Non-adherence and non-persistence with antiretroviral treatment in Belgium: A real-world evaluation using a pharmacy database, 2018-2021.

PURPOSE: To evaluate the real-world rates of non-adherence and non-persistence to antiretroviral therapy (ART) among treatment-naïve adult patients with HIV after a 12-month follow-up period in Belgium. METHODS: A retrospective analysis of longitudinal pharmacy claims was conducted using the Pharmanet database from January 1, 2018, to December 31, 2021. Non-adherence was assessed over 12 months and reported as the proportion of days covered below the 80% threshold. Non-persistence was defined as the first 90-day gap in treatment between the two types of ART dispensed. Poisson regression with robust standard error and Cox proportional hazard models were used to assess the factors associated with non-adherence and non-persistence, respectively. RESULTS: Overall, 2999 patients were initiated on ART between 2018 and 2021. After a 12-month follow-up, the proportions of non-adherence and non-persistence were 35.6% and 15.9%, respectively in 2018, and decreased to 18.7% and 6.8%, respectively in 2021. Non-adherence was higher among women, Brussels residents, and those receiving multiple-tablet regimens (MTRs). Similarly, the prevalence of non-persistence was higher among women and MTR recipients. CONCLUSION: Among treatment-naïve adults with HIV in Belgium, non-adherence, and non-persistence to ART showed improvement over the study period but remained at high levels. Disparities were observed by sex and between geographical regions. Prioritizing strategies targeting women in Brussels and facilitating the transition from MTRs to single-tablet regimens should be emphasized optimize adherence to ART in Belgium.

New Approach for Targeted Treatment of Mild COVID-19 by Honeysuckle through Network Pharmacology Analysis.

OBJECTIVE: To investigate the potential pharmacological value of extracts from honeysuckle on patients with mild coronavirus disease 2019 (COVID-19) infection. METHODS: The active components and targets of honeysuckle were screened by Traditional Chinese Medicine Database and Analysis Platform (TCMSP). SwissADME and pkCSM databases predict pharmacokinetics of ingredients. The Gene Expression Omnibus (GEO) database collected transcriptome data for mild COVID-19. Data quality control, differentially expressed gene (DEG) identification, enrichment analysis, and correlation analysis were implemented by R toolkit. CIBERSORT evaluated the infiltration of 22 immune cells. RESULTS: The seven active ingredients of honeysuckle had good oral absorption and medicinal properties. Both the active ingredient targets of honeysuckle and differentially expressed genes of mild COVID-19 were significantly enriched in immune signaling pathways. There were five overlapping immunosignature genes, among which RELA and MAP3K7 expressions were statistically significant (P < 0.05). Finally, immune cell infiltration and correlation analysis showed that RELA, MAP3K7, and natural killer (NK) cell are with highly positive correlation and highly negatively correlated with hematopoietic stem cells. CONCLUSION: Our analysis suggested that honeysuckle extract had a safe and effective protective effect against mild COVID-19 by regulating a complex molecular network. The main mechanism was related to the proportion of infiltration between NK cells and hematopoietic stem cells.

Trends in Opioid Use Among Cancer Patients in the United States: 2013-2018.

Background: In response to the US opioid epidemic, the Centers for Disease Control and Prevention updated their guideline on prescription opioids for chronic pain management in March 2016. The aim of this study was to provide detailed analysis of trends in opioid claims among cancer patients in the United States during 2013-2018. Methods: We analyzed pharmaceutical dispensing data from Symphony Health's Integrated Dataverse database, which covers approximately 80% of the US population. We examined annual trends in dispensed opioids in cancer patients during 2013-2018. We examined quarterly trends of the prevalence, mean number of days, and dose (stated as morphine milligram equivalents) of opioid dispensing in cancer patients. Results: Dispensing records of an average of over 3.7 million cancer patients contributed to the study annually in 2013-2018. The annual prevalence of opioid dispensing claims declined from 40.2% in 2013 to 34.5% in 2018. Annual declines occurred across cancer sites, and particularly among patients with metastatic cancer (decline of 19.8%), breast cancer (18.2%), and lung cancer (13.8%). By quarter, the prevalence of opioid claims declined statistically significantly from 26.6% in Q1 2013 to 21.2% in Q4 2018; this decline was more pronounced after Q3 2016 (2-sided P = .004). Both quarterly trends in mean days and morphine milligram equivalents of opioids supplied showed a gradual decline from 2013 to 2018, with a slightly larger decline after 2016. Conclusions: We observed a decline in opioid use among cancer patients, particularly after 2016, coinciding with the publication of the Centers for Disease Control and Prevention's guideline on prescription opioids for chronic pain management.

Evaluation of the Expression Profile of Antibiotic-Induced Thrombocytopenia Using the Japanese Adverse Drug Event Report Database.

Drug-induced thrombocytopenia (DITP) can be triggered by antibiotics; however, the details remain unclear. Here, we evaluated the expression profiles of DITP using the Japanese Adverse Drug Event Report (JADER) database. We analyzed reports of DITP between April 2004 and January 2021 from the JADER database. The reporting odds ratio (ROR) and 95% confidence interval (CI) were used to detect DITP signals. Factors thought to affect DITP, such as male sex and an age of at least 60 years, were added as covariates. We evaluated the time-to-onset profile and hazard type using the Weibull shape parameter. The JADER database contained 1,048,576 reports. Twelve of 60 antibiotics showed signals for DITP; the RORs (95% CIs) for ampicillin/sulbactam, ceftazidime, cefozopran, ciprofloxacin, fluconazole, fos-fluconazole, linezolid, pazufloxacin, piperacillin/tazobactam, teicoplanin, trimethoprim/sulfamethoxazole, and voriconazole were 1.75 (1.41-2.16), 1.77 (1.42-2.18), 1.35 (1.06-1.72), 2.56 (2.19-2.98), 1.93 (1.67-2.23), 2.08 (1.76-2.46), 5.29 (2.73-9.60), 1.92 (1.51-2.41), 1.54 (1.05-2.19), 1.47 (1.16-1.84), 1.92 (1.73-2.14), and 2.32 (1.59-3.30), respectively. In multiple logistic regression analysis, 7 and 6 antibiotics were detected for the factors age and male sex, respectively. The median times-to-onset of DITP for ciprofloxacin (oral treatment), fluconazole, linezolid, piperacillin/tazobactam, and trimethoprim/sulfamethoxazole were 91, 91, 11.5, 10, and 9 days, respectively. Furthermore, the 95% CI of the Weibull shape parameter ß for these antibiotics was above and excluded 1, indicating that the antibiotics were the wear out failure type. We revealed the expression profiles of DITP following treatment with 12 antibiotics.

LUNAR :Drug Screening for Novel Coronavirus Based on Representation Learning Graph Convolutional Network.

An outbreak of COVID-19 that began in late 2019 was caused by a novel coronavirus(SARS-CoV-2). It has become a global pandemic. As of June 9, 2020, it has infected nearly 7 million people and killed more than 400,000, but there is no specific drug. Therefore, there is an urgent need to find or develop more drugs to suppress the virus. Here, we propose a new nonlinear end-to-end model called LUNAR. It uses graph convolutional neural networks to automatically learn the neighborhood information of complex heterogeneous relational networks and combines the attention mechanism to reflect the importance of the sum of different types of neighborhood information to obtain the representation characteristics of each node. Finally, through the topology reconstruction process, the feature representations of drugs and targets are forcibly extracted to match the observed network as much as possible. Through this reconstruction process, we obtain the strength of the relationship between different nodes and predict drug candidates that may affect the treatment of COVID-19 based on the known targets of COVID-19. These selected candidate drugs can be used as a reference for experimental scientists and accelerate the speed of drug development. LUNAR can well integrate various topological structure information in heterogeneous networks, and skillfully combine attention mechanisms to reflect the importance of neighborhood information of different types of nodes, improving the interpretability of the model. The area under the curve(AUC) of the model is 0.949 and the accurate recall curve (AUPR) is 0.866 using 10-fold cross-validation. These two performance indexes show that the model has superior predictive performance. Besides, some of the drugs screened out by our model have appeared in some clinical studies to further illustrate the effectiveness of the model.

The effect of generic market entry on antibiotic prescriptions in the United States.

When patented, brand-name antibiotics lose market exclusivity, generics typically enter the market at lower prices, which may increase consumption of the drug. To examine the effect of generic market entry on antibiotic consumption in the United States, we conducted an interrupted time series analysis of the change in the number of prescriptions per month for antibiotics for which at least one generic entered the US market between 2000 and 2012. Data were acquired from the IQVIA Xponent database. Thirteen antibiotics were analyzed. Here, we show that one year after generic entry, the number of prescriptions increased for five antibiotics (5 to 406%)-aztreonam, cefpodoxime, ciprofloxacin, levofloxacin, ofloxacin-and decreased for one drug: cefdinir. These changes were sustained two years after. Cefprozil, cefuroxime axetil and clarithromycin had significant increases in trend, but no significant level changes. No consistent pattern for antibiotic use following generic entry in the United States was observed.

Classification and Design of HIV-1 Integrase Inhibitors Based on Machine Learning.

A key enzyme in human immunodeficiency virus type 1 (HIV-1) life cycle, integrase (IN) aids the integration of viral DNA into the host DNA, which has become an ideal target for the development of anti-HIV drugs. A total of 1785 potential HIV-1 IN inhibitors were collected from the databases of ChEMBL, Binding Database, DrugBank, and PubMed, as well as from 40 references. The database was divided into the training set and test set by random sampling. By exploring the correlation between molecular descriptors and inhibitory activity, it is found that the classification and specific activity data of inhibitors can be more accurately predicted by the combination of molecular descriptors and molecular fingerprints. The calculation of molecular fingerprint descriptor provides the additional substructure information to improve the prediction ability. Based on the training set, two machine learning methods, the recursive partition (RP) and naive Bayes (NB) models, were used to build the classifiers of HIV-1 IN inhibitors. Through the test set verification, the RP technique accurately predicted 82.5% inhibitors and 86.3% noninhibitors. The NB model predicted 88.3% inhibitors and 87.2% noninhibitors with correlation coefficient of 85.2%. The results show that the prediction performance of NB model is slightly better than that of RP, and the key molecular segments are also obtained. Additionally, CoMFA and CoMSIA models with good activity prediction ability both were constructed by exploring the structure-activity relationship, which is helpful for the design and optimization of HIV-1 IN inhibitors.

Development of a Pediatric Relative Bioavailability/Bioequivalence Database and Identification of Putative Risk Factors Associated With Evaluation of Pediatric Oral Products.

Generally, bioequivalence (BE) studies of drug products for pediatric patients are conducted in adults due to ethical reasons. Given the lack of direct BE assessment in pediatric populations, the aim of this work is to develop a database of BE and relative bioavailability (relative BA) studies conducted in pediatric populations and to enable the identification of risk factors associated with certain drug substances or products that may lead to failed BE or different pharmacokinetic (PK) parameters in relative BA studies in pediatrics. A literature search from 1965 to 2020 was conducted in PubMed, Cochrane Library, and Google Scholar to identify BE studies conducted in pediatric populations and relative BA studies conducted in pediatric populations. Overall, 79 studies covering 37 active pharmaceutical ingredients (APIs) were included in the database: 4 bioequivalence studies with data that passed BE evaluations; 2 studies showed bioinequivalence results; 34 relative BA studies showing comparable PK parameters, and 39 relative BA studies showing differences in PK parameters between test and reference products. Based on the above studies, common putative risk factors associated with differences in relative bioavailability (DRBA) in pediatric populations include age-related absorption effects, high inter-individual variability, and poor study design. A database containing 79 clinical studies on BE or relative BA in pediatrics has been developed. Putative risk factors associated with DRBA in pediatric populations are summarized.

Identifying heart failure in patients with chronic obstructive lung disease through the Canadian Primary Care Sentinel Surveillance Network in British Columbia: a case derivation study.

BACKGROUND: Heart failure (HF) poses a substantial global health burden, particularly in patients with chronic obstructive pulmonary disease (COPD). The objective of this study was to validate an electronic medical record-based definition of HF in patients with COPD in primary care practices in the province of British Columbia, Canada. METHODS: We conducted a cross-sectional retrospective chart review from Sept. 1, 2018, to Dec. 31, 2018, for a cohort of patients from primary care practices in BC whose physicians were recruited through the BC node of the Canadian Primary Care Sentinel Surveillance Network. Heart failure case definitions were developed by combining diagnostic codes, medication information and laboratory values available in primary care electronic medical records. These were compared with HF diagnoses identified through detailed chart review as the gold standard. Sensitivity, specificity, negative (NPV) and positive predictive values (PPV) were calculated for each definition. RESULTS: Charts of 311 patients with COPD were reviewed, of whom 72 (23.2%) had HF. Five categories of definitions were constructed, all of which had appropriate sensitivity, specificity and NPV. The optimal case definition consisted of 1 HF billing code or a specific combination of medications for HF. This definition had an excellent specificity (93.3%, 95% confidence interval [CI] 89.4%-96.1%), sensitivity (90.3%, 95% CI 81.0%-96.0%), PPV (80.2%, 95% CI 69.9%-88.3%) and NPV (97.0%, 95% CI 93.8%-98.8%). INTERPRETATION: This comprehensive case definition improves upon previous primary care HF definitions to include medication codes and laboratory data, along with previously used billing codes. A case definition for HF was derived and validated and can be used with data from electronic medical records to identify HF in patients with COPD in primary care accurately.

Major Cardiac-Psychiatric Drug-Drug Interactions: a Systematic Review of the Consistency of Drug Databases.

PURPOSE: Major depressive disorder (MDD) and anxiety disorders (AD) are both highly prevalent among individuals with arrhythmia, ischemic heart disease, heart failure, hypertension, and dyslipidemia. There should be increased support for MDD and AD diagnosis and treatment in individuals with cardiac diseases, because treatment rates have been low. However, cardiac-psychiatric drug interaction can make pharmacologic treatment challenging. METHODS: The objective of the present systematic review was to investigate cardiac-psychiatric drug interactions in three different widely used pharmacological databases (Micromedex, Up to Date, and ClinicalKey). RESULTS: Among 4914 cardiac-psychiatric drug combinations, 293 significant interactions were found (6.0%). When a problematic interaction is detected, it may be easier to find an alternative cardiac medication (32.6% presented some interaction) than a psychiatric one (76.9%). Antiarrhythmics are the major class of concern. The most common problems produced by these interactions are related to cardiotoxicity (QT prolongation, torsades de pointes, cardiac arrest), increased exposure of cytochrome P450 2D6 (CYP2D6) substrates, or reduced renal clearance of organic cation transporter 2 (OCT2) substrates and include hypertensive crisis, increased risk of bleeding, myopathy, and/or rhabdomyolysis. CONCLUSION: Unfortunately, there is considerable inconsistency among the databases searched, such that a clinician's discretion and clinical experience remain invaluable tools for the management of patients with comorbidities present in psychiatric and cardiac disorders. The possibility of an interaction should be considered. With a multidisciplinary approach, particularly involving a pharmacist, the prescriber should be alerted to the possibility of an interaction. MDD and AD pharmacologic treatment in cardiac patients could be implemented safely both by cardiologists and psychiatrists. TRIAL REGISTRATION: PROSPERO Systematic Review Registration Number: CRD42018100424.

High hepatitis C treatment uptake among people with recent drug dependence in New South Wales, Australia.

BACKGROUND & AIMS: High HCV treatment uptake among people at most risk of transmission is essential to achieve elimination. We aimed to characterise subpopulations of people with HCV based on drug dependence, to estimate direct-acting antiviral (DAA) uptake in an unrestricted treatment era, and to evaluate factors associated with treatment uptake among people with recent drug dependence. METHODS: HCV notifications in New South Wales, Australia (1995-2017) were linked to opioid agonist therapy (OAT), hospitalisations, incarcerations, HIV notifications, deaths, and prescription databases. Drug dependence was defined as hospitalisation due to injectable drugs or receipt of OAT, with indicators in 2016-2018 considered recent. Records were weighted to account for spontaneous clearance. Logistic regression was used to analyse factors associated with treatment uptake among those with recent drug dependence. RESULTS: 57,467 people were estimated to have chronic HCV throughout the DAA era. Treatment uptake was highest among those with recent (47%), compared to those with distant (38%), and no (33%) drug dependence. Among those with recent drug dependence, treatment was more likely among those with HIV (adjusted odds ratio [aOR] 1.71; 95% CI 1.24-2.36), recent incarceration (aOR 1.10; 95% CI 1.01-1.19), and history of alcohol use disorder (aOR 1.22; 95% CI 1.13-1.31). Treatment was less likely among women (aOR 0.78; 95% CI 0.72-0.84), patients of Indigenous ethnicity (aOR 0.75; 95% CI 0.69-0.81), foreign-born individuals (aOR 0.86; 95% CI 0.78-0.96), those with outer-metropolitan notifications (aOR 0.90; 95% CI 0.82-0.98), HBV coinfection (aOR 0.69; 95% CI 0.59-0.80), and >1 recent hospitalisation (aOR: 0.91; 95% CI 0.84-0.98). CONCLUSIONS: These data provide evidence of high DAA uptake among people with recent drug dependence, including those who are incarcerated. Enhancing this encouraging initial uptake among high-risk populations will be essential to achieve HCV elimination. LAY SUMMARY: To facilitate HCV elimination, those at highest risk of infection and transmission are a treatment priority. This study shows the successes of Australia's universal provision of DAA therapy in reducing the barriers to treatment which have historically persisted among people who inject drugs. Despite higher DAA therapy uptake among those with recent drug dependence, gaps remain. Strategies which aim to reduce marginalisation and increase treatment uptake to ensure equitable HCV elimination must be advanced.

Scopy: an integrated negative design python library for desirable HTS/VS database design.

BACKGROUND: High-throughput screening (HTS) and virtual screening (VS) have been widely used to identify potential hits from large chemical libraries. However, the frequent occurrence of 'noisy compounds' in the screened libraries, such as compounds with poor drug-likeness, poor selectivity or potential toxicity, has greatly weakened the enrichment capability of HTS and VS campaigns. Therefore, the development of comprehensive and credible tools to detect noisy compounds from chemical libraries is urgently needed in early stages of drug discovery. RESULTS: In this study, we developed a freely available integrated python library for negative design, called Scopy, which supports the functions of data preparation, calculation of descriptors, scaffolds and screening filters, and data visualization. The current version of Scopy can calculate 39 basic molecular properties, 3 comprehensive molecular evaluation scores, 2 types of molecular scaffolds, 6 types of substructure descriptors and 2 types of fingerprints. A number of important screening rules are also provided by Scopy, including 15 drug-likeness rules (13 drug-likeness rules and 2 building block rules), 8 frequent hitter rules (four assay interference substructure filters and four promiscuous compound substructure filters), and 11 toxicophore filters (five human-related toxicity substructure filters, three environment-related toxicity substructure filters and three comprehensive toxicity substructure filters). Moreover, this library supports four different visualization functions to help users to gain a better understanding of the screened data, including basic feature radar chart, feature-feature-related scatter diagram, functional group marker gram and cloud gram. CONCLUSION: Scopy provides a comprehensive Python package to filter out compounds with undesirable properties or substructures, which will benefit the design of high-quality chemical libraries for drug design and discovery. It is freely available at https://github.com/kotori-y/Scopy.

COVID19 Drug Repository: text-mining the literature in search of putative COVID19 therapeutics.

The recent outbreak of COVID-19 has generated an enormous amount of Big Data. To date, the COVID-19 Open Research Dataset (CORD-19), lists ∼130,000 articles from the WHO COVID-19 database, PubMed Central, medRxiv, and bioRxiv, as collected by Semantic Scholar. According to LitCovid (11 August 2020), ∼40,300 COVID19-related articles are currently listed in PubMed. It has been shown in clinical settings that the analysis of past research results and the mining of available data can provide novel opportunities for the successful application of currently approved therapeutics and their combinations for the treatment of conditions caused by a novel SARS-CoV-2 infection. As such, effective responses to the pandemic require the development of efficient applications, methods and algorithms for data navigation, text-mining, clustering, classification, analysis, and reasoning. Thus, our COVID19 Drug Repository represents a modular platform for drug data navigation and analysis, with an emphasis on COVID-19-related information currently being reported. The COVID19 Drug Repository enables users to focus on different levels of complexity, starting from general information about (FDA-) approved drugs, PubMed references, clinical trials, recipes as well as the descriptions of molecular mechanisms of drugs' action. Our COVID19 drug repository provide a most updated world-wide collection of drugs that has been repurposed for COVID19 treatments around the world.

DrugCentral 2021 supports drug discovery and repositioning.

DrugCentral is a public resource (http://drugcentral.org) that serves the scientific community by providing up-to-date drug information, as described in previous papers. The current release includes 109 newly approved (October 2018 through March 2020) active pharmaceutical ingredients in the US, Europe, Japan and other countries; and two molecular entities (e.g. mefuparib) of interest for COVID19. New additions include a set of pharmacokinetic properties for ∼1000 drugs, and a sex-based separation of side effects, processed from FAERS (FDA Adverse Event Reporting System); as well as a drug repositioning prioritization scheme based on the market availability and intellectual property rights forFDA approved drugs. In the context of the COVID19 pandemic, we also incorporated REDIAL-2020, a machine learning platform that estimates anti-SARS-CoV-2 activities, as well as the 'drugs in news' feature offers a brief enumeration of the most interesting drugs at the present moment. The full database dump and data files are available for download from the DrugCentral web portal.

DockCoV2: a drug database against SARS-CoV-2.

The current state of the COVID-19 pandemic is a global health crisis. To fight the novel coronavirus, one of the best-known ways is to block enzymes essential for virus replication. Currently, we know that the SARS-CoV-2 virus encodes about 29 proteins such as spike protein, 3C-like protease (3CLpro), RNA-dependent RNA polymerase (RdRp), Papain-like protease (PLpro), and nucleocapsid (N) protein. SARS-CoV-2 uses human angiotensin-converting enzyme 2 (ACE2) for viral entry and transmembrane serine protease family member II (TMPRSS2) for spike protein priming. Thus in order to speed up the discovery of potential drugs, we develop DockCoV2, a drug database for SARS-CoV-2. DockCoV2 focuses on predicting the binding affinity of FDA-approved and Taiwan National Health Insurance (NHI) drugs with the seven proteins mentioned above. This database contains a total of 3,109 drugs. DockCoV2 is easy to use and search against, is well cross-linked to external databases, and provides the state-of-the-art prediction results in one site. Users can download their drug-protein docking data of interest and examine additional drug-related information on DockCoV2. Furthermore, DockCoV2 provides experimental information to help users understand which drugs have already been reported to be effective against MERS or SARS-CoV. DockCoV2 is available at https://covirus.cc/drugs/.

Homotherapy for heteropathy active components and mechanisms of Qiang-Huo-Sheng-Shi decoction for treatment of rheumatoid arthritis and osteoarthritis.

Qiang-Huo-Sheng-Shi decoction (QHSSD), a classic traditional Chinese herbal formula, which has been reported to be effective in rheumatoid arthritis (RA) and osteoarthritis (OA). However, the concurrent targeting mechanism of how the aforementioned formula is valid in the two distinct diseases OA and RA, which represents the homotherapy-for-heteropathy principle in traditional Chinese medicine (TCM), have not yet been clarified. In the present study, network pharmacology was adopted to analyze the potential molecular mechanism, and therapeutic effective components of QHSSD on both OA and RA. A total of 153 active ingredients in QHSSD were identified, 142 of which associated with 59 potential targets for the two diseases were identified. By constructing the protein-protein interaction network and the compound-target-disease network, 72 compounds and 10 proteins were obtained as the hub targets of QHSSD against OA and RA. The hub genes of ESR1, PTGS2, PPARG, IL1B, TNF, MMP2, IL6, CYP3A4, MAPK8, and ALB were mainly involved in osteoclast differentiation, the NF-κB and TNF signaling pathways. Moreover, molecular docking results showed that the screened active compounds had a high affinity for the hub genes. This study provides new insight into the molecular mechanisms behind how QHSSD presents homotherapy-for-heteropathy therapeutic efficacy in both OA and RA. For the first time, a two-disease model was linked with a TCM formula using network pharmacology to identify the key active components and understand the common mechanisms of its multi-pathway regulation. This study will inspire more innovative and important studies on the modern research of TCM formulas.

Comparison of FDA accelerated vs regular pathway approvals for lung cancer treatments between 2006 and 2018.

Regulatory agencies around the world have been using flexible requirements for approval of new drugs, especially for cancer drugs. The US Food and Drug Administration (FDA) is mostly the first agency to approve new drugs worldwide, mainly due to the faster terms of the accelerated pathway and breakthrough therapy designation. Surrogate endpoints and preliminary data (e.g. single-arm and phase 2 studies) are used for these new approvals, however larger effect sizes are expected. We aim to compare FDA Accelerated vs Regular Pathway approvals and Breakthrough therapy designations (BTD) for lung cancer treatments between 2006 and 2018 regarding study design, sample size, outcome measures and effect size. We assessed the FDA database to collect data from studies that formed the basis of approvals of new drugs or indications for lung cancer spanning from 2006 to 2018. We found that accelerated pathway approvals are based on significantly more single-arm studies with small sample sizes and surrogate primary endpoints. However, effect size was not different between the pathways. A large proportion of studies used to support regular pathway approvals also showed these characteristics that are related to low quality and uncertain evidence. Compared to other approvals, BTD were more frequently based on single-arm studies. There was no significant difference in use of surrogate endpoints or sample size. 44% of BTD were based on studies demonstrating large effect sizes, proportionally more than approvals not receiving this designation. In conclusion, based on the indicators of evidence quality we extracted, criteria's for granting accelerated approval and breakthrough therapy designation seen not clear. Faster approvals are in the majority full of uncertainties which should be viewed with caution and the patient have to be communicated to allow shared decision making. Post-marketing validation is essential.

Using Supervised Learning Methods to Develop a List of Prescription Medications of Greatest Concern during Pregnancy.

INTRODUCTION: Women and healthcare providers lack adequate information on medication safety during pregnancy. While resources describing fetal risk are available, information is provided in multiple locations, often with subjective assessments of available data. We developed a list of medications of greatest concern during pregnancy to help healthcare providers counsel reproductive-aged and pregnant women. METHODS: Prescription drug labels submitted to the U.S. Food and Drug Administration with information in the Teratogen Information System (TERIS) and/or Drugs in Pregnancy and Lactation by Briggs & Freeman were included (N = 1,186 medications; 766 from three data sources, 420 from two). We used two supervised learning methods ('support vector machine' and 'sentiment analysis') to create prediction models based on narrative descriptions of fetal risk. Two models were created per data source. Our final list included medications categorized as 'high' risk in at least four of six models (if three data sources) or three of four models (if two data sources). RESULTS: We classified 80 prescription medications as being of greatest concern during pregnancy; over half were antineoplastic agents (n = 24), angiotensin converting enzyme inhibitors (n = 10), angiotensin II receptor antagonists (n = 8), and anticonvulsants (n = 7). DISCUSSION: This evidence-based list could be a useful tool for healthcare providers counseling reproductive-aged and pregnant women about medication use during pregnancy. However, providers and patients may find it helpful to weigh the risks and benefits of any pharmacologic treatment for both pregnant women and the fetus when managing medical conditions before and during pregnancy.

A Modified Skip-Gram Algorithm for Extracting Drug-Drug Interactions from AERS Reports.

Drug-drug interactions (DDIs) are one of the indispensable factors leading to adverse event reactions. Considering the unique structure of AERS (Food and Drug Administration Adverse Event Reporting System (FDA AERS)) reports, we changed the scope of the window value in the original skip-gram algorithm, then propose a language concept representation model and extract features of drug name and reaction information from large-scale AERS reports. The validation of our scheme was tested and verified by comparing with vectors originated from the cooccurrence matrix in tenfold cross-validation. In the verification of description enrichment of the DrugBank DDI database, accuracy was calculated for measurement. The average area under the receiver operating characteristic curve of logistic regression classifiers based on the proposed language model is 6% higher than that of the cooccurrence matrix. At the same time, the average accuracy in five severe adverse event classes is 88%. These results indicate that our language model can be useful for extracting drug and reaction features from large-scale AERS reports.