Abstract
Purpose: To compare planning target volume (PTV) defined by PET combined with 4DCT to 3DCT and 4DCT. Methods: Eighteen (18/30) esophageal cancer patients who underwent 3DCT, 4DCT and 18F-FDG PET-CT thoracic simulation with SUVmax≥2.0 of the primary volume were enrolled. CTV3D was formed on 3DCT by adding a margin of 30 mm in cranial-caudal direction and 5 mm in transversal direction. PTV3D was defined using a 10 mm margin to CTV3D and CTV4D was obtained by fusion of CTV from ten phases of 4DCT. A 5 mm margin for setup errors to CTV4D was to form PTV4D. BTVPET was generated with the assumption that motion was captured in PET images using a thresholding methods: 20% SUVmax. CTVPET 4DCT was calculated by the union of BTVPET and CTV4D, and a 5 mm margin to CTVPET 4DCT was used to form PTVPET 4DCT. The geometrical differences of the targets were evaluated. Results: Statistically significant differences were observed among CTV3D, CTV4D and CTVPET 4DCT (CTVPET 4DCT>CTV4D>CTV3D, P=0.000-0.038). PTV3D, PTV4D, and PTVPET 4DCT also differed significantly from each other (PTVPET 4DCT>PTV4D>PTV3D, P=0.000-0.048). The DI of PTV3D in PTVPET 4DCT was significantly larger than that of PTV3D in PTV 4D (P=0.042). There were no significant differences between the DI of PTV4D in PTV3D and PTVPET 4DCT in PTV3D (P=0.118). Conclusions: As demonstrated by the assessment of the geometrical differences in PET/4DCT-based and 3DCT-based PTV, PET/4DCT could affect not only the volume of PTV but also its shape.
Keywords: Esophageal cancer, 18F-FDG PET-CT, four-dimensional computed tomography, planning target volume
Introduction
Currently, radiotherapy is one of the most important treatment modalities for patients with esophageal cancer. Technologic advances in radiotherapy planning and delivery systems aimed at improving precision of target dosing and decreasing normal tissue exposure have significantly increased the importance of accurate target volume delineation.
Tumor motion is typically included in radiotherapy treatment planning (RTP) of primary thoracic esophageal cancer due to respiratory and cardiovascular motions [1,2]. To minimize geographic misses, the planning target volume (PTV) has to be used to account for setup variations and interfractional as well as intrafractional target motion, and it is this volume that would ideally receive the prescribed dose [3]. This is significant implications that a reduction in PTV could result in dose escalation to the target volume to achieve local tumor control and reduce the normal tissue complications probabilities.
It has been the standard practice for many years to use free-breathing (FB) to simulate and delineate gross target volume (GTV) with subsequent clinical target volume (CTV) and PTV for radiotherapy planning in esophageal cancer patients. Conventionally, a populational margin derived from the clinical experience or published margin guidelines may be used to account for the intrafraction tumor motion in the 3DCT based treatment planning. In this scenario, there could be unnecessary irradiation of normal tissues if tumor motion is smaller than expected. 4DCT is strongly recommended today in PTV definition in lung cancer patients as it could furnish individualized information on lesion motion and allows the use of smaller margins than free breathing-3DCT [4-6]. However, there have been few reports evaluated the use of 4DCT for PTV determination in esophageal cancer.
18F-FDG PET/CT has demonstrated significant value for radiotherapy of esophageal cancer. The incorporation of PET or PET/CT data into the radiation treatment planning has recently gained widespread acceptance as it has been shown to improve PTV definition [7,8]. However, the corresponding PET scan is often used without motion information for PTV definition. In fact, the PET scan is obtained during several breathing cycles, it represents a time-averaged map of the tumor position. Theoretically, high contrast between tumor and background in PET images means that even long acquisitions do not result in the tumor being lost in the background. Therefore, one potential limitation of the use of PET/CT in RTP is that the temporal mismatch between the short (few seconds) CT and the long (minutes) PET scans may generate a spatial misalignment between CT and PET data. If registration of a PET scan is to be used as a means of incorporating PET in RTP simulation, potential solution to overcome this issue is acquisition of the CT imaging with 4DCT information and subsequent registration of the PET scan to the 4DCT scan.
At present, we compare PTV defined on free-breathing (FB) 3DCT to PTV based on PET/4DCT and to evaluate if PET/4DCT could be useful to define the PTV for lesion motion in primary thoracic esophageal cancer patients.
Methods and materials
Patient selection and characteristics
The study was approved by the code of Ethics of the World Medical Association and informed consents were written by every patient before they were enrolled into the study. From November 2012 to February 2014, we enrolled 30 patients with histologically proven esophageal cancer who were candidates for radiotherapy. Excluded were patients with maximal standardized uptake value (SUVmax) on PET of less than 2.0. In total, the image data from 18 patients were available. The patient characteristics are listed in Table 1.
Table 1.
The characteristics of patients enrolled in the study
| Parameters | Number |
|---|---|
| Sex | |
| M | 14 |
| F | 4 |
| Age, median, y (range) | 44-80 (mean 65) |
| Tunor location | |
| Upper | 3 |
| Mid | 9 |
| Distal | 6 |
| SUVmax Pathological type | 7.33-30.46 (mean 10.67) squamous |
CT simulation and image acquisition
During the simulation, all patients were immobilized using thermoplastic mask. For each person, an axial contrast-enhanced 3DCT scan of the thoracic region was performed followed by a 4DCT scan under uncoached free breathing conditions on a 16-slice CT scanner (Philips Brilliance Bores CT). For 3DCT, each scan (360° rotation) took 1s to acquire followed by a 1.8 s dead time with a 2.4-cm coverage. The 3DCT scanning procedure takes about 30 s. During the 4DCT scanning, the respiratory signal was recorded with the Varian Real-time Positioning Management (RPM) gating system by tracking the trajectory of infrared markers placed on the patients’ abdomen. The signal was sent to the scanner to label a time tag on each CT image. GE Advantage 4D software sorts the reconstructed 4DCT images into ten respiratory phases on the basis of these tags, with 0% corresponding to end inhalation and 50% corresponding to end-exhalation. Both the 3DCT and 4DCT images were reconstructed using a thickness of 3 mm and then transferred to MIM software (MIM, 6.1.0, Cleveland, OH).
PET/CT image acquisition
The PET/CT simulation images were scanned on the same day after the 3DCT and 4DCT scans. All patients were asked to fast for at least 4 h before 18F-FDG PET/CT imaging. Each of them received 370 MBq (10 m Ci) of 18F-FDG intravenously 40 min before scanning and rested in a supine position in a quiet and dimly lit room. All images were acquired with an integrated PET/CT scanner. The patients lay on a flat table using thermoplastic mask with arms placed on the side of the body. The laser alignment lines were placed on the same position as the 3DCT, 4DCT infrared markers in order that the patients kept the same position with the 3DCT and 4DCT simulation scans. CT images (4.5 mm slices) were obtained during quiet breathing from the crown to mid-femur. With imaging times of 2 or 3 min per bed position, PET images were scatter corrected and reconstructed by use of an ordered-subsets expectation maximization algorithm with a post-reconstruction Gaussian filter at 5 mm full width at half maximum.
Image registration
The 3DCT, 4DCT and PET, CT images were imported into MIM. The CT images of PET-CT sets were firstly registered and fused with the 3DCT images using Maximum Mutual Information (MMI) method. The ten phases of 4DCT sets which were produced during the same imaging session with 3DCT and the PET images of PET-CT sets were automatically registered to the same coordinate system with the 3DCT and the CT of PET-CT images. And then further adjusted manually referring to the bony anatomic landmarks such as the sternum or the front edge of vertebral bony.
Tumor delineation and PTV definition
Both 4DCT and PET-CT target volumes were delineated using MIM. The gross target volume (GTV) was first delineated on 3DCT images (window width =400 HU, window level =40 HU) setting. The clinical target volume (CTV3D) based on 3DCT images was obtained by adding a 5 mm margin in left-right and anterior-posterior direction and a 30 m margin in cranial-caudal direction. Actually, the esophagus does not run strictly in super-inferior direction, the 30 mm margin in cranial-caudal direction for the GTV to form CTV in this article were manually contoured along the esophageal wall. PTV3D was defined by expanding the CTV3D with 10 mm margin to account for setup uncertainties and motion. GTV was contoured on the end-expiration phase CT (T50) from the 4DCT data, and then the contours were expanded to form the CTV50% (a 5 mm margin in left-right and anterior-posterior direction and a 30 mm margin in cranial-caudal direction). The CTV50% contours were then propagated to the other nine respiratory phases by using an intensity-based free-form deformable registration algorithm with essentially limitless degrees of freedom by using MIM software [9]. The fusion of the CTV from the ten phases of 4DCT was defined as CTV4D (ITV4D), and PTV4D was generated using a margin of 5 mm to CTV4D for setup errors. To define the PET imaged object, only one threshold setting methods (SUV≥20% SUVmax) with the assumption that motion was captured in the PET images were selected, based on recommendations from previously reported study [10] which sought to match PET for 4DCT-MIP target volumes in motive esophageal tumors. The method was one of optimal method for PET contours in target motion. In our previous work, we have also compared nine PET thresholding contours to target volumes combined by ten phases of 4DCT to determine an optimal segmentation method for primary thoracic esophageal cancer, and demonstrated that BTVPET (biological target volume) at SUV2.0, SUV20% correlated well with IGTV10 which combined from GTVs of 10 phases of 4DCT dataset in tumor VR (volume ratio), and CI (conformity index) (VR: 1.12±0.28, 0.90±0.48; CI: 0.54±0.10, 0.52±0.11). Therefore, we selected the threshold of 20% SUVmax for PET contour, and the contour was defined as BTVPET (ITVBTV) CTVPET 4DCT was the union of CTV4D and BTVPET. PTVPET 4DCT were generated using a margin of 5 mm for setup errors to CTVPET 4DCT. The target volumetric size, the centroid coordinates for CTV3D, CTV4D, CTVPET 4DCT, PTV3D, PTV4D and PTVPET could be generated by using the “Statistics Viewer” in MIM software.
Tumor motion
The center of mass (COM) coordinates for 4DCT-defined GTVs for each patient were measured. The peak-to-peak displacement of COM in the left-right (LR), anterior-posterior (AP) and cranial-caudal (CC) directions throughout 10 phases of 4DCT was obtained. Three dimensional tumor motion vector (motion vector) calculated as followed: Motion vector =(LR2+AP2+CC2)1/2.
PTVs comparison
Position, volume, and degree of inclusion (DI) among the PTVs were compared. PTVs positions are defined by center of target coordinates. The definition of DI of volume A included in volume B, [DI (A in B)] is the intersection between volume A and volume B divided by volume A. The formula is as followed:
DI (A in B)=A∩B/A5. Assumed volume B was reference for the standard volume. If the treatment planning was based on volume A, there would be [1-DI (A in B)] of volume A being unnecessary irradiated and [1-DI (B in A)] of volume B missing irradiation.
Statistical analysis
Statistical analysis was performed using the SPSS software package (SPSS 17.0). Mean ± standard deviation (Mean ± SD) was used to represent the quantitative parameters. A paired sample T test was used for the comparison of volume. The degree of associations between PTV values and motion vector was calculated by the Pearson text. Values of P<0.05 were regarded as significant.
Results
The displacement of clinical target volume (CTV) from 4DCT data
The maximum tumor (CTV) displacement in LR, AP, CC and three-dimensional directions from 4DCT are 1.15±0.63 mm, 1.67±0.92 mm, 6.94±3.64 mm, 7.30±3.67 mm, respectively, which are listed in Table 2. There were significant differences among the tumor displacement in LR, AP, CC directions (CC>AP>LR, P=0.000-0.004).
Table 2.
The maximum tumor (CTV) displacement in LR, AP, CC and three-dimensional directions from 4DCT (mm)
| Patient | LR | AP | CC | Motion vector |
|---|---|---|---|---|
| 1 | 0.70 | 2.00 | 11.20 | 11.40 |
| 2 | 0.90 | 1.00 | 4.00 | 4.22 |
| 3 | 2.21 | 1.30 | 3.30 | 4.18 |
| 4 | 1.00 | 1.10 | 5.50 | 5.70 |
| 5 | 2.17 | 3.60 | 9.80 | 10.66 |
| 6 | 2.30 | 3.40 | 16.00 | 16.52 |
| 7 | 0.50 | 1.10 | 4.40 | 4.56 |
| 8 | 0.40 | 0.80 | 3.70 | 3.81 |
| 9 | 0.50 | 0.80 | 9.40 | 9.45 |
| 10 | 0.84 | 1.20 | 4.20 | 4.45 |
| 11 | 1.33 | 1.60 | 11.80 | 11.98 |
| 12 | 0.70 | 1.30 | 5.80 | 5.98 |
| 13 | 1.20 | 2.00 | 6.00 | 6.44 |
| 14 | 0.68 | 2.40 | 3.80 | 4.55 |
| 15 | 0.70 | 1.00 | 4.00 | 4.18 |
| 16 | 2.01 | 3.30 | 9.20 | 9.98 |
| 17 | 1.20 | 0.90 | 3.90 | 4.18 |
| 18 | 1.30 | 1.20 | 9.00 | 9.17 |
| Mean ± SD | 1.15±0.63 | 1.67±0.92 | 6.94±3.64 | 7.30±3.67 |
The variations of target centroid position among PTV3D, PTV4D and PTVPET 4DCT
No significant difference was observed in centroid coordinates among PTV3D, PTV4D and PTVPET 4DCT in all directions (P=0.075-0.832).
The variations of volumes
The volume of CTV3D, CTV4D, CTVPET 4DCT, PTV3D, PTV4D and PTVPET 4DCT are listed in Table 3. Statistically significant differences were observed among CTV3D, CTV4D and CTVPET 4DCT (CTVPET 4DCT>CTV4D>CTV3D, P=0.000-0.038). PTV3D, PTV4D, and PTVPET 4DCT also differed significantly from each other (PTVPET 4DCT>PTV4D>PTV3D, P=0.000, -0.048). Compared to PTV4D and PTVPET 4DCT, PTV3D increase ranged from 17.30% to 55.81% (mean 41.35%) and from 19.53% to 55.83% (mean 38.00%), respectively. Meanwhile, PTVPET 4DCT increased ranged from -2.9% to 18.98% (mean 2.4%) when compared to PTV4D.
Table 3.
The volume of CTV3D, CTV4D, CTVPET 4DCT, PTV3D, PTV4D and PTVPET 4DCT (cm3)
| Patient | CTV (ml) | PTV (ml) | ||||
|---|---|---|---|---|---|---|
|
|
|
|||||
| CTV3D | CTV4D | CTV | PTV3D | PTV4D | PTV | |
| 1 | 49.78 | 69.11 | 69.69 | 203.54 | 151.81 | 152.79 |
| 2 | 107.84 | 130.19 | 130.25 | 346.3 | 246.56 | 247.74 |
| 3 | 47.23 | 49.68 | 53.2 | 207.54 | 113.56 | 120.48 |
| 4 | 68.85 | 85.36 | 87.41 | 261.83 | 175.63 | 185.53 |
| 5 | 101.81 | 136.49 | 143.44 | 329 | 263.82 | 280.48 |
| 6 | 97.08 | 144.72 | 144.18 | 333.33 | 278.87 | 270.83 |
| 7 | 83.52 | 98.93 | 99.24 | 279.47 | 196.12 | 198.54 |
| 8 | 78.52 | 89.13 | 101.07 | 283.77 | 184.39 | 219.39 |
| 9 | 72.85 | 96.52 | 96.6 | 278.99 | 201.3 | 201.56 |
| 10 | 64.28 | 82.49 | 83.22 | 271.11 | 185.08 | 188.56 |
| 11 | 119.2 | 151.45 | 152.7 | 370.98 | 301.98 | 304.43 |
| 12 | 33.37 | 41.4 | 41.4 | 145.23 | 94.64 | 94.64 |
| 13 | 66.2 | 84.8 | 85.47 | 260.01 | 178.58 | 180.74 |
| 14 | 92.38 | 117.61 | 118.04 | 336.28 | 249.08 | 249.59 |
| 15 | 57.86 | 79.94 | 79.95 | 249.52 | 160.12 | 160.14 |
| 16 | 48.88 | 72.89 | 72.91 | 198.19 | 153.61 | 153.61 |
| 17 | 135.19 | 158.62 | 159.18 | 402.22 | 288.09 | 296.01 |
| 18 | 99.33 | 129.13 | 130.15 | 326.98 | 249.76 | 250.85 |
| Mean ± SD | 79.12±27.53 | 101.03±34.65 | 102.67±34.71 | 282.46±66.79 | 204.06±60.09 | 208.66±60.43 |
The variations of DI
Table 4 showed the DI and (1-DI) among PTV3D, PTV4D and PTVPET 4DCT.
Table 4.
The DI and (1-DI) among PTV3D, PTV4D and PTVPET 4DCT (%)
| Patient | DI between PTV3D and PTV4D | DI between PTV3D and PTVPET 4DCT | DI between PTV4D and PTV | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
||||||||||
| DI of PTV3D in PTV4D | DI of PTV4D in PTV3D | 1-DI (PTV3D in PTV4D) | 1-DI (PTV4D in PTV3D) | DI of PTV3D in PTVPET 4DCT | DI of PTVPET 4DCT in PTV3D | 1-DI (PTV3D in PTVPET 4DCT) | 1-DI (PTVPET 4DCT in PTV3D) | DI of PTV4D in PTVPET 4DCT | DI of PTVPET 4DCT in PTV4D | 1-DI (PTV4D in PTVPET 4DCT) | 1-DI (PTV in PTV4D) | |
| 1 | 68.63 | 92.01 | 31.37 | 7.99 | 69.04 | 91.97 | 30.96 | 8.03 | 100 | 99.38 | 0 | 0.62 |
| 2 | 71.04 | 99.78 | 28.96 | 0.22 | 71.29 | 99.65 | 28.71 | 0.35 | 100 | 99.52 | 0 | 0.48 |
| 3 | 54.71 | 99.99 | 45.29 | 0.01 | 55.94 | 96.36 | 44.06 | 3.64 | 100 | 94.26 | 0 | 5.74 |
| 4 | 65.97 | 98.35 | 34.03 | 1.65 | 70.55 | 99.57 | 29.45 | 0.43 | 100 | 94.66 | 0 | 5.34 |
| 5 | 76.45 | 95.34 | 23.55 | 4.66 | 77.39 | 90.77 | 22.61 | 9.23 | 100 | 94.06 | 0 | 5.94 |
| 6 | 77.59 | 92.74 | 22.41 | 7.26 | 77.34 | 95.19 | 22.66 | 4.81 | 97.12 | 100.00 | 2.88 | 0.00 |
| 7 | 68.31 | 97.34 | 31.69 | 2.66 | 68.67 | 96.66 | 31.33 | 3.34 | 100 | 98.78 | 0 | 1.22 |
| 8 | 64.98 | 100.00 | 35.02 | 0.00 | 72.44 | 93.70 | 27.56 | 6.30 | 100 | 84.05 | 0 | 15.95 |
| 9 | 66.64 | 92.35 | 33.36 | 7.65 | 66.73 | 92.36 | 33.27 | 7.64 | 100 | 99.87 | 0 | 0.13 |
| 10 | 66.36 | 97.21 | 33.64 | 2.79 | 66.78 | 96.02 | 33.22 | 3.98 | 100 | 98.15 | 0 | 1.85 |
| 11 | 73.49 | 90.29 | 26.51 | 9.71 | 73.77 | 89.90 | 26.23 | 10.10 | 100 | 99.20 | 0 | 0.80 |
| 12 | 64.96 | 99.68 | 35.04 | 0.32 | 64.96 | 99.68 | 35.04 | 0.32 | 100 | 100.00 | 0 | 0.00 |
| 13 | 65.38 | 95.19 | 34.62 | 4.81 | 66.21 | 95.24 | 33.79 | 4.76 | 100 | 98.80 | 0 | 1.20 |
| 14 | 72.67 | 98.12 | 27.33 | 1.88 | 72.83 | 98.12 | 27.17 | 1.88 | 100 | 99.80 | 0 | 0.20 |
| 15 | 61.19 | 95.35 | 38.81 | 4.65 | 61.20 | 95.35 | 38.80 | 4.65 | 100 | 99.99 | 0 | 0.01 |
| 16 | 72.74 | 93.85 | 27.26 | 6.15 | 72.74 | 93.85 | 27.26 | 6.15 | 100 | 100.00 | 0 | 0.00 |
| 17 | 71.62 | 100.00 | 28.38 | 0.00 | 72.89 | 99.04 | 27.11 | 0.96 | 100 | 97.32 | 0 | 2.68 |
| 18 | 73.01 | 95.58 | 26.99 | 4.42 | 73.10 | 95.28 | 26.90 | 4.72 | 100 | 99.57 | 0 | 0.43 |
| Mean ± SD | 68.65±5.61 | 96.29±3.13 | 31.35±5.61 | 3.71±3.13 | 69.66±5.43 | 95.48±3.02 | 30.34±5.43 | 4.52±3.02 | 99.84±0.68 | 97.63±3.95 | 0.16±0.68 | 2.37±3.95 |
The DI of PTV3D in PTVPET 4DCT was significantly larger than that of PTV3D in PTV4D (P=0.042). There were no significant differences between the DI of PTV4D in PTV3D and PTVPET 4DCT in PTV3D (P=0.118).
Discussion
The 4DPET/CT techniques, by synchronizing PET and CT acquisition to the patients’ respiratory cycle, represents an innovative methodology for accurate imaging of tumors, particularly those located in the thorax and in the upper abdomen, where organ/lesion motion is relevant [11]. It seems to be a valuable tool in improving diagnostic performance of PET/CT and better defining the target volume for radiation therapy. However, Compared to a standard scan, 4DPET/CT technique requires more time for scanner and patient set-up, acquisition and processing of data. Its real benefit in routine clinical setting and its possible impact on patient management has not been established yet, especially for esophageal cancer. In this article, a PET combined with 4DCT was implemented instead of 4DPET/CT to evaluate the PTV in the definition of radiation treatment planning in esophageal cancer. To our knowledge, this is the first study that compared PTV obtained from 3DCT, 4DCT, with PTV defined by using PET/4DCT by tailoring the target volume to the lesion motion in esophageal cancer patients.
As expected, our preliminary data showed that CTV3D was significantly smaller than both CTV4D and CTVPET 4DCT (P=0.000, 0.038), and CTVPET 4DCT was the largest among the three PTVs (P=0.000). Obviously, this is because the CTV4D were from 4DCT images and included motion of the lesion during breathing, and CTVPET 4DCT were formed on both the 4DCT images and PET images which include not only the motion information of the tumor but also the biological information of the target. Similar result was obtained by Aristophanous and his colleagues [12].
Conversely, PTV3D was the largest among PTV3D, PTV4D, and PTVPET 4DCT (P=0.000, 0.000). Compared to PTV4D and PTVPET 4DCT, PTV3D increased ranged from 17.30% to 55.81% (41.35%) and from 19.53% to 55.83% (38.00%), respectively. These results are comparable with the previous studies who demonstrated a significant reduction of PTV by using 4DCT for radiation treatment planning for lung cancer [13,14] when compared with conventional PTV based on 3DCT. It is indicated that both the use of PTV based on 4DCT alone and 4DCT combined with PET for lesion motion could reduce the target being irradiated. However, PTV3D larger than PTV4D and PTVPET 4DCT does not imply that the former geometrically envelops the latter. The mean DI of PTV4D and PTVPET 4DCT in PTV3D were 96.29% and 95.48%, respectively. As a consequence, in these two cases if the conventional PTV3D was used in treatment planning, the possibility of missing the lesion during treatment is not available, and a larger part of normal tissue would be unnecessarily irradiated. Therefore, it would lead to a greater probability of local relapse and normal tissue toxicity. These results are comparable to a respiratory-gated PET/CT study for lung cancer by Guerra et al [15]. In their research, the PTV defined on respiratory-gated PET/CT were compared to PTV based on 4DCT and ungated 3DCT. Their results indicated that although the PTV obtained on ungated 3DCT was significant larger than respiratory-gated PET/CT-based and 4DCT-based PTV, the former could not entirely coverage the latter.
In this article, PTVPET 4DCT was significantly larger than that of PTV4D (P=0.048). Currently, these data can not be used to draw conclusion, but they could indicate that 4DCT combined with PET may add useful information to 4DCT for a better definition of the target volumes in some cases. The results of DI could further confirm our conclusion: although there were no significance between the DI of PTV4D in PTV3D and PTVPET 4DCT in PTV3D (P=0.118), the DI of PTV3D in PTVPET 4DCT were significantly higher than that of PTV4D in PTVPET 4DCT (P=0.042). These data suggested that assuming PTV3D was reference for the standard volume, if the treatment planning was based on PTV4D or PTVPET 4DCT, the volume percentages that PTV3D outside PTV4D was significantly higher than that of PTV3D outside PTVPET 4DCT. Guerra et al had also proved that when gated PET/CT information was added to gated CT, PTV changed in about 23% of cases, and RG PET/CT add additional information to RG-CT.
There were significant differences among the tumor displacement in LR, AP, CC directions (CC>AP>LR, P<0.05). These results may contribute to improving our knowledge of the esophageal clinical target tumor motion characteristics. The standard 3DCT-based CTV expansion (10 mm expansion in all directions to form PTV) could thus be inappropriate to correctly define the internal target volume of the lesion. These results are consistent to the study investigating the GTV motion characteristics form 4DCT data. Patel et al. [16] reported measurements of oesophageal tumour motion by 4D-CT for a series of 30 patients (1 in the proximal, 4 in the middle, and 25 in the distal), found that 2.2 mm in RL, 2.8 mm in AP, 8.0 mm in SI, respectively. The SI tumour motion was greater than the AP or RL motion. Our results were smaller than the studies above in the SI direction, which may be due to our cases mostly located in the upper and middle locations (6/18). Therefore, the intrafractional GTV and CTV centroid displacement was larger in SI direction during radiotherapy, and the use of asymmetric internal margins (GTV-to ITV; CTV-to-PTV) may be beneficial.
In a review by Bettinardi et al [17], they initially presented how the use of RG 4DPET/CT in PTV definition for radiotherapy treatment planning. It is suggested: (1) draw the BTV on each phase of 4DPET, and combine all the BTV to form ITVBTV which represents the volume of space encompassing the metabolic active part of the tumor, accounting for its motion during respiration (2) draw GTV on individual CT phases of 4DCT (3). Expand GTV to CTV for every phase (3) the convolution (Boolean union) of the resulting CTVs defines ITVCTV (4) ITVBTV and ITVCTV can then be combined (Boolean union) to obtain the “anatomical/functional” ITV (5) Expansions for set-up margins are added to ITV to generate PTV. In the present study, although PET/4DCT was used instead of RG 4DPET/CT for PTV construction, our results suggested that 4DCT combined with PET may add useful information to 4DCT for a better definition of the target volumes in some cases for esophageal cancer. It is believed that RG 4D PET/CT technologies, which is to produce “motion free” and well-matched PET and CT images corresponding to specific phases of the patient’s respiratory cycle, would be much beneficial for the target volume definition in radiation treatment planning.
In the present study, we relied on only one method for contouring of PET images with the assumption that motion was captured in the PET images. As we all know, it is the ratio or gradient between the background 18F-FDG uptake and tumor uptake that makes any method of PET-based contouring possible [18]. The size of PET delineation changes significantly depended on its threshold value. In order to encompass the whole tumor motion shape, the image threshold must be decreased. It is indicated that selection of an image threshold that is too low would overestimate the true volume, leading to a risk of increased normal tissue toxicity. Conversely, the selection of a threshold higher would tend to underestimate the extent of the “motion envelope”, leading to increased of geographic miss and diminished chance of cure.
Furthermore, there are other factors that could impact our results: (1) Despite the scans being done sequentially, with the patient in the same position throughout, the patient could move slightly between scans. (2) It is possible that the patients’ breathing pattern changed between the 3DCT, 4DCT and PET/CT scan acquisitions [19]. (3) The False-positive. However, these factors are not specific matters of investigation for the purpose of our study.
Conclusions
In conclusion, our data indicated that PET/4DCT could affect not only the volume of PTV but also its shape when compared with the PTV based on 3DCT and 4DCT. The use of PET/4DCT may be better for delineating PTV by tailoring the target volume to the lesion motion than 3DCT or 4DCT alone in primary esophageal cancer.
Acknowledgements
This study was accepted as a poster-viewing by American Society for Radiation Oncology (ASTRO) in 2015.
Disclosure of conflict of interest
None.
References
- 1.Hashimoto T, Shirato H, Kato M, Yamazaki K, Kurauchi N, Morikawa T, Shimizu S, Ahn YC, Akine Y, Miyasaka K. Real-time monitoring of a digestive tract marker to reduce adverse effects of moving organs at risk (OAR) in radiotherapy for thoracic and abdominal tumors. Int J Radiat Oncol Biol Phys. 2005;61:1559–1564. doi: 10.1016/j.ijrobp.2005.01.006. [DOI] [PubMed] [Google Scholar]
- 2.Yamashita H, Kida S, Sakumi A, Haga A, Ito S, Onoe T, Okuma K, Ino K, Akahane M, Ohtomo K, Nakagawa K. Four-dimensional measurement of the displacement of internal fiducial markers during 320-multislice computed tomography scanning of thoracic esophageal cancer. Int J Radiat Oncol Biol Phys. 2011;79:588–595. doi: 10.1016/j.ijrobp.2010.03.045. [DOI] [PubMed] [Google Scholar]
- 3.International Commission On Radiation Units and Measurements. Prescribing, Recording, and reporting photon beam therapy. Report 50. Washington, DC: ICRU; 1993. [Google Scholar]
- 4.Rietzel E, Liu AK, Doppke KP, Wolfgang JA, Chen AB, Chen GT, Choi NC. Design of 4D treatment planning target volumes. Int J Radiat Oncol Biol Phys. 2006;66:287–295. doi: 10.1016/j.ijrobp.2006.05.024. [DOI] [PubMed] [Google Scholar]
- 5.Li FX, Li JB, Zhang YJ, Liu TH, Tian SY, Xu M, Shang DP, Ma CS. Comparison of the planning target volume based on three-dimensional CT and four-dimensional CT images of non-smallcell lung cancer. Radiother Oncol. 2011;99:176–180. doi: 10.1016/j.radonc.2011.03.015. [DOI] [PubMed] [Google Scholar]
- 6.Wang W, Li JB, Zhang YJ, Li F, Xu M, Fan T, Shao Q, Shang D. Comparison of patient-specific internal gross tumor volume for radiation treatment of primary esophageal cancer based separately on three-dimensional and four-dimensional computed tomography images. Dis Esophagus. 2014;27:348–354. doi: 10.1111/dote.12089. [DOI] [PubMed] [Google Scholar]
- 7.Muijs CT, Beukema JC, Pruim J, Mul VE, Groen H, Plukker JT, Langendijk JA. A systematic review on the role of FDG-PET/CT in tumour delineation and radiotherapy planning in patients with esophageal cancer. Radiother Oncol. 2010;97:165–171. doi: 10.1016/j.radonc.2010.04.024. [DOI] [PubMed] [Google Scholar]
- 8.MacManus M, Nestle U, Rosenzweig KE, Carrio I, Messa C, Belohlavek O, Danna M, Inoue T, Deniaud-Alexandre E, Schipani S, Watanabe N, Dondi M, Jeremic B. Use of PET and PET/CT for radiation therapy planning: IAEA expert report 2006-2007. Radiother Oncol. 2009;91:85–94. doi: 10.1016/j.radonc.2008.11.008. [DOI] [PubMed] [Google Scholar]
- 9.Piper J. Evaluation of an intensity-based freeform deformable registration algorithm. Med Phys. 2007;34:2353–2354. [Google Scholar]
- 10.Guo Y, Li J, Wang W, Wang J, Duan Y, Shang D, Fu Z. Geometrical differences in target volumes based on 18 F-fluorodeoxyglucose positron emission tomography/computed tomography and four-dimensional computed tomography maximum intensity projection images of primary thoracic esophageal cancer. Dis Esophagus. 2014;27:744–750. doi: 10.1111/dote.12247. [DOI] [PubMed] [Google Scholar]
- 11.Nehmeh SA, Erdi YE, Pan T, Pevsner A, Rosenzweig KE, Yorke E, Mageras GS, Schoder H, Vernon P, Squire O, Mostafavi H, Larson SM, Humm L. Four-dimensional (4D) PET/CT imaging of the thorax. Med Phys. 2004;31:3179–3186. doi: 10.1118/1.1809778. [DOI] [PubMed] [Google Scholar]
- 12.Aristophanous M, Berbeco RI, Killoran JH, Yap JT, Sher DJ, Allen AM, Larson E, Chen AB. Clinical utility of 4D FDG-PET/CT scans in radiation treatment planning. Int J Radiat Oncol Biol Phys. 2012;82:e99–105. doi: 10.1016/j.ijrobp.2010.12.060. [DOI] [PubMed] [Google Scholar]
- 13.Hof H, Rhein B, Haering P, Kopp-Schneider A, Debus J, Herfarth K. 4D-CT-based target volume definition in stereotactic radiotherapy of lung tumours: comparison with a conventional technique using individual margins. Radiother Oncol. 2009;93:419–423. doi: 10.1016/j.radonc.2009.08.040. [DOI] [PubMed] [Google Scholar]
- 14.Rietzel E, Liu AK, Doppke KP, Wolfgang JA, Chen AB, Chen GT, Choi NC. Design of 4D treatment planning target volumes. Int J Radiat Oncol Biol Phys. 2006;66:287–295. doi: 10.1016/j.ijrobp.2006.05.024. [DOI] [PubMed] [Google Scholar]
- 15.Guerra L, Meregalli S, Zorz A, Niespolo R, De Ponti E, Elisei F, Morzenti S, Brenna S, Crespi A, Gardani G, Messa C. Comparative evaluation of CT-based and respiratory-gated PET/CT-based planning target volume (PTV) in the definition of radiation treatment planning in lung cancer: preliminary results. Eur J Nucl Med Mol Imaging. 2014;41:702–710. doi: 10.1007/s00259-013-2594-5. [DOI] [PubMed] [Google Scholar]
- 16.Patel AA, Wolfgang JA, Niemierko A, Hong TS, Yock T, Choi NC. Implications of respiratory motion as measured by four-dimensional computed tomography for radiation treatment planning of esophageal cancer. Med Phys. 2009;36:1610–1617. doi: 10.1016/j.ijrobp.2008.12.060. [DOI] [PubMed] [Google Scholar]
- 17.Bettinardi V, Picchio M, Di Muzio N, Gianolli L, Gilardi MC, Messa C. Detection and compensation of organ/lesion motion using 4D-PET/CT respiratory gated acquisition techniques. Radiother Oncol. 2010;96:311–316. doi: 10.1016/j.radonc.2010.07.014. [DOI] [PubMed] [Google Scholar]
- 18.Davis JB, Reiner B, Huser M, Burger C, Székely G, Ciernik IF. Assessment of 18F PET signals for automatic target volume definition in radiotherapy treatment planning. Radiother Oncol. 2006;80:43–50. doi: 10.1016/j.radonc.2006.07.006. [DOI] [PubMed] [Google Scholar]
- 19.Hugo G, Vargas C, Liang J, Kestin L, Wong JW, Yan D. Changes in the respiratory pattern during radiotherapy for cancer in the lung. Radiother Oncol. 2006;78:326–331. doi: 10.1016/j.radonc.2006.02.015. [DOI] [PubMed] [Google Scholar]