Metabolic imaging with hyperpolarized [1-13 C]pyruvate offers the unique opportunity for a minimally invasive detection of cellular metabolism. Efficient and robust acquisition and reconstruction techniques are required for capturing the wealth of information present for the limited duration of the hyperpolarized state (~1 min). In this study, the Dixon/IDEAL type of water-fat separation is expanded toward spectroscopic imaging of [1- 13C]pyruvate and its down-stream metabolites. For this purpose, the spectral-spatial encoding is based on single-shot spiral image encoding and echo-time shifting in between excitations for the chemical-shift encoding. In addition, also a free-induction decay spectrum is acquired and the obtained chemical-shift prior knowledge is efficiently used in the reconstruction. The spectral-spatial reconstruction problem is found to efficiently separate into a chemical-shift inversion followed by a spatial reconstruction. The method is successfully demonstrated for dynamic, multislice [1- Within the past decade, hyperpolarization for in vivo MR imaging and spectroscopy has expanded from gaseous imaging agents toward liquid ones. Among others, dynamic nuclear polarization (DNP) in the amorphous state followed by rapid dissolution has demonstrated as a versatile method to increase the polarization of liquidstate MR imaging agents by more than four orders of magnitude as compared with thermal polarization levels (1,2). The lifetime of the obtained hyperpolarized liquid MR imaging agent is determined by the spin-lattice T 1 relaxation time, which is dependent on the nuclei and its relative position within the molecule considered.Among possible choices, in particular, labeled [1-13 C]pyruvate (noted subsequently as Pyr) has emerged as a promising marker for metabolic MR imaging due to its endogenous character and favorable properties in terms of hyperpolarization, i.e., high polarization levels of up to $50% and long T 1 relaxation times on the order of $30 s in vivo and $65 s ex vivo (2,3). In vivo, hyperpolarized Pyr gets converted into other MR detectable down-stream metabolites, namely lactate (Lac), alanine (Ala), pyruvate-hydrate (PyrH), and bicarbonate (BiC). The metabolites can be distinguished based on their spectral fingerprint, which consists of approximately singlet peaks at well-separated chemical-shift (CS) frequencies. Information about metabolic pathways and the corresponding turn-over ratios can be derived from the time evolution of the individual metabolite concentrations (4,5).Accordingly, hyperpolarized Pyr provides a wealth of detailed metabolic information during limited time duration, which is on the order of the T 1 relaxation time. Various sequences with different trade-offs between temporal and spatial resolution have been suggested for capturing the available information (6,7): small flip-angle (FA), short repetition time (TR), and slice-selective freeinduction decay (FID) acquisition of full spectra provide high temporal resolution with volume selection only along the...

No abstract

Hyperpolarization of [1-13C]pyruvate in solution allows real-time measurement of uptake and metabolism using MR spectroscopic methods. After injection and perfusion, pyruvate is taken up by the cells and enzymatically metabolized into downstream metabolites such as lactate, alanine, and bicarbonate. In this work, we present comprehensive methods for the quantification and interpretation of hyperpolarized 13C metabolite signals. First, a time-domain spectral fitting method is described for the decomposition of FID signals into their metabolic constituents. For this purpose, the required chemical shift frequencies are automatically estimated using a matching pursuit algorithm. Second, a time-discretized formulation of the two-site exchange kinetic model is used to quantify metabolite signal dynamics by two characteristic rate constants in the form of (i) an apparent build-up rate (quantifying the build-up of downstream metabolites from the pyruvate substrate) and (ii) an effective decay rate (summarizing signal depletion due to repetitive excitation, T1-relaxation and backward conversion). The presented spectral and kinetic quantification were experimentally verified in vitro and in vivo using hyperpolarized [1-13C]pyruvate. Using temporally resolved IDEAL spiral CSI, spatially resolved apparent rate constant maps are also extracted. In comparison to single metabolite images, apparent build-up rate constant maps provide improved contrast by emphasizing metabolically active tissues (e.g. tumors) and suppression of high perfusion regions with low conversion (e.g. blood vessels). Apparent build-up rate constant mapping provides a novel quantitative image contrast for the characterization of metabolic activity. Its possible implementation as a quantitative standard will be subject to further studies.