Survival analysis of localized prostate cancer with deep learning.

In recent years, data-driven, deep-learning-based models have shown great promise in medical risk prediction. By utilizing the large-scale Electronic Health Record data found in the U.S. Department of Veterans Affairs, the largest integrated healthcare system in the United States, we have developed an automated, personalized risk prediction model to support the clinical decision-making process for localized prostate cancer patients. This method combines the representative power of deep learning and the analytical interpretability of parametric regression models and can implement both time-dependent and static input data. To collect a comprehensive evaluation of model performances, we calculate time-dependent C-statistics [Formula: see text] over 2-, 5-, and 10-year time horizons using either a composite outcome or prostate cancer mortality as the target event. The composite outcome combines the Prostate-Specific Antigen (PSA) test, metastasis, and prostate cancer mortality. Our longitudinal model Recurrent Deep Survival Machine (RDSM) achieved [Formula: see text] 0.85 (0.83), 0.80 (0.83), and 0.76 (0.81), while the cross-sectional model Deep Survival Machine (DSM) attained [Formula: see text] 0.85 (0.82), 0.80 (0.82), and 0.76 (0.79) for the 2-, 5-, and 10-year composite (mortality) outcomes, respectively. In addition to estimating the survival probability, our method can quantify the uncertainty associated with the prediction. The uncertainty scores show a consistent correlation with the prediction accuracy. We find PSA and prostate cancer stage information are the most important indicators in risk prediction. Our work demonstrates the utility of the data-driven machine learning model in prostate cancer risk prediction, which can play a critical role in the clinical decision system.

Leukocyte-specific protein 1 regulates T-cell migration in rheumatoid arthritis.

Copy number variations (CNVs) have been implicated in human diseases. However, it remains unclear how they affect immune dysfunction and autoimmune diseases, including rheumatoid arthritis (RA). Here, we identified a novel leukocyte-specific protein 1 (LSP1) deletion variant for RA susceptibility located in 11p15.5. We replicated that the copy number of LSP1 gene is significantly lower in patients with RA, which correlates positively with LSP1 protein expression levels. Differentially expressed genes in Lsp1-deficient primary T cells represent cell motility and immune and cytokine responses. Functional assays demonstrated that LSP1, induced by T-cell receptor activation, negatively regulates T-cell migration by reducing ERK activation in vitro. In mice with T-cell-dependent chronic inflammation, loss of Lsp1 promotes migration of T cells into the target tissues as well as draining lymph nodes, exacerbating disease severity. Moreover, patients with RA show diminished expression of LSP1 in peripheral T cells with increased migratory capacity, suggesting that the defect in LSP1 signaling lowers the threshold for T-cell activation. To our knowledge, our work is the first to demonstrate how CNVs result in immune dysfunction and a disease phenotype. Particularly, our data highlight the importance of LSP1 CNVs and LSP1 insufficiency in the pathogenesis of RA and provide previously unidentified insights into the mechanisms underlying T-cell migration toward the inflamed synovium in RA.