EP-1566: Comparison of IMRT and VMAT plans with different energy levels using monte carlo algorithm for prostate cancer

S189

ESTRO 33, 2014 EP-1563 Comparison of 3-dimensional conformal radiotherapy and intensity modulated radiotherapy for breast cancer H. Kivanc1, M. Gultekin1, F. Yildiz1, M. Gurkaynak1 1 Hacettepe University Faculty of Medicine, Radiation Oncology, Ankara, Turkey Purpose/Objective: Chest wall (CW) and lymphatic irradiation for left breast cancer is one of the most difficult techniques in radiotherapy. The aim of this study is to compare left-sided CW and lymphatic conformal and intensity modulated radiotherapy (IMRT) treatment planning techniques. Materials and Methods: Ten patients with left breast cancer who were treated with mastectomy were selected. Three dimensional conformal radiotherapy (3DCRT), forward-planned IMRT, inverse-planned IMRT (7field and 9-field technique), and hybrid 3DCRT/IMRT were compared. The clinical target volume included left CW and internal mammary (IM), supraclavicular, and axillary nodes. Five treatment plans were created for each patient. The critical structures included the left and right lungs, the heart, esophagus, thyroid, humeral head, spinal cord, brachial plexus and contralateral breast. Dose-volume histograms (DVH) obtained and D95% of PTVs; D98%, D2% and Dmax. of CTVs; V20Gy for the left lung, the left and right lung mean doses; V5Gy, Dmax. and mean doses of the right breast; V5Gy, V10Gy, V20Gy and V30Gy and mean doses of the heart were analayzed. On Alderson Rando phantom each technique has been separately simulated and the dosimetric accuracy of each treatment planning technique was tested with placing TLD dosimeters on spots for each volume. Quality control of inverse planning techniques were performed using EPID and universal phantom. For estimation of dosages on surface, gafchromic film dosimeters were used. Results: For inverse and forward-planned IMRT techniques more homogeneous dose distribution was obtained compared to 3DCRT. The heart volumes receiving high radiation dose (V20Gy ve V30Gy) were significantly less with inverse-planned IMRT (9-field technique) (V20 Gy: 4,6±3,3, p<0.001; V30 Gy: 0,9±0,8, p<0.001). However, right breast doses were significantly reduced with forward-planned IMRT compared to other techniques (V5Gy: 4,5±4,9, p<0.001; mean dose: 145,6±83,4, p<0.001). Compared to the 7-field IMRT technique, 9-field IMRT technique reduced exposure to normal tissues. Approximately 0.3-11.1% difference was found between dosimetric evaluation and TDL estimations. The quality control results of inverse planning techniques were satisfactory. Conclusions: As a result of this study, IMRT techniques improve CW and nodal coverage while decreasing doses to the critical organs when compared with 3DCRT. Moreover, the critical organ volumes receiving low radiation doses were larger for inverse-planned IMRT, while volumes receiving high radiation doses were larger for forward-planned IMRT. Hybrid/3DCRT IMRT plans were found to have the advantages of both forward and inverse planning IMRT techniques. EP-1564 Hippocampal-sparing WBRT using VMAT technique R. Bar-Deroma1, I. Fotina2, A. Nevelsky1 1 Rambam Medical Center, Oncology, Haifa, Israel 2 Tomsk Polytechnic University, Applied Physics, Federation

Tomsk,

Russian

Purpose/Objective: In our previous work we showed the feasibility of hippocampal-sparing whole-brain radiotherapy (HS WBRT) using stepand-shoot IMRT plans created with Monaco TPS v. 3.1 for an Elekta Infinity linac with MLCi2 (1 cm leaf width) collimator. However, the VMAT plans did not meet the compliance criteria predefined in RTOG 0933 protocol. Recently, the Monaco TPS v. 3.3 and Elekta Axesse linac with the Agility MLC were installed at our department. The aim of this study was to demonstrate improvements in quality and efficiency of the VMAT plans for HS WBRT created with the newly installed equipment. Materials and Methods: Ten cases used in the previous study were replanned with the Monaco v. 3.3 using a VMAT approach for the Elekta Infinity and Axesse linacs. Dose prescription was 30 Gy in 10 fractions. RTOG0933 recommendations were applied for treatment planning. Plan evaluation was performed using D2% and D98% for the whole-brain PTV (defined as whole brain excluding hippocampus avoidance region), D100% and maximum dose to the hippocampus, and maximum dose to optic nerves and chiasm. Beam-on time was recorded for all plans. Results: Implementation of the new sequencer in Monaco v. 3.3 enabled us to create VMAT plans which met RTOG 0933 compliance criteria for both linacs. For the Infinity linac there were no significant differences between step-and-shoot and VMAT plans. However, average treatment time was reduced from 13.5 min to 6.5 min. For the Axesse linac the hippocampus D100% mean dose was reduced to 6.9 Gy compared to the

D100% mean dose of 8.7 Gy achieved with Infinity linac due to smaller leaf width and better transmission characteristics of the Agility MLC. Average beam-on time was further reduced to 4.8 min thanks to increased leaf speed, interdigitation of the leaves and introduction of continuously variable dose rate mode. Conclusions: With currently available technologies it is possible to use the VMAT approach with significant improvement in dosimetric quality and delivery efficiency of HS WBRT plans. EP-1565 The effect of FFF for patients receiving pelvic radiotheraphy using volumetric modulated arc therapy technique E. Erdogan1, O. Guneyli1, M. Garipagaoglu1, H. Kucucuk2 1 Acibadem Universty, Radiation Oncology, Kozyatagi - Istanbul, Turkey 2 Acibadem Kozyatagi Hospital, Radiation Oncology, Kozyatagi - Istanbul, Turkey Purpose/Objective: Using flattened beams in linear accelerator and radiotherapy planning has been a routine procedure. However, Intensity modulated radiotherapy (IMRT) could modulate beams therefore using flattened beams (FF) is not mandatory for IMRTplans. Recently some studies showed better dose conformality, less peripheral dose values and treatment time for (IMRT) plans without flattening filter (FFF). In this study, the effect of FF in(IMRT) plans using Volumetric modulated arc therapy (VMAT) technique were examined. Materials and Methods: Ten patients with prostate cancer receiving radiotherapy to prostate and pelvic lymph nodes, using IMRT as VMAT technique were included. Target volumes were PTV1 (prostate), PTV2 (Seminal Vesicle) and PTV3 (pelvic Lymph node); rectum and bladder were accepted as organ at risk (OAR). PTV1, 7770cGy, 6660cGy and 5550cGy respectively. Dose restrictions for OAR were: 1. max rectum and bladder doses should be less than 7400cGy. 2. the rectum (not including PTV) volume receiving should be less than 7000cGy was, 3. 5% of rectum volume should receive less than 6500cGy, 4. 40% of rectum volume should receive less than 5000cGy,5. the bladder (not including PTV) volume should receive less than 7000cGy, 50% of bladder should receive less than 5000cGy was. Treatment plans were created in ECLIPS treatment planning system (Varian Medical Systems) which use AAA algorithm and PRO 11.0.31 optimisation software. Plans were created using 2arcs VMAT, 10MV energy. Dose rates 600MU/second for FF and 800MU/second for FFF plans for Truebeam-STx linac. For each patient ,2 different plans as FF and FFF were made. Mean and maximum dose for targets and OAR, used number of monitor units(MU) values were compared. Furthermore dose homogeneity and conformality for PTV1 were studied. Results: Intended doses for targets and dose limits for OAR were obtained for FF and FFF plans. Dmax values were found higher in FFF plans than FF plans, 1.74%, 0.61% and 0.58% for PTV1, PTV2 and PTV3 respectively. Difference was significant for PTV1 (p=0,00) only. Both homogeneity and conformality for PTV1 were more successful in FF in comparison to FFF; difference was significant: 17.6% for homogeneity (p=0,00) and 2.54% for conformality (not significant). Max OAR doses were less for FF than FFF, difference was 1.17% for rectum and 0.51% for bladder; difference was significant for rectum (p=0,039) while not significant for bladder. MU values 40% higher for FFF than FF (p=0,00). Conclusions: According to results of present study, better conformality, homogeneity and OAR protection were achieved for FF plans comparing to FFF plans while higher MU required for FFF plans in patients receiving pelvic radiotherapy. Results of present study is not compatible with some studies reported benefit of FFF plans for OAR protection. We speculate that, reason of differences are wide treatment target, several dose prescriptions values. EP-1566 Comparison of IMRT and VMAT plans with different energy levels using monte carlo algorithm for prostate cancer O.C. Güler1, C. Onal1, G. Arslan1, S. Sonmez1 1 Baskent Universitesi Tip Fakultesi Adana Hastanes, Radiation Oncology, Adana, Turkey Purpose/Objective: To make dosimetric comparisons of volumetricmodulated arc therapy (VMAT) and 7-field intensity-modulated radiotherapy (IMRT) with dynamic MLCs using Monaco treatment planning system with Monte Carlo algorithm. Materials and Methods: Single-arc VMAT and 7-field IMRT treatment plans were compared for 12 intermediate risk prostate cancer patients treated with prostate and seminal vesicle RT. Clinical target volume (CTV) included the prostate and the entire seminal vesicles. The

S190 planning target volume (PTV) was defined as CTV with a margin of 6 mm posterior and 10 mm in other directions. For all patients the prescribed dose was 78 Gy delivered in 39 fractions. The dosimetric data of IMRT and VMAT plans with 6, 10 15 MV energies were compared. The comparison was made for target volume, organs at risk (OAR) doses, and for monitor units (MU). Results: The mean CTV and PTV were 33.5 cm3 (range 18.6 – 73.1 cm3) and 158.6 cm3 (range 102.1 – 251.4 cm3), respectively. In VMAT plans the normal tissue surrounding the target volume doses were significantly lower than IMRT plans. VMAT plans achieved lower doses to all OARs for nearly all dosimetric endpoints. VMAT plans achieved a 9.4%, 9.0% and 7.0% relative decrease in MUs required for RT delivery, for 6, 10 and 15 MV energy levels, respectively. The target volume and OARs dosimetric values did not differ significantly between 6, 10 and 15 MV photon energy. Conclusions: VMAT plans were found to be dosimetrically equivalent to IMRT plans for prostate cancer patients, with better rectum and bladder sparing and less MUs required. For this reason, rather than IMRT with higher energy, VMAT with lower energy is feasible. Additional studies are needed to evaluate whether the improved dose distribution with VMAT results in decreased toxicity, and very long-term follow-up is required to determine the potential for VMAT to decrease the rate of secondary malignancies compared with IMRT. EP-1567 Comparision of IMRT-plans and volumetric modulated arc therapy for an ARTISTE accelerator U. Spahn1, F.J. Prott1 1 RNS-Wiesbaden, Radiotherapy, Wiesbaden, Germany Purpose/Objective: Since February 2012 a Siemens accelerator ARTISTEÒ is in clinical use in our departement. The dose rate with flattened beam is 300MU/min for 6MV photons respectively 500MU/min for 15MV photons. At he same time also IMRT with the step&shoot method was established in our department. For treatment planning of IMRT-plans the Pather TPS (Prowess Inc.,USA) is used. In summer 2012 we got the opportunity to take part in a clinical trial for a new method of rotational IMRT (mARCÒ) at our accelerator. In this investigation, both methodes of IMRT will be compared according treatment time, monitor units and plan quality. Materials and Methods: In the mARC variation of rotational IMRT, the dose is delivered during a continuous rotation of the gantry in bursts over very short gantry angels with static leaf positions.Between these dose bursts, the MLC leaves are moving rapidly in order to form the next field segment. For treatment planning a direct aperture optimization algorithm DAO (Panther DAO, Prowess Inc.) is used.The planning of rotational IMRT is done with the ProArc option of the Panther treatment planning system. In a retrospective study, mARC will be compared with the IMRT plans used for patient treatment. In all cases,for both methods the same dose prescription to the target volumes and the same dose limits for organs at risk are used. The IMRT techniques are compared regarding the nessescary Monitor Units (MU), the treatment time and plan quality (target coverage (TC), conformity indec (CI), conformity number (CN)). Results: The first results for patientes treated for prostate cancer indicate a slight increase (+7,3%) in monitor units while the treatment time is reduced significantly by 21% (4,8±0,35 min with Step&Shoot IMRT compared to 3,8±0,38 min with mARC). For head&neck cases, also a reduction of monitor units by 6% and treatment time by 17,5% could be observed with mArc. In all cases, the plan quality according to TC, CI and CN could be improved slightly with the mARC techniqe (tab. 1).

Conclusions: In comparision to Step&Shoot IMRT, the mARC technique shows comparable or improved dose distributions with a significant reduction in treatment time even for an accelerator with flattening filter. A further reduction of treatment time might be expected in case of flattening filter free beams with a higher dose rate.

ESTRO 33, 2014

EP-1568 Dosimetric comparison between GPU and CPU cluster based Tomotherapy dose engine for different anatomical districts A. Maggio1, A. Di Dia1, S. Bresciani1, C. Cutaia1, A. Miranti1, M. Stasi1 1 FPO IRCCS Candiolo, Medical Physics, Candiolo, Italy Purpose/Objective: To compare helical tomotherapy plan quality and plan characteristics between GPU (Graphics Processing Unit) and CPU (Central Processing Unit) dose engine for Head and Neck (H&N), prostate (PCa) and mesothelioma (MES) tumors. Materials and Methods: 5 H&N, 5 PCa and 5 MES patients (pts) plans were calculated with both engines. For each pts the same number of iteractions (400), optimization parameters and plan characteristics (field width, pitch,modulation factor) were set at the beginning of optimization and not adjusted during the process. Gamma index Γ(3%/3mm)<1 was used to compare the agreement between GPU and CPU dose plans. The plan differences were evaluated by using D20 (dose to 20% of OAR (organ at risk) volume), D50 and D80 for OARs and D1, D50 and D95 for planning target volumes (PTVs). Treatment delivery time and total planning time were also evaluated. The ability of GPU system in order to obtain the same final solution after 400 iterations was also analyzed only for 1 PCa pts. Results: The average difference among the 5 DVH for the same PCa pts was <0.0001 % for all OARs and PTVs investigated. The gamma analysis demonstrated an excellent agreement between the GPU and CPU dose (average Γpass rate for all patients was 98.8±1.4%). The average total optimization/dose calculation time for GPU dose engine (296 ± 57 min) was significantly different(p<0.0001) than CPU dose engine (75.7 ± 21 min). There was no difference in the average treatment delivery time (9.3±1.9 min for both systems). The results showed no significant difference between the GPU and CPU calculated plans. In fact, the average D20, D50, D80 varied by less than 2.1% for OARs (p>0.1);for PTVs D1, D50 and D95 varied by less than 1% (p>0.1). A dose difference increasing from 0.8 % to 2.1 % between GPU and CPU plans was observed when moving from D20 to D80 for OARs. Conclusions: Plan qualities obtained with the two systems were comparable with no dosimetric significant difference for all cases.However important reduction of planning time with GPU engine respect to CPU dose engine, mainly due to the absence of beamlet calculation step, introduce also the possibility to explore alternative plan parameters in a more flexible way and to use fine dose grid for all pts. EP-1569 The dosimetric effects of the breath-hold technique for helical tomotherapy M. Yamamoto1, T. Endo1, N. Koshi1, K. Masumoto1, M. Tanemura2 1 Kure National Hospital Chugoku District Cancer Ce, Radiation Oncology, Kure Hiroshima, Japan 2 Kure National Hospital Chugoku District Cancer Ce, Surgical Oncology, Kure Hiroshima, Japan Purpose/Objective: Breathing-induced organ motion is a problem in radiation therapy, because a moving target needs additional margin and there is a risk of interplay effect in intensity-modulated radiation therapy. To resolve the problem of breathing-induced organ motion, we designed a breath-hold technique for helical tomotherapy (HT). The technique is simple and to block radiation-beam at angles between 270 and 360 degrees when gantry turn around a patient. Patients hold their breath at angles between 1 and 269 degrees and can breathe at angles between 270 and 360 degrees when gantry turn around a patient. This study evaluates the dosimetric effects of the breath-hold technique for HT. Materials and Methods: The CT images of 7 patients with locally advanced pancreas cancer were used in this study. Each CT volume was obtained under breath-holding conditions using monitoring system in CT scanner with a 50-cm field-of-view, 512 x 512 pixels per slice and a 2.5mm slice thickness. The tumor, regional lymph node areas, duodenum, stomach, liver, kidneys, and small intestine were contoured. The structures to block radiation-beam at angles between 270 and 360 degrees were also contoured. We named the structures blocking structures (BS). HT plans were generated under 2 conditions with or without BS. The prescription was used for tumor with 60 Gy (30 fractions) and for regional lymph node areas with 50 Gy (30 fractions) to cover 50% of target volume. Results were compared statistically using Welch's t test. Results: The maximum doses of tumor were 60.8-63.0 Gy (mean, 61.8