Elsevier

Physica Medica

Volume 46, February 2018, Pages 168-179
Physica Medica

Original paper
Computational analysis of interfractional anisotropic shape variations of the rectum in prostate cancer radiation therapy

https://doi.org/10.1016/j.ejmp.2017.12.019Get rights and content

Highlights

  • The shape variations of the rectum varies in every part of the rectum.

  • The shape variations of the overlap (ROP) region of rectum with PTV were analyzed.

  • The dose evaluation indices of the rectum were investigated.

  • The rectum V75 was smaller when using PRV-based plan.

Abstract

Purpose

To analyze the uncertainties of the rectum due to anisotropic shape variations by using a statistical point distribution model (PDM).

Materials and methods

The PDM was applied to the rectum contours that were delineated on planning computed tomography (CT) and cone-beam CT (CBCT) at 80 fractions of 11 patients. The standard deviations (SDs) of systematic and random errors of the shape variations of the whole rectum and the region in which the rectum overlapped with the PTV (ROP regions) were derived from the PDMs at all fractions of each patient. The systematic error was derived by using the PDMs of planning and average rectum surface determined from rectum surfaces at all fractions, while the random error was derived by using a PDM-based covariance matrix at all fractions of each patient.

Results

Regarding whole rectum, the population SDs were larger than 1.0 mm along all directions for random error, and along the anterior, superior, and inferior directions for systematic error. The deviation is largest along the superior and inferior directions for systematic and random errors, respectively. For ROP regions, the population SDs of systematic error were larger than 1.0 mm along the superior and inferior directions. The population SDs of random error for the ROP regions were larger than 1.0 mm except along the right and posterior directions.

Conclusions

The anisotropic shape variations of the rectum, especially in the ROP regions, should be considered when determining a planning risk volume (PRV) margins for the rectum associated with the acute toxicities.

Introduction

Prostate cancer was ranked as the fifth leading cause of death from cancer for men worldwide in 2012 [1]. Incidence rates are increasing every year in the developed countries such as United Kingdom and Japan [2]. Several options are available to treat the prostate cancer including radiation therapy which allowed the prostate to be treated with high dose of radiation while sparring surrounding normal tissues [3].

The quality of radiation therapy in prostate cancer treatment is affected by high dose regions which could be induced by patient movement, internal motion of the organ, and patient set-up errors [4], [5]. Fig. 1 illustrates the anatomical regions of a rectum, bladder, and planning target volume (PTV) determined by radiation oncologists. Anterior parts of the rectum may overlap with the PTV due to large internal margins and/or rectal displacements as shown in Fig. 1. The rectal position uncertainties, which could cause toxicities (e.g., rectal bleeding, fecal incontinence), mainly comes from the rectal motion due to the changes in rectal filling [6], [7], [8], [9], [10]. The two common methods used to study the rectal motion were tracking the changes in rectal volume and evaluating the translation and rotation errors of the rectum [5], [11], [12], [13]. Fontenla et al. [14], however, noted that the more complex problem of internal organ motion involve changes in the shape (shape variations) of the organ especially along the anterior direction of the rectum [5], [15]. Therefore, the shape variations of the rectum, especially along the anterior direction, need to be investigated.

In order to dealt with the position uncertainties of the organs at risks (OARs), the International Commission on Radiation Units and Measurements (ICRU) reports no. 62 [16] and 83 [17] introduced the concept of planning risk volume (PRV) margins. In the case of prostate cancer radiation therapy, the use of PRV dose-volume histograms (DVHs) is recommended to predict acute rectal toxicity [15], [18], [19]. “Recipes” to determine the uniform PRV margins have been developed by McKenzie et al. and Stroom and Heijmen [20], [21]. However, the uniform PRV margins are inadequate to represent the actual rectal variations during treatment, as noted by McKenzie et al. [20] and Prabhakar et al. [22]. Therefore, an application of anisotropic PRV margins of the rectum should be considered.

There have been three studies that dealt with the shape variations of the rectum. Hoogeman et al. [23] analyzed the quantification of local rectal wall displacements by calculating local systematic and random errors of the rectum along three directions where they unfolded the outer surface of the delineated rectal wall and projected the 3-space coordinates of each surface element to a 2D map. Sohn et al. [24] investigated the correlated motion of adjacent organ structures between prostate, bladder and rectum which were parametrized by using sets of corresponding surface points and calculated the displacements between surface points at each fraction. They did not calculate the systematic and random errors that could be used in determining anisotropic PRV margins. Brierley et al. [25] investigated the determination of the PTV based on the rectal shape variations by using finite element modeling. They did not investigate the geometric errors related to the determination of PRV margins.

None of the previously mentioned studies, including ICRU, investigated directly the shape variations of the rectum along each anatomical direction separately (anterior, posterior, superior, inferior, left and right). The investigation along separate anatomical directions is indispensable for determining the anisotropic PRV margins. There have been also no studies on the systematic and random errors of the region in which the rectum overlapped with the PTV along the anterior wall (ROP regions), even though the shape variations of the ROP regions may cause the regions to be included in high dose distributions which can lead to rectum toxicities. Therefore, this study aims to investigate the anisotropic shape variations of the rectum and the ROP regions for prostate cancer radiation therapy along separate anatomical direction (anterior, posterior, superior, inferior, left and right).

Section snippets

Clinical study

This study was performed with the approval of the Institutional Review Board of our university hospital. The clinical data used in this study were obtained from 11 patients (range: 60–75 years; median age: 64 years; stage: T1-T3a, N0, M0), who had undergone intensity modulated radiation therapy (IMRT) for prostate cancer. The planning CT images were acquired from a CT scanner (Mx 8000, Philips, Amsterdam, Netherlands) with 512 × 512-pixel dimensions, 0.98 mm in-plane pixel size, and 2.0 mm

Results

Fig. 8 shows the population SDs of the systematic and random errors of the rectum due to shape variations along each anatomical direction of all patients. The population SDs for systematic errors were 0.6 mm along the left direction, 0.3 mm along the right direction, 1.0 mm along the anterior direction, 0.7 mm along the posterior direction, 2.1 mm along the inferior direction and 2.4 mm along the superior direction. The population SDs for random errors were 1.2 mm along the left direction,

Discussion

The deviation along the superior, inferior, and anterior directions were dominant for systematic and random errors. Brierley et al. and Nuyttens et al. [25], [47] noted similar observations of large deviations along superior and inferior directions of the rectum. The deviations were affected largely by the variabilities of other organ proximal to the rectum such as small bowel [47].

Fig. 11, Fig. 12 illustrate the SDs of the local systematic and random errors visualized on the reference rectum

Conclusions

An analysis of interfractional anisotropic rectum shape variations using a statistical PDM in a computational framework has been presented. The population SDs for the whole rectum calculated by the proposed method were larger than 1.0 mm along all directions for random errors, while for systematic errors the population SDs were smaller than 1.0 mm along the posterior, left, and right directions. The population SDs of systematic errors for ROP regions calculated by the proposed method were

Conflict of interest statement

None.

Acknowledgements

The authors would like to send the utmost gratitude to all members in Arimura laboratory, which had contribute a great deal of efforts in the performance of this study.

References (51)

  • E.J. Rijkhorst et al.

    Strategies for online organ motion correction for intensity-modulated radiotherapy of prostate cancer: prostate, rectum, and bladder dose effects

    Int J Radiat Oncol Biol Phys

    (2009)
  • L.P. Muren et al.

    Testing the new ICRU 62 “planning organ at risk volume” concept for the rectum

    Radiother Oncol

    (2005)
  • A. McKenzie et al.

    Margins for geometric uncertainty around organs at risk in radiotherapy

    Radiother Oncol

    (2002)
  • J.C. Stroom et al.

    Limitations of the planning organ at risk volume (PRV) concept

    Int J Radiat Oncol

    (2006)
  • R. Prabhakar et al.

    A study on planning organ at risk volume for the rectum using cone beam computed tomography in the treatment of prostate cancer

    Med Dosim

    (2014)
  • M.S. Hoogeman et al.

    Quantification of local rectal wall displacements by virtual rectum unfolding

    Radiother Oncol

    (2004)
  • J.D. Brierley et al.

    Rectal motion in patients receiving preoperative radiotherapy for carcinoma of the rectum

    Int J Radiat Oncol Biol Phys

    (2011)
  • M.S. Hoogeman et al.

    A model to simulate day-to-day variations in rectum shape

    Int J Radiat Oncol Biol Phys

    (2002)
  • H. Chui et al.

    A new point matching algorithm for non-rigid registration

    Comput Vis Image Underst

    (2003)
  • J.C. Stroom et al.

    Geometrical uncertainties, radiotherapy planning margins, and the ICRU-62 report

    Radiother Oncol

    (2002)
  • M. Van Herk

    Errors and margins in radiotherapy

    Semin Radiat Oncol

    (2004)
  • J. Wu et al.

    Positioning errors and prostate motion during conformal prostate radiotherapy using on-line isocentre set-up verification and implanted prostate markers

    Radiother Oncol

    (2001)
  • B. Strbac et al.

    Evaluation of set-up errors in head and neck radiotherapy using electronic portalimaging

    Phys Medica

    (2013)
  • Y. Dzierma et al.

    Set-up errors and planning margins in planar and CBCT image-guided radiotherapy using three different imaging systems: a clinical study for prostate and head-and-neck cancer

    Phys Medica

    (2015)
  • E.H. Huang et al.

    Late rectal toxicity: dose-volume effects of conformal radiotherapy for prostate cancer

    Int J Radiat Oncol Biol Phys

    (2002)
  • Cited by (10)

    • Statistical deformation reconstruction using multi-organ shape features for pancreatic cancer localization

      2021, Medical Image Analysis
      Citation Excerpt :

      Unlike physics-based modeling (Fuerst et al., 2015; Nakao and Minato, 2010; Nakao et al., 2007), statistical modeling is a data-driven approach that does not explicitly describe the elasticity and physical conditions of organs. Image-based (Sotiras et al., 2013; Ruhaak et al., 2017), point-based (Haekal et al., 2018; Shibayama et al., 2017), or mesh-based (Rigaud et al., 2019; Magallon-Baro et al., 2019; Nakamura et al., 2019; Nakao et al., 2019) deformable registration techniques have been used to obtain point-to-point correspondence between two datasets. Specifically, respiratory motion (Jud et al., 2017; Fuerst et al., 2015; Ruhaak et al., 2017; Wilms et al., 2016) has mainly been investigated in the field of image-based lung modeling.

    • Effect of accounting for interfractional CTV shape variations in PTV margins on prostate cancer radiation treatment plans

      2018, Physica Medica
      Citation Excerpt :

      This study concluded that the interfractional CTV shape variations were not negligible for the determination of the PTV margins. Also, Haekal et al. analyzed the uncertainties of the rectum shape due to anisotropic variations using the PDMs [23]. For these regions, where the rectum overlapped with the PTV along the anterior wall, the SDs of systematic errors ranged from 0.8 to 1.9 mm, and the SDs of random errors ranged from 0.1 to 1.8 mm.

    • Rotationally resliced 3D prostate segmentation of MR images using Bhattacharyya similarity and active band theory

      2018, Physica Medica
      Citation Excerpt :

      Other methods based on ASM or AAM have also been developed [6,7]. These methods also work with medical images of other organs and tissues, such ascardiac MR images [8], lesions [9], and rectum contour [10]. Machine learning has also been applied to prostate segmentation, and usually segmented data are used to train some classifiers and to classify the images as prostate or background and then cluster the image patch.

    View all citing articles on Scopus
    View full text