How to compare the efficiency of albumin-bound and nonalbumin-bound contrast agents in vivo: the concept of dynamic relaxivity.

RATIONALE AND
OBJECTIVES: The objective of this study was to compare, in a rabbit experimental model that mimics a magnetic resonance (MR) angiographic protocol, the efficiency of the following types of compound on the MR signal: (1) a nonalbumin-bound blood pool contrast agent: P792; (2) a weak albumin-bound extracellular contrast agent: Gd-BOPTA; and (3) a strong albumin-bound blood pool contrast agent: MS325.
METHODS: The 2 main phases of early distribution after contrast agent injection, ie, the bolus phase (0-15 seconds postinjection) and the postbolus phase (1-5 minutes postinjection) were investigated by measuring Gd blood concentrations in the first 5 minutes postinjection. In the case of MS325 and Gd-BOPTA, the percentage of the free and bound forms were calculated throughout the pharmacokinetic profile. The dynamic relaxivity at 60 MHz in plasma of each contrast agent was determined in the 2 phases after contrast agent injection, ie, the bolus phase and the postbolus phase.
RESULTS: Injected under similar conditions, the 3 contrast agents had a comparable profile during the bolus phase (0-15 seconds postinjection). At 1 minute postinjection, only 38% of Gd-BOPTA remained in the blood, whereas 85% of P792 was still present in the blood. MS-325 had an intermediate position with 61% remaining in the blood. During the postbolus phase, the various compounds demonstrated similar behavior: the plasma concentration of P792 was higher than that of MS325 and Gd-BOPTA, ie, Ci/C0 (P792)>Ci/C0 (MS325)>Ci/C0 (Gd-BOPTA). At the peak of the bolus, 75% of MS325 and 93% of Gd-BOPTA was present in free form. This proportion decreased progressively during the postbolus phase, because 5 minutes postinjection, 23% of the free form remained for MS325 and 82% for Gd BOPTA. A significant decrease in dynamic r1 relaxivity was observed at 60 MHz for the products that bind to albumin (Gd-BOPTA and MS325) during the bolus phase. The dynamic relaxivity for MS325 at the bolus phase was 8.6 mMs and 5.2 mMs for Gd-BOPTA. At the postbolus phase, the dynamic relaxivity increased (17.3 mMs for MS325 and 6.7 mMs for Gd-BOPTA). The dynamic relaxivity of P792, which does not bind to albumin, was constantly equal to 26 mMs at each time point of the pharmacokinetic profile (bolus and postbolus phase).
CONCLUSIONS: The physicochemical measurements of relaxivity in plasma are made in vitro at a fixed concentration of gadolinium and the value of relaxivity is not necessarily an accurate reflection of the efficiency of the contrast agent in vivo, especially for contrast agents that bind to albumin. Indeed, in vivo, the proportion of free and bound forms of albumin-binding contrast agents varies according to the pharmacokinetic profile, and the relaxivities of albumin-bound and free contrast agents are different. Consequently, the concept of dynamic relaxivity was introduced to compare the efficiency of MS325, Gd-BOPTA, and P792 in vivo. The variation of the dynamic relaxivity of MS325 and Gd-BOPTA between the bolus and postbolus phase is significant (101% for MS325 and 29% for Gd-BOPTA) as a result of the variation in the quantity of bound and free forms during the pharmacokinetic profile. The blood pool agent P792 has different properties, which result from its intravascular retention and its lack of albumin binding. Indeed, contrary to Gd-BOPTA and MS325, the dynamic relaxivity of P792 is higher at the bolus phase (26 mMs) and does not vary during the pharmacokinetic profile. The impact of these different dynamic relaxivities should be integrated in the analysis of the performance of the different classes of contrast agents in clinical MRA protocols.