Abstract The landscape of Nuclear Medicine is being fundamentally transformed by the theragnostic concept, which integrates diagnosis with radiopharmaceuticals and targeted radioligand therapy (RLT) for a personalized approach to cancer treatment. While European Directive 2013/59/Euratom mandates dosimetry for organs at risk, current clinical practice still largely relies on standardized, fixed-dose regimens for 177LuLu-DOTATATE and 177LuLu-PSMA-617. However, emerging evidence from key studies (such as LUMEN and LUTADOSE) demonstrates that tumor-absorbed dose is a powerful prognostic factor for progression-free survival, suggesting that a shift toward individualized, dosimetry-guided treatment is essential to maximize the therapeutic index. Despite these findings, significant challenges hinder routine implementation, including the methodological heterogeneity of dosimetric approaches and the logistical burden of multiple post-therapy SPECT/CT acquisitions. Innovative solutions, such as Artificial Intelligence (AI) and Deep Learning (DL), are currently being explored to facilitate pre-therapy dose prediction at the voxel level. Ultimately, the successful transition to precision dosimetry requires unprecedented multidisciplinary collaboration among nuclear medicine physicians, physicists, and regulatory bodies. Establishing unified standards and machine protocols is crucial to ensure that every patient receives optimal benefit from evolving therapeutic approaches, which will further facilitate the translation of these findings into future radiopharmaceutical developments.
Serani et al. (Tue,) studied this question.