Nuclear theranostics at a crossroads:from tracer optimization to biologically informed therapy

Abstract

Nuclear theranostics has transitioned from proof-of-concept to clinical practice with the regulatory approval of SSTR- and PSMA-targeted radioligand therapies. However, therapeutic outcomes remain constrained by dose-limiting toxicity, heterogeneous target expression and empirically defined treatment regimens, indicating that the field is approaching the limits of incremental tracer optimization. This review examines the current crossroads of theranostic development from a translational perspective. Molecular design strategies, including plasma protein binding, metabolic stabilization, covalent target ligation, multimerization and alternative radionuclides, are discussed and critically evaluated against the central criterion of whether increased tumor uptake translates into genuine improvement of the therapeutic index. The emerging understanding of RLT radiobiology, its interaction with the tumor microenvironment, and the resulting rationale for combination strategies with immune checkpoint inhibitors, PARP inhibitors and radiosensitizing agents are then addressed. A critical analysis of preclinical models, dosimetry as the primary translational currency, and publication culture highlights persistent translational bottlenecks. Finally, the diagnostic component of theranostics, its applications ranging from target validation and therapy guidance to functional tumor phenotyping, is identified as systematically undervalued relative to the therapeutic side, and its further development is argued to be essential for advancing the field beyond its current limitations.