Prediction of Intrafraction Prostate Motion: Accuracy of Pre- and Post-Treatment Imaging and Intermittent Imaging

Purpose

To evaluate whether pre- and post-treatment imaging (immediately before and after a radiation therapy treatment fraction) and intermittent imaging (at intervals during a treatment fraction) are accurate predictors of prostate motion during the delivery of radiation.

Methods and Materials

The Calypso 4D Localization System was used to continuously track the prostate during radiation delivery in 35 prostate cancer patients, for a total of 1,157 fractions (28–45 per patient). Predictions of prostate motion away from isocenter were modeled for a pre- and post-treatment imaging schedule and for multiple intermittent intrafraction imaging schedules and compared with the actual continuous tracking data. The endpoint was drift of the prostate beyond a certain radial displacement for a duration of more than 30 s, 1 min, and 2 min. Results were used to evaluate the sensitivity and specificity of these models as an evaluation of intrafraction prostate motion.

Results

The sensitivity of pre- and post-treatment imaging in determining 30 s of intrafraction prostate motion greater than 3, 5, or 7 mm for all fractions was low, with values of 53%, 49%, and 39%, respectively. The specificity of pre- and post-treatment imaging was high for all displacements. The sensitivity of intermittent imaging improved with increasing sampling rate.

Conclusions

These results suggest that pre- and post-treatment imaging is not a sensitive method of assessing intrafraction prostate motion, and that intermittent imaging is sufficiently sensitive only at a high sampling rate. These findings support the value of continuous, real-time tracking in prostate cancer radiation therapy.

Organ motion, Prostate cancer, Radiation therapy, Target localization, Tracking