Hyperpolarized ¹⁵N caffeine, a potential probe of liver function and perfusion

Purpose: To demonstrate hyperpolarization of ¹⁵N-caffeine and report exploratory findings as a potential probe of liver function and perfusion. Methods: An amorphous formulation of [1,3-¹⁵N₂]caffeine was developed for hyperpolarization via dissolution dynamic nuclear polarization. Polarizer hardware was augmented to support monitoring of solid-state ¹⁵N MR signals during the buildup of hyperpolarization. Liquid state hyperpolarized ¹⁵N MR signals were obtained in a preclinical 3T magnet by interfacing an external spectrometer console with home-built RF surface coils. ¹⁵N signal decay constants were estimated in H₂O and in vivo in liver and brain regions of rats at 3 T. Decays were also measured at 9.4 T to assess the effect of B₀, and in the presence of albumin to assess the impact of protein binding. Results: Polarization levels of 3.5% and aqueous T₁ relaxation times of nearly 200 s were attained for both N1 and N3 positions at 3 T. Shorter apparent decay constants were observed in vivo, ranging from 25 s to 43 s, with modest extensions possible by exploiting competitive binding of iophenoxate with plasma albumin. Downstream products of caffeine could not be detected on in vivo ¹⁵N-MR spectra of the liver region, even with metabolic stimulation by (Formula presented.) -naphthoflavone treatment. Considering the high perfusion rate of brain, persistence of caffeine signal in this region is consistent with potential value as a perfusion imaging agent. Conclusion: These results establish the feasibility of hyperpolarization of hyperpolarized ¹⁵N-caffeine, but further work is necessary to establish the role of this new agent to probe liver metabolism and perfusion.