This article is mentioned in:

1. Introduction

Lentigo maligna (LM), also known as malignant freckle, Hutchinson's melanotic freckle, Melanosis circumscripta precancerosa of Dubreuilh or melanoma in situ-LM type, is the most common clinicopathological subtype of in situ malignant melanoma (MM), accounting [along with lentigo maligna melanoma (LMM), its invasive counterpart] for 4-15% of all MM cases (1). Its incidence has been recently increasing worldwide, including in Europe (2,3), the USA (1) and Australia (4), as a consequence of ageing and the increased sun exposure of the population over the past decades. Australia seems to have the highest incidence of LM, estimated at 12.2 per 100,000 individuals (4).

Clinically, LM manifests as a flat macule with irregular contours and a variegate colour (Fig. 1). It is light- or dark-brown in colour, often with several hues, including black or whitish areas (reflecting partial regression), and may exceptionally be amelanotic. It develops on skin areas with high cumulative sun exposure, typically the head and neck (namely the cheeks, forehead, temple, periocular skin, nose and ears) and more rarely on extrafacial sun-exposed body zones (limbs) of middle-aged and elderly patients (mean age, 65 years) with fair skin, exhibiting signs of chronic sun damage, particularly in males (5); however, LM may be observed in younger patients in their 4th decade of life (6). In its early stage, LM may clinically mimic other pigmented facial lesions, such as solar lentigo, pigmented actinic keratosis, early seborrheic keratosis, lichenoid keratosis and pigmented basal cell carcinoma. LM grows radially over months or years [radial growth phase (RGP)] and may after months or years, invade the dermis (vertical growth phase), thereby progressing to invasive MM (LMM), which has potential for metastasis and a fatal outcome. The progression of LM to LMM manifests clinically with the appearance of a thickened zone or an infiltrated nodule, and pathologically belongs usually to the desmoplastic MM variant. The estimated lifetime risk of progression to LMM varies greatly (from 5-50%). A recent Australian study estimated the annual risk of progression to LMM at 3.5%, and the average time of progression to LMM at 28.3 years (7). As the delay to progression can be lengthy (>30 years), this progression may not be observed within the lifespan of elderly individuals.

Figure 1 Lentigo maligna on the cheek of an elderly patient manifesting as a large, ill-defined macule with heterogeneous pigmentation.

The diagnosis of LM is performed clinically and may be aided by non-invasive imaging techniques, such as dermatoscopy and in vivo reflectance confocal microscopy (RCM), which increase the sensitivity and specificity of diagnosis vs. simple visual inspection (8). By dermatoscopy, LM presents an atypical pseudonetwork characterized by different patterns, depending on the age of the lesion. These include, in chronological order, an annular-granular pattern due to the aggregation of blue-grey globules around follicular ostia; a greyish pseudonetwork secondary to thickening of the annular-granular structures; the asymmetric pigmentation of follicular openings (due to irregular atypical melanocyte infiltration of follicular ostia); greyish to black-brown rhomboidal pigmentation; homogeneous grey-blackish pigmented blotches due to obliteration of hair follicles; and reddish rhomboidal structures, scar-like whitish and milky-red areas; occasionally increased vascularity (9,10). Grey colour was the most prevalent finding in one study (11). The use of deep learning convolutional neural network seems to have a high-level performance in the diagnosis of LMM (12). The major diagnostic criteria for LM by in vivo RCM include pagetoid infiltration of the epidermis by large (>20 µm) dendritic or round hyperreflective cells and non-edged dermal papillae. Minor criteria include ≥3 round or dendritic atypical cells at the dermal-epidermal junction in 5 fields of 0.5x0.5 mm, the localization of pagetoid and/or atypical junctional cells within hair follicles, and nucleated cells within the dermal papillae. On the basis of these findings, a ‘LM score’ has been proposed, which has a good diagnostic sensitivity (85-93%) and specificity (61-76%) (13). These non-invasive imaging methods, which can help differentiate LM from clinically similar-looking pigmented facial lesions, have gained popularity and are undoubtedly useful in the diagnosis of LM (14,15); however, the histological examination of a skin biopsy still remains the gold standard for the definitive diagnosis of LM. For suspicious lesions, an excisional (elliptical) biopsy with narrow margins (1-3 mm) around the lesion and under the anticipated deep limit of the lesion seems to be the most effective technique, as this minimizes the risk of missing the most invasive zone of the tumour; however, this is often not feasible for large facial lesions, which account for the majority of LM cases. An incisional or punch skin biopsy/ies is/are therefore often preferred, even though this procedure carries a higher risk for misdiagnosis (16). Indeed, an incisional (or punch) biopsy can easily rule out a non-melanocytic pigmented lesion which may clinically mimic LM (such as pigmented actinic keratosis or seborrheic keratosis); however, in the case of LM it may miss an invasive area which, if present, would be synonymous with invasive melanoma. Therefore, the biopsy/ies should be obtained from the clinically most suspicious area (most infiltrated, darkest or with the highest pigmentary irregularity). Dermatoscopy and in vivo RCM can assist in defining the most informative area from which biopsy/ies should be taken and can help to define the microscopic limits of the lesion, which is useful for the surgical treatment of LM, as atypical melanocytes often extend far beyond the clinically-visible limits of the lesion (17,18). The biopsy of more than one site increases the sampling of the lesion and thereby reduces the risk of missing subclinical dermal invasion; indeed, up to 9-32% of cases diagnosed as LM by a partial (diagnostic) biopsy prove to be invasive (i.e., LMM) upon the examination of the completely-excised lesion (19-21).

The main histopathological feature of LM is a lentiginous proliferation of atypical melanocytes within the basal layer of the epidermis (Fig. 2). These cells may be naevoid, epithelioid or spindle-shaped, and often have a vacuolated cytoplasm and basophilic, occasionally multiple, nuclei. Atypical melanocytes are also frequently found within the basal epithelial layer of the hair-follicle epithelium and are responsible for recurrences following treatment modalities that do not penetrate deep enough in the dermis (22). Atypical melanocytes may form irregular junctional nests and may later invade the dermis, initiating a vertical growth phase (LMM). Two major patterns of RGP have been described, i.e. a ‘classic’ pattern with a dense proliferation of atypical melanocytes along the dermal-epidermal junction with few nests, and a naevoid (or dysplastic nevus-like LM) characterized by more prominent nest formation with occasional bridging of adjacent rete ridges. Additional histologic findings include atrophy of the epidermis (that has a flattened dermal-epidermal junction), severe actinic dermal elastosis (reflecting chronic sun-damage) and infiltration of the upper dermis with melanophages and some lymphocytes (23). LM should be distinguished from atypical melanocytic hyperplasia (solar melanocytosis), which is often found in chronically sun-exposed skin of the face; this differentiation may be difficult, particularly in early LM. Immunostaining for the cancer testis antigen, PRAME, seems to be a promising tool to this effect (24).

Figure 2 Microscopic aspect of lentigo maligna: A lentiginous proliferation of atypical melanocytes within the basal epidermal cell layer is seen (haematoxylin-eosin-saffron stain, original magnification, x250).

According to the latest edition of the WHO classification of skin tumours (2018), LM is classified among melanomas with high cumulative sun damage, due to the activation of the signaling pathway II. LM carries predominantly mutations in the NRAS, KIT and BRAF (non-V600E) genes, more rarely of the NF1, TERT, CDKN2a, TP53, PTEN and RACI genes (23,25).

2. Principles of LM treatment

To date, to the best of my knowledge, there are no available prospective or randomised controlled trials allowing the establishment of definite guidelines for the management of LM (26). It is however, generally admitted that, as for other melanoma forms, the most effective treatment for LM is complete surgical excision, since this achieves the highest cure rates. However, clinical factors, such as the patient's age, life expectancy and comorbidities, and the size and location of the lesion affect the manner in which LM is treated. Other destructive or medical treatments can be proposed when surgery is not a feasible option due to patient- or lesion-related factors. The most appropriate treatment should be tailored to each individual patient, if necessary in the setting of a multidisciplinary oncology team, the aim being to offer the optimal risk-benefit ratio to each patient.

3. Surgical excision

Complete surgical excision with clear histological margins is the best treatment of LM. Indeed, this allows the histopathological assessment of excision margins to ensure complete tumour removal, allows the detection of a clinically inapparent invasive component (in which case the management will be more aggressive) and removes hair follicles where atypical melanocytes may exist, thereby decreasing the risk of recurrence. Surgical excision achieves higher clearance rates than those achieved by non-surgical treatments (27) and is therefore considered the first-line treatment of LM. The definite surgical excision should preferably be performed within 4-6 weeks following the diagnosis of LM. The aim is to excise the lesion completely (i.e., with tumour-free margins), while conserving a maximum of healthy tissue, so as to entail the least possible aesthetic and functional prejudice and maximize the patient's quality of life.

Surgical excision is performed with various modalities, such as standard (wide) excision with predefined safety margins, or with techniques with complete peripheral margin assessment [staged surgical excision (SSE)]. These include mainly radial sections, the square method, the perimeter technique, the ‘spaghetti’ technique and Mohs micrographic surgery (MMS). Compared to wide excision (WE), microscopically-controlled SSE achieves better results in terms of recurrence (28), although a recent study did not find better survival rates of patients treated for LM of the trunk and limbs with MMS vs. wide local excision (29).

WE

WE is performed as a one-stage procedure using predefined safety margins. It is suitable for well-defined, relatively small LM and extrafacial LM. A histological examination is performed on routinely-processed (formalin-fixed, paraffin-embedded) tissue sections, usually prepared with the ‘bread-loaf’ technique. However, the size of safety margins is a matter of debate. For in situ MM of the trunk and extremities, the majority of consensus guidelines recommend excision margins of 5 mm around the visible lesion; however, in LM there is often a subclinical extension of atypical melanocytes, which may be present several centimetres beyond the visible lesion, so that in approximately half of all LM cases, 5 mm margins are not sufficient to achieve histologically-negative margins (30). One study found that the mean surgical margin required for complete LM excision was 7.1 mm (31). Another study demonstrated that a clearance rate of 94% could be achieved with a wider (up to 1.5 cm) margin (19). A more recent study suggested that initial surgical margins of at least 12 mm should be used to achieve histologically negative margins in 97% of head/neck primary melanomas (32). In a prospective study on 1,120 patients with MM in situ treated by MMS, surgical margins of 6 and 9 mm resulted in the removal of 86 and 99% of the tumours, respectively (33). On the basis of such studies, the French guidelines recommend margins of 1 cm; when this margin is unfeasible due to anatomic or functional limitations, a margin of 5 mm is acceptable, pending strict microscopic control of the entire tumour margin, either with SSE or MMS prior to definitive would closure (34). The European consensus guidelines recommend margins of 5 up to 10 mm, and for larger lesions, microscopically-controlled surgery (35). The National Comprehensive Cancer Network (36) and the American Academy of Dermatology (37) also recommend margins 5-10 mm for in situ MM; however, it should be acknowledged that complete excision may require wider margins and/or margin control techniques that allow histologic assessment of the entire peripheral margin. The optimal deep margin of excision has not been adequately investigated to date; this commonly includes the deep subcutaneous fat (37), but not the aponeurosis, since removing this structure does not improve the rate of recurrence (or survival, in the case of LMM).

Standard WE with predefined margins achieves clearance rates ranging from 24 to 70%, which are lower than those achieved with techniques with complete peripheral margin assessment (27). Another limitation of standard WE followed by immediate reconstruction is the difficulty to localize the residual tumour if the margins are not tumour-free, particularly following complex tissue reconstruction. This can be avoided with surgical techniques allowing for complete margin assessment. Margin-controlled excisions are indicated for most LM of the face and for large LM with ill-defined clinical borders.

SSE

SSE techniques comprise several consecutive stages of excision, with each stage defined by the histologic findings of the previous stage. The first stage(s) is excision of peripheral margins, if necessary in multiple steps, until histologically tumour-free ones are obtained; this is followed by the excision of the main tumour. SSE is based on the examination of paraffin-embedded tissue specimens, which ensures a more reliable histopathologic interpretation of the margins, if necessary with the addition of immunohistochemical staining for melanocytic antigens (such as Melan-A, HMB-45, SOX-10 and PRAME) (38). These techniques do not leave the patients with an open wound prior to final reconstruction; however, they entail a delay of at least 24 h before the re-excision and closure of the defect/reconstruction can be performed, usually with flaps or grafts. The recurrence rate achieved with SSE is reportedly <10%; one comparative study found it significantly lower than that obtained with MMS (39). Various SSE techniques have been developed: i) SSE with radial vertical sections includes the traditional excision of the main tumour (debulk) followed by the excision of 4 peripheral quadrants (with 5-7 mm margins) which are marked so as to allow their precise mapping (31). The wound is left open until the pathological examination reports clear margins. Each quadrant is routinely-processed and cut radially at 1-2 mm intervals in a clockwise manner, which allows visualization of their external/distant margins. These are considered clear if no atypical melanocytes are found >2 mm from the peripheral margin, otherwise additional 2-3 mm quadrants are excised and processed in the same manner, until clear margins are obtained. Permanent closure of the wound can subsequently by performed. This technique achieves a low recurrence rate (<5% at 5.9 years) (40). ii) The ‘square technique’ consists of defining a square margin around the lesion and excising thin (2-4 mm) strips around it with a double-bladed scalpel, while the tumour is left in place and the resulting defect sutured (41). The strips are routinely-processed and sectioned vertically, which allows to examine the whole periphery of the square. If the margins are not tumour-free, additional (peripheral) skin strips are excised and examined until clear margins are obtained; this is followed by total excision of the tumour. With this technique clearance rate of 97.7% has been reported, with a long follow-up (42). iii) The ‘spaghetti’ technique is similar, but involves the excision of a non-angled strip peripheral to the tumour, parallel to its contour, with a double-edged scalpel; this is simpler than MMS, and can be used to map the lesion before it is finally removed (39).

MMS

MMS is a specialized form of SSE allowing for total margin assessment. This is achieved similar to the SSE techniques, by first debulking the main tumour with margins of 3-6 mm, and then obtaining tissue strips peripheral and deep to the main tumour with immediate examination of frozen sections stained with haematoxylin-eosin. The margins are examined intraoperatively until clear ones are obtained, allowing for the immediate closure of the wound. A limitation of MMS lies in the difficulty in assessing histologically atypical melanocytes (particularly when they are scattered within the basal epidermal layer) due to the poor quality of frozen tissue sections; immunohistochemical stains for melanocytic antigens (such as gp100/HMB-45 or PRAME) can be used to improve the visualization of atypical melanocytes. This problem probably accounts (at least in part) for the rare recurrences following ‘complete’ excision of LM by MMS. An alternative technique (known as ‘slow Mohs’) is the examination of routinely-processed (formalin-fixed, paraffin-embedded) sections, which provide a better visualization of tumour cells (43). As with the above-mentioned SSE techniques, the ‘slow Mohs’ entails a delay of at least 1 day prior to the definitive closure of the wound. MMS allows smaller initial margins than those of standard excision, and therefore has the advantage of sparing healthy tissue (although in LM the difference with SSE regarding post-surgical lesion size may not be significant). The recurrence rate of LM following MMS is usually in the order of 4-5% (44-47), and in some studies, even lower (0.1-0.5%) (48,49).

4. Non-surgical treatments

Non-surgical treatments may be envisaged in some clinical settings, namely when complete excision would entail unacceptable mutilations, or in elderly or frail patients in whom comorbidities or advanced age render surgery impossible. These non-surgical modalities entail lower morbidity and cosmetic impact; conversely, however, they carry a higher risk of recurrence compared with surgical excision, namely due to the lack of histological tumour assessment, a fact also exposing to the risk of missing invasive MM (LMM). Non-surgical treatment options include the following.

Radiotherapy (RT)

RT has been used for decades for the treatment of MM, including LM, although nowadays it is not considered as the primary treatment option. It is however, included in current European treatment guidelines (35,50) as a second-line treatment for selected patients in whom surgical therapies are not possible or undesirable by the patient. A prospective multisite international randomized controlled phase 3 trial of RT vs. imiquimod (IMQ) for histologically-confirmed LM is underway (RADICAL, NCT02394132) and is expected to provide data with a high level of evidence which may influence future guidelines for the management of LM.

A recent literature review identified 14 series of patients with LM mainly (79-100%) of the head/neck area (total of 1,075 lesions) treated with RT in Europe and the USA from 1940 to 2014(51). A wide range of modalities of RT have been applied, and include mainly low-energy (soft) Grenz rays (GR) and superficial X-rays (‘contact RT’). GR (more commonly used by dermatologists) have an energy range 10-30 kV and limited tissue penetration (half-dose depth of about 1 mm), whereas superficial X-rays (preferred by radiation oncologists) have an energy of 30-150 kV and a deeper tissue penetration (half-dose depth of approximately ≥1 cm).

GR RT has been administered at total doses of 40-160 Gy in 3-13 fractions, every 1 to 4 days (single fraction dose of 10-20 Gy). Superficial RT is more commonly delivered at lower total doses (35-60 Gy), in 5-23 fractions every 1 to 4 days (single fraction dose 2.5-7 Gy) (51). A minimum of 3 fractions per week has been recommended (52). The radiation field must be sufficiently large to account for microscopic tumour extension, so as to avoid recurrences; it ranges between 5 and 10 mm around the visible lesion, up to 20 mm in some studies (53). The radiation should reach deeply-seated melanocytes (namely those contained in hair follicles), considering that the radiation energy decreases with increasing depth; 5 mm is considered sufficient in most cases. In this respect, superficial RT (or high energy GR) is favoured by several authors, as this allows the wider coverage of the target volume, including the skin appendages (51,52). GR RT can nevertheless achieve a high rate of control rates (90%) (54,55), although an older study from the USA reported much lower control rates (56). The recurrence rates varied from 0 to 31%, both for GR and superficial RT, with a follow-up time of 15 to 96 months (53-55,57-62). Pigmentation disappeared within 1-24 months following RT. An assessment of the result 6 months following RT completion is advised. Recurrences have been reported up to 9 years post-treatment, therefore continuous follow-up seems necessary.

Additional RT techniques include brachytherapy and external beam RT. Brachytherapy involves the application to the skin of a mould containing radioactive sources (tubes or wires); it can irradiate large areas but necessitates an expert construction and skill. External beam RT is made of either photons (superficial RT) or electrons (52).

The cosmetic results of RT were reportedly in most cases good to excellent, although late complications may develop, such as telangiectasias, atrophy and hyper- or hypo-pigmentation (57,61); however, no fibrosis or ulceration were reported, and the risk of new skin cancer development due to RT in this population of elderly patients does not seem concerning. RT may also be administered as adjuvant treatment following incomplete surgical excision, or when LM is detected post-surgery by dermatoscopy or in vivo RCM. In these cases, RT doses may need to be increased (52).

Topical IMQ

This is a recommended treatment when both surgery and RT are not appropriate or not preferred by the patient (63). IMQ [1-(2-2methylpropyl)-1H-imidazo(4,5-c) quinolin-4-amine] is a synthetic Toll-like receptor 7-agonist approved for the topical treatment of genital warts, actinic keratoses and superficial basal-cell carcinomas. It is an immune-response modifier, stimulating skin macrophages to secrete several cytokines [such as interleukin (IL)-6, IL-1α, IL-8, IL-12, tumour necrosis factor (TNF)α, interferon (IFN)-α, IFN-γ, granulocyte-macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF)] which exert antiviral and antitumour effects. Local inflammatory side-effects (redness, burning sensation and erosions) frequently develop, and may cause burden to the patient; however, inflammation seems to be associated with the efficacy of treatment (64).

IMQ (available in European countries as 5% cream) can be used as an off-label treatment in different settings; i.e., as a neoadjuvant therapy prior to surgery in order to reduce tumour size (65), as primary monotherapy in selected patients who are not candidates to (or refuse) surgical excision, or as adjuvant therapy following excision. The dosing schemes are variable, varying from twice daily to 5 times weekly applications over a period of several weeks (6-44), with treatment margins from 0.5 to 2 cm around the visible lesion. In some studies, an escalation of the dosing scheme was performed so as to reach an inflammatory response, up to twice daily 7 days/week for 10 weeks (64,66). In the majority of studies (67-71) the follow-up time was relatively short, usually <3 years. The clearance rates reportedly range between 72 and 93%; however, approximately 27% of patients exhibit residual LM on histological analysis (66). An ‘intensive' treatment (daily applications >12 weeks) resulted in a high (74%) clinical and microscopic clearance rate (72). A recent study also reported a 72% long-term (4.1-year) clearance rate following IMQ treatment once or twice daily for up to 22 weeks (64). Clinical recurrences following primary IMQ therapy develop in 4-24% of cases after an average of 3.2 years (72-74) and seem to be more common for LM of the nose (75). The development of an inflammatory reaction to IMQ seems to be associated with a better clinical and histological response (64,76-78), contrary to lesion size and treatment duration. Factors found to be predictive of local recurrence include the total number of melanocytes, the number of basal and suprabasal melanocytes and the number of pagetoid spreading melanocytes (73). Responses are associated with an increased programmed death-ligand 1 (PD-L1) expression and the upregulated expression of genes involved in antigen presentation, T-cell activation and type I IFN signalling, which could be used as biomarkers of LM response to IMQ (79). IMQ can be used in association with local applications of tazarotene 0.1% gel twice-weekly; this decreases stratum corneum cohesion, and thereby increases IMQ penetration (70,80,81), and has achieved a slightly better, although statistically not-significant, clearance rate (80). A review of published studies as of 2015 concluded that IMQ as monotherapy for LM achieves clinical and histologic clearance rates of 78.3 and 76.2%, respectively. Clinical recurrences occur in 2.3% of cases, following a mean follow-up of 34 months. A higher likelihood of histologic clearance is obtained with >60 total applications or with >5 weekly applications (82); however, lower pathological and clinical clearance rates (37 and 43%, respectively) were reported in a more recent study (83); furthermore, progression to LMM has also been observed after IMQ treatment (84-86). Surgical excision can be performed in case of persisting LM, and this further reduces the 5-year recurrence rate (66). Patients treated with IMQ should be followed in the long-term (at least for 5 years) and post-treatment biopsies are recommended even if no recurrence is observed clinically (87). RCM is useful to monitor the response to medical treatments, including IMQ (88).

IMQ can also be used before SE in order to reduce the size of the required surgical margins, or as an adjuvant treatment in patients with positive margins after adequate surgery, or after laser ablation. In a recent study, IMQ 5% was applied as neoadjuvant treatment for an average of 2.5 months before staged excisions. This necessitated a mean surgical margin of 3.5 mm and resulted in a recurrence rate of 3.9%, i.e., similar to reported recurrence rates with SSE by either MMS or en face permanent sections; the mean time to recurrence was 4.3 years and the mean follow-up time 5.5 years (81).

5. Other treatments

Other treatment modalities have been used for LM, but are not included in current consensus guidelines due to unproven or low efficacy. These include the following.

Cryotherapy

This technique consists of an application of a cryogenic agent (usually liquid nitrogen) to the skin in order to destroy superficially-located tumour cells. Melanocytes are more sensitive than keratinocytes to freezing, being destroyed at temperatures between-4 and -7˚C. The majority of relevant studies have included small series of patients with LM. A total of 2-3 freeze-thaw cycles (possibly under local anaesthesia) are usually applied with 5-10 mm lesion margins. Although good results have been reported by some authors (89), other researchers have found high (up to 40%) recurrence rates (90). Cryotherapy may make subsequent progression difficult to detect, and has been complicated by the development of amelanotic MM (91). Hyper- or hypo-pigmentation of the treated area is a possible side-effect (90). Some researchers have used a combination of cryotherapy with IMQ (cryo-immunological treatment or ‘immunocryosurgery’) (3).

Photodynamic therapy (PDT)

PDT consists of an application of a photosensitizer (usually aminolevulinate) followed by illumination with red light at 635 nm. It has been used in small patient series and has resulted in an 80% clearance rate (92). More recently, PDT assisted by ablative fractional laser was used in order to increase the absorption of the photosensitizer, and achieved a 70% histological clearance rate (93).

Laser therapy

Various types of lasers (including Q-switched ruby, Q-switched neodymium-doped yttrium aluminium garnet, carbon dioxide, argon, alexandrite and various combinations) have been used for the ablation of LM (94-96). Potential advantages include better cosmetic results compared with surgery, less pain, speed of therapy and less post-treatment care. The clearance rates are variable, 100% in one study (97), but lower in others (60-80%), probably as the laser beam may not reach a sufficient depth to destroy atypical melanocytes seated deeply within hair follicles, which is a drawback of laser therapy. The combination of ablative laser therapy followed by IMQ has recently been advocated, but recurrences seemed to be frequent, particularly on the nose (75).

Curettage and electrodessication

A small number of patients have been treated with this modality, with rather high recurrence rates (98).

Topical azelaic acid (AZA)

AZA is a dicarboxylic acid that competitively inhibits tyrosinase in vitro. On this basis, it has been tried for the treatment of LM (applications of 15-20% cream for up to 3 months). Local disease control has been reported; however, progression to invasive melanoma has also occurred (99).

Topical 5-fluorouracil (5-FU)

This is an antitumour analogue of pyrimidine approved as 5% cream for the topical treatment of actinic keratoses and superficial basal cell carcinomas. It has been tried in the treatment of LM (twice daily applications for 6-13 weeks); however, recurrences were very common (100,101).

Topical cidofovir 1%

This is an acyclic nucleoside analogue of deoxycytidine monophosphate that may block DNA synthesis; it has been used topically for the treatment of DNA viral infections (HPV warts, molluscum contagiosum, herpes simplex) and has proven effective in a small number of basal- and squamous-cell carcinomas and in two cases of recurring LM (102).

Ingenol mebutate (IM)

IM is a diterpene ester obtained from the plant Euphorbia peplus, and was licensed some years ago for the treatment of actinic keratoses. It has been tried in a small number of patients with LM, although the results have been rather disheartening (103,104). IM was recently withdrawn from the European market, although it is still commercially available in other parts of the world.

Finally, elderly patients with facial LM that are not suspicious of being invasive and who are unable (or refuse) to receive aggressive treatments may be closely monitored with non-invasive imaging techniques (macroscopic and dermatoscopic photography, in vivo RCM) and undergo skin biopsy if changes suggestive of progression to LMM occur. This conservative watchful attitude was preferred by almost 20% of European dermatologists for patients >70 years of age (105).

LMM has the same prognosis as other MM subtypes when adjusted for Breslow thickness and other prognostic factors, and should therefore be treated according to the guidelines established for other MM types, even though the localization on the face and the size of the lesion often pose challenging functional and anatomic problems. The therapeutic decision should be tailored to each patient on an individual basis after discussion in the setting of a multidisciplinary oncology team.

6. Follow-up

Patients with MM should undergo a regular clinical follow-up in order to detect as early as possible recurrences and the development of new primary MM (which occur in 10% of patients). Follow-up should also provide psychosocial support and education on self-examination of the patient and his close family members. The extent and frequency of follow-up examinations depend on the risk of each patient. Since LM has no risk of metastasis, some authors advocate no follow-up, apart from a visit after complete excision to check for new primary MM, and to teach self-examination for a new MM. Apart from patient self-examination, however, the majority of guidelines recommend dermatological examination every 6 months for 3 years, and thereafter once yearly for life (or at least during 10 years for the detection of new MM or other skin cancers).

Acknowledgements

Not applicable.

Funding

No funding was received.

Availability of data and materials

Not applicable.

Author's contribution

JK conceived and designed the study, performed the literature search, and wrote and revised the manuscript. The author has read and approved the final manuscript.

Ethics approval and consent to participate

The clinical image depicted in Fig. 1 and the pathological image depicted in Fig. 2 were obtained several years ago at the Department of Dermatology, Ed. Herriot Hospital, Lyon, France, and are part of a medical anonymous sample collection. They are only depicted herein retrospectively as examples of the disease and its pathology.

Patient consent for publication

Not applicable.

Competing interests

The author declares that he has no competing interests.

References