Temporal characterization and in vitro comparison of cell survival following the delivery of 3D-conformal, intensity-modulated radiation therapy (IMRT) and volumetric modulated arc

7/28/2019 Temporal characterization and in vitro comparison of cell survival following the delivery of 3D-conformal, intensity-modulated radiation therapy (IMRT) and volu

1/14

Temporal characterization and in vitrocomparison of cell survival following the delivery of 3D-

conformal, intensity-modulated radiation therapy (IMRT) and volumetric modulated arc

therapy (VMAT)

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2011 Phys. Med. Biol. 56 2445

(http://iopscience.iop.org/0031-9155/56/8/008)

Download details:

IP Address: 128.42.202.150

The article was downloaded on 31/05/2013 at 16:00

Please note that terms and conditions apply.

View the table of contents for this issue, or go to thejournal homepage for more

ome Search Collections Journals About Contact us My IOPscience
http://iopscience.iop.org/page/termshttp://iopscience.iop.org/0031-9155/56/8http://iopscience.iop.org/0031-9155http://iopscience.iop.org/http://iopscience.iop.org/searchhttp://iopscience.iop.org/collectionshttp://iopscience.iop.org/journalshttp://iopscience.iop.org/page/aboutioppublishinghttp://iopscience.iop.org/contacthttp://iopscience.iop.org/myiopsciencehttp://iopscience.iop.org/myiopsciencehttp://iopscience.iop.org/contacthttp://iopscience.iop.org/page/aboutioppublishinghttp://iopscience.iop.org/journalshttp://iopscience.iop.org/collectionshttp://iopscience.iop.org/searchhttp://iopscience.iop.org/http://iopscience.iop.org/0031-9155http://iopscience.iop.org/0031-9155/56/8http://iopscience.iop.org/page/terms


2/14

IOP PUBLISHING PHYSICS IN MEDICINE AND BIOLOGY

Phys. Med. Biol. 56 (2011) 24452457 doi:10.1088/0031-9155/56/8/008

Temporal characterization and in vitro comparison ofcell survival following the delivery of 3D-conformal,intensity-modulated radiation therapy (IMRT) andvolumetric modulated arc therapy (VMAT)

Conor K McGarry1,2, Karl T Butterworth2, Colman Trainor2,

Joe M OSullivan2,3, Kevin M Prise2 and Alan R Hounsell1,2

1 Radiotherapy Physics, Northern Ireland Cancer Centre, Belfast Health and Social Care Trust,Belfast, UK2 Centre for Cancer Research and Cell Biology, Queens University Belfast, Belfast, UK3 Clinical Oncology, Northern Ireland Cancer Centre, Belfast Health and Social Care Trust,Belfast, UK

E-mail: [email protected]

Received 10 December 2010, in final form 4 February 2011

Published 22 March 2011

Online at stacks.iop.org/PMB/56/2445

Abstract

A phantom was designed and implemented for the delivery of treatment plans

to cells in vitro. Single beam, 3D-conformal radiotherapy (3D-CRT) plans,

inverse planned five-field intensity-modulated radiation therapy (IMRT), nine-

field IMRT, single-arc volumetric modulated arc therapy (VMAT) and dual-arc

VMAT plans were created on a CT scan of the phantom to deliver 3 Gy to the

cell layer and verified using a Farmer chamber, 2D ionization chamber array

and gafchromic film. Each plan was delivered to a 2D ionization chamber

array to assess the temporal characteristics of the plan including delivery time

and cells eye view for the central ionization chamber. The effective fraction

time, defined as the percentage of the fraction time where any dose is delivered

to each point examined, was also assessed across 120 ionization chambers.

Each plan was delivered to human prostate cancer DU-145 cells and normal

primary AGO-1522b fibroblast cells. Uniform beams were delivered to each

cell line with the delivery time varying from 0.5 to 20.54 min. Effective fraction

time was found to increase with a decreasing number of beams or arcs. Fora uniform beam delivery, AGO-1552b cells exhibited a statistically significant

trend towards increased survival with increased delivery time. This trend

was not repeated when the different modulated clinical delivery methods were

used. Less sensitive DU-145 cells did not exhibit a significant trend towards

increased survival with increased delivery time for either the uniform or clinical

deliveries. These results confirm that dose rate effects are most prevalent in

more radiosensitive cells. Cell survival data generated from uniform beam

0031-9155/11/082445+13$33.00 2011 Institute of Physics and Engineering in Medicine Printed in the UK 2445
http://dx.doi.org/10.1088/0031-9155/56/8/008mailto:[email protected]://stacks.iop.org/PMB/56/2445http://stacks.iop.org/PMB/56/2445mailto:[email protected]://dx.doi.org/10.1088/0031-9155/56/8/008


3/14

2446 C K McGarry et al

deliveries over a range of dose rates and delivery times may not always be

accurate in predicting response to more complex delivery techniques, such as

IMRT and VMAT.

(Some figures in this article are in colour only in the electronic version)

1. Introduction

Intensity-modulated radiation therapy (IMRT) can enable highly conformal doses to be

delivered to a tumour volume whilst avoiding or limiting dose to critical structures. The

delivery time of IMRT may be longer than conformal therapy (3D-CRT), depending on

planning parameters such as the number of beams or segments (Kuperman et al 2008).

Volumetric modulated arc therapy (VMAT) is a specific type of IMRT where the gantry rotates

around the patient with the MLCs moving during delivery. During gantry rotation, the speed

and dose rate may or may not be varied (Palma et al 2008). VMAT has shown promise

for the delivery of highly conformal plans whilst reducing overall delivery time compared toconventional IMRT (Verbakel et al 2009a, Bedford 2009).

Clinical outcomes predicted by historical dose volume data of cancer patients treated

with radiotherapy have been typically obtained from conventional radiotherapy or 3D-CRT

that may no longer be applicable for IMRT and VMAT where long term data are not yet

available (Blockhuys et al 2010). Variations in the spatio-temporal dose histories of the

tumour have been demonstrated between different radiotherapy delivery techniques including

3D-CRT, dynamic IMRT, static IMRT and tomotherapy (Shaikh et al 2010). Even within a

single treatment fraction the dose history will vary within a tumour volume (Wang et al 2003,

Schafer etal 2005). A detailed characterization of the temporal delivery properties of advanced

techniques such as VMAT has, to date, not been performed. Randomized clinical trials have

yet to clearly show any adverse effects on tumour control due to IMRT delivery (Veldeman

et al 2008) and therefore pre-clinical models are of importance in predicting outcomes of

advanced radiotherapy techniques.

A number of modelling studies have predicted increased cell survival following radiation

delivery over a prolonged time associated with IMRT delivery (Fowler et al 2004, Wang

et al 2003, Paganetti 2005). Recent reports have provided in vitro evidence that the temporal

pattern of dose delivery may have a significant impact on the radiobiological response of

tumour cells (Sterzing et al 2005, Moiseenko et al 2007, Bewes et al 2008, Yang et al 2009,

Altman et al 2009, Joiner et al 2010). A progressive increase in cell survival for increasing

delivery time has been observed (Bewes et al 2008) with a potential increase in the biological

effectiveness of a given dose by up to 20% for a 1 min Rapidarc R delivery, a commercial

version of VMAT delivery, compared to 17 min IMRT delivery being inferred (Verbakel et al

2009b). The analysis shows that the largest temporal effects do appear for more radiosensitive

cells (SF2 < 0.5) with higher / ratios although dependence of the effects on repair half

times and the relative proportion of different repair processes inhibits detailed interpretationof dose rate effects based on / ratios (Paganetti 2005).

Delivery times for IMRT plans can vary. Improved inverse planning systems such as those

utilizing direct aperture optimization (DAO) have resulted in reduced complexity with reduced

monitor units and/or reduced leaf movement resulting in reduced delivery times (Ludlum and

Xia 2008). Studies investigating temporal differences between conformal delivery and IMRT

delivery either assumed that the delivery was in the order of 2045 min (Joiner et al 2010) or

used IMRT plans with less efficient deliveries (Sterzing et al 2005). The delivery of IMRT


4/14

Comparison of characteristics and cell survival following 3D-CRT, IMRT and VMAT delivery 2447

plans in more recent studies has been between 5 and 10 min (Shaikh et al 2010, Yang et al

2009, Bedford 2009, Moiseenko et al 2007).

In a modelling study, Shaikh et al (2010) hypothesized that any reduced biological effect

would be less for helical tomotherapy (HT) than fixed field IMRT. For HT approximately 90%

of the dose was delivered over a short period of time, termed the rapid dose accumulationtime (RDAT), in the middle of treatment time. With the use of this RDAT as the treatment

time, ignoring the small scatter dose, it would appear that tomotherapy times are equivalent

to VMAT times. However, Yang et al 2009 showed in-vitro that cell survival following

tomotherapy may be closer to IMRT delivery over 7 min than 2 min continuous delivery with

no MLC modulation. Therefore a simple comparison between tomotherapy and VMAT does

not appear to apply. Delivery time has been cited as the major factor in differences in cell

survival (Wang et al 2003, Bewes et al 2008); however, the most up to date delivery methods

are yet to be studied in vitro and compared with uniform delivery over the same time.

The aim of this work was to assess the variation in tumour dose history for clinically

relevant inverse planned IMRT and VMAT deliveries and compare these to a conformal plan.

The effect of varying the number of beams in IMRT (five and nine beams) and the number

of arcs in VMAT (single and dual arc) was also investigated. In this paper prostate cancercell (DU-145) and normal primary AGO-1522b fibroblast cell survival are presented for each

3D-CRT, IMRT and VMAT treatment technique following the delivery of a clinically relevant

dose (3 Gy) to a cell monolayer. To assess if the leaf sequencing has an effect on clinical

delivery compared to uniform exposure, a series of single uniform beams, again for a 3 Gy

dose, were delivered over a longer time range (0.520.54 min) to each cell line.

2. Methods and materials

2.1. Phantom design and construction

A cell phantom was constructed of PMMA using a computer controlled (CNC) milling machine

(Hass, US) with the aim of minimizing air gaps to maintain electronic equilibrium within a

T80 flask (Nunc, UK) filled with culture medium. Figure 1(a) shows that the phantom has

dimensions 30 30 6 cm3 with cells located 1.2 cm from the bottom of the phantom. It is

modular in design to facilitate insertion and removal of culture flasks.

2.2. Experimental setup and treatment planning

Figure 1(b) shows the experimental setup including 10 cm thick 30 30 cm2 and 5 cm thick

30 30cm2 Wte solid water phantom (Barts and The London NHS Trust, London, UK) below

and above the phantom respectively to provide a similar thickness to a prostate patient. A

CT scan was acquired of the setup using a Siemens Emotion 6 CT scanner (Siemens Medical

Systems, Forcheim, Germany). OncentraR v3.3sp1 treatment planning system (Nucletron,

Veenendaal, the Netherlands) was used to outline the planning target volume (PTV), ensuring

a margin of at least 1 cm around the cell layer within the flask. Two pseudo-organs-at-risk(OARs) were also outlined as avoidance structures in the planning process. All plans were

created using 6 MV photon beam data from a Varian 600CD linear accelerator (Varian Medical

Systems, Palo Alto, CA, USA).

Treatment plans were created to deliver 3 Gy per fraction for 20 fractions ensuring that

the 95% isodose encapsulated the PTV and the dose to the pseudo-OARs was minimized.

Conformal plans were created using the class solution employed at our centre. This includes

an anterior beam and two lateral wedged fields. IMRT and VMAT deliveries were inverse


5/14


(a)

(b)

(c)

Figure 1. Photograph of (a) the PMMA phantom, (b) phantom setup for clinical irradiations and(c) phantom setup for uniform irradiations.


6/14


planned using the optimization module within Oncentra R. The local class solution for a

five-field IMRT plan includes beam angles of 35, 100, 180, 260 and 325 with a maximum

number of segments of 50 and a minimum field size of 4 cm. The nine-field IMRT plan

used nine equi-spaced beams 40 apart with a starting angle of 0 (maximum number of

segments = 90 and minimum field size = 4 cm). A single-arc VMAT plan used the defaultgantry spacing of 4 with a start angle of 184 and arc length of 356. For the dual arc, all

properties were the same as the first arc but the second arc started at 180 and rotated in the

opposite direction to the first arc. A constant dose rate and gantry speed was assumed and

the couch accounted for (McGarry et al 2010) during the VMAT planning process. A single

uniform 20 20 cm2 beam was also planned and delivered. All plans were delivered using

a 6 MV photon Varian 600CD linear accelerator (Varian Medical Systems, Palo Alto, CA,

USA).

The aim of this work was to replicate the clinical delivery of each modality as closely as

possible. A local audit of ten clinical five-field IMRT plans and ten clinical conformal plans

showed that they would take approximately 5.0 0.8 min and 2.4 0.3 min respectively

between the first beam on time and the last beam off time on days that the patient was not

imaged. The aim of the VMAT planning was to ensure that delivery time was close to thedelivery of a prostate plan as reported in the literature (Bedford 2009, Boylan et al 2010).

2.3. Temporal effect on cell survival

Figure 1(c) shows the experimental setup with the gantry and couch placed at 90 including

10 cm thick 30 30 cm2 and 5 cm thick 30 30 cm2 Wte solid water behind and in front

of the phantom respectively to provide buildup and backscatter. The total dose was kept at

3 Gy but the average dose rate was varied by changing source-to-surface distances (SSDs)

in the range from 0.75 through 3.42 m whilst ensuring that a 20 20 cm2 field size was

maintained at the surface of the solid water for each SSD studied. The extreme distances

represent the minimum and maximum possible SSD in the treatment room. Similar to Bewes

et al (2008), the instantaneous dose rate varies approximately according to the inverse square

of the distance, while a pulse repetition frequency (PRF) of either 200 or 400 MU min 1 was

utilized. The time to deliver the 3 Gy dose varied from 0.55 to 20.54 min. All plans were

delivered using 6 MV photons on a Varian 600CD linear accelerator (Varian Medical Systems,

Palo Alto, CA, USA).

2.4. Dosimetry

Each of the six clinical plans was re-calculated onto a CT scan dataset of the 30 30

19 cm3 solid water phantom with a Farmer ionization chamber (volume 0.6 cc) located at

the isocentre. The active area of the chamber was outlined and the mean volume from each

plan recorded. This was to avoid possible average volume effects associated with taking a

point measurement using the detector. Each plan was delivered to the Farmer chamber on

a Varian 600CD linear accelerator (Varian Medical Systems, Palo Alto, CA, USA). Farmerchamber readings were converted to average dose and these doses were compared to the doses

calculated.

Each of the six plans was re-calculated onto a CT scan dataset of a 2D ionization chamber

array (IBA MatriXX Evolution) positioned between 6 cm slabs of 30 30 cm2 solid water.

Irradiations were measured in movie mode as a sequence of 0.5 s frames with the frames

summed using Omnipro ImRT software (IBA Dosimetry, Schwarzenbruck, Germany) to

produce a cumulative dose image. A 2D ionization chamber array was used for measurements


7/14


at intervals of 0.5 s as this has previously been shown to be accurate to within 1.5% (Evans

et al 2010). The cumulative image for each delivery was compared with the plane calculated

in terms of the percentage of pixels passing a gamma value of 3%/3 mm.

The six clinical plans were delivered to the phantom with calibrated gafchromic EBT

film (International Specialty Products, NJ, USA) placed at the bottom of the flask. Filmarea measurements (to 0.5 cm inside the perimeter of the flask) were compared with those

calculated using Oncentra R. The gafchromic film was placed at the bottom of the flasks at

each SSD to ensure that, although the delivery time changed, the dose delivered was equivalent

to 3 Gy. Measurements showed that doses were accurate to within 0.8%.

2.5. Temporal delivery characteristics of clinical plans

The cells-eye-view (Goitein 2005) was used to describe the approximate cumulative dose

received by the tumour cells over time. The cumulative central axis dose measurements for

each plan were recorded using the 2D array data by integrating the dose over time. Effective

fraction time is defined as the percentage of the treatment fraction time where any dose is

delivered to the point examined (Schafer et al 2005). This was calculated for 120 ionization

chambers which approximately coincided with the cells inside the flask.

2.6. Cell culture and clonogenic assays

Experiments were conducted using two cell lines, the human prostate cancer cell line, DU-145,

andthe human fibroblastcell line, AGO-1522b. Cell lines were obtained from Cancer Research

UK and selected as malignant and transformed models with different radiosensitivity. DU-

145 cells were grown in RPMI-1640 with L- glutamine (Lonza, UK) supplemented with 10%

fetal bovine serum, 1% penicillin/streptomycin (Gibco, UK). AGO-1522b cells were grown

in Eagles minimum essential medium with deoxyribonucleosides and deoxyribonucleotides

(Lonza, UK) supplemented with 20% fetal bovine serum and 1% penicillin/streptomycin. All

cell lines were maintained at 37 C in a humidified atmosphere of 5% CO2.

Cell survival was determined by clonogenic assay as previously reported (Butterworth

et al 2010). Cells were plated and allowed to adhere overnight. Culture flasks were filled withserum-free medium and sealed immediately prior to irradiation. Cells were irradiated at room

temperature (25 2 C). Following irradiation, serum-free medium was removed and replaced

with complete culture medium. Cultures were incubated for 1014 days before staining with

0.5% crystal violet in 50% methanol. DU-145 colonies were scored using a Colcount (Oxford

Optronix, UK) automated counter which optimized for the cell line. AGO-1522b colonies

were scored manually applying a 50 cell exclusion rule. For each experiment unexposed

controls were prepared and treated as sham exposures.

All exposures were performed in duplicate or triplicate and each clinical treatment plan

group had 1219 samples with each uniform point derived from at least nine replicates.

On each occasion unexposed controls were prepared and treated as sham exposures. For

presentation purposes, and to be consistent with Moiseenko et al (2007), cell survival was

normalized to the standard conformal treatment (delivered over 2.45 min). For uniform beams,the cell survival was normalized to cells irradiated over a time close to conformal treatment

(2.79 min).

2.7. Statistical analysis

Statistical significance was assessed using regression analysis based on linear, quadratic,

exponential and logarithmic models with the model accounting for most statistical variance

used for statistical analysis. Data uncertainties were calculated as the standard error of the


8/14


Table 1. Properties of each delivery modality including technique, dose rate, monitor units (MU)and delivery time. Included are local audit measurements of delivery times for ten conformal andten IMRT (five-field) patients and VMAT single-arc deliverytime reported (aBedford 2009, Boylanet al 2010) for prostate patients. Absorbed dose characteristics for each plan measured using 120detectors of an ionization chamber array in the form of effective fraction time.

Nominal Delivery Audit/ Effective

dose rate time literature fraction

Plan Technique (MU min1) MU (mm:ss) (mm:ss) time (%)

Conformal/3D-CRT Static 400 468 2:27 2:48 00:15 49.3 0.0

VMAT single arc Dynamic arc 300 776 2:35 2:30a 100.0 0.0

VMAT dual arc Dynamic arc 200/200 697 3:42 95.4 0.0

IMRT 5 field Step and shoot 400 612 4:30 5:00 00:50 42.2 0.2

IMRT 9 field Step and shoot 400 703 8:17 30.3 0.2

Single field Static 400 342 0:51 100.0 0.0

mean (SEM). Calculations were performed using Statistical Package for Social Sciences

version 15.0.1.1 (SPSS, Chicago, IL, USA).

3. Results

3.1. Treatment planning and dosimetry

Table 1 shows the characteristics of the six plans created. Figure 2 shows the dose distributions

of the conformal, IMRT and VMAT plans. It is clearly demonstrated that the 95% isodose

line fully encapsulated the PTV whilst avoiding the pseudo-OARs. Ionization chamber results

were within 1.8% of the calculated values for all plans. All plans showed excellent agreement

following comparison of delivery to the ionization chamber array to the calculated plan using

gamma criteria of 3%/3 mm (>98.8% pixels passing). Measurements from the calibrated

gafchromic EBT film placed at the bottom of the flasks compared to the same dose areacalculated confirmed the accuracy of each delivery to cells in the phantom to within 1.7%.

3.2. Temporal delivery characteristics of clinical plans

The cells-eye-view dose accumulation at the central axis of all delivery techniques is shown in

figure 3. Conformal and single-arc VMAT plans took approximately the same time to deliver

and were the fastest of all treatment modalities after single-field delivery. The dual-arc VMAT

was the next fastest to deliver with the five- and nine-field IMRT plans delivered over 4 min

and 8 min respectively. These delivery times are also shown in table 1 and coincide with the

current clinical delivery or literature values (Bedford 2009, Boylan et al 2010). Table 1 also

shows that the effective fraction time is related to the number of beams or arcs with values

of 30% for a nine-field plan up to 100% for a single-arc and single-beam delivery. This is

consistent with the fact that the gantry will have to move between exposures and the beammust be loaded for the next exposure which takes time, time when the tumour is not being

irradiated.

3.3. Cell survival

Figure 4(a) shows that AGO-1522b cell survival increases as the time to deliver a uniform

radiation beam increases. Table 2 shows model parameters when the delivery time was


9/14


(a) (b)

(c) (d)

(e) (f)

Figure 2. Isodose distributions with target (blue) and pseudo-organs-at-risk (red) produced usingOncentra R. (a) Coronal slice through the CT scan with calculated isodoses for a 3D-conformalplan. Transverse slices with isodoses also shown for (b) a conformal plan, (c) a five-field IMRTplan, (d) a nine-field IMRT plan, (e) a VMAT single-arc plan and (f) a VMAT dual-arc plan.

logarithmically regressed on cell survival. It was found that the dose delivery time was

a statistically significant predictor of clonogenic cell survival for AGO-1522b cells for a

uniform field (F= 10.73, n = 70, p < 0.01). However, the dose delivery time was not found

to be a statistically significant predictor of clonogenic cell survival for AGO-1522b cells

when all modalities, including modulated deliveries, were analysed using linear regression

(r= 0.020 0.156, t(1, 62) = 0.157, p = 0.876).Figure 4(b) shows that DU-145 cell survival does not change significantly as the time to

deliver a uniform radiation beam increases. It was found that the dose delivery time was not

a statistically significant predictor of clonogenic cell survival for DU-145 cells ( F= 0.325,

n = 75, p = 0.570). As with AGO-1522b cells, the dose delivery time was not found to

be a statistically significant predictor of clonogenic cell survival for DU-145 cells when all

modalities were analysed using linear regression (r= 0.040 0.317, t(1, 107) = 0.410, p =

0.683).


10/14


Figure 3. Cells-eye-view at the central axis for all six plans.

Table 2. Parameters derived from a logarithmic regression analysis showing the relationshipbetween delivery time and cell survival for a uniform beam. Survival fraction at 3 Gy at 2.79 minshown for each cell line is included to indicate radiosensitivity.

Cell / SF3 Model summaryParameterestimates

line range normalization R SE F Significant Constant a

DU-145 2.332.67a 0.484 (0.034) 0.066 (0.190) 0.325 p = 0.57 1.027 0.011

AGO-1522b 10.7526.5a 0.244 (0.013) 0.365 (0.192) 10.73 p < 0.01 0.969 0.064

a For reference, previously derived/ values are included (Butterworth et al 2010).

4. Discussion

In this study a phantom (figure 1) was designed and implemented to compare cell survival

following the delivery of different radiation therapy techniques. Single-beam, conformal,

five-field IMRT, nine-field IMRT, single-arc VMAT and dual-arc VMAT plans were created

and verified following a 3 Gy delivery using each modality to the cell layer within the flask

in the phantom. Variation in the absorbed dose rate at a cells-eye-level was observed using

a 2D ionization chamber array across all five plans. Dose delivery time was found to be

a statistically significant predictor of clonogenic cell survival for AGO-1522b cells but not

DU-145 cells for uniform deliveries over a time range 0.520 min. No dose rate effects were

observed for either cell line when survival was analysed against delivery time using the clinical

deliveries.

All treatment plans (figure 2) ensured that the 95% isodose line encapsulated the PTV.

Delivery times were in accordance with what was expected locally and from the literature

(Bedford 2009, Boylan et al 2010). Delivery patterns at the central axis for IMRT and 3D-CRT plans were consistent with previous studies (Schafer et al 2005, Kuperman et al 2008,

Moiseenko et al 2007, Shaikh et al 2010). This type of analysis had not yet been performed

on inverse planned VMAT plans prior to this study. In addition to measurements of variation

in delivery time, effective fraction time was also calculated across 120 detectors for 3D-CRT,

IMRT and VMAT plans. Effective fraction times ranging from 11.6 to 37.3%, depending on

the number of IMRT beams, have been previously reported by Schafer et al (2005). Similarly,

we have shown the effective fraction time to be inversely proportional to the number of beams


11/14


(a)

(b)

Figure 4. Clonogenic cell survival fractions at 3 Gy (SF3) for (a) AGO-1522b cell line and(b) DU-145 cell line comparing dose delivery from a single uniform beam and five treatmentmodalities. Error bars show the standard error of the mean. Linear trend lines are shown to guidethe eye (dashed line). The normalization condition corresponds to 3D-CRT delivery. Uniformdelivery of 3 Gy over 0.520.54 min range is also plotted for each cell line. Logarithmic trendlines are shown to guide the eye (solid line). The normalization condition corresponds to 2.79 mindelivery.

or arcs as shown in table 1. The nine-field IMRT plan (effective fraction time = 30.3 0.2%)

was least efficient with 100 0.0% efficiency being observed for VMAT single-arc plans.

Less variation in the dose rates across detectors was observed for the IMRT plans compared to

Schafer et al (2005). This may be attributed to the use of a 2D array and not three-dimensional


12/14


dose points although other factors such as the use of a more modern inverse planning system

may also have contributed.

Each of the six clinical plans created were delivered to give a uniform dose of 3 Gy to

prostate tumour cells (DU-145) and normal primary AGO-1522b fibroblast cells grown as a

monolayer at the bottom surface of the T80 flask within the phantom. The framework usedto study the outcome following cell survival was proposed by Moiseenko et al (2007) which

revealed increased cell survival following longer IMRT delivery compared to acute delivery for

more radiosensitive cells. Within this framework it is important that new delivery techniques

are compared to typical clinical techniques. Using a similar method, Yang etal (2009) showed

significantly increased cell kill with a 2 min arc therapy delivery compared to 7 min IMRT or

HT delivery. Rapid MLC modulation had not been simulated in either arc or IMRT delivery.

The data presented in this paper have focussed on cell survival following the delivery of

clinically relevant inverse planned IMRT and VMAT plans compared to a conventional 3D

conformal plan. Dose delivery time was not found to be a statistically significant predictor of

clonogenic cell survival for AGO-1522b or DU-145 cells when all modalities were analysed.

This is consistent with our previous work comparing survival responses to modulated and

non-modulated delivery (Butterworth et al 2010). When MLC modulation was not includedin any plans and a wider time range used, dose delivery time was a statistically significant

predictor of clonogenic cell survival for AGO-1522b cells but not DU-145 cells. As with

reports from other authors (Moiseenko et al 2007, Bewes et al 2008, Yang et al 2009, Joiner

et al 2010) increased survival over time was largest for the more radiosensitive cells. This

effect appears to reduce when absorbed dose properties vary between delivery techniques. The

properties of plans such as effective fraction time vary widely between modalities. As the time

range used for the dose-rate experiments extended well beyond the longest time for the clinical

irradiations investigated here, the dose-rate data were re-fitted excluding time-points beyond

10.24 min to ensure that these data points were not affecting the comparison with the clinical

exposures. Even after excluding these data points there was still a statistically significant

correlation between treatment time and survival fraction following logarithmic regression

(p = 0.04).

Clinical trials have shown no clinical indications that IMRT has led to adverse effects onlocoregional control or survival (Veldeman et al 2008). This is consistent with our findings,

although the small differences observed between different techniques may require a more

sensitive method of detecting outcomes. Further work is required to extrapolate to the in vivo

scenario where there will be a complex relationship between tumour cells and surrounding

normal tissue.

5. Conclusion

A phantom to study the delivery of radiotherapy plans in vitro was designed and validated.

Single-beam, conformal, IMRT and VMAT plans were created for this phantom and theabsorbed dose rates characterized using a 2D ionization chamber array. The delivery time

and effective fraction time were found to vary widely between modalities. For uniform

irradiations, a statistically significant trend towards increased survival with increased delivery

time was observed for AGO-1522b cells, but not for more radioresistant DU-145 cells. No

trend was observed in either cell line when survival was analysed against delivery time using

the modulated clinical plans with widely differing absorbed dose rate histories. Differences in

absorbed dose rate histories of techniques such as 3D-CRT, IMRT and VMAT may not allow


13/14


direct interpolation of cell survival using data generated from uniform beam delivery over a

range of treatment times.

Acknowledgments

We are indebted to Cyril Mitchell for his invaluable expertise in manufacturing the phantom.

We also thank Dr Christina Agnew for proof reading the manuscript. CKM is supported by a

Health & Social Care Research & Development Office of the Public Health Agency Training

Fellowship Award. We wish to acknowledge financial support from Cancer Research UK

(grant number C1513/A7047 to KM Prise).

References

Altman M B, Stinauer M A, Javier D, Smith B D, Herman L C, Pytynia M L, Aydogan B, Pelizzari C A, Chmura S J

and Roeske J C 2009 Validation of temporal optimization effects for a single fraction of radiation in vitro Int. J.

Radiat. Oncol. Biol. Phys. 75 12406

Bedford J L 2009 Treatment planning for volumetric modulated arc therapy Med. Phys. 36 512838

BewesJ M, SuchowerskaN, Jackson M, Zhang M andMcKenzie D R 2008 Theradiobiological effectof intra-fraction

dose-rate modulation in intensity modulated radiation therapy (IMRT) Phys. Med. Biol. 53 356778

Blockhuys S, Vanhoecke B, De Wagter C, Bracke M and De Neve W 2010 From clinical observations of intensity-

modulated radiotherapy to dedicated in vitro designs Mutat. Res. 704 2005

Boylan C J, Golby C and Rowbottom C G 2010 A VMAT planning solution for prostate patients using a commercial

treatment planning system Phys. Med. Biol. 55 N395404

Butterworth K T, McGarry C K, OSullivan J M, Hounsell A R and Prise K M 2010 A study of the biological effects

of modulated 6 MV radiation fields Phys. Med. Biol. 55 160718

Evans P M, Symonds-Tayler J R, Colgan R, Hugo G D, Letts N and Sandin C 2010 Gating characteristics of an Elekta

radiotherapy treatment unit measured with three types of detector Phys. Med. Biol. 55 N20110

Fowler J F, Welsh J S and Howard S P 2004 Loss of biological effect in prolonged fraction delivery Int. J. Radiat.

Oncol. Biol. Phys. 59 2429

Goitein M 2005 The cells-eye view: assessing dose in four dimensions Int. J. Radiat. Oncol. Biol. Phys. 62 9513

Joiner M C, Mogili N, Marples B and Burmeister J 2010 Significant dose can be lost by extended delivery times in

IMRT with x rays but not high-LET radiations Med. Phys. 37 245765

KupermanV Y, Ventura A M andSommerfeldt M 2008 Effect of radiationprotraction in intensity-modulated radiation

therapy with direct aperture optimization: a phantom study Phys. Med. Biol. 53 327992

Ludlum E and Xia P 2008 Comparison of IMRT planning with two-step and one-step optimization: a way to simplify

IMRT Phys. Med. Biol. 53 80721

McGarry C K, OSullivan J M and Hounsell A R 2010 Impact of modeling the treatment couch for VMAT treatments

Med. Phys. 37 3239

Moiseenko V, Duzenli C and Durand R E 2007 In vitro study of cell survival following dynamic MLC intensity-

modulated radiation therapy dose delivery Med. Phys. 34 151420

Paganetti H 2005 Changes in tumor cell response due to prolonged dose delivery times in fractionated radiation

therapy Int. J. Radiat. Oncol. Biol. Phys. 63 892900

PalmaD, VollansE, James K, NakanoS, Moiseenko V, Shaffer R, McKenzie M, Morris J andOtto K 2008 Volumetric

modulated arc therapy for delivery of prostate radiotherapy: comparison with intensity-modulated radiotherapy

and three-dimensional conformal radiotherapy Int. J. Radiat. Oncol. Biol. Phys. 72 9961001

Schafer M, Munter M, Sterzing F, Haring P, Rhein B and Debus J 2005 Measurements of characteristics of time

pattern in dose delivery in step-and-shoot IMRT Strahlenther. Onkol. 181 58794

Shaikh M, Burmeister J, Joiner M, Pandya S, Zhao B and Liu Q 2010 Biological effect of different IMRT delivery

techniques: SMLC, DMLC, and helical tomotherapy Med. Phys. 37 76270

SterzingF, Munter M W,Schafer M, Haering P, Rhein B, Thilmann C andDebus J 2005 Radiobiologicalinvestigation

of dose-rate effects in intensity-modulated radiation therapy Strahlenther. Onkol. 181 428

Veldeman L, Madani I, Hulstaert F, De Meerleer G, Mareel M and De Neve W 2008 Evidence behind use of

intensity-modulated radiotherapy: a systematic review of comparative clinical studies Lancet Oncol. 9 36775

Review
http://dx.doi.org/10.1016/j.ijrobp.2009.06.076http://dx.doi.org/10.1016/j.ijrobp.2009.06.076http://dx.doi.org/10.1016/j.ijrobp.2009.06.076http://dx.doi.org/10.1118/1.3240488http://dx.doi.org/10.1118/1.3240488http://dx.doi.org/10.1118/1.3240488http://dx.doi.org/10.1088/0031-9155/53/13/012http://dx.doi.org/10.1088/0031-9155/53/13/012http://dx.doi.org/10.1088/0031-9155/53/13/012http://dx.doi.org/10.1016/j.mrrev.2010.02.003http://dx.doi.org/10.1016/j.mrrev.2010.02.003http://dx.doi.org/10.1016/j.mrrev.2010.02.003http://dx.doi.org/10.1088/0031-9155/55/14/N01http://dx.doi.org/10.1088/0031-9155/55/14/N01http://dx.doi.org/10.1088/0031-9155/55/14/N01http://dx.doi.org/10.1088/0031-9155/55/6/005http://dx.doi.org/10.1088/0031-9155/55/6/005http://dx.doi.org/10.1088/0031-9155/55/6/005http://dx.doi.org/10.1088/0031-9155/55/8/N02http://dx.doi.org/10.1088/0031-9155/55/8/N02http://dx.doi.org/10.1088/0031-9155/55/8/N02http://dx.doi.org/10.1016/j.ijrobp.2004.01.004http://dx.doi.org/10.1016/j.ijrobp.2004.01.004http://dx.doi.org/10.1016/j.ijrobp.2004.01.004http://dx.doi.org/10.1016/j.ijrobp.2005.03.054http://dx.doi.org/10.1016/j.ijrobp.2005.03.054http://dx.doi.org/10.1016/j.ijrobp.2005.03.054http://dx.doi.org/10.1118/1.3425792http://dx.doi.org/10.1118/1.3425792http://dx.doi.org/10.1118/1.3425792http://dx.doi.org/10.1088/0031-9155/53/12/014http://dx.doi.org/10.1088/0031-9155/53/12/014http://dx.doi.org/10.1088/0031-9155/53/3/018http://dx.doi.org/10.1088/0031-9155/53/3/018http://dx.doi.org/10.1088/0031-9155/53/3/018http://dx.doi.org/10.1118/1.3468628http://dx.doi.org/10.1118/1.3468628http://dx.doi.org/10.1118/1.3468628http://dx.doi.org/10.1118/1.2712044http://dx.doi.org/10.1118/1.2712044http://dx.doi.org/10.1118/1.2712044http://dx.doi.org/10.1016/j.ijrobp.2005.07.953http://dx.doi.org/10.1016/j.ijrobp.2005.07.953http://dx.doi.org/10.1016/j.ijrobp.2005.07.953http://dx.doi.org/10.1016/j.ijrobp.2008.02.047http://dx.doi.org/10.1016/j.ijrobp.2008.02.047http://dx.doi.org/10.1016/j.ijrobp.2008.02.047http://dx.doi.org/10.1007/s00066-005-1289-7http://dx.doi.org/10.1007/s00066-005-1289-7http://dx.doi.org/10.1007/s00066-005-1289-7http://dx.doi.org/10.1118/1.3284369http://dx.doi.org/10.1118/1.3284369http://dx.doi.org/10.1118/1.3284369http://dx.doi.org/10.1007/s00066-005-1290-1http://dx.doi.org/10.1007/s00066-005-1290-1http://dx.doi.org/10.1007/s00066-005-1290-1http://dx.doi.org/10.1016/S1470-2045(08)70098-6http://dx.doi.org/10.1016/S1470-2045(08)70098-6http://dx.doi.org/10.1016/S1470-2045(08)70098-6http://dx.doi.org/10.1016/S1470-2045(08)70098-6http://dx.doi.org/10.1007/s00066-005-1290-1http://dx.doi.org/10.1118/1.3284369http://dx.doi.org/10.1007/s00066-005-1289-7http://dx.doi.org/10.1016/j.ijrobp.2008.02.047http://dx.doi.org/10.1016/j.ijrobp.2005.07.953http://dx.doi.org/10.1118/1.2712044http://dx.doi.org/10.1118/1.3468628http://dx.doi.org/10.1088/0031-9155/53/3/018http://dx.doi.org/10.1088/0031-9155/53/12/014http://dx.doi.org/10.1118/1.3425792http://dx.doi.org/10.1016/j.ijrobp.2005.03.054http://dx.doi.org/10.1016/j.ijrobp.2004.01.004http://dx.doi.org/10.1088/0031-9155/55/8/N02http://dx.doi.org/10.1088/0031-9155/55/6/005http://dx.doi.org/10.1088/0031-9155/55/14/N01http://dx.doi.org/10.1016/j.mrrev.2010.02.003http://dx.doi.org/10.1088/0031-9155/53/13/012http://dx.doi.org/10.1118/1.3240488http://dx.doi.org/10.1016/j.ijrobp.2009.06.076


14/14


Verbakel W F, Cuijpers J P, Hoffmans D, Bieker M, Slotman B J and Senan S 2009a Volumetric intensity-modulated

arc therapy versus conventional IMRT in head-and-neck cancer: a comparative planning and dosimetric study

Int. J. Radiat. Oncol. Biol. Phys. 74 2529

Verbakel W F, Senan S, Lagerwaard F J, Cuijpers J P and Slotman B J 2009b Comments on Single-Arc IMRT?

Phys. Med. Biol. 54 L314 author reply L356

Wang J Z, Li X A, DSouza W D and Stewart R D 2003 Impact of prolonged fraction delivery times on tumor control:

a note of caution for intensity-modulated radiation therapy (IMRT) Int. J. Radiat. Oncol. Biol. Phys. 57 54352

Yang W, Wang L, Larner J, Read P, Benedict S and Sheng K 2009 Tumor cell survival dependence on helical

tomotherapy, continuous arc and segmented dose delivery Phys. Med. Biol. 54 663543
http://dx.doi.org/10.1016/j.ijrobp.2008.12.033http://dx.doi.org/10.1016/j.ijrobp.2008.12.033http://dx.doi.org/10.1016/j.ijrobp.2008.12.033http://dx.doi.org/10.1088/0031-9155/54/8/L01http://dx.doi.org/10.1088/0031-9155/54/8/L01http://dx.doi.org/10.1088/0031-9155/54/8/L01http://dx.doi.org/10.1016/S0360-3016(03)00747-8http://dx.doi.org/10.1016/S0360-3016(03)00747-8http://dx.doi.org/10.1016/S0360-3016(03)00747-8http://dx.doi.org/10.1088/0031-9155/54/21/013http://dx.doi.org/10.1088/0031-9155/54/21/013http://dx.doi.org/10.1088/0031-9155/54/21/013http://dx.doi.org/10.1088/0031-9155/54/21/013http://dx.doi.org/10.1016/S0360-3016(03)00747-8http://dx.doi.org/10.1088/0031-9155/54/8/L01http://dx.doi.org/10.1016/j.ijrobp.2008.12.033

Documents

Temporal characterization and in vitro comparison of cell survival following the delivery of 3D-conformal, intensity-modulated radiation therapy (IMRT) and volumetric modulated arc