Volume effects in radiosurgical spinal cord dose tolerance: how small is too small?

• Published in: Journal of radiation oncology Vol. 8; no. 1; pp. 53 - 61
• Main Authors: Ma, Ting Martin, Emami, Bahman, Grimm, Jimm, Xue, Jinyu, Asbell, Sucha O., Kubicek, Gregory J., Lanciano, Rachelle, Welsh, James, Peng, Luke, Gui, Chengcheng, Das, Indra J., Goldman, Howard Warren, Brady, Luther W., Redmond, Kristin J., Kleinberg, Lawrence R.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2019; Springer Nature B.V
• Subjects: Cancer Research; Coefficient of variation; Histograms; Imaging; Mathematical analysis; Medicine; Medicine & Public Health; Oncology; Original Research; Parameter uncertainty; Radiation therapy; Radiology; Radiotherapy; Reproducibility; Spinal cord; Standard deviation; Surgery; Surgical Oncology; Dmax; Uncertainty; Stereotactic body radiation therapy (SBRT); Monte Carlo; D0.03 cc; Spine
• ISSN: 1948-7894; 1948-7908
• DOI: 10.1007/s13566-018-0371-6

Objective Spinal cord dose constraints are a critical feature for stereotactic body radiation therapy (SBRT). Spinal cord maximum point dose (Dmax) by Monte Carlo (MC) calculations is used as a critical cord tolerance limit for SBRT, but information is lacking about its reproducibility. This study examines uncertainty of MC dose calculations for small volumes in spine SBRT. Methods Seven consecutive spine radiosurgery cases were randomly selected to measure precision of the Dmax calculation in comparison to other volumes. Each plan was calculated five times using MC with a 2% uncertainty objective, and variabilities in dose-volume histogram (DVH) parameters across recalculations were evaluated with coefficient of variation (standard deviation divided by mean). The average ratio of D0.03 cc/Dmax was calculated across a larger series of 130 cases. Results The variability of Dmax was twice as high for D0.03 cc and five times as high for D1 cc across recalculations for the seven cases. For larger volumes, the variability was lower. The standard deviation of Dmax was 0.1959 Gy, compared to 0.0931 Gy, 0.0569 Gy, and 0.0364 Gy for D0.03 cc, D0.1 cc, and D1 cc, respectively. The average D0.03 cc/Dmax among 130 cases was 0.93. Conclusions Dmax has greater variability compared to D0.03 cc, D0.1 cc, and D1 cc, potentially creating risks when used for guidance for spinal cord. D0.03 cc may be an attractive alternative with higher reliability while its limits could be obtained by scaling the reported Dmax limit by a factor of 0.93. This may help guide treatment planning and aid in discovering true dose constraints for spine SBRT.