The application of artificial intelligence in the detection of basal cell carcinoma: A systematic review.

Basal cell carcinoma (BCC) is one of the most common types of cancer. The growing incidence worldwide and the need for fast, reliable and less invasive diagnostic techniques make a strong case for the application of different artificial intelligence techniques for detecting and classifying BCC and its subtypes. We report on the current evidence regarding the application of handcrafted and deep radiomics models used for the detection and classification of BCC in dermoscopy, optical coherence tomography and reflectance confocal microscopy. We reviewed all the articles that were published in the last 10 years in PubMed, Web of Science and EMBASE, and we found 15 articles that met the inclusion criteria. We included articles that are original, written in English, focussing on automated BCC detection in our target modalities and published within the last 10 years in the field of dermatology. The outcomes from the selected publications are presented in three categories depending on the imaging modality and to allow for comparison. The majority of articles (n = 12) presented different AI solutions for the detection and/or classification of BCC in dermoscopy images. The rest of the publications presented AI solutions in OCT images (n = 2) and RCM (n = 1). In addition, we provide future directions for the application of these techniques for the detection of BCC. In conclusion, the reviewed publications demonstrate the potential benefit of AI in the detection of BCC in dermoscopy, OCT and RCM.

[Basal cell nevus syndrome: the interface between dentistry and dermatology]. / Het basaalcelnaevussyndroom: raakvlak tussen tandheelkunde en dermatologie.

Basal cell nevus syndrome is a rare, autosomal dominant disorder, predominantly caused by a mutation in the PTCH1 gene. As basal cell carcinomas and keratocysts are the most common abnormalities, dermatologists, orofacial maxillary surgeons, and dentists play a key role in patient care. From the age of 8, screening for odontogenic keratocysts with an orthopantomogram or MRI is recommended every other year. The intensity increases to annual screening after the development of the first odontogenic keratocyst. If BCNS is caused by an underlying SUFU mutation, screening is not indicated since there are no reports of odontogenic keratocyst in these patients to date. Radiation exposure by, for example, computed tomography, should be minimized as it induces new BCCs. Regular follow-up by a dermatologist for early diagnosis and treatment of (multiple) BCC's is recommended for life.

2021 international consensus statement on optical coherence tomography for basal cell carcinoma: image characteristics, terminology and educational needs.

BACKGROUND: Despite the widespread use of optical coherence tomography (OCT) for imaging of keratinocyte carcinoma, we lack an expert consensus on the characteristic OCT features of basal cell carcinoma (BCC), an internationally vetted set of OCT terms to describe various BCC subtypes, and an educational needs assessment. OBJECTIVES: To identify relevant BCC features in OCT images, propose terminology based on inputs from an expert panel and identify content for a BCC-specific curriculum for OCT trainees. METHODS: Over three rounds, we conducted a Delphi consensus study on BCC features and terminology between March and September 2020. In the first round, experts were asked to propose BCC subtypes discriminable by OCT, provide OCT image features for each proposed BCC subtypes and suggest content for a BCC-specific OCT training curriculum. If agreement on a BCC-OCT feature exceeded 67%, the feature was accepted and included in a final review. In the second round, experts had to re-evaluate features with less than 67% agreement and rank the ten most relevant BCC OCT image features for superficial BCC, nodular BCC and infiltrative and morpheaphorm BCC subtypes. In the final round, experts received the OCT-BCC consensus list for a final review, comments and confirmation. RESULTS: The Delphi included six key opinion leaders and 22 experts. Consensus was found on terminology for three OCT BCC image features: (i) hyporeflective areas, (ii) hyperreflective areas and (iii) ovoid structures. Further, the participants ranked the ten most relevant image features for nodular, superficial, infiltrative and morpheaform BCC. The target group and the key components for a curriculum for OCT imaging of BCC have been defined. CONCLUSION: We have established a set of OCT image features for BCC and preferred terminology. A comprehensive curriculum based on the expert suggestions will help implement OCT imaging of BCC in clinical and research settings.

A trial-based cost-effectiveness analysis of topical 5-fluorouracil vs. imiquimod vs. ingenol mebutate vs. methyl aminolaevulinate conventional photodynamic therapy for the treatment of actinic keratosis in the head and neck area performed in the Netherlands.

BACKGROUND: Actinic keratosis (AK) is a common premalignant skin condition that might have the ability to progress into squamous cell carcinoma. Due to the high incidence of AK, treatment of this disease significantly impacts healthcare spending. OBJECTIVES: To determine which commonly prescribed field-directed treatment is the most cost-effective, when comparing 5-fluorouracil (5-FU) 5%, imiquimod (IMQ) 5%, ingenol mebutate (IM) 0·015% and methyl aminolaevulinate photodynamic therapy (MAL-PDT) for AK in the head and neck region. METHODS: We performed an economic evaluation from a healthcare perspective. Data were collected alongside a single-blinded, prospective, multicentre randomized controlled trial with 624 participants in the Netherlands. The outcome measure was expressed as the incremental cost-effectiveness ratio, which is the incremental costs per additional patient with ≥ 75% lesion reduction compared with baseline. This trial was registered at ClinicalTrials.gov, number NCT02281682. RESULTS: The trial showed that 5-FU was the most effective field treatment for AK in the head and neck region. Twelve months post-treatment, the total mean costs for 5-FU were significantly lower (€433) than the €728, €775 and €1621 for IMQ, IM and MAL-PDT, respectively. The results showed that 5-FU was a dominant cost-effective treatment (more effective and less expensive) compared with the other treatments, 12 months post-treatment. CONCLUSIONS: Based on these results, we consider 5-FU 5% cream as the first-choice treatment option for multiple AKs in the head and neck area. What's already known about this topic? Due to the increasing incidence of actinic keratosis (AK), the recommended treatment results in a considerable socioeconomic burden for (dermatological) healthcare. Although cost-effectiveness modelling studies have been performed in which different treatments for AK were compared, a prospective clinical trial comparing four frequently prescribed treatments on effectiveness and resource consumption within a time horizon of 12 months has never been conducted. What does this study add? This is the first study examining the cost-effectiveness of 5-fluorouracil 5% cream, imiquimod 5% cream, ingenol mebutate 0·015% gel and methyl aminolaevulinate photodynamic therapy, with data collected in a randomized controlled trial over a time horizon of 12 months. We found that 5-fluorouracil was a dominant cost-effective treatment (more effective and less costly), based on data from the Netherlands. Linked Comment: Steeb et al. Br J Dermatol 2020; 183:612.

Genetic profiling of basal cell carcinomas detects postzygotic mosaicism in basal cell naevus syndrome.

Basal cell naevus syndrome (BCNS) is associated with germline mutations in the PTCH1 gene. Postzygotic mosaicism can also cause BCNS. Here we describe two patients, one with multiple basal cell carcinomas (BCCs) and one with clinical BCNS, who had no PTCH1 mutation in DNA extracted from blood. In both patients, we performed genetic analysis on different BCCs, revealing the presence of a shared PTCH1 mutation in all tumours. Our findings show that in patients with symptoms of BCNS and initial absence of a PTCH1 mutation in blood, genetic profiling of BCCs can detect postzygotic mosaicism. What's already known about this topic? Basal cell naevus syndrome (BCNS) is associated with germline mutations in the PTCH1 gene, but it can also be caused by low-grade postzygotic mosaicism in PTCH1. What does this study add? In patients suspected of having BCNS or patients with multiple basal cell carcinomas (BCCs) with a special distribution on the body and no mutation detected in blood, it is worthwhile to search for a shared PTCH1 mutation in their BCCs as this can detect postzygotic mosaicism. This information is important to ensure proper surveillance programmes, choose the right therapy and provide adequate genetic counselling.

Treatment of superficial basal cell carcinoma by topical photodynamic therapy with fractionated 5-aminolaevulinic acid 20% vs. two-stage topical methyl aminolaevulinate: results of a randomized controlled trial.

BACKGROUND: Basal cell carcinoma (BCC) is the most common type of skin cancer and incidence rates are increasing. Photodynamic therapy (PDT) is a frequently used treatment, especially for superficial BCC (sBCC). Two topical photosensitizing agents are currently used to treat sBCC, namely 5-aminolaevulinic acid (ALA) and its ester, methyl aminolaevulinate (MAL). Previous research showed a high efficacy for ALA-PDT using a twofold fractionated illumination scheme in which two light fractions of 20 J cm-2 and 80 J cm-2 were delivered 4 h and 6 h after ALA application. OBJECTIVES: To evaluate whether twofold ALA-PDT is superior to conventional MAL-PDT for sBCC. METHODS: We performed a single-blind, randomized, multicentre trial in the Netherlands. RESULTS: Overall, 162 patients were randomized either to conventional MAL-PDT or twofold ALA-PDT. After 12 months, a total of six treatment failures occurred following ALA-PDT and 13 treatment failures occurred following MAL-PDT. The 12-month cumulative probability of remaining free from treatment failure was 92·3% [95% confidence interval (CI) (83·7-96·5)] for ALA-PDT and 83·4% (95% CI 73·1-90·0) for MAL-PDT (P = 0·091). CONCLUSIONS: The twofold ALA-PDT scheme resulted in fewer recurrences, although the difference between both treatment groups was not statistically significant. However, ALA-PDT resulted in higher pain scores and more post-treatment side-effects compared with MAL-PDT.

Postzygotic mosaicism in basal cell naevus syndrome.

Basal cell naevus syndrome (BCNS) is an autosomal dominant disorder most commonly caused by a germline mutation in the Drosophila homologue of patched-1 gene (PTCH1). Here we describe a patient with clinical signs of BCNS, caused by postzygotic mosaicism of a PTCH1 mutation. We performed restriction fragment length polymorphism analysis and Droplet Digital polymerase chain reaction to determine the degree of mosaicism in different tissues of this patient. Our case shows that a relatively low-grade mosaicism can lead to clinical signs reminiscent of those caused by a germline mutation. This finding has important implications for genetic counselling and therefore is pivotal to recognize for dermatologists, as well as for clinical geneticists and clinical laboratory geneticists.

Reliability of the Patient and Observer Scar Assessment Scale and a 4-point scale in evaluating linear facial surgical scars.

BACKGROUND: In order to evaluate and improve aesthetic outcome for patients undergoing dermatological surgery, a reliable evaluation tool must be used. The 4-point scale and the Patient and Observer Scar Assessment Scale (POSAS) were both developed for this purpose. OBJECTIVE: To compare the reliability of the POSAS scale with the 4-point scale for facial linear surgical scars and to assess the influence of different scar characteristics on the overall impression. METHODS: Patients visiting the outpatient clinics of the Maastricht University Medical Centre with linear facial scars were included. The 4-point scale and the Observer Scar Assessment Scale (OSAS) were completed by three independent observers. The Patient Scar Assessment Scale (PSAS) was completed by the patient on the day of the visit and 2 weeks later. The intra-class correlation coefficient (ICC) with corresponding 95% confidence intervals (CI) were calculated to assess the reliability of the scales. Linear multivariate regression analyses were performed to evaluate how the score on each aspect affected the overall opinion. RESULTS: Fifty scars in 50 patients were included. The ICC of the 4-point scale was 0.819 (95% CI: 0.708-0.892) for multiple observers and 0.602 (95% CI: 0.447-0.734) for single observer. These were superior to the ICCs of total OSAS scores (0.783 95% CI: 0.547-0.888 for multiple observers and 0.546 95% CI: 0.287-0.726 for single observer). ICCs of individual sub-items on the OSAS scored even lower. The sub-items contributed differently to the overall opinion among the different observers or patient. CONCLUSION: In terms of reliability, the overall opinion and the 4-point scale were superior to the total POSAS score or the sub-items. Observers do not weight individual scar characteristics equally to arrive at an overall opinion, which challenges the assumption that calculating a total POSAS score by summing of the scores on the individual items is a valid approach.

Photodynamic therapy vs. topical imiquimod for treatment of superficial basal cell carcinoma: a subgroup analysis within a noninferiority randomized controlled trial.

BACKGROUND: A recent noninferiority randomized controlled trial (RCT) indicated that imiquimod can be considered as superior to methylaminolevulinate photodynamic therapy (MAL-PDT) in the treatment of superficial basal cell carcinoma (sBCC). Knowledge of treatment effectiveness in subgroups of patients is of great value in clinical practice to select the most effective treatment for an individual patient with sBCC. OBJECTIVES: To explore whether the relative treatment effect of MAL-PDT and imiquimod is consistent across subgroups defined by patient and tumour characteristics. METHODS: Data were derived from a single-blinded, noninferiority, multicentre RCT comparing MAL-PDT, topical imiquimod and fluorouracil (ISRCTN79701845). Treatment success was defined as free of tumour recurrence at 12-month follow-up. Subgroup analyses were performed for subgroups defined by sex, age, tumour location and tumour size. RESULTS: Two hundred and two patients received MAL-PDT and 198 received imiquimod. The superiority of imiquimod vs. MAL-PDT was observed in subgroups of females, sBCC on the trunk and large tumours with risk differences in favour of imiquimod of 18·4% [95% confidence interval (CI) 7·8-29·0%], 21·0% (95% CI 10·9-31·1%) and 18·9% (95% CI 7·1-30·7%), respectively. Higher probability of treatment success for imiquimod vs. MAL-PDT was consistently found in all other subgroups with the exception of sBCC localized on the lower extremities in older patients. In the latter subgroup, the risk difference at the expense of imiquimod was -57·3% (95% CI -81·7% to -32·9%). CONCLUSIONS: Imiquimod remains the first-choice treatment for sBCC in terms of effectiveness. In older patients with sBCC on the lower extremities MAL-PDT might be preferred. Results should be interpreted carefully as subgroup analyses were exploratory and not driven by prior hypotheses.

Cost-effectiveness of topical imiquimod and fluorouracil vs. photodynamic therapy for treatment of superficial basal-cell carcinoma.

BACKGROUND: A recent noninferiority randomized trial showed that in terms of clinical effectiveness imiquimod was superior and topical fluorouracil noninferior to methylaminolaevulinate photodynamic therapy (MAL-PDT) for treatment of superficial basal-cell carcinoma (sBCC). Although it was expected that MAL-PDT would be more costly than either cream, a full cost-effectiveness analysis is necessary to determine the balance between effectiveness and costs. OBJECTIVE: To determine whether imiquimod or topical fluorouracil are cost-effective treatments for sBCC compared with MAL-PDT. METHODS: An economic evaluation was performed from a healthcare perspective. Data on resource use and costs were collected alongside the randomized clinical trial. The incremental cost-effectiveness ratio was expressed as the incremental costs per additional patient free of tumour recurrence. RESULTS: At 12 months follow-up, the total mean costs for MAL-PDT were €680, for imiquimod cream €526 and for topical fluorouracil cream €388. Both imiquimod and topical fluorouracil were cost-effective treatments compared with MAL-PDT. Comparing costs and effectiveness of both creams led to a incremental investment of €4451 to achieve an additional patient free of tumour recurrence. The acceptability curve showed that, for a threshold value of €4451, the probability of imiquimod being more cost-effective than topical fluorouracil was 50%. CONCLUSION: Based on the 12 months follow-up results, imiquimod and topical fluorouracil cream are more cost-effective than MAL-PDT for treatment of sBCC. Hence, substituting MAL-PDT with either imiquimod or topical fluorouracil results in cost savings; these savings will be larger for topical fluorouracil. Long-term follow-up effectiveness data are necessary to confirm the cost-effectiveness of imiquimod vs. topical 5-fluorouracil cream.

Aesthetic evaluation after non-invasive treatment for superficial basal cell carcinoma.

BACKGROUND: Cosmetic results following non-invasive treatments are difficult to compare. Although qualified objective scar assessment scales are available, they are not used in dermatological studies. Usually a 4-point scale is used in dermatological scars. The reproducibility of this method has never been evaluated. Moreover, significant specific scar characteristics are lacking. The patient and observer scar assessment scale (POSAS) is a scale qualified for the assessments of surgical scars. It has proven to be as reliable as the widely used Vancouver Scar Scale, but has the advantage that it includes the patient's opinion and specifies different scar characteristics. OBJECTIVE: Both methods were used to evaluate cosmetic results following non-invasive treatments of superficial basal cell carcinoma (BCC). METHODS: A total of 54 lesions following non-invasive treatment for BCC in 54 patients were evaluated with the traditional 4-point scale and the POSAS. RESULTS: The 4-point scale showed the best reproducibility and had an intra-class correlation coefficient (ICC) of 0.66 (95% CI: 0.52-0.77) for a single observer and 0.85 (95% CI: 0.77-0.91) for multiple observers. The ICC of the POSAS was 0.41 (95% CI: 0.21-0.58) for a single observer and 0.67 (95% CI: 0.45-0.81) for three observers. The scar characteristics, vascularity and pigmentation were most decisive for the overall opinion. CONCLUSION: The use of the 4-point scale is a valid method to compare scars of non-invasive dermatological treatments. Supplementary registering vascularity and pigmentation can be useful in future studies.

Agreement between histological subtype on punch biopsy and surgical excision in primary basal cell carcinoma.

BACKGROUND: Diagnosis of clinically suspected basal cell carcinoma (BCC) by histological confirmation with punch biopsy has been recommended before treatment. Even shave biopsy has been proposed as useful to predict the correct subtype in primary BCC in 76-81%, whereas the agreement between histological BCC subtype on punch biopsy and subsequent excision specimens in recurrent BCC is 67.1%. However, no large studies on the agreement between histological BCC subtype seen on punch biopsy and the following surgical excision are performed in primary BCC. OBJECTIVE: The aims of this study were (i) to establish the agreement between histological BCC subtype on punch biopsy and the subsequent surgical excision of primary BCC and; (ii) to investigate the proportion of primary BCCs in which punch biopsy enables identification of the most aggressive growth pattern. METHODS: Retrospective analyses of 243 primary BCCs with both punch biopsy and subsequent surgical excision. Analyses were based on the most aggressive histological subtype of the tumour. RESULTS: The agreement between BCC subtype on punch biopsy and the subsequent surgical excision of primary BCCs was 60.9%. A punch biopsy can predict the most aggressive growth pattern of primary BCCs in 84.4%. Seventy-four percentage of all primary BCCs consisted of more than one histological subtype. CONCLUSION: Dermatologists and other physicians have to be aware of the limited diagnostic value of a punch biopsy to determine the histological BCC subtype of the whole lesion. Misdiagnosis of the subtype will lead to undertreatment in one of six primary BCCs.