Introduction

Lentigo maligna (LM) is a pigmented skin lesion that is often mistaken for a “benign-type” lesion. However, to date, no longitudinal studies have proven this assumption. In fact, recent research is pointing to a continuous clinical and pathological spectrum of skin cancer spanning lentigo, LM, and lentigo-maligna melanoma (LMM) in the same fashion as benign nevus, melanoma in situ, and invasive melanoma. LM is currently considered to be “melanoma in situ”, with a 5–20% lifetime risk of progression to LMM [1], and represents 4–15% of all invasive melanomas [2], with properties of LM plus the metastatic potential of malignant melanoma [1, 3–5]. At this point, it is the dermatologist’s and dermatopathologist’s challenge to assess the risk of invasive potential of each pigmented lesion and to prevent further morbidity and mortality from melanoma transformation. This article is based on previously conducted studies and does not involve any new studies of human or animal subjects performed by any of the authors.

LM and LMM

LM, formerly known as a Hutchinson melanotic freckle, is defined as “melanoma in situ” occurring at a site of chronic sun exposure. In 5–20% of patients, it will develop into an invasive melanoma, most often of the aggressive desmoplastic melanoma subtype [1]. This progression can take anywhere from less than 10 years to more than 50 years, and any time in between. Confusing as these statistics may seem, these data come from studies completed 30 years ago and have not yet been refuted [1, 3–5]. The probability that LM transforms into a melanoma increases if the LM exhibits variation in color, expanding surface area, increasing border irregularity, and/or development of elevated areas [1, 3–5].

LMM, on the other hand, is a subset of melanoma that accounts for 4–15% of melanoma diagnoses [1, 3–5]. Early studies suggested a lifetime risk of developing LMM from LM of 5% [6]; however, more recent work suggests that this number may be as high as 20% [7–9]. Ultraviolet radiation-induced mutation of the hair follicle has been suggested as the cause of LMM, producing a fairly aggressive and deep skin cancer [3, 4]. LMM is so named because it demonstrates features of LM in its in situ component, while also displaying features typical of invasive melanoma [1, 3–5]. Risk factors that predispose a patient to either LM or LMM include light skin that freckles, a history of non-melanoma skin cancer, and a history of sun damage/burn. However, the development of LM/LMM has no reported dose–effect relationship with respect to an individual’s cumulative sun damage or an association with the genetic propensity of an individual to form nevi [10, 11] (Fig. 1).

LM and LMM have similar clinical presentation. The lesions are both found in chronically sun-exposed areas, often of the nose, cheeks, and ears (not often on the upper back, forearms, or legs). The peak incidence for both lesions is 65–80 years of age, but in recent years there has been increasing incidence in younger age groups (40–65 years), as witnessed in a cohort from northern California [1, 3, 12]. Characteristically, an LM/LMM will present as an atypical, non-scaly pigmented patch with ill-defined borders and variable size, shape, or shade (from tan to black to the rare amelanotic). These skin lesions grow radially and may grow/regress in a pattern that makes the LM/LMM appear to “move across” the skin [1, 3]. The skin surrounding the LM/LMM may also show signs of chronic solar damage [solar elastosis, solar lentigines, actinic keratosis (AK)]. If an LM develops dark pigmentation, sharp borders, or elevated areas, these changes may suggest a clinical progression to LMM [1, 3, 13].

Differential Diagnoses of LM/LMM

The differential diagnosis of LM/LMM includes myriad other pigmented skin lesions, including solar lentigo, seborrheic keratosis, lichen planus-like keratosis, junctional dysplastic nevus over sun damage, and pigmented AK [14–16]. However, there are distinct differences that can be used to distinguish between these differential diagnoses and the diagnosis of LM/LMM.

One of the most under-recognized differential diagnosis for LM/LMM is the pigmented AK [14]. Pigmented AKs are areas of skin that are damaged with ultraviolet radiation and are considered “pre-cancerous”. Without treatment, they may progress to either a squamous cell carcinoma (SCC) within 2 years or a basal cell carcinoma (BCC) [17, 18]. Based on reporting as of 2012, it is thought that pigmented AKs arise as a collision between a non-pigmented AK and a separate but coexistent pigmented lesion such as a solar lentigo. Therefore, the pigmentation within a pigmented AK is not believed to be due to changes in the quality or quantity of melanocytes, as in a melanoma [19]. Histologically, a pigmented AK will show signs of apoptotic keratinocytes in the epidermis and upper dermis, hyperkeratosis/parakeratosis, melanophages in the papillary dermis, and increased melanin deposition [20, 21].

Dermoscopy of LM/LMM Versus Pigmented AK

Under the dermatoscope, a pigmented AK will have characteristics such as asymmetric pigmented follicular openings, hyperpigmentation of the follicle rim, light brown pseudo-networks, light rhomboidal structures, light streaks, and peripheral grey dots. Though these features are very similar to those found with LM/LMM dermoscopy, the lighter color pigment differentiates pigmented AK from LM/LMM [13].

There are three dermoscopic criteria that significantly differentiate a pigmented AK from a malignant LM/LMM: dark rhomboidal structures, dark streaks, and black blotches. Dark streaks and blotches are specific for LM/LMM at a rate of 97% and 100%, respectively [13].

LM skin lesions show characteristic black blotches, asymmetric hyperpigmented rims around follicular openings, dark rhomboidal structures (pigmented lines surrounding appendageal openings), asymmetric peripheral dark grey dots (indicative of superficial melanotic lesions, with a greater asymmetry proportional to greater malignancy), and annular-granular structures [13]. LMM skin lesions demonstrate features such as a target-like pattern (there is no definitive pathological correlate to date, but it is believed to be related to pilar infundibulum invasion), increased vascular density, dark streaks, and either an annular-granular or peppering pattern [22]. Four features in particular are hallmarks of LMM: asymmetric pigmented follicular openings, dark rhomboidal structures, slate-grey areas, and slate-grey dots/globules/pepper pattern. If all these structures are found in one lesion, they have 89% sensitivity and 96% specificity for a diagnosis of LMM [1, 13, 23, 24] (Table 1).

Diagnosis of LM/LMM

The gold standard of LM/LMM diagnosis is the skin biopsy [1]; however, this standard is very limited in light of the high rate of diagnostic discordance among dermatopathologists [27]. Excisional biopsy is the optimal method, but may not be possible due to the size of the lesion or its location at a critical site such as the eyelid margin. An incisional biopsy site is chosen based on the most clinically significant areas by dermoscopic and clinical examination; unfortunately, due to site selection, there may be a risk of sampling error. In addition, it is also possible to perform a deep saucerization shave biopsy [1]. Both incisional and saucerization shave biopsies risk transection of the LM/LMM, therefore impacting histological diagnosis, although a recent study showed that melanoma transection does not necessarily impact overall disease-free survival or patient mortality [28]. On pathology, the diagnosis of LM/LMM is very subtle and easily missed; it is often mistaken for a junctional nevus overlying sun damage and therefore underdiagnosed [15].

Another instrument that can be used to determine the actual margins of the LM/LMM lesion is Wood’s lamp, which amplifies the difference in pigmentation between the LM/LMM and the surrounding normal tissue [16, 29]. Ultimately, however, it is often necessary to use clinical, dermoscopic, and histopathologic methods as complementary tools for a definitive diagnosis of LM/LMM.

Dermatopathologic features of both LM and LMM include atypical junctional melanocytic hyperplasia (a sign of chronic sun damage), extension of melanocytes down adnexal structures (LMM shows a characteristic pagetoid appearance), melanocyte cellular atypia (multinucleated with dendritic processes), non-uniform pigmentation and/or distribution of melanocytes, and increased melanocyte density [2, 30]. In addition, biopsies show extensive photodamage consisting of bridging/attenuation of rete ridges, epidermal atrophy, underlying elastosis, and inflammatory infiltrate in the dermis [10, 16, 26]. LM/LMM is notorious for skip areas on biopsy, leading to false-negative margins, and therefore it is often necessary to biopsy a larger area to determine where the true margins of the lesion lie [25]. Sometimes it may be useful to biopsy a “negative control” in an area of sun-damaged skin that appears normal; this will provide an individual’s background level of melanocytic hyperplasia/atypia that can serve as a reference [26]. Unfortunately, the diagnosis of LM/LMM is difficult, and there is not a high degree of concordance among dermatopathologists in interpreting excision margins [27].

To assist in the diagnosis of LM/LMM, a variety of immunostaining is available that can specifically mark melanocytes. HMB-45 (human melanoma black) is a monoclonal antibody that reacts against the antigen Pmel 17 in human melanocytic tumors; MART-1 (protein melan-A or melanoma antigen recognized by T cells) is a melanocyte surface antigen that is useful as a biomarker in melanocytic tumors (however, it is less specific, as it is found in benign nevi as well) [31, 32] (Table 2).

One of the most recent experimental advancements in histological techniques is immunostaining for soluble adenylyl cyclase (sAC). sAC generates cyclic adenosine monophosphate (cAMP), a molecule needed for signaling and regulatory melanocyte function. R21 is a mouse monoclonal antibody that is directed against amino acids 203–213 of the human sAC protein. Research has shown that invasive LM/LMMs exhibit strong pan-nuclear R21 staining. Given this strong staining, studies have reported that sAC theoretically can help define the margins of excision down to the last cell showing nuclear expression of sAC (if margins are ill-defined, a combination of sAC/MART-1 staining can be used to determine positive versus negative margins) [33, 34]. A subsequent study determined that the sensitivity of R21 in detecting LM was 88% [34]. Though R21 immunostaining is very promising, R21 is not widely used due to limited availability [33, 34] (Fig. 2).

Overview of Treating LM/LMM

For treatment of LM/LMM, there are both non-surgical and surgical methods. Non-surgical methods, depending on the type, have shown recurrence rates of 20–100% at 5 years; surgical methods have an average recurrence of 6.8% at 5 years [10]. Because of the high rates of recurrence in LM/LMM, the standard of care is surgical excision, except in cases such as elderly patients with multiple medical conditions who are inoperable or cases of large unresectable lesions. Surgical margins of 5–10 mm are recommended by multiple guidelines to ensure as complete of a resection as possible and to reduce the chance of recurrence. The most common surgical treatments include wide local excision, staged excision, and Mohs micrographic surgery [10, 35].

To date, there have been no formal clinical trials in which wide local excision has been evaluated with regard to optimal margins for LM/LMM, but a report from 2012 determined that 9-mm margins were adequate for complete clearance of the lesion in 99% of cases [36]. However, the recurrence rate with this technique is 6–9% over 3 to 3.5 years. Histological visualization may not be adequate in all cases, and removal of subclinical peripheral tissue may not be possible due to adjacent follicular involvement or the “field effect”, in which an extension of atypical melanocytes leads to a recurrence at the periphery of the treated area [10, 36, 37] (Fig. 3).

Mohs microscopic surgery uses frozen horizontal sections, allowing for faster processing of slides. In a comparison of frozen and permanent sections, only 5 % of margins that were negative in the frozen sections were actually positive by permanent section. Recurrence rates of 0–2% at 29–38 months have been reported with this method of surgical removal. Although frozen sections are more challenging to read on pathology because of the difficulty in interpreting melanocytic atypia, immunostaining with MART-1 or Mel-5 may help in the identification of melanocytes [10, 38, 39]. As it is almost impossible to determine melanocytic atypia on frozen sections, many believe this to be a serious limitation of the Mohs surgical technique for removal of LM/LMM.

Staged excision with rush permanent sections (a.k.a. “slow Mohs”) is thought to be the optimal treatment for LM/LMM, with recurrence of 0–5% at 23–57 months [10]. The most common surgery is geometric excision [40], although there are several variations of this surgical method, including square, perimeter, contoured, and spaghetti [41–44]. It is important to note that although LMM is considered a lower-risk melanoma, it still must be treated as a melanoma and excised with borders, consistent with standard treatment. However, the cons of this type of treatment include the time between each stage of excision and the inability to visualize all margins of the lesion, depending on the type of staged excision used [10]. Unfortunately, with all surgical treatments, the surgical wound can be much larger than the clinical lesion. Surgical wound healing may involve secondary intention, primary repair, skin graft, local flap, or distant flap [1, 10, 45]; the type of healing is based on the dermatologist’s best judgment.

Experimental non-surgical methods of treatment for LM/LMM all include the use of superficial destructive therapy. Though less invasive, non-surgical treatment methods are not preferred, due to the high rate of recurrence [1, 10]. Reasons for the high recurrence risk include insufficient topical application treating an inadequate surface area, topical treatment not reaching the skin depths needed to treat LM/LMM involving hair follicles, and the resistance of atypical melanocytes found in LM/LMM to destructive therapies. Unfortunately, after superficial destructive therapy, melanocytes may lose the ability to produce pigment, leading to hypopigmentation, which often delays the clinical diagnosis of recurrent LM/LMM [1, 10]. Non-surgical treatment options include cryotherapy/cryosurgery [46], radiotherapy [46–48], the use of Grenz rays [49], laser surgery including Q-switched Nd:YAG and CO2 (these treatments have the highest 5-year recurrence rate, at 43%) [1, 35, 50, 51], electrodesiccation with curettage [1, 35], photodynamic therapy [52], and 5% topical imiquimod (which may also be useful in treating amelanotic LM/LMM) [47, 53–56]. Radiotherapy is considered a superior non-invasive treatment method, as it may eradicate potentially invasive components of LM [46–48]. However, if this is not an option, combining non-surgical techniques with superior imaging (such as confocal microscopy) may also allow for appropriate assessment of response to therapy (Table 3).

In the follow-up of LM/LMM treatment, it is important to be aware of the field effect with extension of atypical melanocytes at the periphery of the treated area, which may occur after many years of remission [37]. The National Cancer Institute suggests follow-up every 3 to 6 months for the first 2 years after treatment and annually thereafter [57].

Conclusions

LM is a pigmented lesion that can undergo malignant transformation from low-grade to particularly aggressive forms of melanoma. Perhaps similar to the spectrum of malignancy from AK to SCC [58], LM and LMM lie on their own dermatologic spectrum from melanoma in situ to malignant melanoma.

Due to ambiguous dermoscopy, LM/LMM are often mistaken for junctional dysplastic nevi over sun-damaged skin, pigmented AK, solar lentigo, or seborrheic keratosis, and therefore diagnosis must be made histologically. The gold standard of LM/LMM diagnosis is skin biopsy, allowing visualization of dermatopathologic features consistent with LM/LMM. These include a pagetoid appearance of melanocytes, melanocyte atypia, and non-uniform pigmentation and distribution of melanocytes, all on a background of extensive photodamage. Though traditional melanocyte stains such as Pmel 17, MART-1, and S-100, are often used for immunohistochemical analysis of LM/LMM, new antibodies are becoming available that may aid in the diagnosis of LM/LMM. The emergence of new immunostaining techniques such as sAC/R21 staining have made it easier to determine the true margins of the LM/LMM lesions.

Fig. 1
figure 1

Pigmented lesion of the cheek in an elderly man. Biopsy showed atypical intraepidermal melanocytic proliferation extending to the lateral margins. With anti-melan-A staining, a final diagnosis of melanoma in situ was made

Full size image
Fig. 2
figure 2

sAC staining of a lentigo maligna melanoma from the face of a 64-year-old woman. a H&E staining at ×20; b H&E staining at ×40; c sAC staining at ×40; d sAC staining at ×100. sAC is a novel marker that is expressed in a pan-nuclear pattern in LM and LMM lesions. The detection with antibody R21 has an estimated sensitivity of 88 % and therefore can be used to determine the margins of the LM/LMM down to a single cell. Although this immunohistochemical stain is very promising, sAC is not yet widely used in dermatopathology labs due to its limited availability. H&E hematoxylin and eosin, LM lentigo maligna, LMM lentigo-maligna melanoma, sAC soluble adenylyl cyclase

Full size image
Fig. 3
figure 3

Pigmented lesion of the cheek in a middle-aged woman. a Pigmented lesion before removal; b surgical borders as defined by Wood’s lamp; c surgical removal of the lesion; d H&E staining on low-power view (×10); e H&E staining on higher-power view (×20); f H&E staining on high-power view (×40). After removal of the lesion, dermatopathology revealed atypical intraepidermal melanocytic proliferation overlying dermal elastosis extending to all margins; the final diagnosis was melanoma in situ on sun-damaged skin. Even with borders defined by Wood’s lamp, the dermatopathologist deemed it necessary to go back and surgically remove more at all the margins. H&E hematoxylin and eosin

Full size image

Surgical removal of LM/LMM is the preferred treatment option due to the lower 5-year recurrence rates. Recently, experimental combination therapies such as excision and topical imiquimod [59, 60], laser therapy and topical imiquimod [61], or cryosurgery and topical imiquimod [62] have emerged. Unfortunately, even after treatment, LM/LMM still carries a high risk of recurrence, and therefore patients must be monitored rigorously afterwards. In addition, given the high degree of sun damage in patients with LM/LMM, screening for additional skin lesions/cancers should be completed by a knowledgeable dermatologist on a regular basis.

Table 1 Summary of dermoscopic and histological findings of pigmented AK, LM, and LMM [1, 10, 13, 16, 19–26]
Full size table
Table 2 Overview of markers that may be used for melanocytic immunostaining
Full size table
Table 3 Treatment options for LM/LMM [1, 10, 35, 36, 38–56]
Full size table