Statistical control of processes applied to geometric uncertainties for CTV expansion margins determination in prostate cancer patients treated with VMAT: a prospective study in 57 patients.

Abstract

To study the control graphs applicability for the geometric uncertainties of VMAT treatments in prostate cancer patients, and their use to verify the hypothesis of the data obtained randomness, to apply the margins of Van Herk expression. During the first 5days of treatment, and then once a week, a Kv CBCT was performed, compared with the simulation CT and adjusted the displacements, to determine the inter-fraction errors. Immediately after radiation therapy, another CBCT was performed (for intra-fraction errors). With these data, the X, R position control charts have been made. The patients, not maintained the deviations within the charts control limits, were called "anomalies". Then, we compared the deviations and margins calculated with the van Herk expression for all patients and for those without anomalies. The margins determined show appreciable differences if there were calculated for the total set of patients or for the set of them without anomalies in the control charts. For the overall set of patients, the lateral, longitudinal, and vertical margins were 0.45cm, 0.52cm, 0.56cm, while for the set of patients without anomalies were 0.29cm, 0.35cm, and 0.38cm. The use of control charts allows tracking geometric deviations both inter and intra-fraction, variability real-time control and to detect situations in which it can change for non-random reasons, and require immediate investigation. Maintaining geometric deviations in the control state decreases the margins needed to administer a high dose to CTV in a high percentage of cancer prostate patients.