AAMD 41st Annual Meeting

Poster Abstracts

Analysis of inhomogeneous dose distribution using Volumetric Modulated Arc Therapy (VMAT) for prostate, GBM, and lung patients and the effect on organs at risk
Stacy Peterson

The University of Texas MD Anderson Cancer Center

Abstract:
Analysis of inhomogeneous dose distribution using Volumetric Modulated Arc Therapy (VMAT) for prostate, GBM, and lung patients and the effect on organs at risk

Stacy Peterson, Pasquale Montanaro, M.S., DABR

Purpose: The following study provides an analysis on the difference in dose to normal tissue and organs at risk when limits on dose homogeneity in the target volume are changed.

Methods and Materials: The study used the Pinnacle planning system and VMAT to plan four separate plans for each disease site: prostate, GBM, and lung, each to the prescribed dose. The prescribed dose for each prostate plan was 79.2 Gy/44 fractions. The planning target volume was created by adding a margin of 5 mm to the gross tumor volume. The GTV received 100% of the prescribed dose and the PTV received 98% of the prescribed dose in each plan. The prescribed dose for each GBM plan was 60 Gy/30 fractions. The GTV received 100% of the prescribed dose, the CTV received 98% of the prescribed dose, and the PTV received 95% of the prescribed dose in each plan. The prescribed dose for each lung plan was 63Gy/35 fractions and the planning target volumes varied for each patient. Each plan was examined after the maximum dose allowed was escalated from approximately 105% to 120% of the prescribed dose while maintaining equivalent coverage of the tumor volume for each plan. The maximum dose constraint was then relaxed and priority was given to lowering the dose to normal tissue. The critical structures for each plan were analyzed using data from the dose volume histograms. Metrics used in the evaluation of stereotactic planning will were used for the evaluation of the resulting dose distributions. These include R100 and R50, which are the ratio of the 100% isodose line and 50% isodose line to the planning target volume respectively.

Results: The data for each of the nine plans indicated that as the maximum dose constraint was relaxed the R50 (50% isodose line to the planning target volume) increased, and as a result dose to critical structures and normal tissue decreased. The structures that are pertinent for analysis for the prostate plans are the rectum, urethra, bladder, and femoral heads. The structures that will be analyzed for the GBM plans include the spinal cord, brainstem, right and left orbits, right and left optic nerves, cochlea, optic chiasm, and right and left lenses. The structures that will be analyzed for the lung plans include the heart, both lungs, spinal cord, and esophagus.

Conclusion: When the maximum dose constraint within the PTV was relaxed, the resulting dose distribution exhibited lower doses to the contoured organs at risk, and to normal tissue in general.

Keyword(s): VMAT, Inhomogeneous, Prostate, Glioblastoma, GBM, Lung, Organs at risk

The reader should describe the effect on critical structures when limits on dose homogeneity in the target volume are changed.

The reader should describe how metrics used in the evaluation of stereotactic planning can also be used in the evaluation of resulting dose distributions in VMAT planning.

The reader should describe the effects of the 50% isodose line to the planning target volume escalating in dose as the maximum dose constraint is relaxed.


Efficient delivery of MLC-based grid therapy using a flattening filter free beam
Christopher Beyer
Certified Medical Dosimetrist
Henry Ford Health System

Abstract:

Purpose: To compare and improve upon the efficiency of multi-leaf collimator (MLC) based grid therapy using a flattening filter free (FFF) beam versus a traditional flattened beam. Grid therapy has been proven to be a simple yet effective treatment for bulky tumors but the extended treatment times have a greater risk for intrafraction motion. This study aimed to reduce the treatment time by utilizing a high dose rate FFF beam on the Varian TrueBeam treatment machine, while maintaining comparable plan quality with the regular MLC based grid therapy.

Methods: Five patients were included in this study. Two MLC based grid therapy treatment plans were created for each patients: a FFF 10x treatment plan manually flattened with a field in field technique (FinF) and a flattened 10x treatment plan. The grid sizes were 1cm by 1cm with 1 cm spacing. The dose rates were 2400 MU/min for the FFF beam and 600 MU/min for the flattened beam. Treatment plans were created for treatment sites including head and neck, breast, and extremity.  All treatment plans were created in the commercial Treatment Planning System and delivered on a Varian TrueBeam System with Millennium 120 MLC. Treatment plans were prescribed 15 Gy at depth of 2.0cm in one fraction.

Results: Preliminary data from the 5 cases revealed that the FFF plans were more efficient at delivering MLC-based grid therapy.  Although as a result of FinF, the FFF plans yielded about 36%±22% higher amount of MU, due to the high dose rate of 10xFFF beam, the average beam-on time was significantly reduced by 200%±60% (7.9±3.6 minutes vs. 2.82±1.49 minutes, p<0.01). The FFF plans showed a slightly higher global maximum dose of 1.08±0.91Gy. The valley to peak ratio was 11% and 15% for the FFF plans and flat beam plans respectively.

Conclusion: The preliminary results showed that the treatment time can be significantly reduced in MLC-based grid therapy using FFF beams.  More efficient treatment delivery can potentially reduce the risk of intrafraction motion and increase patient satisfaction.

  1. FFF mode provides much greater efficiency over traditional beam for Grid Therapy
  2. FFF plans can achieve the same plan quality as traditional beam Grid Therapy plans
  3. An increase in MU's does not necessarily equate to longer beam on time

Single-Fraction versus Multi-Fraction Radiotherapy in Palliative Bony Metastasis Patients: Comparison of Primary Malignancy and Demographic Factors
Kristen Vu
Educational Coordinator
Grand Valley State University

Abstract:
Introduction

The intent of this study is to analyze Medicare billing data to discover the differences in patients that are treated with single-fraction and those treated with multi-fraction treatments for the treatment of bony metastases. The factors investigated are the overall number of single versus multiple fraction treatments, primary malignancy, age, gender, race and region in which patients live. The key issues studied are the Radiation Oncologists patterns of utilization of single or multi-fraction treatment.

Methods and Materials

Radiation therapy treatment billing claims for bony metastasis from Medicare beneficiaries in 2014 were analyzed. Beneficiary inclusion criteria included ICD-9 code 198.5 (Secondary Neoplasm of Bone Metastases) and a radiation treatment delivery code 77401-77416. The delivery codes were counted for frequency on consecutive days (one per day and with no more than 3 days off in-between) to determine the total number of days treated. Descriptive statistics were used to summarize and describe the data found.

  1. The reader should be able to describe the three standard dose fractionations of palliative bony metastases most commonly prescribed.
  2. The reader should be able to describe what patient population most commonly receives the single-fraction prescription.
  3. The reader should be able to describe why single-fraction treatment is recommended by ASTRO for most patients.

The dosimetric benefits of voluntary deep inspiration breath hold, V-DIBH, treatments for left breast irradiation
Jamie Christ
Student Dosimetrist
Southern Illinois University Medical Dosimetry Program/Rush University Medical Center

Abstract:

Background: Adjuvant radiation therapy is the standard of care for women following breast-conserving surgery for breast cancer.  Whole breast radiation therapy is typically delivered using two opposed tangential beams. To deliver adequate dose to the breast tissue, portions of the cardiac structures and left lung also receive low doses of radiation due to their proximity to the chest wall. V-DIBH treatment is a technique in which patients hold their breath on inspiration during beam delivery. V-DIBH can be used to reduce dose to the cardiac structures, such as the cardiac ventricles and whole heart. During deep inspiration, the heart is pushed posteriorly and inferiorly from the chest wall, increasing its distance to the radiation fields. The goal of this study is to determine the dosimetric benefits of V-DIBH, expressed as the reduction in dose to the cardiac ventricles, heart, and left lung in comparison to the dose that would be received if the patient were to be treated utilizing standard free breathing (FB) technique.

Methods and Materials: 14 left-sided breast cancer patients that have undergone tangent beam radiation to the whole breast using v-DIBH were retrospectively selected. Each patient had undergone an initial CT simulation. During the simulation, the patient was scanned once using a FB pattern and again using V-DIBH technique. Plans were developed on the V-DIBH scan. These optimized plans used to treat with V-DIBH were transferred and registered to the FB image set. A visual inspection was performed for each case and the dose distribution on the FB scan was considered clinically acceptable. The DVHs for the heart, cardiac ventricles and left lung were extracted for both breathing modalities and processed outside of the treatment planning system. Average, minimum and maximum DVHs were generated for each modality and organ at risk.  Mean heart dose, the volume of the left lung receiving 20Gy (V20) the mean dose to the cardiac ventricles and the volume cardiac ventricles receiving 25Gy (V25) and 35Gy (V35) respectively were tabulated for each modality. The samples were compared using a t-test: two-sample assuming unequal variances statistical analysis.

Results: The composite average DVHs for the cardiac ventricles, heart and left lung shows significantly lower dose using V-DIBH treatment modality compared with FB. Compared with FB plans the V-DIBH plans show a reduction of the mean dose to the cardiac ventricles of 212 cGy (207cGy versus 419 cGy). The statistical analysis shows the following results: the average dose to the cardiac ventricles is reduced by about 130 cGy (p=0.045), V25 by 2.5% (p=0.038), and V35 by 1.5% (p=0.048). For the heart, the mean dose was reduced by 70 cGy (p=0.049) while for the left lung the dose was reduced by 7% (p=0.032).

Conclusions: The benefits of using V-DIBH technique translate into statistically significant dosimetric and theoretical clinical benefits for the cardiac structures as well as the ipsilateral lung.

  1. The reader should be able to answer what effect voluntary deep inspiration breath hold, V-DIBH, technique has on the position of the heart in relation to the treatment beams.
  2. The reader should be able to describe the effect of V-DIBH versus free breathing technique on the mean dose to the cardiac ventricles.
  3. The reader should be able to describe the effect of V-DIBH versus free breathing technique on the mean dose to the heart.

Auto-Planning: A Dosimetric comparison of Phillips Pinnacle Auto-planning tool for Lung, Brain, Abdomen, Head and Neck, and Prostate treatment sites.
Hasti Hesami

University of Texas MD Anderson Cancer Center

Abstract:
Auto-Planning: A Dosimetric comparison of Phillips Pinnacle Auto-planning tool for Lung, Brain, Abdomen, Head and Neck, and Prostate treatment sites.

Hasti Hesami, Amalia Flores, Sanya Makahani, Ancy Mathew, Elizabeth Medrano, Hirra Siddiqui, Peter Balter, PhD, Jordan Sutton, CMD

Medical Dosimetry Program, School of Health Professions, The University of Texas, M,D, Anderson Cancer Center, Houston, Texas, USA.

ABSTRACT

The purpose of this study is to evaluate the dosimetric differences between Phillips Auto-planning tool and the clinically approved and treated Volumetric Modulated Arc Therapy (VMAT) plans for multiple treatment sites. A total of thirty patients, six from each site, are re-optimized using identical plan parameters as the previously treated physician approved plans. The Auto-plans are generated by inserting the dose and normal tissue constraints as requested per physician’s planning directive for each patient. The Auto-plan trials are optimized once and the plan differences are evaluated by comparing the target volumes, organs at risk (OARs), monitor units (MU), and dose conformality to determine overall plan quality. Our results indicate that Auto-planning can generate a clinically acceptable treatment plan based on the parameters that are initially set. The target coverage and organs at risk are achieved in most cases. A greater difference is noticed for more complex treatment sites, such as head and neck and lung. High isodose conformality is reduced compared to clinically approved plans in these cases. On average, the trials created for Auto-planning show less monitor units as opposed to the clinically approved plan. Initial plan creation by Philips Auto-planning tool is shown to produce clinically treatable plans, however further optimization would be advantageous.

Keywords: Phillips Pinnacle, Auto-planning, VMAT

  1. The reader should compare the target coverage, OAR dose, integral dose, and monitor units (MUs) for Volumetric Arc Therapy (VMAT)-Auto-plans to already treated and approved clinical VMAT plans.
  2. The reader will be able to conclude that Auto-planning is an acceptable starting point for further optimization and will save planning time.
  3. The reader will find that the auto-planning tool can
  4. decrease (MUs) by decreasing the possible over-optimization that can occur by the planner.

Preservation of normal brain tissue in patients with multiple metastatic lesions using gamma knife radiosurgery
Marissa Tranquilli
Student
Roswell Park Cancer Institute

Abstract:

Traditionally, patients with metastatic brain lesions have been treated using whole brain radiotherapy (WBRT).   More recently, studies have suggested that treatment with stereotactic radiosurgery (SRS) alone may result in preservation of neurocognition, with no change in overall survival.  The total volume of brain treated may be directly linked to neurocognitive decline, therefore SRS may help preserve patient quality of life by sparing as much normal tissue as possible.  Standard practice has been to favor WBRT if more than four metastatic lesions are present; however, in recent years SRS has been safely used for patients with 10 or more lesions.    Here, we analyze the volume of normal brain tissue treated using gamma knife radiosurgery in patients with varying numbers of brain metastases (from one to twenty nine).   The purpose of the study is to re-evaluate the appropriate number of brain lesions for which SRS may be useful in terms of sparing normal brain tissue, and therefore potentially improving patient quality of life.

Materials and Methods

Approximately 500 patients with brain metastases (of any primary) treated with gamma knife radiosurgery at a single institution were identified using a database search.  Using the patient’s pre-gamma knife MRIs, brain tissue was segmented as accurately as possible using Eclipse software and then exported into the Gamma Knife planning system.   For each plan, the following data points were recorded: total brain volume, number of lesions treated, volume of brain receiving 8 Gy (V8), V10, V12, and V15.  Volume of brain treated was analyzed as a function of number of lesions, as well as total lesion volume.

Results

Using the V12 data set as indicative of the overall results, the volume of brain treated did not correlate with number of lesions treated.  Similar statistics existed for the V8, V10, and V15 data sets. The volume of brain treated did correspond, however, with the total volume of the brain metastases. These relationships are clearly demonstrated through the graphical data representations.

Conclusions

The number of metastatic brain lesions does not correlate with total brain volume treated when using stereotactic radiosurgery (SRS).  Although standard practice has favored using WBRT in patients with greater than 4 lesions, a significant amount of normal brain tissue may be spared by treating these patients with SRS.   This may translate into preservation of neurocognition and quality of life.  SRS should be carefully considered in patients with multiple brain lesions as an alternative to WBRT, with the emphasis on total brain volume involved rather than the number of lesions.

  1. The reader should be able to identify and describe previous WBRT and SRS brain treatment paradigms.
  2. The reader should understand the correlation between number of metastases treated and the volume of brain treated.
  3. The reader should be able to discuss which situations would show preference to SRS  over WBRT.

Dosimetric Comparison Between Pinnacle3 Auto-Planning and Manual Planning for Lung SBRT Treatments
Aaron Bishop

University Texas Health Science Center San Antonio

Abstract:
Introduction:  Stereotactic body radiation therapy (SBRT) has become an effective treatment option for early stage and small metastatic lung tumors. Due to plan complexity with SBRT, traditional, inverse optimized IMRT planning may be time consuming and the quality of plans is likely to vary with the skill and experience of the planner. Auto-Planning, a new functionality developed by Pinnacle3, has been made commercially available with Version 9.10. Auto-Planning is a tool that automates and facilitates inverse optimization of treatment plans. Benefits of Auto-Planning are based on planning efficiency improvements and standardization of plan quality. The aim of this study is to dosimetrically compare the treatment plan quality of Auto-Planning against previously approved clinical plans.

Methods and Materials:  Nine (n=9) lung SBRT patients previously treated using a non-coplanar, 6 MV fix-field IMRT technique with a 120 HDMLC Novalis Tx were replanned in Pinnacle3 (v9.10) with Auto-Planning. Patient plans were normalized to 5000 cGy in 5 fractions such that at least 98% of the PTV received 100% of the prescription dose. The same beam geometries were used for Auto-Planning as the clinical plan. An Auto-Planning template, derived from previously treated patients, was developed based on optimized planning objectives and TG101 dose tolerances to critical structures. The template was applied to all patients. Metrics used for comparison were the dose fall-off (RX = VX%/Vptv) at the 70%, 50%, and 30% isodose lines, conformation number (CN), and homogeneity index (HI = D0.2cc / D98%).  Statistical differences were evaluated using a paired sample Wilcoxon signed rank test with significance level of 0.05.

Results: For the dose fall-off of parameters, a mean percentage difference improvement of 1.6%, 0.4% and 2.9% was found for the R70, R50 and R30, respectively, using Auto-Planning—however no statistically significant difference was noted. The conformation number and homogeneity index showed a mean percentage difference of 2.73% and 4.73%, respectively—no statistically significant difference was noted between Auto-Planning and clinical plans.  Additionally, Auto-Planning did show an increase, on average, of 7.94% (p<0.05) in total monitor units over clinical plans.

Conclusions:  Based on the comparison metrics evaluated, Auto-Planning appears to generate SBRT treatment plans for lung lesions of similar treatment plan quality to the manually optimized, clinical plans. No statistically significant differences were note for the majority of the target dosimetric parameters. Statistically significant differences were only noted in the total treatment plan monitor units. 

  1. To study Pinnacle's Auto-Planning System.
  2. To study planning for Lung SBRT.
  3. To compare Auto-Planning with manual treatment planning.

Evaluating Treatment Plan Quality between Manual Planning and Auto-Planning in Patients with Prostate and Seminal Vesicle Irradiation
Tyler Marston
Student
Department of Radiation Oncology, The University of Texas Health Science Center at San Antonio

Abstract:
Evaluating Treatment Plan Quality between Manual Planning and Auto-Planning in Patients with Prostate and Seminal Vesicle Irradiation                           

T Marston, B.Sc., TY Eng, M.D., D Baacke, B.Sc., N Kirby, Ph.D., N Papanikolaou, Ph.D., AN Gutierrez, Ph.D.          

Department of Radiation Oncology, University of Texas Health Science Center San Antonio, San Antonio TX

INTRODUCTION: Auto-Planning is a tool that automates and facilitates inverse optimization of treatment plans. Benefits of Auto-Planning are based on planning efficiency improvements and standardization of plan quality. With this in mind, the purpose of this treatment planning study is to dosimetrically assess the quality of Auto-Planning plans generated for patients receiving treatment of the prostate and seminal vesicles for low-/intermediate-risk prostate cancer. The primary objective is to compare the quality of Auto-Planning treatment plans against previously approved clinical treatment plans.

METHODS AND MATERIALS: Fifteen (n=15) patients previously treated to the prostate and seminal vesicles using a VMAT technique were replanned with Auto-Planning. All patients were planned in Pinnacle3 (Version 9.10) using single or dual SmartArc plans with either 6 or 10 MV photon energies. Plan settings were equivalent for each patient for both clinical and Auto-Planning. Plans were prescribed to 7920cGy and normalized such that 95% of the PTV received at least 100% of the prescription dose. A template for Auto-Planning was initially developed using the dose constraints to critical structures from QUANTEC. A final template was adjusted based on the preliminary test results of six patients. Patients were evaluated for PTV and organ at risk (OAR) parameters for the bladder, rectum, sigmoid, penile bulb, and femoral heads. Metrics used for PTV comparison were D2cc, homogeneity (HI = D2cc / DRx), conformity (CN), and dose fall-off (Rx = Vx% / VPTV) for the 70%, 50%, and 30% isodose lines. OAR metrics included D2cc and DMEAN along with specific volume based constraints for the bladder and rectum per QUANTEC.

RESULTS: For the PTV, a decrease in the mean percentage difference of 1.7% (p<0.05) was noted in the CN as well as an increase in HI of 1.4% (p<0.05) and total plan MUs of 20.6% (p<0.05) for Auto-Planning. No statistically significant differences were noted for the other PTV parameters. For OARs, statistically significant reductions in the DMEAN of all structures, decrease in the D2CC of the sigmoid of 18.1% (p<0.05), and decrease in the V65Gy of 4.7% (p<0.05) and V50Gy of 11.4% (p<0.05) for the rectum were noted with Auto-Planning. Statistically significant increases in the mean percentage difference in the D2cc of 1.7% (p<0.05) and V80Gy of 21.8% (p<0.05) for the bladder, and D2cc increase of 1.4% (p<0.05) for the rectum was shown.

CONCLUSIONS: The study showed that Auto-Planning produces plans of comparable quality to clinical plans based on the evaluated metrics for the sample of patients studied. Clinical plans showed better homogeneity and less total plan MUs. Auto-Planning had lower average doses (DMEAN) for all critical structures.

  1. To improve IMRT planning efficiency
  2. To standardize treatment plan quality
  3. To test Pinnacle's new Auto-Planning feature

Investigating the Dosimetric Differences between Manual Planning and Auto-Planning in Patients with Prostate and Pelvic Lymph Nodes Irradiation
Aziz Lirani
UTHSCSA

Abstract:

INTRODUCTION: Volumetric modulated arc therapy (VMAT) is now a common technique used in irradiating cancer of the prostate and pelvic lymph nodes due to its improved delivery efficiency over conventional IMRT. However, VMAT planning requires a consistent, proficient, and time-consuming effort from the planner to obtain an optimal plan. Auto-Planning, a new functionality developed by Pinnacle3, has been made commercially available with Version 9.10. Auto-Planning is a tool that automates and facilitates inverse optimization of treatment plans. Benefits of Auto-Planning are based on planning efficiency improvements and standardization of plan quality. The aim of this study is to dosimetrically compare the treatment plan quality of Auto-Planning against previously delivered clinical plans.

METHODS AND MATERIALS: Ten patients (n=10) who were previously treated with a dual or triple arc, 6 MV VMAT technique to the prostate and pelvic lymph nodes were re-planned using Auto-Planning in Pinnacle3 (v9.10). Auto-Planning plans were created with the same beam geometry, isocenter, and dose grid as the clinical plan. An Auto-Planning template was created and utilized for all of the plans. Plans were subsequently modified with a maximum of three warm-starts to further enhance the plan. Both clinical and Auto-Planning plans were normalized such that 95% of the PTV received at least 100% of the prescription dose. Patients were evaluated for PTV and organ at risk (OAR) parameters for the rectum, bladder, sigmoid, small bowel, penile bulb, and the right and left femoral heads. Metrics used for comparison were Dmean, D2cc, homogeneity (HI), conformity (CN), dose fall-off (R50), and total monitor units (MUs) for the PTV, and Dmean, D2cc, V80%, V60%, V40%, and V20% for the OARs. Statistical differences were evaluated with a paired-sample Wilcoxon signed rank test with a significance level of 0.05.

RESULTS:

For the PTV, statistically significant improvements in the median percentage difference of 2.0% were noted for both D2cc and HI for Auto-Planning plans. An increase in the median percentage difference of R50 of 6.5% (p<0.05) was noted for Auto-Planning. With regards to the OARs, all structures except the penile bulb, had a statistically significant reduction in Dmean. The bladder and femoral heads showed an improvement in D2cc. The rectum, bladder, sigmoid, and small bowel all showed improvement in the V40%.

CONCLUSIONS:

Auto-Planning generates similar, if not better, treatment plan quality as compared to previously delivered clinical plans for VMAT based pelvic irradiation based on the metrics evaluated in this study. Auto-Planning appears to facilitate the treatment optimization process all the while making the overall planning process less laborious and time consuming.

  1. To collect and organize clinical data
  2. To become familiar with the Auto-Planning technique
  3. To evaluate the research critically and compare new data with prior findings

Evaluation of Quality and Delivery of Head and Neck treatment plans created with Pinnacle Auto-Plan technique
Kalpana Vaidya
Senior Dosimetrist
Northwell Health system

Abstract:
Evaluation of Quality and Delivery of Head and Neck treatment plans created with Pinnacle Auto-Plan technique

Purpose: This study is to assess the quality and feasibility of selected head and neck treatment plans generated with the Auto-planning module in Pinnacle treatment planning system. The dosimetry and delivery of the Auto-plan VMAT and IMRT after initial optimization were retrospectively compared with previously treated plans.

Materials and Methods: 10 patients with oropharyngeal or hypopharyngeal cancers were reviewed to test the consistency of the Pinnacle auto planning function.  Pinnacle v9.10 auto-plan module was used to generate VMAT (2 Arcs) and IMRT (9 fields) head and neck plans based on the adapted RTOG protocol dosimetric requirements  for target coverage as well as organ at risk (OAR) sparing. Treatment planning CTs were taken with Siemens Somatom AS CT scanner in supine, head-first position using a custom aquaplast immobilization mask. Alignment tattoos were given at the time of simulation. CT images had slice spacing of 3 mm. All the plans were constructed with concurrent boost technique of 3 dose levels of 70, 63, and 58/54Gy corresponding to average volumes of 125, 510 and 256cc respectively. All plans were calculated with grid size of 4mm. These plans were compared with dosimetrist generated plans for dose distribution and delivery assessment.

Results and Discussion: The auto-plans were performed with minimal manual interaction. The planning time required for initial optimizing iteration for each VMAT Auto-plan was between 110 to 120 minutes, each IMRT Auto-plan required 30 to 60 minutes. The delivery time and monitor units decreased for VMAT Auto-plan and increased for IMRT Auto-plan compared with plans constructed manually. The following criteria was used to assess dose distribution: 95% of target volume achieves greater than 95% of prescribed dose and  mean dose to larynx and  both parotids  as well as spinal cord max