Discrepancies in volume calculations between different radiotherapy treatment planning systems

Abstract

It has been determined that, contrary to expectation, there is a clinically significant variation in the volume calculations of different RTPS (Radiotherapy Treatment Planning System) for identical contours. The situation was investigated prior to a multi-centre trial 1 to determine whether tumour volume is an independent prognostic factor in NSCLC (nonsmall cell lung cancer) 2 and included four of the commercially available RTPS. The four RTPS tested were, Theraplan Plus V3.0, Cadplan V6.2, Focus V2.6 and ADAC V3.0. Five randomly chosen clinical target related volumes (3 GTVs, one PTV and one CTV) from the trial database originally marked on the Cadplan system were transferred to the other four systems and the resulting volumes were calculated. It was found that Cadplan consistently underestimated the volume relative to the other three systems by 6-12%. This systematic underestimation was found to be caused by different assumptions made by the Cadplan system about the axial outer slice extension of the volume. Cadplan truncates the volume, while the other three systems extrapolate it by half the slice thickness at each end. A short program was written to apply the same method of volume extension to the Cadplan volume that is utilised by the other systems. This produced calculated volumes that were within r1.0% of the average of the volumes calculated by the other three planning systems, and the maximum deviation from the average for any planning system was then reduced to 1.5%. This program was implemented at all participating trial centres utilising Cadplan, thus reducing the intersystem variability to a negligible factor in comparison to the estimates of inter-physician variation 2 . This unexpected finding has significant implications for the validity of multi-centre trials using dose volume histograms, and indeed the adoption of any clinical protocol employing dose volume histogram constraints derived from experience at another centre employing a different RTPS.

A study on the tumor volume computation between different 3D treatment planning systems in radiotherapy

“…The information provided by the dose volume histogram is based on the definition of tumor and critical structures, and also on the accuracy of the calculation algorithms. The target volume definition is influenced by several parameters such as imaging modality, [9,10] interobserver variability, [11][12][13] patient movement/setup errors, [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] resolution limitation of the scanner [17] and contouring system. [6] In this study, we have assessed the variation of tumor volume between different treatment planning systems and also compared the differences in generating PTV from CTV with different planning systems.…”

“…Studies have shown that there is a significant difference between the volumes computed by the treatment planning systems. [5,6] Currently, Dicom objects are widely used in radiotherapy as a standard protocol for transferring images and RT datasets. We have multiple treatment planning systems in our department and the usual methodology is to contour the structures in one treatment planning system/contouring station and transfer the images to other treatment planning system depending on the workload.…”

Contouring variations and the role of atlas in non-small cell lung cancer radiation therapy: Analysis of a multi-institutional preclinical trial planning study

“…All institutions submitted the dry run planning results, but only the datasets with complete DICOM RT files (RT plan, structure, and dose) available at the time of analysis (this included eleven plans for Case1 and seven plans for Case2) were qualified for this study. The volumetric and dosimetric parameters for the same plan calculated by different software systems may differ [18]. Therefore, all submitted plans were imported into a single software system, MIMvista (Version 5.2.1; MIMvista Corp., Cleveland, OH), for consistent analysis.…”

Methods, software and datasets to verify DVH calculations against analytical values: Twenty years late(r)

“…The structure is automatically expanded from the boundary slices 1 /2 CT slice width inferiorly and superiorly. 10 This was verified by computing the volumes of axially aligned cylinders with different axial contour spacings. Other than that, the details of the DVH calculation algorithm are not documented in public domain.…”