Prediction of Potential Drug-Disease Associations through Deep Integration of Diversity and Projections of Various Drug Features.

Int J Mol Sci. 2019 Aug 22;20(17):

Authors: Xuan P, Song Y, Zhang T, Jia L

Abstract
Identifying new indications for existing drugs may reduce costs and expedites drug development. Drug-related disease predictions typically combined heterogeneous drug-related and disease-related data to derive the associations between drugs and diseases, while recently developed approaches integrate multiple kinds of drug features, but fail to take the diversity implied by these features into account. We developed a method based on non-negative matrix factorization, DivePred, for predicting potential drug-disease associations. DivePred integrated disease similarity, drug-disease associations, and various drug features derived from drug chemical substructures, drug target protein domains, drug target annotations, and drug-related diseases. Diverse drug features reflect the characteristics of drugs from different perspectives, and utilizing the diversity of multiple kinds of features is critical for association prediction. The various drug features had higher dimensions and sparse characteristics, whereas DivePred projected high-dimensional drug features into the low-dimensional feature space to generate dense feature representations of drugs. Furthermore, DivePred's optimization term enhanced diversity and reduced redundancy of multiple kinds of drug features. The neighbor information was exploited to infer the likelihood of drug-disease associations. Experiments indicated that DivePred was superior to several state-of-the-art methods for prediction drug-disease association. During the validation process, DivePred identified more drug-disease associations in the top part of prediction result than other methods, benefitting further biological validation. Case studies of acetaminophen, ciprofloxacin, doxorubicin, hydrocortisone, and ampicillin demonstrated that DivePred has the ability to discover potential candidate disease indications for drugs.

PMID: 31443472 [PubMed - indexed for MEDLINE]

Effectiveness of corticosteroid compared with non-corticosteroid therapy for the treatment of drug-induced acute interstitial nephritis: a systematic review.

Intern Med J. 2019 05;49(5):562-569

Authors: Quinto LR, Sukkar L, Gallagher M

Abstract
Corticosteroids may hasten recovery in drug-induced acute interstitial nephritis (DI-AIN). However, there is no consensus regarding the effectiveness of corticosteroid compared with non-corticosteroid therapy. We conducted a systematic review of the literature according to Preferred Reporting in Systematic Reviews and Meta-Analysis (PRISMA) guidelines using MEDLINE, EMBASE, CINAHL, Scopus and Web of Science from inception until November 2017 using predefined search terms. Studies that compared the effects of corticosteroid therapy versus non-corticosteroid therapy in the treatment of DI-AIN were included. Outcomes were change in serum creatinine, adverse drug reactions, need for renal replacement therapy (RRT) and death. Due to considerable heterogeneity, a meta-analysis was not performed. There were no randomised controlled trials. Eight retrospective studies met inclusion criteria, with 430 patients (300 received and 130 did not receive corticosteroid therapy) and a median age of 57 (range 29-75) and 58 (22-76) years respectively. When treatment details were reported, prednisone was commenced at 40-60 mg daily in five studies, and two studies commenced intravenous methylprednisolone 1 mg/kg with a treatment duration of 1.5-12 weeks. Non-corticosteroid therapy was poorly defined across all studies. Four studies showed no difference in serum creatinine between corticosteroid and comparator arms, while four studies found a benefit. Adverse drug reactions, need for RRT and deaths were infrequently reported. Risk of bias was high across all domains. The limited evidence does not support the use of corticosteroids in the treatment of DI-AIN. Larger, well-designed trials are needed to help guide clinical management of this condition.

PMID: 30129289 [PubMed - indexed for MEDLINE]

Efficacy of cattle encephalon glycoside and ignotin in patients with acute cerebral infarction: a randomized, double-blind, parallel-group, placebo-controlled study.

Neural Regen Res. 2020 Jul;15(7):1266-1273

Authors: Zhang H, Li CL, Wan F, Wang SJ, Wei XE, Hao YL, Leng HL, Li JM, Yan ZR, Wang BJ, Xu RS, Yu TM, Zhou LC, Fan DS

Abstract
Cattle encephalon glycoside and ignotin (CEGI) injection is a compound preparation formed by a combination of muscle extract from healthy rabbits and brain gangliosides from cattle, and it is generally used as a neuroprotectant in the treatment of central and peripheral nerve injuries. However, there is still a need for high-level clinical evidence from large samples to support the use of CEGI. We therefore carried out a prospective, multicenter, randomized, double-blind, parallel-group, placebo-controlled study in which we recruited 319 patients with acute cerebral infarction from 16 centers in China from October 2013 to May 2016. The patients were randomized at a 3:1 ratio into CEGI (n = 239; 155 male, 84 female; 61.2 ± 9.2 years old) and placebo (n = 80; 46 male, 34 female; 63.2 ± 8.28 years old) groups. All patients were given standard care once daily for 14 days, including a 200 mg aspirin enteric-coated tablet and 20 mg atorvastatin calcium, both taken orally, and intravenous infusion of 250-500 mL 0.9% sodium chloride containing 40 mg sodium tanshinone IIA sulfonate. Based on conventional treatment, patients in the CEGI and placebo groups were given 12 mL CEGI or 12 mL sterile water, respectively, in an intravenous drip of 250 mL 0.9% sodium chloride (2 mL/min) once daily for 14 days. According to baseline National Institutes of Health Stroke Scale scores, patients in the two groups were divided into mild and moderate subgroups. Based on the modified Rankin Scale results, the rate of patients with good outcomes in the CEGI group was higher than that in the placebo group, and the rate of disability in the CEGI group was lower than that in the placebo group on day 90 after treatment. In the CEGI group, neurological deficits were decreased on days 14 and 90 after treatment, as measured by the National Institutes of Health Stroke Scale and the Barthel Index. Subgroup analysis revealed that CEGI led to more significant improvements in moderate stroke patients. No drug-related adverse events occurred in the CEGI or placebo groups. In conclusion, CEGI may be a safe and effective treatment for acute cerebral infarction patients, especially for moderate stroke patients. This study was approved by the Ethical Committee of Peking University Third Hospital, China (approval No. 2013-068-2) on May 20, 2013, and registered in the Chinese Clinical Trial Registry (registration No. ChiCTR1800017937).

PMID: 31960812 [PubMed]

Liability and Failure to Warn a Patient.

Continuum (Minneap Minn). 2019 Aug;25(4):1141-1144

Authors: Kass JS, Rose RV

Abstract
This medicolegal article examines a physician's liability when he or she has knowledge of adverse effects associated with a prescription medication and suggests ways to mitigate that liability risk. The article also discusses the circumstances under which pharmaceutical companies face liability for side effects such as tardive dyskinesia.

PMID: 31356297 [PubMed - indexed for MEDLINE]

The safety and immunogenicity of a cell-derived adjuvanted H5N1 vaccine - A phase I randomized clinical trial.

J Microbiol Immunol Infect. 2019 Oct;52(5):685-692

Authors: Cheng A, Hsieh SM, Pan SC, Li YH, Hsieh EF, Lee HC, Lin TW, Lai KL, Chen C, Shi-Chung Chang S, Chang SC

Abstract
BACKGROUND: Development of an efficacious egg-free mock-up H5N1 vaccine is key to our preparedness against pandemic avian flu.
METHODS: This is a single-center, randomized, observer-blinded phase I clinical trial evaluating the safety and immunogenicity of an alum-adjuvanted Madin-Darby canine kidney (MDCK)-derived inactivated whole-virion H5N1 influenza vaccine in healthy adults. Hemagglutination inhibition (HAI) and neutralizing antibody titers were measured using horse and turkey red blood cells (RBCs).
RESULTS: Thirty-six adult subjects were randomized to receive two doses of 0.5 mL of the MDCK-derived H5N1 alum-adjuvanted vaccine containing 7.5, 15, or 30 μg of hemagglutinin (HA) 21 days apart. The candidate vaccine was well tolerated and safe across the three dosing groups. The most frequent adverse event was injection site pain (46.5%). Both HAI and neutralizing antibody titers increased after each vaccination in all three dosing groups. The best HAI responses, namely a seroconversion rate of 91.7% and a geometric mean ratio of 9.51 were achieved with the HA dose of 30 μg assayed using horse RBCs at day 42. HAI titers against H5N1 avian influenza virus was significantly higher when measured using horse RBCs compared with turkey RBCs.
CONCLUSIONS: This Phase I trial showed the MDCK-derived H5N1 candidate vaccine is safe and immunogenic. The source of RBCs has a significant impact on the measurement of HAI titers (ClinicalTrials.gov number: NCT01675284.).

PMID: 31255574 [PubMed - indexed for MEDLINE]

Hypoperfusion in lenticulostriate arteries territory related to unexplained early neurological deterioration after intravenous thrombolysis.

Int J Stroke. 2019 04;14(3):306-309

Authors: Zhou Y, Zhong W, Wang A, Huang W, Yan S, Zhang R, Liu C, Fu J, Jiaerken Y, Lou M

Abstract
BACKGROUND: Early neurological deterioration occurs in approximately 10% acute ischemic stroke patients after thrombolysis. Over half of the early neurological deterioration occurred without known causes and is called unexplained early neurological deterioration.
AIMS: We aimed to explore the development of early neurological deterioration at 24 h after thrombolysis, and whether it could be predicted by the presence of baseline hypoperfusion in lenticulostriate arteries territory in acute ischemic stroke patients.
METHODS: We retrospectively reviewed our prospectively collected database of acute ischemic stroke patients in the unilateral middle cerebral artery territory who had baseline perfusion image and received thrombolysis. Unexplained early neurological deterioration was defined as ≥ 2 points increase of National Institutes of Health Stroke Scale (NIHSS) from baseline to 24 h, without known causes. Hypoperfusion lesions in different territories were identified on perfusion maps.
RESULTS: A total of 306 patients were included in analysis. Patients with pure lenticulostriate arteries hypoperfusion (defined as the presence of hypoperfusion in lenticulostriate artery territory, but not in middle cerebral artery terminal branch territory) were more likely to have unexplained early neurological deterioration than others (27.6% vs. 6.1%; OR, 5.974; p = 0.001), after adjusting for age, baseline NIHSS and onset to treatment time.
CONCLUSIONS: Patients presenting hypoperfusion in pure lenticulostriate arteries territory were easier to experience unexplained early neurological deterioration.

PMID: 30777811 [PubMed - indexed for MEDLINE]

Electronic Systems for Patients to Report and Manage Side Effects of Cancer Treatment: Systematic Review.

J Med Internet Res. 2019 01 24;21(1):e10875

Authors: Warrington L, Absolom K, Conner M, Kellar I, Clayton B, Ayres M, Velikova G

Abstract
BACKGROUND: There has been a dramatic increase in the development of electronic systems to support cancer patients to report and manage side effects of treatment from home. Systems vary in the features they offer to patients, which may affect how patients engage with them and how they improve patient-centered outcomes.
OBJECTIVE: This review aimed to (1) describe the features and functions of existing electronic symptom reporting systems (eg, symptom monitoring, tailored self-management advice), and (2) explore which features may be associated with patient engagement and patient-centered outcomes.
METHODS: The review was registered with the International Prospective Register of Systematic Reviews (PROSPERO) and followed guidelines from the Centre for Reviews and Dissemination (University of York, United Kingdom). Primary searches were undertaken of MEDLINE, Embase, PsycInfo, Web of Science, Cochrane Central Register of Controlled Trials, and the Health Technology Assessment databases. Secondary searches were undertaken by screening reference lists and citations. Two researchers applied broad inclusion criteria to identify and select relevant records. Data were extracted and summarized using Microsoft Excel. In order to meet the aims, the study selection, data extraction, and data synthesis comprised two stages: (1) identifying and characterizing available systems and (2) summarizing data on patient engagement and patient-centered outcomes.
RESULTS: We identified 77 publications relating to 41 distinct systems. In Stage 1, all publications were included (N=77). The features identified that supported clinicians and care were facility for health professionals to remotely access and monitor patient-reported data (24/41, 58%) and function to send alerts to health professionals for severe symptoms (17/41, 41%). Features that supported patients were facility for patients to monitor/review their symptom reports over time (eg, graphs) (19/41, 46%), general patient information about cancer treatment and side effects (17/41, 41%), tailored automated patient advice on symptom management (12/41, 29%), feature for patients to communicate with the health care team (6/41, 15%), and a forum for patients to communicate with one another (4/41, 10%). In Stage 2, only publications that included some data on patient engagement or patient-centered outcomes were included (N=29). A lack of consistency between studies in how engagement was defined, measured, or reported, and a wide range of methods chosen to evaluate systems meant that it was not possible to compare across studies or make conclusions on relationships with system features.
CONCLUSIONS: Electronic systems have the potential to help patients manage side effects of cancer treatment, with some evidence to suggest a positive effect on patient-centered outcomes. However, comparison across studies is difficult due to the wide range of assessment tools used. There is a need to develop guidelines for assessing and reporting engagement with systems, and a set of core outcomes for evaluation. We hope that this review will contribute to the field by introducing a taxonomy for characterizing system features.
TRIAL REGISTRATION: PROSPERO CRD42016035915; www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42016035915.

PMID: 30679145 [PubMed - indexed for MEDLINE]

15 new pubmed citations were retrieved for your search. Click on the search hyperlink below to display the complete search results:

("drug-induced" OR "drug-related") AND ("adverse events" OR "side effects" OR "side-effects")

These pubmed results were generated on 2020/01/21

PubMed comprises more than millions of citations for biomedical literature from MEDLINE, life science journals, and online books. Citations may include links to full-text content from PubMed Central and publisher web sites.

15 new pubmed citations were retrieved for your search. Click on the search hyperlink below to display the complete search results:

("drug-induced" OR "drug-related") AND ("adverse events" OR "side effects" OR "side-effects")

These pubmed results were generated on 2020/01/21

PubMed comprises more than millions of citations for biomedical literature from MEDLINE, life science journals, and online books. Citations may include links to full-text content from PubMed Central and publisher web sites.

Safety and efficacy of nazartinib (EGF816) in adults with EGFR-mutant non-small-cell lung carcinoma: a multicentre, open-label, phase 1 study.

Lancet Respir Med. 2020 Jan 15;:

Authors: Tan DS, Leighl NB, Riely GJ, Yang JC, Sequist LV, Wolf J, Seto T, Felip E, Aix SP, Jonnaert M, Pan C, Tan EY, Ko J, Moody SE, Kim DW

Abstract
BACKGROUND: Resistance to first-generation and second-generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) is mediated by the emergence of the Thr790Met mutation in 50-60% of treated patients with non-small-cell lung cancer (NSCLC). We aimed to assess the safety and activity of nazartinib (EGF816), a third-generation EGFR TKI that selectively inhibits EGFR with Thr790Met or activating mutations (or both), while sparing wild-type EGFR, in patients with advanced EGFR-mutant NSCLC.
METHODS: This phase 1 dose-escalation part of an open-label, multicentre, phase 1/2 study was conducted at nine academic medical centres located in Europe, Asia, and North America. Patients were included if they were aged 18 years or older and had stage IIIB-IV EGFR-mutant NSCLC (with varying statuses of EGFR mutation and previous therapy allowed), at least one measurable lesion, and an Eastern Cooperative Oncology Group (ECOG) performance status of 2 or less. Nazartinib (at seven dose levels between 75 mg and 350 mg, in capsule or tablet form) was administered orally, once daily, on a continuous 28-day dosing schedule. A two-parameter Bayesian logistic regression model, guided by the escalation with overdose control principle, was implemented to make dose recommendations and estimate the maximum tolerated dose or recommended phase 2 dose of nazartinib (the primary outcome). This study is registered with ClinicalTrials.gov (NCT02108964); enrolment to phase 1 is complete and the study is ongoing.
FINDINGS: By Aug 31, 2017, 180 patients (116 [64%] women; median age 60 years (52-69); 116 [64%] with ECOG performance status 1) received nazartinib across seven dose levels: 75 mg (n=17), 100 mg (n=38), 150 mg (n=73), 200 mg (n=8), 225 mg (n=28), 300 mg (n=5), and 350 mg (n=11). Seven dose-limiting toxicities were observed in six (3%) patients who received 150 mg, 225 mg, or 350 mg nazartinib once daily. Although the maximum tolerated dose was not met, the recommended phase 2 dose was declared as 150 mg once daily (tablet). The most common adverse events, regardless of cause, were rash (all subcategories 111 [62%] patients, maculopapular rash 72 [40%], dermatitis acneiform 22 [12%]), diarrhoea (81 [45%]), pruritus (70 [39%]), fatigue (54 [30%]), and stomatitis (54 [30%]), and were mostly grades 1-2. Any-cause grade 3-4 adverse events were reported in 99 (55%) patients across all doses, the most common being rash (all subcategories grouped 27 [15%]), pneumonia (12 [7%]), anaemia (ten [6%]), and dyspnoea (nine [5%]). Serious adverse events suspected to be drug-related occurred in 16 (9%) patients.
INTERPRETATION: Nazartinib has a favourable safety profile, with low-grade skin toxicity characterised by a predominantly maculopapular rash that required minimal dose reductions.
FUNDING: Novartis Pharmaceuticals Corporation.

PMID: 31954624 [PubMed - as supplied by publisher]

Safety of local anesthetics.

An Bras Dermatol. 2019 Dec 18;:

Authors: Cherobin ACFP, Tavares GT

Abstract
Local anesthetics are essential medications for the conduction of dermatological procedures. They stop the depolarization of nerve fibers and are divided into two main categories, the amide and ester types. Systemic toxicity with reflex on the central nervous and cardiovascular systems is their most feared adverse reactions, and the anaphylactic reaction is the most concerning one. Although potentially fatal, these events are extremely rare, so local anesthetics are considered safe for use in in-office procedures.

PMID: 31952994 [PubMed - as supplied by publisher]

Risk of Acute Kidney Injury Associated With Medication Administration in the Emergency Department.

J Emerg Med. 2020 Jan 15;:

Authors: Hinson JS, Ehmann MR, Al Jalbout N, Ortmann MJ, Zschoche J, Klein EY

Abstract
BACKGROUND: Patients who develop acute kidney injury (AKI) have a 2-fold increased risk for major adverse events within 1 year. An estimated 19-26% of all cases of hospital-acquired AKI may be attributable to drug-induced kidney disease (DIKD). Patients evaluated in the emergency department (ED) are often prescribed potentially nephrotoxic drugs, yet the role of ED prescribing in DIKD is unknown.
OBJECTIVE: We sought to measure the association between ED medication administration and development of AKI.
METHODS: This was a retrospective 5-year cohort analysis at a single center. Patients with a serum creatinine measurement at presentation in the ED and 24-168 h later were included. Outcome was incidence of AKI as defined by Kidney Disease Improving Global Outcomes criteria in the 7 days after ED evaluation. Medication administration risk was estimated using Cox proportional hazards model.
RESULTS: There were 46,965 ED encounters by 30,407 patients included in the study, of which 6461 (13.8%) patients met the criteria for AKI. For hospitalized patients, administration of a potentially nephrotoxic medication was associated with increased risk of AKI (hazard ratio [HR] 1.30 [95% confidence interval {CI} 1.20-1.41]). Diuretics were associated with the largest risk of AKI (HR 1.64 [95% CI 1.52-1.78]), followed by angiotensin-converting enzyme inhibitors (HR 1.39 [95% CI 1.26-1.54]) and antibiotics (HR 1.13 [95% CI 1.05-1.22]). For discharged patients, administration of antibiotics was strongly associated with increased risk of AKI (HR 3.19 [95% CI 1.08-9.43]).
CONCLUSION: ED administration of potentially nephrotoxic medications was associated with an increased risk of AKI in the following 7 days. Diuretics, angiotensin-converting enzyme inhibitors, and antibiotics were independently associated with increased risk of AKI. Nephroprotective practices in the ED may mitigate kidney injury and long-term adverse outcomes.

PMID: 31952871 [PubMed - as supplied by publisher]

A systematic review on the emerging association between the occurrence of immune-related adverse events and clinical outcomes with checkpoint inhibitors in advanced cancer patients.

Semin Oncol. 2019 Aug - Oct;46(4-5):362-371

Authors: Cortellini A, Buti S, Agostinelli V, Bersanelli M

Abstract
The correlation between clinical outcomes and treatment-related adverse events (AEs) has always been a debated topic in clinical oncology. Despite toxicities pharmacodynamics effects, the misunderstanding has always been around the corner: AEs themselves could lead to morbidity and mortality; on the other hand, the choice of the clinical outcomes to measure is not univocal. After the advent of immune checkpoint inhibitors (ICIs), such as anti-cytotoxic T-lymphocyte-associated protein 4 and anti-programmed death-1/programmed death ligand-1, new class-specific AEs have emerged, called immune-related AEs (irAEs). With irAEs, the correlation between toxicity and clinical outcomes has suddenly been suggested, but it is still to be proven. We conducted a systematic literature review regarding this emerging association, pointing out all the available data and speculating on the possible underlying mechanisms. Thirty-six studies were included in the analysis, involving different malignancies (mostly melanoma and lung cancer), with different measured clinical outcomes. The most of them were retrospective. Despite the high heterogeneity, and the enormous biases of the revised studies, we can assume that irAEs occurrence is linked to the therapeutic activity of immune checkpoint inhibitors, with a (certain) direct proportionality, maybe subtending the likelihood of an immunogenic phenotype. This phenomenon seems to occur with both anti-cytotoxic T-lymphocyte-associated protein 4 and anti-programmed death-1/programmed death ligand-1, and across different solid malignancies.

PMID: 31727344 [PubMed - indexed for MEDLINE]

Local Anesthetic Systemic Toxicity: A Narrative Literature Review and Clinical Update on Prevention, Diagnosis, and Management.

Plast Reconstr Surg. 2019 09;144(3):783-795

Authors: Gitman M, Fettiplace MR, Weinberg GL, Neal JM, Barrington MJ

Abstract
BACKGROUND: The objective of this narrative review of local anesthetic systemic toxicity is to provide an update on its prevention, diagnosis, and management.
METHODS: The authors used a MEDLINE search of human studies, animal studies, and case reports and summarize findings following the American Society of Regional Anesthesia and Pain Medicine practice advisories on local anesthetic systemic toxicity.
RESULTS: Between March of 2014 and November of 2016, there were 47 cases of systemic toxicity described. Twenty-two patients (47 percent) were treated with intravenous lipid emulsion and two patients (4.3 percent) died. Seizures were the most common presentation. The spectrum of presenting neurologic and cardiovascular symptoms and signs are broad and can be obscured by perioperative processes. Local anesthetic type, dosage, and volume; site of injection; and patient comorbidities influence the rate of absorption from the site of injection and biodegradation of local anesthetics. Consider discussing appropriate dosages as a component of the surgical "time-out." A large-volume depot of dilute local anesthetic can take hours before reaching peak plasma levels. Oxygenation, ventilation, and advanced cardiac life support are the first priorities in treatment. Lipid emulsion therapy should be given at the first sign of serious systemic toxicity with an initial bolus dose of 100 ml for adults weighing greater than 70 kg and 1.5 ml/kg for adults weighing less than 70 kg or for children.
CONCLUSION: All physicians who administer local anesthetics should be educated regarding the nature of systemic toxicity and contemporary management algorithms that include lipid emulsion therapy.

PMID: 31461049 [PubMed - indexed for MEDLINE]

Inotersen treatment for ATTR amyloidosis.

Amyloid. 2019;26(sup1):27-28

Authors: Benson MD

PMID: 31343345 [PubMed - indexed for MEDLINE]

Polypharmacy Rates among Patients over 45 years

Ir Med J. 2019 03 14;112(2):893

Authors: Tatum T, Curry P, Dunne B, Walsh K, Bennett K

Abstract
Polypharmacy, defined as receipt of ≥5 medications in any one month, is often associated with potentially inappropriate prescribing and adverse drug interactions. High levels of polypharmacy have been observed internationally and in Ireland. The Health Service Executive Primary Care Reimbursement Services (HSE-PCRS) pharmacy claims database for the GMS eligible population was used. We conducted Chi-square tests to determine the statistical significance of perceived differences in medication use among patients aged ³ 45 years. Our results establish a national benchmark for polypharmacy in gender and various age categories in the HSE-PCRS. Of the 794,628 individuals aged ≥45 years with at least one claim in 2013, 64.3% (510,946) had polypharmacy, with higher rates among women (67.0% - 293,886 - compared to 60.8% of men - 216,444). Patients aged 45-54 years were less likely to have polypharmacy (38.6% - 69,934) compared to those aged 75 years old (82.6% - 197,565). The high levels of polypharmacy are of interest, and suggest that monitoring and evaluation of patients’ medication regimes may be required to ensure appropriateness.

PMID: 31045333 [PubMed - indexed for MEDLINE]

The 10th International Congress on Cutaneous Adverse Drug Reactions, Shimane, Japan, 2018: Focus on New Discoveries.

Drug Saf. 2019 06;42(6):797-801

Authors: Olteanu C, Shear NH, Morita E, Chung WH, Niihara H, Matsukura S, Hashimoto R, Dodiuk-Gad RP

PMID: 31037651 [PubMed - indexed for MEDLINE]

ADMET modeling approaches in drug discovery.

Drug Discov Today. 2019 05;24(5):1157-1165

Authors: Ferreira LLG, Andricopulo AD

Abstract
In silico prediction of ADMET is an important component of pharmaceutical R&D. Last year, the FDA approved 59 new molecular entities, with small molecules comprising 64% of the therapies approved in 2018. Estimation of pharmacokinetic properties in the early phases of drug discovery has been central to guiding hit-to-lead and lead-optimization efforts. Given the outstanding complexity of the current R&D model, drug discovery players have intensely pursued molecular modeling strategies to identify patterns in ADMET data and convert them into knowledge. The field has advanced alongside the progress of chemoinformatics, which has evolved from traditional chemometrics to advanced machine learning methods.

PMID: 30890362 [PubMed - indexed for MEDLINE]

A Comparison Study of Algorithms to Detect Drug-Adverse Event Associations: Frequentist, Bayesian, and Machine-Learning Approaches.

Drug Saf. 2019 06;42(6):743-750

Authors: Pham M, Cheng F, Ramachandran K

Abstract
INTRODUCTION: It is important to monitor the safety profile of drugs, and mining for strong associations between drugs and adverse events is an effective and inexpensive method of post-marketing safety surveillance.
OBJECTIVE: The objective of our work was to compare the accuracy of both common and innovative methods of data mining for pharmacovigilance purposes.
METHODS: We used the reference standard provided by the Observational Medical Outcomes Partnership, which contains 398 drug-adverse event pairs (165 positive controls, 233 negative controls). Ten methods and algorithms were applied to the US FDA Adverse Event Reporting System data to investigate the 398 pairs. The ten methods include popular methods in the pharmacovigilance literature, newly developed pharmacovigilance methods as at 2018, and popular methods in the genome-wide association study literature. We compared their performance using the receiver operating characteristic (ROC) plot, area under the curve (AUC), and Youden's index.
RESULTS: The Bayesian confidence propagation neural network had the highest AUC overall. Monte Carlo expectation maximization, a method developed in 2018, had the second highest AUC and the highest Youden's index, and performed very well in terms of high specificity. The regression-adjusted gamma Poisson shrinkage model performed best under high-sensitivity requirements.
CONCLUSION: Our results will be useful to help choose a method for a given desired level of specificity. Methods popular in the genome-wide association study literature did not perform well because of the sparsity of data and will need modification before their properties can be used in the drug-adverse event association problem.

PMID: 30762164 [PubMed - indexed for MEDLINE]

A Machine-Learning Algorithm to Optimise Automated Adverse Drug Reaction Detection from Clinical Coding.

Drug Saf. 2019 06;42(6):721-725

Authors: McMaster C, Liew D, Keith C, Aminian P, Frauman A

Abstract
INTRODUCTION: Adverse drug reaction (ADR) detection in hospitals is heavily reliant on spontaneous reporting by clinical staff, with studies in the literature pointing to high rates of underreporting [1]. International Classification of Diseases, 10th Revision (ICD-10) codes have been used in epidemiological studies of ADRs and offer the potential for automated ADR detection systems.
OBJECTIVE: The aim of this study was to develop an automated ADR detection system based on ICD-10 codes, using machine-learning algorithms to improve accuracy and efficiency.
METHODS: For a 12-month period from December 2016 to November 2017, every inpatient episode receiving an ICD-10 code in the range Y40.0-Y59.9 (ADR code) was flagged for review as a potential ADR. Each flagged admission was assessed by an expert pharmacist and, if needed, reviewed at regular ADR committee meetings. For each report, a determination was made about ADR probability and severity. The dataset was randomly split into training and test sets. A machine-learning model using the random forest algorithm was developed on the training set to discriminate between true and false ADR reports. The model was then applied to the test set to assess accuracy using the area under the receiver operating characteristic (AUC).
RESULTS: In the study period, 2917 Y40.0-Y59.9 codes were applied to admissions, resulting in 245 ADR reports after review. These 245 reports accounted for 44.5% of all ADR reporting in our hospital in the study period. A random forest model built on the training set was able to discriminate between true and false reports on the test set with an AUC of 0.803.
CONCLUSIONS: Automated ADR detection using ICD-10 coding significantly improved ADR detection in the study period, with improved discrimination between true and false reports by applying a machine-learning model.

PMID: 30725336 [PubMed - indexed for MEDLINE]