Noncoplanar Volumetric Modulated Arc Therapy for Hepatocellular Carcinoma Based on a Cage-Like Radiotherapy System: A Simulation Study

Abstract

Background

The incorporation of noncoplanar beam arrangements has been proposed in liver radiotherapy modalities, which can reduce the dose in normal tissues compared to coplanar techniques. Noncoplanar radiotherapy techniques for hepatocellular carcinoma treatment based on the Linac design have a limited effective arc angle to avoid collisions.

Purpose

To propose a novel noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system and investigate its performance in hepatocellular carcinoma patients.

Methods

The computed tomography was deflected 90° to meet the structure of a cage-like radiotherapy system and design the noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system plan in the Pinnacle3 planning system. An noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system plan was customized for each of 10 included hepatocellular carcinoma patients, with 6 dual arcs ranging from −30° to 30°. Six couch angles were set with an interval of 36° and distributed along with the longest diameter of planning target volume. The dosimetric parameters of noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system plan were compared with the noncoplanar volumetric modulated arc therapy and volumetric modulated arc therapy plan.

Results

The 3 radiotherapy techniques regarding planning target volume were statistically different for D98%, D2%, conformity index, and homogeneity index with χ² = 9.692, 14.600, 8.600, and 12.600, and P = .008, .001, .014, and .002, respectively. Further multiple comparisons revealed that noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system significantly reduced the mean dose (P = .005) and V5 (P = .005) of the normal liver, the mean dose (P = .005) of the stomach, and V30 (P = .028) of the lung compared to noncoplanar volumetric modulated arc therapy. Noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system significantly reduced the mean dose (P = .005) and V5 (P = .005) of the normal liver, the mean dose (P = .017) of the spinal cord, V50 (P = .043) of the duodenum, the maximum dose (P = .007) of the esophagus, and V30 (P = .047) of the whole lung compared to volumetric modulated arc therapy. The results indicate that noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system protects the normal liver, stomach, and lung better than noncoplanar volumetric modulated arc therapy and protects the normal liver, spinal cord, duodenum, esophagus, and lung better than volumetric modulated arc therapy.

Conclusions

The noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system technique with the arrangement of noncoplanar arcs provided optimal dosimetric gains compared with noncoplanar volumetric modulated arc therapy and volumetric modulated arc therapy, except for the heart. Noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system should be considered in more clinically challenging cases.

Keywords

noncoplanar VMAT coplanar VMAT cage-like radiotherapy hepatocellular carcinoma

Introduction

Hepatocellular carcinoma (HCC), or liver cancer, is the sixth most common cancer worldwide and the third leading cause of cancer deaths in the world in 2020¹ according to a report by the International Agency for Research on Cancer.² Surgical resection is generally regarded as the primary treatment for HCC and has demonstrated effective results.³ However, for selected HCC patients with surgical contraindications and inoperable tumors, radiotherapy could be an effective treatment choice.⁴

With advances in radiotherapy modalities, 3-dimensional conformal radiotherapy can accurately irradiate the target volume while limiting radiation to the healthy liver, perhaps prolonging the lives of some HCC patients.^3,5 Intensity-modulated radiotherapy (IMRT) has the ability to generate complicated spatial dose distributions for HCC patients that conform more closely to the target volume while sparing critical organs by employing an inverse planning algorithm.^6–11 Volumetric modulated arc therapy (VMAT), the newest form of IMRT, can improve the time efficiency of dose delivery and deliver more conformal dose distributions to HCC patients by changing treatment apertures (dynamic multiple leaf collimators) as well as providing a modulated dose rate.^12–14 Recently, the incorporation of noncoplanar beam arrangements has been proposed in liver radiotherapy modalities, which can reduce the dose in normal tissues compared to coplanar techniques. de Pooter et al proved that IMRT with optimized noncoplanar configurations led to the best sparing of organs at risk (OAR) (healthy liver, spinal cord, and bowel).¹⁵ Hsieh et al proposed that noncoplanar IMRT was a potentially effective radiotherapy technique for HCC patients.¹⁶ Dong et al demonstrated that the noncoplanar radiation delivery technique significantly improved dose gradient, reduced high-dose spillage, and improved OAR sparing compared with state-of-the-art VMAT plans. Thibault et al demonstrated that a combination of coplanar full and noncoplanar partial arc in liver VMAT radiotherapy was sufficient for HCC patients.¹⁷ Sharfo et al demonstrated that noncoplanar VMAT (N-VMAT) could reduce the dose received by healthy liver volumes.¹² Lin et al¹⁸ considered that N-VMAT could improve OAR sparing more than VMAT, with only a small and acceptable increase in delivery time.¹⁹

However, to avoid collisions, noncoplanar radiotherapy techniques for HCC treatment based on the C-arm linear accelerator (C-Linac) design have a limited effective arc angle and significantly increased treatment time, which does not reflect the benefits of noncoplanar radiotherapy techniques. Niu et al designed a cage-like radiotherapy system (CRTS) to further promote the clinical application of noncoplanar radiotherapy.²⁰ The CRTS comprises 2 stands, 2 O-rings, several arc girders, an X-ray head, an imaging subsystem, and a treatment couch.²⁰ The X-ray head moves along with the arc girder and rotates using O-rings around the patient's body. In addition, the CRTS enabled precise noncoplanar irradiation without movement of the patient body, automated control of the motions of different parts without the risk of collisions, and continuous radiation across an angle far greater than a full turn.

This study aimed to design a novel noncoplanar radiotherapy technique based on the CRTS (N-VMAT-Cage) and investigate its performance in HCC patients. N-VMAT and VMAT plans based on the C-Linac were designed for comparison.

Materials and Methods

Patient Data

Ten patients with HCC (8 men and 2 women) were selected from our institution between July and November 2021. The prescription dose was 44 to 50 Gy. All plans were normalized to deliver 100% of the prescribed dose to 95% of the planning target volume (PTV). The PTV ranged from 533.13 to 2609.47 cc, and the healthy liver volume, defined as the total liver minus gross tumor volume (Liver-GTV), ranged from 506.18 to 1412.35 cc (see Table 1). The mean patient age was 55 (range, 47-76) years. Planning computed tomography (CT) data were acquired using a CT scanner (Sensation Open, Siemens Healthineers) using 1.3-mm thick slices. A 4-dimensional CT image set was acquired. Details of baseline characteristics of the patients and tumors are listed in Table 1. Our institution does not require ethical approval for reporting individual cases or case series. For this study, anonymized patient data were used. According to our local ethics committee, this does not require ethics approval. This study was a retrospective randomized study.

Table 1.

Patient Characteristics for the Treatment Techniques Being Studied.

No.	Age	Gender	Staging	Dose/fr	GTV (cc)	PTV (cc)	Healthy liver volume (cc)	Liver volume (cc)	Child-Pugh	ECOG
1	55	Man	IVB	50Gy/ 25	301.04	660.57	1122.50	1436.53	A5	1
2	49	Man	IIIB	48Gy/ 24	1500.10	2358.57	1203.86	2716.55	A5	1
3	76	Woman	IIIA	50Gy/ 25	691.54	1302.90	1307.12	2009.29	A5	1
4	73	Man	IIIC	46Gy/ 23	437.87	814.46	912.97	1331.47	A5	1
5	57	Woman	IIB	50Gy/ 25	206.96	544.88	1105.37	1325.14	A5	1
6	47	Man	IIIB	50Gy/ 25	198.92	533.13	1412.35	1203.36	A5	1
7	67	Man	IIIB	46Gy/ 23	1095.26	1885.70	947.59	2021.98	A5	1
8	67	Man	IIIB	50Gy/ 25	956.66	1611.59	844.05	1761.96	A5	1
9	54	Man	IIIB	50Gy/ 25	1856.00	2609.47	971.76	2751.85	A5	1
10	72	Man	IIIB	44Gy/ 22	541.79	980.28	506.18	1051.89	A5	1

Abbreviations: ECOG, Eastern Cooperative Oncology Group; PTV, planning target volume; GTV, gross tumor volume.

Treatment Planning

All plans were designed using the Pinnacle³ treatment planning system (version 16.2, Philips Healthcare) with Collapsed Cone Convolution Superposition algorithm and delivered using an Elekta Versa HD accelerator (Elekta Oncology Systems). All plans used 6-MV FFF X-rays and were calculated on a dose grid with 4-mm³ voxels using a collapsed cone convolution algorithm. The prescribed dose was 2 Gy delivered per fraction for all patients.

Noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system

The structure of CRTS is somewhat similar to the Tomotherapy machine (Accuray).²¹ The gantry of CRTS is mainly composed of several arc girders that stretch across the O-rings at −45° to 45° of rotation and rotate around the longitudinal axis that passes through the isocenter. The details of the structure of CRTS were shown in Appendix. In the Pinnacle³ TPS (version 16.3), a CT deflection of 90° was needed to meet the gantry of CRTS. Then, the N-VMAT-Cage plan was designed in TPS by rotating the couch and gantry. Deflection CT was mainly divided into 5 steps: (1) Export CT DICOM image from TPS; (2) Open the exported DICOM image using Image J software (Image J 1.50v, NIH) and use the “Transform J Rotate” command to deflect 90°; (3) Export the deflected image, with a raw format; (4) Use the program simulated in MATLAB (R2020b, Mathworks, Inc.) to convert the format to DICOM; and (5) Import DICOM images into the Pinnacle planning system, deflect CT images to re-register, replicate regions of interest, and design the N-VMAT-Cage plans. Moreover, the gantry of CRTS rotating around the longitudinal axis was determined based on the couch rotation (Linac). In this study, 6 couch angles with 60° arcs (6*60° = 360°) were designed to approximately simulate a full arc.

The design of the N-VMAT-Cage plan was primarily divided into 5 steps. Step 1, 90° CT deflection (realize the setting of the CRTS in the TPS) image was imported in the TPS to design an N-VMAT-Cage plan.; Step 2, the gantry angle was initially determined on the transverse slice, and the arc size was −30° to 30° to reduce high doses in OARs; Step 3, the couch angle was selected on the coronal slice, which was distributed along with the longest diameter of the PTV. Six couch angles were set with an interval of 36°, and the couch range was 180° ((6−1) × 36 = 180); Step 4, in the BRV view on the coronal slice (the orientation was Y1/Y2), the gantry angle (arc size was 60°) was finely adjusted to minimize exposure to the OAR (all arcs were dual arcs); Step 5, dosimetric parameters were optimized in the TPS.

Noncoplanar volumetric modulated arc therapy

The design of the N-VMAT plan was primarily divided into 2 steps. Step 1: set 6 couch angles of 30°, 60°, 90°, 330°, 330°, and 360° referring to Yuan et al²² to better compare with the N-VMAT-Cage, and Step 2: design arc size and ensure that the gantry angle can reach the maximum range without collision (all arcs are dual arcs). To determine the range of the gantry angle with different couch angles, the thermoplastic mask of an HCC patient was buckled on the couch plate of the Elekta Versa HD accelerator (Elekta Oncology Systems). The gantry was rotated at 6 couch angles. The range of the gantry angle without collision between the gantry, couch, or thermoplastic mask is shown in Table 2. All operations conformed to clinical practice.

Table 2.

Range of the Gantry Angle With 6 Couch Angles.

Beam	Couch	Gantry 1	Gantry 2	Orient
1	30.0°	78.1°	278.2°	ACW
2	60.0°	62.5°	298.2°	ACW
3	90.0°	60.5°	303.5°	ACW
4	300.0°	61.5°	297.2°	ACW
5	330.0°	78.5°	280.9°	ACW
6	360.0°	180.0°	180.1°	ACW

Abbreviation: ACW, anticlockwise.

Volumetric modulated arc therapy

The patients were clinically treated using VMAT plans with either dual coplanar full or dual coplanar partial arcs entering the body proximal to the tumor, depending on the tumor laterality. Five VMAT plans used full arcs, and others used partial arcs.

Planning Evaluation

Several parameters, such as conformity index (CI) and homogeneity index (HI), of PTV were evaluated. CI was based on Paddick's formula²³: $C I = \frac{T V_{P V}^{2}}{T V \times P V}$ , where TV is the PTV volume, PV is the volume covered by the prescribed dose, and $T V_{P V}$ is the volume of PTV covered by the prescribed dose. A CI value close to 1 represents better conformity. HI was defined as follows²⁴: $H I = \frac{D_{2 %} - D_{98 %}}{D_{50 %}}$ , where $D_{2 %}$ , $D_{98 %}$ , and $D_{50 %}$ are doses to 2%, 98%, and 95% of the target volume, and an HI value close to 1 indicates better homogeneity. V_95%, D_98%, D_2%, and D_50% of PTV were evaluated.

Regarding OARs, mean dose (D_mean), max dose (D_max), and volume of structure receiving 5 Gy (V₅(cc)) were calculated for the healthy liver (Liver-GTV). In addition, D_mean, D_max, and percentage of structure volume receiving 20 Gy (V₂₀(%)) for the left kidney (Kidney L) and right kidney (Kidney R); D_mean and D_max to the cord, cord PRV, and esophagus; D_mean and D_max to and V₅₀(cc) of the intestine, stomach, and duodenum; D_mean and D_max to and V₅₂(cc) of the colon; D_mean and D_max to and V₃₀ (%) and V₄₀ (%) of the heart; and D_mean and D_max to and V₂₀ (%), V₅ (%), and V₃₀ (%) of the entire lung (Lung All) were also evaluated.

Statistical Analysis

The data were analyzed using SPSS 25.0 software. The data from the radiotherapy techniques could not conform to both positive distribution and χ², so Friedman nonparametric test was used. The descriptive statistics were expressed as median (interquartile spacing) with P < .05. Then the multiple analysis used Wilcoxon paired signed-rank test, with Bonferroni correction P values. A significant level should be P < .0166 as the result with a statistical difference.

Results

Plan Comparison

Table 3 compares the dosimetric parameters of the 3 radiotherapy techniques about PTV and OAR parameters, and the descriptive statistics are presented as median and interquartile spacing. The 3 radiotherapy techniques regarding PTV were statistically different for D98%, D2%, CI and HI with χ² = 9.692, 14.600, 8.600, and 12.600 and P = .008, .001, .014, and .002, respectively. Further multiple analyses showed that the D98% of N-VMAT was significantly higher than N-VMAT-Cage (P = .009), and the D98% of N-VMAT was significantly higher than VMAT (P = .011). The D2% of N-VMAT-Cage was significantly higher than N-VMAT (P = .013). The D2% of VMAT was significantly higher than N-VMAT-Cage (P = .007). The D2% of VMAT was significantly higher than N-VMAT (P = .005). The CI of N-VMAT was significantly better than VMAT (P = .013). The HI of N-VMAT-Cage was significantly better than N-VMAT (P = .007). The HI of VMAT was better than N-VMAT (P = .005). All other parameters were not statistically different. Overall, the dose parameters of PTV with all 3 radiotherapy techniques meet the clinical requirements, but it might not be possible to clearly compare the advantages and disadvantages.

Table 3.

Summarizes the Dosimetric and Plan Quality Results for All Patients.

Parameters		N-VMAT-Cage			N-VMAT			VMAT			χ²	P
Parameters		Q1	Q2	Q3	Q1	Q2	Q3	Q1	Q2	Q3	χ²	P
PTV	V95% (%)	93.559	94.787	95.383	94.611	95.314	96.572	93.470	94.618	94.789	5.600	.061
	Mean dose	1.168	1.199	1.282	1.158	1.181	1.244	1.181	1.212	1.248	4.200	.122
	D98%	0.760	0.866	0.901	0.883	0.919	0.953	0.743	0.879	0.922	9.692	.008^a
	D2%	1.301	1.347	1.409	1.284	1.328	1.384	1.344	1.400	1.465	14.600	.001^a
	D50%	5652.750	5973.000	6478.250	5656.250	5981.500	6209.000	5781.250	5977.500	6432.750	3.436	.179
	CI	0.725	0.744	0.798	0.781	0.795	0.809	0.704	0.744	0.762	8.600	.014^a
	HI	0.340	0.421	0.483	0.307	0.344	0.376	0.365	0.441	0.563	12.600	.002^a
Liver-GTV	Mean dose (cGy)	1868.525	1963.700	2336.425	1999.225	2110.450	2442.525	1994.025	2087.700	2497.300	15.800	.000^a
	Max dose (cGy)	6530.225	6635.850	7010.750	6002.500	6349.200	6654.375	6360.325	6646.500	7264.100	7.800	.020
	V5 (cc)	405.255	587.866	654.579	464.140	647.774	702.373	431.340	637.673	691.810	15.000	.001^a
	Min dose (cGy)	60.100	61.000	73.300	54.800	68.800	111.900	42.600	81.300	123.900	1.143	.565
Kidney L	Mean dose (cGy)	386.150	620.250	905.925	484.050	614.600	807.925	463.225	717.850	913.350	2.400	.301
	Max dose (cGy)	1651.000	2315.700	2828.475	1547.075	2081.150	2828.775	1625.150	2740.800	2992.050	2.600	.273
	V20 (%)	0.000%	0.157%	3.250%	0.000%	0.353%	3.110%	0.000%	4.641%	6.850%	1.867	.393
	Min dose (cGy)	63.300	77.800	110.100	70.800	100.700	163.900	37.900	58.800	271.300	2.571	.276
Kidney R	Mean dose (cGy)	366.950	1024.800	1751.475	521.575	1550.850	1797.600	407.500	1180.100	1812.500	3.800	.150
	Max dose (cGy)	3793.025	5537.350	6051.425	3315.775	5486.100	6083.800	3392.550	5439.400	6048.325	1.400	.497
	V20 (%)	2.690%	17.641	29.850	2.973%	23.946	28.286	2.345	20.336	30.832	2.250	.325
	Min dose (cGy)	51.900	96.000	225.400	52.000	67.900	361.500	32.900	60.600	113.800	5.429	.066
Cord	Mean dose (cGy)	596.100	851.450	1199.925	520.225	939.350	1241.075	532.650	878.700	1244.525	1.400	.497
	Max dose (cGy)	2624.000	3607.350	3838.275	2465.775	3369.650	3718.625	2669.700	3576.400	3865.375	8.600	.014^a
Cord PRV	Mean dose (cGy)	561.025	830.400	1170.925	483.875	918.200	1188.000	497.825	852.500	1194.500	1.400	.497
	Max dose (cGy)	2960.175	3884.550	4276.150	2807.150	3757.150	4195.950	2931.800	3879.050	4216.675	2.600	.273
Intestine	Mean dose (cGy)	127.025	425.650	760.375	193.075	403.200	724.225	142.350	295.000	868.525	1.400	.497
	Max dose (cGy)	1452.975	2593.550	5381.875	1323.350	1962.250	5472.325	1351.825	2635.700	5495.550	2.400	.301
	V50 (cc)	0.000	0.000	0.599	0.000	0.000	1.605	0.000	0.000	2.749	3.500	.174
Colon	Mean dose (cGy)	392.375	604.850	1005.250	339.600	749.700	1225.025	283.050	652.950	1245.825	0.200	.905
	Max dose (cGy)	3245.150	5588.350	5786.975	3748.725	5464.300	5558.850	3832.700	5595.250	5729.600	3.200	.200
	V52 (cc)	0.000	0.648	2.052	0.000	0.268	2.187	0.000	0.658	3.328	1.852	.396
Stomach	Mean dose (cGy)	361.475	578.250	850.150	392.500	661.800	803.775	546.525	895.100	1687.275	15.000	.001^a
	Max dose (cGy)	4230.850	4755.450	5428.325	3487.175	4914.450	5263.425	3930.550	4871.400	5491.375	1.400	.497
	V50 (cc)	0.000	0.000	0.186	0.000	0.000	0.130	0.000	0.000	2.988	2.000	.368
Duodenum	Mean dose (cGy)	212.850	436.050	1627.475	280.375	811.500	1832.050	253.775	900.200	2011.675	2.400	.301
	Max dose (cGy)	1113.625	3494.850	5476.650	1759.900	3932.650	5704.275	2573.075	4476.400	5564.425	4.200	.122
	V50 (cc)	0.000	0.000	1.268	0.000	0.000	2.255	0.000	0.975	4.261	7.684	.021
Esophagus	Mean dose (cGy)	189.900	452.100	1509.475	191.450	608.900	1402.025	247.775	790.650	1560.550	3.200	.202
	Max dose (cGy)	849.975	2755.050	5253.000	787.975	2997.300	5157.850	1051.175	3740.100	5299.425	7.400	.025
Heart	Mean dose (cGy)	111.700	520.900	1287.000	130.275	337.800	1303.950	143.225	396.800	1298.700	0.600	.741
	Max dose (cGy)	741.525	4650.050	6050.725	997.450	3167.500	6189.550	874.675	3500.400	6294.425	0.600	.741
	V30 (%)	0.000	0.385	10.029	0.000	0.355	10.223	0.000	0.950	10.224	1.600	.449
	V40 (%)	0.000	0.020	6.000	0.000	0.129	6.261	0.000	0.254	6.010	1.600	.449
Lung All	Mean dose (cGy)	438.075	508.200	813.525	357.450	538.900	853.450	336.875	519.150	908.475	2.600	.273
	Max dose (cGy)	5827.175	6503.300	6818.325	5606.175	6304.700	6718.575	5743.400	6571.500	7215.700	7.800	.020
	V20 (%)	5.133	7.457	12.175	5.424	9.725	13.539	5.050	10.961	16.419	7.400	.025
	V5 (%)	15.825	19.355	35.350	15.358	23.509	37.332	13.202	21.758	34.710	2.600	.273
	V30 (%)	3.700	5.326	9.28	3.324	6.846	10.659	3.560	7.299	9.793	7.400	.025

Abbreviations: VMAT, volumetric modulated arc therapy; N-VMAT, noncoplanar volumetric modulated arc therapy; N-VMAT-Cage, noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system; PTV, planning target volume; Liver-GTV, liver minus gross tumor volume; CI, conformity index; HI, homogeneity index.

Statistically significant, P < .0166. Q1: first quartile, Q2: median, Q3: third quartile.

The differences in mean dose (χ² = 15.800, P = .000), and V5 (χ² = 15.000, P = .001) for Liver-GTV were statistically significant, respectively. Maximum dose for spinal cord (χ² = 8.600, P = .014) and mean dose for stomach (χ² = 15.000, P = .001) were statistically significant, respectively. Further multiple comparisons revealed that N-VMAT-Cage significantly reduced the mean dose (P = .005) and V5 (P = .005) of the normal liver, and the mean dose (P = .005) of the stomach compared to N-VMAT. N-VMAT-Cage significantly reduced the mean dose (P = .005) and V5 (P = .005) of the normal liver, V50 (P = .043) of the duodenum, and the maximum dose (P = .007) of the esophagus compared to VMAT. The results indicate that N-VMAT-Cage protects the normal liver, stomach, and lung better than N-VMAT and protects the normal liver, spinal cord, duodenum, esophagus, and lung better than VMAT.

The spread of dosimetric parameters in PTV of the 3 radiotherapy plan moderations is shown in Figure 1. The spread of mean and maximum dose in OAR of the 3 radiotherapy plan moderations relative to plan a prescription dose is shown in Figure 2.

Figure 1.

Comparison of the dosimetric parameters of planning target volume (PTV) for the 3 different radiotherapy techniques.

Figure 2.

Noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system (N-VMAT-Cage) versus noncoplanar volumetric modulated arc therapy (N-VMAT) and volumetric modulated arc therapy (VMAT) dosimetric comparison, dose relative to the prescription dose (a) Max dose, (b) Mean dose, and (c) V%.

Dose Distribution

The dose distribution and beam arrangement for N-VMAT-Cage, N-VMAT, and VMAT of the representative patient (No.1) are shown in Figure 3. N-VMAT-Cage had the lowest V₅ of Liver-GTV compared with N-VMAT and VMAT. The dose–volume histogram (DVH) of the representative patient is shown in Figure 4. These reductions in OAR dose were evident in the DVH.

Figure 3.

Dose distribution and beam arrangement for (A) noncoplanar volumetric modulated arc therapy technique based on a cage-like radiotherapy system (N-VMAT-Cage), noncoplanar volumetric modulated arc therapy (N-VMAT), and volumetric modulated arc therapy (VMAT). (A) Transverse, (B) sagittal, (C) coronal slices (color wash: green, planning target volume PTV; red, gross tumor volume GTV).

Figure 4.

Dose–volume histograms of planning target volume and organs at risk.

Discussion

In this study, we compared 3 radiotherapy techniques for HCC. Our study showed that the N-VMAT-Cage technique with the arrangement of noncoplanar arcs provided optimal dosimetric gains over the N-VMAT and VMAT techniques. By utilizing more freedom in the noncoplanar beam geometry solution space and beam orientation selection, N-VMAT-Cage achieved dose reductions to OARs, particularly the healthy liver, kidney, cord, colon, esophagus, stomach, and duodenum, as well as a significantly more compact dose distribution. N-VMAT-Cage was optimal in protecting the normal liver, which can reduce the incidence of radiation-induced liver injury and improve the local control and long-term survival rates of liver cancer patients. N-VMAT had advantages over VMAT in protecting the kidney, cord, intestine, stomach, and duodenum. However, the dosimetric advantage of N-VMAT with the finite noncoplanar arc was not obvious to avoid a collision.²⁵

There is a reason for the limited improvement in plan quality observed when using the N-VMAT technique as opposed to the VMAT technique. It is not guaranteed that arcs between anchoring beams will be optimal when using the N-VMAT technique. Because of the continuity of arcs, the range of favorable noncoplanar angles is restricted to avoid collision.^26,27 It has been demonstrated that attempting to put arc angles in a rotational IMRT system that is not in the same plane as each other does not improve the plan.²⁸ In addition, the N-VMAT technique would simultaneously rotate arcs that need both the couch and gantry, which would be impossible to achieve in existing treatment implementation.¹⁸ Internal organ motions, which may be significantly affected by couch rotation and are certainly more difficult to monitor, are another potential problem irrelevant to the current clinical investigation.

The N-VMAT-Cage technique, where the collision-free beam geometry solution space is individually mapped and the beam orientations are optimized, is an effective method for overcoming these restrictions. Two degrees of freedom are available for translating the X-ray head in the CRTS: it can rotate around as well as slide along with the longitudinal axis of the machine. This allows the body to be irradiated from different directions around the treatment volume as the machine rotates.²⁰ A patient does not need to be moved at any stage throughout the noncoplanar radiation treatment process. The X-ray head is moved immediately from one noncoplanar radiation position to another without moving the couch. Therefore, the N-VMAT-Cage technique eliminates the collision problem between gantry and couch/patient and internal organ motion caused by rotating the couch. Therefore, even though N-VMAT is currently available, it is not a reasonable substitution for N-VMAT-Cage. Due to the special setting of CRTS, it is convenient to design a noncoplanar plan. However, since the optimization algorithm is still under development, the preliminarily selected noncoplanar optimization trajectory has shown better dosimetric characteristics than VMAT. In the future, with the addition of mature optimization algorithms, caged plans will show better dosimetric characteristics than the present.

The common treatment methods include hepatectomy, liver transplantation, ablation therapy, TACE, radiotherapy, and systemic antitumor therapy. Choosing reasonable treatment methods for patients with different stages of HCC can maximize the therapeutic effect. Due to the deep location of liver, radiation is more damaging to normal liver cells, so radiotherapy is not a routine treatment for patients with liver cancer. However, for some patients with good liver function index, limited tumor, distant metastasis, and not suitable for surgical resection, radiotherapy is also one of the alternative treatments.

Moreover, N-VMAT-Cage demonstrates that the arrangement of noncoplanar arcs helped create safer treatment plans without compromising the PTV coverage. Our study was applicable to the Pinnacle 16.3 platform with 6-MV FFF photon beams and single liver lesions. There is a wide range of potential N-VMAT arc configurations examined for HCC, which is a significant strength of this work. Finally, we would like to highlight that our study was limited to single liver tumors and that the findings should not be extrapolated to more complex treatment situations, such as previous liver irradiation or multiple liver cancers. This study provides a novel noncoplanar field technique whose dosimetric advantage has been verified using the N-VMAT-Cage plan for HCC.

Because the CRTS is still a simulation study, there is currently no ready-made planning system that conforms to this accelerator structure. The CT was deflected 90° to enable the corresponding planning design in the Pinnacle³ planning system. In the future, a TPS suitable for the cage system should be developed. One major drawback of N-VMAT-Cage may be prolonged treatment time. In HCC treatment, treatment delivery speed is very critical to reduce the impact of intrafraction motion on the treatment outcome. Although VMAT is a time-saving delivery technique, it may result in lower plan quality. N-VMAT-Cage is expected to achieve a higher plan quality than VMAT with only a minor and acceptable increase in delivery time, whereas N-VMAT-Cage with superior plan quality will have significantly shorter delivery times than N-VMAT, according to the theory of the CRTS. However, since no TPS is suitable for the cage system, specific delivery times have not been compared among the 3 radiotherapy techniques. The N-VMAT-Cage technique does not require couch rotation, which reduces movement between organs and the delivery time, compared with N-VMAT. Moreover, N-VMAT-Cage is superior to the 2 other techniques in terms of dosimetric advantages. In the future, we will further enhance the application of CRTS on different types of tumor and a greater number of patients. What's more, we will distinguish the difference between large PTV and small PTV in treating HCC patient with CRTS.

Conclusion

In this study, N-VMAT-Cage plans with the arrangement of noncoplanar arcs provided significant dosimetric gains over N-VMAT and VMAT plans, except for the heart and lung. N-VMAT-Cage significantly reduced the healthy liver dose compared with N-VMAT and VMAT and should be used in more clinically challenging cases.

Footnotes

Abbreviations

Authors’ Note

For this study, anonymized patient data was used. According to our local ethics committee this does not require ethics approval.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (No.11875320).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This work was supported by the National Natural Science Foundation of China (No.11875320).

ORCID iD

Jianrong Dai

Appendix

A cage-like radiotherapy system (CRTS) has 2 stands, which are fixed to the ground in parallel, and their circular through-holes are located coaxially along with the longitudinal axis. A free treatment volume is formed between the 2 stands. Thereby, irradiating the body with sliding angle range of −45° to 45° can be done from different directions around the treatment volume. The structure of CRTS is shown in Figure A1. The sphere is constructed with the source axis distance as the radius. The stereoscopic irradiation angle achievable with the C-Linac is marked on the surface of the sphere with a thin line. The stereoscopic irradiation angle achievable with the cage radiotherapy unit is marked on the surface of the sphere with a thick line.

References

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209-249.

Cancer IAfRo. Latest global cancer data: Cancer burden rises to 19.3 million new cases and 10.0 million cancer deaths in 2020. Press release. 2020(292).

Chen

Chiang

C-L

Dawson

. Efficacy and safety of radiotherapy for primary liver cancer. Chin Clin Oncol. 2021;10(1):9.

Liang

Zhu

, et al. Radiation-induced liver disease in three-dimensional conformal radiation therapy for primary liver carcinoma: the risk factors and hepatic radiation tolerance. Int J Radiat Oncol Biol Phys. 2006;65(2):426-434.

Toya

Murakami

Baba

, et al. Conformal radiation therapy for portal vein tumor thrombosis of hepatocellular carcinoma. Radiother Oncol. 2007;84(3):266-271.

Wang

Zhang

Dong

, et al. Effectiveness of noncoplanar IMRT planning using a parallelized multiresolution beam angle optimization method for paranasal sinus carcinoma. Int J Radiat Oncol Biol Phys. 2005;63(2):594-601.

Verhey

. Comparison of three-dimensional conformal radiation therapy and intensity-modulated radiation therapy systems. Semin Radiat Oncol. 1999;9(1):78-98.

Dong

Lee

Ruan

, et al. 4pi non-coplanar liver SBRT: a novel delivery technique. Int J Radiat Oncol Biol Phys. 2013;85(5):1360-1366.

Lyu

Ruan

, et al. A novel optimization framework for VMAT with dynamic gantry couch rotation. Phys Med Biol. 2018;63(12):125013.

10.

Smyth

Evans

Bamber

, et al. Recent developments in non-coplanar radiotherapy. Br J Radiol. 2019;92(1097):20180908.

11.

Smyth

Evans

Bamber

, et al. Non-coplanar trajectories to improve organ at risk sparing in volumetric modulated arc therapy for primary brain tumors. Radiother Oncol. 2016;121(1):124-131.

12.

Woods

Nguyen

Tran

, et al. Viability of non-coplanar VMAT for liver SBRT as compared to coplanar VMAT and beam orientation optimized 4π IMRT. Adv Radiat Oncol. 2016;1(1):67-75.

13.

Teoh

Clark

Wood

, et al. Volumetric modulated arc therapy: a review of current literature and clinical use in practice. Br J Radiol. 2011;84(1007):967-996.

14.

Matuszak

Yan

Grills

, et al. Clinical applications of volumetric modulated arc therapy. Int J Radiat Oncol Biol Phys. 2010;77(2):608-616.

15.

de Pooter

Romero

Wunderink

, et al. Automated non-coplanar beam direction optimization improves IMRT in SBRT of liver metastasis. Radiother Oncol. 2008;88(3):376-381.

16.

Hsieh

C-H

Liu

C-Y

Shueng

P-W

, et al. Comparison of coplanar and noncoplanar intensity-modulated radiation therapy and helical tomotherapy for hepatocellular carcinoma. Radiat Oncol. 2010;5:1-8.

17.

Thibault

Wronski

Korol

, et al. Dosimetric comparison of volumetric modulated arc therapy and intensity-modulated radiation therapy for liver stereotactic ablative body radiotherapy. J Radiat Oncol. 2016;5(1):63-69.

18.

Lin

Ozawa

Miura

, et al. Split-VMAT technique to control the expiratory breath-hold time in liver stereotactic body radiation therapy. Phys Med. 2017;40:17-23.

19.

Sharfo

AWM

Dirkx

MLP

Breedveld

, et al. VMAT plus a few computer-optimized non-coplanar IMRT beams (VMAT plus) tested for liver SBRT. Radiother Oncol. 2017;123(1):49-56.

20.

Niu

C-M

M-H

Dai

J-R

. Cage-like radiotherapy system for noncoplanar radiotherapy. Nucl Sci Tech. 2021;32(2).

21.

Langen

Papanikolaou

Balog

, et al. QA For helical tomotherapy: report of the AAPM Task Group 148a). Med Phys. 2010;37(9):4817-4853.

22.

, et al. Non-coplanar volumetric modulated arc therapy for locoregional radiotherapy of left-sided breast cancer including internal mammary nodes. Radiol Oncol. 2021;55(4):499-507.

23.

Paddick

. A simple scoring ratio to index the conformity of radiosurgical treatment plans. Technical note. J Neurosurg. 2000;93(Suppl 3):219-222.

24.

Patel

Mandal

Choudhary

, et al. Plan evaluation indices: a journey of evolution. Rep Pract Oncol Radiother. 2020;25(3):336-344.

25.

Smyth

Bamber

Evans

, et al. Trajectory optimization for dynamic couch rotation during volumetric modulated arc radiotherapy. Phys Med Biol. 2013;58(22):8163.

26.

Meedt

Alber

Nusslin

. Non-coplanar beam direction optimization for intensity-modulated radiotherapy. Phys Med Biol. 2003;48(18):2999-3019.

27.

Tran

Zhang

Woods

, et al. Treatment planning comparison of IMPT, VMAT and 4pi radiotherapy for prostate cases. Radiat Oncol. 2017;12(1):10.

28.

Yang

Jones

, et al. Feasibility of non-coplanar tomotherapy for lung cancer stereotactic body radiation therapy. Technol Cancer Res Treat. 2011;10(4):307-315.