Malignant melanoma (MM) is the most common cause of death from skin cancer in Caucasians. The incidence of MM has considerably increased in most countries in the recent decades. However, the incidence rate of melanoma demonstrates a racial difference. The incidence rates range between 20 and 40 per 100,000 people in Australia and United States each year [1, 2]. In contrast, the incidence in Asians is clearly lower, approximately one per 100,000 [3, 4]. However, the most common clinical subtype of melanoma in Asians is acral lentiginous melanoma (ALM), which occurs at a rate of approximately 40%–65% of all cutaneous melanomas [4, 5] compared with only 2%–3% in Caucasians . This subtype of melanoma, first described by Reed in 1976, is characterized by its predilection for the acral regions such as the soles, palms, and nail apparatus and by a pattern of radial lentiginous growth phase that evolves over months or years without any solar elastosis to a dermal (vertical) invasive stage. This subgroup has been designated as “plantar lentiginous melanoma,” which corresponds to ALM . Thereafter, MM was classified into the following four subtypes by Clark et al. in 1986 according to histological features; nodular melanoma (NM), superficial spreading melanoma (SSM), lentigo maligna melanoma (LMM), and ALM .
The prognosis of ALM is generally considered to be poorer than other subtypes such as SMM and LMM. In particular, the lesions of the sole are often overlooked by patients. In addition, ALM has a likelihood of being misdiagnosed as a benign melanocytic nevus, which leads to the development of tumor and delay in treatment . This problem has improved with the advent of dermoscopy, which enables the dermatologists to distinguish between the early stages of ALM and benign melanocytic nevus.
Acral sites such as the palms and soles are not sun-exposured areas, and therefore overexposure to ultraviolet light has not been shown to be a risk factor for the development of ALM . On the other hand, ultraviolet light can play an important role in the development of LMM . A recent study showed that 81% of non-chronic sun-damaged melanomas such as SSM had a high likelihood of BRAF or N-RAS mutations and the other subtypes, including ALM, had a low rate of mutation in either gene. In contrast, ALM had a frequent mutation or amplification of the KIT gene .
Because of the biological and unique anatomical specificity of ALM, there are some controversies in the treatment of ALM. In this chapter, we review the published articles regarding the surgical treatment of primary ALM lesions, including our paper and experiences, assess the current role of surgery, and discuss several major controversies of ALM surgical treatment.
2. Problems of excision margins in ALM
The World Health Organization Melanoma Program undertook a randomized trial and compared lateral margins of 1 and 3 cm for 612 melanoma patients with thicknesses of <2 mm . Disease-free and overall survival rates did not differ between the two groups. Two subsequent trials in Europe compared the results of treating melanomas with lateral margins of 2 or 5 cm. The Scandinavian Melanoma Group Study and French Cooperative Group Trial compared lateral margins of 2 cm with 5 cm for patients with 0.8–2.0 mm-thick and <2.1 mm-thick melanomas, respectively [14, 15]. Neither study showed any evidence that lateral margins of 5 cm reduced the local recurrence rate or improved survival rates. The Intergroup Melanoma Surgical Trial also reported the results of a randomized prospective trial that compared lateral margins of 2 cm with 4 cm for 740 melanoma patients with thicknesses of 1.01–4.0 mm . This trial also demonstrated that the local recurrence and survival rates were similar for the two groups. This evidence suggests that lateral margins of at least 2 cm are suitable for patients with melanoma with thicknesses of >2 mm.
Although some uncertainties remain and further trial-based evidence is required for clarification, the general consensus is that at least margins of 1 cm should be adequate for melanomas with a thickness of ≤1 mm . For 1.01–2mm-thick tumors, some evidence suggests that lateral margins > 1 cm are desirable . As for in situ lesions, a lateral margin of 5 mm is generally recommended in some national guidelines, including Japan. However, a recent prospective, comparative study demonstrated that 86% of 1120 in situ lesions were successfully excised with lateral margins of 6 mm and 98.9% were successful with lateral margins of 9 mm and concluded that lateral margins of 9 mm were appropriate for in situ lesions .
The appropriate depth of excision for melanoma is also still controversial. The depth of excision has been recommended to be at least to the level of muscle fascia; deeper excisions have not been shown to improve outcomes [20, 21]. On the other hand, a recent study demonstrated that there was no advantage for a resection of the deep muscular fascia but a potential for an increased risk of intralymphatic recurrences .
Based on the evidence described above, current major melanoma guidelines such as the National Comprehensive Cancer Network Guideline recommends adequate lateral margins alone, depending on the maximum thickness of the tumor. However, most of the available evidence has been obtained from clinical trials that included melanomas that were located mainly on the trunk or proximal extremities. Most studies and trials have not included melanomas in the head and neck region and on distal extremities, including ALMs. Acral sites are locations where more complex reconstructive procedures may be required if wider and deeper excision margins are applied. It is unclear whether the application of current recommended excision margins to the treatment of ALM that is biologically different from other subtypes is appropriate.
3. Surgical management of subungual melanoma (SUM)
SUM is a rare manifestation of MM. The incidence of SUM also has a racial difference. It represents approximately 2%–3% and 20% of all cutaneous MMs in Caucasians [23-26] and Asians [4, 27], respectively.
The anatomy of the nail apparatus is complex and is composed of the proximal nail fold, nail matrix, nail bed, and hyponychium. The proximal nail fold is continuous with the dorsal aspect of the nail matrix distally and the extensor aspect of the digital skin proximally. The cuticle separates the ventral portion of the proximal nail fold from the underlying nail plate. The nail plate is the horny end product of the nail matrix. The hyponychium is the distal portion of the nail bed, which represents the site of union between the nail bed and tip of the digits .
3.1. Invasive SUM
In the surgical treatment of SUM, amputation has been traditionally performed . Comparative studies of various amputation levels have not shown any advantages for metacarpal amputations over metacarpophalangeal (MP), proximal interphalangeal (PIP), or distal interphalangeal (DIP) amputations ; this means that the prognosis of SUM does not depend on the amputation level but on the clinical stage . Therefore, the recent trend of surgery is to utilize more distal amputations, without compromising recurrence or survival [25, 26]. In particular, in the thumb, a more distal amputation is preferable (Fig. 1). Although amputation at the interphalangeal joint of the thumb results in only a 10% loss of function of a useful hand, amputation at the MP joint of the thumb results in a 40% loss of function . In contrast, from the viewpoint of the functional and cosmetic aspects, ray amputation may be more appropriate than MP, PIP, or DIP amputation for the treatment of the other fingers (Fig. 2).
3.2. In situ or minimally-invasive SUM
Recently, several authors have proposed conservative surgery, which involves a narrow surgical margin without amputation followed by skin grafting for in situ SUM and minimally-invasive SUM (Fig. 3), which is defined as a Breslow tumor thickness (TT) ≤0.5 mm [33-38]. There have also been a few case reports of two-step surgeries using artificial dermis tissue for the tentative coverage of the defect after local excision, followed by skin graft coverage after the histopathological confirmation of negative surgical margins [39, 40]. These strategies are becoming more common, although the total number of such surgeries still remains small.
3.3. Possibility of conservative surgery for invasive SUM
The unique anatomy of the nail apparatus described above also demonstrates that there is scant soft tissue between the nail unit and bony surface of the distal phalanx. Haneke  studied the distance from the tip of the nail matrix to the extensor tendon insertion of the middle finger of a young male, and the measured distance was 0.8 mm. Kim et al.  also studied the shortest distances between the deepest base of the nail matrix and surface of the distal phalanx in cadavers. The mean distances were 0.90 mm for thumbs, 0.72 mm for fifth fingers, 0.87 mm for first toes, and 1.09 mm for fifth toes. The mean distance of all digits was 0.90 mm. These studies have suggested that wide local excision was insufficient for the eradication of invasive SUM with a safe, deep surgical margin and appeared to be sufficient for only in situ or minimally-invasive SUM.
Dermatologists sometimes encounter patients with invasive SUMs that do not invade or attach to the distal phalanx. Therefore, we measured the shortest distances of 30 surgical specimens of invasive SUM between the deepest base of the tumor and the surface of the distal phalanx . There were no bone invasion cases with 4 mm-thick SUMs and the shortest tumor-to-bone distances exceeded 0.9 mm in all specimens with thicknesses of <4 mm (Fig. 4). In statistical analysis, the Pearson chi-square test showed that there was a higher likelihood of bone attachment or invasion when TT exceeded 4 mm, which was statistically significant (P=0.009). Both univariate and multivariate analyses also revealed that thick TT alone had a statistically significant effect [odds ratio 1.807 and 1.865 (95% CI 1.1085–3.008 and 1.111–3.13, P=0.023 and 0.018)]. This study only focused on the histologic evaluation of deep margins and has limitations including the lack of prognostic information; however, there have also been several reports of patients with >0.5-mm-thick SUMs who underwent non-amputative surgeries and who had no local recurrence or metastasis [36, 40, 44].
These studies indicate that the patients with invasive SUMs of intermediate thicknesses may be candidates for non-amputative surgery. However, the main problem is that it is still difficult to evaluate tumor thickness by inspection or dermoscopy preoperatively. There is a possibility of the presence of an unexpectedly thicker area in the tumor than the thickness identified from an incisional biopsy, which can consequently change the surgical strategy. High-resolution ultrasound is a noninvasive examination that can evaluate tumor thickness preoperatively, and there have been several studies that reported its accuracy for the determination of tumor thickness [45-47]. However, these studies excluded SUM; therefore, its accuracy for tumor thickness evaluation in SUM remains unclear. Furthermore, an ultrasound at 20 MHz tends to overestimate thickness because of lymphocytic infiltration around the tumor or nevus remnant [48, 49]. Reflectance confocal microscopy permits the acquisition of dynamic images of the epidermis and papillary dermis with resolution to a cellular level  and has been experimentally used for intraoperative imaging of the nail matrix . However, its limitation is the imaging depth, which cannot be deeper than 200–300 μm .
For other surgical procedures, Chow et al. has reported on wide local excision procedures involving the removal of the nail apparatus with a layer of underlying bone of 1-mm depth running parallel to the nail bed before reconstruction with skin grafting . Moehrle et al. reported on the excision of SUM with the distal part of the distal phalanx (processus unguinatus), which has led to its denotation as a “functional” surgery .
4. Surgical management of ALM on the volar skin
In general, ALMs start as in situ lesions, with brown macules that enlarge slowly and form irregularly pigmented, asymmetric macular lesions over the years, corresponding to the radial growth phase. Thereafter, indurated nodules appear within the macular lesions and sometimes the nodules ulcerate in the so-called vertical growth phase.
4.1. Problems of excision margins
Because of this very slow clinical course, ALMs on the volar skin often contain in situ lesions at the periphery (Fig. 5). Based on the recommended lateral margins described above, the dermatological surgeons are often confused regarding the decision for lateral margins. It is still unclear whether lateral margins of 0.5–1 cm from the peripheral border of the lesion should be suitable as the peripheral lesion is considered an in situ lesion, or lateral margins of 2 cm from the periphery of the lesion should be selected as the entire tumor itself is regarded as an invasive lesion.
4.2. ALM of the palm
In general, most ALMs on the palm are diagnosed at an early stage because the palm is the site where the patient easily notices the lesion at an earlier stage. Therefore, wide local excision is easily accomplished and the full-or split-thickness skin graft is adequate for the coverage of the surgical defect. The main point that the surgeon should pay attention to is avoiding intraoperative injuries of the palmar digital nerves, which are easily exposed during tumor excision because of the scant subcutaneous fatty tissue in some areas of the palm (Fig. 6). As for skin grafting, the skin obtained from non-weight-bearing areas of the foot is preferable because of the similarity to the quality of the palmar skin; however, the skin quality difference between the palmar skin and the skin obtained from other areas may be inconspicuous in elderly patients (Fig. 6).
4.3. ALM of the sole
ALM of the sole has been traditionally difficult to find and diagnose because of its “out-of sight” location. The lesion often precludes primary closure because of the lack of mobility of the skin on the sole and horizontal-growth of the tumor, which increases the reconstructive complexity. Therefore, after wide local excision, substantial defects have to be repaired using various methods such as skin grafting, local and distant flaps, and secondary intention healing depending on the size and site of the defect and the medical condition and lifestyle of the patient . Unlike the palm of the hands, reconstruction of the sole of the foot is difficult because of its anatomy. The sole has weight-bearing areas such as the heel. In addition, a defect on the foot may result in insufficient vascular flow .
4.3.1. Split-thickness or full-thickness skin graft
Although split-thickness skin grafting is considered to be adequate for non-weight-bearing areas, it has been considered to be inappropriate for the heel and distal plantar area and could lead to dismal results . If the surgical defect is located on the non-weight-bearing areas such as the foot and distal plantar area, or if the patient has severe medical comorbidities or is less active, the defect may be covered by split-or full-thickness skin grafting. Although skin grafting has advantages such as being much simpler than flap reconstruction, it is considered that defects of the weight-bearing areas should be repaired using local or distant flaps in ambulatory patients.
By authors’ experience, the indication for skin grafting on the heel depends on the amount of the remaining subcutaneous fat pad after tumor excision. If the primary tumor is a thinner lesion and there is no need to excise the enormous subcutaneous fat pad together with the tumor, both split-or full-thickness skin grafts may be acceptable for weight-bearing reconstruction (Fig. 7A). In contrast, neither split-nor full-thickness skin grafting is appropriate when wide local excision including the subcutaneous fat beneath the tumor is required for oncologic purposes, which may lead to exposure of the calcaneum. Insufficient cushion because of the paucity of the fat pad eventually produces an erosion or ulceration of the skin graft (Fig. 7B).
As for the distal plantar area, the weight pressure on this area is not as great as on the heel. Therefore, we think that skin grafting for this area is more acceptable than that for the heel. Even through a large amount of the subcutaneous fat may be dissected with the tumor, full-thickness skin grafting may be durable for weight pressure over a long-term postoperative course (Fig. 7C). When there is remaining subcutaneous fat, reconstruction using the split-thickness skin grafting is also acceptable (Fig. 7D).
Skin grafts are generally taken from the contralateral limb because the melanoma cells were thought to primarily metastasize via lymphatic routes . In contrast, a recent study has shown that there was no difference in the rates of donor site recurrence between the ipsilateral and contralateral limbs .
4.3.2. Local or distant flaps
A flap reconstruction is usually recommended for the closure of weight-bearing areas of the feet. The coverage of weight-bearing areas provides well-padded tissue superior to skin grafting. Among the various flaps, the medial plantar flap is the optimal reconstructive procedure. The plantar flap can provide the same quality of skin cover for weight-bearing areas and can also provide some sensation. In particular, using heel reconstruction, the donor site is located distally and away from the lymphatics of the primary site (Fig. 8).
The distally-based sural artery neurocutaneous flap is an optional procedure for heel reconstruction. Because the donor site is the posterior aspect of the lower leg, the quality of the skin is different from that of the sole. A comparative study between the medial plantar flap and distally-based sural artery flap demonstrated that the postoperative complications were higher in patients who underwent sural artery neurocutaneous flap procedures .
4.3.3. Secondary intention healing
Secondary intention healing requires a long-term healing period, prolonged care such as regular and frequent dressing changes, and careful observation. According to previous reports, it took approximately 12–18 weeks to close the defects (mean defect size: 32.6–36.5 cm2) [54, 59]. Despite these drawbacks, it has the advantages of avoiding a secondary wound for tissue harvesting, a smaller scar because of the natural contraction of the wound, and granulation tissue that acts as a cushion to absorb impact while walking . Recently, the effect of negative pressure wound therapy in addition to secondary intention healing has been investigated . Compared with secondary intention healing alone, there was no difference in time to complete wound healing. However, the vascularity score and height of the scars was significantly better, and no wound infections during the course of treatment had occurred in the former group.
These methods clearly require a more extended period of healing time than skin grafting or flaps. Therefore, it would be appropriate to apply these methods to small-to medium-sized surgical defects.
Increased understanding of the molecular biology and pathogenesis of melanoma may lead to the development of novel therapeutic agents and treatment plans for melanoma. However, surgery is still the mainstay of treatment as there are no proven effective adjuvant systemic treatments.
Although several reliable national guidelines for melanoma have been produced using the evidence-based method, there are still some controversies in the treatment of ALM. Because of the biological and unique anatomical specificity of ALM, It is still unclear whether the thickness of ALM and recommended lateral margins are the same as that with other subtypes. Large randomized prospective studies with long-term follow up are necessary to fully evaluate surgery for ALM in the future. The accumulation of evidence produced by such studies will lead to the development of novel ALM treatments in the melanoma guidelines.
This work was partly supported by the National Cancer Center Research and Development Fund (26-A-4).