Since the inception of radionuclide therapies by John Lawrence and Saul Hertz over 80 y ago, little has changed regarding dosing strategies. We tend to use empirical amounts of therapeutic radionuclide administered at fixed intervals or as a single treatment, with little regard to an individual patient’s disease characteristics and uptake. The reasons given for this radiotherapeutic empiricism can often be traced to radioiodine therapy for thyroid malignancy, where a one-size-fits-all approach has been used to great success for many decades. The appeal to industry of producing therapeutic treatments in fixed, standardized amounts may have also led to the popularity of this approach.
However, newer therapeutic radiopharmaceuticals do not demonstrate such near-ideal targeting specificity as radioiodine, meaning that maximization of the therapeutic index to achieve curative effect balanced against potential toxicity may be considered to optimize clinical outcomes. Common examples in our field include the somatostatin receptor subtype 2 in neuroendocrine neoplasia or meningiomas and prostate-specific membrane antigen in adenocarcinoma of the prostate. When peptides directed at these targets are radiolabeled with an imaging agent, such as the PET radiometal 68Ga, a whole-body imaging assessment of the extent and avidity of the disease is obtained. The replacement of 68Ga with the β-emitting radiometal 177Lu, for example, provides a potent therapeutic agent that produces excellent outcomes in many patients, causes few side effects, and can be delivered as an outpatient treatment. This theranostic paradigm with neuroendocrine tumor and prostate cancer therapies is now widely used but has generally been implemented without departing from radiotherapeutic empiricism—that is, treatments delivered as standard amounts of radioactivity at fixed intervals (often 6–8 wk). This approach results in most patients being underdosed, and some being overdosed, compared with what could be safely achieved through personalized treatments (1–3).
Treatment personalization and optimization using radiation dosimetry have been proposed as ways of improving patient outcomes. However, progress has been impeded because of technical challenges, the lack of prospective clinical trial data demonstrating efficacy, and reimbursement and sustainability issues. Initiatives such as the Society of Nuclear Medicine and Molecular Imaging (SNMMI) Dosimetry Challenge (4) have sought to address some of these technical barriers, whereas the SNMMI Dosimetry Task Force has proposed pathways to enable reimbursement (5), laying important groundwork for broader adoption. Furthermore, the lack of reproducible and traceable measurements of administered activity and quantitative image measurements, compounded by a wide range of nonstandardized dosimetry practices, has led to the inability to compare data across drug development and therapeutic platforms in some cases, even within the paradigm of empirical fixed-activity treatments. These obstacles must be tackled sequentially and systematically. The proposed project in this commentary intends to address and resolve the major technical issues, clearing the road for the collection of quality dosimetry data to support prospective clinical trials.
IMPROVING TREATMENT PLANNING AND OUTCOMES
The Foundation for the National Institutes of Health (FNIH) Biomarkers Consortium has recently launched the Precision Dosimetry Imaging Biomarker (PDIB) project, an international consortium-based initiative focused on solving the technical issues that currently limit the effective use of dosimetry-informed therapeutic radiation dosimetry. The PDIB project kicked off in April 2025—entirely funded by industry partners (Table 1) who see standardization and harmonization as both critical to the effective use of radiation dosimetry and laying firmly in the precompetitive space—and aims to enable reliable and reproducible standardized dosimetry of targeted radionuclide therapies. This work is expected to support the definition of absorbed dose toxicity limits for therapeutic radiopharmaceuticals as well as enable more accurate assessments of tumor absorbed dose. More information can be found at https://fnih.org/our-programs/biomarkers-consortium/precision-dosimetry-imaging-biomarker/.
TABLE 1.
FNIH PDIB Project Partners and Funders
| Funding type | Partner/funder name |
|---|---|
| Public sector partners | Australian Nuclear Science & Technology Organisation, Belgian Nuclear Research Centre, Conseil National de Recherches Canada, Food and Drug Administration, National Cancer Institute, National Physical Laboratory, National Institute of Standards and Technology |
| Academic partners | BC Cancer, Oslo University Hospital, Institute of Nuclear Medicine, University College (London), Peter MacCallum Cancer Centre, University of Alabama at Birmingham, University of Iowa, University of Sydney |
| Medical societies | Australia and New Zealand Society of Nuclear Medicine, Society for Nuclear Medicine and Molecular Imaging, American Society for Radiation Oncology |
| Radiopharmaceutical innovators partners | Eli Lilly and Company, Fusion Pharmaceuticals (a member of the AstraZeneca Group), Novartis US, RayzeBio (a Bristol Meyers Squibb Company) |
| Biotechnology companies | ITM, Lantheus, NorthStar, Perspective Therapeutics, Telix Pharmaceuticals |
| Start-ups, software and device developers | Capintec (a Mirion Medical Company), Hermes Medical Solutions, MIM Software (a GE HealthCare Company), Rapid, RayMed, VOXIMETRY |
The PDIB project consists of 2 distinct phases. The first phase is a 3-y, $5 M project consisting of 3 subprojects, each aimed at providing workable technical solutions to primary challenges facing the field (Fig. 1). The nascent second phase is anticipated to implement the infrastructure and standardization methodologies from the first phase into 2 radiopharmaceutical-agnostic clinical trials specifically targeting better definition of kidney and bone marrow toxicity in relation to absorbed dose.
FIGURE 1.
Three subprojects that comprise phase 1 of PDIB project are shown.
Phase 1
Subproject 1: Establishment of Secondary Standards Calibration Laboratories (SSCLs)
The first subproject consists of setting up an international network of SSCLs to act as a resource for accurate radioactivity measurements of therapeutic radionuclides. As a field, and in stark contrast to norms in radiation oncology, there is generally an absence of routine accuracy verification for radionuclide measurement devices, leading to systemic, concerning, and long-standing calibration errors (6,7). In subproject 1, 4 SSCLs will be established worldwide. They will work hand in hand with 2 well-recognized metrology institutes (National Institute of Standards and Technology in the United States and National Physical Laboratory in the United Kingdom) to develop robust procedures for traceable radioactivity measurements using high-resolution γ-spectrometry. These SSCLs will generate rapid turnaround (<1 wk) and affordable radioactivity measurements (or radioactivity standards) of common theranostic radionuclides (177Lu, 131I, 225Ac, 111In, 203Pb, 212Pb) with an expected accuracy within ±2% in clinically relevant geometries and with associated uncertainties. The aim is to support clinical trials and routine clinical practice by improving the accuracy of activity measurement and establishing traceability to internationally recognized standards. Similar projects in radionuclide metrology have been initiated by bodies such as the European Federation of Organisations for Medical Physics, the National Physical Laboratory, and the National Institute of Standards and Technology to improve the accuracy of radionuclide calibration for therapeutic applications. Members of the PDIB working groups are engaged in these activities as well.
Subproject 2: Standardization of SPECT/CT Sensitivity Calibration
The second subproject targets the sensitivity calibration of SPECT/CT systems for the radionuclides that will be optimized in subproject 1 (177Lu, 131I, 225Ac, 111In, 203Pb, 212Pb), which provide images defined in terms of radioactivity concentration. SPECT/CT is the imaging workhorse for most therapeutic radionuclides, with posttreatment biodistribution of the radiopharmaceutical measured at several time points to estimate the radiation dosimetry to normal tissues and tumors. This requires an accurate, quantitative SPECT platform. The clinical trial arms of 3 of the main global nuclear medicine organizations (Australasian Radiopharmaceutical Trials Network of the Australian and New Zealand Society of Nuclear Medicine, European Association of Nuclear Medicine Research Ltd., and the SNMMI Clinical Trials Network) have started SPECT/CT calibration programs with their own phantoms and methodologies, independent of each other. Each approach is slightly different but conceptually similar. We therefore need to investigate whether these methods generate equivalent calibration results (in relation to the image conversion factor or sensitivity factor) for the range of current SPECT/CT systems across the range of therapeutic radionuclides being optimized. Standardized best practice phantom preparation and calibration methodologies for quantitative calibration and validation for SPECT/CT systems that can be used uniformly in clinical trials and clinical practice will be developed. Imaging sites from Europe, Australia, and North America will contribute new and well-curated data using metrologically traceable quantities of radioactivity provided by the SSCLs from subproject 1. Additionally, to better understand the quantitative performance characteristics of each model of SPECT/CT system as a function of object size for more complex geometries (to aid in understanding the accuracy of tumor absorbed dose), a series of phantom-based experiments across the range of SPECT/CT systems are planned.
Subproject 3: Standardizing Dosimetry Calculation Workflows
The third subproject will generate standardization of dosimetry calculation workflows and will establish data-driven uncertainty or reproducibility estimates when using these refined methods. Recent results from the SNMMI 177Lu Dosimetry Challenge have demonstrated that the current lack of standardized approaches is leading to an unacceptably high variability of 20% for organs and of greater than 25% for tumors between users on multiple-time-point SPECT/CT dosimetry, making it difficult to pool data from different trials to generate meaningful radiopharmaceutical absorbed dose–effect relationships (4,8,9). An international team of 5 expert medical physicists have been tasked with developing and testing standardized dosimetry approaches using a range of commercial and publicly available dosimetry software to measure and report variability (between operators as well as for repeat analysis by the same operator) of the calculated organ at risk and tumor absorbed dose when following standardized best practices. In addition to elucidating the variability of dosimetry results when following best practices, the subproject team intends to publish the detailed standard operating procedures and implementation guides for common dosimetry software packages. Finally, the team intends to make publicly available a series of multiple-time-point SPECT/CT datasets alongside dosimetry results using the techniques agreed upon by the medical physicists. These products, alongside detailed dosimetry methods, should allow for the training of new physicists and software operators, credentialing of dosimetry service providers, and validation of new dosimetry tools as they become available.
Each of these 3 initiatives in the first 3 y aims to eliminate the reproducibility and accuracy crisis that exists today in the practice of radionuclide therapy dosimetry. If these subprojects deliver as expected, a second phase is planned to run radiopharmaceutical-agnostic paired prospective clinical trials for patients with neuroendocrine tumors and of prostate cancer treatments to examine kidney, bone marrow, and tumor radiation dosimetry estimates and patient outcomes.
International in scope and input, the PDIB project aims to both improve and standardize the use of radiation dosimetry estimates in radionuclide therapies to enable broader adoption of dosimetry-based personalized radionuclide therapies leading to better outcomes for our patients.
DISCLOSURE
Support for the FNIH Biomarkers Consortium PDIB project was provided by Eli Lilly and Company; Fusion Pharmaceuticals, a member of the AstraZeneca Group; Novartis US; RayzeBio, a Bristol Meyers Squibb Company; ITM; Lantheus; Rapid; RayMed; and Telix Pharmaceuticals. In-kind donations of anonymized imaging data, phantoms, software, and radionuclides to support the project were provided to the FNIH by Capintec (a Mirion Medical Company); Hermes Medical Solutions; ITM; MIM Software (a GE HealthCare Company); NorthStar; Perspective Therapeutics; and VOXIMETRY. No other potential conflict of interest relevant to this article was reported.
REFERENCES
- 1. Eberlein U, Cremonesi M, Lassmann M. Individualized dosimetry for theranostics: necessary, nice to have, or counterproductive? J Nucl Med. 2017;58(suppl 2):97S–103S. [DOI] [PubMed] [Google Scholar]
- 2. Sandström M, Freedman N, Fröss-Baron K, Kahn T, Sundin A. Kidney dosimetry in 777 patients during 177Lu-DOTATATE therapy: aspects on extrapolations and measurement time points. EJNMMI Phys. 2020;7:73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Del Prete M, Buteau FA, Beauregard JM. Personalized 177Lu-octreotate peptide receptor radionuclide therapy of neuroendocrine tumours: a simulation study. Eur J Nucl Med Mol Imaging. 2017;44:1490–1500. [DOI] [PubMed] [Google Scholar]
- 4. Uribe C, Peterson A, Van B, et al. An international study of factors affecting variability of dosimetry calculations, part 1: design and early results of the SNMMI Dosimetry Challenge. J Nucl Med. 2021;62(suppl 3):36S–47S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Graves SA, Bageac A, Crowley JR, Merlino DAM. Reimbursement approaches for radiopharmaceutical dosimetry: current status and future opportunities. J Nucl Med. 2021;62(suppl 3):48S–59S. [DOI] [PubMed] [Google Scholar]
- 6. Bailey DL, Hofman MS, Forwood NJ, et al.; the ProPSMA Trial Investigators. Accuracy of dose calibrators for gallium-68 PET imaging: unexpected findings in a multi-centre clinical pre-trial assessment. J Nucl Med. 2018;59:636–638. [DOI] [PubMed] [Google Scholar]
- 7. Graves SA, Martin M, Tiwari A, Merrick M, Sunderland J. SIR-Spheres activity measurements reveal systematic miscalibration. J Nucl Med. 2022;63:1131–1135. [DOI] [PubMed] [Google Scholar]
- 8. Brosch-Lenz J, Ke S, Wang H, et al. An international study of factors affecting variability of dosimetry calculations, part 2: overall variabilities in absorbed dose. J Nucl Med. 2023;64:1109–1116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Brosch-Lenz J, Kurkowska S, Frey E, Dewaraja YK, Sunderland J, Uribe C. An international study of factors affecting variability of dosimetry calculations, part 3: contribution from calculating absorbed dose from time-integrated activity. J Nucl Med. 2024;65:1166–1172. [DOI] [PMC free article] [PubMed] [Google Scholar]