Reproducibility of total choline/water ratios in mouse U87MG xenograft tumors by 1H-MRS.

PURPOSE: To evaluate the reproducibility of the measurement of the total choline-to-water ratio, and the effect of repositioning the subject between scans, using (1) H-magnetic resonance spectroscopy in a mouse U87MG xenograft model. MATERIALS AND METHODS: In vivo single-voxel MR spectra at 7T from xenograft tumors were obtained using both a water-suppressed and a nonwater-suppressed point-resolved spectroscopy (PRESS) sequence. Reproducibility of the total choline/water ratio was evaluated under the conditions of immediate rescan with no change in position of the animal or voxel, immediate reposition, and reposition after 1 or 7 days. RESULTS: Total choline-to-water ratios in U87MG tumor xenografts averaged ≈0.018 across all of the groups. The average percent difference between the two scans in each condition was always less than ≈3.0%, and the coefficient of variation was always less than ≈12%. Bias was unrelated to the testing condition and relatively negligible in magnitude (<3%). Due to heteroscedasticity in the ratios, the limits of agreement were calculated after log transformation of the data and ranged from ≈12% when animals were maintained in the same position and immediately rescanned to ≈52% when the two scans were 7 days apart. CONCLUSION: The total choline-to-water ratio provides a reproducible measure of choline-containing metabolites in subcutaneous U87MG xenograft tumors in mice.