TURKISH JOURNAL OF ONCOLOGY

Summary

OBJECTIVE
This study is a comparison of contoured diagnostic images derived from computed tomography (CT) and magnetic resonance imaging (MRI) by both a radiation oncologist (RO) and a radiologist (R) using volumetric modulated arc therapy (VMAT) and intensity modulated radiation therapy (IMRT) techniques.

METHODS
CT and MRI sections of 16 patients were contoured by the RO and the R. Planning target volume (PTV) criteria assessed were conformity index (CI), homogeneity index (HI), volume covered by 98% isodose line (V 98%) and maximum dose (D_max). In critical organs, 40 Gy organ area volume (V₄₀), 65 Gy organ area volume (V₆₅), and D_mean criteria were evaluated. Paired samples t-test was used for statistical analysis.

RESULTS
PTV and critical organs were compared. MRI PTV and bladder volume drawn by R were lower. Comparison of CT images revealed IMRT plans were superior in terms of Dmax and CI, while V₄₀ and D_mean values for rectum and bladder were lower in MRI-based VMAT plans. In MRI plans, IMRT was superior in terms of PTV, D_max, CI, V₆₅, and D_mean for critical organs; however, critical organs were well preserved with both planning techniques.

CONCLUSION
There was some difference between contouring of the R and the RO, which was reflected in the treatment plans.

Summary

Introduction

It is well established that high-dose radical radiation therapy (RT) for localized prostate cancer improves disease control.[1?3] It is of the utmost importance that the RT planning process accurately defines gross tumor volume and organs at risk for successful patient management. Many centers register magnetic resonance imaging (MRI) to computed tomography (CT) usdata sets to take advantage of the superior soft-tissue contrast of MRI and electron density information of CT.[4] Changes in anatomical and tumor definition as a result of using MRI data compared with CT have been reported for prostate patients.[5?9] Studies comparing CT- versus MRI-derived volume found that in general, prostrate volume delineated on CT was approximately 1.3 times larger than MRI-derived volume.[10] This advantage of MRI is based on better soft-tissue visualization and availability of multiplanar image acquisition. As a consequence, MRI in addition to CT has been recommended for RT planning for the prostate.[9,11]

Successful RT depends on high geometric and dosimetric accuracy and precision. Intensity modulated radiation therapy (IMRT) has become the standard technique to deliver external beam RT treatment to the prostate due to its greater ability to deliver higher-dose treatment to the planning target volume (PTV) while reducing dose delivered to surrounding critical organs and healthy tissue.[12,13] A novel form of IMRT called volumetric modulated arc therapy (VMAT) is delivered using a cone beam that rotates around the patient.[14]

The aim of this study was to determine difference between organs and target volumes drawn using CT and MRI, and to evaluate interobserver variability between radiation oncologist (RO) and radiologist (R). A further goal of this research was to examine how differences in target volume calculated and affected RT planning.

Introduction

Methods

This study was approved by the Ethical Committee of Marmara University School of Medicine with 09.2015.314 protocol number. Sixteen patients with early stage prostate cancer treated in our clinic were included.

All patients were diagnosed with low-risk prostate cancer. Patients had empty rectum, drank 1 L of water, and waited half an hour to achieve full bladder prior to acquiring images. CT and MRI scans of cross-sectional area of 3.75 cm were taken at the same position for all patients. After these images were transferred to treatment planning system (Eclipse version11; Varian Medical Systems, Palo Alto, CA, USA), they were matched 3-dimensionally. After image pairing, prostate, rectum and bladder were independently contoured by the RO and the R. PTV was created in prostate volume with 0.5 cm posterior wall and 1 cm margin in all directions.

For each patient, in addition to VMAT (2 full arc area) technique used in our clinic, 7-field (51° interval) IMRT plans were created (Figure 1a,b). Six MV photon energy was used. Calculations for both planning techniques were made using same optimization criteria and analytical anisotropic algorithm. Calculation grid of 0.25 cm was selected. Total of 78 Gy doses in 39 fractions of 2 Gy were to be delivered to PTV in all plans and all plans were normalized in such a way that at least 95% of PTV would receive entire defined dose.

Fig 1: (a) VMAT 2 full arc field display. (b) IMRT 7-field display.

Dosimetry of VMAT and IMRT plans were assessed with respect to PTV and critical organs. For PTV, volume covered by 98% isodose line (V 98%) and maximum dose (D_max), conformity index (CI), and homogeneity index (HI) criteria were analyzed, and for critical organs, 40 Gy organ volume (V₄₀), 65 Gy area organ volume (V₆₅), and average dose (D_mean) criteria were examined.

For CI, the following equation was used: CI=TV² _PIV / TV x PIV. TVPIV represents volume of PTV within prescription isodose line, TV denotes volume of PTV (prostate volume), and PIV denotes volume encompassed by prescription isodose line. Optimal CI value is 1. CI greater than 1 indicates that volume of 98% isodose line is greater than PTV 98%. HI value was obtained using following equality: HI = (D_max-D_min)/D_rx. D_max is 1% of PTV dose, Dmin is 99% of PTV dose, and D_rx is the prescribed dose. HI value should be 0 for ideal treatment. To evaluate critical organ doses, 40 Gy organ area volume (V₄₀), 65 Gy organ area volume (V₆₅), and average organ dose (D_mean) values in the bladder and rectum were used. Dosimetric differences obtained were assessed using paired samples t-test and values below p<0.05 were considered statistically significant.

Methods

Results

The target volume and volume for critical organs obtained from the contours independently drawn on CT and MRI images by 2 different radiology specialists were compared. In the comparison of CT images, seminal vesicle drawn by the RO was larger (p=0.01). In comparison of MRI images, contours of PTV and bladder volume of RO were greater. When volumes drawn were compared in terms of technique, it was determined that drawings of both specialists had larger PTV and bladder volume on CT images (p=0.004, p<0.001, respectively) (Table 1). Statistically significant difference in bladder volume was likely due to length of time elapsed before MRI was performed.

Table 1: Mean volume value of contours generated by radiation oncologist and radiologist on computed tomography and magnetic resonance images

Dosimetric comparisons were first made between the 2 specialists and then between imaging techniques. Comparison of IMRT and VMAT plans, critical organs, and target tissues for each patient using contours drawn on CT images is presented in Table 2a.

Table 2a: Summary of dosimetric comparison of volumetric modulated arc therapy and intensity modulated radiation therapy plans on computed tomography images with contours drawn by radiation oncologist and radiologist

Both the R and the RO had better contour results in IMRT plans than VMAT plans according to D_max and CI.

When critical organs were evaluated, doses of V₄₀ and V₆₅ were lower when VMAT technique was used. IMRT plans were superior according to Dmax and CI in the contours plotted on MRI images, while the VMAT plans were superior to the V₄₀ and V₆₅ doses for rectum and bladder (Table 2b). When comparing the plans based on volumes defined by the R, no difference was found in target volume contour plotted on MRI using either technique, but better protection of critical organs was observed in MRI volumes (Table 3a).

Table 2b: Summary dosimetric comparison of volumetric modulated arc therapy and intensity modulated radiation therapy plans on magnetic resonance images with contours drawn by radiation oncologist and radiologist

Table 3a: Summary of dosimetric comparison of volumetric modulated arc therapy and intensity modulated radiation therapy plans on computed tomography and magnetic resonance images with contours drawn by radiologist

Similar results were obtained in the volume-based plan comparison of the RO (Table 3b).

Table 3b: Summary of dosimetric comparison of volumetric modulated arc therapy and intensity modulated radiation therapy plans on computed tomography and magnetic resonance images with contours drawn by radiation oncologist

Results

Discussion

Correct targeting of target tissue in RT is very important, both for tumor control and the protection of nearby healthy tissue. It is more likely that desired high doses can be delivered while reducing risk of complications during and after treatment with smaller, precise targets. Many studies have indicated that use of MRI in conjunction with CT to clinically identify the prostate and seminal vesicle is the gold standard. Villers et al. compared delineations of CT alone and CT with MRI in combination of 3 ROs and clinical target volume (CTV) plotted on CT were larger than CTV volumes plotted on CT with MRI.[15] In our study, when we compared volumes of the RO on CT and MRI, prostate volume was clearer and volume was smaller because MRI images provided more detail of the soft tissue.

We were careful to ensure that MRI and CT images were taken at the same position and within the same cross-sectional area in order to ensure that the images were recorded with the least possible amount of error. Hanvey et al. also emphasized the necessity of same MRI and CT position.[16] CT and MRI images taken at RT position were noted to have significantly smaller volume in the prostate seminal vesicle and bone structure recordings than MRI in diagnostic position (p=0.001).

It has been proven in many studies that advanced RT techniques in prostate radiotherapy are superior to conventional RT techniques in terms of target conformity and critical organ protection.[17,18] There have been many studies comparing advanced techniques. Fontenot et al. compared single-field VMAT technique and 7 to 9-field IMRT technique and no significant difference was found in terms of target conformity, target homogeneity, or critical organs (p>0.005).[19] In our study, it was observed that IMRT plans were superior to VMAT plans in terms of target conformation and target homogeneity. Both techniques yielded similar results for critical organs; however R had lower average contour volume, which yields better critical organ doses.

Chow et al. compared IMRT and VMAT plans in prostate patient suffering from weight loss and phantom and reported that VMAT results were superior and preferable to IMRT plans.[20] Elith et al. compared 5-field IMRT with single-field and double-field VMAT plans. It was determined that IMRT plans produced better results in terms of target homogeneity and VMAT plans were better in terms of target conformation and critical organ doses.[21]

There was a difference in the PTV volumes plotted by R and RO, and this difference was reflected in the treatment plans made; meanwhile both specialists had smaller PTV volumes based on MRI. Doses to critical organs were low. While IMRT plans are advantageous for the target dose conformation, critical organs were well preserved with both techniques. Both planning techniques were clinically relevant and results were consistent with the literature.

Disclosure Statement
The authors declare no conflicts of interest.

Discussion