3'-Deoxy-3'-(18)F-fluorothymidine ((18)F-FLT) is used as a biomarker of cell proliferation. We investigated the kinetics of (18)F-FLT during treatment of malignant glioma with bevacizumab and irinotecan.
Methods: Fifteen patients with recurrent high-grade brain tumors (2 grade III, 13 grade IV) were studied at baseline (study 1 [S1]), after 1 course of therapy (2 wk, study 2 [S2]), and at the end of therapy (6 wk, study 3 [S3]). (18)F-FLT (1.5 MBq/kg) was administered intravenously, and dynamic PET was performed for 1 h. Curves representing blood clearance and tumor uptake were derived from factor images and summed frames with thresholding techniques or with a fixed cube. The standard (18)F-FLT model was used to estimate the rate constants. (18)F-FLT uptake was measured at 2 time points (early standardized uptake value [SUV(early)] and late SUV [SUV(late)]).
Results: Parameters appeared similar for curves derived from factor images and summed frames; the steepest drop occurred between S1 and S2 for transport, influx, SUV(early), and SUV(late). Three groups were distinguished on the basis of clinical outcome: patients who died within 6 mo (group 1 [G1], n = 4), survived 6-12 mo (group 2 [G2], n = 6), and survived more than 1 y (group 3 [G3], n = 5). None of the rate constants was significantly different between the groups. Long-term survivors (G3) showed a significantly different SUV change (in percentage) between S1 and S3, whereas short-term survivors (G1 and G2) did not.
Conclusion: Overall, the relative SUV change from S1 to S3 predicted a favorable clinical outcome, whereas the SUV change from S1 to S2 did not. Long-term survivors (G3) showed a significant drop in SUV from S1 to S2 and from S1 to S3. Significant correlations were found between SUV and both the rate constant and the influx rate. The correlation coefficient between SUV(late) and influx rate was 0.91, permitting response monitoring by the measurement of (18)F-FLT uptake changes.