At the brink of a paradigm shift in early cancer detection: insights and directions for the modeling community

Abstract

Multicancer early detection (MCED) tests are more than a new class of blood-based tests; they are complex medical innovations representing an integrated diagnostic platform combining molecular and computational technologies. They embody a paradigm shift in how to conceptualize, detect, and manage cancer-carrying the potential to improve outcomes and reduce disparities, yet also the risk of exacerbating them. Real-world evidence remains limited, and existing evidence point to substantial heterogeneity even in standard-of-care screening practices-reflecting patterns of overuse and underuse, fluctuations, and practice variation-despite notable advances in cancer treatment and technology over time. Integrating complex medical innovations into equally complex health systems poses significant challenges, underscoring the urgent need for model-based policy guidance to support their incorporation as a complement to population-based screening within standard-of-care pathways. In this editorial, existing policy-oriented dynamic simulation models on MCED tests are summarized, and insights on how modeling frameworks should evolve in parallel with the growing complexity of medical technologies are offered. Traditional approaches often rest on the implicit assumption that evidence reviews lead linearly to interpretation, policy, and adoption-without accounting for feedback between these stages. Evidence-based guideline formation as a feedback process is revisited as is how modelers develop a suite of flexible models tailored to distinct policy questions. Models that coexist and evolve iteratively as new evidence emerges, thereby capturing the adaptive and evolving nature of the problem itself. Such an approach must transcend disciplinary silos, enabling the integration of diverse data sources and supporting innovative portfolio approaches with methodological flexibility.