Evaluation of geometrical uncertainties on localized prostate radiotherapy of patients with bilateral metallic hip prostheses using 3D-CRT,IMRT and VMAT: A planning study

Abstract

OBJECTIVE:

Since most radiation treatment plans are based on computed tomography (CT) images, which makes it difficult to define the targeted tumor volume located near a metal implant, this study aims to evaluate and compare three treatment plans in order to optimally reduce geometrical uncertainty in external radiation treatment of localized prostate cancer.

METHODS:

Experimental subjects were three prostate patients with bilateral hip prosthesis who had undergone radical radiotherapy. The treatment plans were five-field three-dimensional conformal radiation therapy (3D-CRT), fixed 5-field intensity-modulated radiation therapy (IMRT) using similar gantry angles, and single-arc volumetric modulated arc therapy (VMAT). The monitor units (MUs), dose volume histograms (DVHs), the dose indices of planning target volume (PTV), clinical target volume (CTV) and rectum were compared among the three techniques. The geometrical uncertainties were evaluated by shifting the iso-center (2– 10 mm in the anterior, posterior, left, right, superior, and inferior directions). The CTV and rectum dose indexes with and without the iso-center shifts were compared in each plan.

RESULTS:

The Conformity Index of PTV were 1.35 in 3D-CRT, 1.12 in IMRT, and 1.04 in VMAT, respectively. The rectum doses in 3D-CRT are also higher than those in IMRT and VMAT. The iso-center shift little affected the CTV dose when smaller than the margin size. The rectum dose increased especially after a posterior shift. Additionally, this dose increase was larger in the VMAT plan than in the 3D– CRT plan. However, the VMAT achieved a superior rectum DVH to that of 3D– CRT, and this effect clearly exceeded the rectum-dose increase elicited by the iso-center shift.

CONCLUSION:

For radiotherapy treatment of localized prostate cancer in patients with hip prosthesis, the dose distribution was better in the VMAT and Metal Artifact Reduction (MAR)–CT image methods than the conventional methods. Because the anatomical structure of the male pelvic region is relatively constant among individuals, we consider that VMAT is a valid treatment plan despite analyzing just three cases.

Keywords

Radiation therapy metal artifact reduction technique localized prostate cancer computed tomography Volumetric Modulated Arc Therapy

1 Introduction

Most recent radiation treatment plans are based on computed tomography (CT) images. However, if a metal implant is located near a tumor, the target volume is difficult to define. Metal artifact reduction (MAR) techniques can reduce the influence of metal artifacts and improve the accuracy of the target volume calculation [1 –6]. Among the MAR methods are single-energy projection on CT images [2, 3], and monochromatic images based on dual-energy CT [4, 5]. Figure 1 compares a CT image before and after applying a MAR technique. Note that the metal artifacts are significantly reduced by the MAR technique. Therefore, MAR techniques on CT images are useful for planning radiation treatment regimens and are expected to extend the use of CT images in future.

Fig. 1

Comparison of CT images before and after applying the MAR technique. Metal artifacts are significantly reduced by the MAR technique.

In conventional methods, any metal implants contained in the CT slices must be excluded from the radiation field by restricting the beam angle [7]. Additionally, the metal artifacts increase the difficulty of accurately calculating the dose distribution, which is usually performed without a heterogeneous correction. CT-image-based dose calculation using MAR techniques obtains more accurate dose distributions than conventional approaches. A technique called volumetric modulated arc therapy (VMAT) [8 –10] applies inverse planning without restricting the beam angles.

This study plans external radiation treatments of localized prostate cancer for patients with metallic hip prostheses. Prostate cancer, the most common cancer in men, can be effectively treated by high-dose radiotherapy [11 –13]. Prostheses are usually composed of high-density material such as titanium or steel, which (as mentioned above) degrades the accuracy of the dose distributed to localized prostate cancer. As the population ages, the number of patients presenting for radiotherapy with hip prostheses is expected to increase, hastening the need for accurate dose distribution in such cases. This problem might be solved by MAR and inverse planning techniques.

VMAT for localized prostate cancer of patients with metallic hip prostheses has been previously reported [14, 15]. However, no report has directly compared the dose volume histograms (DVHs) among VMAT, three-dimensional conformal radiation therapy (3D– CRT), and fixed multi-field intensity-modulated radiation therapy (IMRT) [9]. Additionally, because VMAT is a rotational irradiation method, the beam passes through the metal implants at some gantry angles. The dose distribution is optimized assuming a constant patient position. When treating localized prostate cancer in patients with metallic hip prostheses, changes in the patients’ positions might particularly affect the dose distribution. Therefore, the effects of geometrical uncertainties must be investigated. The geometric effect has been investigated in multi-field IMRT [16], but not in VMAT. To bridge this knowledge gap, we compare the dose distributions of three techniques— 3D-CRT, IMRT, and VMAT— using MAR– CT images. We also evaluate the geometrical uncertainties on the localized prostate radiotherapy of patients with metallic hip prostheses.

2 Materials and Methods

2.1 Creation of plans based on the 3D– CRT, IMRT, and VMAT techniques

The radiation treatment planning system was implemented in Varian Eclipse ver. 13.5. The linear accelerator was a Varian TrueBeam (10 MV X-ray) and Millennium 120 MLC type. The MAR– CT images were acquired by a GE Optima CT system [17]. The experimental subjects were three prostate patients who had undergone radical radiotherapy, and who presented with bilateral hip prosthesis. In this study, the patients with bilateral hip prosthesis were selected because they probably had the largest geometrical uncertainties on localized prostate radiotherapy. The hip prosthesis used were mainly made of titanium. The patients were classified into the risk groups according to the National Comprehensive Cancer Network (NCCN) criteria. There were 1 intermediate risk patient (T2c) and 2 high risk patients (both T3a) in this study. All procedures were approved by the Ethical Committee of our institution. The created plans were 5-field 3D– CRT (selecting gantry angles that avoided the hip prostheses), fixed 5-field IMRT (with similar gantry angles as 3D– CRT), and single-arc VMAT (with gantry angles of 179– 181° CCW). Generally, any metal implants must be excluded from the radiation field. Because the restricting of gantry angle was large in patients with bilateral metallic hip prostheses, dose distributions of IMRT and VMAT avoiding hip prostheses were expected to become the same. Additionally, geometrical uncertainties on localized prostate radiotherapy of patients with bilateral metallic hip prostheses have been not investigated in full-arc VMAT. Therefore, full-arc VMAT was used in this study. The doses were calculated by the AAA [18, 19] (3D– CRT and IMRT) and AcurosXB [20, 21] (VMAT) algorithms with heterogeneous correction. The target volume was determined based on the risk group. The clinical target volume (CTV) for the intermediate risk group was the prostate + seminal vesicle surrounding 1 cm of the prostate, for the high risk group was the prostate + seminal vesicle surrounding 2 cm of the prostate. The planning target volume (PTV) was defined by the CTV + an 8-mm margin (rectum side: 4 mm). Table 1 shows the patient characteristics. The hip prosthesis was contoured by an auto contour function equipped to the RTP. The prosthetic hip material in the RTP was assumed as titanium with its nominal density value (4.5 g/cm³). In the IMRT and VMAT plans, the dose distributions were optimized to satisfy the constraints on the doses applied to the PTV and organs at risk, which are defined in our in-house protocols. The dose constraints are listed in Table 2. In the 3D– CRT plans, the radiation fields were optimized to fit the PTV with a multi-leaf collimator (MLC) margin of 3 mm. The prescribed dose to 95% (D_95 %) of the PTV was 74 Gy in 37 fractions. In all examples, the dose-calculation grid size was set to 2 mm. The monitor units (MUs), DVH, and dose indices (PTV: D_98 %, D_mean, D_2 %, CI, CTV: D_98 %, D_mean, D_2 %, rectum: V_75Gy, V_70Gy, V_50Gy, V_30Gy) were compared among the three techniques.

Table 1
Patient characteristics

Patient 1 Patient 2 Patient 3

Age 76 61 82

T stage T2c T3a T3a

CTV volume 14.62 cm³ 25.22 cm³ 30.52 cm³

PTV volume 55.53 cm³ 85.54 cm³ 91.75 cm³

Bladder volume 40.90 cm³ 47.99 cm³ 103.93 cm³

	Patient 1	Patient 2	Patient 3
Age	76	61	82
T stage	T2c	T3a	T3a
CTV volume	14.62 cm³	25.22 cm³	30.52 cm³
PTV volume	55.53 cm³	85.54 cm³	91.75 cm³
Bladder volume	40.90 cm³	47.99 cm³	103.93 cm³

Table 2

Dose constraints applied in this study

Contour	Index	Optimal
PTV	D _95 %	=74 Gy
	Mean dose	74.74– 76.22 Gy
	D _max	<79.18 Gy
	V _98 %	>98%
Rectum	V _65Gy	<17%
	V _40Gy	<35%
	V _20Gy	<60%
Rectum wall	V _78Gy	<1%
	V _70Gy	<20%
	V _60Gy	<30%
	V _40Gy	<60%
Bladder	V _54Gy	<25%
	V _33Gy	<50%
Bladder wall	V _70Gy	<35%
	V _40Gy	<60%
Small bowel	V _60Gy	<0.5 cm³
Large bowel	V _65Gy	<0.5 cm³

2.2 Verification of geometrical uncertainties

The geometrical uncertainties were evaluated by shifting the iso-center in six directions (anterior, posterior, left, right, superior, and inferior). The shift was varied as 2, 4, 6, 8, and 10mm. The deviations of the does indices of CTV (D_98 %, D_mean, and D_2 %) and rectum (V_75Gy, V_70Gy, V_50Gy, and V_30Gy) between the plans with and without the iso-center shift were respectively calculated by the following formulas: $Deviations of CTV dose [%] = \frac{D_{shift} - D_{original}}{D_{original}},$ $Deviations of rectum dose [%] = V_{shift} - V_{original} .$

3 Results

Table 3 compares the total MUs among the three techniques. The MU was smallest in the 3D– CRT. Compared with 3D-CRT, MU was large in IMRT and VMAT because beam output efficiency was reduced due to complex MLC motion. The MU in VMAT was smaller than the MU of IMRT. These results are consistent with previous reports on external beam radiation treatment planning of prostate cancer without hip prostheses [22, 23]. In general, irradiation time is decreased with decrease of MU, and intra-fractional error can be reduced by short irradiation time. However, in evaluation of treatment plans, not only the irradiation time but also the dose distribution is important.

Table 3
Comparison of total MUs among the three techniques

3D-CRT IMRT VMAT

Total MU 290.4±5.5 854.8±63.4 525.2±13.9

	3D-CRT	IMRT	VMAT
Total MU	290.4±5.5	854.8±63.4	525.2±13.9

Panels (a), (b) and (c) of Fig. 2 show the average DVHs of the PTV, CTV, and rectum, respectively, in the 3D– CRT, IMRT, and VMAT plans. Table 4 compares the dose indices calculated by DVH of each technique. The CI of PTV were 1.35 in 3D-CRT, 1.12 in IMRT, and 1.04 in VMAT, respectively. The rectum dose V_75Gy, V_70Gy, V_60Gy, V_50Gy, V_40Gy, and V_30Gy were 16.59, 29.13, 40.82, 49.39, 57.24, and 61.88 in 3D-CRT, 1.07, 10.18, 21.34, 32.84, 43.04, and 53.32 in IMRT, 2.70, 7.64, 13.97, 20.14, 28.08, and 39.80 in VMAT, respectively. The VMAT delivered a better DVH than IMRT and 3D-CRT. Panels (a), (b) and (c) of Fig. 3 show comparison of dose distributions in iso-center plane among three techniques. The rectal dose of VMAT was lower than 3D-CRT and IMRT in visual as well as quantitative evaluation.

Fig. 2

Comparison of dose volume histograms among the various plans. DVH was better in VMAT than in IMRT and 3D-CRT.

Table 4

Comparison of dose indexes among the three techniques

	Index	3D-CRT	IMRT	VMAT
PTV	D _98 %	72.73±0.11	73.17±0.09	73.52±0.13
	D _mean	77.55±0.44	75.87±0.37	75.31±0.23
	D _2 %	79.76±0.67	77.12±0.54	77.15±0.34
	CI	1.35±0.10	1.12±0.04	1.04±0.02
CTV	D _98 %	76.70±1.09	75.02±0.23	74.16±0.17
	D _mean	78.71±0.33	76.20±0.62	75.13±0.37
	D _2 %	79.76±0.60	76.97±0.60	76.31±0.48
Rectum	V _75Gy	16.59±4.74	1.07±0.84	2.70±1.87
	V _70Gy	29.13±10.26	10.18±4.40	7.64±3.74
	V _60Gy	40.82±16.26	21.34±8.54	13.97±6.01
	V _50Gy	49.39±20.11	32.84±12.84	20.14±8.45
	V _40Gy	57.24±19.64	43.04±15.90	28.08±11.46
	V _30Gy	61.88±19.54	53.32±19.54	39.80±14.55

Fig. 3

Comparison of dose distributions among three techniques (e.g. pink: 100%, yellow: 90%, lemon green: 80%, green: 65%, and cyan: 50% iso-dose lines). In VMAT, the dose in posterior side of rectum was lower than 3D-CRT and IMRT visually.

Table 5 shows the deviations in the CTV dose indexes after the iso-center shifts. With 2 mm shift, all deviations were less than |0.5% |. When the iso-center shift was smaller than the margin size, its effect on the CTV dose was small (fundamentally,< |1.5 % |) regardless of technique. At shifts in the superior or inferior direction, the effect was slightly more pronounced in the 3D– CRT than in the other plans (With 6 mm shift,> |1.5% | deviations were found).

Table 5

Deviations of CTV dose indexes elicited by iso-center shifts. With 2 mm shift, all deviations were less than |0.5% |. Values described in left side of vertical line were the deviations when the iso-center shifted by less than margin size, and values described in right side were the deviations when the iso-center shifted by more than margin size

CTV	Anterior	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	– 0.4	– 0.6	– 1.1	– 1.7
	IMRT	– 0.7	– 1.5	– 3.4	– 7.3
	VMAT	– 0.2	– 1.7	– 6.8	– 15.2
D _mean	3D-CRT	– 0.1	– 0.3	– 0.4	– 0.6
	IMRT	0.4	0.4	0.3	0.0
	VMAT	0.1	0.0	– 0.5	– 1.5
D _2 %	3D-CRT	0.1	0.1	0.1	0.1
	IMRT	0.6	1.1	1.6	2.0
	VMAT	0.3	0.3	0.4	0.4
CTV	Posterior	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	0.2	0.1	– 0.1	– 0.3
	IMRT	– 0.6	– 1.0	– 1.3	– 2.0
	VMAT	– 0.2	– 0.4	– 0.7	– 1.5
D _mean	3D-CRT	0.0	0.0	– 0.1	– 0.3
	IMRT	– 0.4	– 0.6	– 0.9	– 1.2
	VMAT	– 0.1	– 0.2	– 0.3	– 0.5
D _2 %	3D-CRT	– 0.2	– 0.2	– 0.4	– 0.5
	IMRT	– 0.3	– 0.5	– 0.7	– 0.8
	VMAT	– 0.1	0.0	0.0	0.0
CTV	Left	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	0.0	– 0.3	– 0.7	– 1.3
	IMRT	– 0.3	– 0.4	– 0.6	– 0.9
	VMAT	– 0.2	– 0.2	– 0.3	– 0.5
D _mean	3D-CRT	– 0.1	– 0.2	– 0.3	– 0.5
	IMRT	– 0.1	– 0.1	– 0.2	– 0.3
	VMAT	– 0.1	– 0.2	– 0.2	– 0.2
D _2 %	3D-CRT	0.0	0.0	0.0	0.0
	IMRT	0.1	0.1	0.1	0.1
	VMAT	– 0.1	0.0	0.0	0.1
CTV	Right	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	– 0.3	– 0.7	– 1.2	– 2.0
	IMRT	0.2	0.2	0.2	0.2
	VMAT	0.2	0.2	0.2	0.1
D _mean	3D-CRT	– 0.1	– 0.2	– 0.3	– 0.5
	IMRT	0.1	0.2	0.3	0.4
	VMAT	0.2	0.3	0.5	0.5
D _2 %	3D-CRT	0.0	0.0	– 0.1	– 0.1
	IMRT	0.2	0.4	0.7	0.9
	VMAT	0.2	0.5	0.7	0.9
CTV	Superior	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	– 0.7	– 1.7	– 3.6	– 7.1
	IMRT	– 0.8	– 0.9	– 1.2	– 1.6
	VMAT	– 0.5	– 0.8	– 1.1	– 1.4
D _mean	3D-CRT	– 0.2	– 0.4	– 0.7	– 1.2
	IMRT	– 0.2	– 0.3	– 0.4	– 0.6
	VMAT	– 0.3	– 0.5	– 0.6	– 0.8
D _2 %	3D-CRT	– 0.1	– 0.2	– 0.3	– 0.4
	IMRT	– 0.2	– 0.2	– 0.2	– 0.3
	VMAT	– 0.2	– 0.2	– 0.1	– 0.2
CTV	Inferior	4 mm	6 mm	8 mm	10 mm
D _98 %	3D-CRT	– 1.1	– 2.1	– 3.6	– 6.3
	IMRT	0.2	0.0	– 0.1	– 0.4
	VMAT	0.3	0.4	0.5	0.4
D _mean	3D-CRT	– 0.1	– 0.3	– 0.5	– 1.0
	IMRT	0.1	0.2	0.2	0.1
	VMAT	0.5	0.8	1.1	1.4
D _2 %	3D-CRT	0.1	0.1	0.1	0.2
	IMRT	0.3	0.3	0.4	0.5
	VMAT	1.2	2.3	3.4	4.1

Table 6 shows the deviations in the rectal dose indexes after the iso-center shifts. The absolute value of the deviation in the rectum dose was increased with increase of shift from the iso-center. The rectum dose was increased most prominently by an iso-center shift in the posterior direction (Range of deviations with 10 mm shift; 3D-CRT: 5.8 to 16.6, IMRT: 4.5 to 20.4, and VMAT: 17.3 to 24.3), and was decreased by an iso-center shift in the anterior (3D-CRT: – 19.5 to – 10.5, IMRT: – 22.1 to – 1.1, and VMAT: – 24.7 to – 2.7) and superior (3D-CRT: – 7.8 to 0.9, IMRT: – 5.9 to – 0.7, and VMAT: – 4.8 to – 1.9) directions. An iso-center shift in the left, right, and inferior directions elicited a small response in the rectum dose. The rectum dose little deviated after a shift in the left, right, or inferior direction, and all deviations were smaller than |6.5% | when the iso-center shifted by less than 10 mm. The deviations in the rectum dose after a posterior shift were tended to be larger in the IMRT and VMAT plans than in 3D– CRT.

Table 6

Deviations of Rectal dose indexes elicited by iso-center shifts

Rectum	Anterior	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	– 2.9	– 8.1	– 12.2
	IMRT	– 0.5	– 1.0	– 1.1
	VMAT	– 1.7	– 2.7	– 2.7
V _70Gy	3D-CRT	– 3.8	– 11.5	– 18.2
	IMRT	– 2.9	– 7.4	– 9.9
	VMAT	– 3.2	– 7.1	– 7.6
V _50Gy	3D-CRT	– 3.5	– 11.4	– 19.5
	IMRT	– 4.6	– 13.7	– 22.1
	VMAT	– 4.9	– 13.2	– 18.8
V _30Gy	3D-CRT	– 1.6	– 5.5	– 10.5
	IMRT	– 3.8	– 12.3	– 21.8
	VMAT	– 4.8	– 14.7	– 24.7
Rectum	Posterior	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	3.0	8.6	13.7
	IMRT	0.7	2.6	4.5
	VMAT	2.3	9.0	17.3
V _70Gy	3D-CRT	3.7	10.7	16.6
	IMRT	3.2	10.3	17.6
	VMAT	3.8	13.0	23.0
V _50Gy	3D-CRT	3.0	7.6	11.1
	IMRT	4.5	13.0	20.4
	VMAT	5.0	15.1	24.3
V _30Gy	3D-CRT	1.4	3.9	5.8
	IMRT	3.4	8.9	12.5
	VMAT	4.5	12.5	19.2
Rectum	Left	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	– 0.2	– 1.5	– 3.9
	IMRT	– 0.1	– 0.1	– 0.1
	VMAT	0.0	– 0.2	0.0
V _70Gy	3D-CRT	– 0.1	– 2.0	– 5.5
	IMRT	– 0.2	– 0.9	– 2.3
	VMAT	0.1	0.1	– 0.1
V _50Gy	3D-CRT	– 0.4	– 2.6	– 6.3
	IMRT	0.4	1.5	2.8
	VMAT	0.4	1.5	2.1
V _30Gy	3D-CRT	– 0.1	– 0.1	0.2
	IMRT	0.2	1.4	3.3
	VMAT	0.5	2.5	3.8
Rectum	Right	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	– 0.2	– 1.6	– 4.0
	IMRT	0.1	0.2	0.5
	VMAT	– 0.1	– 0.4	– 0.7
V _70Gy	3D-CRT	– 0.4	– 2.6	– 6.3
	IMRT	0.0	– 0.5	– 1.7
	VMAT	– 0.2	– 0.6	– 1.3
V _50Gy	3D-CRT	– 0.1	– 2.1	– 5.9
	IMRT	– 0.2	0.1	0.9
	VMAT	– 0.2	– 0.3	0.0
V _30Gy	3D-CRT	0.2	0.8	1.6
	IMRT	0.1	1.1	3.1
	VMAT	– 0.1	0.7	2.6
Rectum	Superior	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	– 1.5	– 4.6	– 7.8
	IMRT	– 0.2	– 0.6	– 0.7
	VMAT	– 0.5	– 1.5	– 1.9
V _70Gy	3D-CRT	– 1.2	– 4.0	– 7.3
	IMRT	– 0.9	– 2.7	– 4.3
	VMAT	– 0.8	– 2.4	– 3.7
V _50Gy	3D-CRT	0.0	– 0.9	– 2.7
	IMRT	– 1.0	– 3.3	– 5.9
	VMAT	– 0.9	– 2.8	– 4.8
V _30Gy	3D-CRT	0.6	1.4	0.9
	IMRT	0.0	– 0.8	– 2.8
	VMAT	– 0.7	– 2.4	– 4.7
Rectum	Inferior	2 mm	6 mm	10 mm
V _75Gy	3D-CRT	1.4	3.5	4.3
	IMRT	0.3	1.2	2.3
	VMAT	0.7	2.1	3.6
V _70Gy	3D-CRT	1.0	2.5	3.1
	IMRT	1.0	3.1	4.9
	VMAT	0.8	2.6	4.0
V _50Gy	3D-CRT	– 0.2	– 1.2	– 2.9
	IMRT	0.8	2.1	2.7
	VMAT	0.9	2.4	3.5
V _30Gy	3D-CRT	– 0.8	– 2.9	– 5.7
	IMRT	– 0.2	– 1.1	– 2.8
	VMAT	0.5	1.0	0.7

4 Discussion

The VMAT plans required fewer and more MUs than the IMRT and 3D– CRT plans, respectively. Moreover, the VMAT achieved a better DVH than IMRT and 3D– CRT. These results are consistent with previous reports. Because the IMRT and VMAT techniques optimize the dose distributions by inverse planning, they can potentially improve the dose distribution over that of 3D– CRT. In prostate cancer patients without hip prosthesis, the DVHs of VMAT and IMRT are reportedly equivalent [22, 23]. However, in patients with hip prostheses, the gantry angle is limited to the regions not affected by the prosthesis, so the dose distribution might be worse in IMRT than in VMAT. In fixed multi-field IMRT, a greater proportion of the beam passes through the hip prosthesis than in VMAT, so the iso-center shift exerted a larger effect in IMRT than in VMAT. Mirroring the MUs, the irradiation time was smaller in the VMAT plan than in the IMRT plan. VMAT offered improved throughput and a reduced intra-fractional error compared with IMRT. In addition, the dose distribution and DVH were better in VMAT than in IMRT. The MU of VMAT was larger than the MU of 3D-CRT, however, dose distribution and DVH were greatly improved.

When the iso-center shift was smaller than the margin size, its influence on the CTV dose was small in all techniques (even in VMAT, where the beam partly passes through the hip prostheses). At shifts above the margin size, the CTV dose tended to decrease because the high-dose region was optimized to fit the PTV. Like the conventional method, appropriate margin setting, and image guidance in VMAT ensures that the delivered dose approximates the target dose. The iso-center shift in the superior or inferior direction affected a slightly higher influence on the CTV dose in 3D– CRT than in the other plans, probably because the MLC margin (3 mm) was set smaller than the MLC width (5 mm).

The influences of the iso-center shift on rectum dose were similar in all techniques. The larger the shift, the larger was the absolute value of the dose deviation. The rectum dose little deviated after a shift in the left, right, or inferior direction. The rectum dose was greatly reduced due to leave from high dose region by a shift in the anterior direction. Similarly, as the upper rectum is often bent backward, the rectum dose was decreased by a shift in the superior direction. The dose increase after a shift in the posterior direction was larger in the VMAT plan than in the 3D– CRT plan. However, as evidenced in Tables 4 and 6, the VMAT achieved a superior rectum DVH to that of 3D– CRT, and this effect clearly exceeded the rectum-dose increase elicited by the iso-center shift. For example, rectum V_75Gy of 3D-CRT and VMAT were 16.59% and 2.70% whereas the deviations of rectum V_75Gy by a 10 mm shift in the posterior direction were 13.7% and 17.3%, respectively. And, rectum V_30Gy of 3D-CRT and VMAT were 61.88% and 39.80% whereas the deviations of rectum V_30Gy by a 10 mm shift in the posterior direction were 5.8% and 19.2%, respectively.

From above, MU and irradiation time of VMAT were small than these of IMRT. In addition, compared with 3D-CRT, the VMAT plan greatly improved the dose distribution beyond the increase in geometrical uncertainties. Thus, the VMAT technique is effective for treating localized prostate cancer in patients with hip prosthesis.

The present study is limited by the small sample size (only three cases). As few applicable patients are available for analysis, increasing the number of eligible cases is not immediately feasible. However, the anatomical structure of the male pelvic region is relatively constant among individuals, and radiotherapy of localized prostate cancer targets the whole pelvis volume. Given the small variation between patients, we trust the utility of VMAT even after analyzing just three cases. Additionally, bladder was not discussed in this study because 2 of 3 cases had bladder volume smaller than recommended values [24]. Rectum is more important than bladder for localized prostate radiotherapy, however, further investigation is required.

5 Conclusion

In external radiation treatment planning of localized prostate cancer in patients with hip prosthesis, VMAT, and MAR– CT image methods achieved better dose distributions than the conventional methods.

Conflict of interest

The authors have no financial relationships to disclose.

References

Giantsoudi

, De Man

, Verburg

, et al., Metal artifacts in computed tomography for radiation therapy planning: Dosimetric effects and impact of metal artifact reduction, Phys Med Biol 21; 62(8) (2017), R49–R80.

Teixeira

P.A.

, Meyer

J.B.

, Baumann

, et al., Total hip prosthesis CT with single-energy projection-based metallic artifact reduction: Impact on the visualization of specific periprosthetic soft tissue structures, Skeletal Radiol 43 (2014), 1237–1246.

Jeong

, Kim

S.H.

, Hwang

E.J.

, et al., Usefulness of a metal artifact reduction algorithm for orthopedic implants in abdominal CT: Phantom and clinical study results, Am J Roentgenol 204 (2015), 307–317.

Wang

, Qian

, Li

, et al., Metal artifacts reduction using monochromatic images from spectral CT: Evaluation of pedicle screws in patients with scoliosis, Eur J Radiol 82 (2013), e360–366.

, Leng

and McCollough

C.H.

, Dual-energy CT-based monochromatic imaging, Am J Roentgenol 199 (2012), S9–15.

Jeon

and Lee

C.O.

, A CT metal artifact reduction algorithm based on sonogram surgery, J Xray Sci Technol 26(3) (2018), 413–434.

Reft

, Alecu

, Das

I.J.

, et al., Dosimetric considerations for patients with HIP prostheses undergoing pelvic irradiation. Report of the AAPM Radiation Therapy Committee Task Group 63, Med Phys 30(6) (2003), 1162–1182.

Otto

, Volumetric modulated arc therapy: IMRT in a single gantry arc, Med Phys 35 (2008), 310–317.

Hatanaka

, Tamaki

, Endo

, et al., Utility of Smart Arc CDR for intensity-modulated radiation therapy for prostate cancer, J Radiat Res 55(4) (2014), 774–779.

10.

Wali

, Helal

, Darwesh

and Attar

, A dosimetric comparison of volumetric modulated arc therapy (VMAT) and high dose rate brachytherapy (HDR) in localized cervical cancer radiotherapy, J Xray Sci Technol 27(3) (2019), 473–483.

11.

Viani

G.A.

, da Silva

L.G.

and Stefano

E.J.

, High-dose conformal radiotherapy reduces prostate cancer-specific mortality: Results of a meta-analysis, Int J Radiat Oncol Biol Phys 83 (2012), 619–625.

12.

Zelefsky

M.J.

, Fuks

, Hunt

, et al., High dose radiation delivered by intensity modulated conformal radiotherapy improves the outcome of localized prostate cancer, J Urol 166(3) (2001), 876–881.

13.

Pahlajani

, Ruth

K.J.

, Buyyounouski

M.K.

, et al., Radiotherapy doses of 80Gy and higher are associated with lower mortality in men with Gleason score 8 to 10 prostate cancer, Int J Radiat Oncol Biol Phys 82(5) (2012), 1949–1956.

14.

Nga

W.L.

, Brunt

, Temple

, et al., Volumetric modulated arc therapy in prostate cancer patients with metallic hip prostheses in a UK centre, Rep Pract Oncol Radiother 20(4) (2015), 273–277.

15.

Prabhakar

, Kumar

, Cheruliyil

, et al., Volumetric modulated arc therapy for prostate cancer patients with hip prosthesis, Rep Pract Oncol Radiother 18(4) (2013), 209–213.

16.

Hwang

A.B.

, Kinsey

and Xia

, Investigation of the dosimetric accuracy of the isocenter shifting method in prostate cancer patients with and without hip prostheses, Med Phys 36(11) (2009), 5221–5227.

17.

GE Healthcare. Smart metal artifact reduction (MAR). https://www.gehealthcare.com/products/computed-tomography/radiation-therapy-planning/metal-artifact-reduction

18.

Sievinen

, Ulmer

and Kaissl

, AAA photon dose calculation model in Eclipse, Palo Alto, CA: Varian Medical Systems 118 (2005), 2894.

19.

Esch

A.V.

, Tilikainen

, Pyykkonen

, et al., Testing of the analytical anisotropic algorithm for photon dose calculation, Med Phys 33 (2006), 4130–4148.

20.

Failla

G.A.

, Wareing

, Archambault

, Thompson

, Acuros XB advanced dose calculation for the Eclipse treatment planning system, Palo Alto CA: Varian Medical Systems. 2010.

21.

Vassiliev

O.N.

, Wareing

, McGhee

, et al., Validation of new grid based Boltzmann equation solver for dose calculation in radiotherapy with photon beam, Phys Med Biol 55 (2010), 581–598.

22.

Zhang

, Happersett

, Hunt

, et al., Volumetric modulated arc therapy: Planning and evaluation for prostate cancer cases, Int J Radiat Oncol Biol Phys 76 (2010), 1456–1462.

23.

Palma

, Vollans

, James

, et al., Volumetric modulated arc therapy for delivery of prostate radiotherapy: Comparison with intensity-modulated radiotherapy and three-dimensional conformal radiotherapy, Int J Radiat Oncol Biol Phys 72 (2008), 996–1001.

24.

Nakamura

, Shikama

, Takahashi

, et al., The relationship between the bladder volume and optimal treatment planning in definitive radiotherapy for localized prostate cancer, Acta Oncol 51(6) (2012), 730–734.