Dosimetric dependences of bone heterogeneity and beam angle on theunflattened and flattened photon beams: A Monte Carlo comparison

James C.L. Chow a,b,n, Amir M. Owrangi c

a Department of Radiation Oncology, University of Toronto, Toronto, Canadab Department of Radiation Physics, Princess Margaret Cancer Center, Toronto, Canadac Department of Radiation Oncology, University of Michigan Health System, Ann Arbor, MI 48109, USA

H I G H L I G H T S

� Dosimetric impact of bone heterogeneity in the presence of unflattened and flattened photon beams was compared.� Surface and depth dose of unflattened and flattened photon beams varying with the beam angle were compared.� Depth dose deviation due to the presence of bone was sensitive to the beam obliquity.� Surface dose deviation between the unflattened and flattened beams became smaller with an increase of beam angle.� Surface dose and range of depth dose ratios (unflattened to flattened beam) decreased with an increase of beam angle.

a r t i c l e i n f o

Article history:Received 10 February 2014Accepted 29 March 2014Available online 13 April 2014

Keywords:Unflattened photon beamSurface doseBeam obliquityBone heterogeneityMonte Carlo simulation

a b s t r a c t

The variations of depth and surface dose on the bone heterogeneity and beam angle were comparedbetween unflattened and flattened photon beams using Monte Carlo simulations. Phase-space files ofthe 6 MV photon beams with field size of 10�10 cm2 were generated with and without the flatteningfilter based on a Varian TrueBeam linac. Depth and surface doses were calculated in a bone and waterphantoms using Monte Carlo simulations (the EGSnrc-based code). Dose calculations were repeated withangles of the unflattened and flattened beams turned from 01 to 151, 301, 451, 601, 751 and 901 in thebone and water phantoms. Monte Carlo results of depth doses showed that compared to the flattenedbeam the unflattened photon beam had a higher dose in the build-up region but lower dose beyond thedepth of maximum dose. Dose ratios of the unflattened to flattened beams were calculated in the rangeof 1.6–2.6 with beam angle varying from 01 to 901 in water. Similar results were found in the bonephantom. In addition, higher surface doses of about 2.5 times were found with beam angles equal to 01and 151 in the bone and water phantoms. However, surface dose deviation between the unflattened andflattened beams became smaller with increasing beam angle. Dose enhancements due to the bonebackscatter were also found at the water–bone and bone–water interfaces for both the unflattened andflattened beams in the bone phantom. With Monte Carlo beams cross-calibrated to the monitor unit insimulations, variations of depth and surface dose on the bone heterogeneity and beam angle wereinvestigated and compared using Monte Carlo simulations. For the unflattened and flattened photonbeams, the surface dose and range of depth dose ratios (unflattened to flattened beam) decreased withincreasing beam angle. The dosimetric comparison in this study is useful in understanding thecharacteristics of unflattened photon beam on the depth and surface dose with bone heterogeneity.

& 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The use of flattening filter free or unflattened photon beam hasbecome popular in radiotherapy recently (Cashmore et al., 2011;

Fu et al., 2004; O’Brien et al., 1991; Titt et al., 2006; Vassiliev et al.,2009). Traditionally, in external beam conformal radiotherapy,flattened photon beams are created by a flattening filter in thelinac head to produce a homogeneous dose distribution at thetumor target. However, for newer radiation dose delivery techni-ques such as intensity modulated radiotherapy and volumetricmodulated arc therapy, the segmental field fluence is spatially andtemporally modulated by the dose rate and field shape generatedby the multileaf collimator. Therefore, flattened photon beam

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/radphyschem

Radiation Physics and Chemistry

http://dx.doi.org/10.1016/j.radphyschem.2014.03.0410969-806X/& 2014 Elsevier Ltd. All rights reserved.

n Corresponding author at: Department of Radiation Physics, Princess MargaretCancer Center, University Health Network, 610 University Avenue, Toronto, ON,Canada M5G 2M9. Tel.: þ1 416 946 4501; fax: þ1 416 946 6566.

E-mail address: james.chow@rmp.uhn.on.ca (J.C.L. Chow).

Radiation Physics and Chemistry 101 (2014) 46–52

seems to be unnecessary (Vassiliev et al., 2006a, 2006b; Nicoliniet al., 2011). Moreover, for stereotactic radiotherapy, whether thephoton beam is flattened or not does not give significant deviationin dose distribution, because the peak of the unflattened beamprofile is significant merely in large field size (Mok et al., 2011;Georg et al., 2011).

When the flattening filter of photon beam is removed, theweight of low-energy photon fluence in the beam increases (Sixeland Faddegon, 1995; Titt et al., 2006; Mesbahi and Nejad, 2008).This makes the unflattened photon beam less preventive, resultingin a lower depth dose compared to the flattened beam in water. Onthe other hand, in the absence of flattening filter at the centralbeam axis, the head scatter and leakage are reduced (Cashmore,2008; Kragl et al., 2009). This leads to a lower out-of-field orperipheral dose. However, the main advantage of using theunflattened beam is still the increase of dose rate by removingthe flattening filter. It is reported that with the newly developedbeam generation system, the Varian TrueBeam linac is able todeliver dose rates up to 2400 monitor unit per minute (Fu et al.,2004). This would definitely decrease the treatment time andincrease the patient throughput in radiotherapy. On the otherhand, such high dose per pulse of unflattened beams would affectthe current radiobiological models on the evaluation of cancer cellsurvival (Lohse et al., 2011). As removing the flattening filter canincrease the radiation output, and decrease the head scatter andleakage, cancer treatment deliveries using intensity modulatedradiotherapy, volumetric modulated arc therapy and stereotacticradiotherapy technique with unflattened photon beams can bebenefited by the improvement in dose delivery efficiency.

The increase in weight of low-energy photon and decrease inhead scatter and leakage also affect the surface dose of patient.The increase in the number of low-energy photons, which shouldbe removed by the flattening filter, from the unflattened beamcontributes energy deposition in the build-up region of thepatient, and therefore increases the surface dose compared tothe flattened beam. However, decreases of head scatter andleakage due to the absence of flattening filter from the unflattenedbeam lower the surface dose. It is interesting to investigate whichof the above factors is more significant, leading to an increase ordecrease of surface dose for the unflattened photon beam (Wanget al., 2011). Apart from the presence of flattening filter whichwould affect the surface dose, beam obliquity and bone hetero-geneity in the build-up region would also have impact on thesurface dose (Chow and Grigorov, 2007; Chow and Owrangi, 2012).It is noted that the photon fluence in the depth of patient would beaffected by the obliquity of central beam axis, which results in ahigher surface dose (Chow et al., 2010). Furthermore, the presenceof bone heterogeneity would produce bone dose enhancementtowards the patient surface (Chow and Owrangi, 2011). To date,though there are studies on the surface dose in unflattened photonbeams (Wang et al. 2011; Huang et al. 2012), there is no relatedstudy concerning surface dose variations on the bone heteroge-neity and beam angle.

In this study, Monte Carlo simulation was used to predict thedepth and surface dose from the unflattened and flattened photonbeams. The Varian TrueBeam linac was modeled here because itcan produce unflattened and flattened megavoltage photon beamsby selecting the flattening filter absence or presence in the centralbeam axis. Although there were studies on Monte Carlo simula-tions based on unflattened photon beams, dose calculations weremostly based on a simulation model by removing the flatteningfilter from a typical linac, which did not have the feature ofproducing a real unflattened beam (Titt et al., 2006; Vassilievet al., 2006a; Ponisch et al., 2006; Dalaryd et al., 2010; Parasaiet al., 2007). In addition, the Monte Carlo beams were cross-calibrated to the machine monitor unit in simulations of the

unflattened and flattened photon beams. This is importantbecause the Monte Carlo results can inform us the deviation ofradiation output between the two beams with the same sourceoutput. In measurement, this is difficult to determine because theoutput of unflattened photon beam would have been calibrated bythe dosimetric protocol as if the flattened beam is in water(Hrbacek et al., 2011). The aim of this study is to comparevariations of depth and surface doses on bone heterogeneity andbeam angle, between unflattened and flattened photon beams.Monte Carlo simulation using the EGSnrc-based code was used(Kawrakow and Rogers, 2000).

2. Experimental

2.1. Phantom and calculation geometry

Fig. 1 shows the bone heterogeneity phantom and beamgeometry in this study. In Fig. 1, unflattened and flattened photonbeams of 6 MV with field size equal to 10�10 cm2, used through-out this study, irradiated a phantom with a bone layer of 2 cmthickness. The bone layer was positioned under a 1 cm thick waterlayer on top of the phantom. This made the bone heterogeneity inthe build-up region of the photon beams with depth of maximumdose equal to 1.5 cm. The bone density was equal to 1.75 g/cm3

and the ICRPBONE700ICRU bone was used containing elements H,C, N, O, Mg, P, S, Ca and Zn in ratios of 4.69, 1.2, 0.29, 2.79, 0.0091,0.34, 0.098, 0.52 and 0.00015, respectively (ICRP, 1975). Theisocenter was set at a depth of 10 cm and the source-to-surfacedistance was equal to 90 cm. Apart from the photon beam angle of01 as shown in Fig. 1, the beamwas turned to 151, 301, 451, 601, 751

Fig. 1. Schematic diagram (not to scale) showing the calculation geometry of thebone phantom using the unflattened and flattened photon beams. The isocenter isat a depth of 10 cm from the phantom surface. The photon beams were rotatedfrom 01 to 901 clockwise and the thickness of bone was equal to 2 cm. Dosecalculations were repeated using the same beam geometry but a phantomwith thebone replaced by water for dosimetric comparison.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–52 47

and 901 clockwise. Depth and surface doses were calculated at thephantom surface and central beam axis (vertical broken line inFig. 1) using Monte Carlo simulations. For dosimetric comparison,calculations were repeated using a water phantom with the bonelayer substituted by water.

2.2. Monte Carlo simulation

Dose calculations were carried out by Monte Carlo simulationusing the EGSnrc-based code developed by the National CouncilCanada (Kawrakow and Rogers, 2000). The head of the VarianTrueBeam linac was modeled by the Varian Monte Carlo teamusing the Geant4 Monte Carlo code and related works werepublished elsewhere (Constantin et al., 2011a, 2011b). In thisstudy, phase-space files of the unflattened and flattened photonbeams generated by Constantin et al. (2011b) were used in oursimulations (Constantin, 2011c). These phase-space files generatedby the Geant4 code simulated the linac head from the sourcedown to the space above the jaw. Based on this data, the BEAMnrcwas used to create phase-space files of the unflattened andflattened photon beams by including the X and Y jaws in thesimulation model (Rogers et al., 2011). The field size was set to10�10 cm2 at the isocenter. The phase-space files generated bythe BEAMnrc contained 1�109 particles and were used in ourdose calculations.

Depth and surface doses of the bone (Fig. 1) and correspondingwater phantom (bone replaced by water) were calculated usingthe DOSXYZnrc code (Walters and Rogers, 2009). In Monte Carlosimulations, voxel size of 0.2�0.2�0.2 mm3 corresponding to thex-, y- and z-axis was used. One hundred million histories were runin dose calculations per beam geometries as shown in Fig. 1. TheMonte Carlo beams were calibrated to allow the incorporation ofmonitor units based on the method by Popescu et al. (2005). Theenergy cutoffs for electron and photon transport were set to700 keV and 1 keV, respectively. Using this parameter setting,the relative dose error or statistical uncertainty as a fraction ofdose in the voxel was found to be around 1% for all calculations.

Verification of the Monte Carlo model was carried out bycomparing the commissioning data of the Varian TrueBeam linacusing the ionization chamber and scanning water tank. Fig. 2shows the percentage depth doses of the 6 MV unflattened andflattened photon beams calculated and measured by Monte Carlosimulation and ionization chamber, respectively. The field size ofphoton beams was 10�10 cm2 and the source-to-surface distance

was equal to 100 cm. It is seen in Fig. 2 that the unflattened photonbeam has a higher depth dose in the build-up region than that ofthe flattened beam. This is because the low-energy photons, whichhave been removed by the flattening filter, contribute to the dosein the build-up region. This dosimetric effect is stronger thanlosses of head scatter and leakage due to the absence of theflattening filter. The increase of dose in the build-up region of theunflattened beam makes the dose build up quicker and it has asmaller depth of maximum dose than that of the flattened beam.However, the flattened photon beam is more penetrative than theunflattened in water. It is because of the beam hardening effect inthe presence of flattening filter absorbing low-energy photonsfrom the beam.

3. Results

Fig. 3(a) and (b) shows respectively relative depth doses of theunflattened and flattened photon beams varying with differentbeam angles (0–901) in the water phantom. All depth doses werenormalized to the maximum doses of the unflattened (Fig. 3(a))and flattened (Fig. 3(b)) photon beams with beam angle equal to01. According to the beam geometry in Fig. 1, it is seen that thedepth dose of photon beam at angle equal to 901 is half of thebeam profile of the beam with angle equal to 01. Fig. 3(c) and(d) shows relative depth doses of the photon beams with the samebeam geometries and normalizations as those of Fig. 3(a) and (b),but the bone phantom (Fig. 1) is used, respectively. The boneheterogeneity is marked at depths between 1 and 3 cm as shownin Fig. 3(c) and (d). Fig. 4(a) shows relative depth doses of theunflattened and flattened photon beams in water and bonephantoms. The beam angle was equal to 01 and all depth doseswere normalized to the maximum doses of the flattened photonbeams in the water phantom. Fig. 4(b) shows the same photonbeam geometries and phantoms as those of Fig. 4(a) except thatthe beam angle is changed to 301. The depth dose ratios of theunflattened to flattened photon beams in a variation of beam angleare shown in Fig. 5(a) and (b) using the water and bone phantoms.The relative surface doses for the unflattened and flattened photonbeams varying with the beam angle are shown in Fig. 6(a) and (b),using the water and bone phantoms. In Fig. 6(a) and (b), all dosesare normalized to the maximum surface doses of the flattenedphoton beams in the water and bone phantoms, respectively.

4. Discussion

In conventional radiotherapy using megavoltage photon beams,surface dose may not be a clinical concern due to the skin-sparingeffect of the beams. However, unflattened photon beams are usedfor stereotactic radiotherapy with a hypofractionated scheme. Thisinvolves a very high dose (46 Gy) delivered to the patient perfraction, and it would result in an acute skin reaction in the courseof treatment (Wang et al., 2011). Therefore, it is clinically sig-nificant to investigate the depth and surface dose characteristicsusing the unflattened and flattened photon beams.

4.1. Depth dose variation with beam angle in the presenceof flattening filter

In Fig. 3(a) showing the variation of relative depth doses withthe beam angle, it is seen that the depth of maximum doseincreased with an increase of the beam angle (0–301). On theother hand, the depth doses in Fig. 3(a) decrease with the beamangle because the central beam axes of oblique beams (angle401)are moved away from the beam axis with angle equal to 01

Fig. 2. Percentage depth doses of the unflattened and flattened photon beams bymeasurements and Monte Carlo simulations in water.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–5248

(i.e. vertical broken line in Fig. 1). With the photon beam turnedfrom 01 to 901 clockwise, the relative depth doses along the centralbeam axis at angle 01 became half of the beam profiles at 10 cmdepth in water, with another half in the air above the phantomsurface. Therefore, the depth doses of beams with angles larger

than 451 increased with depth, which in fact were the horns of thebeam profiles. Comparing the relative depth doses of unflattenedphoton beams (Fig. 3(b)) with those of flattened, it is found thatthe deviation of depth doses at beam angle 01 and 151 for theunflattened beams was larger than that of the flattened beams.

Fig. 3. Relative depth doses of the (a) flattened (water phantom), (b) unflattened (water phantom), (c) flattened (bone phantom) and (d) unflattened (bone phantom) photonbeams with beam angles of 01, 151, 301, 451, 601, 751, and 901. All doses were normalized to the maximum doses of the photon beams (flattened or unflattened) with angleequal to 01.

Fig. 4. Relative depth doses of the unflattened and flattened photon beams in the water and bone phantoms. The beam angles were set to (a) 01 and (b) 301. For both (a) and(b), all doses were normalized to the maximum dose of the flattened photon beams with beam angle equal to 01 in the water phantom.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–52 49

The more significant decrease of relative depth doses with anincrease of beam angle of the flattened beam was due to thecontribution of more high-energy photons along the depth inwater. For depth doses near the phantom surface, it is seen thatthe unflattened beams were higher (about 49%) than those of theflattened beams (about 40%) with beam angles equal to 01 and 151.This is due to the presence of low-energy photons, which wereremoved by the flattening filter, deposited dose at the phantomsurface.

When the water phantom is replaced by bone phantom (Fig. 1),it is seen in Fig. 3(c) that depth doses increase above the water–bone interface at a depth of 1 cm for the flattened beams withangles equal to 01 and 151. Similar results are found just above thewater–bone interface in Fig. 3(d) for the unflattened beams, andthis is because of the bone backscatters. At the bone–waterinterface at a depth of 3 cm in Fig. 3(c) and (d), depth doses inbone decrease due to the decrease of backscatter from watercompared to bone beyond the depth larger than 3 cm. For relativedepth doses with beam angles between 601 and 901, depth dosesranging from 7 to 10 cm are higher than those in Fig. 3(a). This isdue to the presence of bone as shown in Fig. 1. In Fig. 3(d), theincrease of dose due to bone backscatter at 1 cm depth is morethan that in Fig. 3(c), as the backscatter of the unflattened beamcontained higher weight of low-energy photons than that of

the flattened beam. For beams with angles equal to 01 and 151,their depth dose deviations in Fig. 3(d) are larger than those inFig. 3(c) in the bone. It shows that the presence of flattening filteraffects the photon energy distribution with bone heterogeneityresulting in a variation of depth dose.

4.2. Depth dose variation with bone heterogeneity in the presenceof flattening filter

Depth doses of photon beams with and without a flatteningfilter can be found in Fig. 4 using the bone phantom (Fig. 1). Fordosimetric comparison, depth doses in water using the same beamgeometry are also shown with beam angles equal to 01 (Fig. 4(a))and 301 (Fig. 4(b)). All doses in the figures are normalized to themaximum dose of depth dose for the flattened beam in water. It isseen that the depth dose of the unflattened photon beam wasabout two times of the flattened beam in both the water and bonephantoms. This result agreed with the radiation output deter-mined by Hrbacek et al. (2011). In Fig. 4, it is found that depthdoses are higher in the water phantom than bone phantomwhether the photon beams are turned to 301 or not. Doses inwater near the water–bone and bone–water interfaces werehigher due to backscatters of both the unflattened and flattenedbeams. In Fig. 4(a), doses in the bone are lower than those in

Fig. 5. Relationship between the dose ratio (unflattened to flattened photon beams) and depth in the (a) water and (b) bone phantoms using the photon beams withdifferent beam angles (0–901).

Fig. 6. Relationship between the relative surface dose and beam angle for the unflattened and flattened photon beams in the (a) water and (b) bone phantoms. All doseswere normalized to the surface dose of the flattened photon beams with angle equal to 01.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–5250

water, due to the higher relative electron density of bone. How-ever, depth doses increased in water (depth43 cm) beyond thebone–water interface. This is because of the contribution of bonebackscatter to water. In Fig. 4, with beam angle equal to 01, it isseen that depth doses did not vary significantly whether theflattening filter was present or not in the bone phantom. Whenthe beam angle was turned to 301, the depth of maximum dosewas shifted to larger depth in the bone as shown in Fig. 4(b). Themaximum dose in water for the flattened photon beam was lowerthan 100% because all doses were normalized to the maximumdose of the flattened beam at angle 01 in water (Fig. 4(a)). Thisshows that the beam path from the central axis of the obliquebeam is longer than the beam at angle 01. In Fig. 4(b), it is also seenthat the dosimetric variation due to the bone heterogeneity ishigher for beams at angle equal to 301 compared to 01, showingthat the dosimetric variation due to the bone heterogeneitybecomes more significant in oblique photon beams.

4.3. Depth dose ratio variation with beam angle

Fig. 5(a) and (b) shows depth dose ratios of unflattened toflattened photon beams varying with the beam angle in the boneand water phantoms. With the beam angle varying, doses of theunflattened photon beams increased between 1.6 and 2.6 timesalong the depth compared to the flattened beams in water.However, when the depth increased, the dose ratio rangedecreased with a variation of beam angle. Similar results couldbe found in the bone phantom. It showed that for larger depth,deviations between the unflattened and flattened beams becamesmaller with the beam angle variation. In Fig. 5(a), it is seen forbeam angles equal to 01 and 151 that depth doses increasedsignificantly within the build-up region. This can also be seen inthe bone phantom as shown in Fig. 5(b). Comparing the depthdose ratios of the water to bone phantom varying with the beamangle, it is found that they are very similar except that the doseratio range is slightly decreased in the bone (ranging betweendepths at 1 and 3 cm) as shown in Fig. 5(b).

4.4. Relative surface dose variation with the beam angle in thepresence of flattening filter

When the beam angle is equal to 01, the surface dose ofunflattened beam increases more than 2.5 times in water asshown in Fig. 6(a). However, such surface dose enhancement isdue to the absence of flattening filter, which decreases when thebeam angle increases. For beam angle larger than 451, it is foundthat deviation of surface doses between the unflattened andflattened beams is not significant in the water and bone phantoms.Since results in Fig. 6(b) are similar to those in Fig. 6(a), the boneheterogeneity does not affect the increase of surface dose due tothe presence of flattening filter. In addition, the bone heterogene-ity did not affect the deviation of surface dose due to the beamangle. From Fig. 4(a), it can be seen that the bone heterogeneityaffects doses only close to the bone–water interface but not thephantom surface.

5. Conclusions

Dosimetric comparisons on the bone heterogeneity and beamangle using the unflattened and flattened photon beams werestudied by Monte Carlo simulation. For the bone heterogeneity,dose enhancements due to the bone backscatter were found inboth the unflattened and flattened beams in water adjunct to thewater–bone and bone–water interfaces. For beam angles varyingfrom 01 to 901, depth dose ratios of the unflattened to flattened

beams were found in ranges of 1.6–2.6 to 1.9–2.0, when the depthincreased from 0 to 10 cm in water. For the relative surface dose,the unflattened photon beam was about 2.6 times higher than theflattened beam with angle equal to 01. The deviation of relativesurface dose between the unflattened and flattened beamsdecreased with an increase of beam angle in both the bone andwater phantoms.

Acknowledgments

The authors would like to acknowledge the Varian MonteCarlo research team for generating the phase-space data of theTrueBeam linac available in this study.

References

Cashmore, J., 2008. The characterization of unflattened photon beams from a 6 MVlinear accelerator. Phys. Med. Biol. 53, 1933–1946.

Cashmore, J., Ramtohul, M., Ford, D., 2011. Lowering whole-body radiation doses inpediatric intensity-modulated radiotherapy through the use of unflattenedphoton beams. Int. J. Radiat. Oncol. Biol. Phys. 80, 1220–1227.

Chow, J.C.L., Grigorov, G.N., 2007. Surface dosimetry for oblique tangential photonbeams: a Monte Carlo simulation study. Med. Phys. 35, 70–76.

Chow, J.C.L., Jiang, R., Leung, M.K.K., 2010. Dosimetry of oblique tangential photonbeams calculated by superposition/convolution algorithms: a Monte Carloevaluation. J. Appl. Clin. Med. Phys. 12, 108–121.

Chow, J.C.L., Owrangi, A.M., 2011. Monte Carlo study on mucosal dose in oral andnasal cavity using photon beams with small field. J. Radiother. Pract. 10,261–271.

Chow, J.C.L., Owrangi, A.M., 2012. Dependences of mucosal dose on photon beamsin head-and-neck intensity-modulated radiation therapy: a Monte Carlo study.Med. Dosim. 37, 195–200.

Constantin, M., Sawkey, D., Mansfield, S., Salop, A., 2011a. Extended True-Beampatient-independent phase space library for the 6� , 6� FFF, 10� , and 10� FFFradiotherapy beams calculations. Med. Phys. 38, 3642.

Constantin, M., Perl, J., LoSasso, T., Salop, A., Whittum, D., Narula, A., Svatos, M.,Keall, P.J., 2011b. Modeling the TrueBeam linac using CAD to Geant4 geometryimplementation: dose and IAEA-compliant phase space calculations. Med. Phys.38, 4018–4024.

Constantin, M., 2011c ⟨http://my.varian.com/montecarlo_reg/2⟩.Dalaryd, M., Kragl, G., Ceberg, C., Georg, D., McClean, B., Wetterstedt, S., Wieslander,

E., Knoos, T., 2010. A Monte Carlo study of a flattening filter-free linearaccelerator verified with measurements. Phys. Med. Biol. 55, 7333–7344.

Fu, W.H., Dai, J.R., Hu, Y.M., Han, D.S., Song, Y.X., 2004. Delivery time comparison forintensity-modulated radiation therapy with/without flattening filter: a plan-ning study. Phys. Med. Biol. 49, 1535–1547.

Georg, D., Knoos, T., McClean, B., 2011. Current status and future perspective offlattening filter free photon beams. Med. Phys. 38, 1280–1293.

Hrbacek, J., Lang, S., Klock, S., 2011. Commissioning of photon beams of a flatteningfilter-free linear accelerator and the accuracy of bam modelling using ananisotropic analytical algorithm. Int. J. Radiat. Oncol. Biol. Phys. 80, 1228–1237.

Huang, Y., Siochi, R.A., Bayouth, J.E., 2012. Dosimetric properties of a beam quality-matched 6 MV unflattened photon beam. J. Appl. Clin. Med. Phys. 13, 71–80.

ICRP, 1975. Report on the Task Group on Reference Man: Anatomical, Physiologicaland Metabolic Characteristics. ICRP, New York (ICRP Report 23).

Kawrakow, I., Rogers, D.W.O., 2000. EGSnrc Code System: Monte Carlo Simulationof Electron and Photon Transport. National Research Council of Canada, Ottawa,Canada (Technical Report PIRS-701).

Kragl, G., af Wetterstedt, S., Knausl, B., et al., 2009. Dosimetric characteristics of6 and 10 MV unflattened photon beams. Radiother. Oncol. 93, 141–146.

Lohse, I., Lang, S., Hrbacek, J., Scheidegger, S., Bodis, S., Macedo, N.S., Feng, J., Lutolf,U.M., Zaugg, K., 2011. Effect of high dose per pulse flattening filter-free beamson cancer cell survival. Radiother. Oncol. 101, 226–232.

Mesbahi, A., Nejad, F.S., 2008. Monte Carlo study on a flattening filter-free 18-MVphoton beam of a medical linear accelerator. Radiat. Med. 26, 331–336.

Mok, E., Kielar, K., Wang, L., Hsu, A., 2011. Evaluation of the accuracy of eclipse AAAmodeling for flattening filter free beam used for SBRT. Med. Phys. 38,3711–3712.

Nicolini, G., Ghosh-Laskar, S., Shrivastava, S.K., et al., 2011. Volumetric modulationarc radiotherapy with flattening filter-free beams compared with static gantryIMRT and 3D conformal radiotherapy for advanced esophageal cancer: afeasibility study. Int. J. Radiat. Oncol. Biol. Phys. 84, 553–560.

O’Brien, P.F., Gillies, B.A., Schwartz, M., Young, C., Davey, P., 1991. Radiosurgery withunflattened 6-MV photon beams. Med. Phys. 18, 519–521.

Parasai, E.I., Pearson, D., Kvale, T., 2007. Consequences of removing the flatteningfilter from linacs in generating high dose rate photon beams for clinicalapplications: a Monte Carlo study verified by measurement. Nucl. Instrum.Methods Phys. Res. B 261, 735–759.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–52 51

Ponisch, F., Titt, U., Vassiliev, O.N., Kry, S.F., Mohan, R., 2006. Properties ofunflattened photon beams shaped by a multileaf collimator. Med. Phys. 33,1738–1746.

Popescu, I.A., Shaw, C.P., Zavgorodni, S.F., Beckham, W.A., 2005. Absolute dosecalculations for Monte Carlo simulations of radiotherapy beams. Phys. Med.Biol. 50, 3375–3392.

Rogers, D.W.O., Walters, B.R.B., Ding, G.X., Sheikh-Bagheri, D., Zhang, G., 2011.BEAMnrc User Manual. National Research Council of Canada, Ottawa (NRCCReport No. PIRS-0509(a)Rev L).

Sixel, K.E., Faddegon, B.A., 1995. Calculation of x-ray spectra for radiosurgicalbeams. Med. Phys. 22, 1657–1661.

Titt, U., Vassiliev, O.N., Ponisch, F., Dong, L., Liu, H., Mohan, R., 2006. A flatteningfilter free photon treatment concept evaluation with Monte Carlo. Med. Phys.33, 1595–1602.

Vassiliev, O.N., Kry, S.F., Chang, J.Y., Balter, P.A., Titt, U., Mohan, R., 2009. Stereotacticradiotherapy for lung cancer using a flattening filter free Clinac. J. Appl. Clin.Med. Phys. 10, 2880.

Vassiliev, O.N., Titt, U., Kry, S.F., Ponish, F., 2006a. Monte Carlo study of photon fieldsfrom a flattening filter-free clinical accelerator. Med. Phys. 33, 820–827.

Vassiliev, O.N., Titt, U., Ponisch, F., 2006b. Dosimetric properties of photon beamsfrom a flattening filter free clinical accelerator. Phys. Med. Biol. 51, 1907–1917.

Walters, B.R.B., Rogers, D.W.O., 2009. DOSXYZnrc User Manual. National ResearchCouncil of Canada, Ottawa (NRCC Report No. PIRS-794Rev B).

Wang, Y., Khan, M.K., Ting, J.Y., Easterling, S.B., 2011. Surface dose investigation ofthe flattening filter-free photon beams. Int. J. Radiat. Oncol. Biol. Phys. 83,281–285.

J.C.L. Chow, A.M. Owrangi / Radiation Physics and Chemistry 101 (2014) 46–5252