Dynamic contrast enhanced (DCE) MRI non-invasively measures the uptake kinetics of a contrast agent in tumor tissues. The vascular parameters calculated from the uptake kinetics can be used as biomarkers to characterize individual tumors. These biomarkers can be used for non-invasive tumor grade and evaluation of tumor response to anticancer therapies. Low molecular weight contrast agents, which are exclusively used in clinical studies, rapidly extravasate through both tumor blood vessels and normal blood vessels and are not ideal for tumor characterization. Macromolecular contrast agents with molecular weights larger than 20 KDa have limited extravasation through normal blood vessels and can selectively pass through porous tumor blood vessels. They are considered to be effective for more accurate tumor characterization with DCE-MRI. Unfortunately, macromolecular contrast agents are not available for clinical application because of safety concerns. Polydisulfide based biodegradable macromolecular MRI contrast agents have been developed to alleviate the safety concerns (25
). The biodegradable macromolecular contrast agents can be rapidly excreted from the body after the MRI studies and have showed minimal long-term tissue accumulation comparable that of low molecular weight clinical contrast agents. These agents have shown promises to be further developed for clinical applications, including tumor characterization with DCE-MRI.
In this study, the effectiveness of different biodegradable macromolecular MRI contrast agents in tumor characterization with DCE-MRI were preliminarily investigated in two different mouse tumor models. The results obtained with the biodegradable macromolecular contrast agents were compared to those obtained with a clinical contrast agent, Gd(DTPA-BMA), and a prototype macromolecular contrast agent, albumin-(Gd-DTPA). In both tumor models, the values of the calculated vascular parameters, particularly microvascular permeability Ktrans, showed a dependence on size and degradability of the contrast agents (). The values estimated by biodegradable macromolecular contrast agents were between those by the small molecular weight contrast agent and those by the non-degradable macromolecular contrast agent. With the same high molecular weight, GDCC-70 resulted in a higher Ktrans values than GDCP-70 in both tumors due to high degradability of the former. However, the difference of biodegradability between GDCC-20 and GDCP-20, which had similar low molecular weight, had not significant effect on the Ktrans values possibly due to their relatively small size and high diffusion rate from plasma to tumor extracellular space.
It appears that Ktrans was a more sensitive parameter than fPV in tumor characterization as shown in . There was a significant difference in Ktrans between PC-3 and MDA PCa 2b prostate tumor xenografts estimated by most of the tested contrast agents, including Gd(DTPA-MBA), GDCC-20, GDCC-70, GDCP-70, and albumin-(Gd-DTPA). No significant difference was observed in Ktrans between the two tumor models when assessed by GDCP-20, probably due to a relatively small sample size and large experimental errors among the samples (). In contrast, the difference of fPV between two tumor models was not significant for most agents except for GDCC-70. The result was consistent to the histological observation that the microvessel density was not significantly different between two tumor models. No clear trend was observed on the influence of the size and degradability of the contrast agents on the fPV values.
The correlation of the Ktrans
values to physiology and histology of the tumor models is not clear. Although the MDA PCa 2b tumor xenografts grew slower than the PC-3 tumor xenografts, the Ktrans
values of the MDA PCa 2b tumor xenografts were significantly larger than those of the PC-3 tumor xenografts estimated with most of the agents. Similar uptake kinetics was reported previously studies by Kim et al. for two tumor with different agents (31
). PC-3 cells are androgen-independent and poorly differentiated cells, while MDA PCa 2b cells are androgen-sensitive and relatively well-differentiated cells (37
). There might be an association of tumor vascular permeability with the androgen-sensitivity of the cancer. We also observed the MDA PCa 2b tumor xenografts were softer than the PC-3 tumor xenografts when palpated. Further studies are needed to understand the correlation of the correlation of the Ktrans
values of the tumor models to their physiology and histology.
The degradation of the biodegradable macromolecular contrast agents might affect accurate data analysis and calculation of vascular parameters due to the change of the relaxivities of the degradation products since the contrast agents degraded into smaller chelates during the process of the DCE-MRI data acquisition. As shown in , the difference in the T1 relaxivities of biodegradable macromolecular contrast agents of two different molecular weights (20 and 70 KDa) was relatively small and may not be significant in DCE-MRI data analysis because DCE-MRI is not a perfectly quantitative method. The vascular parameters calculated from the DCE-MRI data with the biodegradable macromolecular contrast agents, particularly those with a relatively large size and slow degradation rate, were comparable to those estimated by albumin-(Gd-DTPA) in both tumors without consideration of the change of relaxivities. The biodegradable macromolecular MRI contrast agents, particularly those with high molecular weights, were effective to characterize individual tumors with DCE-MRI.