Interpretable machine learning model predicting immune checkpoint inhibitor-induced hypothyroidism: A retrospective cohort study.

Hypothyroidism is a known adverse event associated with the use of immune checkpoint inhibitors (ICIs) in cancer treatment. This study aimed to develop an interpretable machine learning (ML) model for individualized prediction of hypothyroidism in patients treated with ICIs. The retrospective cohort of patients treated with ICIs was from the First Affiliated Hospital of Ningbo University. ML methods applied include logistic regression (LR), random forest classifier (RFC), support vector machine (SVM), and extreme gradient boosting (XGBoost). The area under the receiver-operating characteristic curve (AUC) was the main evaluation metric used. Furthermore, the Shapley additive explanation (SHAP) was utilized to interpret the outcomes of the prediction model. A total of 458 patients were included in the study, with 59 patients (12.88%) observed to have developed hypothyroidism. Among the models utilized, XGBoost exhibited the highest predictive capability (AUC = 0.833). The Delong test and calibration curve indicated that XGBoost significantly outperformed the other models in prediction. The SHAP method revealed that thyroid-stimulating hormone (TSH) was the most influential predictor variable. The developed interpretable ML model holds potential for predicting the likelihood of hypothyroidism following ICI treatment in patients. ML technology offers new possibilities for predicting ICI-induced hypothyroidism, potentially providing more precise support for personalized treatment and risk management.

ADENet: a novel network-based inference method for prediction of drug adverse events.

Identification of adverse drug events (ADEs) is crucial to reduce human health risks and improve drug safety assessment. With an increasing number of biological and medical data, computational methods such as network-based methods were proposed for ADE prediction with high efficiency and low cost. However, previous network-based methods rely on the topological information of known drug-ADE networks, and hence cannot make predictions for novel compounds without any known ADE. In this study, we introduced chemical substructures to bridge the gap between the drug-ADE network and novel compounds, and developed a novel network-based method named ADENet, which can predict potential ADEs for not only drugs within the drug-ADE network, but also novel compounds outside the network. To show the performance of ADENet, we collected drug-ADE associations from a comprehensive database named MetaADEDB and constructed a series of network-based prediction models. These models obtained high area under the receiver operating characteristic curve values ranging from 0.871 to 0.947 in 10-fold cross-validation. The best model further showed high performance in external validation, which outperformed a previous network-based and a recent deep learning-based method. Using several approved drugs as case studies, we found that 32-54% of the predicted ADEs can be validated by the literature, indicating the practical value of ADENet. Moreover, ADENet is freely available at our web server named NetInfer (http://lmmd.ecust.edu.cn/netinfer). In summary, our method would provide a promising tool for ADE prediction and drug safety assessment in drug discovery and development.

Disproportionality Analysis of Stomatitis Associated with Anticancer Drugs Using the Japanese Adverse Drug Event Report Database.

INTRODUCTION: Anticancer drug-induced stomatitis can affect a patient's quality of life and the continuation of drug treatment. Although there have been reports of the occurrence of stomatitis associated with anticancer agents in clinical trials, few Japanese participants have been enrolled in clinical trials and have not been sufficiently investigated. In addition, there has been little attention on research on anticancer drugs associated with stomatitis by patient stratification with different carcinogenic sites. Therefore, the aim of this study was to determine the disproportionality associated with stomatitis for various types of anticancer drugs in different types of cancer patients using the Japanese Adverse Drug Event Report (JADER) database. METHODS: The aim of this study was to identify the disproportionality of stomatitis by analyzing the type of anticancer drug and cancer patients using the Japanese Pharmacovigilance Database. Data obtained from spontaneous reports of adverse events with more than 10 stomatitis outbreaks reported in the JADER database between April 2004 and March 2023 were analyzed. The safety signal for an adverse event was defined as the lower limit of the 95% confidence interval of the reported odds ratio of >1. RESULTS: There were 6,178 reports of drugs associated with stomatitis. Among these, 41 drugs were suggested to be associated with stomatitis, and 41 drugs were detected as signals. These drugs were classified based on their efficacy: antipyrimidines (six drugs), folate metabolism antagonists (three drugs), alkylating agents (four drugs), platinum (three drugs), topoisomerase inhibitors (three drugs), microtubule inhibitors (three drugs), mammalian target of rapamycin (mTOR) inhibitors (two drugs), kinase inhibitors (seven drugs), anti-growth factor antibodies (five drugs) immune checkpoint inhibitors (one drug), and others (four drugs). CONCLUSION: The drugs that may be associated with stomatitis were cell cycle-dependent drugs, epidermal growth factor receptor-tyrosine kinase inhibitors, and mTOR inhibitors. Moreover, this study suggested that anti-growth factor antibodies and immune checkpoint inhibitors may be associated with stomatitis development.

Discontinuation risk from adverse events: immunotherapy alone vs. combined with chemotherapy: a systematic review and network meta-analysis.

BACKGROUND: While immunotherapy combined with chemotherapy (Chemo-IO) is generally recognized for providing superior outcomes compared to monotherapy (mono-IO), it is associated with a higher incidence of treatment-related adverse events (TRAEs), which may lead to treatment discontinuation. In this study, we compared the rates of treatment discontinuation between mono-IO and Chemo-IO as first-line treatments for various solid tumors. METHODS: We systematically reviewed clinical trials from databases (PubMed, Embase, Cochrane Library, and an additional source) published from January 1, 2018, to July 10, 2023. We included phase III randomized controlled trials (RCTs) that utilized immunotherapy agents in at least one arm as first-line treatments for a variety of solid tumors. Data extraction followed the Preferred Reporting Items for Systematic Reviews (PRISMA) extension statement for network meta-analysis. A random effects model was used for the network meta-analysis, with the risk of bias assessed using the Cochrane risk-of-bias tool II. The primary outcomes encompassed treatment discontinuation rates due to TRAEs among patients who underwent immunotherapy, either alone or combined with chemotherapy, for various solid tumors. Pooled relative risks (RRs) with 95% confidence intervals (CIs) were calculated to compare between treatment groups. RESULTS: From 29 RCTs, a total of 21,677 patients and 5 types of treatment were analyzed. Compared to mono-IO, Chemo-IO showed a significantly higher rate of discontinuation due to TRAEs (RR 2.68, 95% CI 1.98-3.63). Subgroup analysis for non-small cell lung cancer (NSCLC) patients also exhibited a greater risk of discontinuation due to TRAEs with Chemo-IO compared to mono-IO (RR 2.93, 95% CI 1.67-5.14). Additional analyses evaluating discontinuation rates due to either treatment emergent adverse events (TEAEs) or AEs regardless of causality (any AEs) consistently revealed an elevated risk associated with Chemo-IO. CONCLUSIONS: Chemo-IO was associated with an elevated risk of treatment discontinuation not only due to TRAEs but also any AEs or TEAEs. Given that the treatment duration can impact clinical outcomes, a subset of patients might benefit more from mono-IO than combination therapy. Further research is imperative to identify and characterize this subset.

Unintended Consequences of Immune Therapy for Immune-Mediated Diseases: Paradoxical Psoriasis and Dupilumab-Associated Musculoskeletal Syndrome.

Two presentations at the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA) 2023 annual meeting focused on unintended consequences of immunomodulatory therapy for psoriasis (PsO). Dr. Elizabeth Wallace presented on unintended consequences of tumor necrosis factor inhibitors for treating PsO and other inflammatory disorders. These consequences include paradoxical PsO, which is defined as unexpected new PsO cases or worsening PsO symptoms seemingly induced by treatment. Dr. Bruce Kirkham focused on unintended consequences of dupilumab treatment, which can include a musculoskeletal syndrome similar to psoriatic arthritis.

Validating methods used to identify non-adherence adverse drug events in Canadian administrative health data.

AIMS: Medication non-adherence is a type of adverse drug event that can lead to untreated and exacerbated chronic illness, and that drives healthcare utilization. Research using medication claims data has attempted to identify instances of medication non-adherence using the proportion of days covered or by examining gaps between medication refills. We sought to validate these measures compared to a gold standard diagnosis of non-adherence made in hospital. METHODS: This was a retrospective analysis of adverse drug events diagnosed during three prospective cohorts in British Columbia between 2008 and 2015 (n = 976). We linked prospectively identified adverse drug events to medication claims data to examine the sensitivity and specificity of typical non-adherence measures. RESULTS: The sensitivity of the non-adherence measures ranged from 22.4% to 37.5%, with a proportion of days covered threshold of 95% performing the best; the non-persistence measures had sensitivities ranging from 10.4% to 58.3%. While a 7-day gap was most sensitive, it classified 61.2% of the sample as non-adherent, whereas only 19.6% were diagnosed as such in hospital. CONCLUSIONS: The methods used to identify non-adherence in administrative databases are not accurate when compared to a gold standard diagnosis by healthcare providers. Research that has relied on administrative data to identify non-adherent patients both underestimates the magnitude of the problem and may label patients as non-adherent who were in fact adherent.

Disproportionality analysis of cardiac adverse events associated with lenvatinib using the Japanese Adverse Drug Event Report database.

AIMS: This study was conducted to examine disproportionality, times to onset, incidence rates and outcomes of lenvatinib-associated cardiac adverse events (AEs) using the Japanese Adverse Drug Event Report database. METHODS: We analysed data for the period between April 2004 and May 2023. Data on cardiac AEs were extracted and the relative disproportionality of AEs was estimated using reporting odds ratios (RORs). Furthermore, Weibull distribution parameters were calculated. RESULTS: Of the 2 230 863 reports analysed, we identified 7684 reports of AEs associated with lenvatinib, including 317 cardiac AEs. Signals were detected for eight cardiac AEs: hypertension, cardiac failure, myocarditis, myocardial infarction, immune-mediated myocarditis, cardiomyopathy, angina unstable and cardiotoxicity. Among these, fatal outcomes were observed for cardiac failure, myocarditis and myocardial infarction. Histograms of median times to onset for the eight detected cardiac AE signals showed that AEs occurred at a median of 3.5-134.5 days after lenvatinib administration. The Weibull distributions showed that cardiac failure occurred early after administration (early failure type), myocarditis occurred in a dose-dependent manner (wearout failure type), and myocardial infarction occurred constantly throughout the exposure period (random failure type). CONCLUSIONS: We focused on cardiac AEs associated with lenvatinib as post-marketing AEs. Serious outcomes can arise after lenvatinib administration. Patients should be monitored for signs of onset of these AEs not only at the start of administration, but also over an extended period.

Validating use of diagnostic codes in Canadian administrative data for identification of adverse drug events.

AIMS: While diagnostic codes from administrative health data might be a valuable source to identify adverse drug events (ADEs), their ability to identify unintended harms remains unclear. We validated claims-based diagnosis codes for ADEs based on events identified in a prospective cohort study and assessed whether key attributes predicted their documentation in administrative data. METHODS: This was a retrospective analysis of 3 prospective cohorts in British Columbia, from 2008 to 2015 (n = 13 969). We linked prospectively identified ADEs to administrative insurance data to examine the sensitivity and specificity of different diagnostic code schemes. We used logistic regression to assess which key attributes (e.g., type of event, symptoms and culprit medications) were associated with better documentation of ADEs in administrative data. RESULTS: Among 1178 diagnosed events, the sensitivity of the diagnostic codes in administrative data ranged from 3.4 to 52.6%, depending on the database and codes used. We found that documentation was worse for certain types of ADEs (dose-related: odds ratio [OR]: 0.32, 95% confidence interval [CI]: 0.15, 0.69; nonadherence events (OR: 0.35, 95% CI: 0.20, 0.62), and better for those experiencing arrhythmias (OR: 4.19, 95% CI: 0.96, 18.28). CONCLUSION: ADEs were not well documented in administrative data. Alternative methods should be explored to capture ADEs for health research.

Analysis of Atypical Antipsychotics-Induced Adverse Events Related to Diabetes Mellitus in Patients With Dementia Using the Japanese Adverse Drug Event Report Database.

BACKGROUND: Patients with dementia are prescribed low-dose atypical antipsychotics (AAPs) to treat psycho-behavioral symptoms. Although AAPs are known to cause diabetes mellitus-related adverse events (DMAEs), information regarding AAPs-induced DMAEs experienced by patients with dementia is lacking. OBJECTIVE: To use the Japan Adverse Drug Event Report (JADER) database to assess the onset tendencies and patterns of DMAEs attributable to AAPs prescribed to patients with dementia. METHODS: We performed an analysis using dementia cases from the JADER database that were registered from April 2004 to December 2022. Data in the JADER database are completely anonymized; thus, we did not require institutional review board approval for using the JADER database in our study. The reporting odds ratio and proportional reporting ratio (PRR) were used to assess the onset tendencies of DMAEs with AAPs. In addition, Weibull shape parameters were used to assess the patterns of DMAEs that occur with the use of AAPs. RESULTS: We identified AAPs associated with DMAEs. In particular, low doses of quetiapine showed the potential to induce DMAEs. An analysis of the onset of DMAEs showed the early failure patterns for AAPs (median onset = 38 days). CONCLUSION AND RELEVANCE: The AAPs may cause DMAEs in patients with dementia. Low doses of quetiapine may induce DMAEs. Health care workers should focus on the development of DMAEs during the early administration period of AAPs. These results may assist with the safe management of patients with dementia who use AAPs.

Artificial intelligence-powered pharmacovigilance: A review of machine and deep learning in clinical text-based adverse drug event detection for benchmark datasets.

OBJECTIVE: The primary objective of this review is to investigate the effectiveness of machine learning and deep learning methodologies in the context of extracting adverse drug events (ADEs) from clinical benchmark datasets. We conduct an in-depth analysis, aiming to compare the merits and drawbacks of both machine learning and deep learning techniques, particularly within the framework of named-entity recognition (NER) and relation classification (RC) tasks related to ADE extraction. Additionally, our focus extends to the examination of specific features and their impact on the overall performance of these methodologies. In a broader perspective, our research extends to ADE extraction from various sources, including biomedical literature, social media data, and drug labels, removing the limitation to exclusively machine learning or deep learning methods. METHODS: We conducted an extensive literature review on PubMed using the query "(((machine learning [Medical Subject Headings (MeSH) Terms]) OR (deep learning [MeSH Terms])) AND (adverse drug event [MeSH Terms])) AND (extraction)", and supplemented this with a snowballing approach to review 275 references sourced from retrieved articles. RESULTS: In our analysis, we included twelve articles for review. For the NER task, deep learning models outperformed machine learning models. In the RC task, gradient Boosting, multilayer perceptron and random forest models excelled. The Bidirectional Encoder Representations from Transformers (BERT) model consistently achieved the best performance in the end-to-end task. Future efforts in the end-to-end task should prioritize improving NER accuracy, especially for 'ADE' and 'Reason'. CONCLUSION: These findings hold significant implications for advancing the field of ADE extraction and pharmacovigilance, ultimately contributing to improved drug safety monitoring and healthcare outcomes.

Signal detection of adverse events associated with gabapentinoid use for chronic pain.

INTRODUCTION: Gabapentinoids (GABA) prescribing as a potential and conceivably safer substitute for opioids has substantially increased. Understanding all potential adverse drug events (ADEs) associated with GABA will guide clinical decision-making for pain management. METHODS: A 20% sample of Medicare enrollees with new chronic pain diagnoses in 2017-2018 was selected. GABA users were those with >=30 consecutive days prescription in a year without opioid prescription. Opioid users were similarly defined. The control group used neither of these drugs. Propensity score match across three groups based on demographics and comorbidity was performed. We used proportional reporting ratio (PRR), Gamma Poisson Shrinker, and tree-based scan statistic (TBSS) to detect ADEs within 3, 6, and 12 months of follow-up. RESULTS: Immunity disorder was detected within 3 months of follow-up by PRR compared to opioid use (PRR:2.33), and by all three methods compared to controls. Complications of transplanted organs/tissues and schizophrenia spectrum/other psychotic disorders were consistently detected by PRR and TBSS within 3 months. Skin disorders were detected by TBSS; and stroke was detected by PRR within 3 months compared to opioid use (PRR:4.74). Some malignancies were detected by PRR within 12 months. Other signals detected in GABA users were neuropathy and nerve disorders. CONCLUSIONS: Our study identified expected and unexpected ADE signals in GABA users. Neurological signals likely related to indications for GABA use. Signals for immunity, mental/behavior, and skin disorders were found in the FDA adverse event reporting system database. Unexpected signals of stroke and cancer require further confirmatory analyses to verify.

Re-Exposure to Culprit Medication Following Adverse Drug Event Diagnosis in Canadian Emergency Department Patients: A Cohort Study.

PURPOSE: The magnitude of repeat exposures to culprit medications after hospital discharge is not well studied. We combined prospective cohort data with administrative health data to understand the frequency of repeat exposures to culprit medications after discharge and the risk factors for their occurrence. METHODS: This was a retrospective analysis of three prospective cohorts of patients who presented to the hospital with an adverse drug event in British Columbia, from 2008 to 2015 (n = 849). We linked prospectively identified adverse drug events to administrative data to examine patterns of redispensing of culprit medications. We used Cox regression to assess risk factors for re-exposure, and conducted subgroup analyses for essential vs. nonessential medications. RESULTS: Among 849 diagnosed adverse drug events, 45.2% had subsequent culprit medication redispensing within a year of hospital discharge. The factors associated with re-exposures included atrial fibrillation, adverse drug event type (e.g. adverse reaction), culprit medication type, and longer historical duration of medication use. CONCLUSIONS: Re-exposures to culprit medications occurred in almost half of the adverse drug events diagnosed in emergency departments. Many of these were appropriate re-exposures to essential medications for indications in which the risk of uncontrolled disease likely outweighed the risk of a repeat adverse event. More research is needed to understand re-exposures to nonessential medications or medications with safer alternatives.

Effect of Enoxaparin and Daikenchuto Coadministration on Hepatic Disorder Markers in Gynecological Cancer Patients after Abdominal Surgery.

Enoxaparin and daikenchuto are commonly administered to prevent venous thromboembolism and intestinal obstruction after gynecological malignancy surgery. However, the effects of their combined use on hepatic function are not well studied. This study aimed to clarify the effects of the coadministration of enoxaparin and daikenchuto on hepatic function. First, Japanese Adverse Drug Event Report (JADER) data were analyzed to identify signals of hepatic disorders. Second, a retrospective observational study of patients who underwent surgery for gynecological malignancies was conducted. This study defined hepatic disorders as an increase in aspartate aminotransferase (AST) or alanine aminotransaminase (ALT) levels above the reference values, using 1-h postoperative values as the baseline. The analysis of JADER data revealed an increased risk for hepatic disorders with the coadministration of enoxaparin and daikenchuto. An observational study also showed higher odds ratios (95% confidence intervals) for the occurrence of hepatic disorders in the coadministration group (4.27; 2.11-8.64) and enoxaparin alone group (2.48; 1.31-4.69) than in the daikenchuto alone group. The median increase in the ALT level was also higher in the coadministration group (34; 15-59) than in the enoxaparin alone (19; 6-38) and daikenchuto alone groups (8; 3-33). In conclusion, our study suggests that compared with the use of enoxaparin or daikenchuto alone, enoxaparin and daikenchuto coadministration increases the risk of hepatic disorders, with more significant increases in AST and ALT levels. Healthcare workers need to be aware of these potential side effects when combining these drugs after surgery for gynecological malignancies.

Cross-Cancer Type Evaluation of Potential Interstitial Lung Disease Complications of Immune Checkpoint Inhibitors Using JADER.

Interstitial lung disease (ILD) is a serious adverse event caused by the administration of immune checkpoint inhibitors (ICIs). However, only few large-scale studies have explored the association among ICI use, underlying cancer type, and ILD complications. This study aimed to analyze the association between the primary cancer type and ICI-induced ILD in a cross-sectional manner using the Japanese Adverse Drug Event Report (JADER) database. Nivolumab and pembrolizumab (anti-programmed cell death 1 (PD-1) antibodies) and durvalumab, avelumab, and atezolizumab (anti-programmed cell death ligand 1 (PD-L1) antibodies) were included as ICIs in this study. Adverse events were identified based on the preferred terms of Medical Dictionary for Regulatory Activities (MedDRA) version 27.0/J listed in the Standardized MedDRA Queries (SMQ) "interstitial lung disease." The reporting odds ratio was calculated to detect the association between ICI use and ILD complications, and a signal was detected if the lower limit of the 95% confidence interval exceeded 1. In the analysis of all cancer types, a signal was detected for all ICIs except avelumab. An association between ICI and ILD was detected for all cancer types with nivolumab. However, pembrolizumab exhibited a signal only in colorectal cancer. In contrast, anti-PD-L1 antibodies displayed signals in five cancer types, excluding head and neck cancer, which was not reported in JADER. Among these cancer types, atezolizumab exhibited a signal only in breast cancer. The results of this study will help guide the safe use of ICIs based on the underlying cancer type in terms of ILD complications.

Probable effects of polypharmacy and equivalent doses of psychotropic drugs on prevalence of adverse drug events among psychiatric inpatients in a general hospital in Japan.

OBJECTIVE: In psychiatry, polypharmacy or high psychotropic drug doses increase adverse drug event (ADE) prevalence. However, the full relationship between polypharmacy and ADEs is unclear, and few studies have evaluated dose equivalents for psychotropic drugs for ADEs. Thus, we conducted a retrospective analysis to clarify the effects of polypharmacy and chlorpromazine (CP)-, diazepam (DAP)-, and imipramine- equivalent doses on all ADEs in inpatients. METHODS: Psychiatric inpatients in a Japanese hospital from April 1, 2016 to March 31, 2018, were enrolled. ADE severity and causality were assessed. Multiple logistic regression analyses were performed to evaluate ADE risk factors. RESULTS: Among 462 patients analyzed, out of 471 patients enrolled, 145 (31.4%) experienced ADEs. The causality assessment determined that "possible" was 96.5%. The most common ADEs were nervous system disorders (35%). Multiple logistic regression analyses indicated an increase in ADE prevalence with the number of drugs used (≥5; p = 0.026); CP-equivalent dose (p = 0.048); and endocrine, nutritional, and metabolic disorders (p = 0.045). DAP-equivalent dose; infectious and parasitic diseases; and injury, poisoning, and consequences of other external causes decreased ADE prevalence (p = 0.047, 0.022, and 0.021, respectively). CONCLUSIONS: Avoiding polypharmacy in psychiatric inpatients and adjusting drug regimens to safe equivalent doses could reduce ADEs during hospitalization.

Inhaled analgesics for the treatment of prehospital acute pain-A systematic review.

BACKGROUND: Many prehospital emergency patients receive suboptimal treatment for their moderate to severe pain. Various factors may contribute. We aim to systematically review literature pertaining to prehospital emergency adult patients with acute pain and the pain-reducing effects, adverse events (AEs), and safety issues associated with inhaled analgetic agents compared with other prehospital analgesic agents. METHODS: As part of an initiative from the Scandinavian Society of Anaesthesia and Intensive Care Medicine, we conducted a systematic review (PROSPERO CRD42018114399), applying the PRISMA guidelines, Grading of Recommendations Assessment, Development, and Evaluation (GRADE), and Cochrane methods, searching the Cochrane Library, Epistemonikos, Centre for Reviews and Dissemination, PubMed, and EMBASE databases (updated March 2024). Inclusion criteria were the use of inhaled analgesic agents in adult patients with acute pain in the prehospital emergency care setting. All steps were performed by minimum of two individual researchers. The primary outcome was pain reduction; secondary outcomes were speed of onset, duration of effect, and relevant AEs. RESULTS: We included seven studies (56,535 patients in total) that compared inhaled agents (methoxyflurane [MF] and nitrous oxide [N2O]) to other drugs or placebo. Study designs were randomized controlled trial (1; n = 60), randomized non-blinded study (1; n = 343), and randomized open-label study (1; n = 270). The remaining were prospective or retrospective observational studies. The evidence according to GRADE was of low or very low quality. No combined meta-analysis was possible. N2O may reduce pain compared to placebo, but not compared to intravenous (IV) paracetamol, and may be less effective compared to morphine and MF. MF may reduce pain compared to paracetamol, ketoprofen, tramadol, and fentanyl. Both agents may be associated with marked but primarily mild AEs. CONCLUSION: We found low-quality evidence suggesting that both MF and N2O are safe and may have a role in the management of pain in the prehospital setting. There is low-quality evidence to support MF as a short-acting single analgesic or as a bridge to IV access and the administration of other analgesics. There may be occupational health issues regarding the prehospital use of N2O.

SARS-CoV-2 mRNA vaccine-related myocarditis and pericarditis: An analysis of the Japanese Adverse Drug Event Report database.

BACKGROUND: The association between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mRNA vaccines and myocarditis/pericarditis in the Japanese population has not been systematically investigated. This study was aimed at clarifying the association between SARS-CoV-2 mRNA vaccines (BNT162b2 and mRNA-1273) and myocarditis/pericarditis as well as influencing factors by using the Japanese Adverse Drug Event Report database. METHODS: Reporting odds ratios (RORs) and 95 % confidence intervals (95 % CIs) for the association between the vaccines and myocarditis/pericarditis were calculated using data from the database (April 2004-December 2023). Age, sex, onset time, and outcomes in symptomatic patients were evaluated. RESULTS: The total number of reports was 880,999 (myocarditis: 1846; pericarditis: 761). The adverse events associated with the vaccines included myocarditis (919 cases) and pericarditis (321 cases), with the ROR [95 % CIs] being significant for both (myocarditis: 30.51 [27.82-33.45], pericarditis: 21.99 [19.03-25.40]). Furthermore, the ROR [95 % CIs] of BNT162b2 and mRNA-1273 were 15.64 [14.15-17.28] and 54.23 [48.13-61.10], respectively, for myocarditis, and 15.78 [13.52-18.42] and 27.03 [21.58-33.87], respectively, for pericarditis. Furthermore, most cases were ≤30 years or male. The period from vaccination to onset was ≤8 days, corresponding to early failure type based on analysis using the Weibull distribution. Outcomes were recovery or remission for most cases; however, they were severe or caused death in some cases. CONCLUSION: In the Japanese population, SARS-CoV-2 mRNA vaccination was significantly associated with the onset of myocarditis/pericarditis. The influencing factors included age of ≤30 years and male. Furthermore, although most adverse events occurred early after vaccination, overall outcomes were good.

Developing large language models to detect adverse drug events in posts on x.

Adverse drug events (ADEs) are one of the major causes of hospital admissions and are associated with increased morbidity and mortality. Post-marketing ADE identification is one of the most important phases of drug safety surveillance. Traditionally, data sources for post-marketing surveillance mainly come from spontaneous reporting system such as the Food and Drug Administration Adverse Event Reporting System (FAERS). Social media data such as posts on X (formerly Twitter) contain rich patient and medication information and could potentially accelerate drug surveillance research. However, ADE information in social media data is usually locked in the text, making it difficult to be employed by traditional statistical approaches. In recent years, large language models (LLMs) have shown promise in many natural language processing tasks. In this study, we developed several LLMs to perform ADE classification on X data. We fine-tuned various LLMs including BERT-base, Bio_ClinicalBERT, RoBERTa, and RoBERTa-large. We also experimented ChatGPT few-shot prompting and ChatGPT fine-tuned on the whole training data. We then evaluated the model performance based on sensitivity, specificity, negative predictive value, positive predictive value, accuracy, F1-measure, and area under the ROC curve. Our results showed that RoBERTa-large achieved the best F1-measure (0.8) among all models followed by ChatGPT fine-tuned model with F1-measure of 0.75. Our feature importance analysis based on 1200 random samples and RoBERTa-Large showed the most important features are as follows: "withdrawals"/"withdrawal", "dry", "dealing", "mouth", and "paralysis". The good model performance and clinically relevant features show the potential of LLMs in augmenting ADE detection for post-marketing drug safety surveillance.

Real-world experience of immune checkpoint inhibitors in patients with solid tumours in the Top End of the Northern Territory, Australia from 2016 to 2021: a retrospective observational cohort study.

BACKGROUND: Use of immune checkpoint inhibitors is growing, but clinical trial data may not apply to Indigenous patients or patients living in remote areas. AIMS: To provide real-world incidence of immune-related adverse events (irAE) in the Top End of the Northern Territory and compare incidence between demographic subgroups. METHODS: This retrospective, observational, cohort study collected data from electronic records of patients living in the Top End with solid organ cancer treated with immunotherapy between January 2016 and December 2021. The primary outcome was cumulative incidence of any-grade and severe irAE. Secondary outcomes were overall survival, treatment duration and reason for treatment discontinuation. RESULTS: Two hundred and twenty-six patients received immunotherapy. Forty-eight (21%) lived in a remote or very remote area, and 36 (16%) were Indigenous. Cumulative incidence of any-grade irAE was 54% (122/226 patients); incidence of severe irAE was 26% (59/226 patients). Rates were similar between Indigenous and non-Indigenous patients of any-grade (42% vs 56%, P = 0.11) and severe (11% vs 18%, P = 0.29) irAE. However, Indigenous patients had shorter treatment duration, more frequently discontinued treatment due to patient preference and appeared to have shorter median overall survival than non-Indigenous patients (17.1 vs 30.4 months; hazard ratio (HR) = 1.5, 95% confidence interval (CI) = 0.92-2.66). There was no difference in mortality between remote and urban patients (median overall survival 27.5 vs 30.2 months; HR = 1.1, 95% CI = 0.7-1.7). CONCLUSIONS: Rates of irAE in our cohort are comparable to those in the published literature. There was no significant difference in any-grade or severe irAE incidence observed between Indigenous and non-Indigenous patients.

Providing information about medication changes upon discharge from a geriatric unit: the community healthcare professionals' point of view.

INTRODUCTION: It is well known that polypharmacy is associated with adverse drug events. Accordingly, specialist geriatric units have to pay particular attention to the appropriateness of prescription and the withdrawal of potentially inappropriate medications. Even though community healthcare professionals are keen to received medication reconciliation results, the literature data show that the quality of communication between the hospital and the community needs to be improved. OBJECTIVE: To assess community healthcare professionals' opinions about the receipt of medication reconciliation results when a patient is discharged from a specialist geriatric unit. METHOD: We performed a qualitative study of general practitioners, community pharmacists and retirement home physicians recruited by phone in the Indre-et-Loire region of France. A grounded theory method was used to analyze interviews in multidisciplinary focus groups. RESULTS: The 17 community healthcare professionals first explained why the receipt of medication reconciliation results was important to them: clarifying the course and outcomes of hospital stays and reducing the lack of dialogue with the hospital, so that the interviewees could provide the care expected of them. The interviewees also described mistrust of the hospital and uncertainty when the modifications were received; these two concepts accentuated each other over time. Lastly, they shared their opinions about the information provided by the hospital, which could improve patient safety and provide leverage for treatment changes but also constituted a burden. PERSPECTIVES: Our participants provided novel feedback and insight, constituting the groundwork for an improved medication reconciliation form that could be evaluated in future research. Exploring hospital-based professionals' points of view might help to determine whether the requested changes in the medication reconciliation form are feasible and might provide a better understanding of community-to-hospital communication.