Cancer Science & Therapy

Assessing the Dosimetric Consequence of Inter-fractional Setup Shifts on Helical TomoTherapy Plans with Independent Dose Calculation

Abstract

Weihong He, Luis A Vazquez Q, Eric Dzintars and Nikos Papanikolaou

This work studied on dosimetric impact due to inter-fractional uncertainties for one hundred patients from five different treatment sites (30 prostate, 26 head & neck, 18 lung, 17 pelvis, and 9 brain patients) for Tomotherapy modality. Daily setup shifts were quantified and grouped into systematic (mean daily setup shifts) and random shifts (fractionbased shifts with corresponding systematic shift subtraction). Both systematic and random shifts were incorporated into in-house independent point dose calculation software, MU-Tomo, to separately evaluate the systematic and random dosimetric variations. Systematic dosimetric variations showed large dose deviation, with the largest difference at -10.02% compared to the planned dose and 3% standard deviation. Mean random dosimetric variations showed relatively small dose deviation with the largest at -5.65% compared to the planned dose and 1.9% standard deviation. Furthermore, different treatment sites were sorted into the head & neck and brain group, and the body group including lung, pelvis, and prostate cancers. According to ANOVA analyses, random dosimetric variations were found significantly different between patients treated at the same treatment site, while systematic dosimetric variations were significantly different between the head & neck and brain group and the body group. No significant differences were discovered among specific patients for systematic dosimetric variations, and no significant differences were observed within each of the two groups for random dosimetric variations. Dosimetric consequences are not significantly correlated with treatment fraction number according to the Pearson correlation analysis. By comparing doses without any shift against those with the random shift, overall dosimetric impacts to each patient were found to be very small with the mean value -0.0053% and standard deviation of 1.11%. Ninety-nine percentage of the averaged variation results were within 3.5%. This implies that overall dosimetric impact from random variations is small; instead, dosimetric impact is more affected by systematic shifts.