Pauline Schiro

Purpose Pharmacovigilance Centres (PVCs) provide independent drug information to healthcare professionals in addition to their pharmacovigilance activities. However, their impact on clinical decision-making has been poorly studied. This study assessed the impact of a French PVC on clinical management in a tertiary university hospital (Toulouse University Hospital). Methods We conducted a retrospective observational study including all drug-related enquiries submitted to the Toulouse PVC during 2024. Impact was evaluated according to three predefined criteria: whether the enquiry was submitted before or after a clinical management decision had been made, concordance between PVC recommendations and subsequent clinical management, and documentation of the advice in the electronic patient record. Based on these criteria, impact was classified as certain, probable, or uncertain for enquiries submitted before clinical management, and as confirmatory (validating an already implemented decision) or uncertain for enquiries submitted after clinical management had already been initiated. Results A total of 128 enquiries were analysed. Most concerned adverse drug reactions (111/128), involved hospitalised patients (100/128), and originated from a broad range of clinical departments, mainly geriatrics (14/128), gastroenterology (11/128), neurology (11/128), infectious diseases (10/128) and internal medicine (10/128). Overall, clinical management was concordant with PVC recommendations in 87% of cases. Among the 92 pre-decision enquiries, PVC impact was classified as certain in 62% and probable in 24% of cases. Among the 36 post-decision enquiries, PVC advice was confirmatory in 44% of cases. The enquiries also generated 50 pharmacovigilance reports, including one potential safety signal. Conclusion PVCs support both clinical decision-making and drug safety surveillance.

Aim: Machine learning (ML) applications in pharmacovigilance remain limited and underexplored. Using data from the French National pharmacovigilance database (FNPV), this proof‐of‐concept study aimed to assess the feasibility of using a ML algorithm—eXtreme Gradient Boosting (XGBoost)—combined with SHapley Additive exPlanations (SHAP) analysis, to classify a well‐defined drug–adverse reaction pair (warfarin‐associated gastrointestinal bleeding) and identify risk factors for reporting this adverse reaction. Methods: We extracted individual case safety reports (ICSRs) involving warfarin from the FNPV. Following data preprocessing, the dataset was randomly split into a training set (75%) and an independent test set (25%). The XGBoost algorithm was trained to classify reports as either gastrointestinal bleeding or non‐gastrointestinal bleeding adverse reactions. Model performance was assessed on the test set using multiple metrics: accuracy, recall, precision, F1‐score and the area under the receiver operating characteristic curve (AUC‐ROC). Feature importance and directional effects were interpreted using SHAP values. Results: A total of 2025 ICSRs involving warfarin were extracted from the FNPV. After preprocessing, 1045 ICSRs were retained. On the test set, the XGBoost model demonstrated an accuracy of 0.68, recall of 0.80, precision of 0.68, F1‐score of 0.74 and an AUC‐ROC of 0.716. SHAP analysis revealed the four most influential features in predicting gastrointestinal bleeding reporting: age, gastrointestinal bleeding risk, antiplatelet and history of renal dysfunction. Conclusion: We demonstrated the feasibility of using an XGBoost algorithm combined with SHAP analysis to pharmacovigilance data for the classification of adverse drug reactions and identify predictive features of reporting, following post hoc model interpretation by pharmacologists.

Background and Objectives Despite increasing global use of cannabis and cannabinoids, comprehensive data on their safety profile remain limited. This study aimed to examine global patterns in pharmacovigilance reports associated with cannabis and cannabinoids, irrespective of regulatory status or context of use. Methods VigiBase^®, the World Health Organisation’s global database of individual case safety reports (ICSRs), was analysed in a retrospective cross-sectional study from 1 January 2020 to 31 December 2024 (data extraction: 19 February 2025). “Cannabis ICSRs” included at least one mention of Cannabis sativa or cannabinoids. Temporal and geographical distributions, reported substances, and suspected adverse drug reactions (ADRs) were assessed. A single ICSR may include multiple substance mentions and several suspected ADRs. Results Among 7,007,689 records, 9983 cannabis ICSRs were identified across 44 countries, comprising 12,538 cannabis or cannabinoid mentions and 27,593 suspected ADRs. The median age was 29 years, with 62.7% males. Most cases were serious (68.5%), including hospitalisation (45.9%) and death (28.6%). The highest reporting rates relative to national ICSR numbers were observed in France, Greece, and the USA. The most frequently reported substances were Cannabis sativa (n = 5951) and cannabidiol (n = 2967). Suspected ADRs were predominantly psychiatric (29.2%) and nervous system disorders (16.8%), with gastrointestinal, respiratory, and cardiac disorders less common (6.4%, 3.6%, and 2.8%). Conclusions This study provides the first comprehensive global overview of suspected ADRs associated with cannabis and cannabinoids. The predominance of serious cases, particularly psychiatric and nervous system reactions, highlights important safety considerations, offering key insights to inform clinical practice and risk assessment worldwide.

Context: Bispecific antibodies (bsAbs) recently reshaped the treatment landscape of multiple myeloma (MM). Rare cases of progressive multifocal leukoencephalopathy (PML) have been reported in MM patients, and more recently in bsAbs-treated patients. Objectives: We aimed to investigate a potential safety signal for PML in patients with MM treated with bsAbs. Methods: We included all patients with MM treated with bsAbs identified in the French prospective registry of JC polyomavirus infections of the central nervous system and/or from the French pharmacovigilance database (FPVD). We also performed a systematic review. We then assessed whether a disproportionality signal for PML reporting exists, comparing bsAbs and other MM therapies in VigiBase®, the world pharmacovigilance database. Results: In the registry-FPVD cohort, 14 patients were included; all had at least triple-refractory MM, with a median disease duration of 13 years (IQR, 10-14) and were exposed to BCMA-targeting bsAbs (median time to PML diagnosis: 22 months, IQR, 12-24); 10 (71%) were in complete response. Main clinical manifestations included motor deficit, cognitive impairment and visual symptoms (8/14 patients each, 57%). Twelve (86%) had died from PML by month 12. Similar features and outcomes were observed in the systematic review (n=14 cases). PML reporting was significantly higher with bsAbs than with other MM therapies (Reporting Odds Ratio 17.7; 95% CI, 12.1-26.1). Interpretation: PML may represent an emerging complication in patients with MM receiving bsAbs. Vigilant monitoring and rapid assessment of new neurological deficits are critical in this population.