Effect of patient setup errors on simultaneously integrated boost head and neck IMRT treatment plans

Jeffrey V. Siebers, Paul J. Keall, Qiuwen Wu, Jeffrey F. Williamson, Rupert K. Schmidt-Ullrich

Research output: Contribution to journalArticlepeer-review

78 Scopus citations


Purpose: The purpose of this study is to determine dose delivery errors that could result from random and systematic setup errors for head-and-neck patients treated using the simultaneous integrated boost (SIB)-intensity- modulated radiation therapy (IMRT) technique. Methods and Materials: Twenty-four patients who participated in an intramural Phase I/II parotid-sparing IMRT dose-escalation protocol using the SIB treatment technique had their dose distributions reevaluated to assess the impact of random and systematic setup errors. The dosimetric effect of random setup error was simulated by convolving the two-dimensional fluence distribution of each beam with the random setup error probability density distribution. Random setup errors of σ = 1, 3, and 5 mm were simulated. Systematic setup errors were simulated by randomly shifting the patient isocenter along each of the three Cartesian axes, with each shift selected from a normal distribution. Systematic setup error distributions with Σ = 1.5 and 3.0 mm along each axis were simulated. Combined systematic and random setup errors were simulated for σ = Σ = 1.5 and 3.0 mm along each axis. For each dose calculation, the gross tumor volume (GTV) received by 98% of the volume (D98), clinical target volume (CTV) D90, nodes D90, cord D2, and parotid D50 and parotid mean dose were evaluated with respect to the plan used for treatment for the structure dose and for an effective planning target volume (PTV) with a 3-mm margin. Results: Simultaneous integrated boost-IMRT head-and-neck treatment plans were found to be less sensitive to random setup errors than to systematic setup errors. For random-only errors, errors exceeded 3% only when the random setup error σ exceeded 3 mm. Simulated systematic setup errors with Σ = 1.5 mm resulted in approximately 10% of plan having more than a 3% dose error, whereas a Σ = 3.0 mm resulted in half of the plans having more than a 3% dose error and 28% with a 5% dose error. Combined random and systematic dose errors with σ = Σ = 3.0 mm resulted in more than 50% of plans having at least a 3% dose error and 38% of the plans having at least a 5% dose error. Evaluation with respect to a 3-mm expanded PTV reduced the observed dose deviations greater than 5% for the σ = Σ = 3.0 mm simulations to 5.4% of the plans simulated. Conclusions: Head-and-neck SIB-IMRT dosimetric accuracy would benefit from methods to reduce patient systematic setup errors. When GTV, CTV, or nodal volumes are used for dose evaluation, plans simulated including the effects of random and systematic errors deviate substantially from the nominal plan. The use of PTVs for dose evaluation in the nominal plan improves agreement with evaluated GTV, CTV, and nodal dose values under simulated setup errors. PTV concepts should be used for SIB-IMRT head-and-neck squamous cell carcinoma patients, although the size of the margins may be less than those used with three-dimensional conformal radiation therapy.

Original languageEnglish
Pages (from-to)422-433
Number of pages12
JournalInternational Journal of Radiation Oncology Biology Physics
Issue number2
StatePublished - Oct 1 2005


  • Intensity-modulated radiation therapy
  • Setup errors
  • Simultaneous integrated boost (SIB)


Dive into the research topics of 'Effect of patient setup errors on simultaneously integrated boost head and neck IMRT treatment plans'. Together they form a unique fingerprint.

Cite this