutqvist LE, Andersson R, et al. Long term results of a randomized study by the Swedish Melanoma Study Group on 2-cm versus 5-cm resection margins for patients with cutaneous melanoma with a tumor thickness of 0.8-2.0 mm. Cancer. 2000;89:1495-1501.

30.  Balch CM, Soong SJ, Smith T, et al. Long-term results of a prospective surgical trial comparing 2 cm vs. 4 cm excision margins for 740 patients with 1-4 mm melanomas. Ann Surg Oncol. 2001;8:101-108.

31.  Karakousis CP, Balch CM, Urist MM, Ross MM, Smith TJ, Bartolucci AA. Local recurrence in malignant melanoma: long-term results of the multiinstitutional randomized surgical trial. Ann Surg Oncol. 1996;3: 446-452.

32.  Stebbins WG, Garibyan L, Sober AJ. Sentinel lymph node biopsy and melanoma: 2010 update. Part I. J Am Acad Dermatol. 2010;62:723-734; quiz 735-736.

33.  Hussussian CJ. Surgical treatment of advanced melanoma. Clin Plast Surg. 2010;37:161-168.

34.  Allen PJ, Bowne WB, Jaques DP, Brennan MF, Busam K, Coit DG. Merkel cell carcinoma: prognosis and treatment of patients from a single institution. J Clin Oncol. 2005;23:2300-2309.

35.  Eng TY, Boersma MG, Fuller CD, Cavanaugh SX, Valenzuela F, Herman TS. Treatment of Merkel cell carcinoma. Am J Clin Oncol. 2004;27:510-515.

36.  McAfee WJ, Morris CG, Mendenhall CM, Werning JW, Mendenhall NP, Mendenhall WM. Merkel cell carcinoma: treatment and outcomes. Cancer. 2005;104:1761-1764.

37.  Bowne WB, Antonescu CR, Leung DH, et al. Dermatofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution. Cancer. 2000;88:2711-2720.

38.  Johnson-Jahangir H, Ratner D. Advances in management of dermatofibrosarcoma protuberans. Dermatol Clin. 2011;29:191-200.