Generating amorphous target margins in radiation therapy to promote maximal target coverage with minimal target size

https://doi.org/10.1016/j.cmpb.2018.09.003Get rights and content

Abstract

Background and significance

This work provides proof-of-principle for two versions of a heuristic approach that automatically creates amorphous radiation therapy planning target volume (PTV) margins considering local effects of tumor shape and motion to ensure adequate voxel coverage with while striving to minimize PTV size. The resulting target thereby promotes disease control while minimizing the risk of normal tissue toxicity.

Methods

This work describes the mixed-PDF algorithm and the independent-PDF algorithm which generate amorphous margins around a radiation therapy target by incorporating user-defined models of target motion. Both algorithms were applied to example targets – one circular and one “cashew-shaped.” Target motion was modeled by four probability density functions applied to the target quadrants. The spatially variant motion model illustrates the application of the algorithms even with tissue deformation. Performance of the margins was evaluated in silico with respect to voxelized target coverage and PTV size, and was compared to conventional techniques: a threshold-based probabilistic technique and an (an)isotropic expansion technique. To demonstrate the algorithm's clinical utility, a lung cancer patient was analyzed retrospectively. For this case, 4D CT measurements were combined with setup uncertainty to compare the PTV from the mixed-PDF algorithm with a PTV equivalent to the one used clinically.

Results

For both targets, the mixed-PDF algorithm performed best, followed by the independent-PDF algorithm, the threshold algorithm, and lastly, the (an)isotropic algorithm. Superior coverage was always achieved by the amorphous margin algorithms for a given PTV size. Alternatively, the margin required for a particular level of coverage was always smaller (8–15%) when created with the amorphous algorithms. For the lung cancer patient, the mixed-PDF algorithm resulted in a PTV that was 13% smaller than the clinical PTV while still achieving ≥99.9% coverage.

Conclusions

The amorphous margin algorithms are better suited for the local effects of target shape and positional uncertainties than conventional margins. As a result, they provide superior target coverage with smaller PTVs, ensuring dose delivered to the target while decreasing the risk of normal tissue toxicity.

Introduction

Modern radiation therapy requires high geometric accuracy in treatment planning and delivery. Motion and positional uncertainties can compromise the desired dose distribution and must be addressed to deliver adequate dose to the treatment target and to minimize the risk of normal tissue toxicities. The most widely used technique for considering positional uncertainties is to add safety margins around the target. The standard sets of margins are those described by the International Commission on Radiation Units and Measurements (ICRU) [1], [2], [3]. The reports from the ICRU present the gross tumor volume (GTV) consisting of observable disease, along with expansions of the GTV: the clinical target volume (CTV) to account for subclinical disease, the internal target volume (ITV) to account for anatomic motion, and the planning target volume (PTV) to account for setup error and limited machine precision. The underlying principle is that by designing a treatment plan to deliver the desired dose to the PTV, the CTV will receive adequate dose regardless of the effects of positional uncertainties.

When implementing safety margins to address uncertainties, the challenge is determining the appropriate margin magnitude; too small and target coverage is compromised, too large and the PTV will include excessive amounts of normal tissue, increasing the risk of toxicity. Margin magnitude has been a major theme during the development of image-guided radiation therapy. Numerous reports have presented the observed motion of various tumor sites using a large number of imaging techniques [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]. In addition, several formulas have been created to calculate the necessary margins based on estimates of positional uncertainties [22], [23], [24]. A well-known approach is that presented by van Herk et al., which determines the theoretical CTV-to-PTV margin required to provide a minimum CTV dose of 95% of the prescription for 90% of patients [23].

While margins are ubiquitous in radiation therapy, their implementation typically remains limited to either isotropic or anisotropic morphological dilations along three orthogonal anatomic axes [5], [10], [15], [16], [18], [25]. Some have refined the precision of anisotropic margins through registration of selective anatomic regions-of-interest when estimating setup uncertainties [16], [19], [21], [26], [27]. Others have made efforts to provide more local estimates of motion on the voxel or sector level [28], [29], [30], [31]. Local estimates are subsequently averaged to create expansions in a few directions, likely to correspond with the functionality of current treatment planning systems. However, simple expansions in a few directions cannot account for spatial complexities in target shape or motion. As a result, any given margin will be suboptimal, compromising CTV coverage and/or normal tissue sparing.

Other efforts at local margins are rooted in statistical descriptions of target coverage. This can be achieved by thresholding a probabilistic description of the CTV [22], [27], [32], or by iterative evaluation of the margins based on dosimetric target coverage [33]. While these techniques provide the advantage of a statistical evaluation of coverage to the target as a whole, they do not preclude generating a margin that results in under-dosing a particular region of the target. Some prior work was focused on a voxelized measurement of target coverage. However, this retrospective analysis only considered margin expansions in 6 directions and its ability to generalize the observed margins to other patients may be limited [34].

The purpose of the work presented here was to describe and evaluate two alternative margin-generating algorithms. These algorithms are heuristic approaches that, unlike the above-described approaches to margins, are designed to be highly flexible, prospective methods for generating PTVs that ensure adequate target coverage. They automatically consider local characteristics of the shape of the target and an unrestricted model of target motion. The resulting PTVs provide the desired amount of coverage on a voxel level with a more minimal size. To illustrate the application and efficacy of these approaches, PTVs were generated for two example CTVs and a previously treated lung cancer patient. The PTVs were compared with, and consistently outperformed, probabilistic threshold-based margins and conventional (an)isotropic margins.

Section snippets

Amorphous PTV margin algorithms

Presented here are two algorithms used to generate amorphous PTV margins. The PTVs are amorphous in that their boundary is highly flexible to accommodate local effects of the CTV shape and motion. The two amorphous margin algorithms are denoted as the mixed-PDF and independent-PDF algorithms. Both are based on a user-defined model of CTV motion. The CTV motion model is comprised of a probability density function (PDF) for each CTV voxel that represents the relative likelihood that the voxel

Results

PTV margins were generated for the two example CTVs at various levels of desired target coverage using the four techniques described above (mixed-PDF, independent-PDF, threshold, and (an)isotropic margin techniques). The performance of these margin techniques is presented for CTV1 and CTV2 in Fig. 3. For all margin techniques, the minimum coverage provided to a CTV voxel increased with the size of the PTV, though this effect diminished with increasing PTV size. Also, for a given level of

Discussion

This work describes the mixed-PDF and independent-PDF amorphous margin-generating algorithms and demonstrates that the performance of these heuristic approaches was consistently better than margins created by thresholding the CTV probability distribution or by an (an)isotropic expansion of the CTV. For a given level of minimum CTV voxel coverage, the mixed-PDF margin achieved the coverage with the smallest PTV, followed by the independent-PDF margin, the threshold margin, and the (an)isotropic

Conclusion

Presented are heuristic PTV margin-generating algorithms that 1) are automatic with the inclusion of a model of motion and/or uncertainty, 2) are generalizable to motion/uncertainty models of all forms, including those describing non-rigid tissue deformation, 3) result in amorphous margins determined by local features of the CTV and of the motion/uncertainty model, and 4) ensure a minimum level of coverage to each CTV voxel with a PTV consistently smaller than those created using conventional

Conflict of interests

I have no real or perceived conflicts of interest to disclose.

Acknowledgments

I would like to thank Mich Price, Ph.D., Manuel Morales, Ph.D., and George Ding, Ph.D. for insightful comments regarding the preparation of this manuscript.

References (44)

  • W. Li et al.

    Setup reproducibility for thoracic and upper gastrointestinal radiation therapy: influence of immobilization method and on-line cone-beam CT guidance

    Med. Dosim.

    (2010)
  • A. Sandhu et al.

    Prostate bed localization with image-guided approach using on-board imaging: reporting acute toxicity and implications for radiation therapy planning following prostatectomy

    Radiother. Oncol

    (2008)
  • J.-J. Sonke et al.

    Frameless stereotactic body radiotherapy for lung cancer using four-dimensional cone beam CT guidance

    Int. J. Radiat. Oncol. Biol. Phys.

    (2009)
  • S. van Kranen et al.

    Setup uncertainties of anatomical sub-regions in head-and-neck cancer patients after offline CBCT guidance

    Int. J. Radiat. Oncol. Biol. Phys.

    (2009)
  • O.A. Zeidan et al.

    Evaluation of image-guidance protocols in the treatment of head and neck cancers

    Int. J. Radiat. Oncol. Biol. Phys.

    (2007)
  • L. Zhang et al.

    Multiple regions-of-interest analysis of setup uncertainties for head-and-neck cancer radiotherapy

    Int. J. Radiat. Oncol. Biol. Phys.

    (2006)
  • J.C. Stroom et al.

    Inclusion of geometrical uncertainties in radiotherapy treatment planning by means of coverage probability

    Int. J. Radiat. Oncol. Biol. Phys.

    (1999)
  • M. van Herk et al.

    The probability of correct target dosage: dose-population histograms for deriving treatment margins in radiotherapy

    Int. J. Radiat. Oncol. Biol. Phys.

    (2000)
  • M. van Herk

    Errors and margins in radiotherapy

    Semin. Radiat. Oncol.

    (2004)
  • J. Wang et al.

    The clinical feasibility and effect of online cone beam computer tomography-guided intensity-modulated radiotherapy for nasopharyngeal cancer

    Radiother. Oncol.

    (2009)
  • Y. Seppenwoolde et al.

    Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat, measured during radiotherapy

    Int. J. Radiat. Oncol. Biol. Phys.

    (2002)
  • G.J. Meijer et al.

    What CTV-to-PTV margins should be applied for prostate irradiation? Four-dimensional quantitative assessment using model-based deformable image registration techniques

    Int. J. Radiat. Oncol. Biol. Phys.

    (2008)
  • View full text