Utility of Pre-clinical Dosimetry Evaluation in Estimating Human Absorbed Doses for First-in-Human Studies with 11C- and 18F-labeled Radiopharmaceuticals: an FDA Perspective

1561 Background: For radioactive drugs, 21 CFR 312.23(a)(10)(ii) requires that IND submissions include sufficient data from animal or human studies to allow a reasonable estimation of absorbed-dose (AD) to the whole-body and critical organs upon administration to a human subject. For first-in-human (FIH) studies, IND submissions include AD estimates for humans that are often extrapolated from pre-clinical (animal) biodistribution data. The extrapolation techniques make assumptions on the differences in metabolism, anatomy and biodistribution between animals and humans. However, calculation of human doses from animals is imprecise, and this presents a regulatory challenge. Two letters to the editor from Zannoti-Fregonara et al. have previously suggested pathways to assess radiation safety of 11C and 18F tracers recommending that an upper limit of injected activity for FIH studies can be derived from the max reported effective dose (ED) in humans rather than based on pre-clinical dosimetry. This study aims to gain a better understanding of the degree of uncertainty in calculations of human-organ AD and ED from animal data for 11C and 18F imaging agents. Methods: Two separate literature searches were conducted using PubMed, Embase, SciFinder and WoS for 11C and 18F radiopharmaceuticals developed since the 1990s. Articles were selected based on available human-organ dose estimates from both pre-clinical and clinical studies and analyzed according to MIRD or related methodology. FDA internal platforms were also used to retrieve information from new drug applications on 11C and 18F tracers. Human-organ AD and ED extrapolated from animal data and calculated from measurements in humans were collected from a selected number of articles. Ratios between human dose estimates derived from animals of different species and from measurements in humans were calculated. Results: The literature searches resulted in 114 (11C) and 186 (18F) articles. Dosimetry data were collected for thirteen 11C and three 18F radiotracers. The 11C radiotracers in clinical use have relatively low ED compared to 18F (range, 3.0-6.9 µSv/MBq for selected tracers), except for 11C-WAY100635 (14.1 µSv/MBq). Mean ED from clinical studies of the selected tracers was 4.6±1.1 µSv/MBq (11C) and 22±0.2 µSv/MBq (18F). The mean ratio in human ED estimate from animals to human-measured data was 1.0±0.5 (11C) and 1.0±0.2 (18F). Organs that exhibited the highest AD differ between radiotracers. The mean ratio between animal-derived and human-measured AD for human organs was: 1.1±0.6 (mouse), 1.0±0.2 (rat), 0.9±0.6 (monkey) and 1.1±0.2 (pig) for 11C; 1.0±0.5 (mouse) and 1.1±0.6 (monkey) for 18F, with AD to individual organs being both over- and under-estimated. Conclusions: In general, the ED measured in humans is quite constant across the different tracers of 11C and those of 18F, suggesting that a mean ED value from published human studies can be derived for each radioisotope. The generally modest radiation dose profile of C-11 tracers is likely due to the shorter half-life of 11C that is the major determinant of ED. The degree of over- and under-estimation in the human ED and AD values to individual human organs derived from animal biodistribution data is being analyzed to evaluate the utility of pre-clinical dosimetry for FIH trials involving short-lived PET imaging agents.[1] Zanotti-Fregonara, P., et al. (2012). EJNMMI, 39(3), 544-547[2] Zanotti-Fregonara, P., et al. (2013). EJNMMI, 40(11), 1781-1783