SHIELD: Semantic Heterogeneity Integrated Embedding for Latent Discovery in Clinical Trial Safety Signals

TL;DR

SHIELD is a new method that combines statistical disproportionality analysis with semantic clustering of adverse event terms, enhanced by language models, to improve safety signal detection and interpretation in clinical trials.

Contribution

It introduces an integrated framework that merges statistical disproportionality measures with semantic clustering and language model annotations for better safety signal analysis.

Findings

Successfully recovers known safety signals

Generates interpretable safety profiles

Provides a network and hierarchical visualization of safety data

Abstract

We present SHIELD, a novel methodology for automated and integrated safety signal detection in clinical trials. SHIELD combines disproportionality analysis with semantic clustering of adverse event (AE) terms applied to MedDRA term embeddings. For each AE, the pipeline computes an information-theoretic disproportionality measure (Information Component) with effect size derived via empirical Bayesian shrinkage. A utility matrix is constructed by weighting semantic term-term similarities by signal magnitude, followed by spectral embedding and clustering to identify groups of related AEs. Resulting clusters are annotated with syndrome-level summary labels using large language models, yielding a coherent, data-driven representation of treatment-associated safety profiles in the form of a network graph and hierarchical tree. We implement the SHIELD framework in the context of a single-arm incidence summary, to compare two treatment arms or for the detection of any treatment effect in a multi-arm trial. We illustrate its ability to recover known safety signals and generate interpretable, cluster-based summaries in a real clinical trial example. This work bridges statistical signal detection with modern natural language processing to enhance safety assessment and causal interpretation in clinical trials.