A Risk-Factor Model for Antineoplastic Drug-Induced Serious Adverse Events in Cancer Inpatients: A Retrospective Study Based on the Global Trigger Tool and Machine Learning.

The objective of this study was to apply a machine learning method to evaluate the risk factors associated with serious adverse events (SAEs) and predict the occurrence of SAEs in cancer inpatients using antineoplastic drugs. A retrospective review of the medical records of 499 patients diagnosed with cancer admitted between January 1 and December 31, 2017, was performed. First, the Global Trigger Tool (GTT) was used to actively monitor adverse drug events (ADEs) and SAEs caused by antineoplastic drugs and take the number of positive triggers as an intermediate variable. Subsequently, risk factors with statistical significance were selected by univariate analysis and least absolute shrinkage and selection operator (LASSO) analysis. Finally, using the risk factors after the LASSO analysis as covariates, a nomogram based on a logistic model, extreme gradient boosting (XGBoost), categorical boosting (CatBoost), adaptive boosting (AdaBoost), light-gradient-boosting machine (LightGBM), random forest (RF), gradient-boosting decision tree (GBDT), decision tree (DT), and ensemble model based on seven algorithms were used to establish the prediction models. A series of indicators such as the area under the ROC curve (AUROC) and the area under the PR curve (AUPR) was used to evaluate the model performance. A total of 94 SAE patients were identified in our samples. Risk factors of SAEs were the number of triggers, length of stay, age, number of combined drugs, ADEs occurred in previous chemotherapy, and sex. In the test cohort, a nomogram based on the logistic model owns the AUROC of 0.799 and owns the AUPR of 0.527. The GBDT has the best predicting abilities (AUROC = 0.832 and AUPR = 0.557) among the eight machine learning models and was better than the nomogram and was chosen to establish the prediction webpage. This study provides a novel method to accurately predict SAE occurrence in cancer inpatients.

Antifungal Therapy with Azoles Induced the Syndrome of Acquired Apparent Mineralocorticoid Excess: a Literature and Database Analysis.

We aimed to estimate the risk of varied antifungal therapy with azoles causing the syndrome of acquired apparent mineralocorticoid excess (AME) in real-world practice. First, we conducted a disproportionality analysis based on data from the FDA Adverse Event Reporting System (FAERS) database to characterize the signal differences of triazoles-related AME. Second, a systematic review was conducted, and clinical features of AME cases reported in clinical practice were described. In the FAERS database, we identified 27 cases of triazoles-AME, posaconazole [ROR = 865.37; 95%CI (464.14; 1613.45)], and itraconazole [ROR = 556.21; 95% (303.05; 1020.85)] significantly increased the risk of AME events, while fluconazole, voriconazole, and isavuconazole did not affect any of the mineralocorticoid excess targets. Eighteen studies with 39 cases raised evidence of AME following posaconazole and itraconazole treatment, and another 27 cases were identified by analysis of the description of clinical features in the FAERS database. The average age of 66 patients was 55.5 years (6-87 years). AME mainly occurs in patients with posaconazole concentrations above 3 µg/mL (mean = 4.4 µg/mL, range 1.8∼9.5 µg/mL), and is less likely to occur when levels are below 2 µg/mL (6%). The median time to event onset was 11.5 weeks, and 50% of the adverse events occurred within 3 months for posaconazole. The presented study supports very recent findings that posaconazole and itraconazole, but not the other three azole antifungals investigated, are associated with AME and that the effects are dose-dependent, which allows for a dose de-escalation strategy and for substitution with fluconazole, isavuconazole, or voriconazole to resolve the adverse effects.

Human papillomavirus vaccine-associated premature ovarian insufficiency and related adverse events: data mining of Vaccine Adverse Event Reporting System.

We detected disproportionate reports of premature ovarian insufficiency (POI) and related events, including amenorrhea, menstruation irregular, FSH increased, and premature menopause, following human papillomavirus (HPV) vaccine from FDA Vaccine Adverse Event Reporting System (VAERS). The signal was detected by the methods of Bayesian Confidence Propagation Neural Network (BCPNN) and Multi-item Gamma Poisson Shrinker (MGPS). When both methods detected a positive result, a signal was generated. Besides, time-scan map is drawn based on the IC value and 95%CI of BCPNN, if the IC curve showed a steady upward trend and the 95%CI narrowed, the signal was stable and strong association.The results showed that there were not POI reports of HPV vaccine, but VAERS received a total of 2, 389, 27 POI related events for HPV2, HPV4, HPV9 respectively from the year of marketed to 2018. No signal was detected for HPV2. HPV4-POI ralated events were all detected as signals by two methods. There was only one signal of menstruation irregular for HPV9. Time scan of HPV4-POI ralated events showed those signals were stability and strong association, but not for HPV9. Our results only represent statistical association between HPV vaccine and POI related events, causal relationship needs further investigation.

Adverse Event Profiles of Anti-CTLA-4 and Anti-PD-1 Monoclonal Antibodies Alone or in Combination: Analysis of Spontaneous Reports Submitted to FAERS.

BACKGROUND AND OBJECTIVE: Immune checkpoint inhibitors (ICIs)-cytotoxic T-lymphocyte antigen-4 (CTLA-4) and programmed death receptor-1 (PD-1) monoclonal antibodies (mAbs)-either as single agents or in combination have become the standard of care for an increasing number of indications. Understanding both the ICI-associated adverse events (AEs) and the possible rank-order of these drugs in terms of susceptibility is essential if we are to improve the curative effect and reduce toxicity. METHODS: We detected signals of the AEs of ICIs by data mining using the US Food and Drug Administration (FDA) AEs Reporting System (FAERS) database. The definition relied on the preferred terms (PTs) and the standardized MedDRA Queries (SMQs) provided by the Medical Dictionary for Regulatory Activities (MedDRA). Disproportionality analysis was performed by calculating the reporting odds ratios (ROR) with 95% confidence intervals (CIs). RESULTS: Adverse effects of CTLA-4 and PD-1 mAbs were most commonly observed in the skin, gastrointestinal tract, endocrine systems, liver, and lung, and they included rash, diarrhea, colitis, and thyroid dysfunction. Thyroid dysfunction, type 1 diabetes mellitus, and pneumonitis were more closely associated with the use of anti-PD-1, whereas colitis, diarrhea, hypophysitis, and adrenal insufficiency were more closely associated with anti-CTLA-4; rash and hepatitis occurred similarly in both. Disproportionality signals for less common AEs in other organ systems, including the renal, neurological, cardiac, ocular, musculoskeletal, and hematologic systems, were also detected. Nivolumab and pembrolizumab have very similar safety profiles, but the signal strength of AEs increased when combined with ipilimumab. CONCLUSIONS: The results of this study are in agreement with clinical observations, suggesting the usefulness of pharmacovigilance in "real-world" safety monitoring.

Adverse drug events in Chinese pediatric inpatients and associated risk factors: a retrospective review using the Global Trigger Tool.

Understanding the epidemiology and risk factors of adverse drug events (ADEs) in pediatric inpatient is essential if we are to prevent, reduce or ameliorate the harm experienced. The Global Trigger Tool (GTT) is a method of retrospective medical record review that measures harm in hospitalized children. We employed a three-stage retrospective chart review of random samples of 1800 pediatric inpatients discharged from January 2013 to December 2015. 31 kinds of pediatric-specific triggers were made based on the previous trigger tool studies developed for use in adult or pediatric. Positive predictive value (PPV) of individual triggers, as well as ADEs detection rates were calculated. Stepwise logistic regression was performed to investigate risk factors associated with ADEs. Of 1746 patients, detected in 221 patients (12.7%) with 247 ADEs. The PPV of the trigger tool was 13.3%. Of the 247 ADEs, 82.6% were identified as category E, 11.7% category F and 5.7% category H. The pediatric-focused trigger tool is a feasible and useful tool for detecting pediatric ADEs. Especially for patients who have had more drugs, more doses or more admissions which needs to be closely monitored as triggers to improve the safety.