Early switch from run-in treatment with vemurafenib plus cobimetinib to atezolizumab after 3 months leads to rapid loss of tumour control in patients with advanced BRAFV600-positive melanoma: The ImmunoCobiVem phase 2 randomised trial.

Aim: ImmunoCobiVem investigated whether a planned switch to atezolizumab after achieving tumour control during run-in with vemurafenib + cobimetinib improves progression-free survival (PFS) and overall survival (OS) compared to continuous targeted therapy (TT) in patients with previously untreated advanced BRAF-mutated melanoma.

Methods: In this multicenter phase 2 study, patients received vemurafenib plus cobimetinib. After 3months, patients without progressive disease (PD) were randomly assigned (1:1) to continue vemurafenib + cobimetinib (Arm A) or switch to atezolizumab (Arm B) until first documented PD (PD1).

Medical Needs and Therapeutic Options for Melanoma Patients Resistant to Anti-PD-1-Directed Immune Checkpoint Inhibition.

Available 4- and 5-year updates for progression-free and for overall survival demonstrate a lasting clinical benefit for melanoma patients receiving anti-PD-directed immune checkpoint inhibitor therapy. However, at least one-half of the patients either do not respond to therapy or relapse early or late following the initial response to therapy. Little is known about the reasons for primary and/or secondary resistance to immunotherapy and the patterns of relapse.

Early decrease of blood myeloid-derived suppressor cells during checkpoint inhibition is a favorable biomarker in metastatic melanoma.

Background: The need for reliable clinical biomarkers to predict which patients with melanoma will benefit from immune checkpoint blockade (ICB) remains unmet. Several different parameters have been considered in the past, including routine differential blood counts, T cell subset distribution patterns and quantification of peripheral myeloid-derived suppressor cells (MDSC), but none has yet achieved sufficient accuracy for clinical utility.

Methods: Here, we investigated potential cellular biomarkers from clinical routine blood counts as well as several myeloid and T cell subsets, using flow cytometry, in two independent cohorts of a total of 141 patients with stage IV M1c melanoma before and during ICB.

Clinical and pathological characteristics of familial melanoma with germline TERT promoter variants.

Around 10% of melanoma occurs in patients with a suspected familial predisposition. TERT promoter mutations are the most common somatic hotspot mutations in human cancers. However, only two families with germline mutations have been identified to date.

Dynamics of Melanoma-Associated Epitope-Specific CD8+ T Cells in the Blood Correlate With Clinical Outcome Under PD-1 Blockade.

Immune checkpoint blockade (ICB) is standard-of-care for patients with metastatic melanoma. It may re-invigorate T cells recognizing tumors, and several tumor antigens have been identified as potential targets. However, little is known about the dynamics of tumor antigen-specific T cells in the circulation, which might provide valuable information on ICB responses in a minimally invasive manner.

Explainable artificial intelligence in skin cancer recognition: A systematic review.

Background: Due to their ability to solve complex problems, deep neural networks (DNNs) are becoming increasingly popular in medical applications. However, decision-making by such algorithms is essentially a black-box process that renders it difficult for physicians to judge whether the decisions are reliable. The use of explainable artificial intelligence (XAI) is often suggested as a solution to this problem.

Skin cancer classification via convolutional neural networks: systematic review of studies involving human experts.

Background: Multiple studies have compared the performance of artificial intelligence (AI)-based models for automated skin cancer classification to human experts, thus setting the cornerstone for a successful translation of AI-based tools into clinicopathological practice.

Objective: The objective of the study was to systematically analyse the current state of research on reader studies involving melanoma and to assess their potential clinical relevance by evaluating three main aspects: test set characteristics (holdout/out-of-distribution data set, composition), test setting (experimental/clinical, inclusion of metadata) and representativeness of participating clinicians.

Methods: PubMed, Medline and ScienceDirect were screened for peer-reviewed studies published between 2017 and 2021 and dealing with AI-based skin cancer classification involving melanoma.

A benchmark for neural network robustness in skin cancer classification.

Background: One prominent application for deep learning-based classifiers is skin cancer classification on dermoscopic images. However, classifier evaluation is often limited to holdout data which can mask common shortcomings such as susceptibility to confounding factors. To increase clinical applicability, it is necessary to thoroughly evaluate such classifiers on out-of-distribution (OOD) data.

Integrating Patient Data Into Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review.

Background: Recent years have been witnessing a substantial improvement in the accuracy of skin cancer classification using convolutional neural networks (CNNs). CNNs perform on par with or better than dermatologists with respect to the classification tasks of single images. However, in clinical practice, dermatologists also use other patient data beyond the visual aspects present in a digitized image, further increasing their diagnostic accuracy.

Combining CNN-based histologic whole slide image analysis and patient data to improve skin cancer classification.

Background: Clinicians and pathologists traditionally use patient data in addition to clinical examination to support their diagnoses.

Objectives: We investigated whether a combination of histologic whole slides image (WSI) analysis based on convolutional neural networks (CNNs) and commonly available patient data (age, sex and anatomical site of the lesion) in a binary melanoma/nevus classification task could increase the performance compared with CNNs alone.

Methods: We used 431 WSIs from two different laboratories and analysed the performance of classifiers that used the image or patient data individually or three common fusion techniques.

Robustness of convolutional neural networks in recognition of pigmented skin lesions.

Background: A basic requirement for artificial intelligence (AI)-based image analysis systems, which are to be integrated into clinical practice, is a high robustness. Minor changes in how those images are acquired, for example, during routine skin cancer screening, should not change the diagnosis of such assistance systems.

Objective: To quantify to what extent minor image perturbations affect the convolutional neural network (CNN)-mediated skin lesion classification and to evaluate three possible solutions for this problem (additional data augmentation, test-time augmentation, anti-aliasing).

Targeted and Checkpoint Inhibitor Therapy of Metastatic Malignant Melanoma in Germany, 2000-2016.

Targeted therapies (TT) and immune checkpoint inhibitors (ICI) have become increasingly important in the treatment of metastatic malignant melanoma in recent years. We examined implementation and effectiveness of these new therapies over time in Germany with a focus on regional differences. We analyzed data from 12 clinical cancer registries in 8 federal states in Germany over the period 2000-2016.

Melanoma brain metastases - Interdisciplinary management recommendations 2020.

Melanoma brain metastases (MBM) are common and associated with a particularly poor prognosis; they directly cause death in 60-70% of melanoma patients. In the past, systemic treatments have shown response rates around 5%, whole brain radiation as standard of care has achieved a median overall survival of approximately three months. Recently, the combination of immune checkpoint inhibitors and combinations of MAP-kinase inhibitors both have shown very promising response rates of up to 55% and 58%, respectively, and improved survival.

Effects of Label Noise on Deep Learning-Based Skin Cancer Classification.

Recent studies have shown that deep learning is capable of classifying dermatoscopic images at least as well as dermatologists. However, many studies in skin cancer classification utilize non-biopsy-verified training images. This imperfect ground truth introduces a systematic error, but the effects on classifier performance are currently unknown.