Panvigilance: integrating biomarkers in clinical trials for systems pharmacovigilance

Abstract

Drug safety and pharmacovigilance are rapidly changing with biomarkers and new technologies such as artificial intelligence. However, we need new ideas and application contexts for integration of biomarkers and emerging technologies in modern pharmacovigilance. A new concept, panvigilance, has been recently introduced for proactive "stress testing" of new drug candidates in panels of patients or healthy volunteers identified by biomarkers, and who are situated in population edges in terms of pharmacokinetic (PK) and/or molecular target interindividual variability. Panvigilance aims to provide upper and lower bound estimates for drug performance under conditions that mimic population edges. Subsequently, it becomes easier to extrapolate pharmacovigilance signals with regard to individuals who reside in between the population edges. In this expert review, we explain that the prefix "pan," meaning everything or all, refers to the three-pronged panvigilance goals to (1) decipher the full population scale variability in medicinal product PKs and molecular target variability, (2) empower forecasting of pharmacovigilance signals within and across populations through knowledge of biomarker variations worldwide, and (3) integration of pharmacovigilance signals across government ministries, civil society organizations, and other stakeholders through, for example, institutional innovation such as centers for panvigilance. We note that panvigilance and pharmacovigilance are complementary, and underscore the added value of panvigilance for global clinical trials. Panvigilance offers a new opportunity for meaningful biomarker application in clinical trials beyond traditional contexts such as personalized medicine. In sum, panvigilance is a systems approach to pharmacovigilance and poised to innovate risk governance in medicinal product development and clinical trials.