Abstract
Background
Lentigo maligna (LM), a form of melanoma in situ, has no risk of causing metastasis unless dermal invasive melanoma (LMM) supervenes. Furthermore, the detection of invasion impacts prognosis and management.
Objective
To assess the accuracy of RCM for the detection of invasion component on LM/LMM lesions.
Methods
In the initial case–control study, the performance of one expert in detecting LMM at the time of initial RCM assessment of LM/LMM lesions was recorded prospectively (n = 229). The cases were assessed on RCM‐histopathology correlation sessions and a panel with nine RCM features was proposed to identify LMM, which was subsequently tested in a subset of initial cohort (n = 93) in the matched case–control study by two blinded observers. Univariable and multivariable logistic regression models were performed to evaluate RCM features predictive of LMM. Reproducibility of assessment of the nine RCM features was also evaluated.
Results
A total of 229 LM/LMM cases evaluated by histopathology were assessed blindly and prospectively by an expert confocalist. On histopathology, 210 were LM and 19 were LMM cases. Correct identification of an invasive component was achieved for 17 of 19 LMM cases (89%) and the absence of a dermal component was correctly diagnosed in 190 of 210 LM cases (90%). In the matched case–control (LMM n = 35, LM n = 58), epidermal and junctional disarray, large size of melanocytes and nests of melanocytes were independent predictors of LMM on multivariate analysis. The interobserver analysis demonstrated that these three features had a fair reproducibility between the two investigators (K = 0.4). The multivariable model including those three features showed a high predictive performance AUC = 74% (CI 95% 64–85%), with sensitivity of 63% (95% CI 52–78%) and specificity of 79% (CI 95% 74–88%), and likelihood ratio of 18 (p‐value 0.0026).
Conclusion
Three RCM features were predictive for identifying invasive melanoma in the background of LM.
INTRODUCTION
Lentigo maligna (LM) is a subtype of melanoma in situ where the malignant cells are confined to the epidermis and usually occurs in the context of chronic sun‐damaged skin. 1 LM represents the most prevalent form of melanoma in situ in Australia with a rising rapidly incidence. 2 LM can be a precursor lesion to invasive melanoma, lentigo maligna melanoma (LMM), with the latter estimated to occur in up to 3.5% of cases per year. 3
In many cases, LM is not a straightforward clinical or pathological diagnosis. It can vary as a spectrum from an early precursor stage that is often classified as atypical melanocytic hyperplasia (AMH) to fully developed LM to a dermal invasive stage (LMM). 4 , 5 , 6 Only the latter stage is capable of causing metastasis and death. LMM is managed with surgery with wide margins while LM may be managed with narrower margins. When complex, LM may require a more conservative approach and non‐surgical options may be considered. When non‐surgical management is being considered and complete excision has not been obtained with initial surgical treatment, excluding the possibility of dermal invasion is even more important since differentiation between LM and LMM is crucial in order to establish the most appropriate management.
Clinical and dermoscopic aspects are not always reliable to differentiate the non‐invasive (LM) from the invasive component (LMM). Reflectance confocal microscopy (RCM) has been used to differentiate between malignant and benign melanocytic lesions, to delineate the margins of the melanocytic lesions and to detect recurrence. 7 , 8 In this study, we sought to evaluate the diagnostic accuracy of RCM for detecting invasion in LM type lesions.
MATERIALS AND METHODS
Study design
The hypothesis of this study was that RCM could detect the presence of focal dermal invasion/invasive melanoma within LM/LMM lesions and it was established before the data collection.
This study consisted of two phases: a prospective study performed on a cohort of 229 cases and a matched case–control study performed on a subset of this cohort composed by 93 cases to test the RCM features established by the first phase of the study (Figure 2b).
FIGURE 2.
(a) Flow chart of the study population with prospectively collected data analyzed by single reader (P.G.). (b) Study population for the matched case–control study analyzed by two blinded readers (B.G. and C.S.)
The case–control cohort included LM/LMM lesions recruited from 2005 to 2014 and aimed to evaluate the performance of a single reader (P.G.) for detecting invasion at the time of initial diagnosis prospectively. The confocal expert performed confocal assessment and targeted a biopsy guided by confocal on the areas suspicious for invasion. All cases suspicious for LMM, regardless of the previous histopathological diagnosis obtained by the referring doctor, were targeted‐biopsied with confocal guidance. All the targeted biopsies were performed at the same time as the confocal assessment. Every case had the histopathology reviewed for precise RCM‐histopathology correlation and establishment of the relevant features to differentiate the invasive from the in‐situ component.
All LMM cases were treated with wide local excision as standard of care and these specimens have been reviewed by experienced pathologists (RV, PF, RAS) for this study. The LM cases were treated with surgical and non‐surgical options and there were regularly followed up for 7 years to check for recurrence or disease progression.
The matched case–control study involved retrospective assessment of a subset of the first cohort enriched in new LMM cases. Nine RCM features established during the RCM‐pathology correlation sessions were scored independently by two other blinded investigators (BMG and CS). Reproducibility and inter‐observer agreement were also evaluated.
Participants and lesions
From 2005 to 2014, patients with lesions suspicious of LM/LMM, referred to two Australian tertiary referral centres (Sydney Melanoma Diagnostic Centre and Melanoma Institute Australia) for confocal assessment, who consented to have their data collected, were included in this study.
Confocal assessment was performed prospectively and blinded to pathology. Subsequently, histopathology slides were reviewed for each case of LM/LMM and patients who did not fulfil criteria for LM/LMM were excluded. Some lesions had partial biopsies prior to confocal assessment; however, the confocal expert was blinded from the pathology results at the time of confocal examination.
Nine RCM criteria, established based on histopathology correlation during the first phase of the study, were scored independently by two observers. They were established a priori to allow a quantification of the features seen in pathology and not just presence or absence like in the LM score. These criteria are already described in the literature 9 by our team and are reported in Table 1.
TABLE 1.
RCM features stratified by LMM cases & LM controls.
| Characteristics | Overall (N = 93) | LMM cases (N = 35) | LM controls (N = 58) | p‐value |
|---|---|---|---|---|
| Degree of lentiginous proliferation | ||||
| Scattered | 16 (17.2%) | 4 (11.4%) | 12 (20.7%) | 0.2011 |
| Cluster | 32 (34.4%) | 10 (28.6%) | 22 (37.9%) | |
| Sheet | 45 (48.4%) | 21 (60.0%) | 24 (41.4%) | |
| Pleomorphism | ||||
| Mild | 7 (7.5%) | 2 (5.7%) | 5 (8.6%) | 0.2250 |
| Moderate | 20 (21.5%) | 11 (31.4%) | 9 (15.5%) | |
| Marked | 66 (71.0%) | 22 (62.9%) | 44 (75.9%) | |
| Predominant melanocyte size | ||||
| Small | 4 (4.3%) | 1 (2.9%) | 3 (5.2%) | 0.0124 |
| Moderate | 43 (46.2%) | 10 (28.6%) | 33 (56.9%) | |
| Large | 46 (49.5%) | 24 (68.6%) | 22 (37.9%) | |
| Nests of melanocytes | ||||
| Absent | 44 (47.3%) | 11 (31.4%) | 33 (56.9%) | 0.0158 |
| Present | 28 (30.1%) | 11 (31.4%) | 17 (29.3%) | |
| Florid | 21 (22.6%) | 13 (37.1%) | 8 (13.8%) | |
| Adnexal spread | ||||
| Absent | 45 (48.9%) | 20 (57.1%) | 25 (43.9%) | 0.4335 |
| Present | 27 (29.3%) | 8 (22.9%) | 19 (33.3%) | |
| Florid | 20 (21.7%) | 7 (20.0%) | 13 (22.8%) | |
| Pagetoid spread | ||||
| Absent | 9 (9.7%) | 4 (11.4%) | 5 (8.6%) | 0.9056 |
| Focal | 35 (37.6%) | 13 (37.1%) | 22 (37.9%) | |
| Florid | 49 (52.7%) | 18 (51.4%) | 31 (53.4%) | |
| Epidermal and junctional disarray | ||||
| Absent | 31 (33.3%) | 5 (14.3%) | 26 (44.8%) | 0.0025 |
| Present | 62 (66.7%) | 30 (85.7%) | 32 (55.2%) | |
| Pigment incontinence | ||||
| Absent | 26 (28.3%) | 10 (28.6%) | 16 (28.1%) | 0.9587 |
| Present | 66 (71.7%) | 25 (71.4%) | 41 (71.9%) | |
| Dermal component | ||||
| Absent | 20 (21.7%) | 6 (17.1%) | 14 (24.6%) | 0.1352 |
| Destroyed | 67 (72.8%) | 25 (71.4%) | 42 (73.7%) | |
| Nucleated | 5 (5.4%) | 4 (11.4%) | 1 (1.8%) | |
Note: The bold values indicates statistically significant (p‐value <0.05).
The research protocol was approved with Protocol No X15‐0392 & 2019/ETH06871. Consent was obtained prior to enrollment and all clinical investigations were conducted according to the Declaration of Helsinki Principles.
In vivo RCM evaluation
A commercially available in‐vivo RCM device (Vivascope® 1500, Calibre ID, USA) and a hand‐held probe (Vivascope® 3000, Calibre ID, USA) were used for imaging. A detailed description of these techniques and devices used has previously been published. 10 High‐resolution images from the superficial epidermis, dermal‐epidermal junction, and upper dermis captured as mosaics, stacks, or individual field of views were acquired. All cases were also reviewed by an experienced dermatopathologist to confirm the presence or absence of an invasive component.
Blinded confocal scoring
Two investigators (BG and CZ) were blinded to participant name, age, sex, and clinical and pathological diagnosis and were randomly given the RCM images. Before scoring, the investigators were instructed in the interpretation of RCM images by the senior investigator (PG) using a tutorial with representative images of each individual RCM parameter cited in Table 1. For each patient, images of each skin layer, as well as complete Vivablocks® and Vivastacks®, were blindly assessed by each investigator. Thereby the score was assigned to each individual parameter using a scoring sheet listed in Table 1.
Statistical analysis
In the case–control study performed from 2004 to 2015, the presence/absence of invasion on RCM assessed by a single investigator (PG) was compared to the presence/absence of invasion on pathological evaluation as the reference. For analysis of the matched case–control study, RCM features were described using frequencies and proportions stratified by LMM and LM. The chi‐square test or Fisher's exact test was used to test RCM feature differences between LMM and LM.
In the matched case–control study, the association of each feature with a diagnosis of LMM was assessed using univariable and multivariable logistic regression model. Odds ratio (OR) and its associated 95% confidence interval (CI) was calculated for each feature. The final multivariable model included features that showed a statistically significant associations with LMM from the univariate results. The predictive performance of the final model was assessed using area under the receiver operating characteristics curve (AUC), sensitivity, specificity, and likelihood.
Finally, the inter‐observer reproducibility of each feature was measured by the concordance level between the two blinded investigators using both percent agreement and Kappa value.
Statistical analysis was performed using statistical software SPSS for Windows Version 6.0 (SPSS, Inc.), SAS version 9.4, and R version 3.6.3.
RESULTS
Prospective case–control study from 2005 to 2014
A total of 229 LM/LMM were analyzed prospectively by RCM by a single observer the primary investigator (PG).
This cohort originated only from Australia and all patients were low phototype (I‐II). The lesions were all located on the head and neck region. The LM cases studied were from patients with a mean age of 75 years and 39% were females while 61% were males.
The LMM cases were from patients with a mean age of 73.5 years and 43% were female while 57% were male. The average tumour thickness of the LMM series was 0.55 mm, with a range of 0.18–2.2 mm. The median Breslow thickness was 0.4 mm. Most of the tumours (76%) were thin melanomas (<0.75 mm). A LMM case with confocal and respective pathology correlation is illustrated in Figure 1.
FIGURE 1.
LMM 0.6 mm Breslow located on the scalp (a) Confocal image (Vivascope 3000 1 × 1 mm): Floating cells distributed on DEJ (b) Histopathology image demonstrating invasive melanoma in the superficial dermis with associated dermal fibrosis and chronic inflammation. In situ melanoma is present in the overlying epidermis. (×400).
The correct diagnosis of ‘invasion’ was obtained in 17 of 19 of the LMM cases (89%) and it was not deemed to be invasive in 190 of 210 of the LM cases (90%). A total number of 22 lesions were misdiagnosed, whereby two LMM cases were misclassified as LM and 20 cases of LM were misclassified as LMM on RCM, as illustrated in the Flow Chart, Figure 2. These two misdiagnosed LMM cases were amelanotic LM.
The LM cases were confirmed to be truly LM after being followed up for 7 years and no recurrence or disease progression to LMM was observed.
Retrospective matched case–control study
A total of 93 cases diagnosed pathologically as either LM (n = 58) or LMM (n = 35) on excisions were retrospectively evaluated by two blinded confocal readers. These 93 cases were a subset of the original cohort that has been enriched with 18 new consecutive LMM cases diagnosed from 2015 to 2020, in order to improve our statistical analysis.
These new LMM lesions did not follow the same process of analysis as the LMM lesions in the initial cohort of the study. They were not analyzed prospectively by a blinded single‐reader to assess for an invasive component. The new 18 LMM cases were added to enrich the LMM population for statistical analysis of the matched case–control study in which two blinded confocal experts scored the RCM features retrospectively.
As demonstrated in Table 1, nine RCM features were scored and the most predictive RCM features for the diagnosis of LMM were epidermal and junctional disarray, large size of melanocytes, and nests of melanocytes. All three RCM features were independent predictors of LMM (p value <0.05).
The association of each feature with LMM diagnosis were assessed using univariate logistic regression model (Table 2). Epidermal and junctional disarray, large size of melanocytes, and nests of melanocytes were associated with LMM diagnosis on uni and multivariable regression analysis. The confocal images to illustrate these three LMM features are demonstrated in Figure 3, as well as their corresponding pathology.
TABLE 2.
Univariable and multivariable regression of Lentigo maligna melanoma.
| Variable | Univariate | Multivariate (†) | ||
|---|---|---|---|---|
| OR (95% CI) | p‐value | OR (95% CI) | p‐value | |
| Degree of lentiginous proliferation | ||||
| Scattered | 1 | 0.2074 | ||
| Cluster | 1.36 (0.35, 5.29) | |||
| Sheet | 2.62 (0.73, 9.39) | |||
| Pleomorphism | ||||
| Mild | 1 | 0.1995 | ||
| Moderate | 3.05 (0.47, 19.65) | |||
| Marked | 1.25 (0.22, 6.96) | |||
| Predominant melanocyte size | ||||
| Small | 1 | 0.0195 | 1 | 0.0753 |
| Moderate | 0.91 (0.08, 9.74) | 0. 26 (0. 02, 3.52) | ||
| Large | 3.27 (0.32, 33.84) | 0.83 (0.07, 10.53) | ||
| Nests of melanocytes | ||||
| Absent | 1 | 0.0204 | 1 | 0.6325 |
| Present | 1.94 (0.70, 5.38) | 1.33 (0.41, 4.35) | ||
| Florid | 4.87 (1.60, 14.85) | 1.91 (0. 51, 7.23) | ||
| Adnexal spread | ||||
| Absent | 1 | 0.4371 | ||
| Present | 0.53 (0.19, 1.45) | |||
| Florid | 0.67 (0.23, 2.00) | |||
| Pagetoid spread | ||||
| Absent | 1 | 0.9061 | ||
| Focal | 0.74 (0.17, 3.25) | |||
| Florid | 0.73 (0.17, 3.06) | |||
| Epidermal and Junctional disarray | ||||
| Absent | 1 | 0.0040 | 1 | 0.0366 |
| Present | 4.87 (1.66, 14.34) | 4.13 (1.09, 15.58) | ||
| Pigment incontinence | ||||
| Absent | 1 | 0.9586 | ||
| Present | 0.98 (0.38, 2.48) | |||
| Dermal component | ||||
| Absent | 1 | 0.1868 | ||
| Destroyed | 1.39 (0.47, 4.08) | |||
| Nucleated | 9.33 (0.85, 102.0) | |||
Note: The bold values indicates statistically significant (p‐value <0.05).
FIGURE 3.
(a) Confocal image (Vivascope 3000 1 × 1 mm) – normal architecture is partially lost by the presence of multiple dendritic cells, linear shape, distributed on the epidermis and DEJ (b) Confocal image (Vivascope 3000 1 × 1 mm) – large melanocytes: big dendritic cells (twice the size of keratinocytes) with linear shape (red circle) and round shape (blue circle) (c) Confocal image (Vivascope 3000 1 × 1 mm): Nests ‐ Oval to round bright aggregate with well‐defined borders, composed of clustered cells or hyperrefrective round structures at the junction and upper dermis (d) Histopathology image (×400)‐ Intraepidermal melanocytes show pagetoid scatter and expansile nests of large, atypical melanocytes invade the dermis with a fibrotic response.
Blinded analysis
Concordance between the RCM investigators for the RCM feature were also assessed by the percent agreement and the Kappa value have been calculated on the LMM population recruited from 2015 to 2020 as the cases were available for both blinded readers, as described in Table 3.
TABLE 3.
Inter‐observer agreement analysis.
| RCM feature | Agreement (percent) | Kappa |
|---|---|---|
| Degree of lentiginous proliferation | 15 out of 17 (88.2%) | 0.433 |
| Epidermal and junctional disarray | 14 out of 17 (82.4%) | 0.485 |
| Large melanocytes | 13 out of 17 (76.5%) | 0.468 |
| Dermal component present | 13 out of 17 (76.5%) | 0.443 |
| Nests of melanocytes | 13 out of 17 (76.5%) | 0.433 |
RCM score for LMM
The three RCM features most‐predictive of a diagnosis of LMM, epidermal and junctional disarray, large size of melanocytes, and nests of melanocytes, were combined for a ROC curve analysis. Those three independent features demonstrated an area under the curve of 74% (95% CI 65–85%), with sensitivity (63%, 95% CI 52–78%) and specificity (79% 95% CI 74–88%) at the optimal threshold and likelihood ratio of 18 with p‐value 0.0026 for the LMM diagnosis. The ROC curve is illustrated in the Figure 4.
FIGURE 4.
ROC curve analysis of the RCM features predictive of LMM.
DISCUSSION
This study was designed to assess the accuracy of RCM for the diagnosis of microinvasion in LM lesions. Gender and age were similar between the LM and the LMM cases and no statistically significant difference in the demographic data between patients with LM and those with invasive variant, LMM. In the literature, the age has been described higher ‐ 10 years on average ‐ in the LM/LMM group compared to other melanoma types. 2
Regarding the Breslow thickness, the micro‐invasive component was on average 0.5 mm, which is thinner than the Breslow reported on LMM cases without RCM (average 1.1 mm). 4 This result suggests that biopsies guided by RCM are more accurate for detecting early invasive melanoma as it is possible to target suspicious areas precisely. The precision biopsy technique and accuracy has been previously reported in the literature. 11 , 12
RCM has demonstrated high accuracy to identify invasive features on LM/LMM lesions. The most predictive RCM features on multivariate analysis for the diagnosis of LMM were epidermal and junctional disarray, the large size of melanocytes and the nests of melanocytes. These criteria are not part of the RCM score. Previous publications reported RCM features to differentiate LM from benign macules while, in this study, we used criteria based on pathology correlation to differentiate the in situ and the invasive components in the LM spectrum.
Epidermal and junctional disarray is a feature that is easily detected when a large area is assessed from a horizontal perspective. Therefore, confocal microscopy assessment is ideal to characterize this feature due to its horizontal view. As this feature was highly prevalent in the LMM cases, it potentially has applicability in clinical practice. In our experience, epidermal disarray may be the only clue to targeting the biopsy in the area suspected of invasion, especially when deeper components are difficult to detect by confocal, particularly in hypertrophic lesions with thick epidermis or lesions without pigmentation such as amelanotic melanoma or with deep components such as desmoplastic melanoma. In the literature, epidermal disarray is a non‐specific term which covers multiple epidermal changes that can vary from irregular honeycomb to complete loss of honeycomb. While irregular honeycomb is described as ‘keratinocytes of variable size and shape’ and is part of the natural aging process, 13 , 14 the decohesion and then the loss of honeycomb is more associated with malignancy. 15 The epidermal and junctional disarray was also the main feature described in the literature for extra mammary Paget disease. 16 In our study, we observed that the epidermal disarray (severe distortion to complete loss of honeycomb) was associated with junctional disarray that we described as discohesive cells (‘floating cells’) on RCM images, illustrated in Figure 1. This junctional disarray on RCM correlated to junctional decohesion seen on histopathology. 15 We classified these features as epidermal and junctional disarray because it is hard to define precisely the levels on RCM when the architecture is disrupted, especially on facial skin which is very thin. This is especially the case in chronic sun‐damaged skin. Epidermal and junctional disarray is frequent in LM 6 , 9 but statistically more associated with LMM in our analysis.
A second feature predictive of LMM, the large size of melanocytes, is also easily detected by RCM. Large cells were already defined in the literature as more than 20 microns. 10 Of note, it may not correspond to large cells on the pathology as RCM images are horizontally oriented, different from the vertical cuts on pathology. In our 1 to 1 correlation, we could observe that usually, the large cells on confocal correspond to the most atypical pleomorphic ones on histopathology. In daily practice, the advantage of this feature is that the cellular size is easier to detect and more reproducible feature than cellular morphology on RCM (Figure 3).
Finally, the third feature predictive of LMM, nests of melanocytes are more frequent and florid in the invasive areas. Nests can be found on LM with no invasion when localized strictly to the DEJ. However, it may be hard to establish how deep the nests are and a targeted biopsy is recommended in cases of doubt.
Our data is consistent with a previously published study (n = 12 patients) by Ahlgrimm‐Siess et al. which suggested that junctional nests were predictive of the presence of LMM on LM lesions. The aforementioned study also highlighted changes at the suprabasal layer that we specified herein as epidermal and junctional disarray as a feature predictive of LMM. In contrast to our study, the size of the melanocytes was not formally scored by Ahlgrimm‐Siess et al. However, they did mention that pleomorphism/atypical cells were detected by the amount of pigmentation captured by RCM. 17
In another study, Navarrete‐Dechent et al. evaluated the impact of dermoscopic and confocal LMM/LM findings on the pathologist's interpretation of specimens. It compared 42 biopsies guided by dermoscopy with 30 biopsies targeted by RCM. RCM was more sensitive to select sites containing LM/LMM to biopsy. RCM also detected more features consistent to LMM, being more often able to identify the invasion component on a presurgical staging biopsy. 18 To the best of our knowledge, there have been no previous studies evaluating confocal features for detecting LM progression to invasive LMM.
Our study tested quantitative RCM features and identified the three most predictive ones for LMM diagnosis (Figure 3). This study aimed to address the spectrum of LM disease and allow better pathology correlation, but these quantitative features had low inter rate correlation (kappa) most likely due to the small LMM sample.
Another limitation of this study is the setting from two Australian centres. Moreover, we did not review the lesions classified as lichenoid keratosis, pigmented actinic keratosis, and solar lentigo from our confocal database. The only lesions included on this study were histopathology proven LM/LMM. Our design was to determine whether it might be potentially feasible to target the biopsy appropriately in LM/LMM lesions.
The method needs to be tested on a bigger sample prospectively, including all the differentials, to analyze other potential false positive cases.
In conclusion, clinicians can use these three features to guide partial biopsies and/or mark the area of concern with a punch to guide the histopathologist on the lesions that will be completely excised (‘punch scoring’) as RCM is a valuable tool which can assist in diagnosing dermal invasion arising in LM thereby, facilitating the optimal management of these patients.
AUTHOR CONTRIBUTIONS
Bruna Melhoranse Gouveia: MD, MPhil / Dermatologist, Confocalist at Melanoma Institute Australia, Sydney, Australia, PhD candidate at The University of Sydney, Australia. Giuliana Carlos: MBBS, The University of Sydney, Sydney, Australia. Andreanne Wadell: MD / Dermatologist at CHUS ‐ Hôtel‐Dieu, Sherbrooke, Québec, Research fellow at Melanoma Institute Australia, Sydney, Australia. Christoph Sinz: MD/ Dermatologist at Department of Dermatology, Medical University of Vienna, Vienna, Austria; Research fellow at Sydney Melanoma Diagnostic Centre, Royal Prince Alfred Hospital, Sydney, Australia. Robert Rawson: MBBS FCRPA / Staff Specialist at Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Sydney, Australia, and at Melanoma Institute Australia, Faculty of Medicine and Health, The University of Sydney, Sydney, Australia. Peter Ferguson: MBChB, PhD FRCPA/ Staff Specialist at Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital, Australia, and at Melanoma Institute Australia, Faculty of Medicine and Health, The University of Sydney, Sydney, Australia. Tasnia Ahmed: MS / Biostatistician at Melanoma Institute Australia, Sydney Medical School, The University of Sydney, Sydney, Australia. Serigne N Lo, PhD / Senior Biostatistician at Melanoma Institute Australia, Sydney Medical School, The University of Sydney, Sydney, Australia. Richard A. Scolyer: BMedSci, MBBS, MD, FRCPA, FRCPath/ Professor at Tissue Pathology and Diagnostic Oncology, Royal Prince Alfred Hospital and NSW Health Pathology, Sydney, Australia and at Melanoma Institute Australia, Faculty of Medicine and Health,The University of Sydney, Sydney, Australia; Charles Perkins Centre, The University of Sydney, Sydney, Australia. Pascale Guitera: PhD, FACD / Prof at Melanoma Institute Australia, and Sydney Medical School, The University of Sydney; and Sydney Melanoma Diagnostic Centre, Royal Prince Alfred Hospital, Sydney, Australia.
FUNDING INFORMATION
RAS is supported by an NHMRC Practitioner Fellowship (APP1141295). RR is supported by a grant from Sydney Research. SL and PG are supported by Melanoma Institute Australia. Support from The Cameron Family through Melanoma Institute Australia, as well as from colleagues at Melanoma Institute Australia and Royal Prince Alfred Hospital is also gratefully acknowledged.
CONFLICT OF INTEREST STATEMENT
RAS has received fees for professional services from MetaOptima Technology Inc., Hoffmann‐La Roche Ltd., Evaxion, Provectus Biopharmaceuticals Australia, Qbiotics, Novartis, Merck Sharp & Dohme, NeraCare, AMGEN Inc., Bristol‐Myers Squibb, Myriad Genetics, GlaxoSmithKline. PG has received honoraria from Metaoptima. All the other authors do not have any conflict of interest.
ACKNOWLEDGEMENTS
To all patients that kindly consented to this study and all staff members from Sydney Melanoma Diagnostic Centre and Melanoma Institute Australia. To Dr. Cheng Huang on facilitating the data collection. Open access publishing facilitated by The University of Sydney, as part of the Wiley ‐ The University of Sydney agreement via the Council of Australian University Librarians.
Gouveia BM, Carlos G, Wadell A, Sinz C, Ahmed T, Lo SN, et al. In vivo reflectance confocal microscopy can detect the invasive component of lentigo maligna melanoma: Prospective analysis and case–control study. J Eur Acad Dermatol Venereol. 2023;37:1293–1301. 10.1111/jdv.18998
DATA AVAILABILITY STATEMENT
Raw data were generated at The University of Sydney. Derived data supporting the findings of this study are available from the corresponding author B.G. on request.
References
- 1. Balch CM, Gershenwald JE, Soong SJ, Thompson JF, Atkins MB, Byrd DR, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol. 2009;27(36):6199–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Guitera P, Collgros H, Madronio CM, Goumas C, Mann GJ, Watts CG, et al. The steadily growing problem of lentigo maligna and lentigo maligna melanoma in Australia: population‐based data on diagnosis and management. Australas J Dermatol. 2019;60(2):118–25. [DOI] [PubMed] [Google Scholar]
- 3. Menzies SW, Liyanarachchi S, Coates E, Smith A, Cooke‐Yarborough C, Lo S, et al. Estimated risk of progression of lentigo maligna to lentigo maligna melanoma. Melanoma Res. 2020;30(2):193–7. [DOI] [PubMed] [Google Scholar]
- 4. Tannous ZS, Lerner LH, Duncan LM, Mihm MC Jr, Flotte TJ. Progression to invasive melanoma frommalignant melanoma in situ, lentigo maligna type. Hum Pathol. 2000;31(6):705–8. [DOI] [PubMed] [Google Scholar]
- 5. Shiau CJ, Thompson JF, Scolyer RA. Controversies and evolving concepts in the diagnosis, classification and management of lentigo maligna. Expert Rev Dermatol. 2013;8(2):195–214. [Google Scholar]
- 6. Guitera P, Pellacani G, Crotty KA, Scolyer RA, Li LXL, Bassoli S, et al. The impact of in vivo reflectance confocal microscopy on the diagnostic accuracy of lentigo maligna and equivocal pigmented and nonpigmented macules of the face. J Investig Dermatol. 2010;130(8):2080–91. [DOI] [PubMed] [Google Scholar]
- 7. Ensslin CJ, Hibler BP, Lee EH, Nehal KS, Busam KJ, Rossi AM. Atypical melanocytic proliferations: a review of the literature. Dermatol Surg. 2018;44(2):159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Star P, Guitera P. Lentigo maligna, macules of the face, and lesions on sun‐damaged skin: confocal makes the difference. Dermatol Clin. 2016;34(4):421–9. [DOI] [PubMed] [Google Scholar]
- 9. Rocha LK, Vilain RE, Scolyer RA, Lo SN, Drummond M, Star P, et al. Confocal microscopy, dermoscopy, and histopathology features of atypical intraepidermal melanocytic proliferations associated with evolution to melanoma in situ. Int J Dermatol. 2021;61(2):167–74. [DOI] [PubMed] [Google Scholar]
- 10. Hofmann‐Wellenhof R, Pellacani G, Malvehy J, Peter Soyer H, editors. Reflectance confocal microscopy for skin diseases. Berlin, Germany: Springer Science & Business Media; 2012. [Google Scholar]
- 11. Navarrete‐Dechent C, Cordova M, Sahu A, Liopyris K, Rishpon A, Chen C, et al. Optical imaging guided‐‘precision'biopsy of skin tumors: a novel approach for targeted sampling and histopathologic correlation. Arch Dermatol Res. 2021;313(7):517–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Schwartz RJ, Ho G, Gouveia BM, Martin LK, Guitera P. An adapter window for precise skin marking during handheld reflectance confocal microscopy scanning. J Am Acad Dermatol. 2022;S0190–9622. [DOI] [PubMed] [Google Scholar]
- 13. Longo C, Casari A, Beretti F, Cesinaro AM, Pellacani G. Skin aging: in vivo microscopic assessment of epidermal and dermal changes by means of confocal microscopy. J Am Acad Dermatol. 2013;68(3):e73–82. [DOI] [PubMed] [Google Scholar]
- 14. Guida S, Pellacani G, Ciardo S, Longo C. Reflectance confocal microscopy of aging skin and skin cancer. Dermatol Pract Concept. 2021;11(3):e2021068. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Gill M, Longo C, Farnetani F, Cesinaro AM, González S, Pellacani G. Non‐invasive in vivo dermatopathology: identification of reflectance confocal microscopic correlates to specific histological features seen in melanocytic neoplasms. J Eur Acad Dermatol Venereol. 2014;28(8):1069–78. [DOI] [PubMed] [Google Scholar]
- 16. Navarrete‐Dechent C, Aleissa S, Cordova M, Hibler BP, Erlendsson AM, Polansky M, et al. Treatment of extramammary Paget disease and the role of reflectance confocal microscopy: a prospective study. Dermatol Surg. 2021;47(4):473–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ahlgrimm‐Siess V, Massone C, Scope A, Fink‐Puches R, Richtig E, Wolf IH, et al. Reflectance confocal microscopy of facial lentigo maligna and lentigo maligna melanoma: a preliminary study. Br J Dermatol. 2009;161(6):1307–16. [DOI] [PubMed] [Google Scholar]
- 18. Navarrete‐Dechent C, Cordova M, Liopyris K, Rishpon A, Aleissa S, Rossi AM, et al. Reflectance confocal microscopy and dermoscopy aid in evaluating repigmentation within or adjacent to lentigo maligna melanoma surgical scars. J Eur Acad Dermatol Venereol. 2020;34(1):74–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Raw data were generated at The University of Sydney. Derived data supporting the findings of this study are available from the corresponding author B.G. on request.