Encrypted login | home

Program Information

Treatment Plan Complexity as a Factor of IROC Head and Neck Phantom Performance

M Carson

M Carson1*, J Kerns2 , S Zhou1 , D Followill1 , S Kry1 , (1) The University of Texas MD Anderson Cancer Center, Houston, TX, (2) CAMC Cancer Center, Charleston, WV


TU-FG-702-9 (Tuesday, August 1, 2017) 1:45 PM - 3:45 PM Room: 702

Purpose: To determine whether IMRT plan complexity influences IROC Houston’s anthropomorphic head and neck (H&N) phantom performance among institutions undergoing credentialing.

Methods: 354 H&N phantom irradiations (September 2011 – December 2016) with a range of total MU delivered (1162-3761) and segments used (16-1485) were evaluated using IROC Houston’s standard protocols for H&N phantom point-dose analysis. Irradiations were categorized by both delivery technique (step-and-shoot, dynamic, and VMAT) and machine class (Varian Base Class, Varian TrueBeam, Varian 2100, and Elekta Agility). Dose errors were determined as the percentage difference between the measured TLD doses (6 TLD per phantom) and their corresponding TPS estimations. Spearman’s rank-order correlation coefficients were determined for the study sample (excluding 6 outliers) as well as for each of the subgroups, where treatment plan complexity (measured by total MU delivered or number of segments) was compared with average or extreme (worst agreement) absolute dose errors.

Results: For this sample, treatment plan complexity and average dose inaccuracy showed no clear relationship and were not correlated statistically (p>0.63). Extreme dose errors were correlated with total segment number for only the total sample (p=0.05) and VMAT plans (p=0.01). All other subgroups and metrics showed negative results: no correlation coefficient exceeded 0.12, and none of these relationships were determined to be statistically significant (p>0.05).

Conclusion: The extent of disagreement between measured and calculated doses in the IROC Houston H&N phantom is not correlated with increasing complexity of the treatment plan. These results indicate that other factors such as dosimetric modeling inaccuracy or MLC performance should be investigated to determine a more likely cause for dose delivery errors.

Funding Support, Disclosures, and Conflict of Interest: This work was supported by Public Health Service Grant No. CA180803 awarded by the National Cancer Institute, United States Department of Health and Human Services.

Contact Email: