The medical dictionary for regulatory activities (MedDRA)

Identification of Pregnancy Adverse Drug Reactions in Pharmacovigilance Reporting Systems: A Novel Algorithm Developed in EudraVigilance

INTRODUCTION

There is a need to strengthen the evidence base regarding medication use during pregnancy and to facilitate the early detection of safety signals. EudraVigilance (EV) serves as the primary system for managing and analysing information concerning suspected adverse drug reactions (ADRs) within the European Economic Area. Despite its various functionalities, the current format for electronic submissions of safety reports lacks a specific data element indicating medicine exposure during pregnancy.

OBJECTIVE

This paper aims to address the limitations of existing approaches by developing a rule-based algorithm in EV that more reliably identifies cases that are truly representative of an ADR during pregnancy.

METHODS

The study utilised the standardised MedDRA query (SMQ) 'Pregnancy and neonatal topics' (PNT) as a benchmark for comparison. Recognising that the SMQ PNT also retrieves healthy pregnancy outcomes, contraceptive failure, failed abortifacients as well as ADRs not associated with pregnancy, a novel algorithm was tailored to improve the accuracy of identifying suspected ADRs occurring during pregnancy.

RESULTS

Upon testing, the algorithm demonstrated superior performance, correctly predicting 90% of cases reporting an ADR during pregnancy, compared to 54% achieved by the SMQ PNT. The implementation of the algorithm in EV led to the retrieval of 202,426 cases.

CONCLUSION

The development and successful testing of the novel algorithm represents a step forward in pregnancy-specific signal detection in EV. Because signals associated with pregnancy may be diluted in a large database such as EV, this study lays the groundwork for future research to evaluate the effectiveness of disproportionality methods on a more refined subset of pregnancy-related ADR reports.

The clinical and cost-effectiveness of elective primary total knee replacement with PAtellar Resurfacing compared to selective patellar resurfacing: a pragmatic multicentre randomized controlled Trial (PART)

Aims

During total knee replacement (TKR), surgeons can choose whether or not to resurface the patella, with advantages and disadvantages of each approach. Recently, the National Institute for Health and Care Excellence (NICE) recommended always resurfacing the patella, rather than never doing so. NICE found insufficient evidence on selective resurfacing (surgeon's decision based on intraoperative findings and symptoms) to make recommendations. If effective, selective resurfacing could result in optimal individualized patient care. This protocol describes a randomized controlled trial to evaluate the clinical and cost-effectiveness of primary TKR with always patellar resurfacing compared to selective patellar resurfacing.

Methods

The PAtellar Resurfacing Trial (PART) is a patient- and assessor-blinded multicentre, pragmatic parallel two-arm randomized superiority trial of adults undergoing elective primary TKR for primary osteoarthritis at NHS hospitals in England, with an embedded internal pilot phase (ISRCTN 33276681). Participants will be randomly allocated intraoperatively on a 1:1 basis (stratified by centre and implant type (cruciate-retaining vs cruciate-sacrificing)) to always resurface or selectively resurface the patella, once the surgeon has confirmed sufficient patellar thickness for resurfacing and that constrained implants are not required. The primary analysis will compare the Oxford Knee Score (OKS) one year after surgery. Secondary outcomes include patient-reported outcome measures at three months, six months, and one year (Knee injury and Osteoarthritis Outcome Score, OKS, EuroQol five-dimension five-level questionnaire, patient satisfaction, postoperative complications, need for further surgery, resource use, and costs). Cost-effectiveness will be measured for the lifetime of the patient. Overall, 530 patients will be recruited to obtain 90% power to detect a four-point difference in OKS between the groups one year after surgery, assuming up to 40% resurfacing in the selective group.

Conclusion

The trial findings will provide evidence about the clinical and cost-effectiveness of always patellar resurfacing compared to selective patellar resurfacing. This will inform future NICE guidelines on primary TKR and the role of selective patellar resurfacing.

Clinical Effectiveness of Newborn Screening for Spinal Muscular Atrophy: A Nonrandomized Controlled Trial

Importance

There is increasing evidence that early diagnosis and treatment are key for outcomes in infants with spinal muscular atrophy (SMA), and newborn screening programs have been implemented to detect the disease before onset of symptoms. However, data from controlled studies that reliably confirm the benefits of newborn screening are lacking.

Objective

To compare data obtained on patients with SMA diagnosed through newborn screening and those diagnosed after clinical symptom onset.

Design, Setting, and Participants

This nonrandomized controlled trial used data from the SMARTCARE registry to evaluate all children born between January 2018 and September 2021 with genetically confirmed SMA and up to 3 SMN2 copies. The registry includes data from 70 participating centers in Germany, Austria, and Switzerland. Data analysis was performed in February 2023 so that all patients had a minimal follow-up of 18 months.

Exposure

Patients born in 2 federal states in Germany underwent screening in a newborn screening pilot project. All other patients were diagnosed after clinical symptom onset. All patients received standard care within the same health care system.

Main Outcomes

The primary end point was the achievement of motor milestones.

Results

A total of 234 children (123 [52.6%] female) were identified who met inclusion criteria and were included in the analysis: 44 (18.8%) in the newborn screening cohort and 190 children (81.2%) in the clinical symptom onset cohort. The mean (SD) age at start of treatment with 1 of the approved disease-modifying drugs was 1.3 (2.2) months in the newborn screening cohort and 10.7 (9.1) months in the clinical symptom onset cohort. In the newborn screening cohort, 40 of 44 children (90.9%) gained the ability to sit independently vs 141 of 190 (74.2%) in the clinical symptom onset cohort. For independent ambulation, the ratio was 28 of 40 (63.6%) vs 28 of 190 (14.7%).

Conclusions and Relevance

This nonrandomized controlled trial demonstrated effectiveness of newborn screening for infants with SMA in the real-world setting. Functional outcomes and thus the response to treatment were significantly better in the newborn screening cohort compared to the unscreened clinical symptom onset group.

Trial Registration

German Clinical Trials Register: DRKS00012699.

The REporting of A Disproportionality Analysis for DrUg Safety Signal Detection Using Individual Case Safety Reports in PharmacoVigilance (READUS-PV): Explanation and Elaboration

In pharmacovigilance, disproportionality analyses based on individual case safety reports are widely used to detect safety signals. Unfortunately, publishing disproportionality analyses lacks specific guidelines, often leading to incomplete and ambiguous reporting, and carries the risk of incorrect conclusions when data are not placed in the correct context. The REporting of A Disproportionality analysis for drUg Safety signal detection using individual case safety reports in PharmacoVigilance (READUS-PV) statement was developed to address this issue by promoting transparent and comprehensive reporting of disproportionality studies. While the statement paper explains in greater detail the procedure followed to develop these guidelines, with this explanation paper we present the 14 items retained for READUS-PV guidelines, together with an in-depth explanation of their rationale and bullet points to illustrate their practical implementation. Our primary objective is to foster the adoption of the READUS-PV guidelines among authors, editors, peer reviewers, and readers of disproportionality analyses. Enhancing transparency, completeness, and accuracy of reporting, as well as proper interpretation of their results, READUS-PV guidelines will ultimately facilitate evidence-based decision making in pharmacovigilance.

Association between voriconazole-induced visual hallucination and dopamine in an analysis of the food and drug administration (FDA) adverse event reporting system database

Voriconazole is a second-generation azole used to treat serious fungal infections. Visual hallucinations constitute a representative adverse event caused by voriconazole. However, its mechanism of action remains unclear. In patients with schizophrenia or Parkinson's disease, the frequency of visual hallucinations is associated with brain dopamine levels. This study investigated the frequency of visual hallucinations in patients treated with voriconazole alone or in combination with dopaminergic medicines or dopamine antagonists, using data collected from the Food and Drug Administration Adverse event Reporting System (FAERS). The frequency of visual hallucinations with voriconazole alone and in combination with a dopaminergic medicine (levodopa) or dopamine antagonists (risperidone and chlorpromazine) was compared using data from the FAERS between 2004 and 2023, using the reporting odds ratio (ROR) with relevant 95% confidence intervals (CI). The reference group comprised patients who had been administered voriconazole without dopaminergic medication or dopamine antagonists. Of the patients, 22,839, 90,810, 109,757, 6,435, 20, 83, and 26, respectively were treated with voriconazole, levodopa, risperidone, chlorpromazine, voriconazole plus levodopa, voriconazole plus risperidone, and voriconazole plus chlorpromazine. The occurrence of visual hallucinations increased when used in combination with levodopa (ROR = 12.302, 95% CI = 3.587-42.183). No increase in incidence was associated with the concomitant use of dopamine antagonists (risperidone, ROR = 1.721, 95% CI = 0.421-7.030; chlorpromazine, ROR = none, 95% CI = none). Dopaminergic medicine may increase the risk of visual hallucinations in patients treated with voriconazole. Whether voriconazole positively modulates dopamine production warrants further investigation using a translational research approach.

An analysis of suicidal and self-injurious behavior reports with antiseizure medications in the FDA adverse event database

BACKGROUND

Pharmacovigilance systems such as the FDA Adverse Event Reporting System (FAERS), are established models for adverse event surveillance that may have been missed during clinical trials. We aimed to analyze twenty-five anti-seizure medications (ASMs) in FAERS to assess for increased reporting of suicidal and self-injurious behavior.

METHODS

Twenty-five ASMs were analyzed: brivaracetam, cannabidiol, carbamazepine, clobazam, clonazepam, diazepam, eslicarbazepine, felbamate, gabapentin, lacosamide, lamotrigine, levetiracetam, oxcarbazepine, perampanel, phenobarbital, phenytoin, pregabalin, primidone, rufinamide, stiripentol, tiagabine, topiramate, valproate, vigabatrin, zonisamide. Reports of "suicidal and self-injurious behavior" were collected from January 1, 2004, to December 31, 2020, using OpenVigil 2.1 tool with indication as "Epilepsy". Relative reporting ratio, proportional reporting ratio, and reporting odds ratio were calculated utilizing all other drug reports for epilepsy patients as a control.

RESULTS

Significant relative operating ratio, ROR (greater than 1, p<0.05) were observed for diazepam (2.909), pregabalin (2.739), brivaracetam (2.462), gabapentin (2.185), clonazepam (1.649), zonisamide (1.462), lacosamide (1.333), and levetiracetam (1.286).

CONCLUSIONS

Of the 25 ASMs that were analyzed in this study, 4 (16%) were identified to have been linked with a likely true adverse event. These drugs included diazepam, brivaracetam, gabapenetin, and pregabalin. Although several limitations are present with the FAERS database, it is imperative to closely monitor patient comorbidities for increased risk of suicidality with the use of several ASMs.

Post-marketing safety concerns with elagolix: a disproportionality analysis of the FDA adverse event reporting system

OBJECTIVE

Elagolix is approved for the treatment of moderate-to-severe pain associated with endometriosis. However, the long-term safety of elagolix in a large sample of real-world patients is unknown.

METHODS

The U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) reports were collected and analyzed from January 2019 to June 2023. Disproportionality analyses, including the reporting odds ratio (ROR), the proportional reporting ratio (PRR), the Bayesian confidence propagation neural network (BCPNN), and the multi-item gamma Poisson shrinker (MGPS) algorithms, were employed in data mining to quantify the signals of elagolix-related adverse events (AEs).

RESULTS

After removing the non-drug-related AE signals, we detected several AE signals such as hot flushes, bone pain, suicidal ideation, depression, and increased liver enzymes, which were known during the clinical trial phase. In addition to this, we detected several unexpected important AEs that were not mentioned in the drug insert, including cystitis interstitial, parosmia, and epiploic appendagitis. The median onset time of elagolix-associated AEs was 28.5 days.

CONCLUSION

Our study provides a comprehensive picture of the safety of elagolix in the post-marketing setting, while also identifying potential new AE signals. These findings emphasize the importance of continued monitoring of the potential risks of elagolix.

Patient perspectives in adverse event reporting after vaginal apical prolapse surgery

BACKGROUND

Many clinical trials use systematic methodology to monitor adverse events and determine grade (severity), expectedness, and relatedness to treatments as determined by clinicians. However, patient perspectives are often not included in this process.

OBJECTIVE

This study aimed to compare clinician vs patient grading of adverse event severity in a urogynecologic surgical trial. Secondary objectives were to estimate the association of patient grading of adverse events with decision-making and quality of life outcomes and to determine if patient perspective changes over time.

STUDY DESIGN

This was a planned supplementary study (Patient Perspectives in Adverse Event Reporting [PPAR]) to a randomized trial comparing 3 surgical approaches to vaginal apical prolapse. In the parent trial, adverse events experienced by patients were collected per a standardized protocol every 6 months during which clinicians graded adverse event severity (mild, moderate, severe/life-threatening). In this substudy, we obtained additional longitudinal patient perspectives for 19 predetermined "PPAR adverse events." Patients provided their own severity grading (mild, moderate, severe/very severe/life-threatening) at initial assessment and at 12 and 36 months postoperatively. Clinicians and patients were masked to each other's reporting. The primary outcome was the interrater agreement (kappa statistic) for adverse event severity between the initial clinician and patient assessment, combining patient grades of mild and moderate. The association between adverse event severity and the Decision Regret Scale, Satisfaction with Decision Scale, the 12-Item Short-Form Health Survey, and Patient Global Impression of Improvement scores was assessed using the Spearman correlation coefficient (ρ) for continuous scales, the Mantel-Haenszel chi-square test for Patient Global Impression of Improvement, and t tests or chi-square tests comparing the assessments of patients who rated their adverse events or symptoms as severe with those who gave other ratings. To describe patient perspective changes over time, the intraobserver agreement was estimated for adverse event severity grade over time using weighted kappa coefficients.

RESULTS

Of the 360 randomly assigned patients, 219 (61%) experienced a total of 527 PPAR adverse events (91% moderate and 9% severe/life-threatening by clinician grading). Mean patient age was 67 years; 87% were White and 12% Hispanic. Among the patients reporting any PPAR event, the most common were urinary tract infection (61%), de novo urgency urinary incontinence (35%), stress urinary incontinence (22%), and fecal incontinence (13%). Overall agreement between clinician and participant grading of severity was poor (kappa=0.24 [95% confidence interval, 0.14-0.34]). Of the 414 adverse events that clinicians graded as moderate, patients graded 120 (29%) as mild and 80 (19%) as severe. Of the 39 adverse events graded as severe by clinicians, patients graded 15 (38%) as mild or moderate. Initial patient grading of the most severe reported adverse event was mildly correlated with worse Decision Regret Scale (ρ=0.2; P=.01), 12-Item Short-Form Health Survey (ρ=-0.24; P<.01), and Patient Global Impression of Improvement (P<.01) scores. There was no association between adverse event severity and Satisfaction with Decision Scale score. Patients with an initial grading of "severe" had more regret, lower quality of life, and poorer global impressions of health than those whose worst severity grade was mild (P<.05). Agreement between the patients' initial severity ratings and their ratings at 12 months (kappa=0.48 [95% confidence interval, 0.39-0.58]) and 36 months (kappa=0.45 [95% confidence interval, 0.37-0.53]) was fair.

CONCLUSION

Clinician and patient perceptions of adverse event severity are discordant. Worse severity from the patient perspective was associated with patient-centered outcomes. Including the patient perspective provides additional information for evaluating surgical procedures.

Comparators in Pharmacovigilance: A Quasi-Quantification Bias Analysis

BACKGROUND AND OBJECTIVE

It is unclear which comparator is the most appropriate for bias reduction in disproportionality analyses based on spontaneous reports. We conducted a quasi-quantitative bias analysis using two well-studied drug-event combinations to assess how different comparators influence the directionality of bias in pharmacovigilance.

METHODS

We used the US Food and Drug Administration Adverse Event Reporting System focusing on two drug-event combinations with a propensity for stimulated reporting: rivaroxaban and hepatotoxicity, and canagliflozin and acute kidney injury. We assessed the directionality of three disproportionality analysis estimates (reporting odds ratio, proportional reporting ratio, information component) using one unrestricted comparator (full data) and two restricted comparators (active comparator, active comparator with class exclusion). Analyses were conducted within two calendar time periods, defined based on external events (approval of direct oral anticoagulants, Food and Drug Administration safety warning on acute kidney injury with sodium-glucose cotransporter 2 inhibitors) hypothesized to alter reporting rates.

RESULTS

There were no false-positive signals for rivaroxaban and hepatotoxicity irrespective of the comparator. Restricting to the initial post-approval period led to false-positive signals, with restricted comparators performing worse. There were false-positive signals for canagliflozin and acute kidney injury, with restricted comparators performing better. Restricting to the period before the Food and Drug Administration warning weakened the false-positive signal for canagliflozin and acute kidney injury across comparators.

CONCLUSIONS

We could not identify a consistent and predictable pattern to the directionality of disproportionality analysis estimates with specific comparators. Calendar time-based restrictions anchored on relevant external events had a considerable impact.

Gastrointestinal infections and gastrointestinal haemorrhage are underestimated but serious adverse events in chimeric antigen receptor T-cell recipients: A real-world study

Chimeric antigen receptor T-cell (CAR-T) therapy has achieved durable response in patients with hematological malignancies, however, therapy-associated multisystem toxicities are commonly observed. Here, we systematically analyzed CAR-T-related gastrointestinal adverse events (GAEs) using the U.S. Food and Drug Administration Adverse Event Reporting System (FAERS) between January 2017 and December 2021. Disproportionality analyses were performed using reporting odds ratios (ROR) and information component (IC). Among 105,087,611 reports in FAERS, 1518 CAR-T-related GAEs reports were identified. 23 GAEs (n = 281, 18.51%) were significantly overreported following CAR-T therapy compared with the full database, of which 11 GAEs (n = 156, 10.28%) were associated with gastrointestinal infections (GI), such as clostridium difficile colitis (n = 44 [2.90%], ROR = 5.55), enterovirus infection (n = 23 [1.52%], ROR = 20.02), and mucormycosis (n = 15 [0.99%], ROR = 3.09). Overall, the fatality rate of 11 GI-related AEs was 29.49%, especially mucormycosis causing substantial mortality with 60%. In addition, 4 of 23 overreported GAEs were related to haemorrhage and the mortality of gastrointestinal haemorrhage was 73.17%. Lastly, 29 death-related GAEs were identified. These findings could help clinicians early alert those rarely reported but lethal GAEs, thus reducing the risk of severe toxicities.

Hospital admissions among patients with Comorbid Substance Use disorders: a secondary analysis of predictors from the NavSTAR Trial

BACKGROUND

Individuals with substance use disorders (SUDs) frequently use acute hospital services. The Navigation Services to Avoid Rehospitalization (NavSTAR) trial found that a patient navigation intervention for hospitalized patients with comorbid SUDs reduced subsequent inpatient admissions compared to treatment-as-usual (TAU).

METHODS

This secondary analysis extends previous findings from the NavSTAR trial by examining whether selected patient characteristics independently predicted hospital service utilization and moderated the effect of the NavSTAR intervention. Participants were 400 medical/surgical hospital patients with comorbid SUDs. We analyzed 30- and 90-day inpatient readmissions (one or more readmissions) and cumulative incidence of inpatient admissions through 12 months using multivariable logistic and negative binomial regression, respectively.

RESULTS

Consistent with primary findings and controlling for patient factors, NavSTAR participants were less likely than TAU participants to be readmitted within 30 (P = 0.001) and 90 (P = 0.03) days and had fewer total readmissions over 12 months (P = 0.008). Hospitalization in the previous year (P < 0.001) was associated with cumulative readmissions over 12 months, whereas Medicaid insurance (P = 0.03) and index diagnoses of infection (P = 0.001) and injuries, poisonings, or procedural complications (P = 0.004) were associated with fewer readmissions. None of the selected covariates moderated the effect of the NavSTAR intervention.

CONCLUSIONS

Previous findings showed that patient navigation could reduce repeat hospital admissions among patients with comorbid SUDs. Several patient factors were independently associated with readmission. Future research should investigate risk factors for hospital readmission among patients with comorbid SUDs to optimize interventions.

TRIAL REGISTRATION

NIH ClinicalTrials.gov NCT02599818, Registered November 9, 2015 https://classic.

CLINICALTRIALS

gov/ct2/show/NCT02599818 .

Comparison of Statistical Signal Detection Methods in Adverse Events Following Immunization - China, 2011-2015

Introduction

The current study aims to assess the performance of data mining techniques in detecting safety signals for adverse events following immunization (AEFI) using routinely obtained data in China. Four different methods for detecting vaccine safety signals were evaluated.

Methods

The AEFI data from 2011 to 2015 was collected for our study. We analyzed the data using four different methods to detect signals: the proportional reporting ratio (PRR), reporting odds ratio (ROR), Bayesian confidence propagation neural network (BCPNN), and multi-item gamma Poisson shrinker (MGPS). Each method was evaluated at 1-3 thresholds for positivity. To assess the performance of these methods, we used the published signal rates as gold standards to determine the sensitivity and specificity.

Results

The number of identified signals varied from 602 for PRR1 (with a threshold of 1) to 127 for MGPS1. When considering the common reactions as the reference standard, the sensitivity ranged from 0.9% for MGPS1/2 to 38.2% for PRR1/2, and the specificity ranged from 85.2% for PRR1 and ROR1 to 96.7% for MGPS1. When considering the rare reactions as the reference standard, PRR1, PRR2, ROR1, ROR2, and BCPNN exhibited the highest sensitivity (73.3%), while MGPS1 exhibited the highest specificity (96.9%).

Discussion

For common reactions, the sensitivities were modest and the specificities were high. For rare reactions, both the sensitivities and specificities were high. Our study provides valuable insights into the selection of signal detection methods and thresholds for AEFI data in China.

Janus kinase inhibitors and tumour necrosis factor inhibitors show a favourable safety profile and similar persistence in rheumatoid arthritis, psoriatic arthritis and spondyloarthritis: real-world data from the BIOBADASER registry

OBJECTIVES

To compare the safety of Janus kinase inhibitors (JAKi) with that of tumour necrosis factor inhibitors (TNFi) and determine drug persistence among patients with rheumatoid arthritis (RA) and spondyloarthritis (SpA).

METHODS

We analysed data from patients included in BIOBADASER 3.0 and treated with JAKi or TNFi from 2015 to 2023 and estimated the incidence rate ratio (IRR) of adverse events and persistence.

RESULTS

A total of 6826 patients were included. Of these, 52% had RA, 25% psoriatic arthritis and 23% axial SpA. Treatment was with TNFi in 86%. The mean duration of treatment was 2.2±2.0 years with TNFi versus 1.8±1.5 with JAKi. JAKis were prescribed in older patients with longer term disease, greater comorbidity and later treatment lines and more frequently as monotherapy. The IRR of all infections and gastrointestinal events was higher among patients with RA treated with JAKi. Drug persistence at 1, 2 and 3 years was 69%, 55% and 45% for TNFi and 68%, 54% and 45% for JAKi. Multivariate regression models showed a lower probability of discontinuation for JAKi (HR=0.85; 95% CI 0.78-0.92) and concomitant conventional synthetic disease-modifying antirheumatic drugs (HR=0.90; 95% CI 0.84-0.96). The risk of discontinuation increased with glucocorticoids, comorbidities, greater disease activity and later treatment lines.

CONCLUSIONS

Infections, herpes zoster and gastrointestinal adverse events in patients with RA tended to be more frequent with JAKi. However, prognosis was poor in patients receiving JAKi. Persistence was similar for TNFi and JAKi, although factors associated with discontinuation differed by diagnostic group.

Breastfeeding-Related Adverse Drug Reactions of Triptans: A Descriptive Analysis Using Four Pharmacovigilance Databases

Objective: Migraine attacks are common in women of reproductive age. Although attacks are often less severe and less frequent during pregnancy, they regularly reoccur shortly after delivery. When first-line analgesic treatment is insufficient, triptans may be used for acute treatment of migraine attacks. Milk levels of occasional triptan use have shown to be low, and no adverse effects in breastfed infants have been reported. However, the available knowledge on the safety of triptans during breastfeeding is still limited. Methods: Four (inter)national pharmacovigilance databases were searched for breastfeeding related adverse drug reactions of triptans. These included the Dutch Pregnancy Drug Register and three databases of spontaneous reports (Netherlands Pharmacovigilance Centre Lareb, the European Medicines Agency [EudraVigilance], and the World Health Organization [VigiBase]). Results: A total of 26 reports on 27 breastfeeding related adverse drug reactions were identified (one report involved two separate adverse drug reactions). These involve three main complaints: painful breasts and/or nipples, painful milk ejection reflex, and a decrease in milk production. Discussion and Conclusion: The hypothesized pharmacological mechanism relates to the serotonin-receptor agonistic properties of triptans. These may lead to vasoconstriction in the breasts and nipples, including the vasculature surrounding the milk ducts and alveoli, and may also influence the hormonal function and levels of prolactin. The reported adverse drug reactions do not negatively impact the overall compatibility of triptans with breastfeeding. However, breastfeeding women may experience them as unsettling. Awareness of these potential adverse drug reactions is essential and should be weighed against the potential adverse effects of (untreated) symptoms of migraine attacks.

Sex differences in adverse events from systemic treatments for psoriasis: A decade of insights from the Swiss Psoriasis Registry (SDNTT)

BACKGROUND

Psoriasis is a disease that often requires prolonged systemic treatment. It is important to determine the safety of available therapies. There is currently little insight into sex-specific differences in the safety of systemic psoriasis therapies.

OBJECTIVES

To examine the real-world, long-term safety of systemic psoriasis therapies with sex stratification in drug-related adverse events (ADRs).

METHODS

Ten-year data from adults with moderate-to-severe psoriasis requiring systemic treatment (conventional systemic therapies [CST], biologics) were obtained from the Swiss psoriasis registry (SDNTT). ADRs were categorized according to the international terminology Medical Dictionary for Regulatory Activities (MedDRA). Safety was assessed by calculating event rates per 100 patient-years (PY). We used descriptive statistics for patient and disease characteristics, and binomial and t-tests to compare treatment groups and sex.

RESULTS

In total, 791 patients (290 females) were included with a mean age of 46 years. 358 (45%) received CSTs and 433 (55%) biologics; both groups had similar baseline characteristics except for more joint involvement in patients using biologics (26.86% vs. 14.8%, p < 0.0001). CSTs were associated with a 2.2-fold higher ADR rate (40.43/100 PY vs. 18.22/100 PY, p < 0.0001) and an 8.0-fold higher drug-related discontinuation rate than biologics (0.16/PY vs. 0.02/PY, p < 0.0001). Trends showed non-significant higher serious adverse event rates per 100 PY for biologics (8.19, CI 6.87-9.68) compared to CSTs (7.08, CI 5.39-9.13) (p = 0.3922). Sex stratification revealed a significantly higher overall ADR rate for all treatments in females (1.8-fold for CSTs [57.30/100 PY vs. 31.69/100 PY] and 2.0-fold for biologics [27.36/100 PY vs. 13.9/100 PY], p < 0.0001), and drug-related discontinuation rates for most CSTs in females.

CONCLUSION

Females were associated with a significantly higher rate of ADRs and drug-related discontinuation rates. Sex stratification should be taken into consideration when designing studies in the patient-tailored management of psoriasis.

A phase I trial of caffeine to evaluate safety in infants with hypoxic-ischemic encephalopathy

OBJECTIVE

Caffeine provides neuroprotection following hypoxic-ischemic injury in animals. We characterized the safety of escalating doses of caffeine in infants with hypoxic-ischemic encephalopathy (HIE) receiving therapeutic hypothermia.

STUDY DESIGN

Phase I trial of infants undergoing therapeutic hypothermia for HIE receiving IV caffeine 20 mg/kg followed by up to two daily doses of 5 mg/kg (n = 9) or 10 mg/kg (n = 8). Safety was evaluated based on adverse events and frequency of pre-specified outcomes compared to data from the Whole-Body Hypothermia for HIE trial (Shankaran, 2005).

RESULTS

Twelve of 17 (71%) infants had ≥1 adverse event during the study period. The frequency of clinical outcomes related to HIE were not statistically different from outcomes in infants receiving hypothermia in the Whole-Body Hypothermia for HIE trial.

CONCLUSION

Caffeine administration was well tolerated. A larger study is required to determine the optimal dose and evaluate drug safety and efficacy.

CLINICAL TRIAL

ClinicalTrials.gov Identifier: NCT03913221.

IBDTransDB: a manually curated transcriptomic database for inflammatory bowel disease

Inflammatory Bowel Disease (IBD) therapies are ineffective in at least 40% patients, and transcriptomic datasets have been widely used to reveal the pathogenesis and to identify the novel drug targets for these patients. Although public IBD transcriptomic datasets are available from many web-based tools/databases, due to the unstructured metadata and data description of these public datasets, most of these tools/databases do not allow querying datasets based on multiple keywords (e.g. colon and infliximab). Furthermore, few tools/databases can compare and integrate the datasets from the query results. To fill these gaps, we have developed IBDTransDB (https://abbviegrc.shinyapps.io/ibdtransdb/), a manually curated transcriptomic database for IBD. IBDTransDB includes a manually curated database with 34 transcriptomic datasets (2932 samples, 122 differential comparisons) and a query system supporting 35 keywords from 5 attributes (e.g. tissue and treatment). IBDTransDB also provides three modules for data analyses and integration. IBDExplore allows interactive visualization of differential gene list, pathway enrichment, gene signature and cell deconvolution analyses from a single dataset. IBDCompare supports comparisons of selected genes or pathways from multiple datasets across different conditions. IBDIntegrate performs meta-analysis to prioritize a list of genes/pathways based on user-selected datasets and conditions. Using two case studies related to infliximab treatment, we demonstrated that IBDTransDB provides a unique platform for biologists and clinicians to reveal IBD pathogenesis and identify the novel targets by integrating with other omics data. Database URL: https://abbviegrc.shinyapps.io/ibdtransdb/.

Predicting drug outcome of population via clinical knowledge graph

Optimal treatments depend on numerous factors such as drug chemical properties, disease biology, and patient characteristics to which the treatment is applied. To realize the promise of AI in healthcare, there is a need for designing systems that can capture patient heterogeneity and relevant biomedical knowledge. Here we present PlaNet, a geometric deep learning framework that reasons over population variability, disease biology, and drug chemistry by representing knowledge in the form of a massive clinical knowledge graph that can be enhanced by language models. Our framework is applicable to any sub-population, any drug as well drug combinations, any disease, and to a wide range of pharmacological tasks. We apply the PlaNet framework to reason about outcomes of clinical trials: PlaNet predicts drug efficacy and adverse events, even for experimental drugs and their combinations that have never been seen by the model. Furthermore, PlaNet can estimate the effect of changing population on the trial outcome with direct implications on patient stratification in clinical trials. PlaNet takes fundamental steps towards AI-guided clinical trials design, offering valuable guidance for realizing the vision of precision medicine using AI.

Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES): a method for populating knowledge bases using zero-shot learning

MOTIVATION

Creating knowledge bases and ontologies is a time consuming task that relies on manual curation. AI/NLP approaches can assist expert curators in populating these knowledge bases, but current approaches rely on extensive training data, and are not able to populate arbitrarily complex nested knowledge schemas.

RESULTS

Here we present Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES), a Knowledge Extraction approach that relies on the ability of Large Language Models (LLMs) to perform zero-shot learning and general-purpose query answering from flexible prompts and return information conforming to a specified schema. Given a detailed, user-defined knowledge schema and an input text, SPIRES recursively performs prompt interrogation against an LLM to obtain a set of responses matching the provided schema. SPIRES uses existing ontologies and vocabularies to provide identifiers for matched elements. We present examples of applying SPIRES in different domains, including extraction of food recipes, multi-species cellular signaling pathways, disease treatments, multi-step drug mechanisms, and chemical to disease relationships. Current SPIRES accuracy is comparable to the mid-range of existing Relation Extraction methods, but greatly surpasses an LLM's native capability of grounding entities with unique identifiers. SPIRES has the advantage of easy customization, flexibility, and, crucially, the ability to perform new tasks in the absence of any new training data. This method supports a general strategy of leveraging the language interpreting capabilities of LLMs to assemble knowledge bases, assisting manual knowledge curation and acquisition while supporting validation with publicly-available databases and ontologies external to the LLM.

AVAILABILITY AND IMPLEMENTATION

SPIRES is available as part of the open source OntoGPT package: https://github.com/monarch-initiative/ontogpt.

Drug-induced hyperglycemia and diabetes

BACKGROUND

Drug-induced hyperglycemia and diabetes have negative and potentially serious health consequences but can often be unnoticed.

METHODS

We reviewed the literature searching Medline database for articles addressing drug-induced hyperglycemia and diabetes up to January 31, 2023. We also selected drugs that could induce hyperglycemia or diabetes according official data from drug information databases Thériaque and Micromedex. For each selected drug or pharmacotherapeutic class, the mechanisms of action potentially involved were investigated. For drugs considered to be at risk of hyperglycemia or diabetes, disproportionality analyses were performed using data from the international pharmacovigilance database VigiBase. In order to detect new pharmacovigilance signals, additional disproportionality analyses were carried out for drug classes with more than 100 cases reported in VigiBase, but not found in the literature or official documents.

RESULTS

The main drug classes found to cause hyperglycemia are glucocorticoids, HMG-coA reductase inhibitors, thiazide diuretics, beta-blockers, antipsychotics, fluoroquinolones, antiretrovirals, antineoplastic agents and immunosuppressants. The main mechanisms involved are alterations in insulin secretion and sensitivity, direct cytotoxic effects on pancreatic cells and increases in glucose production. Pharmacovigilance signal were found for a majority of drugs or pharmacological classes identified as being at risk of diabetes or hyperglycemia. We identified new pharmacovigilance signals with drugs not known to be at risk according to the literature or official data: phosphodiesterase type 5 inhibitors, endothelin receptor antagonists, sodium oxybate, biphosphonates including alendronic acid, digoxin, sartans, linosipril, diltiazem, verapamil, and darbepoetin alpha. Further studies will be needed to confirm these signals.

CONCLUSIONS

The risks of induced hyperglycemia vary from one drug to another, and the underlying mechanisms are multiple and potentially complex. Clinicians need to be vigilant when using at-risk drugs in order to detect and manage these adverse drug reactions. However, it is to emphasize that the benefits of appropriately prescribed treatments most often outweigh their metabolic risks.

Real-world reported adverse events related to systemic immunomodulating therapy in patients with atopic dermatitis: Results from the TREAT NL (TREatment of ATopic eczema, the Netherlands) registry

BACKGROUND

Evidence on the (long-term) safety of systemic immunomodulating therapies in atopic dermatitis (AD) generated by real-world data is sparse.

OBJECTIVES

To describe real-world reported adverse drug reactions (AEs) related to systemic immunomodulating therapy in patients with AD and to compare the incidence rates of AEs with the Summaries of Product Characteristics (SmPCs).

METHODS

We conducted an observational prospective multi-centre cohort study, using the TREAT NL registry. All severe AEs, AEs of special interest and serious AEs in adult and paediatric patients on systemic immunomodulating treatment (ciclosporin, methotrexate, azathioprine, mycophenolic acid, dupilumab, tralokinumab, baricitinib and upadacitinib) were assessed. Incidences rates of all (potentially) drug-related AEs were standardized in patient years and compared to the cumulative incidences in the associated SmPCs.

RESULTS

We collected 422 patient years of safety data from 266 patients, of whom 129 (48.5%) reported a total of 224 (potentially) drug-related AEs. Compared to dupilumab's SmPC, higher incidence rates were found for four AEs (reported ≥5 times): eosinophilia, blepharitis, dry eyes and head and neck erythema (i.e. dupilumab facial redness). A higher incidence rate of fatigue was found in patients on oral methotrexate in our cohort compared to the SmPC. Two new drug-related AEs (reported ≥5 times) were found in patients on dupilumab, including non-infectious conjunctivitis and meibomian gland dysfunction.

CONCLUSIONS

Real-world reported AEs captured in AD patient registries can add information on the estimated incidence of AEs and benefit clinical decision aids. Future studies using data derived from the TREAT NL registry combined with data from other registries within the TREAT Registry Taskforce will provide more information on (rare) AEs associated with immunomodulating therapy in AD patients.

Impact of Canagliflozin on Kidney and Cardiovascular Outcomes by Type 2 Diabetes Duration: A Pooled Analysis of the CANVAS Program and CREDENCE Trials

OBJECTIVE

The study was undertaken because it was unknown whether the duration of type 2 diabetes modifies the effects of sodium-glucose cotransporter 2 inhibitor canagliflozin on cardiovascular (CV) and kidney outcomes.

RESEARCH DESIGN AND METHODS

This post hoc analysis of the Canagliflozin Cardiovascular Assessment Study (CANVAS) Program (N = 10,142) and Evaluation of the Effects of Canagliflozin on Renal and Cardiovascular Outcomes in Participants With Diabetic Nephropathy (CREDENCE) trial (N = 4,401) evaluated hazard ratios and 95% CIs using Cox proportional hazards for the effects of canagliflozin on CV and kidney outcomes, including progression and regression of albuminuria over 5-year intervals of disease duration.

RESULTS

Canagliflozin had ranges of benefit across intervals of diabetes duration, with no heterogeneity for major adverse CV events, CV death or heart failure hospitalization, and kidney failure requiring therapy or doubling serum creatinine. Furthermore, canagliflozin reduced albuminuria progression and increased albuminuria regression with no interaction across all diabetes duration subgroups.

CONCLUSIONS

Our findings suggest that earlier treatment with canagliflozin confers consistent cardiorenal benefits to individuals with type 2 diabetes.

A Real-World Pharmacovigilance Study of Ceftazidime/Avibactam: Data Mining of the Food and Drug Administration Adverse Event Reporting System Database

Ceftazidime/avibactam (CAZ/AVI) is a combination of a well-known third-generation, broad-spectrum cephalosporin with a new beta-lactamase inhibitor that has been approved for the treatment of various infectious diseases (especially multidrug-resistant Gram-negative bacterial infections) by the Food and Drug Administration (FDA). The current study extensively assessed CAZ/AVI-related adverse events (AEs) in the real world through data mining of the FDA Adverse Event Reporting System (FAERS) database to better understand toxicities. The signals of CAZ/AVI-related AEs were quantified using disproportionality analyses, including the reporting odds ratio, the proportional reporting ratio, the Bayesian confidence propagation neural network, and the multi-item gamma Poisson shrinker algorithms. Out of 10,114,815 records retrieved from the FAERS database, 628 cases were identified, where CAZ/AVI was implicated as the primary suspect drug. A total of 61 preferred terms with significant disproportionality that simultaneously met the criteria of all four algorithms were retained. Several unexpected safety signals may also occur, including melena, hypernatremia, depressed level of consciousness, brain edema, petechiae, delirium, and shock hemorrhagic. The median onset time for AEs associated with CAZ/AVI was 4 days, with most cases occurring within 3 days after CAZ/AVI initiation.

Burden of non-serious infections during biological use for rheumatoid arthritis

INTRODUCTION

Biologicals have become a cornerstone in rheumatoid arthritis (RA) treatment. The increased risk of serious infections associated with their use is well-established. Non-serious infections, however, occur more frequently and are associated with a high socioeconomic burden and impact on quality of life but have not received the same attention in the literature to date. The aim of this study was to gain insight into the various non-serious infections reported in RA patients using biologicals and their experienced burden.

MATERIALS AND METHODS

The Dutch Biologic Monitor was a prospective observational study that included adults with rheumatoid arthritis and biological use who answered bimonthly questionnaires on the adverse drug reactions (ADRs) they experienced from their biological and reported the associated impact score (ranging from 1, no impact, to 5, very high impact). ADRs were assigned a MedDRA code by pharmacovigilance experts and labeled as definite, probable, possible or no infection by infectious disease professionals. Descriptive statistics were performed using medians and interquartile ranges.

RESULTS

A total of 586 patients were included in the final analysis. Eighty-five patients (14.5%) reported a total of 421 ADRs labeled as probable or definite infections by the experts. Patient-assigned burden was ADR-specific. Upper respiratory tract infections were most frequently reported and had a high rate of recurrence or persistence, with a median impact score of 3.0 (IQR 2.0-3.0) which remained stable over time.

DISCUSSION

Non-serious infections significantly outnumbered serious infections in this real-life cohort of RA patients using biologicals (77.1 non-serious infections and 1.3 serious infections per 100 patient years, respectively). Infections in the upper respiratory tract were rated as having an average burden, which remained constant over a long period of time. Awareness of the impact of recurrent and chronic non-serious infections may enable healthcare professionals to timely treat and maybe even prevent them, which would lessen the associated personal and socioeconomic burden.

From genome to clinic: The power of translational bioinformatics in improving human health

Translational bioinformatics (TBI) has transformed healthcare by providing personalized medicine and tailored treatment options by integrating genomic data and clinical information. In recent years, TBI has bridged the gap between genome and clinical data because of significant advances in informatics like quantum computing and utilizing state-of-the-art technologies. This chapter discusses the power of translational bioinformatics in improving human health, from uncovering disease-causing genes and variations to establishing new therapeutic techniques. We discuss key application areas of bioinformatics in clinical genomics, such as data sources and methods used in translational bioinformatics, the impact of translational bioinformatics on human health, and how machine learning and artificial intelligence are being used to mine vast amounts of data for drug development and precision medicine. We also look at the problems, constraints, and ethical concerns connected with exploiting genomic data and the future of translational bioinformatics and its potential impact on medicine and human health. Ultimately, this chapter emphasizes the great potential of translational bioinformatics to alter healthcare and enhance patient outcomes.

Design of a Remote Time-Restricted Eating and Mindfulness Intervention to Reduce Risk Factors Associated with Early-Onset Colorectal Cancer Development among Young Adults

Early-onset colorectal cancer (EOCRC) is defined as a diagnosis of colorectal cancer (CRC) in individuals younger than 50 years of age. While overall CRC rates in the United States (US) decreased between 2001 and 2018, EOCRC rates have increased. This research project aims to evaluate the feasibility and acceptability of Time-Restricted Eating (TRE), Mindfulness, or TRE combined with Mindfulness among young to middle-aged adults at risk of EOCRC. Forty-eight participants will be randomly assigned to one of four groups: TRE, Mindfulness, TRE and Mindfulness, or Control. Data on feasibility, adherence, and acceptability will be collected. Measures assessed at baseline and post-intervention will include body weight, body composition, dietary intake, physical activity, sleep behavior, circulating biomarkers, hair cortisol, and the gut microbiome. The effects of the intervention on the following will be examined: (1) acceptability and feasibility; (2) body weight, body composition, and adherence to TRE; (3) circulating metabolic, inflammation, and oxidative stress biomarkers; (4) intestinal inflammation; and (5) the gut microbiome. TRE, combined with Mindfulness, holds promise for stress reduction and weight management among individuals at risk of EOCRC. The results of this pilot study will inform the design and development of larger trials aimed at preventing risk factors associated with EOCRC.

Sunitinib for Metastatic Renal Cell Carcinoma: Real-World Data from the STAR-TOR Registry and Detailed Literature Review

INTRODUCTION

We evaluated the effectiveness and safety profile of the tyrosine kinase inhibitor sunitinib in patients with advanced or metastatic renal cell carcinoma (a/mRCC) in a real-world setting.

METHODS

We analyzed data of adult a/mRCC patients treated with sunitinib. Data were derived from the German non-interventional post-approval multicenter STAR-TOR registry (NCT00700258). Progression-free survival (PFS), overall survival (OS), and adverse events (AEs) were evaluated using descriptive statistics and survival analyses for the entire cohort and patient subgroups.

RESULTS

A total of 116 study sites recruited 702 patients treated with sunitinib (73.1% male; median age 68.0 years; median Karnofsky index 90%) between November 2010 and May 2020. The most frequent histological subtype was clear cell RCC (81.6%). Sunitinib was administered as first-line treatment in 83.5%, as second line in 11.7%, and as third line or beyond in 4.8% of the patients. Drug-related AEs and serious AEs were reported in 66.3% and 13.9% of the patients, respectively (most common AE: gastrointestinal disorders; 39.7% of all patients).

CONCLUSIONS

This study adds further real-world evidence of the persisting relevance of sunitinib for patients with a/mRCC who cannot receive or tolerate immune checkpoint inhibitors. The study population includes a high proportion of patients with unfavorable MSKCC poor-risk score, but shows still good PFS and OS results, while the drug demonstrates a favorable safety profile. The STAR-TOR registry is also registered in the database of US library of medicine (NCT00700258).

Ocular toxicities in chimeric antigen receptor T-cell therapy: a real-world study leveraging FAERS database

Aim: The purpose of this study was to comprehensively explore the ocular toxicity associated with chimeric antigen receptor (CAR) T-cell therapy. Materials & methods: Data were assembled from the US FDA's Adverse Event Reporting System (FAERS) database from 2017 to 2023. Information component and reporting odds ratio methods were used for signal detection in total/categorized CAR T-cell therapy. Results: A total of 17 positive signals (preferred term) were detected, yet none of them were documented in the product information. Some adverse events were with death outcomes and overlapped a lot with cytokine-release syndrome. Conclusion: The ocular adverse events associated with CAR-T cell therapy are noteworthy, and it is imperative to maintain increased alertness and institute early intervention strategies.

Use of Medical Cannabis in Patients with Gilles de la Tourette's Syndrome in a Real-World Setting

Objective: Tourette's syndrome (TS) is a neurodevelopmental disorder characterized by vocal and motor tics and other comorbidities. Clinical recommendations for the use of medical cannabis are established, yet further guidance is needed. The aim of this study was to describe the experience of patients with TS with medical cannabis. Materials and Methods: TS patients were recruited from a registry of patients ("Tikun Olam" company). Questionnaires were answered before and after 6 months of treatment. Patients were divided into two groups: (A) patients who responded and (B) patients who did not respond to the follow-up questionnaire. In group A, an analysis was made to evaluate the presence and frequency of motor and vocal tics. The patients' general mood, employment status, quality of life, and comorbidities were also included in the analysis. Results: Seventy patients were identified. The tetrahydrocannabinol and cannabidiol mean daily dose was 123 and 50.5 mg, respectively. In group A, a statistically significant improvement was identified in quality of life (p<0.005), employment status (p=0.027), and in the reduction of the number of medications (p<0.005). Sixty-seven percent and 89% of patients with obsessive-compulsive disorder and anxiety comorbidities, respectively, reported an improvement. No statistically significant improvement was identified in motor tics (p=0.375), vocal tics (p>0.999), tics frequency (p=0.062), or general mood (p=0.129). The most frequent adverse effects were dizziness (n=4) and increased appetite (n=3). Conclusion: Subjective reports from TS patients suggest that medical cannabis may improve their quality of life and comorbidities. More studies are needed to evaluate the efficacy and safety of medical cannabis. Registry in the MOH: https://www.moh.gov.sg/ (Trial number: 0185-19-ASF).

A Descriptive Analysis of Adverse Event Reports from the Quebec Cannabis Registry

INTRODUCTION

Published data on the safety of natural medical cannabis (MC) when used in the real-world clinical practice setting are lacking. This study aimed to describe adverse events (AEs) reported across three years following MC initiation.

METHODS

The Quebec Cannabis Registry (QCR) was a prospective registry of adults enrolled through participating physicians when they initiated MC between May 2015 and October 2018. Follow-up ended at MC discontinuation, loss to follow-up, three years, or end of data collection (May 2019). Data were collected at baseline and at follow-up visits every three months for the first two years, then once in the third year. Physicians filled adverse event (AE) reports, which were coded using MedDRA® preferred terms (PTs), and descriptive analyses were conducted.

RESULTS

A total of 2991 patients were enrolled (mean age 50.9 years, 50.2% females). During follow-up, 108 patients (3.6%) experienced moderate or severe AEs, yielding 111 AE reports (three patients had two reports) and 214 AEs (average 1.9 AEs per report). Mild AEs were recorded as a reason for MC discontinuation for nine patients, but no AE reports were available. The most common PTs for ingested MC (62 reports) were dizziness (12.9%), nausea (11.3%), somnolence (9.7%), and vomiting (8.1%), and for inhaled MC (23 reports), headache (13.0%) was the most common. The most frequent PTs associated with tetrahydrocannabinol (THC)-dominant MC (25 reports) were dizziness and somnolence (12.0% each); for cannabidiol (CBD)-dominant MC (20 reports), vomiting (20.0%) was most common; and dizziness (17.2%), nausea (13.8%), somnolence (10.3%), and headache (8.6%) were the most frequent for balanced MC (58 reports).

CONCLUSION

No new safety concerns were identified relative to the published literature, although notable differences in AE profile between modes of administration and cannabinoid content ratios should be considered by health professionals. Further work identifying and managing risk factors for AEs is warranted to maintain a favorable benefit-risk balance for MC.

Utilizing temporal pattern of adverse event reports to identify potential late-onset adverse events

OBJECTIVES

Through the use of FDA adverse event reporting system (FAERS) dataset, this study analyzes the pattern of time-to-event (TTE) for drugs and adverse events, and suggest ways to identify candidate late-onset events for monitoring.

METHODS

The duration between administration date of the drug and the onset of adverse events was explored with using FAERS data from 2012-2021. The fold change of proportional reporting ratios or reporting odds ratios were calculated to identify enriched events in the later period and to suggest the late-onset events for further monitoring. To compare the findings, we used the claims database of the Korean National Health Insurance Service (NHIS).

RESULTS

A total of 1,426,781 reports were included. The median TTE was 10 days (interquartile range [IQR]: 0-98 days), with 11.5% (n = 164,093) reporting events that occurred at least one year after administration. TTE and fold change analysis captured historical cases of late-onset events, while generating an additional less-explored list of events. The results for tumor necrosis factor (TNF) inhibitors were compared using the NHIS dataset.

CONCLUSION

Our study provides a comprehensive analysis of the FAERS dataset, focusing on TTE data. Periodic summarization of reports would be helpful in monitoring the late-onset events.

Real-world reporting rates of administration-site reactions with on-demand treatment of hereditary angioedema attacks

Background: Hereditary angioedema (HAE) is characterized by recurrent and unpredictable episodes of subcutaneous and/or submucosal swelling. Objective: To characterize the real-world treatment burden associated with existing on-demand therapies, we analyzed administration-site adverse drug reactions (ADR) associated with approved on-demand HAE therapies reported in the U.S. Food and Drug Administration's (FDA) Adverse Event Reporting System (FAERS). Methods: We searched the FAERS database from October 1, 2009, to March 31, 2022, for reports of all FDA-approved on-demand therapies for HAE: plasma-derived C1-inhibitor (pdC1-INH), ecallantide, icatibant, and recombinant C1-inhibitor (rhC1-INH). ADRs in which the drug was listed as the "primary suspect" were recorded for each drug. ADR preferred terms were grouped into 18 ADR domains based on semantic and/or clinical similarity, and the number of reports for each drug was calculated per year from the time of approval through March 2022, and descriptive results were presented. Preferred terms associated with administration-site ADRs identified from clinical trials and denoted on approved HAE drug U.S. package inserts were examined in a complementary analysis. Results: The highest reported rates of administration-site ADRs per year were site pain (17.9 reports per year), site erythema (7.4 per year), and site swelling (6.7 per year). RhC1-INH was the only drug for which access-site complications and/or malfunctions were reported (9.5 per year). PdC1-INH had the highest rate of incorrect route of product administration (3.7 per year). PdC1-INH showed statistically significant elevated reporting rate of injection-site reactions (reporting odds ratio [ROR] 3.59 [2.36-5.46]; empirical Bayesian geometric mean [EBGM] 1.97 [1.39]). Icatibant and rhC1-INH showed a statistical trend toward an increased reporting rate of administration-site reactions. Conclusion: Real-world data from FAERS were generally consistent with adverse events reported in clinical trials and suggest that patients experience substantial treatment burden associated with FDA-approved parenteral on-demand therapies for HAE attacks. It should be noted that ADR rates are not exposure adjusted and are based on spontaneous reporting.

Evaluation of a simplified pharmacovigilance tool for general practitioners: 5 years of insight

Spontaneous reporting of adverse drug reactions (ADRs) is the cornerstone of pharmacovigilance. However, major underreporting exists. The main objective of this study was to assess the use of a pharmacovigilance simplified reporting tool (PSRT) by general practitioners (GPs) and, secondarily, to describe the quality of ADR reports during this period. The PSRT was proposed on June 1st, 2015, for the 1290 GPs in the Western Normandy Region. The number and quality of ADRs reported monthly by GPs were prospectively collected from June 1st, 2015, to May 31st, 2020 (Period 2), and compared to those reported during a control period (June 1st, 2010, to May 31st, 2015, Period 1). During all the periods, 920 reports were made by 307 GPs (198 reports in Period 1 and 722 reports in Period 2), with 477 reports (51.8%) using the PSRT. During Period 2, the monthly number of reports was multiplied by 3.5 (p < 0.0001), and the number of GPs was 1.4 compared to that in Period 1 (p = 0.01). Our PSRT showed effectiveness in quantitative and qualitative terms. It must now go further and be integrated into GP software to facilitate ADR reporting nationwide.

Enhancing antidepressant safety surveillance: comparative analysis of adverse drug reaction signals in spontaneous reporting and healthcare claims databases

Background/Objective: Spontaneous reporting systems (SRS) such as the Korea Adverse Event Reporting System (KAERS) are limited in their ability to detect adverse drug reaction (ADR) signals due to their limited data on drug use. Conversely, the national health insurance claim (NHIC) data include drug use information for all qualifying residents. This study aimed to compare ADR signal profiles for antidepressants between KAERS and NHIC, evaluating the extent to which detected signals belong to common ADRs and labeling information. Materials and Methods: ADR signal detection in KAERS and NHIC databases, spanning January to December 2017, employed disproportionality analysis. Signal classes were determined based on System Organ Class (SOC) of the Medical Dictionary for Regulatory Activities (MedDRA). Also, Common ADR Coverage (CAC), the proportion of detected signals deemed common ADRs, and labeling information coverage (LIC) represented by mean average precision (mAP) were calculated. Additionally, protopathic bias and relative risk (RR) evaluation were performed to check for signal robustness. Results: Signal detection revealed 51 and 62 signals in KAERS and NHIC databases, respectively. Both systems predominantly captured signals related to nervous system disorders, comprising 33.3% (N = 17) in KAERS and 50.8% (N = 31) in NHIC. Regarding the type of antidepressants, KAERS predominantly reported signals associated with tricyclic antidepressants (TCAs) (N = 21, 41.2%), while NHIC produced most signals linked to selective serotonin reuptake inhibitors (SSRIs) (N = 22, 35.5%). KAERS exhibited higher CAC (68.63% vs. 29.03%) than NHIC. LIC was also higher in KAERS than in NHIC (mAP for EB05: 1.00 vs. 0.983); i.e., NHIC identified 5 signals not documented in drug labeling information, while KAERS found none. Among the unlabeled signals, one (Duloxetine-Myelopathy) was from protopathic bias, and two (duloxetine-myelopathy and tianeptine-osteomalacia) were statistically significant in RR. Conclusion: NHIC exhibited greater capability in detecting ADR signals associated with antidepressant use, encompassing unlabeled ADR signals, compared to KAERS. NHIC also demonstrated greater potential for identifying less common ADRs. Further investigation is needed for signals detected exclusively in NHIC but not covered by labeling information. This study underscores the value of integrating different sources of data, offering substantial regulatory insights and enriching the scope of pharmacovigilance.

Constructing a finer-grained representation of clinical trial results from ClinicalTrials.gov

Randomized controlled trials are essential for evaluating clinical interventions; however, selective reporting and publication bias in medical journals have undermined the integrity of the clinical evidence system. ClinicalTrials.gov serves as a valuable and complementary repository, yet synthesizing information from it remains challenging. This study introduces a curated dataset that extends beyond the traditional PICO framework. It links efficacy with safety results at the experimental arm group level within each trial, and connects them across all trials through a knowledge graph. This novel representation effectively bridges the gap between generally described searchable information and specifically detailed yet underutilized reported results, and promotes a dual-faceted understanding of interventional effects. Adhering to the "calculate once, use many times" principle, the structured dataset will enhance the reuse and interpretation of ClinicalTrials.gov results data. It aims to facilitate more systematic evidence synthesis and health technology assessment, by incorporating both positive and negative results, distinguishing biomarkers, patient-reported outcomes, and clinical endpoints, while also balancing both efficacy and safety outcomes for a given medical intervention.

A patient safety knowledge graph supporting vaccine product development

BACKGROUND

Knowledge graphs are well-suited for modeling complex, unstructured, and multi-source data and facilitating their analysis. During the COVID-19 pandemic, adverse event data were integrated into a knowledge graph to support vaccine safety surveillance and nimbly respond to urgent health authority questions. Here, we provide details of this post-marketing safety system using public data sources. In addition to challenges with varied data representations, adverse event reporting on the COVID-19 vaccines generated an unprecedented volume of data; an order of magnitude larger than adverse events for all previous vaccines. The Patient Safety Knowledge Graph (PSKG) is a robust data store to accommodate the volume of adverse event data and harmonize primary surveillance data sources.

METHODS

We designed a semantic model to represent key safety concepts. We built an extract-transform-load (ETL) data pipeline to parse and import primary public data sources; align key elements such as vaccine names; integrated the Medical Dictionary for Regulatory Activities (MedDRA); and applied quality metrics. PSKG is deployed in a Neo4J graph database, and made available via a web interface and Application Programming Interfaces (APIs).

RESULTS

We import and align adverse event data and vaccine exposure data from 250 countries on a weekly basis, producing a graph with 4,340,980 nodes and 30,544,475 edges as of July 1, 2022. PSKG is used for ad-hoc analyses and periodic reporting for several widely available COVID-19 vaccines. Analysis code using the knowledge graph is 80% shorter than an equivalent implementation written entirely in Python, and runs over 200 times faster.

CONCLUSIONS

Organizing safety data into a concise model of nodes, properties, and edge relationships has greatly simplified analysis code by removing complex parsing and transformation algorithms from individual analyses and instead managing these centrally. The adoption of the knowledge graph transformed how the team answers key scientific and medical questions. Whereas previously an analysis would involve aggregating and transforming primary datasets from scratch to answer a specific question, the team can now iterate easily and respond as quickly as requests evolve (e.g., "Produce vaccine-X safety profile for adverse event-Y by country instead of age-range").

Sex-biased gene expression and gene-regulatory networks of sex-biased adverse event drug targets and drug metabolism genes

BACKGROUND

Previous pharmacovigilance studies and a retroactive review of cancer clinical trial studies identified that women were more likely to experience drug adverse events (i.e., any unintended effects of medication), and men were more likely to experience adverse events that resulted in hospitalization or death. These sex-biased adverse events (SBAEs) are due to many factors not entirely understood, including differences in body mass, hormones, pharmacokinetics, and liver drug metabolism enzymes and transporters.

METHODS

We first identified drugs associated with SBAEs from the FDA Adverse Event Reporting System (FAERS) database. Next, we evaluated sex-specific gene expression of the known drug targets and metabolism enzymes for those SBAE-associated drugs. We also constructed sex-specific tissue gene-regulatory networks to determine if these known drug targets and metabolism enzymes from the SBAE-associated drugs had sex-specific gene-regulatory network properties and predicted regulatory relationships.

RESULTS

We identified liver-specific gene-regulatory differences for drug metabolism genes between males and females, which could explain observed sex differences in pharmacokinetics and pharmacodynamics. In addition, we found that ~ 85% of SBAE-associated drug targets had sex-biased gene expression or were core genes of sex- and tissue-specific network communities, significantly higher than randomly selected drug targets. Lastly, we provide the sex-biased drug-adverse event pairs, drug targets, and drug metabolism enzymes as a resource for the research community.

CONCLUSIONS

Overall, we provide evidence that many SBAEs are associated with drug targets and drug metabolism genes that are differentially expressed and regulated between males and females. These SBAE-associated drug metabolism enzymes and drug targets may be useful for future studies seeking to explain or predict SBAEs.

Safety profile of methotrexate used off-label in ectopic pregnancy: an active monitoring study based on a Chinese hospital pharmacovigilance system

OBJECTIVE

Methotrexate (MTX) is characterized as first-line therapy although its indication of ectopic pregnancy is off-label use. We aimed to conduct a retrospective cohort study to investigate the incidence, characteristics of adverse drug reactions (ADRs) of MTX, provide valuable insights for medical workers.

METHODS

Basing on China Hospital Pharmacovigilance System (CHPS), a retrospective analysis was performed to evaluate the safety of MTX (n = 672). An active monitoring model was set to detect ADR signals from the hospital information system. Frequency, Common Terminology Criteria for Adverse Events (CTCAE) grade proportion and association of dose exposure with ADRs were presented as outcomes.

RESULTS

The total incidence of ADRs was 54.0%. Anaemia (37.6%) was the most frequent ADR, followed by hepatic function abnormal (11.3%), hyperuricemia (6.1%), neutropenia (4.6%), leukopenia (4.0%), and dyslipidaemia (2.5%). For the composition of all ADRs, CTCAE grade one, two and three dominated for 86.3%, 12.1% and 1.6%, respectively. The severity of hepatic function abnormal was more serious in the two-dose exposed group (p = .021), while other types of ADRs had no statistical or clinical differences. Logistic regression analysis showed the incidence of any ADRs (OR 1.87 [1.31-2.64]; p = .001), hepatic function abnormal (OR 2.75 [1.69-4.48]; p < .001), dyslipidaemia (OR 5.15 [1.87-14.13]; p = .001) were significantly higher in the two-dose exposed group. After adjusted, the positive associations were still maintained.

CONCLUSIONS

MTX is quite safe in ectopic pregnancy, despite its mild to moderate hematotoxicity, hepatotoxicity and nephrotoxicity. Taking CHPS can present the accurate denominator of the incidence of adverse drug reactions into account, our study advocates that it may have great potential to be used as an active monitoring tool for off-label drug use risk management.

Unpacking adverse events and associations post COVID-19 vaccination: a deep dive into vaccine adverse event reporting system data

INTRODUCTION

The rapid development of COVID-19 vaccines has provided crucial tools for pandemic control, but the occurrence of vaccine-related adverse events (AEs) underscores the need for comprehensive monitoring.

METHODS

This study analyzed the Vaccine Adverse Event Reporting System (VAERS) data from 2020-2022 using statistical methods such as zero-truncated Poisson regression and logistic regression to assess associations with age, gender groups, and vaccine manufacturers.

RESULTS

Logistic regression identified 26 System Organ Classes (SOCs) significantly associated with age and gender. Females displayed especially higher odds in SOC 19 (Pregnancy, puerperium and perinatal conditions), while males had higher odds in SOC 25 (Surgical and medical procedures). Older adults (>65) were more prone to symptoms like Cardiac disorders, whereas those aged 18-65 showed susceptibility to AEs like Skin and subcutaneous tissue disorders. Moderna and Pfizer vaccines induced fewer SOC symptoms compared to Janssen and Novavax. The zero-truncated Poisson regression model estimated an average of 4.243 symptoms per individual.

CONCLUSION

These findings offer vital insights into vaccine safety, guiding evidence-based vaccination strategies and monitoring programs for precise and effective outcomes.

Interventions for Dry Eye: An Overview of Systematic Reviews

Importance

Dry eye is a common ocular disease that can have substantial morbidity. Systematic reviews provide evidence for dry eye interventions and can be useful for patients, clinicians, and clinical guideline developers. Overviews of reviews use explicit and systematic methods to synthesize findings from multiple systematic reviews, but currently, there are no overviews of systematic reviews investigating interventions for dry eye.

Objective

To summarize the results of reliable systematic reviews of dry eye interventions and to highlight the evidence gaps identified.

Evidence Review

We searched the Cochrane Eyes and Vision US satellite database and included reliable systematic reviews evaluating dry eye interventions published from 2016 to 2022. We reported the proportion of systematic reviews that were reliable with reasons for unreliability. Critical and important outcomes from reliable systematic reviews were extracted and verified. Critical outcomes included dry eye-related patient-reported outcome measures. Results were synthesized from reliable systematic reviews to provide summaries of evidence for each intervention. Evidence for each intervention was defined as conclusive or inconclusive depending on whether high-certainty evidence across systematic reviews was available according to Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) criteria and whether findings reached statistical or clinical significance. Recommendations were made for further research.

Findings

Within the Cochrane Eyes and Vision US satellite database, 138 potentially relevant systematic reviews were identified, 71 were considered eligible, and 26 (37%) were assessed as reliable. Among reliable systematic reviews, no conclusive evidence was identified for any dry eye intervention. Inconclusive evidence suggested that environmental modifications, dietary modifications, artificial tears and lubricants, punctal occlusion, intense pulsed light therapy, vectored thermal pulsation therapy (Lipiflow), topical corticosteroids, topical cyclosporine A, topical secretagogues, and autologous serum may be effective. Only unreliable systematic reviews evaluated lifitegrast, oral antibiotics, and moisture chamber devices.

Conclusions and Relevance

This overview of systematic reviews found some evidence that dry eye interventions may be effective, but no conclusive evidence was available. The conduct and reporting of most systematic reviews for dry eye interventions warrant improvement, and reliable systematic reviews are needed to evaluate lifitegrast, oral antibiotics, and moisture chamber devices.

A Real-World Data Derived Pharmacovigilance Assessment on Drug-Induced Nephropathy: Implication on Gaps in Patient Care

This retrospective cross-sectional study aims to investigate the prevalence and seriousness of drug-induced nephrotoxicity and to identify clinical predictors intensifying the seriousness of nephrotoxicity. Adverse drug events (ADEs) reported to the Korean Adverse Event Reporting System Database (KAERS DB) from January 2012 to December 2021 were investigated. The association between the seriousness and the etiologic drug was estimated in reporting odds ratio (ROR) based on disproportionality analysis. Logistic regression was utilized to recognize predictors associated with serious nephrotoxicity. The majority of ADEs were reported in ages 30 to 59, and immunosuppressants were the most etiologic medications. ADEs involving antibiotics, including vancomycin (ROR 0.268; 95% CI 0.129-0.557), were less likely to be serious. More than 93% of cyclosporine-related ADEs were serious nephrotoxicity, whereas tacrolimus was less likely to report serious nephrotoxicity (ROR 0.356; 95% CI 0.187-0.680). The risk of serious nephrotoxicity was decreased with aging (ROR 0.955; 95% CI 0.940-0.972) while increased in women (OR 2.700; 95% CI 1.450-5.008). Polypharmacy was associated with increased risk of interstitial nephritis (OR 1.019; 95% CI 1.001-1.038). However, further studies investigating the impact of clinical practice on ADE incidences as well as clinical prognosis related to nephrotoxicity are obligated.

Understanding the Medical Dictionary for Regulatory Activities (MedDRA) reporting of herpes simplex virus (HSV) adverse events in atopic dermatitis clinical trials

Drug efficacy is best evaluated by randomized, controlled, double-blind clinical trials; however, safety is harder to assess. The Medical Dictionary for Regulatory Activities (MedDRA) is used to track and categorize adverse events (AE) during clinical trials. Recent atopic dermatitis (AD) clinical trials were reviewed to illustrate how an understanding of MedDRA may be helpful when evaluating the rates and nature of adverse events related to herpes simplex virus (HSV) infection. All completed AD clinical trials (excluding phase I studies) with results on clinicaltrials.gov (01/01/2018-01/31/2023) were queried in January 2023. MedDRA version, preferred term (PT) for AEs captured as "HSV", PTs for other AEs possibly related to HSV, frequency thresholds for reporting non-serious AEs, and route of treatment were recorded. Of the 46 clinical trials, 17 had PTs for AEs captured as "HSV". Among all studies, 11 different versions of MedDRA were utilized, from versions 10 to 24.1. In all studies, PTs for AEs captured as "HSV" were listed in the Infections and Infestations system organ class (SOC) classification of MedDRA. PTs varied from "herpes simplex", "herpes virus infection", "herpes ophthalmic", "ophthalmic herpes simplex", "nasal herpes", "oral herpes", "herpes dermatitis", "eczema herpeticum", "genital herpes simplex", and "genital herpes." While one clinical trial of dupilumab (NCT03359356) simply reported the PT "oral herpes" as an AE, a clinical trial of DS107 (NCT03817190) reported the PTs "oral herpes", "herpes simplex", and "herpes virus infection" separately. In the DS107 clinical trial, it is unclear if the same adverse event was reported under multiple PTs or if multiple HSV-related AEs occurred. Although the definition of HSV is unchanged from 2018 to 2023 and there are few changes between MedDRA versions, coding for HSV is complex. HSV events can be reported in different ways, which may impact the interpretation of a drug's safety profile.

Harms were detected but not reported in six clinical trials of gabapentin

OBJECTIVES

We sought to assess and report harms that were observed but not disclosed previously in clinical trials of gabapentin.

STUDY DESIGN AND SETTING

We reanalyzed individual participant data from six randomized parallel trials of gabapentin for neuropathic pain, and we conducted meta-analyses. Between 1996 and 2003, adult participants were assigned to gabapentin (600 mg-3,600 mg per day) or placebo for 7-14 weeks. We calculated the proportion of participants with: one or more adverse events (AEs), one or more serious AEs, discontinued, discontinued because AEs. We also estimated effects on AEs at three levels of aggregation using COSTART, a hierarchical system for classifying AEs: body system, midlevel system, preferred term.

RESULTS

We found evidence of important harms that were neither included in published trial reports nor included in systematic reviews. Aggregating related harms led to greater confidence that gabapentin might harm the nervous system and possibly the digestive, metabolic and nutritional, respiratory, sensory, and urogenital body systems. Nervous system harms were more common than previous reports suggest.

CONCLUSION

Clinical trials identified harms that were not reported publicly, and journal articles overstated uncertainty about harms. Relying on journal articles to evaluate gabapentin's harms could contribute to incomplete and misleading conclusions in systematic reviews and guidelines. To prevent bias arising from the selective nonreporting of results, journal articles should describe how to access data for all harms observed in clinical trials (e.g., by sharing individual participant data).

Combining varenicline preloading with Acceptance and Commitment Therapy (ACT) in persons with serious mental illness who smoke: The randomized ACTSLow pilot feasibility trial

BACKGROUND

People with serious mental illness (SMI; bipolar [BD] or schizophrenia spectrum disorders [SSD]) who smoke have 30-60% lower odds of quitting and are more prone to experience neuropsychiatric adverse events (NPSAEs) when quitting than smokers without SMI. We pilot-tested the feasibility of combining two different dosing strategies of varenicline preloading with Acceptance and Commitment Therapy (ACT) in persons with SMI in an attempt to bolster quit rates without increasing NPSAEs.

METHODS

Twelve-week, single center, randomized, double-blind, pilot feasibility trial of low (0.5mg twice daily, slower titration) versus standard dose (1.0mg twice daily, standard titration) varenicline in persons with BD or SSD with a 12-week follow-up. All participants received up to 10 sessions of ACT for smoking cessation. Participants were asked to preload with varenicline while still smoking and set a flexible target quit day (TQD) by day 35.

RESULTS

Recruitment was hampered by shutdowns related to COVID-19 and the worldwide varenicline recall, respectively. Retention goals were met. Treatment satisfaction was high across both dosing and diagnostic groups. Most participants (92.9%) adhered to preloading instructions and the flexible TQD. Seven-day point prevalence abstinence at week 12 was highest in BD participants (37.5%) but lowest in SSD participants (16.7%) who received the standard dose. Medication was well tolerated.

CONCLUSIONS

Although recruitment was hindered by unanticipated world events, feasibility was demonstrated. Participants adhered to and were highly satisfied with the combination of pre-cessation varenicline plus ACT. Findings support testing this combined treatment approach in a fully powered trial of persons with BD who smoke.

Patient safety classifications, taxonomies and ontologies, part 2: A systematic review on content coverage

BACKGROUND

Content coverage of patient safety ontology and classification systems should be evaluated to provide a guide for users to select appropriate ones for specific applications. In this review, we identified and compare content coverage of patient safety classifications and ontologies.

METHODS

We searched different databases and ontology/classification repositories to identify these classifications and ontologies. We included patient safety-related taxonomies, ontologies, classifications, and terminologies. We identified and extracted different concepts covered by these systems and mapped these concepts to international classification for patient safety (ICPS) and finally compared the content of these systems.

RESULTS

Finally, 89 papers (77 classifications or ontologies) were analyzed. Thirteen classifications have been developed to cover all medical domains. Among specific domain systems, most systems cover medication (16), surgery (8), medical devices (3), general practice (3), and primary care (3). The most common patient safety-related concepts covered in these systems include incident types (41), contributing factors/hazards (31), patient outcomes (29), degree of harm (25), and action (18). However, stage/phase (6), incident characteristics (5), detection (5), people involved (5), organizational outcomes (4), error type (4), and care setting (3) are some of the less covered concepts in these classifications/ontologies.

CONCLUSION

Among general systems, ICPS, World Health Organization's Adverse Reaction Terminology (WHO-ART), and Ontology of Adverse Events (OAE) cover most patient safety concepts and can be used as a gold standard for all medical domains. As a result, reporting systems could make use of these broad classifications, but the majority of their covered concepts are related to patient outcomes, with the exception of ICPS, which covers other patient safety concepts. However, the ICPS does not cover specialized domain concepts. For specific medical domains, MedDRA, NCC MERP, OPAE, ADRO, PPST, OCCME, TRTE, TSAHI, and PSIC-PC provide the broadest coverage of concepts. Many of the patient safety classifications and ontologies are not formally registered or available as formal classification/ontology in ontology repositories such as BioPortal. This study may be used as a guide for choosing appropriate classifications for various applications or expanding less developed patient safety classifications/ontologies. Furthermore, the same concepts are not represented by the same terms; therefore, the current study could be used to guide a harmonization process for existing or future patient safety classifications/ontologies.

Knowledge and data-driven prediction of organ failure in critical care patients

Purpose

The early detection of organ failure mitigates the risk of post-intensive care syndrome and long-term functional impairment. The aim of this study is to predict organ failure in real-time for critical care patients based on a data-driven and knowledge-driven machine learning method (DKM) and provide explanations for the prediction by incorporating a medical knowledge graph.

Methods

The cohort of this study was a subset of the 4,386 adult Intensive Care Unit (ICU) patients from the MIMIC-III dataset collected between 2001 and 2012, and the primary outcome was the Delta Sequential Organ Failure Assessment (SOFA) score. A real-time Delta SOFA score prediction model was developed with two key components: an improved deep learning temporal convolutional network (S-TCN) and a graph-embedding feature extraction method based on a medical knowledge graph. Entities and relations related to organ failure were extracted from the Unified Medical Language System to build the medical knowledge graph, and patient data were mapped onto the graph to extract the embeddings. We measured the performance of our DKM approach with cross-validation to avoid the formation of biased assessments.

Results

An area under the receiver operating characteristic curve (AUC) of 0.973, a precision of 0.923, a NPV of 0.989, and an F1 score of 0.927 were achieved using the DKM approach, which significantly outperformed the baseline methods. Additionally, the performance remained stable following external validation on the eICU dataset, which consists of 2,816 admissions (AUC = 0.981, precision = 0.860, NPV = 0.984). Visualization of feature importance for the Delta SOFA score and their relationships on the basic clinical medical (BCM) knowledge graph provided a model explanation.

Conclusion

The use of an improved TCN model and a medical knowledge graph led to substantial improvement in prediction accuracy, providing generalizability and an independent explanation for organ failure prediction in critical care patients. These findings show the potential of incorporating prior domain knowledge into machine learning models to inform care and service planning.

Supplementary Information

The online version of this article contains supplementary material available 10.1007/s13755-023-00210-5.

Acute kidney injury associated with nephrotoxic drugs in critically ill patients: a multicenter cohort study using electronic health record data

Background

Nephrotoxic drugs frequently cause acute kidney injury (AKI) in adult intensive care unit (ICU) patients. However, there is a lack of large pharmaco-epidemiological studies investigating the associations between drugs and AKI. Importantly, AKI risk factors may also be indications or contraindications for drugs and thereby confound the associations. Here, we aimed to estimate the associations between commonly administered (potentially) nephrotoxic drug groups and AKI in adult ICU patients whilst adjusting for confounding.

Methods

In this multicenter retrospective observational study, we included adult ICU admissions to 13 Dutch ICUs. We measured exposure to 44 predefined (potentially) nephrotoxic drug groups. The outcome was AKI during ICU admission. The association between each drug group and AKI was estimated using etiological cause-specific Cox proportional hazard models and adjusted for confounding. To facilitate an (independent) informed assessment of residual confounding, we manually identified drug group-specific confounders using a large drug knowledge database and existing literature.

Results

We included 92 616 ICU admissions, of which 13 492 developed AKI (15%). We found 14 drug groups to be associated with a higher hazard of AKI after adjustment for confounding. These groups included established (e.g. aminoglycosides), less well established (e.g. opioids) and controversial (e.g. sympathomimetics with α- and β-effect) drugs.

Conclusions

The results confirm existing insights and provide new ones regarding drug associated AKI in adult ICU patients. These insights warrant caution and extra monitoring when prescribing nephrotoxic drugs in the ICU and indicate which drug groups require further investigation.

Drug-drug interaction prediction: databases, web servers and computational models

In clinical treatment, two or more drugs (i.e. drug combination) are simultaneously or successively used for therapy with the purpose of primarily enhancing the therapeutic efficacy or reducing drug side effects. However, inappropriate drug combination may not only fail to improve efficacy, but even lead to adverse reactions. Therefore, according to the basic principle of improving the efficacy and/or reducing adverse reactions, we should study drug-drug interactions (DDIs) comprehensively and thoroughly so as to reasonably use drug combination. In this review, we first introduced the basic conception and classification of DDIs. Further, some important publicly available databases and web servers about experimentally verified or predicted DDIs were briefly described. As an effective auxiliary tool, computational models for predicting DDIs can not only save the cost of biological experiments, but also provide relevant guidance for combination therapy to some extent. Therefore, we summarized three types of prediction models (including traditional machine learning-based models, deep learning-based models and score function-based models) proposed during recent years and discussed the advantages as well as limitations of them. Besides, we pointed out the problems that need to be solved in the future research of DDIs prediction and provided corresponding suggestions.

Hospital registration of adverse drug reactions in electronic health records: importance and contribution to pharmacovigilance

BACKGROUND

Information on registered adverse drug reactions (ADRs) in hospitals may provide a large real-world data source that can be used to ensure patients' safety. This study aimed to assess the potential contribution of hospital registration of ADRs in electronic health records (EHR) to pharmacovigilance.

RESEARCH DESIGN AND METHODS

An observational retrospective descriptive study using data from the Jeroen Bosch Hospital in the Netherlands in 2019. 'Serious and/or severe' and 'previously unknown' ADRs registered systematically in the corresponding field of EHRs were assessed.

RESULTS

ADR data concerning 1010 patients were included. In total, 1630 ADRs were registered in EHRs. Fifty-eight serious and/or severe ADRs (5.2%) were registered. Tubulointerstitial nephritis was the most frequently registered severe ADR and was mainly associated with antibacterials for systemic use. A total of 82 previously unknown ADRs (5%) were registered. 'Migraine' and 'chest pain' were the most frequently registered unknown ADRs. Additionally, 25 ADRs (1.5%) were registered that may be attributable to 10 drugs 'under additional monitoring.'

CONCLUSIONS

Hospital registrations of ADRs in EHRs provide information on ADRs, which are challenging to assess during clinical trials. However, improvements are required to optimize this registration before it can serve as a valuable data source for pharmacovigilance purposes.

Safety and effectiveness of bDMARDs during pregnancy in patients with rheumatic diseases: Real-world data from the BIOBADASER registry

INTRODUCTION

Inflammatory rheumatic diseases usually affect women of childbearing age treated with biologic drugs. However, there is a lack of literature on the efficacy and toxicity of biologic disease-modifying drugs during pregnancy. The aim of this study was to determine the presence of pregnant patients treated with bDMARDs in a real-world dataset and to examine the impact of pregnancy and lactation on the evolution of rheumatic disease in a registry of Spanish patients.

METHOD

This was a multicentre prospective study with a real-world setting. Information was obtained from BIOBADASER registry. Patients included are women who got pregnant until November 2020 from 19 rheumatology units. We conducted proportions, means, and standard deviations (SD) to describe the study population and the use of treatments. T-test and Chi-square test were applied to assess differences between groups.

RESULT

Ninety cases of pregnancy were registered (n=68 full-term pregnancies; n=22 spontaneous miscarriages). Most of the cases discontinued bDMARDs during pregnancy (78.9%) but 13 cases continued treatment during pregnancy, mainly using certolizumab pegol. These cases were obtaining better management of rheumatic disease, although the differences were not statistically significant [DAS28-CRP, 2.9 (SD: 1.6) vs. 2.0 (1.2), p=.255; DAS28-ESR, 2.2 (1.0) vs. 1.7 (.5), p=.266]. No serious adverse events were reported during pregnancy and lactation.

CONCLUSION

Being pregnant is still an uncommon condition in patients with rheumatic diseases and using bDMARDs. Our results show that rheumatic disease tended to progress better during pregnancy in patients who continued to take bDMARDs.