Intratumor hemodynamics using contrast-enhanced MRI in intra-arterial chemotherapy for head and neck cancer.

OBJECTIVE: The objectives of the study were to estimate the perfusion of tumors by drugs used in intra-arterial chemotherapy for head and neck cancer with magnetic resonance imaging and to establish the factors involved in determining the optimal dose. STUDY DESIGN: Contrast agent was administered intra-arterially into either the lingual or maxillary artery in 43 patients. Triple-phase continuous fast spin echo magnetic resonance imaging was performed. Changes in blood water longitudinal relaxation rate (⊿R1) were measured in relation to imaging phase, type of artery, measurement site, and tumor size. RESULTS: ⊿R1 was significantly higher at the tumor margin than at the center for both arteries, except in the first phase for the lingual artery. ⊿R1 was greatest in the third phase for the lingual artery and in the second phase for the maxillary artery. For both arteries, as the tumor size increased, there was a significant decrease in ⊿R1 at the center of the tumor compared with the margin. CONCLUSIONS: The factors associated with ⊿R1 were imaging phase, type of artery, measurement site, and tumor size. When determining a drug's optimal dose, the type of artery and tumor size must be taken into consideration.

Drug concentration estimation using contrast-enhanced MRI in intra-arterial chemotherapy for head and neck cancers.

OBJECTIVE: In cases of head and neck cancer treated with intra-arterial chemotherapy, no objective indices are available for determining the distribution of anticancer drugs administered to multiple arteries. To establish such indices, noninvasive measurements of drug concentrations are required in the arterial perfusion area of each artery. In MRI, changes in 1/T1 (Δ1/T1) are correlated with the contrast agent concentration. We focused on these properties and investigated whether it is possible to estimate anticancer drug concentrations within tissue based on Δ1/T1. METHODS: We employed the fast spin echo (FSE) sequence to determine optimum imaging parameters using a phantom. Subsequently, contrast agent was administered via the lingual and external carotid arteries for seven cases of tongue cancer. Δ1/T1 were then measured in tumor and nontumor tissues. The results of this study were compared with those of a previous study in which intratumor concentrations of anticancer agent were measured in excised specimens. RESULTS: The optimum imaging parameters for the FSE was two repetition times (TR, 500 and 1000 ms). When compared with the external carotid artery administration, the lingual artery administration of contrast agent resulted in significantly higher Δ1/T1 in both tumor and nontumor tissues (2.13 and 2.62 times, respectively). The multiplying factor for the nontumor tissue and high homogeneity of the contrast agent concentration were reasonably consistent with the results of the previous study. CONCLUSION: This method can be applied to estimating intratissue concentrations of intra-arterially administered anticancer drugs, thus possibly providing useful information in determining the distribution of anticancer drugs.