Physics Contribution
Setup Uncertainties of Anatomical Sub-Regions in Head-and-Neck Cancer Patients After Offline CBCT Guidance

https://doi.org/10.1016/j.ijrobp.2008.11.035Get rights and content

Purpose

To quantify local geometrical uncertainties in anatomical sub-regions during radiotherapy for head-and-neck cancer patients.

Methods and Materials

Local setup accuracy was analyzed for 38 patients, who had received intensity-modulated radiotherapy and were regularly scanned during treatment with cone beam computed tomography (CBCT) for offline patient setup correction. In addition to the clinically used large region of interest (ROI), we defined eight ROIs in the planning CT that contained rigid bony structures: the mandible, larynx, jugular notch, occiput bone, vertebrae C1-C3, C3-C5, and C5-C7, and the vertebrae caudal of C7. By local rigid registration to successive CBCT scans, the local setup accuracy of each ROI was determined and compared with the overall setup error assessed with the large ROI. Deformations were distinguished from rigid body movements by expressing movement relative to a reference ROI (vertebrae C1-C3).

Results

The offline patient setup correction protocol using the large ROI resulted in residual systematic errors (1 SD) within 1.2 mm and random errors within 1.5 mm for each direction. Local setup errors were larger, ranging from 1.1 to 3.4 mm (systematic) and 1.3 to 2.5 mm (random). Systematic deformations ranged from 0.4 mm near the reference C1-C3 to 3.8 mm for the larynx. Random deformations ranged from 0.5 to 3.6 mm.

Conclusion

Head-and-neck cancer patients show considerable local setup variations, exceeding residual global patient setup uncertainty in an offline correction protocol. Current planning target volume margins may be inadequate to account for these uncertainties. We propose registration of multiple ROIs to drive correction protocols and adaptive radiotherapy to reduce the impact of local setup variations.

Introduction

The neck is a complex site for radiotherapy. Often the dose to the target is compromised by the nearby organs at risk (OAR). The introduction of intensity-modulated radiotherapy for this group of patients is highly beneficial (1). The steep dose gradients allow for a more conformal target dosage while optimally sparing OARs.

In practice, however, the benefits of conformal radiation techniques may be limited by the accuracy with which the setup of the patient during treatment delivery can be reproduced. Alignment accuracy depends on the skill and experience of physicians or technicians while at the same time skin marks have a limited ability in predicting internal anatomy. Furthermore, the neck is flexible and needs immobilization with masks, bite blocks, or vacuum pillows 2, 3, 4. Apart from patient setup accuracy, head-and-neck cancer patients show substantial anatomical changes associated with tumor regression and weight loss (5). Intrafraction organ motion introduces additional uncertainty (6).

As the position and shape of the tumor vary over the course of treatment, safety margins are applied to avoid target underdosage (7). A reduction in geometrical uncertainties allows smaller margins, enabling a higher deliverable dose to the target, a lower dose to OARs, and improves clinical outcome (8).

To reduce geometrical inaccuracies many institutes have implemented setup correction protocols based on electronic portal imaging devices 9, 10. Ideally, the position of the tumor and OARs are determined in these images and repositioned to their planned locations immediately before treatment. Because of poor image quality of portal images and overlapping anatomy, identification of the tumor and OARs is generally not possible and nearby bony structures are taken as a surrogate. This is justified by the assumption that movement of soft tissue is highly correlated to movement of nearby bony structures.

Patient setup protocols currently are based on the assumption of rigid body motion: the patient is considered as nondeformable and table shifts and optional rotations are used to correct patient setup errors. Under this assumption, an online patient setup protocol for interfractional movement will result in small margins. In practice, however, patients with head-and-neck cancer show changes in posture and anatomy, giving rise to residual geometrical uncertainties, for which larger margins should be used.

In our institute, isotropic margins of 5 mm are used for patients with head-and-neck cancer in combination with an offline setup correction protocol. Since 2004, cone beam computed tomography (CBCT) image guidance (Elekta Synergy, Elekta Oncology Systems Ltd, Crawley, UK) replaced portal imaging for setup verification. The volumetric imaging capacity of CBCT revealed frequent changes in anatomy and posture that were previously hard to detect with planar imaging. There is no general approach to cope with such changes: deformations are simply ignored or in severe cases an ad-hoc decision for replanning is made. This study aims at quantifying the occurrence and magnitude of such changes during treatment and to assess the consequences for our currently used margins.

Section snippets

Patient group

In this study, 38 patients with head-and-neck cancer were retrospectively selected irrespective of tumor stage in consecutive order. Patients were immobilized with a five-point thermoplastic fixation mask with shoulder fixation (Civco Medical Solutions, Kolona, USA). To maximize the distance between radiation field and the parotid glands, the head was tilted back as far as possible with the help of a patient-specific neck rest. All patients were given a knee support for stability and comfort.

Residual errors after correction protocol

The residual rigid body setup error assessed by registration of the clinically applied ROI is shown in Table 1. The group mean error of –0.5 mm in the anteroposterior (AP) direction was statistically significant (p = 0.020), but small compared with the systematic error. The group mean rotations over the left-right (LR) and CC axes were also statistically significant (p < 0.001 and p = 0.006). Based on 312 registrations, a total of 24 corrections were clinically applied; in the LR, CC, and AP

Local setup accuracy

Rigid registration of sub-regions of bony anatomy showed that local setup errors exceed the residual global patient setup errors in an offline correction protocol. Consequently, quantifying the overall setup accuracy overestimates the precision of radiotherapy in head-and-neck cancer patients.

The bony anatomy was registered with chamfer matching using segmented bone in the reference and localization scan. The clinical ROI containing multiple relevant structures underestimated the resulting

Conclusions

Head-and-neck cancer patients show large local setup variations, resulting from deformations, exceeding residual global patient setup uncertainty in an offline correction protocol, despite the use of immobilization devices. Margins based on global patient setup accuracy may therefore be unsafe. Both for correction strategies and margins design, deformations should be taken into account.

Cited by (150)

View all citing articles on Scopus

Conflict of Interest: None.

Supported by a grant from the Dutch Cancer Society (NKI:2005-3378).

View full text