The Impact of Metal Artifacts on IMRT Dose Distributions in TCP and NTCP

Purpose: To investigate the effects of distorted head-and-neck (H&N) IMRT dose distributions (hot and cold spots) on normal tissue complication probability (NTCP) and tumor control probability (TCP) due to dental-metal artifacts. Methods : Five patients' IMRT treatment plans have been analyzed, employing five different planning image data-sets: (a) uncorrected (UC); (b) homogeneous uncorrected (HUC); (c) sinogram completion corrected (SCC); (d) minimum-value-corrected (MVC), and (e) streak-artifact-reduction including minimum-value-correction (SAR-MVC), that has been taken as the reference data set. The effects on NTCP and TCP were evaluated using the Lyman-NTCP model and the Logistic TCP model, respectively. Results: When compared to the predicted NTCP using the reference data set, the treatment plan based on the original CT data-set (UC) yielded an increase of 3.2 %, and 2.0 % in NTCP for the spared parotid gland and the spinal cord. While the treatment plans based on the MVC CT data set NTCP increased by 1.1 % and 0.1% for the spared parotid and the spinal cord. In addition, the MVC correction method showed reduced Delta TCP (MVC: 0.6% vs. UC.: 1.3%) with respect to that of the reference CT data set. Conclusion: Our results indicate that the presence of dental-metal-artifacts in H&N planning CT data sets has an impact on the estimates of TCP and NTCP. In particular dental-metal-artifacts lead to an increase in NTCP for the spared parotid and a slight decrease in TCP for TV.