A reproducible dynamic phantom for sequence testing in hyperpolarised 13C-magnetic resonance

Abstract

Objective: To develop a phantom system which can be integrated with an automated injection system, eliminating the experimental variability that arises with manual injection; for the purposes of pulse sequence testing and metric derivation in hyperpolarised ¹³C-MR.

Methods: The custom dynamic phantom was machined from Ultem and filled with a nicotinamide adenine dinucleotide and lactate dehydrogenase mixture dissolved in phosphate buffered saline. Hyperpolarised [1-¹³C]-pyruvate was then injected into the phantom (n = 8) via an automated syringe pump and the conversion of pyruvate to lactate monitored through a ¹³C imaging sequence.

Results: The phantom showed low coefficient of variation for the lactate to pyruvate peak signal heights (11.6%) and dynamic area-under curve ratios (11.0%). The variance for the lactate dehydrogenase enzyme rate constant (kP) was also seen to be low at 15.6%.

Conclusion: The dynamic phantom demonstrates high reproducibility for quantification of ¹³C-hyperpolarised MR-derived metrics. Establishing such a phantom is needed to facilitate development of hyperpolarsed ¹³C-MR pulse sequenced; and moreover, to enable multisite hyperpolarised ¹³C-MR clinical trials where assessment of metric variability across sites is critical.

Advances in knowledge: The dynamic phantom developed during the course of this study will be a useful tool in testing new pulse sequences and standardisation in future hyperpolarised work.

Figure 1.

Important downstream metabolites from pyruvate and their respective enzyme pathways. Pyruvate can be converted into lactate via LDH, alanine through ALT, carbon dioxide by PDH and subsequently bicarbonate produced from CA. ALT, alanine aminotransferase; CA, carbonic anhydrase; LDH, lactate dehydrogenase; PDH, pyruvate dehydrogenase

Figure 2.

Visualisation of the dynamic phantom, created and rendered in Autodesk’s Fusion 360 – (a) main body of the phantom, with a hole for (b) a ¼” threaded male Luer lock outlet and a groove for (c) a nitrile O-ring. (d) The lid is fixed to the main body via (e) four 6,6-Nylon M3 screws, whilst an additional hole was added to the lid to allow for (f) a custom inlet to be fitted.

Figure 3.

Change in the metabolite signals observed in the non-localised spectra after the injection of hyperpolarised [1-¹³C] pyruvate into the dynamic phantom. The spectra were normalised to the strongest signal ([1-¹³C] pyruvate at 172 ppm) observed in the first spectrum (19 s). The spectra are zoomed in upon to show all the signals - [¹³C] Urea (165 ppm), [1-¹³C] pyruvate (172 ppm), [1-¹³C] pyruvate hydrate (181 ppm) and [1-¹³C] lactate (185 ppm) signals.

Figure 4.

Change in [1-¹³C] lactate (blue) and [1-¹³C] pyruvate (red) signals, within the dynamic phantom, over time, across all HYP-MR experiments (n = 8). The signal from the [¹³C] urea reference phantom (yellow) within the endorectal coil is also shown.

Figure 5.

(a) Distribution of [1-¹³C] pyruvate and [1-¹³C] lactate within the dynamic phantom, at t = 12 s are shown. The metabolite maps were overlaid on a set of T ₂ weighted reference images. The [¹³C] urea signal from within the endorectal receive coil is also shown. (b) The change in [1-¹³C] pyruvate and [1-¹³C] lactate signals at the centre of the phantom are shown up to 94 s after the start of injection. Note, that slice no. 4 is shown in( b) with a slice thickness 10 mm. It is likely that this slice includes parts of the two injection nozzles which divide the pyruvate injection to the left and right part of the chamber, shown by the increased intensities to left and right of the chamber. The images are scaled to the maximum signal intensity (arbitrary units) of pyruvate and lactate in their respective time courses.