Volumetric modulated arc therapy (VMAT) is superior to intensity modulated radiotherapy (IMRT) for liver sparing in SBRT for hepatocellular carcinoma

Background: The best delivery modality of stereotactic body radiotherapy (SBRT) for hepatocellular carcinoma (HCC) from a technical point of view is still a matter of debate. The purpose of this study was to compare planning parameters with volumetric modulated arc therapy (VMAT) to static intensity modulated radiotherapy (IMRT) in treatment of HCC treated with SBRT. Methods: Twenty patients (pts) with localized HCC who were treated with SBRT were re-planned using two different radiation techniques: IMRT and VMAT. Patients with Child A cirrhosis received 45-54 Gy in 3 fractions and 5 pts with Child B cirrhosis received 30 Gy in 5 fractions. Planning was optimized to minimize doses to organs at risk (OAR) without compromising coverage of the planning treatment volume (PTV). VMAT and IMRT plans were compared using the conformity index (CI), homogeneity index of the PTV and monitor units (MUs) for time of treatment delivery, and other dose volume histogram (DVH) metrics. Results: The CI of VMAT plans were superior to those of IMRT(1.11±0.05 vs 1.18±0.06 (p <0.05) The MUs were signicantly lower for VMAT (423.78±50.65) than for IMRT (890±160.68) (p <0.01). Stomach max dose, normal liver V15 Gy, normal liver mean dose were also decreased with VMAT planning. This was achieved without increased :V30 Gy of duodenum and small bowel , 10 cc(Gy) of duodenum, mean dose of right kidney and cord max dose . Conclusion: PTV coverage was more conformal with VMAT planning, with lower MUs and shorter delivery time compared to IMRT in all pts. Moreover VMAT planning was more effective than IMRT planning in the sparing of normal liver and stomach.


Abstract
Background: The best delivery modality of stereotactic body radiotherapy (SBRT) for hepatocellular carcinoma (HCC) from a technical point of view is still a matter of debate. The purpose of this study was to compare planning parameters with volumetric modulated arc therapy (VMAT) to static intensity modulated radiotherapy (IMRT) in treatment of HCC treated with SBRT.
Methods: Twenty patients (pts) with localized HCC who were treated with SBRT were re-planned using two different radiation techniques: IMRT and VMAT. Patients with Child A cirrhosis received 45-54 Gy in 3 fractions and 5 pts with Child B cirrhosis received 30 Gy in 5 fractions. Planning was optimized to minimize doses to organs at risk (OAR) without compromising coverage of the planning treatment volume (PTV). VMAT and IMRT plans were compared using the conformity index (CI), homogeneity index of the PTV and monitor units (MUs) for time of treatment delivery, and other dose volume histogram (DVH) metrics.
Results: The CI of VMAT plans were superior to those of IMRT(1.11±0.05 vs 1.18±0.06 (p <0.05) The MUs were signi cantly lower for VMAT (423.78±50.65) than for IMRT (890±160.68) (p <0.01). Stomach max dose, normal liver V15 Gy, normal liver mean dose were also decreased with VMAT planning. This was achieved without increased :V30 Gy of duodenum and small bowel , 10 cc(Gy) of duodenum, mean dose of right kidney and cord max dose .
Conclusion: PTV coverage was more conformal with VMAT planning, with lower MUs and shorter delivery time compared to IMRT in all pts. Moreover VMAT planning was more effective than IMRT planning in the sparing of normal liver and stomach.

Background
Hepatocellular carcinoma represents the sixth most commonly occurring cancer in the world and the third largest cause of cancer mortality. Treatment options include liver transplantation, surgical resection, radiofrequency ablation, percutaneous ethanol injection, transarterial embolization/chemoembolization, sorafenib and radiation therapy. In the past, radiation induced liver disease (RILD) has greatly limited the role of radiation due to the relationship between irradiated liver volume and toxicity [1].
Developments of radiotherapy techniques, mainly IMRT, have gradually made radiotherapy more relevant in the treatment of HCC and indications for its use have expanded in the last few years. SBRT has been shown as an alternative option in early stage HCC patients, patients not eligible for other ablative procedures or as a bridge to liver transplant.
SBRT delivers high doses of radiation in a few fractions with high degree of accuracy and has shown encouraging results. [2]. Studies comparing IMRT and VMAT to 3DCRT SBRT in patients with lung cancer, pancreatic cancer, spinal and abdominal metastases showed improvements in dose conformity and homogeneity with VMAT [3,4].

Page 3/10
The best delivery modality for HCC SBRT from a technical point of view is a matter of debate. We compared two IMRT techniques: (1) Sliding window IMRT, in which modulated multileaf collimator in multiple static radiation elds is used and (2) VMAT, which delivers an entire IMRT treatment in a single gantry rotation around the patient. Current literature comparing VMAT and IMRT for lung tumors shows that both techniques could provide comparable target coverage and dose conformity however, OAR results do not show a clear preference. The purpose of this study was to better ascertain whether one technique is preferable to the other in terms of coverage, conformality, homogeneity and dose to OAR in SBRT for HCC.

Patient population
Twenty patients (pts) with single lesion HCC who were treated with SBRT at our center were entered into this planning study, after obtaining Helsinki committee approval. Median age at RT was 62.2 (range, 43.0-83.9) years, 70% were male, 70% were viral carriers, and all had liver cirrhosis. Fifteen pts had Child-Pugh A cirrhosis and 5 pts had Child-Pugh B cirrhosis. ECOG performance status was ≤ 2. All patients had early stage disease (BCLC stage A). Five patients had left lobe tumors and 15 patients had right lobe tumors. Immobilization, simulation, and target delineation The patients were immobilized using vacuum casts in a supine position with arms raised above their heads. Contrast multiphase 4-dimentional CT simulation was performed with a 2.5-mm slice thickness and included whole liver and bilateral kidneys. Respiratory control and abdominal compression were used. Images were reconstructed on the Advantage Workstation (GE Healthcare, Chicago, IL). The internal target volumes were created using all 10 phases to account for maximal tumor excursion.
The gross tumor volume (GTV) was de ned as the contrast enhancing tumor volume on a triphasic CT or MRI scan. The PTV was de ned as a 2-3 mm margin around the GTV. The PTV was reduced in case of proximity to vital normal tissue.
Organs at risk were also delineated, including whole liver, non-target liver (whole liver minus PTV), stomach, duodenum, kidney, spinal cord and small bowel. Before each treatment, cone beam CT was used to position the patient. Treatment planning and dose delivery In our study, we prescribed 95% of total dose to cover ≥ 95% of the PTV coverage while keeping the minimum dose ≥ 93% of total dose and maximum dose ≤ 107% of total dose and normalized all plans to the mean dose of PTV.
Fifteen patients with Child A cirrhosis received 45-54 Gy in 3 fractions and ve patients with Child B cirrhosis received 30 Gy in 5 fractions. Patients were treated every other day. For the purposes of the planning study, each patient case was replanned with both IMRT and VMAT techniques. The plans were calculated using the ECLIPSE planning system (Varian medical systems, Palo Alto, CA).
The VMAT plans were optimized using 1 to 3 arcs with a total arc span varying between 190 to 490 degrees. For IMRT, (6-8 xed elds were used and) almost all plans were optimized with 7 xed elds. Maximum dose rate was used for all plans: 600 MU/min for 6 MV and 10 MV plans and 1400 MU/min and 2400 MU/min for 6 MV-FFF and 10 MV-FFF plans, respectively.
For patients with Child A cirrhosis, one third of the uninvolved liver was restricted to 10 Gy, and 500 cc of uninvolved liver received < 7 Gy. For patients with Child B cirrhosis, one third of the uninvolved liver was restricted to 18 Gy, and 500 cc of uninvolved liver received < 12 Gy. The complete RT constraints can be viewed in Table 1  The dose to the PTV and OAR was evaluated using DVHs. PTV coverage was compared between VMAT and IMRT using the following parameters: 1. Conformity index (CI, CI = VTV/VPTV). Where VTV is the treatment volume enclosed by the prescribed isodose surface, and VPTV is the volume of the PTV. A greater CI indicates lower conformity.
2. The homogeneity index (HI) = D5%/D95%, where D5% and D95% are the minimum doses delivered to 5% and 95% of the PTV. HI is a ratio that is used to evaluate the homogeneity of the PTV. The smaller and closer the value of HI is to 1, the better the homogeneity of the PTV [3].

Statistical analyses
Non-parametric statistical analyses were performed using the paired, two-tailed Wilcoxon signed-rank test, with a p-value < 0.05 de ned as signi cant. We compared each patient's IMRT plan with their VMAT plan.
T-test was used for testing differences in mean coverage of PTV, and to test for differences in the dose to the OAR, among the different planning techniques.
The dosimetric differences among the two treatments for the 20 pts were analysed using the Friedman test. When a signi cant difference (p < 0.05) was found, the difference between two treatments for each effect was further examined by Wilcoxon signed-rank test. All analyses were performed using SPSS software, version 15.0 (SPSS Inc., Chicago, IL).
IMRT had twice the MUs (890 ± 160.68) of VMAT (423.78 ± 50.65) (p < 0.01). Although the estimated treatment time delivery was 25% shorter with VMAT plans (p < 0.046) The full data for PTV coverage can be seen in Table 2.

Discussion
In the past decade, radiotherapy has gradually become a treatment option for HCC patients and is currently a standard in several guidelines. Klein and Dawson [5] concluded that SBRT in HCC can have comparable results with other therapies and suggested that it might be offered as a treatment option for early stage HCC patients or patients not eligible to other ablative procedures. SBRT uses up to ve fractions utilizing a steep dose fall-off enabling the delivery of high dose to the target volume while limiting the dose to adjacent normal tissue.
The recent results with SBRT for HCC are very encouraging. Bujold et al. [6] in two sequential trials, showed a 1-year local control of 87% with a clear association of better outcome with higher SBRT doses, although grade 3 toxicity was observed in one third of the patients. Jang et al. [7] reported that 2-year local control /overall survival (OS )rates in SBRT treated patients with doses of > 54, 45-54, and < 45 Gy were 100/71, 78/64, and 64%/30%, respectively (p = .009/p < .001). SBRT dose was found to be a statistically signi cant prognostic factor for OS. In our own experience with SBRT for HCC, we reported favorable outcomes using SBRT as de nitive treatment or as a bridge to liver transplantation in pts with early stage inoperable HCC [9]. Median prescribed dose to the tumor was 54 Gy (range, 30-54Gy) with median overall-survival for the transplanted patients not reached and 23 months for the patients who were not transplanted. Liver explant revealed 81% response rate (27% complete and 54 partial). Only one patient developed RILD.
This demonstrates that not only is SBRT effective but that it is critical to maximize the therapeutic window in these patients and therefore the importance of the best planning technique.
There is no large-scale trial for assessment the best radiation planning technique for HCC SBRT both for target coverage and normal tissue avoidance.
In reviewing studies of SBRT delivery techniques in other tumors, such as lung cancer, pancreatic cancer ,spinal and abdominal metastases, IMRT and VMAT for SBRT and compared those techniques to 3D -CRT with improvement in dose conformality and homogeneity [3]. However, when compared to IMRT dose conformality was greatest with VMAT. Treatment time was markedly quicker with VMAT versus 3DCRT and IMRT. Overall both IMRT and VMAT were able to meet all planning constraints in the studies reviewed; however, treatment e ciency was greatest with VMAT.
However, to our knowledge, this is the rst report comparing IMRT to VMAT in SBRT treated HCC patients. Our data suggests that PTV coverage is more conformal with VMAT planning in all pts. VMAT tended to be more effective in all groups for the sparing of normal liver. For pts with PTV volume < = 46 cc the value of VMAT was better for normal liver V15 and CI, for PTV volume > = 46 cc no difference was seen. For right lobe tumor VMAT had lower stomach maximum dose, liver V15, CI.
The study has several limitations; It is a small retrospective study, with different prescribed dose to tumor according to Child type of cirrhosis. In addition, the limited sample size makes it di cult to distinguish variable related to the location of disease. In addition, the relationship between PTV size and volumetric sparing e ciency should be further examined to form a conclusive understanding of this relationship.
In conclusion based on our data we suggest further exploration into the use of VMAT strategies for treatment of HCC with SBRT.