Background
While the excisional biopsy and histological examination of suspicious lesions remains the current gold standard for diagnosing cutaneous melanoma (CM), there is a demand for more ...objective and non‐invasive examination methods that may support clinicians in their decision when to biopsy or not.
Methods
This review is based on publications and guidelines retrieved by a selective search in PubMed and MEDLINE and focused on non‐invasive diagnostic strategies for detecting melanoma.
Results
Ten different non‐invasive techniques were compared with regard to applicability, status of development, and resources necessary for introduction into clinical routine (dermoscopy, sequential digital dermoscopy, total body photography, computer‐aided multispectral digital analysis, electrical impedance spectroscopy, Raman spectroscopy, reflectance confocal microscopy, multiphoton tomography, stepwise two‐photon‐laser spectroscopy, quantitative dynamic infrared imaging). In an effort to create a classification based on our analyses, we suggest to differentiate i) tools for screening of patients in daily clinical routine, ii) tools for examination of a restricted number of preselected lesions that produce an automated diagnostic score, iii) tools for examination of a restricted number of preselected lesions at specialized centers requiring extensive training, iv) devices at an experimental stage of development.
Conclusion
None of the discussed examination techniques is able to provide a definite and final diagnosis or to completely replace the histopathological examination. Up to date, the need for fully automated devices offering a complete skin cancer screening has not been satisfied.
Summary
Background
The Psoriasis Area and Severity Index (PASI) represents the gold standard for psoriasis severity assessments but is limited by its subjectivity and low intra‐ and inter‐rater ...consistency.
Objectives
To investigate the precision and reproducibility of automated, computer‐guided PASI measurements (ACPMs) in comparison with three trained physicians.
Methods
This was a comparative observational study assessing ACPMs attained by automated total‐body imaging and computerized digital image analysis in a cohort of 120 patients affected by plaque psoriasis of various severities. The level of agreement between ACPMs and physicians’ PASI measurements was calculated by the intraclass correlation coefficient (ICC). The reproducibility of ACPMs in comparison with physicians’ PASI measurements was investigated by performing two successive ‘repeat PASI calculations’ in the same patients.
Results
The agreement between ACPMs and physicians’ PASI calculations in 120 fully evaluable patients was high (ICC 0·86, 95% confidence interval 0·80–0·90, mean absolute difference 2·5 PASI points). Repeat ACPMs to measure the reproducibility showed an excellent ICC of 0·99 (95% confidence interval 0·98–0·99) with a mean absolute difference of 0·5 PASI points. The ACPMs thus outperformed the three physicians for intrarater reliability (mean ICC 0·86).
Conclusions
The results of this first clinical study investigating ACPMs in 120 patients with psoriasis indicate a similar precision and higher reproducibility in comparison with trained physicians. Limitations arise from poorly observable body sites and from patients unable to attain predefined postures during automated image acquisition.
What's already known about this topic?
The Psoriasis Area and Severity Index (PASI) is the standard for psoriasis severity assessments but is limited by its subjectivity and low reproducibility.
What does this study add?
Automated computer‐guided PASI measurements (ACPMs) agreed with the results of trained physicians and outperformed the physicians in terms of reproducibility.
ACPMs provide a first step towards a higher standardization and reproducibility of PASI measurements.
Linked Comment: Garcia‐Doval and Albrecht. Br J Dermatol 2019; 180:260–261.
Respond to this article
Summary
Background
Early detection is a key factor in improving survival from melanoma. Today, the clinical diagnosis of cutaneous melanoma is based mostly on visual inspection and dermoscopy. ...Preclinical studies in freshly excised or paraffin‐embedded tissue have shown that the melanin fluorescence spectra after stepwise two‐photon excitation, a process termed dermatofluoroscopy, differ between cutaneous melanoma and melanocytic naevi. However, confirmation from a larger prospective clinical study is lacking.
Objectives
The primary end point of this study was to determine the diagnostic accuracy of dermatofluoroscopy in melanoma detection. Secondary end points included the collection of data for improving the computer algorithm that classifies skin lesions based on melanin fluorescence and the assessment of safety aspects.
Methods
This was a prospective, blinded, multicentre clinical study in patients with pigmented skin lesions (PSLs) indicated for excision either to rule out or to confirm cutaneous melanoma. All included lesions underwent dermoscopy and dermatofluoroscopy in vivo before lesions were excised and subjected to histopathological examination.
Results
In total, 369 patients and 476 PSLs were included in the final analysis. In 101 of 476 lesions (21·2%) histopathology revealed melanoma. The observed sensitivity of dermatofluoroscopy was 89·1% (90 of 101 melanomas identified), with an observed specificity of 44·8%. The positive and negative predictive values were 30·3% and 93·9%, respectively. No adverse events occurred.
Conclusions
Dermatofluoroscopy is a safe and accurate diagnostic method to aid physicians in diagnosing cutaneous melanoma. Limitations arise from largely amelanotic or regressing lesions lacking sufficient melanin fluorescence.
What's already known about this topic?
Dermatofluoroscopy is the analysis of melanin fluorescence spectra after stepwise two‐photon excitation.
In preclinical studies dermatofluoroscopy showed a high diagnostic accuracy for detecting cutaneous melanoma.
What does this study add?
This is the first prospective, multicentre, clinical study investigating the diagnostic accuracy of dermatofluoroscopy for the diagnosis of melanoma in preselected pigmented skin lesions with the indication for excision to rule out or confirm melanoma.
Linked Comment: Cinotti and Rubegni. Br J Dermatol 2018; 179:255–256.
Respond to this article
Convolutional neural networks (CNNs) efficiently differentiate skin lesions by image analysis. Studies comparing a market-approved CNN in a broad range of diagnoses to dermatologists working under ...less artificial conditions are lacking.
One hundred cases of pigmented/non-pigmented skin cancers and benign lesions were used for a two-level reader study in 96 dermatologists (level I: dermoscopy only; level II: clinical close-up images, dermoscopy, and textual information). Additionally, dermoscopic images were classified by a CNN approved for the European market as a medical device (Moleanalyzer Pro, FotoFinder Systems, Bad Birnbach, Germany). Primary endpoints were the sensitivity and specificity of the CNN’s dichotomous classification in comparison with the dermatologists’ management decisions. Secondary endpoints included the dermatologists’ diagnostic decisions, their performance according to their level of experience, and the CNN’s area under the curve (AUC) of receiver operating characteristics (ROC).
The CNN revealed a sensitivity, specificity, and ROC AUC with corresponding 95% confidence intervals (CI) of 95.0% (95% CI 83.5% to 98.6%), 76.7% (95% CI 64.6% to 85.6%), and 0.918 (95% CI 0.866–0.970), respectively. In level I, the dermatologists’ management decisions showed a mean sensitivity and specificity of 89.0% (95% CI 87.4% to 90.6%) and 80.7% (95% CI 78.8% to 82.6%). With level II information, the sensitivity significantly improved to 94.1% (95% CI 93.1% to 95.1%; P < 0.001), while the specificity remained unchanged at 80.4% (95% CI 78.4% to 82.4%; P = 0.97). When fixing the CNN’s specificity at the mean specificity of the dermatologists’ management decision in level II (80.4%), the CNN’s sensitivity was almost equal to that of human raters, at 95% (95% CI 83.5% to 98.6%) versus 94.1% (95% CI 93.1% to 95.1%); P = 0.1. In contrast, dermatologists were outperformed by the CNN in their level I management decisions and level I and II diagnostic decisions. More experienced dermatologists frequently surpassed the CNN’s performance.
Under less artificial conditions and in a broader spectrum of diagnoses, the CNN and most dermatologists performed on the same level. Dermatologists are trained to integrate information from a range of sources rendering comparative studies that are solely based on one single case image inadequate.
•A market-approved convolutional neural network (CNN) trained on dermoscopic images was tested against 96 dermatologists.•Test data included a broad range of skin lesions and was compiled from external sources not involved in CNN training.•Dermatologists indicated their management decisions after reviewing clinical, dermoscopic, and textual case information.•In this setting dermatologists performed on par with the CNN's classifications based on dermoscopic images alone.
Background
Metabolic reprogramming and altered gene expression mediated by hypoxia‐inducible factors play crucial roles during tumour growth and progression. Nevertheless, studies analysing the ...expression of hypoxia‐inducible factor‐1α and its downstream targets in Merkel cell carcinoma (MCC) are lacking but are warranted to shed more light on MCC pathogenesis and to potentially provide new therapeutic options.
Objectives
To analyse the immunohistochemical expression of hypoxia‐inducible factor‐1α (HIF‐1α), vascular endothelial growth factor‐A (referred to as VEGF throughout the manuscript), VEGF receptor‐2 (VEGFR‐2), VEGF receptor‐3 (VEGFR‐3), glucose transporter‐1 (Glut‐1), monocarboxylate transporter 4 (MCT4) and carbonic anhydrase IX (CAIX) in primary cutaneous MCC.
Methods
The 16 paraffin‐embedded primary cutaneous MCCs (Merkel cell polyomavirus (McPyV) positive/negative: 11/5) were analysed by immunohistochemistry, namely HIF‐1α, VEGF, VEGFR‐2 (KDR), VEGFR‐3 (FLT4), Glut‐1, MCT4 and CAIX. An established quantification score (QS) was applied to quantitate the protein expression by considering the percentage of positive tumour cells (0: 0%; 1: up to 1%; 2: 2–10%; 3: 11–50%; 4: >50%) in relation to the staining intensity (0: negative; 1: low; 2: medium; 3: strong).
Results
HIF‐1α was expressed in all MCCs and predominantly found at the invading edges of tumour margins. The HIF‐1α downstream factors Glut‐1, MCT4 and CAIX were expressed in 13 of 16 MCC (81%), 14 of 16 MCC (88%) and 16 of 16 MCC (100%), respectively. Interestingly, VEGF and VEGFR‐2 were not expressed in tumour cells, whereas VEGFR‐3 was expressed in all MCCs. HIF‐1α was expressed significantly stronger in McPyV+ tumours (QS: 10.36 ± 2.41) than in McPyV− tumours (QS: 5.40 ± 1.34; P = 0.002). Similarly, VEGFR‐3 was also expressed significantly stronger in McPyV+ tumours (QS: 10.00 ± 2.52) than in McPyV− tumours (QS: 5.40 ± 3.43, P = 0.019).
Conclusions
Our data provide first evidence for a role of HIF‐1α in induced metabolic reprogramming contributing to MCC pathogenesis. The metabolic signatures of McPyV+ and McPyV− tumours seem to show relevant differences.
Background
The Psoriasis Area and Severity Index (PASI) is the standard for psoriasis severity assessment. However, PASI measurement is complex and subjective, frequently leading to a high intra‐ and ...interobserver variability. To date, the precise extent of variability in PASI measurements and its underlying causes remain unknown.
Objective
To determine the inter‐ and intrarater variability of image‐based PASI measurements by calculating Intra‐Class‐Correlation‐Coefficients (ICCs) and to investigate the impact of the different PASI components and specific anatomic regions on the extent of variability.
Methods
First, the methodology of ‘image‐based’ vs. commonly used ‘live’ PASI measurements was validated in a pilot study. Next, in an observational cohort study, PASI scores of 120 patients affected by plaque psoriasis were prospectively evaluated by three formally trained physicians by means of total body images (TBI). Each observer independently performed two rounds of image‐based PASI calculations in all patients at two different time points.
Results
Overall, 720 image‐based PASI scores were calculated with a mean PASI of 8.8 (range 0.7–34.8). An interrater variability with an ICC of 0.895 and mean absolute difference (MAD) of 3.3 PASI points were observed. Intrarater variability showed a mean ICC of 0.877 and a MAD of 2.2 points. When considering specific PASI components, the highest agreement was found for the assessment of the involved body surface area (BSA), while the lowest ICCs were calculated for severity scoring of ‘scaling’ and ‘induration’. As BSA scores serve as a multiplier in the calculation of PASI, minor inaccuracies were capable of inducing a large share of variability.
Conclusion
The overall inter‐ and intrarater reliability of image‐based PASI measurements in this study was good. However, physicians were formally trained and experienced, which frequently is not the case in a real‐life clinical setting. Therefore, new strategies for higher standardization and objectivity of PASI calculations are needed.
Background
Deep learning convolutional neural networks (CNN) may assist physicians in the diagnosis of melanoma. The capacity of a CNN to differentiate melanomas from combined naevi, the latter ...representing well‐known melanoma simulators, has not been investigated.
Objective
To assess the diagnostic performance of a CNN when used to differentiate melanomas from combined naevi in comparison with dermatologists.
Methods
In this study, a CNN with regulatory approval for the European market (Moleanalyzer‐Pro, FotoFinder Systems GmbH, Bad Birnbach, Germany) was used. We attained a dichotomous classification (benign, malignant) in dermoscopic images of 36 combined naevi and 36 melanomas with a mean Breslow thickness of 1.3 mm. Primary outcome measures were the CNN's sensitivity, specificity and the diagnostic odds ratio (DOR) in comparison with 11 dermatologists with different levels of experience.
Results
The CNN revealed a sensitivity, specificity and DOR of 97.1% (95% CI 82.7–99.6), 78.8% (95% CI 62.8–89.1.3) and 34 (95% CI 4.8–239), respectively. Dermatologists showed a lower mean sensitivity, specificity and DOR of 90.6% (95% CI 84.1–94.7; P = 0.092), 71.0% (95% CI 62.6–78.1; P = 0.256) and 24 (95% CI 11.6–48.4; P = 0.1114). Under the assumption that dermatologists use the CNN to verify their (initial) melanoma diagnosis, dermatologists achieve an increased specificity of 90.3% (95% CI 79.8–95.6) at an almost unchanged sensitivity. The largest benefit was observed in ‘beginners’, who performed worst without CNN verification (DOR = 12) but best with CNN verification (DOR = 98).
Conclusion
The tested CNN more accurately classified combined naevi and melanomas in comparison with trained dermatologists. Their diagnostic performance could be improved if the CNN was used to confirm/overrule an initial melanoma diagnosis. Application of a CNN may therefore be of benefit to clinicians.
Linked Commentary: M.V. Heppt et al. J Eur Acad Dermatol Venereol 2020; 34: 1134–1135. https://doi.org/10.1111/jdv.16577.
Cold atmospheric plasma (CAP, i.e. ionized air) is an innovating promising tool in reducing bacteria.
We conducted the first clinical trial with the novel PlasmaDerm® VU-2010 device to assess safety ...and, as secondary endpoints, efficacy and applicability of 45 s/cm(2) cold atmospheric plasma as add-on therapy against chronic venous leg ulcers.
From April 2011 to April 2012, 14 patients were randomized to receive standardized modern wound care (n = 7) or plasma in addition to standard care (n = 7) 3× per week for 8 weeks. The ulcer size was determined weekly (Visitrak® , photodocumentation). Bacterial load (bacterial swabs, contact agar plates) and pain during and between treatments (visual analogue scales) were assessed. Patients and doctors rated the applicability of plasma (questionnaires).
The plasma treatment was safe with 2 SAEs and 77 AEs approximately equally distributed among both groups (P = 0.77 and P = 1.0, Fisher's exact test). Two AEs probably related to plasma. Plasma treatment resulted in a significant reduction in lesional bacterial load (P = 0.04, Wilcoxon signed-rank test). A more than 50% ulcer size reduction was noted in 5/7 and 4/7 patients in the standard and plasma groups, respectively, and a greater size reduction occurred in the plasma group (plasma -5.3 cm(2) , standard: -3.4 cm(2) ) (non-significant, P = 0.42, log-rank test). The only ulcer that closed after 7 weeks received plasma. Patients in the plasma group quoted less pain compared to the control group. The plasma applicability was not rated inferior to standard wound care (P = 0.94, Wilcoxon-Mann-Whitney test). Physicians would recommend (P = 0.06, Wilcoxon-Mann-Whitney test) or repeat (P = 0.08, Wilcoxon-Mann-Whitney test) plasma treatment by trend.
Cold atmospheric plasma displays favourable antibacterial effects. We demonstrated that plasma treatment with the PlasmaDerm® VU-2010 device is safe and effective in patients with chronic venous leg ulcers. Thus, larger controlled trials and the development of devices with larger application surfaces are warranted.