Trial architecture in contemporary cardio-oncology: Why parallel-group randomized controlled trials may not fit every cardiotoxicity phenotype

Abstract

Cardio-oncology has evolved rapidly with immune checkpoint inhibitors, chimeric antigen receptor T-cell therapies, antibody-drug conjugates, bispecific antibodies, and targeted kinase inhibitors, yet cardiovascular prevention strategies remain largely informed by evidence generated during the anthracycline era. This critical narrative review examines why cardioprotection trials have failed to keep pace with these evolving toxicity mechanisms, analyzing trial populations, endpoint selection, biomarker strategies, imaging approaches, and overall trial design across traditional and newer cancer therapies, rather than summarizing therapeutic efficacy. The evidence shows a persistent mismatch between contemporary toxicity phenotypes and existing trial design: Most studies continue to focus on anthracycline-associated left ventricular dysfunction and surrogate imaging endpoints, while immune-mediated myocarditis, cytokine release syndrome-related cardiomyopathy, vascular toxicities, arrhythmias, and other emerging phenotypes remain inadequately represented, compounded by limited biological stratification, one-size-fits-all prevention strategies, heterogeneous outcome definitions, and insufficient mechanism-specific monitoring. Progress in cardio-oncology prevention is constrained not only by the availability of candidate cardioprotective interventions but by the near-uniform persistence of a single trial architecture across mechanistically distinct toxicities. Future studies should adopt phenotype-driven, mechanism-informed, adaptive trial designs using clinically meaningful outcomes that reflect the diversity of contemporary cardiovascular toxicities.

Keywords: Cardio-oncology; Cardiotoxicity; Cardioprotection; Clinical trial design; Adaptive clinical trials; Biomarker-enriched trial design; Master protocols; Registry-based randomized trial; Immune checkpoint inhibitor myocarditis; Cancer therapy-related cardiovascular toxicity

Core Tip: This minireview is concerned with how cardioprotection trials are designed, not with the pharmacological efficacy of any particular cardioprotective agent. These trials have relied almost exclusively on the conventional parallel-group randomized controlled trial, regardless of which cardiotoxicity phenotype is studied. This minireview maps six contemporary phenotypes – from anthracycline-related left ventricular dysfunction to radiation-induced cardiovascular disease – against established trial-design principles (biomarker-enriched, adaptive, master-protocol, and registry-embedded designs) already used elsewhere in medicine. Only anthracycline-related left ventricular dysfunction is currently supported by conducted trials using an alternative architecture; the remaining five phenotypes require prospective evaluation of untested conceptual directions.