A Natural Language Processing Approach Towards Harmonized Communication of Uncertainties Identified During the European Medicine Authorization Process.

Within the European Union, the European Medicines Agency's (EMA's) European Public Assessment Report (EPAR) is an important source of information for healthcare professionals and patients that allows them to understand important risks and uncertainties associated with the use of a medicine. However, the EPAR sections describing such important uncertainties can differ substantially in wording, length, and detail, thereby potentially limiting understanding. In this study, we therefore present a natural language processing approach to cluster sentences extracted from the sections on uncertainties in EPARs of centrally authorized medicines, as a steppingstone to harmonization of text describing uncertainties. We used a BERT language model together with dimensionality reduction (Uniform Manifold Approximation and Projection (UMAP)) and clustering (Density-Based Spatial Clustering of Applications with Noise (DBSCAN)) to identify semantic similarities between sentences. Clusters were labeled according to an overarching topic by reviewing the semantically similar sentences. Each cluster was also characterized according to medicine-related characteristics, such as efficacy or side effects. In total, 1,648 medicines were included in this study. For 573 of these medicines (authorized July 27, 2010 to December 31, 2022), we identified an EPAR that described a complete regulatory dossier and contained sections on uncertainties. Of these, 553 EPARs could be attributed to unique active substance-indication combinations. In these 553 EPARs, we identified 13,105 sentences in sections on uncertainties, leading to 26 clusters of which 2 were labeled as noise. The clusters and associated topics provided in this article can be used by regulators and medicine developers as a steppingstone toward a unified way of communicating uncertainties identified during the EMA process to the broader public.

Variability of in vivo potency assays of whole-cell pertussis, inactivated polio, and meningococcal B vaccines.

For the batch release of vaccines, potency release assays are required. Non-animal in vitro tests have numerous advantages and are preferred; however, several vaccines are still released using in vivo assays. Their major drawback is the inherent variability with its practical implications. We quantified the variability of in vivo potency release assays for whole-cell pertussis, inactivated polio and meningococcal B (MenB) vaccines which showed large CV (Coefficient of Variation) ranging from 34% to 125%. As inherent variability might potentially be attributed to the highly variable immune system between individual animals, we evaluated the antibody titres to four MenB antigens in 344 individual outbred mice. These varied strongly, with more than 100-fold differences in antibody titres in responsive mice. Furthermore, within individual mice there was generally no correlation between the strengths of the responses to the four antigens. A mouse with a very low or no response to one antigen in many cases exhibited a strong response to another antigen. The large differences between individual animals is likely a considerable contributor to the inherent variability of in vivo potency assays. Our data again support the notion that it is preferred to move away from in vivo potency assays for monitoring batch to batch consistency as part of vaccine batch release testing.

Repurposed drug studies on the primary prevention of SARS-CoV-2 infection during the pandemic: systematic review and meta-analysis.

OBJECTIVE: Current evidence on the effectiveness of SARS-CoV-2 prophylaxis is inconclusive. We aimed to systematically evaluate published studies on repurposed drugs for the prevention of laboratory-confirmed SARS-CoV-2 infection and/or COVID-19 among healthy adults. DESIGN: Systematic review. ELIGIBILITY: Quantitative experimental and observational intervention studies that evaluated the effectiveness of repurposed drugs for the primary prevention of SARS-CoV-2 infection and/or COVID-19 disease. DATA SOURCE: PubMed and Embase (1 January 2020-28 September 2022). RISK OF BIAS: Cochrane Risk of Bias 2.0 and Risk of Bias in Non-Randomised Studies of Interventions tools were applied to assess the quality of studies. DATA ANALYSIS: Meta-analyses for each eligible drug were performed if ≥2 similar study designs were available. RESULTS: In all, 65 (25 trials, 40 observational) and 29 publications were eligible for review and meta-analyses, respectively. Most studies pertained to hydroxychloroquine (32), ACE inhibitor (ACEi) or angiotensin receptor blocker (ARB) (11), statin (8), and ivermectin (8). In trials, hydroxychloroquine prophylaxis reduced laboratory-confirmed SARS-CoV-2 infection (risk ratio: 0.82 (95% CI 0.74 to 0.90), I2=48%), a result largely driven by one clinical trial (weight: 60.5%). Such beneficial effects were not observed in observational studies, nor for prognostic clinical outcomes. Ivermectin did not significantly reduce the risk of SARS-CoV-2 infection (RR: 0.35 (95% CI 0.10 to 1.26), I2=96%) and findings for clinical outcomes were inconsistent. Neither ACEi or ARB were beneficial in reducing SARS-CoV-2 infection. Most of the evidence from clinical trials was of moderate quality and of lower quality in observational studies. CONCLUSIONS: Results from our analysis are insufficient to support an evidence-based repurposed drug policy for SARS-CoV-2 prophylaxis because of inconsistency. In the view of scarce supportive evidence on repurposing drugs for COVID-19, alternative strategies such as immunisation of vulnerable people are warranted to prevent the future waves of infection. PROSPERO REGISTRATION NUMBER: CRD42021292797.

Implementing multiple intervention strategies in Dutch public health-related policy networks.

Improving public health requires multiple intervention strategies. Implementing such an intervention mix is supposed to require a multisectoral policy network. As evidence to support this assumption is scarce, we examined under which conditions public health-related policy networks were able to implement an intervention mix. Data were collected (2009-14) from 29 Dutch public health policy networks. Surveys were used to identify the number of policy sectors, participation of actors, level of trust, networking by the project leader, and intervention strategies implemented. Conditions sufficient for an intervention mix (≥3 of 4 non-educational strategies present) were determined in a fuzzy-set qualitative comparative analysis. A multisectoral policy network (≥7 of 14 sectors present) was neither a necessary nor a sufficient condition. In multisectoral networks, additionally required was either the active participation of network actors (≥50% actively involved) or active networking by the project leader (≥monthly contacts with network actors). In policy networks that included few sectors, a high level of trust (positive perceptions of each other's intentions) was needed-in the absence though of any of the other conditions. If the network actors were also actively involved, an extra requirement was active networking by the project leader. We conclude that the multisectoral composition of policy networks can contribute to the implementation of a variety of intervention strategies, but not without additional efforts. However, policy networks that include only few sectors are also able to implement an intervention mix. Here, trust seems to be the most important condition.