Search

BackgroundSkin cancers are known for their strong immunogenicity, which may contribute to a high treatment efficacy of immune checkpoint inhibition (ICI). However, a considerable proportion of patients with skin cancer is immuno-compromised by concomitant diseases. Due to their previous exclusion from clinical trials, the ICI treatment efficacy is poorly investigated in these patients. The present study analyzed the ICI treatment outcome in advanced patients with skin cancer with a concomitant hematological malignancy.MethodsThis retrospective multicenter study included patients who were treated with ICI for locally advanced or metastatic melanoma (MM), cutaneous squamous cell carcinoma (cSCC), or Merkel cell carcinoma (MCC), and had a previous diagnosis of a hematological malignancy irrespective of disease activity or need of therapy at ICI treatment start. Comparator patient cohorts without concomitant hematological malignancy were extracted from the prospective multicenter skin cancer registry ADOREG. Treatment outcome was measured as best overall response, progression-free (PFS), and overall survival (OS).Results84 patients (MM, n=52; cSCC, n=15; MCC, n=17) with concomitant hematological malignancy were identified at 20 skin cancer centers. The most frequent concomitant hematological malignancies were non-Hodgkin’s lymphoma (n=70), with chronic lymphocytic leukemia (n=32) being the largest entity. While 9 patients received ICI in an adjuvant setting, 75 patients were treated for advanced non-resectable disease (55 anti-PD-1; 8 anti-PD-L1; 5 anti-CTLA-4; 7 combinations). In the latter 75 patients, best objective response (complete response+partial response) was 28.0%, disease stabilization was 25.3%, and 38.6% showed progressive disease (PD). Subdivided by skin cancer entity, best objective response was 31.1% (MM), 26.7% (cSCC), and 18.8% (MCC). Median PFS was 8.4 months (MM), 4.0 months (cSCC), and 5.7 months (MCC). 1-year OS rates were 78.4% (MM), 65.8% (cSCC), and 47.4% (MCC). Comparison with respective ADOREG patient cohorts without hematological malignancy (n=392) revealed no relevant differences in ICI therapy outcome for MM and MCC, but a significantly reduced PFS for cSCC (p=0.002).ConclusionsICI therapy showed efficacy in advanced patients with skin cancer with a concomitant hematological malignancy. Compared with patients without hematological malignancy, the observed ICI therapy outcome was impaired in cSCC, but not in MM or MCC patients.

Background Recent studies have successfully demonstrated the use of deep-learning algorithms for dermatologist-level classification of suspicious lesions by the use of excessive proprietary image databases and limited numbers of dermatologists. For the first time, the performance of a deep-learning algorithm trained by open-source images exclusively is compared to a large number of dermatologists covering all levels within the clinical hierarchy. Methods We used methods from enhanced deep learning to train a convolutional neural network (CNN) with 12,378 open-source dermoscopic images. We used 100 images to compare the performance of the CNN to that of the 157 dermatologists from 12 university hospitals in Germany. Outperformance of dermatologists by the deep neural network was measured in terms of sensitivity, specificity and receiver operating characteristics. Findings The mean sensitivity and specificity achieved by the dermatologists with dermoscopic images was 74.1% (range 40.0%–100%) and 60% (range 21.3%–91.3%), respectively. At a mean sensitivity of 74.1%, the CNN exhibited a mean specificity of 86.5% (range 70.8%–91.3%). At a mean specificity of 60%, a mean sensitivity of 87.5% (range 80%–95%) was achieved by our algorithm. Among the dermatologists, the chief physicians showed the highest mean specificity of 69.2% at a mean sensitivity of 73.3%. With the same high specificity of 69.2%, the CNN had a mean sensitivity of 84.5%. Interpretation A CNN trained by open-source images exclusively outperformed 136 of the 157 dermatologists and all the different levels of experience (from junior to chief physicians) in terms of average specificity and sensitivity.

Background: Recent studies have demonstrated the use of convolutional neural networks (CNNs) to classify images of melanoma with accuracies comparable to those achieved by board-certified dermatologists. However, the performance of a CNN exclusively trained with dermoscopic images in a clinical image classification task in direct competition with a large number of dermatologists has not been measured to date. This study compares the performance of a convolutional neuronal network trained with dermoscopic images exclusively for identifying melanoma in clinical photographs with the manual grading of the same images by dermatologists. Methods: We compared automatic digital melanoma classification with the performance of 145 dermatologists of 12 German university hospitals. We used methods from enhanced deep learning to train a CNN with 12,378 open-source dermoscopic images. We used 100 clinical images to compare the performance of the CNN to that of the dermatologists. Dermatologists were compared with the deep neural network in terms of sensitivity, specificity and receiver operating characteristics. Findings: The mean sensitivity and specificity achieved by the dermatologists with clinical images was 89.4% (range: 55.0%-100%) and 64.4% (range: 22.5%-92.5%). At the same sensitivity, the CNN exhibited a mean specificity of 68.2% (range 47.5%-86.25%). Among the dermatologists, the attendings showed the highest mean sensitivity of 92.8% at a mean specificity of 57.7%. With the same high sensitivity of 92.8%, the CNN had a mean specificity of 61.1%. Interpretation: For the first time, dermatologist-level image classification was achieved on a clinical image classification task without training on clinical images. The CNN had a smaller variance of results indicating a higher robustness of computer vision compared with human assessment for dermatologic image classification tasks. (C) 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).