Summary Basis of Decision for Primovist ®

3.2.1 Pharmacodynamics

Primary Pharmacodynamics

Gadoxetate disodium is a paramagnetic compound that develops a magnetic moment when placed in a magnetic field. The relatively large magnetic moment produced results in a local magnetic field, yielding enhanced relaxation rates (shortening of relaxation times) of water protons in the vicinity of the paramagnetic agent. This leads to an increase in signal intensity (brightening) of blood and tissue.

In MRI, visualization of normal and pathological tissue depends in part on variations in the radiofrequency signal intensity that occur with differences in proton density; differences of the spin-lattice or longitudinal relaxation time (T₁); and differences in the spin-spin or transverse relaxation time (T₂). When placed in a magnetic field, gadoxetate disodium decreases the T₁- and T₂-relaxation time in target tissue. At the recommended dose, the effect is observed with greatest sensitivity in T₁-weighted magnetic resonance (MR) sequences.

In non-clinical pharmacodynamic (PD) studies, the relaxation times of protons in water molecules were used to assess the paramagnetic effect of gadoxetate, in water, plasma, bile, and liver tissue. The T₁- and T₂-relaxation times in water and plasma (bovine) were measured in samples containing 0, 0.25, 0.50, and 1.0 mmol/mL of gadoxetate as its disodium salt or a reference. The T₁-relaxation times in liver were determined following a single IV injection of 0.25 mmol/kg gadoxetate as its dimeglumine salt in rats. Some animals were sacrificed without receiving an injection. All other animals were sacrificed at 5, 10, 20, 30, or 60 minutes after administration. The livers were excised and divided into three parts. The T₁-relaxation time and the gadolinium concentration were determined in each fraction of the livers immediately after excision. The T₁-relaxation times in bile were determined after single IV injection of 0.5 mmol/kg gadoxetate as its dimeglumine salt in rats. The bile was collected in fractions while the animals were under anaesthesia. The T₁-relaxation time and the gadolinium concentration were determined in the bile fractions.

Both relaxation times (T₁ and T₂) were markedly reduced in all the studies, demonstrating the pronounced paramagnetic effect of gadoxetate. A further increase in T₁ relaxivity (the slope of the linear plot of relaxation rate in sec -1 versus the concentration of contrast agent in mmol/L) was observed in liver tissue as gadoxetate disodium is selectively taken up by hepatocytes. As liver lesions have no or minimal hepatocyte function (cysts, metastases, and the majority of hepatocellular carcinomas), the concentration of Primovist is much lower than in the surrounding tissue, resulting in an increased contrast image of normal liver tissue versus liver lesions lacking normal hepatocyte function in MRI.

In a separate study, the relaxivity of gadoxetate disodium was measured in a 2.0-Tesla field. The contrast agent decreased the relaxation time of protons of water and increased the relaxivity. As a reference, the relaxivities of Magnevist^® (gadopentetate dimeglumine) in water and plasma were determined using the same procedure. Gadoxetate disodium decreased both T₁- and T₂-relaxation times in water and plasma. The relaxivities of gadoxetate disodium were higher than those of Magnevist^® in water and plasma.

The efficacy of gadoxetate disodium to enhance the liver was studied in female rats by means of MRI images taken using a 2.0-Tesla animal imager. Standard imaging sequences (T₁-weighted spin-echo sequences) were used. Even after the lowest dose of 10 µmol gadolinium (Gd)/kg, there was a marked increase in contrast between the tumour and the liver parenchyma. The contrast increased further as the dose was increased and only after the highest dose of 60 µmol Gd/kg was there no additional increase in contrast.

A pharmacological characterization study was conducted to characterize the individual gadoxetate disodium enantiomers. There was no statistically significant difference between either enantiomer in binding to human plasma. In rats, the median lethal dose (LD₅₀) was estimated to be between 7.5 and 10.0 mmol/kg for both enantiomers. The relaxivity of both enantiomers was not seen to differ.

Secondary Pharmacodynamics

Pharmacological parameters such as protein binding, influence on the morphology of erythrocytes, enzyme inhibition, haemodialysis and complement activation were studied in in-vitro tests.

The protein binding of gadoxetate was found to be exceedingly low and vary between 7.7 and 9.1%. A study was conducted to evaluate if the low protein binding of gadoxetate disodium would interfere with hemodialysis. Gadoxetate disodium is quickly and completely removable from plasma by dialysis through a commercial hemodialysis device. In a separate study, it was determined that there was no difference between the effects of gadoxetate disodium and Magnevist^® in terms of their influence on the morphology of erythrocytes.

Safety Pharmacology

Safety pharmacological investigations studied the potential effects of gadoxetate disodium on several organ functions such as blood coagulation (rat), kidney (rat, rabbit), cardiovascular (in-vitro and in-vivo, dog), central nervous system (CNS) (mouse, rat) and respiratory (rabbit).

Gadoxetate disodium did not exhibit any major effect on renal function; however, the single-dose study which examined renal function had no control group making the results difficult to interpret. Urine flow was increased at 2 hours and returned to baseline at 6 hours (2.7 to 6.9 mL/h). Proteinuria was increased with four of five rabbits having increased urine protein content (30 mg/dL). An increase in creatinine clearance was observed at 2 hours which returned to baseline at 6 hours (4.92 mL/min to 17.14 mL/min). As well, lactate dehydrogenase (LDH) was increased in urine from 2 to 48 hours. Gadolinium is taken up by bone tissue, as seen by the low concentrations found in bone at Day 7 of the study.

Gait disturbances were noted in mice at 800 mg/kg 30 minutes post dose. These effects were not observed in the control group.

There was a dose-dependent increase in action potential duration, and a significant change in membrane potential (p = 0.015) of the action potential at 30% repolarization (ADP30) at a dose of 10 mmol/L and at 1 Hz stimulation in isolated guinea pig papillary muscle. This concentration is about 38-times higher than the clinical exposure (0.26 mmol/L at 2 minutes).

An increase in QTcF and QTcQ was seen with the use of gadoxetate disodium at 0.1 and 0.5 mmol/kg; however, the study only used four animals. Nevertheless, the effect of gadoxetate disodium on the prolongation of the QTc interval cannot be excluded by this study. On further investigation and consultation with a clinical expert, it was concluded that Primovist has a minimal effect on the QTc interval.

An effect of drug interaction on the MRI enhancement properties of gadoxetate disodium was investigated in rats. The results demonstrated that compounds belonging to the class of rifamycins block the hepatic uptake of gadoxetate disodium, thus reducing the hepatic contrast effect. In this case, the expected benefit of an injection of Primovist may be limited. Interactions with other medications are not known.

3.2.2 Pharmacokinetics

Biodistribution studies conducted in rats using radiolabelled gadoxetate disodium demonstrated that the highest levels of gadoxetate were found in the liver, kidney, and bowel (due to biliary excretion). Gadoxetate was still detectable in the femoral bone and testis 72 hours post-administration.

In rats, the maximum concentration of radioactivity in the foetus and amniotic fluid were below one one-hundredth of that in maternal plasma, and radioactivity disappeared completely by 24 h. There was evidence that gadoxetate disodium is transferred via milk. The amount transferred to a neonate via milk was below 0.5% of the dose administered to the dam.

Plasma binding in both rat and dog was observed to be ~10%. The data presented suggests that the study agent is dialyzable.

There was no incidence of biotransformation of gadoxetate disodium as examined by mass spectroscopy, with high-performance liquid chromatography (HPLC), or inductively coupled plasma atomic emission spectroscopy (ICP-AES).

Both dog and rat studies showed non-linear and dose-dependent excretion by renal and hepatic routes. The renal excretion was linear and only limited by the glomerular filtration rate (GFR). The biliary excretion route (uptake by hepatocytes, excretion into bile, then to faeces) was the rate-limiting step. In dog studies, recovery of gadoxetate disodium was approximately 97% after 7 days.

Rat studies showed that the hepatic elimination was blocked with sulfobromophthalein, suggesting that the organic ion transport mechanism is involved. The enterohepatic circulation in rats was reported to be 3.89% (± 2.12%) of the injected dose.

Pharmacokinetic (PK) data derived from the pharmacologic characterization study indicated that the enantiomers were bioequivalent with respect to the amount eliminated in faeces. Complete bioequivalence was not seen in area under the concentration time curve (AUC), renal clearance, and amounts recovered in urine; these were not statistically bioequivalent. This could be due to the higher renal excretion rate of the R-enantiomer; however, this is not likely of clinical relevance.

3.2.3 Toxicology

Toxicity studies were conducted with gadoxetate disodium supplied in 0.25 mmol/mL and 0.5 mmol/mL formulations.

Single-dose Toxicity

Single-dose toxicity studies were conducted in mice, rats, and dogs. The results indicate that following IV administration, gadoxetate disodium was tolerated without lethality at doses up to 7.5 mmol/kg (mice), 10.0 mmol/kg (rats), 5.0 mmol/kg (weaned rats), and 3.0 mmol/kg (dogs). Higher doses were not tested in dogs as the dose of 3.0 mmol/kg already exceeded the envisaged diagnostic dose in humans by approximately 120-fold on the basis of body weight. Mortality was observed at doses of 10.0 mmol/kg (mice), 12.5 mmol/kg (rats), and 7.5 mmol/kg (weaned rats).

Repeat-dose Toxicity

In 4-week, repeated IV dose toxicity studies conducted with rats and dogs, relatively minor clinical signs and effects on body weight, food consumption, and clinical pathology were observed mainly with mid-doses (0.3 mmol/kg) and high-doses (1.0 mmol/kg) in dogs; however, the most significant finding was dose-dependent reversible renal tubular vacuolation in mid- (0.6 mmol/kg) and high-dose (2.0 mmol/kg) rats and high-dose dogs. This reversible vacuolation has also been observed with other gadolinium-based contrast agents (GBCAs).

The no observable effect level (NOEL) in rats (0.2 mmol/kg) and dogs (0.1 mmol/kg) to human (0.025 mmol/kg) dose ratios were 8-fold based on body weight and 1.3-fold based on body surface area in rats and 4-fold based on body weight and 2.2-fold based on body surface area in dogs.

The margins of safety are considered acceptable as Primovist is to be used as a single-dose agent.

Mutagenicity

Gadoxetate disodium was not mutagenic or clastogenic in a battery of in vitro and in vivo tests that exceed current ICH guideline recommendations.

Reproductive and Developmental Toxicity

The reproductive and developmental studies were consistent with current ICH guidelines. Gadoxetate disodium was not teratogenic in rats, at multiples of the 0.025 mmol/kg human dose up to 200-fold based on body weight and 32-fold based on body surface area. Further testing in rabbits also demonstrated that gadoxetate disodium was not teratogenic; however, it did cause embryo-foetal toxicity at multiples of the human dose of 80-fold based on body weight and 26-fold based on body surface area, with the NOEL at 20-fold based on body weight and 6.5-fold based on body surface area.

No effects on fertility were observed in rats at doses up to 40-times the human dose based on body weight or 6.5-fold based on surface area. In a peri- and post-natal development study, maternal toxicity was observed, but no developmental effects on their progeny were seen at doses up to 48-times the human dose based on body weight or 7.8-fold based on surface area.

Local Tolerance

Evaluation of local tolerance revealed that the 0.25 mmol/mL formulation was generally well tolerated when administered to rabbits and/or rats by the IV, paravenous, and intra-arterial routes. This was in contrast to the 0.5 mmol/mL formulation which caused reactions (slight to moderate reddening of the skin with vessel injection) in the congested vein. No intolerance reactions were seen in the uncongested vein. Intramuscular (IM) injection of either formulation caused injection site reactions. Direct injection of the 0.5 mmol/mL formulation into the liver, mammary gland, and prostate gland did not cause any local reactions. Dermal application of gadoxetate disodium was well tolerated by rabbits as was 10 mg, but not 100 mg, applied to the conjuctival sac of rabbits.

The 0.25 mmol/mL formulation of gadoxetate disodium did not show antigenic potential in a mouse-rat test system and in guinea pigs or delayed hypersensitivity potential in guinea pigs.

3.2.4 Summary and Conclusion

The results of the non-clinical PD studies support the proposed indication for Primovist for the detection and characterization of lesions in the liver. Safety pharmacology studies suggest the potential for gadoxetate disodium to affect vital functions in humans is low.

The PK profile of gadoxetate disodium has been adequately characterized in the non-clinical studies. It is excreted primarily by renal and hepatic routes.

The toxicity profile of gadoxetate disodium has been characterized. The results of the toxicity studies support the intended single IV dose of a 0.25 mmol/mL formulation for diagnostic imaging purposes.