Researchers have modelled how the choice of radioactive isotope and the precise subcellular location where a radiopharmaceutical binds jointly shape the radiation dose delivered to cancer-cell nuclei, in a computational study published in EJNMMI Physics. The work, led by Arvin Haj-Mirzaian, Victor V. Onecha, Umar Mahmood, Alejandro Bertolet and Pedram Heidari, is presented as an open-access original research article and was published on 9 October 2026.
The study addresses a central question in radiopharmaceutical therapy dosimetry: the dose a cell nucleus receives depends not only on the emission physics of the radionuclide itself but also on the spatial distribution of the radioactive activity. To disentangle these two factors, the team focused on a single human breast-cancer region of interest and asked how radionuclide identity and protein-defined subcellular localization influenced nuclear absorbed-dose distributions in that tissue.
Methodologically, the researchers took multiplex immunofluorescence images from one breast-cancer region of interest and segmented them into 41,454 individual cells. Protein-expression maps corresponding to five compartments—nuclear, nuclear-membrane, cytoplasmic, plasma-membrane and extracellular-matrix locations—were then used as source maps in Monte Carlo simulations performed with the TOPAS/Geant4 toolkit. Four radionuclides were compared: terbium-161, lutetium-177, lead-212 and actinium-225, spanning beta emitters and alpha-emitting chains.
The simulated endpoints were deliberately varied. The team computed a similarity dose, denoted Dsim,99, along with simulated surrogate IC50 and IC90 threshold-crossing doses over hypothetical per-cell thresholds ranging from 0.2 to 5.0 Gy. They also calculated the percentage of cancer nuclei exceeding a submitted threshold of 0.5 Gy for the beta emitters terbium-161 and lutetium-177, and a lower 0.1-Gy threshold for the alpha emitters lead-212 and actinium-225. For the alpha-emitter IC analysis, the submitted whole-chain factor of 5 was applied, while coverage calculations used unmultiplied scored dose with the factor-five-equivalent 0.1-Gy threshold described in the methods.
The numerical results highlight how strongly the emitter choice matters. For nuclear-membrane localization, the Dsim,99 value was 0.030 Gy for terbium-161, 0.074 Gy for lutetium-177, 0.412 Gy for lead-212 and 0.580 Gy for actinium-225. Across all five localization compartments, Dsim,99 ranged from 0.030 to 0.074 Gy for terbium-161, from 0.063 to 0.089 Gy for lutetium-177, from 0.412 to 1.023 Gy for lead-212, and from 0.580 to 1.116 Gy for actinium-225.
Localization also shaped the dose-response behaviour. Extracellular-matrix locations generally produced the largest simulated surrogate IC slopes, although the authors note that the magnitude of this effect and its dependence on the endpoint varied by isotope. In addition, some mixed-chain IC90 fits for extracellular-matrix compartments were numerically unstable, a caveat the researchers report openly rather than smoothing over.
When the radioactive source was targeted at the nucleus itself, the descriptive fractions of cancer nuclei receiving doses above the relevant thresholds were 20.25 percent for terbium-161, 15.76 percent for lutetium-177, 18.32 percent for lead-212 and 18.82 percent for actinium-225. As with the other endpoints, the beta and Auger coverage figures used the 0.5-Gy threshold, whereas the alpha-emitter panels used the 0.1-Gy threshold, so the four percentages are not directly interchangeable.
The study’s conclusion is deliberately restrained. The authors state that radionuclide identity and the modelled subcellular localization jointly influenced the submitted nuclear-dose summaries, but that no universal optimal compartment was established across endpoints. In other words, the best binding location depends on which isotope is used and which dosimetric endpoint is being considered, and no single strategy emerged as uniformly superior.
The limitations are equally explicit. These are single-region-of-interest computational results, described by the authors as hypothesis-generating. Before any clinical application, the findings would require particle-resolved transport validation, pharmacokinetic data, radiobiological assessment, normal-tissue evaluation and replication across multiple patients. The article is currently shared as an accepted manuscript that is citable and carries a permanent DOI but remains subject to further editorial edits before the final version of record replaces it.
Within the limits of one breast-cancer region and a simulation-based design, the answer reported here is that it does—but the size and direction of that effect depend on the isotope, and the search for an optimal coupling strategy remains open.
Subject of Research: Sub-tissue dosimetry modelling of how radionuclide choice and subcellular localization affect cancer-cell nuclear absorbed dose in radiopharmaceutical therapy
Article Title: Coupling target and radioactive emitter matters: a sub-tissue dosimetric study to identify optimal strategies
Article References: Haj-Mirzaian, A., Onecha, V. V., Mahmood, U., Bertolet, A., & Heidari, P. (2026). Coupling target and radioactive emitter matters: a sub-tissue dosimetric study to identify optimal strategies. EJNMMI Physics. https://doi.org/10.1186/s40658-026-00952-7
Image Credits: AI Generated
DOI: 10.1186/s40658-026-00952-7
Keywords: radiopharmaceutical therapy, dosimetry, TOPAS, Geant4, terbium-161, lutetium-177, lead-212, actinium-225, subcellular localization, breast cancer, Monte Carlo simulation, nuclear absorbed dose
News Source: Nathaniel Bowman. (October 28, 2026). Where a radioactive emitter sits inside a tumour cell changes simulated dose. Scienmag.
