Target volume delineation variation in radiotherapy for early stage rectal cancer in the Netherlands

Radiother Oncol. 2012 Jan;102(1):14-21. doi: 10.1016/j.radonc.2011.08.011. Epub 2011 Sep 6.

Abstract

Purpose: The aim of this study was to measure and improve the quality of target volume delineation by means of national consensus on target volume definition in early-stage rectal cancer.

Methods and materials: The CTV's for eight patients were delineated by 11 radiation oncologists in 10 institutes according to local guidelines (phase 1). After observer variation analysis a workshop was organized to establish delineation guidelines and a digital atlas, with which the same observers re-delineated the dataset (phase 2). Variation in volume, most caudal and cranial slice and local surface distance variation were analyzed.

Results: The average delineated CTV volume decreased from 620 to 460 cc (p<0.001) in phase 2. Variation in the caudal CTV border was reduced significantly from 1.8 to 1.2 cm SD (p=0.01), while it remained 0.7 cm SD for the cranial border. The local surface distance variation (cm SD) reduced from 1.02 to 0.74 for anterior, 0.63 to 0.54 for lateral, 0.33 to 0.25 for posterior and 1.22 to 0.46 for the sphincter region, respectively.

Conclusions: The large variation in target volume delineation could significantly be reduced by use of consensus guidelines and a digital delineation atlas. Despite the significant reduction there is still a need for further improvement.

Publication types

  • Multicenter Study

MeSH terms

  • Atlases as Topic
  • Female
  • Humans
  • Magnetic Resonance Imaging
  • Male
  • Netherlands
  • Patient Positioning
  • Practice Guidelines as Topic
  • Quality Assurance, Health Care
  • Radiation Oncology / standards*
  • Rectal Neoplasms / radiotherapy*
  • Reproducibility of Results
  • Retrospective Studies
  • Tomography, X-Ray Computed