Keywords

Introduction

Lentigo maligna (LM) and lentigo maligna melanoma (LMM) are diagnostically and therapeutically challenging due to their diverse presentations, often with poorly defined, irregular borders on a background of photodamaged skin. Moreover, these lesions often occur on cosmetically-sensitive areas of the head and neck. As such, “blind” mapping biopsies for diagnosis, or potentially disfiguring surgical excision as treatment, may not be welcomed [1–3]. Patients may have had prior treatment that further obscures clinical borders, or the lesion may be multiply recurrent due to inadequate excision or failure of nonsurgical therapy. Initial biopsies diagnostic for LM may miss areas of occult invasion, due to sampling error. In all of these cases, there is a critical need for better visualization and evaluation of the lesion pre-treatment to inform the patient and physician, and to guide optimal therapy. There is a need for novel imaging modalities to better visualize these lesions for diagnosis and improve monitoring for recurrence after both surgical and non-surgical treatment. Advances in newer imaging technologies, including reflectance confocal microscopy (RCM), have improved our ability to better diagnose and manage these challenging lesions. Herein, this chapter will discuss the advantages and limitations of confocal microscopy for LM and review other emerging, non-invasive tools employed to manage cases of LM.

Reflectance Confocal Microscopy

RCM is a non-invasive imaging tool that uses a low-power laser system to provide real-time imaging of the epidermis and superficial papillary dermis with cellular-level resolution, and has been demonstrated to improve diagnostic accuracy for melanocytic lesions [4–7]. Imaging can be readily obtained at a defined depth (up to 200 μm) and captured in an en face orientation.

At any chosen depth (z plane), a two dimensional sequence or matrix of neighboring images can be captured and then stitched into a mosaic to display extended areas of skin (x-y plane). Thus, capturing vertical stacks of images combined with multiple mosaics at different depths can allow for 3D approximation of lesion margins [8, 9].

Studies have demonstrated the effectiveness of RCM for delineating surgical margins, assessing physiologic responses to therapy, and evaluating the response to non-surgical treatments in vivo for LM [1, 3, 10–13]. One major advantage of this non-invasive imaging modality is the ability to repeat studies on the same area of skin over time. This opens the door for longitudinal studies of cutaneous responses to nonsurgical therapies and the ability to non-invasively monitor for recurrence in vivo.

RCM for LM Diagnosis

Traditionally, non-invasive tools for assessment of LM have included dermoscopy and Wood’s light. These instruments have been utilized to assist in the clinical examination of LM and better define the extent of the lesion. However, our ability to reliably discriminate LM from benign pigmented lesions can be challenging, especially in the context of recurrent disease or background of significantly photodamaged skin; therefore, histopathology remains the gold standard in diagnosis [14–16]. Because LM may be large and tends to occur on cosmetically and functionally sensitive areas, biopsies for diagnosis of LM or demarcation of margins may add morbidity for patients in the form of pain, infection, and scarring. It is important to note that even with an adequate tissue biopsy, sampling error or a high degree of background melanocytic hyperplasia may result in imprecise histopathologic interpretation of suspicious lesions. Ultimately, improved technologies and techniques are warranted to augment existing methods for pigmented lesion analysis.

Reflectance confocal microscopy (RCM) provides real-time, non-invasive imaging of intact skin at a resolution comparable to conventional histology. Several studies have demonstrated that RCM may improve diagnostic accuracy of melanoma compared to dermoscopy or Wood’s light examination [5, 7, 17–19]. RCM correlates of dermoscopic findings have also been shown to be helpful in distinguishing LM from pigmented nonmelanocytic macules [20].

Guitera et al. looked at various RCM features identified in a sample of clinically equivocal macules of the face, of which 81 were LM and 203 were benign [4]. From this study, they determined features suggestive of LM and created a scoring algorithm to distinguish LM from benign macules. The two major features (scoring +2 points each) were nonedged papillae (Fig. 14.1) and round, large pagetoid cells (Fig. 14.2) greater than 20 um. The three minor features (+1 point) included: atypical cells at the dermoepidermal junction (three or more found in five 0.5 × 0.5 mm2 fields) as seen in Fig. 14.3, follicular localization of atypical cells (Fig. 14.4), and nucleated cells within the dermal papillae. A broadened honeycomb pattern was less indicative of LM, and scored −1 point. Overall, an LM score of 2 or greater yielded a sensitivity of 85 % and specificity of 76 % for the diagnosis of LM (Table 14.1). Other studies have found comparable features including nests of atypical melanocytes surrounding and/or infiltrating adnexal structures, sheets of dendritic melanocytes, and cord-like rete ridges at the dermoepidermal junction to be suggestive of facial LM/LMM [21]. Similar features have held true in the case of amelanotic LM as well [22].

Fig. 14.1
figure 1

RCM feature of LM: nonedged dermal papillae

Full size image
Fig. 14.2
figure 2

RCM feature of LM: round large pagetoid cells

Full size image
Fig. 14.3
figure 3

RCM feature of LM: atypical cells at the dermal-epidermal junction

Full size image
Fig. 14.4
figure 4

RCM feature of LM: follicular localization of atypical cells

Full size image
Table 14.1 RCM score for diagnosis of lentigo maligna
Full size table

Rossi et al. studied the use of a handheld confocal microscope and compared it to histopathology in the diagnosis of 60 equivocal pigmented lesions in patients concerning for LM. In this study, RCM and histopathology interpretations were concordant in 89 % of cases (56/63). While there were no false-negative outcomes on RCM, 7 false-positive results were seen, a majority being diagnosed on histopathology as pigmented actinic keratotis. Features suggestive of LM in the false-positive group include the presence of numerous hyperreflectile large cells at the dermoepidermal junction and follicular localization of these cells [23].

A recent meta-analysis found a sensitivity and specificity of 93 % and 76 %, respectively, for RCM when used as a second-level test for diagnosing pigmented lesions that are clinically equivocal [24]. Others have also reported sensitivities of 100 % using RCM to detect LM, further supporting the idea that RCM is a reliable method for diagnosing LM or monitoring for treatment failure in vivo [12, 14]. Using RCM to non-invasively identify LM without biopsy is an exciting improvement in the management of patients with chronically sun-exposed skin.

Another important function of RCM is to improve the ability to hone in on optimal areas for mapping biopsies and detect possible occult invasion in LM lesions. Blind mapping biopsies of LM are prone to sample bias and depend greatly on biopsy technique. Even adequate biopsies of LM can be challenging to definitively interpret under standard hematoxylin and eosin histology due to its occurrence in areas with a background of melanocytic hyperplasia. Studies have demonstrated that occult invasion in LM with standard biopsy technique was not consistently apparent until complete surgical excision was performed. For example, Agarwal-Antal et al. reported on 92 cases of LM of which 16 % were found to have unsuspected invasion on final excisional pathology [25]. Due to the cosmetically-sensitive nature of the lesions, physicians may feel discouraged to take numerous mapping biopsies, even in cases of large lesions. This makes it quite difficult to adequately evaluate the breadth of the lesion or detect occult invasion. Moreover, biopsies are subject to sampling error due to the heterogeneous nature of LM and its characteristic subclinical extension. The costs and morbidity associated with multiple biopsies in patients with a high burden of actinic disease can be substantial. Utilizing real-time video imaging of the dermoepidermal junction at the margin and within the lesion has allowed for the detection of deep atypical melanocytes suspicious for invasion to better hone in on suspicious areas and guide mapping biopsies. Being able to detect the relative depth of invasion pre-treatment through RCM imaging or by guiding mapping biopsies is essential for not only counseling the patient about disease risk but also imperative for choosing an appropriate treatment modality.

RCM for LM Management (Surgical)

Surgery is considered the first line treatment for LM; however, it is not without associated morbidity. Wide surgical margins, especially on cosmetically-sensitive areas such as the face, are not always possible to obtain, and become further complicated when trying to maintain adequate functional and aesthetic outcomes. The margins required for surgical clearance may not be straightforward for facial lesions. A study by Hazan et al. reviewed 117 cases of LM and LMM and found that the total surgical margin required for excision of LM was 7.1 mm and for LMM was 10.3 mm. Moreover, of the tumors that were initially diagnosed as LM on biopsy, 16 % were found to have unsuspected invasion [26].

As surgical excision remains the standard of care for LM, it is important to optimize surgical methods and because there may be extensive subclinical extension, there is a need for better pre-treatment margin evaluation in LM. RCM is emerging as an adjunct to existing technologies, including dermoscopy and Wood’s lamp, to better delineate borders. Utilizing RCM pre-surgically offers the benefit of surgical planning, as it helps define the extent of subclinical spread prior to initiating the surgery. This informs both the surgeon and the patient to assist in reconstructive design and patient expectations. While RCM may be used to show that margins need to be increased due to subclinical spread, it may also allow for confirming narrower surgical margins in critical anatomical areas, facilitating reconstruction and decreasing patient morbidity. Thus, RCM provides valuable clinical information to potentially guide surgical management, and may lead to favorable cosmetic outcomes and a better prognosis.

One approach to using RCM to guide surgical management of LM is to first demarcate the lesion clinically with the aid of Wood’s lamp and dermoscopy, followed by placing appropriate surgical margins at 5–10 mm depending on clinical and histologic criteria. RCM may then be used within the lesion to identify features of the melanoma, thus serving as a control. An imaging “map” (Fig. 14.5) may be made by dividing the lesion into quadrants and capturing RCM video imaging along the periphery of surgical margins of each quadrant at the level of the dermoepidermal junction (main region to detect features of LM and LMM). In areas where positive findings including hyperreflective dendritic cells, large, round pagetoid cells, and epidermal disarray are seen, the margins are extended out radially. Video capture can be used to recreate video mosaics by stitching together sequences of images captured to re-create a larger field of view. As such, RCM is a valuable adjunct to the clinical exam and dermoscopy to determine clinical margins and define the gross tumor volume.

Fig. 14.5
figure 5

RCM “map” created to delineate surgical margins of LM

Full size image

Advancements have been made in RCM technology, overcoming limitations of earlier iterations of the device. The newer, handheld Vivascope 3000 (Caliber ID, Rochester, NY) offers the advantage of real-time assessment in areas that may not have been amenable to previous versions of the device. Employing RCM during the initial consultation may help clinicians characterize subclinical spread of LM and therefore better counsel patients about the extent of their lesion. Additionally, Hibler et al. described the use of the handheld Vivascope 3000 intraoperatively to provide the surgeon with real-time assessment of tumor margins in vivo [27]. This may be a valuable approach for large cases of LM being performed in the operating room under general anesthesia, where the benefits of obtaining immediate visual confirmation of margins to ensure clearance may prevent a return trip to the operating room, saving costs and avoiding risks of additional anesthesia. Using RCM in this mapping fashion could ultimately allow for improved clearance of LM, thereby decreasing the likelihood of recurrence and the need for re-excision, while also maximizing tissue conservation and lowering morbidity.

RCM for LM Management (Non-surgical)

While surgical excision is the treatment of choice for LM, factors including advanced patient age, multiple comorbidities, large lesion size in functionally or aesthetically-sensitive areas, and indiscriminate borders on photodamaged skin may make surgical excision complicated or not a feasible option. For patients unable to pursue surgical treatment and in cases where surgery would cause excess morbidity or deformity, multiple nonsurgical treatment options have been pursued. The use of superficial radiation or off label use of topical therapies, i.e. Imiquimod, has been reported in the literature as alternative non-surgical treatment options [28, 29]. However, the lack of histological confirmation, and possibility for undetected invasive spread have been limits to these modalities. Similarly, close monitoring for disease recurrence and progression is of utmost importance. Typically this is carried out by clinical examination, without adjunctive imaging beyond dermoscopy. RCM is emerging as an imaging technology that is proving useful to aid in the assessment of disease extent, treatment response and disease recurrence for LM after non-surgical therapy [1]. This is illustrated in Fig. 14.6.

Fig. 14.6
figure 6

RCM used to detect recurrent LM. (a) Pre-confocal mapping of brown pigmentation along scar from excision of invasive lentigo maligna melanoma 15 years prior. ‘V’ indicates sites where images were captured in the z-plane. (+) indicates features of lentigo maligna on confocal microscopy. (b) Reflectance confocal image: Yellow circles indicate suspicious features for lentigo maligna: hyper reflective dendritic cells surrounding hair follicles. (c) Shave biopsies (blue circles) guided by confocal all showed melanoma in situ and patient opted for treatment with Imiquimod to avoid surgical morbidity

Full size image

In the same way that RCM may provide enhanced delineation of lesion margins for surgical intervention, it may also be capable of better defining a treatment field for radiation or topical therapies (Fig. 14.7). LMs treated with radiation or non surgical treatment modalities need close follow-up to detect recurrences [28]. Detecting recurrence can be a challenge clinically, as the lesion may recur as an amelanotic lesion, or can be further obscured by radiation-induced inflammation and post-radiation pigment changes. Because RCM allows for the same area of skin to be re-examined over time, this technology can also be applied to monitor for recurrence in LMs [30]. Changes in tissue architecture have been observed in LMs after radiation, including: superficial necrosis and apoptotic cells, dilated vessels, and increased inflammatory cells in both the dermis and epidermis [10]. After radiation, LM-specific large pagetoid cells were decreased or even resolved in the epidermis, dermal-epidermal junction, and in the follicles [10]. When using RCM to monitor for recurrence post-treatment, it is important to wait until the inflammation and post treatment changes have subsided to ensure any acute radiation-induced changes in skin architecture have resolved and will not cause false positives [31]. Epidermal regeneration post-radiation therapy begins 3–5 weeks after treatment and heals within 1–3 months, suggesting that radiation-induced changes on RCM might persist for this duration of time, although this has not been formally studied [32]. The ability to visualize and define changes during and after RT suggest RCM may be useful for monitoring for treatment failure. Examination with RCM may augment our ability to better define the radiation field pre-treatment and has been shown to be capable of detecting areas concerning for residual or recurrent disease post-treatment before clinical repigmentation [33].

Fig. 14.7
figure 7

RCM used to plan radiation treatment margins. (a) Initial lesion with irregular pigmentation confirmed as lentigo maligna on biopsy. Patient elected treatment with Imiquimod due to her advancing age and medical comorbidities. Follow-up biopsies found melanoma invasive to 0.37 mm and patient underwent surgery to excise the invasive melanoma but in situ LM remained at surgical margins. (b) Reflectance confocal mapping for radiation therapy planning. Yellow circles indicate areas of dendritic pagetoid hyper reflecticle cells suspicious for lentigo maligna. (c) RCM map at 1 cm and 2 cm margins from surgical scar created to guide further radiation planning. ‘v’ indicates stacks of images captured in the z-plane. (+) indicates findings suspicious for lentigo maligna

Full size image

In a similar manner, RCM may be utilized to monitor response after treatment with off label Imiquimod cream [34]. While the use of Imiquimod for LM has been well documented in the literature, the application, duration of therapy, and response to treatment vary greatly. Furthermore, factors accurately predicting a positive response to treatment have yet to be fully elucidated, as the degree of inflammatory response and erythema have not correlated well with overall clearance. The benefit of RCM after topical therapy is that it represents a non-invasive modality to monitor response to treatment and may help assess the need for increased duration of treatment. Moreover, similar to the changes induced post-radiation, treatment with Imiquimod may cause an alteration of the clinically apparent pigment, and it is therefore difficult to assess treatment success by clinical inspection alone. The use of RCM before, during, and after treatment provides a longitudinal assessment of the lesion, and may augment our ability to determine treatment success or failure.

RCM Limitations and Future Directions

As outlined above, RCM is a non-invasive technology with the potential to significantly augment our ability to counsel and treat patients regarding their skin cancer diagnoses, management, and expected outcome. Yet, a number of limitations of this technology currently exist, including the time needed to image, limited depth of imaging, technology access and cost, and associated learning curve. The field of view for RCM is limited, so for larger lesions it may take time to assess the entirety of the lesion. The advent of video mosaicing and the handheld RCM has improved upon the time required to assess lesions [9], yet it may still be time consuming in the case of large lesions. Moreover, the restricted depth of imaging (~200 μm) restricts evaluation of the dermis to the superficial papillary dermis. Additionally, widespread adoption of this technology is limited by its high cost relative to dermoscopy and associated learning curve [35].

There is a learning curve associated with RCM imaging; however, the training required for accurate RCM interpretation has been reported to be less than that of dermoscopy [36]. Importantly, studies have shown that key RCM diagnostic criteria for lesions including melanoma and basal cell carcinoma are reproducibly recognized among RCM users, and that diagnostic accuracy increases with experience [37]. Although more onerous and time-consuming than dermoscopy, RCM provides detailed images of live tissue with cellular-level resolution and can reconstruct 3D areas for evaluation, critical for assessing heterogeneous lesions such as LM with poorly defined borders that may have significant subclinical extension. Due to these limitations, the use of RCM is highly individualized depending on the size and nature of the lesion, its location, and patient comorbidities.

Handheld and Video Mosaicing

The handheld Vivascope 3000 overcomes limitations of the stationary Vivascope 1500 device, and offers advantages such as being able to assess lesions in difficult locations on the face [38, 39]. Compared to previous non-handheld RCM devices the use of the HRCM does not need to attach a ring to the skin and is less bulky. This permits its use at the bedside of the patient or even intraoperatively [27]. Furthermore, the ability to create video mosaics overcomes the limited field of view provided by standard RCM imaging, and allows for rapid and accurate assessment of large lesions in real time [9]. This may permit complete examination of the periphery of lesions, critical for evaluation of subclinical extension of LM and verifying clearance after surgery. Indeed, studies have found good correlation between handheld RCM findings and histological findings after surgery for LM/LMM, suggesting that it is a valuable technique to guide surgical excision [40]. Handheld RCM is a noteworthy ancillary tool as it can be readily performed at the bedside of the patient or even intraoperatively, and may represent a faster approach than conventional RCM in cases where large areas need to be mapped.

Other Non-invasive Tools for LM

Apart from RCM, there are other non-invasive imaging modalities that have been applied for melanoma detection. Depending on the imaging technology, the practical in-vivo use of such compared to RCM may be limited. Technologies such as optical coherence tomography and ultrasound have been applied to melanoma diagnosis and while they are able to penetrate deeper than RCM, the level of resolution is macroscopic compared to RCM’s cellular resolution. Therefore reliable diagnostic criteria have not been fully elucidated in regards to melanoma diagnosis.

Devices that are based on multispectral imaging (MSI) have also become available for the diagnosis of melanoma. MSI works through using multiple wavelengths, ranging from 400 to 1000 nm, to enhance detection of dermoscopic features within the lesion. Different skin chromophores absorb and reflect different wavelengths to create an “image” which is then analyzed algorithmically [41]. Spectrophotometric intracutaneous analysis (SIAscopy), is one of two types of MSI devices applied for the detection of melanoma, however, there have been multiple studies that reveal differing results in the ability to detect melanoma in vivo [42, 43].

Smartphone (Melanoma in General)

The mobile market has experienced a rapid expansion in the number of dermatological applications marketed to educate individuals and monitor lesions. There are over 200 dermatology-related mobile applications, with the most common being general dermatology references, self-surveillance/diagnosis tools, disease guides, and educational aids [44]. Most of these applications (51 %) are targeted towards patients, while 41 % of applications are targeted towards medical professionals, and 8 % target both. Consumers are able to rapidly access and use mobile applications. Moreover, the mobile market makes it possible to reach more remote locations with these educational resources and diagnostic aids.

While this technology is widely distributed, few studies have evaluated the accuracy of smartphone applications. Of major concern is that diagnostic inaccuracy may result in delayed treatment due to false reassurance that a lesion is benign. For example, 3 of 4 smartphone applications incorrectly identified at least 30 % of melanomas as “unconcerning,” and the sensitivity of such applications ranged from 6.8 to 98.1 %, highlighting the drastic variability among current applications [45]. A recent review of dermatology-related mobile applications found that none of the applications that provided risk-assessment of lesions appeared to have been validated for diagnostic accuracy, and there was limited information regarding the credentials of those involved with making the application—some applications were not updated in over 3 years [46]. As such, creators may make unsubstantiated assertions in order to influence users to download and use their application, and have even been fined by the Federal Trade Commission for unproven claims [47]. Therefore, regular appraisal of dermatology-related mobile applications may be warranted to objectively review the spectrum of applications available and to make recommendations.

Conclusion

LM and LMM present diagnostic and therapeutic challenges due to the heterogeneous nature of the lesions, occurrence on cosmetically and anatomically sensitive areas, and indistinct clinical margins. As such, the need for non-invasive devices to detect and diagnose LM is clear. While many different technologies have been applied to this task, RCM has had the most promising results thus far for real time in vivo use. RCM has been utilized to diagnose challenging lesions, “map” out subclinical margins, and detect recurrence of LM. With the advent of newer technologies, improved laser/light optics, and enhanced algorithmic capacities, there will continue to be much progress in this arena.