Isointense model for the evaluation of tumor-specific MRI contrast agents

An isointense model has been developed to evaluate the applicability of putative tumor-specific MRI contrast agents. Data for tissue relaxation measurements in the presence of Mn(III)TPPS₄ are used to illustrate the model. The concentration of contrast agent in tumor tissue required for a tumor/normal tissue signal difference-to-noise ratio of 5 (ΔSNR = 5) is determined for a T₁ weighted pulse sequence and several hypothetical tumor/normal tissue pairs. The impact of various contrast agent characteristics including initial tumor/normal tissue relaxation values, differential uptake of contrast agent, and in vivo relaxivity are considered. Isointense tumor/normal tissue with longer initial relaxation times are shown to be more affected by the presence of contrast agent. In addition those with initially longer relaxation times have less rigorous requirements for tumor specificity. Typically, a normal tissue/tumor uptake ratio of 1:2 increases the concentration required for ΔSNR = 5 by a factor of two compared to that of exclusive uptake in tumor. For the T₁ weighted pulse sequence employed, the concentration required for ΔSNR = 5 is shown to be linear with the inverse of in vivo relaxivity for the hypothetical tissues considered. The isointense model is also extended to predict the field dependence of tumor-specific contrast enhancement by Mn(III)TPPS₄.