Implementation of preemptive pharmacogenetic testing: progress, puzzles and priorities from an implementation science perspective.

An implementation science reading of the barriers, facilitators and evidence gaps that separate the clinical utility of pre-emptive genotyping from its integration into routine practice.

Pre-emptive pharmacogenetic testing, which uses genetic variation to predict drug response and toxicity, promises to improve drug safety and efficacy, yet its integration into routine practice still faces a substantial translational gap. This review examines it through an implementation science lens, moving beyond clinical utility to synthesise the implementation landscape and identify systemic integration challenges; over the past decade, a growing body of pragmatic implementation studies has systematically catalogued barriers and facilitators across multiple levels. Adoption metrics are now relatively well documented, but evidence on fidelity and long-term sustainability remains scarce, and most cost-effectiveness data derive from modelling rather than implementation trials. The authors consider that synergistic strategies across several domains are required: robust evidence from large-scale pragmatic trials, health informatics infrastructure, multidisciplinary services, standardised clinical workflows, comprehensive education for clinicians and patients, and active participation in collaborative networks. An equity gap persists, nearly all published pre-emptive implementation programmes originating from high-income countries, hence the need for context-adapted approaches in low- and middle-income settings, for which emerging implementation frameworks might guide the embedding of equity into implementation design.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The most useful finding is twofold: adoption can be measured, but implementation fidelity and sustainability remain poorly documented, and the cost-effectiveness argument rests mostly on models rather than on implementation trials. The priorities set out (pragmatic trials, health informatics, standardised workflows, education) are general recommendations, with no threshold or quantified course of action, and the abstract states neither the number of studies examined nor the search method, so the completeness of the synthesis and the risk of selection bias cannot be judged. The geographical imbalance, with nearly all programmes coming from high-income countries, further limits how far these lessons transfer to other health systems.

Analysis by Dr Thibaut Benquey