Assessment of hepatic transporter function in rats using dynamic gadoxetate-enhanced MRI: a reproducibility study

Objective Previous studies have revealed a substantial between-centre variability in DCE-MRI biomarkers of hepatocellular function in rats. This study aims to identify the main sources of variability by comparing data measured at different centres and field strengths, at different days in the same subjects, and over the course of several months in the same centre. Materials and methods 13 substudies were conducted across three facilities on two 4.7 T and two 7 T scanners using a 3D spoiled gradient echo acquisition. All substudies included 3–6 male Wistar-Han rats each, either scanned once with vehicle (n = 76) or twice with either vehicle (n = 19) or 10 mg/kg of rifampicin (n = 13) at follow-up. Absolute values, between-centre reproducibility, within-subject repeatability, detection limits, and effect sizes were derived for hepatocellular uptake rate (Ktrans) and biliary excretion rate (kbh). Sources of variability were identified using analysis of variance and stratification by centre, field strength, and time period. Results Data showed significant differences between substudies of 31% for Ktrans (p = 0.013) and 43% for kbh (p < 0.001). Within-subject differences were substantially smaller for kbh (8%) but less so for Ktrans (25%). Rifampicin-induced inhibition was safely above the detection limits, with an effect size of 75 ± 3% in Ktrans and 67 ± 8% in kbh. Most of the variability in individual data was accounted for by between-subject (Ktrans = 23.5%; kbh = 42.5%) and between-centre (Ktrans = 44.9%; kbh = 50.9%) variability, substantially more than the between-day variation (Ktrans = 0.1%; kbh = 5.6%). Significant differences in kbh were found between field strengths at the same centre, between centres at the same field strength, and between repeat experiments over 2 months apart in the same centre. Discussion Between-centre bias caused by factors such as hardware differences, subject preparations, and operator dependence is the main source of variability in DCE-MRI of liver function in rats, closely followed by biological between-subject differences. Future method development should focus on reducing these sources of error to minimise the sample sizes needed to detect more subtle levels of inhibition. Supplementary Information The online version contains supplementary material available at 10.1007/s10334-024-01192-5.


THE TRISTAN RAT ASSAY: DEFINITION
The TRISTAN rat assay measures the hepatic plasma clearance rate, K trans [mL/min/mL] and biliary efflux rate, kbh [mL/min/mL] of gadoxetate in rats.Guidelines including detailed technical specifications and SOPs on animal handling, data acquisition, and image processing for the TRISTAN rat assay are provided below.Relevant software for image analysis [1] and modelling [2] are also made available.

SUBJECTS AND PREPARATION
Subjects: male Wistar-Han rats.
Number of subjects: Investigators are permitted to choose the number of rats, balancing the ethical need to minimise animal use while entering sufficient rats to obtain estimates of the uncertainty adequate for the context of use.
Preparation: Animals are allowed to acclimate for 1 week prior to initiation of the imaging study with adequate husbandry.For the scan, animals are anaesthetised with a suitable inhalation anaesthetic such as isoflurane to place a tail vein catheter and maintained under anaesthesia using approximately 2% isoflurane in the air mixture.Animals should be monitored during the scan for respiratory rate and body temperature and provided external heat, e.g., through forced hot air or circulating warm water.Before the start of the exam, animals receive a saline or vehicle control, or a test compound.

DATA ACQUISITION
MRI is performed on an MRI system suitable for high-quality dynamic rat MRI [3].Such systems typically operate between 3T and 14T.MRI sequences include localizers, followed by a T2-weighted (T2W) spin echo sequence for delineation of the anatomy.For DCE, heparinized | 2 saline is placed in the tail vein catheter prior to the bolus of gadoxetate.Gadoxetate is diluted 1:5 in saline and administered at 0.5 mL/kg (25 μmol/kg bodyweight).
Dynamic imaging is performed at approximately 60 sec per time point, and 30 time points are acquired.The first five time points are considered baseline, and the IV bolus of gadoxetate is injected at the beginning of the sixth measurement at 4 min and 45 seconds into the acquisition.
The slice selections of the T2W, T1 map, and dynamic series all use the same slice orientation to facilitate ROI selection.

IMAGE PROCESSING
MRI images are exported, ideally in DICOM format, and analysed using any suitable software package which permits the operations below.The location of the liver is confirmed on T2W images, and three circular ROIs are placed in three separate slices of liver parenchyma, ventral to the caudate processes and approximately 10 voxels in diameter; the locations of the ROIs are then used to record liver signal intensities averaged over the ROI on the dynamic series.
RoIwise rather than voxelwise analysis minimises the risk of biased data at the Rician noise floor.Signal time curves are extracted in csv format and further processed using a python script available here [2].The script fits the time curves to a model (details below).The three ROI values are then averaged producing a single value for K trans and kbh for each subject at each visit.

DATA MODELLING
The kinetic model assumes that two separate liver compartments can be distinguished, the extracellular space, e and the intrahepatocytic compartment, h (defined by their respective volumes, ve and vh, the mean transit time in h, Th, and the extraction fraction of gadoxetate from the extracellular to the hepatocyte compartment, E).The mean transit time in the | 3 extracellular space is assumed to be negligible at a sampling time of 58s.With a given arterial input function ca(t), this produces the following solution for the concentrations in e and h, ce(t) and ch(t): where Fp is the liver plasma flow in mL/min/mL.
Considering the difficulty in measuring an arterial input function reliably in small animals, the TRISTAN rat assay implements a standardised ca(t) derived from a simplified model of the rat circulation [4] as shown in Figure 1.Within this model, literature values for rats are used for all constants as follows: whole body permeability-surface for gadoxetate (PS) was estimated from the rat reproducibility data as 0.172 mL/sec, using the PBPK model from Table 3 in Scotcher et al. [5]; whole body plasma volume (Vp) = (1 -Hct)(VB) [mL], where Hct is the hematocrit equal to 0.418 [6] and VB is the whole body blood volume, estimated as 15.8 mL when assuming a body weight of 250 g [7]; whole body extracellular volume (VE) was derived as 30 mL from Table 2 in the supplementary material in Scotcher et al. [5]; K = GFR + (K trans )(VL), where GFR is the glomerular filtration rate equal to 0.023 mL/sec [8], K trans is the hepatic plasma clearance rate [mL/min/mL], and VL is the liver volume equal to 8.47 mL [5].
The relaxation rate of liver tissue is derived from the concentrations as follows: | 4 A single consensus pre-contrast longitudinal relaxation rate R10 is adopted for each field strength (R10 = 1.281 s -1 at 4.7T; R10 = 1.109 s -1 at 7T).This is based on a weighted mean of rat liver R1 values aggregated with published values [9].Values for hepatocyte and blood plasma relaxivity (r1,h and r1,p) are also obtained from literature [10]: 7.6 s -1 /mM and 6.0 s -1 /mM for hepatocytes at 4.7T and 7T, respectively, and 6.4 s -1 /mM and 6.2 s -1 /mM for blood plasma at 4.7T and 7T, respectively.ve is equally fixed to the literature value of 0.230 [10].
Finally, signals, S are derived from liver relaxation rates assuming the standard signal model of a variable flip angle sequence in steady state: | 5 where  is the flip angle, TR is the repetition time, S is the signal amplitude in arbitrary units (a.u.) and S0 (in the same a.u.) is the signal asymptote at infinite TR.
Inserting Eq. ( 1) into Eq.( 2) and Eq. ( 2) into Eq.( 3) produces a model for the signal time course S(t).Previous sensitivity analysis [11] has demonstrated that aside from S0, only two parameters E and Th can be reliably determined from the given data.Hence, Fp is fixed to a value of 1.32 mL/min/mL, assuming Fp = (1-Hct)*Fb, where Fb is the blood flow = (hepatic artery flow [mL/min] + portal vein blood flow [mL/min]) / mass of the liver [mL], and is calculated as 2.27 mL/min/mL using literature values from Tables 3-4 in the supplementary material of Scotcher et al. [5].Model fitting for the remaining free parameters S0, E and Th is performed by the Python script accessible on Zenodo [2].The hepatic plasma clearance rate parameter, K trans is derived from K trans = E*Fp and gadoxetate efflux, kbh, is calculated as kbh = vh/Th, using 0.722 as the hepatocellular extracellular volume fraction, vh (calculated from vh = vtvi, where vt is the tissue volume fraction = (1blood volume fraction) = 0.115 and vi is the interstitial volume fraction = 0.163 [12]).

DATA ANALYSIS AND REPORTING
Analysts blinded to treatment perform a quality assessment of the data, considering artefacts or missing data in the scans and goodness-of-fit of the model to the signal time curves.If a single dataset shows strong artefacts or poor goodness-of-fit, the subject should be excluded.
For the remaining subjects, K trans and kbh values are averaged to produce a final set of values, and 1.96 x standard error on the mean is returned as the 95% confidence interval (CI).The variance in the standard error must be calculated using Bessel's correction.These numbers are reported to the requesting entity as the rate constants of the drug, with centre-specific uncertainty, as measured by the TRISTAN rat assay.In addition, error intervals must be derived | 6 using the percentage errors measured in this report (reproducibility error).Benchmark values in saline derived in this report should be included as reference.

Figure 1 .
Figure 1.Tracer kinetic model for blood concentrations in rats.PS = whole body permeability-surface for Gadoxetate; VP = whole body plasma volume; VE = whole body extracellular volume; K = GFR + (K trans )(VL),where GFR is the glomerular filtration rate, K trans is the hepatic plasma clearance rate, and VL is the liver volume.