Effect of translational and rotational errors on complex dose distributions with off-line and on-line position verification.

Abstract

To investigate the influence of translational and rotational errors on prostate intensity-modulated radiotherapy (IMRT) with an integrated boost to the tumor and to evaluate the effect of the use of an on-line correction protocol. For 19 patients, who had been treated with prostate IMRT and fiducial marker-based position verification, highly inhomogeneous IMRT plans, including an integrated tumor boost, were made using varying margins (2, 4, 6, and 8 mm). The measured translational and rotational errors were used to calculate the dose using two positioning strategies: an off-line and an on-line protocol to correct the translational shifts. The estimated dose to the targets and the organs at risk was compared with the intended dose. Residual deviations after off-line correction led to statistically significant, but very small, reductions in dose coverage. Even when a 2-mm margin was used, the average reduction in dose to 99% of the volume was 1.4 +/- 1.9 Gy for the tumor, 1.5 +/- 1.5 Gy for the prostate without seminal vesicles (boost volume), and 4.3 +/- 4.6 Gy, including the seminal vesicles (clinical target volume). Patients with large systematic rotational errors demonstrated a substantial decrease in dose, especially for the clinical target volume. If an on-line correction protocol was used, the average mean dose and dose to 99% of the volume of the targets improved. However, the extensive dose reduction for patients with large rotational errors barely recovered with on-line correction. For complex prostate IMRT with an integrated tumor boost, the use of an on-line correction protocol yields little improvement without the correction of rotational errors.