No abstract

The recent, rapid advances in immuno-oncology have revolutionized cancer treatment and spurred further research into tumor biology. Yet, cancer patients respond variably to immunotherapy despite mounting evidence to support its efficacy. Current methods for predicting immunotherapy response are unreliable, as these tests cannot fully account for tumor heterogeneity and microenvironment. An improved method for predicting response to immunotherapy is needed. Recent studies have proposed radiomics—the process of converting medical images into quantitative data (features) that can be processed using machine learning algorithms to identify complex patterns and trends—for predicting response to immunotherapy. Because patients undergo numerous imaging procedures throughout the course of the disease, there exists a wealth of radiological imaging data available for training radiomics models. And because radiomic features reflect cancer biology, such as tumor heterogeneity and microenvironment, these models have enormous potential to predict immunotherapy response more accurately than current methods. Models trained on preexisting biomarkers and/or clinical outcomes have demonstrated potential to improve patient stratification and treatment outcomes. In this review, we discuss current applications of radiomics in oncology, followed by a discussion on recent studies that use radiomics to predict immunotherapy response and toxicity.

Though overall death from opioid overdose are increasing in the United States, the death rate in some states and population groups is stabilizing or even decreasing. Several states have enacted a Naloxone Accessibility Laws to increase naloxone availability as an opioid antidote. The extent to which these laws permit layperson distribution and possession varies. The aim of this study is to investigate differences in provisions of Naloxone Accessibility Laws by states mainly in the Northeast and West regions, and the impact of naloxone availability on the rates of drug overdose deaths. This cross-sectional study was based on the National Vital Statistics System multiple cause-of-death mortality files. The average changes in drug overdose death rates between 2013 and 2017 in relevant states of the Northeast and West regions were compared according to availability of naloxone to laypersons. Seven states in the Northeast region and 10 states in the Western region allowed layperson distribution of naloxone. Layperson possession of naloxone was allowed in 3 states each in the Northeast and the Western regions. The average drug overdose death rates increased in many states in the both regions regardless of legalization of layperson naloxone distribution. The average death rates of 3 states that legalized layperson possession in the West region decreased (-0.33 per 100,000 person); however, in states in the West region that did not allow layperson possession and states in the Northeast region regardless of layperson possession increased between 2013 and 2017. The provision to legalize layperson possession of naloxone was associated with decreased average opioid overdose death rates in 3 states of the West region.

Background: Despite common use in clinical practice, the impact of blood transfusions on prognosis among patients with lung cancer remains unclear. The purpose of the current study is to perform an updated systematic review and meta-analysis to evaluate the influence of blood transfusions on survival outcomes of lung cancer patients.Methods: We searched PubMed, Embase, Cochrane Library, and Ovid MEDLINE for publications illustrating the association between blood transfusions and prognosis among people with lung cancer from inception to November 2019. Overall survival (OS) and disease-free survival (DFS) were the outcomes of interest. Pooled hazard ratios (HRs) with 95% confidence intervals (CIs) were computed using the randomeffects model. Study heterogeneity was evaluated with the I 2 test. Publication bias was explored via funnel plot and trim-and-fill analyses.Results: We included 23 cohort studies with 12,175 patients (3,027 cases and 9,148 controls) for metaanalysis. Among these records, 22 studies investigated the effect of perioperative transfusions, while one examined that of transfusions during chemotherapy. Two studies suggested the possible dose-dependent effect in accordance with the number of transfused units. In pooled analyses, blood transfusions deleteriously

Over the past decades, the field of machine learning (ML) has made great strides in medicine. Despite the number of ML-inspired publications in the clinical arena, the results and implications are not readily accepted at the bedside. Although ML is very powerful in deciphering hidden patterns in complex critical care and emergency medicine data, various factors including data, feature generation, model design, performance assessment, and limited implementation could affect the utility of the research. In this short review, a series of current challenges of adopting ML models to clinical research will be discussed.