BACKGROUND: Cancers are known to elude the immune system, for example, by MHC loss, FAS up-regulation, or increased secretion of TGF-beta. Recently, ligands of coinhibitory receptors like programmed cell death ligand-1 (PD-L1, B7-H1) have come to attention for their role in tumor immune escape. Various tumors have been tested for PD-L1 expression, and conflicting results were obtained regarding its correlative impact on patient survival. This study aimed to determine the prognostic relevance of PD-L1 expression for the survival of melanoma patients. METHODS: Paraffin-embedded nevi, primary melanoma, and in-transit, lymph node, and distant organ metastases from a set of 63 stages III-IV melanoma patients referred to the Netherlands Cancer Institute between 2000 and 2004 for a sentinelnode procedure or systemic therapy were studied. A large effort was invested in validating specific PD-L1 staining. In addition to immunological factors such as T-cell infiltration (CD8, CD4, and regulatory T cells), TGF-beta and MHC-I expression were assessed. RESULTS: Longitudinal analysis revealed no relevant PD-L1 expression on primary melanoma compared with metastatic disease. No significant correlations with prognosis were found regarding immunological factors, whereas known prognostic markers such as Breslow thickness and sex could be confirmed. Analyses of the overall survival of our patient cohort did not reveal a negative association with PD-L1 expression. CONCLUSIONS: Correlation of overall survival with PD-L1 expression by melanoma cells remains controversial, and future clinical studies should focus on antibody validation and time of analysis in respect to disease progression.
Artificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent validation. With this in mind, we organized the PANDA challenge—the largest histopathology competition to date, joined by 1,290 developers—to catalyze development of reproducible AI algorithms for Gleason grading using 10,616 digitized prostate biopsies. We validated that a diverse set of submitted algorithms reached pathologist-level performance on independent cross-continental cohorts, fully blinded to the algorithm developers. On United States and European external validation sets, the algorithms achieved agreements of 0.862 (quadratically weighted κ, 95% confidence interval (CI), 0.840–0.884) and 0.868 (95% CI, 0.835–0.900) with expert uropathologists. Successful generalization across different patient populations, laboratories and reference standards, achieved by a variety of algorithmic approaches, warrants evaluating AI-based Gleason grading in prospective clinical trials.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.