Context.—Acral lentiginous melanoma is the most prevalent clinical presentation of melanoma in ethnic groups other than whites and also occurs in significant numbers in North America and Europe. Despite a clear-cut clinical picture, histologic findings seen in partial biopsies may be too subtle and deceive pathologists dealing with such cases.
Objectives.—To make pathologists aware of the histologic findings during early phases of acral lentiginous melanoma (including the in situ phase), to compare those findings with what is seen in acral junctional nevus, and to highlight their similarities and differences. This review will also emphasize the important clinical and dermatoscopic findings to be considered when diagnosing acral lentiginous melanoma.
Data Sources.—Review of published articles on the epidemiology; the clinical, dermatoscopic, and histopathologic findings; and the molecular biology of acral lentiginous melanoma as well as the personal experience of the authors when dealing with such cases.
Conclusions.—Acral lentiginous melanoma is a clinicopathologic entity with a clear-cut clinical picture: a diameter larger than 0.7 mm; ill-defined, darkly pigmented, flat lesion with irregular borders on acral locations; and the presence of mostly single-cell proliferations of melanocytes along the dermo-epidermal junction. Along with a few additional criteria, these findings should be sufficient to allow the pathologist to make the diagnosis and to recommend complete excision. Fluent communication between clinician and pathologist will facilitate a correct diagnosis.
One of the major challenges in dermatopathology is the differential diagnosis between junctional nevus and melanoma in situ in acral locations. The term acral lentiginous melanoma (ALM), coined by Reed in 1976, applies to melanoma developing on palms, soles, and nail apparatus, which presents with clinical, histologic, and even biologic differences from melanoma developing elsewhere in the body. This review will focus on the distinction between junctional nevus and early acral lentiginous melanoma occurring on palms and soles and on the diagnostic criteria required for early forms of ALM.
EPIDEMIOLOGY OF ALM
Knowledge of ALM is relevant because of its variable frequency in different populations worldwide. The disease occurs mostly in elderly patients. Although initially thought to be rare in whites, the difference is more a matter of relative frequency when compared with other types of malignant melanoma (MM) occurring in areas continuously or intermittently exposed to sunlight. When the incidence of ALM is compared among different races, the number seems quite similar.1 It is the low frequency of MM elsewhere in the body that makes ALM such an important subject in nonwhite populations, in areas as diverse as Asia, Latin America, Africa, and the Middle East. Hispanic populations seem to have the highest proportion of ALM, when compared with other types of melanoma.2 In Peru, in patients seen at public hospitals, the acral region is by far the most common location for MM. In a retrospective study3 done at the Instituto Nacional de Enfermedades Neoplasicas (Lima, Peru), the main cancer center for the country, 60% of melanomas detected during a 4-year period were located on the lower extremities. In another retrospective study,4 from the Peruvian Air Force Medical Center (Lima, Peru), 52% of all melanomas seen during a 9-year period were in acral locations. A more recent study,5 done at the Hospital Militar (Lima, Peru), the main referral hospital for Peruvian military personnel, found 52% of all MMs were in acral locations; 79% of those MM were Clark level II or higher, and 46% were Clark levels IV or V. A slight predominance in women was detected in both studies, a finding observed in other studies of ALM as well. The advanced stage of the disease at diagnosis, described in North American and European series and especially affecting nonwhite ethnic groups,6–8 may be explained, at least in part, on the basis of hesitation by pathologists to make a definitive diagnosis in the presence of subtle histologic changes, despite an obvious clinical picture.
GENETICS AND MOLECULAR BIOLOGY OF ALM
Considerable effort has been made to better explain the clinicopathologic differences between ALM and MM in other parts of the body. Using comparative genomic hybridization, Bastian et al9 compared 15 acral melanomas and superficial spreading melanoma, demonstrating that all acral melanomas had at least 1 (mean, 2.0) gene amplification, significantly more than the superficial spreading melanoma, in which only 2 of 15 patients (13%) had 1 amplification each. The most frequently amplified regions in acral melanomas occurred at bands 11q13 (47%), 22q11–22q13 (40%), and 5p15 (20%). Amplification of band 11q13 in 3 of 5 additional cases of ALM in situ (60%) further supported the notion that amplifications arise early in the progression of ALMs. Isolated melanocytes with amplifications in the epidermis up to 3 mm beyond the histologic recognizable extent of the melanomas were found in 5 of 15 invasive ALMs. Genetic differences among acral melanomas, melanomas from skin with intermittent and chronic sun exposure, and mucosal melanomas were pointed out in a seminal paper published in 2003.10 We agree with the opinion that the differences reported in this study are significant enough to warrant a new classification of melanoma based on genetic changes.
There are no specific reports evaluating the presence of chromosomal changes in acral nevi; however, in a study11 that included 54 benign nevi from different parts of the body, only 7 (13%) showed chromosomal aberrations, and most of those (6 of 7; 86%) were Spitz nevi with isolated gains involving the entire short arm of 11. Interestingly, CCND1, an oncogene that encodes cyclin D1 protein and is assumed to play a significant role in the pathogenesis of melanoma,12 is located in the 11q13 region, which is frequently amplified in acral melanomas. In an attempt to correlate molecular detection of early melanoma with the parallel ridge pattern frequently found in the dermatoscopy of melanoma, Yamaura et al13 used fluorescent in situ hybridization to detect cyclin D1 gene amplification in 33 acral lesions clinically suspected of being early melanomas. Histologically, the lesions were classified as melanoma in situ (8 cases; 24%) and melanocytic nevi (9 cases; 27%), with 16 cases (48%) too difficult to classify in either of the 2 groups because they showed only a slight increase of nonatypical melanocytes in the basal layer of the epidermis. Yamaura et al13 found amplification of cyclin D1 in 2 melanomas in situ (2 of 8; 25%), in none of the melanocytic nevi, and in 4 of the 9 cases (44%) of histologically ambiguous cases showing the parallel ridge pattern. They concluded that cyclin D1 gene amplification identified a very early progression phase of acral melanoma that precedes histopathologically defined melanoma in situ.
Recently, mutations involving the gene encoding the c-Kit protein were evaluated in the subsets of melanoma. Using quantitative polymerase chain reaction, c-Kit mutations were detected in 23% (3 of 13) of acral melanomas, 15.6% (7 of 45) of mucosal melanomas, 1.7% (1 of 58) of melanomas from other parts of the skin, 7.7% (1 of 13) of conjunctival melanomas, and 0% (0 of 60) of choroidal melanomas.14 These findings may explain some of the dramatic responses seen in patients with very high c-Kit expression and/or the documented mutations that were activated when the patient was treated with imatinib mesilate.15
CLINICAL AND DERMATOSCOPIC FINDINGS IN ACRAL JUNCTIONAL NEVUS AND ALM
In the past few years, major advances have been made in detection of melanocytic lesions through dermatoscopy, a technique now quite common among dermatologists. The use of this imaging tool, which allows melanocytic lesions to be seen under polarized light, has created another tier in the diagnostic workup to the clinical and histopathologic procedures.
Because of the rarity of other lesions, such as seborrheic keratosis, actinic keratosis, or basal cell carcinomas on acral areas, nonmelanocytic lesions on the differential diagnosis include diverse entities, such as tinea nigra or intracorneal hemorrhage. Tinea nigra is a superficial mycosis seen in tropical areas, caused by the fungus Hortaea werneckii. It produces a light brown pigmentation on acral areas, usually with a small diameter, and with a reticulated pattern on dermatoscopy. Clinicians can establish the diagnosis by scraping the lesion off because the pigment is located only on the stratum corneum. Intracorneal hemorrhage is a common event due to trauma to the hands and feet and is particularly common in sports (black heel). Unfortunately, hemorrhage, by itself, does not exclude ALM. An observer can be fooled by seeing blood in the stratum corneum if the biopsy is too superficial and fails to show the basal layer, where the subtle melanocytic proliferation is located. Once hemorrhage has been ruled out, the differential diagnosis of any pigmented lesion in acral locations narrows down to benign (nevus) or malignant (ALM) melanocytic proliferations. The most common differential diagnosis is between junctional nevus and ALM in situ. The diagnosis is clear when dealing with compound nevus, where an intradermal component of well-differentiated, small melanocytes clearly favors a benign lesion, the exception being MM associated with a preexisting nevus (how common that is, is still a matter of debate). Although early studies showed preexistent nevus associated with ALM to be rare, recent publications tend to underline the presence of nevus cells in some cases, and a case control study of ALM, done in Australia and Scotland, found patients with ALM did have higher total body and acral nevi counts.16
Acral lentiginous melanoma is easy to diagnose clinically in advanced stages; the flat, darkly pigmented lesion on the sole of a foot that is several centimeters in diameter can hardly be anything but a melanoma. All clinicians and pathologists should be aware of the importance of the diameter as a critical factor when diagnosing ALM. Studies in Japan17 have clearly defined acquired, large (>7 mm in diameter), flat, heavily pigmented lesions as almost always being melanoma, even considering the subtle changes seen in partial biopsies of such lesions, findings that can easily fool an unwary pathologist. We consider the term atypical melanosis of the foot to be confusing and more a product of a pathologist being unfamiliar with the subtle histopathologic findings of early ALM, even in the presence of a clinically clear-cut malignant lesion. This misconception may also be due to the slowing, evolving natural history of ALM: lesions may grow in diameter for many years but still keep their flat appearance. Clinically, nodular MM on the hands and feet are rare, and when seen, it is most likely the vertical component of a flat lesion that went misdiagnosed for many years, rather than a true rapidly growing lesion. Unfortunately, amelanotic forms do not seem to be rare in AMM; in those cases, the disease may go unsuspected even by the best clinician.
HISTOLOGIC FINDINGS OF ACRAL JUNCTIONAL NEVUS
Junctional nevus on acral locations start, like those in any other area, as a proliferation of individual melanocytes. The classic epidermal hyperplasia of a lentigo simplex, with thin, elongated rete ridges, may be absent. Instead, the epidermis may show the regular acanthosis associated with the acral area. Many junctional nevi in acral areas will show marked pigmentation of the deeper part of rete ridges; the melanocytes, still as solitary units, tend to group in the same areas (Figure 1). A well-defined, symmetrically distributed nest is always an excellent criterion in favor of nevus and against MM. However, many times, acral nevi do not form nests until they reach 2 or 3 mm in diameter, and the whole lesion will be a proliferation of single melanocytes along the dermo-epidermal junction. Then, it becomes very hard to evaluate the symmetry or circumscription in the absence of nesting. Differentiating early melanoma in situ from a small junctional nevus then becomes more difficult. When dealing with small, flat pigmented lesions on acral locations that are suspicious either clinically or dermatoscopically for ALM, clinicians should be advised that the best sampling method is excisional biopsy with adequate margins, avoiding the need for conservative re-excision.
The tendency of solitary-unit proliferation has also been described in lesions near, but not quite in, an acral location, such as the ankle.18 A similar, but even more dramatic, situation arises when analyzing acral congenital nevus shortly after birth: Proliferation of individual melanocytes is clearly greater than proliferation in a nest.19 In such a case, even large lesions (>5 mm in diameter) are usually solitary units, sometimes with a complete absence of nesting.
Another interesting caveat with acral-located nevi is the tendency of melanocytes in such lesions to be scattered throughout the entire thickness of the epidermis, giving the impression of pagetoid spreading. This phenomenon, more commonly seen in young patients, has been playfully called MANIAC nevus (melanocytic acral nevus with intraepidermal ascent of cells) by LeBoit.20 Proliferation through higher levels of the epidermis may extend laterally beyond the limits of the intradermal component.
Pathologists should be aware of the singular, anatomic aspects of the skin surface in acral locations. Rather than flat, it is composed of sinuous ridges and furrows, as in the pattern seen in fingerprints. Ridges are wider than furrows, and at their center, the opening of the eccrine ducts can be identified. This particular anatomic feature of the acral skin often causes nevi in these locations to adopt a striped arrangement, which dermatoscopists call the parallel furrow pattern, because of the thin, parallel, pigmented lines.21 Such peculiar organization becomes crucial when dealing with acral nevus. Several studies have shown the importance of doing microscopic sections of acral nevus perpendicular to the furrow pattern; following this technical step gives the pathologist an advantage in evaluating the symmetry and circumscription of the lesion.22 Nests and individual cells commonly concentrate at the base of the furrows (cristae limitantes), sparing the areas at the base of the ridge around the acrosyrinx (cristae intermedia). Unfortunately, the furrow pattern is not the only pattern seen in acral nevi. Other dermatoscopic patterns have descriptive names, such as latticelike, fibrillar, homogeneous, globular, or reticular, and they make it more difficult to evaluate the symmetry and nesting formation.23
Dermatoscopic examinations describe a clear, distinct pattern in ALM, called the parallel ridge pattern, which shows up more as bands than as thin lines, with eccrine pores in a line at the center of the bands. From these descriptions, an interesting observation emerges: In nevus, solitary units or nested melanocytic proliferations of melanocytes at the junctions tend to be located between sweat ducts, whereas in ALM, solitary units start to proliferate coincidental with the acrosyrinx.24 Occasionally, nevi can have the parallel ridge pattern, but microscopically, the melanocytes will be arranged in a nest.
Following the concept of symmetry, pigment of acral nevus is commonly present in the stratum corneum but tends to be arranged in columns rather that randomly through the entire width of the lesion, and many times, it is located on top of the junctional nest at the base of the furrow.
From a cytologic examination, melanocytes in nevi tend to be small and roundish, and the cells may exhibit visible dendrites. However, as a rule, those dendrites should be short, thin, and the same length throughout the entire lesion.
HISTOLOGIC FINDINGS OF ALM IN SITU AND EARLY ALM
Whereas establishing the diagnosis of invasive MM on acral locations is not difficult, efforts should be directed to make the diagnosis as early as possible, even at an in situ state. In theory, this should be possible, considering the lesion's prolonged period of horizontal extension before it invades the dermis. The main goal is to recognize MM in acral locations as a clinicopathologic entity, rather than solely a histologic diagnosis. When a clinician is confronted with a noncongenital, flat, irregularly shaped, dark brown to black pigmented lesion on the palm or sole, which is larger than 1 cm in diameter, the diagnosis can only be malignant melanoma. Because the large size makes it more difficult to excise the lesion in toto, 3 to 4 mm punch biopsies are usually taken, which is, therefore, the type of sample the pathologist will receive. If subtle changes are not recognized, the final report may be phrased in a descriptive diagnosis, such as proliferation of mostly single cells or atypical melanocytic proliferation; the clinician, expecting a more definitive diagnosis, will be at a loss, and many times, he or she will either discharge the patient or postpone more radical surgery for a later time, when the lesion is already invasive. In a sense, this is like delaying the diagnosis of mycosis fungoides until the tumor stage is reached. Therefore, fluid communication between clinician and pathologist is of utmost importance.
Clinicians should make an effort to excise the lesion completely, but, because of the large diameter of ALM, partial biopsies are usually the rule. The classic picture will be one of marked melanocytic proliferation as solitary units along the dermo-epidermal junction of a 3 or 4 mm punch biopsy. The epidermis looks acanthotic (Figure 2). The cells tend to be elongated, somewhat plump, with darkly pigmented nucleus, and a surrounding halo (Figure 3). Dendrites will be visible, ascending into high levels of the epidermis (Figure 4). The dendrites will be thick, of different lengths, and have a tendency to form a web around the basal cells (Figure 5). Although pagetoid cells (ample cytoplasm and dusty melanin) are rare, compared with the spindle pattern, pagetoid spreading is common; however, because of the small size of the nucleus, that spreading is not as obvious as it is in other locations.
Pigment is present along wide areas of the stratum corneum (as opposed to being in columns in a nevus). Nesting is rarely seen in 3 to 4 mm punch samples but can be detected in wider biopsies. It is common to see a split effect or separation at the dermo-epidermal junction in areas where basal melanocytes overwhelm the number of basal keratinocytes.
The presence of individual melanocytes crowding around the crista intermedia (around the acrosyrinx), favors ALM, especially if individual cells are also detected along the junction between the cristas, and the larger the diameter of the lesion, the larger the number of individual melanocytes at the basal layer and above. Skip areas may be present, suggesting an asymmetric distribution. Nests will appear later, and they will tend toward a lateral confluence and spindle configuration. Ulceration and inflammatory infiltrates favor melanoma over nevus.
A particularly difficult subject is the occurrence of ALM in continuity with a preexisting nevus. Reported cases confirm this possibility,25 whereas other authors emphasize the presence of a benign nevus cell component in the dermis as favoring a diagnosis of a nevus, with only a rare diagnosis of ALM in continuity with a nevus.26 Benign nevi are not rare in acral locations, which was found in mole distribution studies in patients with atypical nevus syndrome. Unfortunately, that favors neither of the 2 positions. The published reports are not conclusive, but in our experience, it is uncommon to see benign-looking areas of nevus in advanced cases of ALM.
After years, the lesion, once limited to the epidermis, becomes invasive. The dermal component may be composed of either spindle or epithelioid cells. The atypical features become more obvious, and a clear diagnosis of melanoma can be easily made; that delay, unfortunately, goes against the best interest of the patient.
We propose that ALM in situ should be considered a clinicopathologic entity: In a classic clinical picture of ALM (an irregularly shaped, darkly pigmented, flat lesion with diameter > 7 mm), pathologists should not hesitate in making such a diagnosis in the presence of a proliferation of single melanocytes along the dermo-epidermal junction. All criteria previously mentioned here should be taken into account (Table); additional effort, including serial cutting, second opinions, and even the suggestion of another biopsy, should be used to establish the appropriate clinicopathologic correlation, bearing in mind that failing to transmit the likely possibility of melanoma to the clinician may result in a delayed diagnosis. Even in an era of great advances at the molecular and genetic levels, fluid communication between the clinician and the pathologist may be the most useful tool for an early diagnosis of ALM.
From the Department of Pathology, Alberto Hurtado School of Medicine, Universidad Peruana Cayetano Heredia, Lima, Peru.
The authors have no relevant financial interest in the products or companies described in this article.