In regard to "Setup error analysis in helical tomotherapy based image-guided radiation therapy treatments".

Sir,

I read with interest the paper of Dr. Thondykandy et al. in the latest issue of your journal.[1] The current rapid introduction of pretreatment (tomotherapy megavoltage computed tomography (CT) and cone-beam CT on Varian and Elekta linacs) and in-treatment (CyberKnife, ViewRay) image guidance (IG) needs a careful assessment of these technologies in order to establish their optimal usage. The authors investigated the position correction shifts for 102 patients treated with helical tomotherapy using megavoltage CT for matching the patient position of the day to the planning CT study. There is no information on how planning CT studies were obtained, but for patients with a significant target motion, the usage of fast helical CT studies for planning may be suboptimal for IG purpose, and untagged average studies should be recommended. Systematic and random errors were evaluated and used for treatment margin calculation as per van Herk et al.[2] However, the authors’ concluded that the clinical margins used in their hospital were adequate enough for the brain, head and neck, and lung cancer patients while being out of clinical margins for the pelvis and cervical spine injury patients, may be confusing. The margins calculated by Thondykandy et al.[1] correspond only to a part of the total planning target volume (PTV) construction accounting for interfraction motion as explained in BIR publication.[3] These margin values can be used to account for positioning errors if no pretreatment IG is performed.

We need to establish comprehensive recommendations of the PTV margins that include disease site, planning CT parameters, immobilization devices, patient preparation protocols, imaging schedules (no-IG, IG for a limited number of fractions, daily IG), and patient-specific features. The latter is an important factor that can personalize margins. In the case of prostate cancer, Piotrowski et al. have shown that daily IG is needed for obese patients while the patients with body mass index <30 will benefit from IG procedures performed only during a few first treatment fractions.[4] A limited number of pretreatment IG sessions presents as an interesting compromise that allows for savings on labor, time, and imaging dose while taking into account specific information about target motion with the construction of personalized margins for the subsequent fractions without IG.[5]

Compiling a clearly defined instructions for IG procedures and corresponding PTV margins needs a multi-institutional collaboration (task group, workshop, or a special issue of J Med Phys) followed by a careful analysis of clinical outcomes for robust evaluation.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.