FDA Reconsiders Rules Around Radiation Dosimetry for First-in-Human Studies of Investigational PET Radiopharmaceuticals

TO THE EDITOR: Radiopharmaceuticals for PET imaging are regulated as drugs by the U.S. Food and Drug Administration (FDA) through several offices within the Center for Drug Evaluation and Research, including the Office of New Drugs and the Office of Generic Drugs. Within this framework, radiopharmaceuticals follow the same general drug development pathway as pharmaceuticals. Thus, investigational new drug (IND) applications are required for first-in-human (FIH) studies, as well as subsequent phase 2 and 3 clinical trials intended to demonstrate safety and efficacy. Marketing authorization is then obtained through approval of a new drug application. Alternatively, in the case of generic drugs, marketing authorization is obtained via approval of an abbreviated new drug application.

Despite using the same general regulatory framework as nonradioactive drugs, because of the unique characteristics of radiopharmaceuticals (radioactive as well as pharmaceutical doses need considering, short half-lives, microdosing [≤100 μg], low administered activity [AA], final product testing of every batch, etc.), there are some contents of regulatory filings such as INDs that are specific to this class of drugs. For example, dosimetry is a means to quantify the absorbed dose (AD), expressed in Grays or rads, associated with the AA of a PET drug. Historically, 21CFR§312.23(a)(10)(ii) has required that IND applications include sufficient data from animal (or human) studies to allow reasonable estimation of AD to critical organs and effective dose to the patient to justify the lowest AA that can be given to obtain a usable PET image (1).

Dosimetry estimates are typically obtained by conducting biodistribution studies in rodents and using these data, in conjunction with radiation dosimetry software such as OLINDA/EXM (Hermes Medical Solutions) (2), to estimate the radiation AD for different organs within the body from the proposed AA for a given radiopharmaceutical. After approval of an IND, the FDA requires that the first few human studies (e.g., n = 4; 2 men/2 women) consist of whole-body scans to determine actual human dosimetry. However, stakeholders in new PET drug development have expressed that pre-IND animal dosimetry studies are burdensome, contain assumptions or uncertainties given the extrapolation from animal data to human estimates (e.g., rats do not have a gallbladder), and are potentially unnecessary given both the established safety record of PET drugs and the availability of human dosimetry immediately on initiating FIH studies.

In response to this, the FDA convened a meeting of its Medical Imaging Drugs Advisory Committee in August 2023 to reevaluate what dosimetry data are needed to support the initial clinical study in an original IND for certain new PET drugs (the meeting summary is provided at http://jnm.snmjournals.org) (3). Their goals were to conduct a systematic review of clinical dosimetry estimates of PET drugs derived from preclinical and clinical dosimetry studies and to determine AA amounts that could be used safely in FIH studies of new PET drugs without prior animal dosimetry studies. This review included human organ AD and whole-body effective dose estimates extrapolated from animal dosimetry studies and also calculations based on human dosimetry studies, from a total of 322 radiopharmaceuticals, including several approved by FDA. The study found good agreement between animal-derived and human-measured dosimetry.

Analysis of recommended AA values (mean AA from package inserts of approved radiopharmaceuticals) and published dosimetry data suggest that if the planned AA of a FIH study with a new radiopharmaceuticals labeled with ¹⁸F, ¹¹C, ⁶⁸Ga, ⁶⁴Cu, ⁸²Rb, and ¹³N is no more than 299 MBq (8 mCi), 555 MBq (15 mCi), 158 MBq (4.3 mCi), 148 MBq (4 mCi), 1440 MBq (39 mCi), and 552 MBq (15 mCi), respectively, sufficient data are available to justify omitting preclinical dosimetry studies and proceeding directly to required phase 1 human dosimetry studies to establish radiation safety. Although the recommendations in the dossier do not necessarily represent the final position of the agency, several of us have recently used this new approach, filing an IND application for a FIH study with a ⁶⁴Cu-labeled radiopharmaceutical without a preclinical dosimetry package, and were gratified that the FDA is following this approach and approved the study to proceed. Notably, radiopharmaceuticals labeled with longer-lived radionuclides such as ⁸⁹Zr and ¹²⁴I were also considered but are not presently included in the exemption of preclinical dosimetry requirements because of the potential for higher radiation risks. ¹⁵O was also excluded because there are currently no ¹⁵O-labeled PET drugs with FDA approval.

Besides human studies under an IND, many PET radiopharmaceuticals are used under the auspices of a local Radioactive Drug Research Committee (RDRC) (4). Conducting research in humans using radiopharmaceuticals under RDRC approval requires that the research is basic science and not intended to demonstrate safety or efficacy or for diagnostic purposes, that the pharmacologic dose to be administered is known to cause no clinically detectable pharmacologic effect in humans, and that the radiation dose is justified and within specified annual limits (≤3 rem for whole body, active blood-forming organs, lens of the eye, and gonads, and ≤5 rem for other organs for a single dose).

In many instances, when dosimetry has been previously published (5), RDRC approval is straightforward to obtain. However, in the event that a radiopharmaceutical has been used in humans previously, but dosimetry has not been published and cannot be accessed (e.g., legacy dosimetry has been lost, contact cannot be made with original lab, or parties are unwilling or unable to share), then sites are forced to undertake burdensome preclinical estimates of dosimetry themselves, negating benefits offered by 21CFR§361 and the RDRC mechanism in the first place. Given that radiopharmaceuticals used under RDRC must have already been in humans at or above the proposed pharmacologic dose and have shown no adverse events, they are as safe as new radiopharmaceuticals being translated under an IND application. Therefore, we close this letter by respectfully requesting that the FDA also consider adopting this new flexibility in omitting animal dosimetry studies in the case of PET drugs used in basic clinical research under the RDRC mechanism.

• © 2025 by the Society of Nuclear Medicine and Molecular Imaging.

• Received for publication March 11, 2025.
• Accepted for publication March 18, 2025.