Smartphone applications ("apps") with artificial intelligence (AI) algorithms are increasingly used in healthcare. Widespread adoption of these apps must be supported by a robust evidence-base and ...app manufacturers' claims appropriately regulated. Current CE marking assessment processes inadequately protect the public against the risks created by using smartphone diagnostic apps.
AbstractObjectiveTo examine the validity and findings of studies that examine the accuracy of algorithm based smartphone applications (“apps”) to assess risk of skin cancer in suspicious skin ...lesions.DesignSystematic review of diagnostic accuracy studies.Data sourcesCochrane Central Register of Controlled Trials, MEDLINE, Embase, CINAHL, CPCI, Zetoc, Science Citation Index, and online trial registers (from database inception to 10 April 2019).Eligibility criteria for selecting studiesStudies of any design that evaluated algorithm based smartphone apps to assess images of skin lesions suspicious for skin cancer. Reference standards included histological diagnosis or follow-up, and expert recommendation for further investigation or intervention. Two authors independently extracted data and assessed validity using QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2 tool). Estimates of sensitivity and specificity were reported for each app.ResultsNine studies that evaluated six different identifiable smartphone apps were included. Six verified results by using histology or follow-up (n=725 lesions), and three verified results by using expert recommendations (n=407 lesions). Studies were small and of poor methodological quality, with selective recruitment, high rates of unevaluable images, and differential verification. Lesion selection and image acquisition were performed by clinicians rather than smartphone users. Two CE (Conformit Europenne) marked apps are available for download. SkinScan was evaluated in a single study (n=15, five melanomas) with 0% sensitivity and 100% specificity for the detection of melanoma. SkinVision was evaluated in two studies (n=252, 61 malignant or premalignant lesions) and achieved a sensitivity of 80% (95% confidence interval 63% to 92%) and a specificity of 78% (67% to 87%) for the detection of malignant or premalignant lesions. Accuracy of the SkinVision app verified against expert recommendations was poor (three studies).ConclusionsCurrent algorithm based smartphone apps cannot be relied on to detect all cases of melanoma or other skin cancers. Test performance is likely to be poorer than reported here when used in clinically relevant populations and by the intended users of the apps. The current regulatory process for awarding the CE marking for algorithm based apps does not provide adequate protection to the public.Systematic review registrationPROSPERO CRD42016033595.
Image-based skin cancer diagnostic technologies utilizing artificial intelligence (AI) are increasing in popularity in dermatology. The World Health Organization has cautioned against overestimating ...the benefits of AI at the expense of core investment and strategies required to achieve universal health coverage. In the UK, the National Health Service has a fixed budget, and expenditure decisions regarding new AI technologies need to be fully justified. Health economic evaluations can be challenging, but such evaluations can also help to elucidate the factors that determine suitability of a new AI diagnostic tool beyond its immediate costs and purported benefits.
Background
Checkpoint inhibitor (CPI) therapy has significantly improved overall survival in several cancers including metastatic melanoma (MM) and in the adjuvant setting. Cutaneous immune‐related ...adverse events (irAEs) secondary to CPIs are commonly observed; however, autoimmune blistering disorders such as bullous pemphigoid (BP) are rare.
Objectives
To review the prevalence, incidence risk, clinicopathological features and management of toxicity in bullous cutaneous irAEs associated with CPI therapy.
Methods
A multicentre, retrospective, observational study of CPI‐associated bullous irAEs in adults with all cancers across four UK specialist centres between 2006 and 2019.
Results
In total, 7391 patients were identified. CPI‐associated bullous irAEs including BP (n = 16) occurred in 0·3% (n = 22). The median age of onset was 76 years, and there was a male predominance. Most patients had cutaneous melanoma (73%, n = 16), of which 81% (13 of 16) were BRAF wildtype. Grade 1, 2, 3 and 4 skin toxicity occurred in 9%, 45%, 41% and 5%, respectively. The mucosae were involved in 27%, and 25% of confirmed cases of BP did not present with bullae. The median time to onset of bullous irAEs was 12 months, with a median total symptom duration of 6 months. Single PD‐1/PD‐L1 agents had a longer time to onset of symptoms than combination therapy (median 12 vs. 7 months, respectively). Overall, 91%, 64% and 9% of patients required one, two or three lines of treatment, respectively. Two cases occurred after completion of CPIs (1 and 3 months). Of the 20 cases that presented while on CPIs this was permanently discontinued in 55% (11 of 20) and temporarily held in 20% (four of 20). In the four held cases of CPI, bullous eruption reflared in 50%.
Conclusions
CPI‐associated bullous skin toxicity is a rare cutaneous irAE occurring in approximately 0·3% of cases over 13 years of treated patients in this series. Not all cases are diagnostic of BP, but management remains the same. There is a prolonged latency of onset compared with other cutaneous irAEs, with a median time of 12 months, and they can occur after cessation of therapy. Discontinuation of CPIs may be required. Recognizing bullous irAEs promptly and referral to dermatology are essential to optimize management and improve patient outcomes and tumour responses.
What is already known about this topic?
Checkpoint inhibitor (CPI)‐associated bullous pemphigoid is a rare dermatological immune‐related adverse event (irAE) that has been reported in small case series and reports.
What does this study add?
This is the largest multicentre, observational study conducted in the UK over the longest period of 13 years, which demonstrates an overall incidence of bullous cutaneous irAEs secondary to CPIs of 0·3%.
Clinical presentation is variable, with one‐quarter of patients with bullous pemphigoid presenting without bullae, and mucosal involvement was noted in 27%. Prolonged pruritus is frequently a prodromal symptom.
The median time to diagnosis is 12 months and irAEs rarely present after cessation of treatment. Time to onset of symptoms is longer with a single CPI, but with a shorter duration of symptoms compared with combination CPI therapy.
Most patients had cutaneous melanoma, of which 81% were BRAF wildtype.
This is the largest multicentre, observational study conducted in the UK, demonstrating an overall incidence of bullous cutaneous irAEs associated with CPI of 0·3%. Clinical presentation may vary, with pruritus and non‐bullous variants appearing similar to other cutaneous toxicities. Time to onset appears later than with other cutaneous toxicities (12 months) and can occur after discontinuation of therapy. Interruption of CPI is often needed. Our study demonstrates this occurs mostly in men, older age groups and is associated with BRAF wildtype in MM patients. It is important for clinicians to recognize this condition promptly, so early optimal management is initiated to reduce interruptions to CPI and improve overall patient outcomes and tumour response.
Linked Comment: K. Kridin and C.M. Hammers. Br J Dermatol 2022; 187:843–844.
Plain language summary available online
Immunodeficiency-associated lymphoproliferative disorders (IA-LPDs) constitute a diverse range of conditions including posttransplant lymphoproliferative disorders, other iatrogenic IA-LPDs, and ...lymphoproliferative disorders associated with an underlying primary immune disorder or HIV infection. IA-LPDs are clinically and pathologically heterogeneous, and there is a lack of standardization of diagnostic terminology. They can represent a potential serious diagnostic pitfall because the histological features of clinically indolent proliferations may mimic those of high-grade lymphoma. However, correct identification of these entities is essential given that complete remission may occur upon reversal of the underlying cause of immunosuppression without the need for systemic therapy. IA-LPDs presenting in the skin are rare but well documented. One form of iatrogenic IA-LPD, methotrexate-associated lymphoproliferative disorder (MTX-LPD), can present with cutaneous nodules, plaques, or ulcers. Predominantly, MTX-LPD develops in the context of long-term treatment of autoimmune conditions, such as rheumatoid arthritis, dermatomyositis, and Sjögren syndrome, and may be associated with underlying Epstein-Barr virus (EBV) infection. We present 4 cases of cutaneous EBV-positive B-cell MTX-LPD and describe their clinical and morphological findings. Comparison of our histological findings to the diagnostic criteria for EBV-positive mucocutaneous ulcer (EBVMCU) revealed significant overlap, highlighting the intersection between MTX-LPD and EBVMCU. Withdrawal of methotrexate resulted in healing of all lesions at a mean time of 2 months. In summary, close clinicopathological correlation is vital to identify MTX-LPD presenting as cutaneous EBVMCU given that the initial treatment strategy is that of withdrawal of methotrexate without the need for immediate systemic therapy.
Background
Melanoma has one of the fastest rising incidence rates of any cancer. It accounts for a small percentage of skin cancer cases but is responsible for the majority of skin cancer deaths. ...Although history‐taking and visual inspection of a suspicious lesion by a clinician are usually the first in a series of ‘tests’ to diagnose skin cancer, dermoscopy has become an important tool to assist diagnosis by specialist clinicians and is increasingly used in primary care settings. Dermoscopy is a magnification technique using visible light that allows more detailed examination of the skin compared to examination by the naked eye alone. Establishing the additive value of dermoscopy over and above visual inspection alone across a range of observers and settings is critical to understanding its contribution for the diagnosis of melanoma and to future understanding of the potential role of the growing number of other high‐resolution image analysis techniques.
Objectives
To determine the diagnostic accuracy of dermoscopy alone, or when added to visual inspection of a skin lesion, for the detection of cutaneous invasive melanoma and atypical intraepidermal melanocytic variants in adults. We separated studies according to whether the diagnosis was recorded face‐to‐face (in‐person), or based on remote (image‐based), assessment.
Search methods
We undertook a comprehensive search of the following databases from inception up to August 2016: CENTRAL; MEDLINE; Embase; CINAHL; CPCI; Zetoc; Science Citation Index; US National Institutes of Health Ongoing Trials Register; NIHR Clinical Research Network Portfolio Database; and the World Health Organization International Clinical Trials Registry Platform. We studied reference lists and published systematic review articles.
Selection criteria
Studies of any design that evaluated dermoscopy in adults with lesions suspicious for melanoma, compared with a reference standard of either histological confirmation or clinical follow‐up. Data on the accuracy of visual inspection, to allow comparisons of tests, was included only if reported in the included studies of dermoscopy.
Data collection and analysis
Two review authors independently extracted all data using a standardised data extraction and quality assessment form (based on QUADAS‐2). We contacted authors of included studies where information related to the target condition or diagnostic threshold were missing. We estimated accuracy using hierarchical summary receiver operating characteristic (SROC),methods. Analysis of studies allowing direct comparison between tests was undertaken. To facilitate interpretation of results, we computed values of sensitivity at the point on the SROC curve with 80% fixed specificity and values of specificity with 80% fixed sensitivity. We investigated the impact of in‐person test interpretation; use of a purposely developed algorithm to assist diagnosis; observer expertise; and dermoscopy training.
Main results
We included a total of 104 study publications reporting on 103 study cohorts with 42,788 lesions (including 5700 cases), providing 354 datasets for dermoscopy. The risk of bias was mainly low for the index test and reference standard domains and mainly high or unclear for participant selection and participant flow. Concerns regarding the applicability of study findings were largely scored as ‘high’ concern in three of four domains assessed. Selective participant recruitment, lack of reproducibility of diagnostic thresholds and lack of detail on observer expertise were particularly problematic.
The accuracy of dermoscopy for the detection of invasive melanoma or atypical intraepidermal melanocytic variants was reported in 86 datasets; 26 for evaluations conducted in person (dermoscopy added to visual inspection), and 60 for image‐based evaluations (diagnosis based on interpretation of dermoscopic images). Analyses of studies by prior testing revealed no obvious effect on accuracy; analyses were hampered by the lack of studies in primary care, lack of relevant information and the restricted inclusion of lesions selected for biopsy or excision. Accuracy was higher for in‐person diagnosis compared to image‐based evaluations (relative diagnostic odds ratio (RDOR) 4.6, 95% confidence interval (CI) 2.4 to 9.0; P < 0.001).
We compared accuracy for (a), in‐person evaluations of dermoscopy (26 evaluations; 23,169 lesions and 1664 melanomas),versus visual inspection alone (13 evaluations; 6740 lesions and 459 melanomas), and for (b), image‐based evaluations of dermoscopy (60 evaluations; 13,475 lesions and 2851 melanomas),versus image‐based visual inspection (11 evaluations; 1740 lesions and 305 melanomas). For both comparisons, meta‐analysis found dermoscopy to be more accurate than visual inspection alone, with RDORs of (a), 4.7 (95% CI 3.0 to 7.5; P < 0.001), and (b), 5.6 (95% CI 3.7 to 8.5; P < 0.001). For a), the predicted difference in sensitivity at a fixed specificity of 80% was 16% (95% CI 8% to 23%; 92% for dermoscopy + visual inspection versus 76% for visual inspection), and predicted difference in specificity at a fixed sensitivity of 80% was 20% (95% CI 7% to 33%; 95% for dermoscopy + visual inspection versus 75% for visual inspection). For b) the predicted differences in sensitivity was 34% (95% CI 24% to 46%; 81% for dermoscopy versus 47% for visual inspection), at a fixed specificity of 80%, and predicted difference in specificity was 40% (95% CI 27% to 57%; 82% for dermoscopy versus 42% for visual inspection), at a fixed sensitivity of 80%.
Using the median prevalence of disease in each set of studies ((a), 12% for in‐person and (b), 24% for image‐based), for a hypothetical population of 1000 lesions, an increase in sensitivity of (a), 16% (in‐person), and (b), 34% (image‐based), from using dermoscopy at a fixed specificity of 80% equates to a reduction in the number of melanomas missed of (a), 19 and (b), 81 with (a), 176 and (b), 152 false positive results. An increase in specificity of (a), 20% (in‐person), and (b), 40% (image‐based), at a fixed sensitivity of 80% equates to a reduction in the number of unnecessary excisions from using dermoscopy of (a), 176 and (b), 304 with (a), 24 and (b), 48 melanomas missed.
The use of a named or published algorithm to assist dermoscopy interpretation (as opposed to no reported algorithm or reported use of pattern analysis), had no significant impact on accuracy either for in‐person (RDOR 1.4, 95% CI 0.34 to 5.6; P = 0.17), or image‐based (RDOR 1.4, 95% CI 0.60 to 3.3; P = 0.22), evaluations. This result was supported by subgroup analysis according to algorithm used. We observed higher accuracy for observers reported as having high experience and for those classed as ‘expert consultants’ in comparison to those considered to have less experience in dermoscopy, particularly for image‐based evaluations. Evidence for the effect of dermoscopy training on test accuracy was very limited but suggested associated improvements in sensitivity.
Authors' conclusions
Despite the observed limitations in the evidence base, dermoscopy is a valuable tool to support the visual inspection of a suspicious skin lesion for the detection of melanoma and atypical intraepidermal melanocytic variants, particularly in referred populations and in the hands of experienced users. Data to support its use in primary care are limited, however, it may assist in triaging suspicious lesions for urgent referral when employed by suitably trained clinicians. Formal algorithms may be of most use for dermoscopy training purposes and for less expert observers, however reliable data comparing approaches using dermoscopy in person are lacking.
Primary cutaneous leiomyosarcoma (LMS) is a rare soft tissue tumour type with two subtypes, dermal and subcutaneous. As deeper tumours confer a worse prognosis, they require a more aggressive ...approach. Conversely, a more conservative surgical approach for dermal LMS has been suggested. Few studies have comprehensively reported both clinical surgical and histological excision margins. Therefore, we sought to provide margin recommendations based on our experience and review of the existing literature. We undertook a retrospective case-note review (1998-2019) of cutaneous LMS management to establish histological/surgical margins using pathology/electronic patient records. The diagnosis was made and classified by an experienced dermatopathologist according to the World Health Organization classification. In the dermal LMS cohort (n = 35), mean peripheral and deep histological margins were 5.4 mm (range 0.5-20) and 5.6 mm (range 0.1-14.5), respectively. The incomplete excision rate was 31% (11 of 35). There were no recurrences. In the subcutaneous LMS cohort (n = 10), mean peripheral and deep histological margins were 5.7 mm (range 0.2-14) and 1.1 mm (range 0.2-1.7), respectively. The incomplete excision rate was 40% (4 of 10). The recurrence rate was 20% (2 of 10) despite achieving histological clearance after 1 year. One lung metastasis occurred 1 year following an adequately excised primary scalp LMS. Thus, for dermal LMS we propose a clinical margin of 5-10 mm (depending on lesion size) at the initial excision or at scar re-excision following involved/close histological peripheral and/or deep margins (i.e. < 1 mm). For subcutaneous LMS, we suggest a clinical margin of 15-20 mm (depending on lesion size) to achieve a peripheral histological clearance of 10 mm and negative deep margin (i.e. > 1 mm), down to the periosteum/fascia/muscle according to anatomical site. If this is not achieved, a re-excision would be recommended. However, prospective studies are needed for optimal guidance.
Providing detailed skin cancer statistics, including incidence and survival, by tumour type and patient characteristics is important for up-to-date epidemiological information.
To create a new ...clinically relevant consensus-based classification for registered skin tumours using tumour type and patient characteristics and to describe its application to all registered tumours in England between 2013 and 2019.
Tumours with skin topographical codes (ICD-10) and morphology and behaviour (ICD-O3) were grouped together in an iterative process creating a hierarchical tree structure. The primary-level grouping partitioned skin tumours into skin cancer, melanoma in situ, extramammary Paget disease (EMPD) and tumours of uncertain malignant potential. Second-level groups split skin cancer into keratinocyte cancer (KC), melanoma and rare cancers. The third-level group split KC into basal cell carcinoma (BCC) and squamous cell carcinoma (cSCC). Further groups were split into genital or non-genital, first or subsequent tumour, age, gender, stage, or National Health Service (NHS) region. Incidence counts, Kaplan-Meier and net survival estimates and referral routes two-week wait (TWW), general practitioner (GP), outpatient categorisations were calculated for each grouping across all years.
A total of 1 445 377 skin cancers and 49 123 precancerous lesions and undefined entities were registered in England between 2013 and 2019. Skin tumours and skin cancer incidence rates are increasing for most tumour types. The most common type of skin cancer was BCC with an incidence rate of 282.36 per 100 000 person-years (PYs) n = 158 934, 95% confidence interval (CI) 280.98-283.76 in 2019, followed by cSCC with an incidence rate of 85.24 per 100 000 PYs (n = 47 977, 95% CI 84.48-86.00) and melanoma with 27.24 (n = 15 332, 95% CI 26.81-27.67) per 100 000 PYs. Each year approximately 1800 rare skin cancers, 1500 genital cSCCs and 100 cases of EMPD are registered. Of 15 000 melanoma cases, 120 cases of melanoma occur in individuals aged < 25 years annually. One-year and five-year overall net survival varies by tumour type. cSCC 5-year net survival (89.8%, 95% CI 88.8-90.9) was comparable to the net survival of all melanomas (89.6%, 95% CI 88.7-90.6). BCC had excellent survival (overall net survival > 100%). Patients with late-stage melanoma, Merkel cell carcinoma and genital cSCC have a 5-year net survival < 60%. Older patients received fewer TWW referrals than their younger counterparts with the same tumour type at the same location. Patients with acral lentiginous melanoma had fewer TWW referrals and more standard GP referrals than patients with common melanomas.
'Get Data Out' Skin provides detailed and up-to-date statistics on all registrable skin tumours in England, including for the first time precancerous lesions and rare subtypes of common cancers. These data can be used by clinicians, researchers and commissioners to better understand skin cancer and improve resource allocation.