This work summarizes the results of the largest skin image analysis challenge in the world, hosted by the International Skin Imaging Collaboration (ISIC), a global partnership that has organized the world's largest public repository of dermoscopic images of skin. The challenge was hosted in 2018 at the Medical Image Computing and Computer Assisted Intervention (MICCAI) conference in Granada, Spain. The dataset included over 12,500 images across 3 tasks. 900 users registered for data download, 115 submitted to the lesion segmentation task, 25 submitted to the lesion attribute detection task, and 159 submitted to the disease classification task. Novel evaluation protocols were established, including a new test for segmentation algorithm performance, and a test for algorithm ability to generalize. Results show that top segmentation algorithms still fail on over 10% of images on average, and algorithms with equal performance on test data can have different abilities to generalize. This is an important consideration for agencies regulating the growing set of machine learning tools in the healthcare domain, and sets a new standard for future public challenges in healthcare.

No abstract

PURPOSE The aim of the current study was to report the efficacy of topical and systemic treatments for immune-related cutaneous adverse events (ircAEs) attributed to checkpoint inhibitors in an uncontrolled cohort of patients referred to oncodermatology clinics. METHODS A retrospective analysis of patients with ircAEs evaluated by dermatologists from January 1, 2014, to December 31, 2017, at three tertiary care hospitals and cancer centers were identified through electronic medical records. Clinicopathologic characteristics, dermatologic therapy outcome, and laboratory data were analyzed. RESULTS A total of 285 patients (median age, 65 years [range, 17 to 89 years]) with 427 ircAEs were included: pruritus (n = 138; 32%), maculopapular rash (n = 120; 28%), psoriasiform rash (n = 22; 5%), and others (n = 147; 34%). Immune checkpoint inhibitor class was associated with ircAE phenotype ( P = .007), where maculopapular rash was predominant in patients who received combination therapy. Severity of ircAEs was significantly reduced (mean Common Terminology Criteria for Adverse Events grade: 1.74 v 0.71; P < .001) with dermatologic interventions, including topical corticosteroids, oral antipruritics, and systemic immunomodulators. A total of 88 ircAEs (20%) were managed with systemic immunomodulators. Of these, 22 (25%) of 88 persisted or worsened. In seven patients with corticosteroid-refractory ircAEs, improvement resulted from targeted biologic immunomodulatory therapies that included rituximab and dupilumab. Serum interleukin-6 (IL-6) was elevated in 34 (52%) of 65 patients; grade 3 or greater ircAEs were associated with increased absolute eosinophils (odds ratio, 4.1; 95% CI, 1.3 to 13.4) and IL-10 (odds ratio, 23.8; 95% CI, 2.1 to 262.5); mean immunoglobulin E serum levels were greater in higher-grade ircAEs: 1,093 kU/L (grade 3), 245 kU/L (grade 2), and 112 kU/L (grade 1; P = .043). CONCLUSION Most ircAEs responded to symptom- and phenotype-directed dermatologic therapies, whereas biologic therapies were effective in patients with corticosteroid-refractory disease. Increased eosinophils, IL-6, IL-10, and immunoglobulin E were associated with ircAEs, and they may represent actionable therapeutic targets for immune-related skin toxicities.

Patients detected almost half of first relapses. Our data suggest that routine physical examinations beyond 3 years for stage IIIA, 2 years for stage IIIB, and 1 year for stage IIIC patients and radiologic imaging beyond 3 years for stages IIIA and IIIB and 2 years for stage IIIC patients would be expected to detect few first systemic relapses.

Prophylaxis with oral minocycline may be useful in decreasing the severity of the acneiform rash during the first month of cetuximab treatment. Topical tazarotene is not recommended for management of cetuximab-related rash.