Reply: Pharmacokinetic and Pharmacodynamic Modifiers of EF5 Uptake and Binding

REPLY: We thank Drs. Koch and Evans for their insightful comments on our recent article on the hypoxia imaging agent ¹⁸F-EF5 (2-(2-nitro-1H-imidazol-1-yl)-N-(2,2,3,3,3-pentafluoropropyl)-acetamide) (1). In their letter, Koch and Evans have suggested possible reasons for the significant retention of unbound ¹⁸F-EF5 in H460 tumor xenografts in rats compared with that in tumors grown in mice as described in our article. We agree that the differences in drug half-life between rats and mice could be the major factor contributing to higher retention of unbound ¹⁸F-EF5 in rat tumors, especially when the radiotracer is coadministered with its nonradioactive analog for immunohistochemical analysis of bound EF5 adducts on tumor sections. We used 2.5 h for single-time-point imaging (3 h after injection for tumor collection and autoradiography) to enable direct comparison among the tumor models, and based on the literature reports suggesting that 2–3 h is generally an optimal time window for imaging after ¹⁸F-EF5 injection (2–4). For comparison of autoradiography and immunohistochemical images of ¹⁸F-EF5/EF5 binding in tumors, we agree that fixation of tumor sections may remove unbound ¹⁸F activity and yield autoradiography images that may closely match the EF5-immunohistochemical images. In our studies, we used a standard method of comparing images derived from whole tumor sections (untreated) with the hypoxia profile determined from EF5-bound adducts in immunohistochemical images because the purpose of this analysis was to study the distribution (intratumoral) of the radiotracer and corroborate the small-animal PET image findings at the selected time point (2.5 h) (5,6).

In our article, we did not intend to make any suggestions on the metabolism of EF5 or ¹⁸F-EF5, including nonhypoxic metabolism in vivo (7). We think that the observed effect of lower intratumoral contrast in H460 tumors at 2.5 h after injection of ¹⁸F-EF5 in our study could be due to the presence of excess drug or due to slower clearance of the radiotracer from nonhypoxic tumor regions (areas not positive for EF5 adducts) when the radiotracer was coadministered with unlabeled EF5 at a 30 mg/kg dose. We note that this is in line with the suggestion of Koch and Evans that the 10-fold difference in drug concentration between the group of animals receiving radiotracer alone and the group receiving radiotracer coinjected with EF5 (30 mg/kg) could have caused changes in drug half-life and possibly affect the pharmacokinetic loss of unbound drug (¹⁸F-EF5) in H460 tumors in rats. Given the longer half-life of EF5 in rats, imaging at later time points (e.g., >3 h) may allow better clearance of the unbound radiotracer and further improve the contrast between hypoxic and nonhypoxic tumor regions in tumors grown in rats and at the 30 mg/kg dose (100 μM).

With regard to the statement “the authors suggest that uptake of 2-nitroimidazoles such as EF5 selects for tissues that have a partial pressure of oxygen less than 10 mm Hg,” again, we would like to clarify that we used “partial pressure of oxygen < 10 mm Hg” only in the introduction section (as a parenthesis to a sentence) to provide general information that tumor retention of 2-nitroimidazole–based hypoxia tracers typically reflects partial pressure of oxygen values less than 10 mm Hg, as the binding rate of 2-nitroimidazole hypoxia markers increases sharply at partial pressure of oxygen values less than 10 mm Hg (8–10). The full sentence reads as follows: “With the exception of ⁶⁴Cu-diacetyl-bis(N⁴-methylthiosemicarbazone), current small-molecule PET hypoxia tracers consist of a 2-nitroimidazole moiety that forms the basis for their selective uptake in hypoxic tumor cells (partial pressure of oxygen < 10 mm Hg).” In our studies of the 3 tumor models, PC3 tumors displayed a distinctive pattern of hypoxia as indicated by large regions of EF5 binding in immunohistochemical images. In some tumors, the intensity of EF5 binding increased from the center to the outer margin of hypoxic regions. This binding pattern of EF5 in PC3 tumors appears consistent with the macroscopic regions of hypoxia reported by the Koch group in rat 9L gliosarcoma tumors (11).

• © 2015 by the Society of Nuclear Medicine and Molecular Imaging, Inc.