Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy:

estimation of endothelial dose enhancement

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

2010 Phys. Med. Biol. 55 6533

(http://iopscience.iop.org/0031-9155/55/21/013)

Download details:

IP Address: 131.170.6.51

The article was downloaded on 21/03/2013 at 09:22

Please note that terms and conditions apply.

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

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING PHYSICS IN MEDICINE AND BIOLOGY

Phys. Med. Biol. 55 (2010) 6533–6548 doi:10.1088/0031-9155/55/21/013

Applying gold nanoparticles as tumor-vasculardisrupting agents during brachytherapy: estimationof endothelial dose enhancement

Wilfred Ngwa, G Mike Makrigiorgos and Ross I Berbeco

Department of Radiation Oncology, Division of Medical Physics and Biophysics, Brigham andWomen’s Hospital, Dana-Farber Cancer Institute and Harvard Medical School, Boston,MA 02115, USA

E-mail: mmakrigiorgos@lroc.harvard.edu

Received 11 August 2010, in final form 22 September 2010Published 19 October 2010Online at stacks.iop.org/PMB/55/6533

AbstractTumor vascular disrupting agents (VDAs) represent a promising approachto the treatment of cancer, in view of the tumor vasculature’s pivotal rolein tumor survival, growth and metastasis. VDAs targeting the tumor’sdysmorphic endothelial cells can cause selective and rapid occlusion ofthe tumor vasculature, leading to tumor cell death from ischemia andextensive hemorrhagic necrosis. In this study, the potential for applyinggold nanoparticles (AuNPs) as VDAs, during brachytherapy, is examined.Analytic calculations based on the electron energy loss formula of Cole werecarried out to estimate the endothelial dose enhancement caused by radiation-induced photo/Auger electrons originating from AuNPs targeting the tumorendothelium. The endothelial dose enhancement factor (EDEF), representingthe ratio of the dose to the endothelium with and without gold nanoparticleswas calculated for different AuNP local concentrations, and endothelial cellthicknesses. Four brachytherapy sources were investigated, I-125, Pd-103,Yb-169, as well as 50 kVp x-rays. The results reveal that, even at relativelylow intra-vascular AuNP concentrations, ablative dose enhancement to tumorendothelial cells due to photo/Auger electrons from the AuNPs can be achieved.Pd-103 registered the highest EDEF values of 7.4–271.5 for local AuNPconcentrations ranging from 7 to 350 mg g−1, respectively. Over the sameconcentration range, I-125, 50 kVp and Yb-169 yielded values of 6.4–219.9,6.3–214.5 and 4.0–99.7, respectively. Calculations of the EDEF as a function ofendothelial cell thickness showed that lower energy sources like Pd-103 reachthe maximum EDEF at smaller thicknesses. The results also reveal that thehighest contribution to the EDEF comes from Auger electrons, apparently dueto their shorter range. Overall, the data suggest that ablative dose enhancementto tumor endothelial cells can be achieved by applying tumor vasculature-targeted AuNPs as adjuvants to brachytherapy, with lower energy sources. Such

0031-9155/10/216533+16$30.00 © 2010 Institute of Physics and Engineering in Medicine Printed in the UK 6533

6534 W Ngwa et al

ablative magnitude dose enhancement in a relatively small endothelial volumemay rapidly disrupt or cause severe biological damage to tumor endothelialcells, without increased toxicity to healthy tissues not containing AuNPs. Thefindings provide significant impetus for considering the application of AuNPsas VDAs during brachytherapy.

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

1. Introduction

The application of gold nanoparticles (AuNPs) to enhance radiation therapy is blossominginto a promising frontier in cancer research (Hainfeld et al 2008, 2010, Cho et al 2009, Roeskeet al 2007, Zhang et al 2009, Jones et al 2010, Herold et al 2000). Following the pioneeringwork by Herold et al (2000) with gold microspheres, Hainfeld et al performed pre-clinicalexperiments in mouse models to establish proof-of-principle of the therapeutic benefit ofintravenous injection of gold nanoparticles targeting tumors during radiotherapy (Hainfeldet al 2004). More recent experimental work further shows that gold nanoparticles enhancethe radiotherapy of a radioresistant mouse squamous cell carcinoma (Hainfeld et al 2010).Beside the experimental work, theoretical work by Roeske et al (2007) showed that low energyx-rays and brachytherapy sources provide the highest degree of dose enhancement when usedwith high Z materials like gold. Furthermore, Cho et al introduced the term ‘GNRT’ (goldnanoparticle-aided radiation therapy) while performing Monte Carlo studies of external beamand brachytherapy sources (Cho et al 2009, Cho 2005). They concluded that GNRT may besuccessfully implemented via brachytherapy with low energy gamma-/x-ray sources. Morerecent theoretical work (Jones et al 2010) has further predicted a remarkable microscopic doseenhancement around AuNPs employed with low energy photon sources.

Meanwhile, tumor vascular targeting has recently garnered considerable attention, inrecognition of the tumor vasculature’s pivotal role in tumor survival, growth and spread. Tothis end, vascular disrupting agents (VDAs) targeting tumor endothelial cells are designed toselectively occlude or disrupt established tumor vasculature. Such selective shutdown of thevasculature leads to tumor cell death from ischemia and widespread central necrosis (Gridelliet al 2009, Denekamp 1993, Cooney et al 2006).

Motivated by the concomitantly growing interests and research in the development andapplication of gold nanoparticles and VDAs in cancer treatment, this study investigates thedosimetric potential of applying gold nanoparticles as VDAs during brachytherapy. Analyticcalculations based on the empirically determined electron energy loss formulae of Cole (1969)are carried out to quantify the local dose enhancement due to photo/Auger electrons fromAuNPs targeting the tumor endothelium during brachytherapy. In the previous theoreticalworks cited above (Cho et al 2009, Roeske et al 2007, Zhang et al 2009, Jones et al 2010),the common assumption has been that the AuNPs will be distributed throughout the tumor,with all parts of the tumor constituting a target of equal importance. However, in consideringpotential clinical application of AuNPs as VDAs, the tumor endothelial cells are of particularimportance. Hence, instead of calculating the dose enhancement to the entire tumor volume,only the local dose enhancement to the endothelial cells is calculated—taking into account thespecific dimensions of endothelial cells and restricted distribution of the AuNPs. Assumingthat the AuNPs are targeted and restricted to the tumor endothelium, and since the photo/Augerelectrons from the AuNPs are inherently short range, the dose to the endothelial cells due

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6535

to these electrons should be greater than that to the rest of the tumor. To provide moreperspective, the endothelial dose enhancement is investigated as a function of realistic localAuNP concentrations, and different endothelial cell thicknesses. The thickness dependencestudy is motivated by the fact that various endothelial cell thicknesses have been reported inthe literature (Hashizume et al 2000, Warrell et al 2005, Stoletov et al 2007, Shen et al 2009).

The approach for applying AuNPs as VDAs, investigated in this work, is particularlymotivated by the same concept that makes vascular targeting appealing. This underlyingconcept is that, in contrast to targeting all tumor cells, the killing of relatively few vascularendothelial cells could result in the death of a large area of tumor, and the suppression ofmetastasis (Denekamp 1982, 1984). In fact, the study by Garcia-Barros et al (2003) indicatesthat damage of tumor endothelial cells is a prime determinant of tumor cell response toradiation at the clinically relevant dose range. Pre-clinical studies by Boerman et al (1992)with radionuclides, as well as recent computational modeling (Zhu et al 2010), also suggest thata more restricted or concentrated distribution of radio-pharmaceuticals around blood vessels ismore effective than distributing them homogenously within the tumor. Furthermore, targetingthe tumor endothelial cells with AuNPs may be less challenging than targeting large solidtumors since the endothelial cells are more accessible and genetically stable than tumor cellsthemselves. Application-wise, AuNPs may also present unique advantages in potential VDAapplications due to the fact that gold is non-toxic (Mukherjee et al 2005) and biocompatible(Shukla et al 2005). In view of all these, the results of this study will provide useful insightsinto potential application of AuNPs as VDAs during brachytherapy.

2. Materials and methods

2.1. Endothelial cell model

In this study, a tumor vascular endothelial cell (EC) is modeled as a slab, similar to previousstudies (Desgrosellier and Cheresh 2010; Warrell et al 2005), only here with relatively moreconservative default dimensions of 2 μm (thickness) × 10 μm (length) × 10 μm (width)(figure 1(a)). The EC is surrounded on the four short sides by other ECs. In the calculation,a spherical nanoparticle (AuNP) is simulated attached to the lumen-side surface of theendothelial cell. Here, the exact position of the AuNP is not of particular importance as severalnanoparticles, assumed to be evenly spread along the lumen wall of the vasculature, can beattached on a single endothelial cell. Furthermore, photo/Auger electrons from nanoparticlesattached near the edge of an endothelial cell may deposit energy to the adjacent cell, and viceversa, thereby providing lateral electron equilibrium. Such lateral compensation of the emittedenergy allows energy emissions from AuNPs in the periphery of endothelial cells to be treatedsimilar to centrally located AuNPs. A nanoparticle size of 400 nm diameter was chosen tokeep AuNPs small relative to the endothelial cells but large enough to be restricted within thevasculature (Unezaki et al 1996, Maeda et al 2000). The dose enhancement from a singlenanoparticle to the closest endothelial cell is calculated, and the results are then extended toother concentrations of nanoparticles.

2.2. Gold nanoparticle concentration

The AuNPs concentrations that are often cited in the literature are 18 and 7 mg g−1 (Hainfeldet al 2004, 2008, Cho 2005). The value of 18 mg g−1 was cited as the content of AuNPs in blood2 min after injection of a mouse with 2.7 g Au Kg−1 body weight. In previous theoreticalcalculations, the 7 mg g−1 is usually assumed as the average concentration throughout the

6536 W Ngwa et al

(a)

(b)

Figure 1. (a) Simplified slab model (2 μm (thickness) × 10 μm (length) × 10 μm (width))of the endothelial cell layer between the lumen and the tumor cells. The gold nanoparticles areattached to the lumen side of the endothelial cells. The range of photoelectrons is shown as a‘sphere of interaction’ with the nanoparticle at the center. Only the dose deposited within theadjacent endothelial cell (shaded region) is used to calculate the dose enhancement. (b) Transverseview of an idealized micro-vascular segment with gold nanoparticles distributed along the interiorendothelial cell wall.

entire tumor, while in the current work this is restricted to the vasculature. Assuming that thevasculature constitutes 5% of the tumor volume (Ryschich et al 2004), and the AuNPs arerestricted to the vasculature, the average concentration in the vasculature will be 20-fold theoverall tumor concentration (i.e. 140 mg g−1 for a total tumor concentration of 7 mg g−1).

Considering a micro-segment of the tumor vasculature with a diameter of 20 μm anda thickness of about one endothelial cell length (10 μm) (see figure 1(b)), the mass of thevasculature micro-segment is about 15.7 times the mass of one endothelial cell (EC). But thereare roughly 6.3 ECs per micro-segment, given the dimensions of the EC and micro-segment.This implies that for a given vascular concentration, the average local concentration targetingthe endothelial cell is about 2.5-fold the overall vasculature concentration. Hence, altogether,a 7 mg g−1 overall tumor concentration is tantamount to up to 350 mg g−1 (i.e. 140 mg g−1 ×2.5) average local concentration at the targeted endothelial cell. However, limiting the AuNPsstrictly near the endothelial cells in vivo (even with active targeting) may not be achievable.The present calculations are, therefore, performed for nanoparticle concentrations in the rangeof 7–350 mg g−1 to account for less-than-perfect targeting.

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6537

The number of AuNPs within the segment is related to the overall nanoparticleconcentration:

NAuNP = mH2O[nanoparticle concentration (mg g−1)]

mAuNP= 4.9 × 103[concentration]. (1)

Here, mH2O represents the mass of water and mAuNP the mass of gold, and NAuNP isthe number of AuNPs. Assuming that all AuNPs are surface-bound, the maximum allowablenumber of AuNPs given the surface area and the size of the AuNPs in this example is 5000. Thiscorresponds to a maximum theoretical AuNP concentration of about 1064 mg g−1, with theAuNPs distributed ‘shoulder to shoulder’ for the specified geometry within the endothelium.

2.3. Estimation of dose enhancement from photoelectrons

In the presence of charged particle equilibrium, the absorbed dose to any medium is given by(Khan 2010)

D =∑E

�Ep

(μen

ρ

)E

. (2)

μen

ρ= mass energy absorption coefficient (cm2 g−1) is obtained from tables (Hubbell and

Seltzer 1996); linear interpolation was used to estimate the mass energy absorption coefficientfor energies not included in the tables; � is the photon fluence (photons cm−2), Ep is the energyper photon (J), and the summation is carried out over all energies in the source spectrum.

For an arbitrary dose to the tissue without AuNP, equation (2) and the brachytherapysource spectra (see below) were used to estimate the incident photon fluence for each energyin the brachytherapy source spectrum. This fluence is then multiplied by the cross-sectionalarea of the spherical AuNP (πrAuNP

2, where rAuNP is the AuNP radius), to obtain the numberof incident photons on the AuNP. For the photons incident on the AuNP, the probability ofphotoelectric interaction is given by

N(x)

N0= 1 − e−(

μPEρ

)ρAuNPdAuNP ≈(

μPE

ρ

)ρAuNPdAuNP (3)

where(

μPE

ρ

)is the photoelectric absorption coefficient for gold obtained using the XCOM

software (Berger et al 2005), dAuNP is the average distance traversed by photons through aspherical AuNP (diameter × 2/3), and ρAuNP is the density of gold (19.32 g cm−3). The numberof interactions is then obtained by multiplying the probability for photoelectric interaction bythe number of incident source photons. The number of interactions equal the number ofphotoelectrons emitted.

The kinetic energy E of an emitted photoelectron is given by

E = Ep − Eedge (4)

where Eedge corresponds to the appropriate photoelectric absorption edge of gold (i.e.∼81 keV for K-edge, 12–14 keV for L-edge and 3 keV for M-edge). Each photoelectronemitted from a gold nanoparticle will deposit energy locally as a function of this kineticenergy. The energy loss for a statistical sample of photoelectrons will occur in a ‘sphere ofinteraction’, centered on the nanoparticle (figure 1(a).

To estimate the energy loss, the electron energy loss equation by Cole (1969) wasemployed. Cole determined that for electrons of 20 eV to 20 MeV, there is an empiricalrelation between the electron energy loss dE

dr(keV μm−1) and the residual range R (μm), in

unit density materials:dE

dR= 3.316(R + 0.007)−0.435 + 0.0055R0.33. (5)

6538 W Ngwa et al

Here R = Rtot − r, where r is the distance from the photoelectron emission site, and Rtot is thetotal range of the photoelectron for a kinetic energy, E:

Rtot = 0.431(E + 0.367)1.77 − 0.007. (6)

In calculating the energy lost by the photoelectron in the endothelium, the volume of thesphere of interaction (figure 1(a) that lies outside the endothelial cell is ignored. The methodof Makrigiorgos et al (1989, 1990a, 1990b) to calculate dose from a spherical source to aspherical target is used, albeit modified for the present slab geometry. The energy depositedwithin a volume is calculated by integration (equation (7)) over the differential energy loss(dE/dr) from the surface of the gold nanoparticle (Rn) to the distal side of the endothelial cell(DE):

Eendothelial =∫ Rn+DE

Rn

shellhemisphere − shellspherical cap beyond the endothelial cell

shellentire sphere× dE

drdr. (7)

In the integration, the hemispherical shell in the blood vessel is excluded, as is the sphericalshell beyond the endothelial cell. The absorbed dose is given by the energy deposited in theendothelial cell divided by its mass. It is assumed that each neighboring endothelial cellhas a similar nanoparticle attached; therefore energy that is deposited in an adjacent cell(‘cross-fire’) is accounted for. This calculation does not consider dose to endothelial cellsfrom nanoparticles located across the opposite side of the blood vessel wall. Given the energyof the emitted photoelectrons, the additional dose contribution from these AuNPs should notbe significant.

The calculated endothelial dose enhancement factor (EDEF) is defined as the doseabsorbed by the endothelial cell in the presence of AuNP divided by the absorbed dosewithout AuNP:

EDEF = Absorbed Dosewith AuNP

Absorbed Dosewithout AuNP.

Here, in calculating the absorbed dose (without AuNP) in the denominator of the EDEF, thenanoparticle is essentially replaced by water. An EDEF of 1.0 corresponds to no enhancement.An EDEF of 2.0 means that for every 1 Gy of dose that would have been delivered withoutAuNPs, 2 Gy is absorbed by the endothelium. Expressed alternatively, an EDEF of 2 istantamount to a 100% dose enhancement, etc. The effective EDEF for a polyenergetic beam(such as for the brachytherapy sources used in this calculation) includes a summation over allof the energy bins included in the energy spectrum.

Three brachytherapy sources were investigated: I-125, Pd-103 and Yb-169. The majordecay modes for these radionuclides are summarized in table 1. In addition a 50 kVp x-raysource (Cho et al 2009) was investigated. These sources were chosen as a representativesample with Pd-103 and I-125 representing common lower energy brachytherapy sources,Yb-169 representing a relatively higher energy–high dose rate source, and the 50 kVp sourcerepresenting an electronic source.

To estimate the energy deposited in the endothelial cell due to Auger electrons, spectraobtained by Monte Carlo simulations (Cho et al 2009) for tumors loaded with goldnanoparticles at 7 mg g−1 were employed. These spectra are shown in figure 2(a) for I-125, Yb-169 and 50 kVp brachytherapy sources. Figure 2(b) shows the Auger electron spectrum for Pd-103 calculated by the deterministic approach (Stepanek 2000). For perspective, the spectrumfor I-125 calculated with the deterministic approach is also shown in figure 2(b). From thespectra, the number of Auger electrons per source photon at higher gold concentrations maybe obtained by scaling the number of Auger electrons per source photon at 7 mg g−1 Au. Thisassumes that, for higher concentrations, additional nanoparticles attached to the endothelial

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6539

Table 1. Brachytherapy source spectra used for endothelial dose enhancement calculations (Choet al 2009, Roeske et al 2007).

Isotope Half-life (days) Energy (keV) Relative intensity

Pd-103 17.2 20.1 0.65623.0 0.125

I-125 59.4 22.1 0.2525.2 0.0727.4 1.0031.4 0.2535.5 0.06

Yb-169 32 49.5 0.53250.7 0.94057.6 0.29559.1 0.08263.1 0.44293.6 0.026

109.8 0.175118.2 0.019130.5 0.113177.2 0.222198.0 0.358261.1 0.017307.7 0.101

cell experience the same photon fluence. This is the case if the considered tumor vasculatureis within the planning target volume (PTV), prescribed to receive spatially uniform dose.

To calculate the energy deposited due to Auger electrons, the photon fluence wasdetermined as described for the photoelectrons. The calculated fluence was then used todetermine the number of source photons reaching an AuNP with cross-sectional area πrAuNP

2.For each Auger electron energy bin, the number of Auger electrons emitted was calculatedby multiplying the number of source photons by the number of Auger electrons per sourcephoton. Equations (5)–(7) were then used to calculate the energy deposited in the endothelialcell for each Auger electron energy bin. The total energy deposited for each energy bin equalsthe number of Auger electrons emitted multiplied by the corresponding energy deposited.Summing over all energy bins then yields the total energy deposited in the endothelial cell.

3. Results

Results for the endothelial dose enhancement, due to AuNP-emitted photo-electrons, as afunction of local AuNP concentration are shown in figure 3. As expected, the EDEF increaseswith AuNP concentration. The results also show that the lower energy brachytherapy sourceslike Pd-103 engender the highest EDEF, followed by I-125 and 50 kVp; Yb-169 yields thelowest EDEF relative to the lower energy sources. This is understandable when considered inthe light of the decay energies in table 1. For example, for Pd-103 with an average energy ofabout 21 KeV, the kinetic energy of a photoelectron emitted by a gold particle after interactionwould have an average energy of about 8 keV (= 21 keV − 13 keV, where 13 keV is the

6540 W Ngwa et al

(a)

0 10 20 30 40 50 60 70 8010

-6

10-5

10-4

10-3

10-2

10-1

100

Electron Energy (KeV)

Auger

ele

ctro

ns

per

sourc

e p

hoto

n I-125 and 7 mg AuYb-169 and 7 mg Au50 kVp and 7 mg Au

(b)

0 2 4 6 8 10 12 1410

-6

10-5

10-4

10-3

10-2

10-1

100

Electron Energy (keV)

Aug

er e

lect

rons

per

sou

rce

phot

on

I-125 and 7 mg AuPd-103 and 7 mg Au

Figure 2. (a) Auger electron spectra for I-125, Yb-169 and 50 kVp x-ray sources. Monte Carlo-generated spectral data are obtained from Cho et al (2009). (b) Auger electron spectra for Pd-103obtained by the deterministic approach (Stepanek 2000). The spectrum of I-125 calculated by thedeterministic approach is also shown.

average L-edge of gold). According to equation (6), the range of such an electron would beabout 1.84 μm. I-125 has an average photon energy of about 29 keV, corresponding to a rangeof 6.02 μm. Despite the fact that an emitted photoelectron would carry a kinetic energy ofabout 16 keV, the energy loss (equation (5)) within the EC (of thickness 2 μm) is less thanthat for Pd-103. However, employing I-125 source yields a relatively much higher EDEFcompared to Yb-169. The latter has an average energy of 93 keV but the spectral energieswith a high magnitude of relative intensities are at 50.7, 49.5 and 63.1 keV, which are belowthe K-edge of gold, ∼81 KeV, and far above the L-edge. Hence these energies have lowerphotoelectric absorption coefficients; these typically decrease with higher energy between theabsorption edges, with higher values for the lower energy sources. The highest values arefor Pd-103, followed by I-125 and then Yb-169. So while the EDEF for Yb-169 may besignificant, the commonly used lower energy sources like Pd-103 and I-125 yield much higherEDEFs. Meanwhile, the EDEF for the 50 kVp x-ray source is very close to the values obtainedfor I-125. This is likely due to the fact that the 50 kVp source has an average energy close

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6541

0 50 100 150 200 250 300 3500

5

10

15

20

25

Concentration (mg/g)

ED

EF

Pd-103I-125Yb-16950 kVp

Figure 3. Endothelial dose enhancement factor (EDEF) versus local Au nanoparticle concentrationat the endothelium due to photoelectrons. AuNP-emitted Auger electrons are not included in thecalculation of the EDEF.

to that of I-125. Overall, these results suggest that lower energy brachytherapy sources likePd-103 would provide maximal endothelial dose enhancement from photoelectrons, and maybe more advantageous for potential VDA applications.

While the focus in this study is on the endothelial cell, one may, with a calculationaladjustment, be able to estimate how results obtained using the approach compare with otherstudies, assuming homogenous AuNP distribution through out the tumor. This adjustment ismade by calculating the integral in equation (7) from Rn to Rtot. This means that one includesthe dose deposited in the entire sphere of interaction, as opposed to restricting the calculationto the dose deposited on the endothelial cell on which AuNPs are attached. This adjustmentyields EDEF values of about 57%, 51% and 41% for I-125, 50 kVp and Yb-169, respectively,for 7 mg g−1 concentration. These results compare well with those from Cho et al estimatingthe macroscopic dose enhancement from a homogenous distribution of gold nanoparticles inthe entire tumor (Cho et al 2009). They concluded that the macroscopic dose enhancementfor I-125, 50 kVp and Yb-169 ranges from 40% to 70% at a uniform gold concentration of7 mg g−1.

EDEF dependence on endothelial cell thickness up to 5 μm is shown in figure 4, forconstant 18 mg g−1 AuNP concentration. The results illustrate that the EDEF peaks atdifferent thicknesses for the different brachytherapy sources e.g. the EDEF for Pd-103 peaksrelatively earlier at about 3 μm thickness, while I-125 peaks beyond 5 μm. Among thesources investigated, Pd-103 exhibits the highest EDEF values for thicknesses up to about4 μm, while I-125 post the highest EDEFs when endothelial thicknesses between 4 and 5 areconsidered. Again, these results reinforce the idea that lower energy sources like Pd-103 maybe, dosimetric-wise, better suited for potential VDA applications than higher energy sourceslike Yb-169.

Turning to Auger electrons, the EDEF due to Auger electrons as a function of AuNPconcentration is shown in figure 5. As expected the EDEF increases with the AuNPconcentration. Here, substantially higher dose enhancement is also observed for the lowerenergy sources.

6542 W Ngwa et al

1 1.5 2 2.5 3 3.5 4 4.5 51

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

thickness (microns)

ED

EF

Pd-103 and 18 mg/g concentrationI-125 and 18 mg/g concentrationYb-169 and 18 mg/g concentration50 kVp and 18 mg/g concentration

Figure 4. Endothelial dose enhancement factor (EDEF) versus endothelium wall thickness at18 mg g−1 local concentration. AuNP-emitted Auger electrons are not included in the calculationof the EDEF.

0 50 100 150 200 250 300 3500

50

100

150

200

250

300

Concentration (mg/g)

ED

EF

Pd-103I-125Yb-16950 kVp

Figure 5. Endothelial dose enhancement factor (EDEF) due to Auger electrons from gold versuslocal gold nanoparticle concentration.

With respect to thickness, it is observed (figure 6) that the EDEFs due to Auger electronsapparently peak at smaller thicknesses compared to those for photoelectrons (figure 4). Thisclearly illustrates the fact that Auger electrons are of shorter range and hence may depositmore of their energy within the endothelial cell compared to photoelectrons.

Table 2 summarizes the results for EDEFs due to photo/Auger electrons at 7, 18 and350 mg g−1 concentrations from figures 3 and 6, for the endothelial thickness of 2 μm. Thetotal EDEF is also shown for the different brachytherapy sources. The total EDEF is given by

Total EDEF = DPE + DAuger

DW

.

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6543

1 1.5 2 2.5 3 3.5 4 4.5 55

10

15

20

Cell thickness (microns)

ED

EF

Pd-103 and 18 mg AuI-125 and 18 mg AuYb-169 and 18 mg Au50 kVp and 18 mg Au

Figure 6. Endothelial dose enhancement factor (EDEF) due to Auger electrons from gold versusendothelium wall thickness at 18 mg g−1 local concentration.

Here, DW is the dose to the endothelial cell without AuNPs; DPE is the dose to the endotheliumwhen AuNPs are present, with dose enhancement due to photoelectrons; DAuger is the dose tothe endothelium when AuNPs are present, with dose enhancement due to Auger electrons. Itis evident from the results in table 2 that Pd-103 scores the highest total EDEF values of about7.41–271.48 for a concentration range of 7–350 mg g−1. Though Yb-169 yields the lowesttotal EDEFs (3.95–99.66), these values are still substantial. The results (table 2) also indicatethat the Auger electron contribution to the total endothelial dose enhancement is higher thanthat due to photoelectrons. This is apparently due to the shorter range of Auger electrons,allowing them to deposit their energies more locally

The average number of photoelectrons per incident photon produced by the 400 nmparticle was estimated to be about 0.037, 0.018, 0.003 and 0.019 for Pd-103, I-125, Yb-169and 50 kVp sources, respectively. Meanwhile, the average number of Auger electrons persource photon at 7 mg g−1 concentration was 1.11, 0.62, 0.12 and 0.56, for Pd-103, I-125,Yb-169 and 50 kVp sources, respectively, consistent with Cho et al (2009). It should also bementioned that the calculated dose deposited in the endothelial cells by these photo/Augerelectrons represent average dose estimates. In particular, for lower concentrations, stochasticeffects may be important.

In perspective, Jones et al (2010) have shown that the microscopic dose enhancementwithin a tumor loaded with gold is enhanced by factors ranging from 2 to 20 within 5 μm ofthe nanoparticles, for a test case based on a realistic gold nanoparticle distribution in tissue.In the current study, a slight adjustment of the calculation (as done above), to include the dosedeposited in the entire sphere of interaction, shows that for 7 mg g−1, the total EDEFs are about:7.5, 6.8, 4.5, 6.7, for Pd-103, I-125, Yb-169 and 50 kVp sources, respectively. Though slightlydifferent, these values are in concordance with Jones et al since a non-uniform distributionwithin the tumor will result in sub-regions of varying concentrations.

Overall, the data in figures 3–6 indicate that even traces (∼1 mg g−1) of AuNPs specificallytargeted to the tumor endothelium would elicit substantial brachytherapy dose enhancement,while modest to higher concentrations (>18 mg g−1) would result in ablative doses. Recentwork reveals that endothelial response in mouse and human tumor specimens both display an

6544W

Ngw

aetal

Table 2. Summary of endothelial dose enhancement due to photo/Auger electrons at 7, 18 and 350 mg g−1 for the investigated brachytherapy sources. The results shown are forendothelial cell thickness of 2 μm, and nanoparticle diameter of 400 nm.

EDEF due to photoelectrons EDEF due to Auger electrons Total EDEF

7 mg g−1 18 mg g−1 350 mg g−1 7 mg g−1 18 mg g−1 350 mg g−1 7 mg g−1 18 mg g−1 350 mg g−1

Pd-103 1.42 2.09 22.15 5.99 13.82 250.33 7.41 15.91 271.48I-125 1.26 1.66 13.79 5.10 11.55 206.12 6.36 13.21 219.91Yb-169 1.03 1.09 2.71 2.92 5.93 96.95 3.95 7.02 99.6650 kVp 1.25 1.65 13.63 5.00 11.28 200.83 6.25 12.93 214.46

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6545

apparent threshold of 8–10 Gy and a maximal response at 20–25 Gy of external beam irradiation(Fuks and Kolesnick 2005). This suggests that, depending on the AuNP concentrationsbound to the endothelium, such maximal response may be achieved with the use of AuNPs,potentially even during a single brachytherapy treatment fraction. The possibility for suchmaximal response may also help elucidate the observations by Hainfeld et al in recent pre-clinical experiments with mouse models employing AuNPs during radiotherapy (Hainfeldet al 2004, 2010). The authors of the earlier work (Hainfeld et al 2004) noted that tumorsbecame hemorrhagic before shrinking, indicating ‘catastrophic endothelial damage’, whenAuNPs were included.

4. Discussion

VDAs as originally proposed by Denekamp (1982, 1984) are designed to selectively targetthe established tumor blood vessels. This is the main presupposition of this study, whichassumes the feasibility of specifically targeting the endothelium with gold nanoparticles atthe investigated concentration levels. This assumption is justifiable given that pre-clinicalwork has already demonstrated the feasibility for 7 and 18 mg g−1 concentrations throughpassive targeting (Hainfeld et al 2004). Besides, a significant body of research has alreadydemonstrated active targeting of tumors with gold nanoparticles associated with peptides,antibodies, oligonucleotides and liposomes (Sokolov et al 2003, Choi et al 2010, Brownet al 2010, Dabbas et al 2008, Cai et al 2006, Diagaradjane et al 2008, McCarthy andWeissleder 2008, Qian et al 2008). Given the greater accessibility to the endothelium, andits unique features compared to healthy tissue endothelium, it may be possible to adaptsuch active targeting strategies for potential application of AuNPs as VDAs. So while thenanoparticle size of 400 nm was chosen to restrict the AuNPs within the vasculature, it couldbe possible to employ active targeting strategies to restrict smaller size gold nanoparticlesto the endothelium to achieve similar concentration levels. The results of this work mayprovide further momentum for developing such active targeting strategies. Such advances forAuNPs may also have spin-off benefits for other applications like imaging, e.g. in enhancingvisualization of the tumor vasculature (Hainfeld et al 2006).

VDAs are designed to be administered acutely to secure more rapid effects (Gridelliet al 2009). The high EDEFs in this study would certainly constitute acute administration ofbrachytherapy dose. In fact, the high EDEF results in this study suggest that for realistic AuNPconcentrations, it is theoretically possible to deliver ablative dose (equivalent to dose deliveredduring a complete HDR brachytherapy treatment course) to the tumor endothelial cells in onetreatment fraction. Such ablative dose to a small endothelial cell volume would more rapidlydisrupt the tumor vasculature. This could, in turn, result in massive ‘downstream’ tumourcell killing similar to VDAs. A study by Garcia-Barros et al (2003) revealed that exposure ofmouse MCA129 fibrosarcoma and B16 melanoma to single doses of 15–20 Gy is followed bya rapid wave of endothelial apoptosis at 1–6 h as the earliest anatomical evidence of radiationdamage. The current study indicates that such 15–20 Gy endothelial dose enhancementmay be attained even during a single brachytherapy treatment fraction. It would thereforebe theoretically feasible to induce rapid endothelial apoptosis, in consonance with expectedresponse from typical VDAs.

VDAs should, ideally, not engender additional toxicity to healthy tissue. The concern isthat more than just the tumor vasculature may be targeted by systemic exposure to VDAs, withpotential damage to vascular compartments outside the tumor. This may contribute to acutecoronary syndromes and thromboembolic events (van Heeckeren et al 2006). In fact closemonitoring of patients receiving VDAs for any cardiovascular toxicity is usually imperative

6546 W Ngwa et al

(Cooney et al 2006). In this regard, AuNPs may actually have an advantage in potentialapplications as VDAs since gold is non-toxic (Mukherjee et al 2005) and biocompatible(Shukla et al 2005). Indeed, AuNPs at the investigated concentration levels have been shownto be relatively inert with insignificant toxicity in pre-clinical experiments (Hainfeld et al2004). However, toxicity characterization in humans would be a crucial determinant forapplying AuNPs as VDAs.

There is growing consensus that VDAs are more effective as adjuvants to other therapies,e.g. radiotherapy (Siemann and Shi 2003; Cooney et al 2006). This is partly because in VDAapplications, the rapid induction of central necrosis may still leave a rim of viable and possiblymalignant cells persisting at the tumor-normal tissue interface (Skliarenko et al 2006). In theapproach considered in this study, the AuNPs would essentially serve as such adjuvants—tobrachytherapy. Also, AuNPs have been shown to exhibit antiangiogenic properties (Mukherjeeet al 2005), so their applications as adjuvants or in combination therapy may be feasible andin line with the current thinking. In such combination therapy, timing when to administer theAuNPs may be important in determining the effectives of treatment outcomes.

Concerning the different brachytherapy sources used in this study, it is evident that Pd-103provides the highest endothelial dose boost. So from a dosimetric standpoint it may be themost appealing source for use with AuNPs in potential VDA application. However, the choiceof source may depend on more than just how much endothelial dose enhancement it provides.In potential applications, one may have to also consider the source’s half-life in conjunctionwith the biological half-life of the AuNPs in vivo, etc. With respect to the source’s half-life,x-ray sources like the 50 kVp may hold an advantage over the other low energy sources.Active targeting would also be a factor to consider, here. All in all, it is evident that morestudies, especially experimental, are required to further corroborate the current results anddetermine optimal potential application design parameters. Experiments at the preclinicallevel are planned.

5. Conclusion

We examined the potential for applying gold nanoparticles as tumor vascular disruptingagents, during brachytherapy. The results suggest that targeting AuNPs to the endothelium,in conjunction with lower energy brachytherapy sources, may yield very substantial doseenhancement to endothelial cells targeted by AuNPs. Such dose enhancement may be sufficientto rapidly elicit endothelial response and consequential apoptosis, with potentially no increasedtoxicity to healthy tissue not containing AuNPs. The findings provide significant impetus forconsidering the application of gold nanoparticles as tumor-vascular disrupting agents duringbrachytherapy.

References

Berger M J, Hubbell J H, Seltzer S M, Chang J, Coursey J S, Sukumar R and Zucker D S 2005 XCOM: Photo CrossSection Database version 1.3 (Gaithersburg, MD: NIST) http://physics.nist.gov/xcom

Boerman O C, Sharkey R M, Blumenthal R D, Aninipot R L and Goldenberg D M 1992 The presence of a concomitantbulky tumor can decrease the uptake and therapeutic efficacy of radiolabeled antibodies in small tumors Int. J.Cancer 51 470–5

Brown S D, Nativo P, Smith J, Stirling D, Edwards P R, Venugopal B, Flint D J, Plumb J A, Graham D and WheateN J 2010 Gold nanoparticles for the improved anticancer drug delivery of the active component of oxaliplatinJ. Am. Chem. Soc. 132 4678

Cai W, Shin D W, Chen K, Gheysens O, Cao Q, Wang S X, Gambhir S S and Chen X 2006 Peptide-labelednear-infrared quantum dots for imaging tumor vasculature in living subjects Nano Lett. 6 669–76

Applying gold nanoparticles as tumor-vascular disrupting agents during brachytherapy 6547

Cho S H 2005 Estimation of tumour dose enhancement due to gold nanoparticles during typical radiation treatments:a preliminary Monte Carlo study Phys. Med. Biol. 50 N163–73

Cho S H, Jones B L and Krishnan S 2009 The dosimetric feasibility of gold nanoparticle-aided radiation therapy(GNRT) via brachytherapy using low-energy gamma-/x-ray sources Phys. Med. Biol. 54 4889–905

Choi C H, Alabi C A, Webster P and Davis M E 2010 Mechanism of active targeting in solid tumors with transferrin-containing gold nanoparticles Proc. Natl Acad. Sci. USA 107 1235–40

Cole A 1969 Absorption of 20-eV to 50,000-eV electron beams in air and plastic Radiat. Res. 38 7–33Cooney M M, van Heeckeren W, Bhakta S, Ortiz J and Remick S C 2006 Drug insight: vascular disrupting agents

and angiogenesis—novel approaches for drug delivery Nat. Clin. Pract. Oncol. 3 682–92Dabbas S, Kaushik R R, Dandamudi S, Kuesters M and Campbell R B 2008 Importance of the lipoosomal cationic

lipid content and type in tumor vascular targeting: physicochemical characterization and in vitro studies usinghuman primary and transformed endothelial cells Endothelium 15 189–201

Denekamp J 1982 Endothelial cell proliferation as a novel approach to targeting tumour therapy Br. J. Cancer45 136–9

Denekamp J 1984 Vascular endothelium as the vulnerable element in tumours Acta Radiol. Oncol. 23 217–25Denekamp J 1993 Review article: angiogenesis, neovascular proliferation and vascular pathophysiology as targets

for cancer therapy Br. J. Radiol. 66 181–96Desgrosellier J S and Cheresh D A 2010 Integrins in cancer: biological implications and therapeutic opportunities

Nat Rev Cancer 10 9–22Diagaradjane P, Orenstein-Cardona J M, Colon-Casasnovas N E, Deorukhkar A, Shentu S, Kuno N, Schwartz D L,

Gelovani J G and Krishnan S 2008 Imaging epidermal growth factor receptor expression in vivo: pharmacokineticand biodistribution characterization of a bioconjugated quantum dot nanoprobe Clin. Cancer Res. 14 731–41

Fuks Z and Kolesnick R 2005 Engaging the vascular component of the tumor response Cancer Cells 8 89–91Garcia-Barros M, Paris F, Cordon-Cardo C, Lyden D, Rafii S, Haimovitz-Friedman A, Fuks Z and Kolesnick R 2003

Tumor response to radiotherapy regulated by endothelial cell apoptosis Science 300 1155–9Gridelli C, Rossi A, Maione P, Rossi E, Castaldo V, Sacco P C and Colantuoni G 2009 Vascular disrupting agents: a

novel mechanism of action in the battle against non-small cell lung cancer Oncologist 14 612–20Hainfeld J F, Dilmanian F A, Slatkin D N and Smilowitz H M 2008 Radiotherapy enhancement with gold nanoparticles

J. Pharm. Pharmacol. 60 977–85Hainfeld J F, Dilmanian F A, Zhong Z, Slatkin D N, Kalef-Ezra J A and Smilowitz H M 2010 Gold nanoparticles

enhance the radiation therapy of a murine squamous cell carcinoma Phys. Med. Biol. 55 3045–59Hainfeld J F, Slatkin D N, Focella T M and Smilowitz H M 2006 Gold nanoparticles: a new x-ray contrast agent Br.

J. Radiol. 79 248–53Hainfeld J F, Slatkin D N and Smilowitz H M 2004 The use of gold nanoparticles to enhance radiotherapy in mice

Phys. Med. Biol. 49 N309–15Hashizume H, Baluk P, Morikawa S, McLean J W, Thurston G, Roberge S, Jain R K and McDonald D M 2000

Openings between defective endothelial cells explain tumor vessel leakiness Am. J. Pathol. 156 1363–80Herold D M, Das I J, Stobbe C C, Iyer R V and Chapman J D 2000 Gold microspheres: a selective technique for

producing biologically effective dose enhancement Int. J. Radiat. Biol. 76 1357–64Hubbell J H and Seltzer S M 1996 Tables of x-ray mass attenuation coefficients and mass energy-absorption coefficients

from 1 keV to 20 MeV for elements Z = 1 to 92 and 48 additional substances of dosimetric interest (NIST)Jones B L, Krishnan S and Cho S H 2010 Estimation of microscopic dose enhancement factor around gold nanoparticles

by Monte Carlo calculations Med. Phys. 37 3809–16Khan F M 2010 The Physics of Radiation Therapy (Baltimore, MD: Wolters Kluwer)Maeda H, Wu J, Sawa T, Matsumura Y and Hori K 2000 Tumor vascular permeability and the EPR effect in

macromolecular therapeutics: a review J. Control. Release 65 271–84Makrigiorgos G M, Adelstein S J and Kassis A I 1989 Limitations of conventional internal dosimetry at the cellular

level J. Nucl. Med. 30 1856–64Makrigiorgos G M, Adelstein S J and Kassis A I 1990a Cellular radiation dosimetry and its implications for

estimation of radiation risks. Illustrative results with technetium 99m-labeled microspheres and macroaggregatesJAMA 264 592–5

Makrigiorgos G M, Ito S, Baranowska-Kortylewicz J, Vinter D W, Iqbal A, Van den Abbeele A D, Adelstein S Jand Kassis A I 1990b Inhomogeneous deposition of radiopharmaceuticals at the cellular level: experimentalevidence and dosimetric implications J. Nucl. Med. 31 1358–63

McCarthy J R and Weissleder R 2008 Multifunctional magnetic nanoparticles for targeted imaging and therapy Adv.Drug Deliv. Rev. 60 1241–51

Mukherjee P, Bhattacharya R, Wang P, Wang L, Basu S, Nagy J A, Atala A, Mukhopadhyay D and Soker S 2005Antiangiogenic properties of gold nanoparticles Clin. Cancer Res. 11 3530–4

6548 W Ngwa et al

Qian X, Peng X H, Ansari D O, Yin-Goen Q, Chen G Z, Shin D M, Yang L, Young A N, Wang M D and Nie S2008 In vivo tumor targeting and spectroscopic detection with surface-enhanced Raman nanoparticle tags Nat.Biotechnol. 26 83–90

Roeske J C, Nunez L, Hoggarth M, Labay E and Weichselbaum R R 2007 Characterization of the theorectical radiationdose enhancement from nanoparticles Technol. Cancer Res. Treat. 6 395–401

Ryschich E, Schmidt E, Maksan S M, Klar E and Schmidt J 2004 Expansion of endothelial surface by an increase ofvessel diameter during tumor angiogenesis in experimental and hepatocellular and pancreatic cancer World J.Gastroenterol. 10 3171–4

Shen Y, Zhang A and Xu L 2009 A study on mechanical damage of tumor microvasculature induced by alternatecooling and heating J. Ther. Sci. Eng. Appl. 1 031002

Shukla R, Bansal V, Chaudhary M, Basu A, Bhonde R R and Sastry M 2005 Biocompatibility of gold nanoparticlesand their endocytotic fate inside the cellular compartment: a microscopic overview Langmuir 21 10644–54

Siemann D W and Shi W 2003 Targeting the tumor blood vessel network to enhance the efficacy of radiation therapySemin. Radiat. Oncol. 13 53–61

Skliarenko J V, Lunt S J, Gordon M L, Vitkin A, Milosevic M and Hill R P 2006 Effects of the vascular disruptingagent ZD6126 on interstitial fluid pressure and cell survival in tumors Cancer Res. 66 2074–80

Sokolov K, Follen M, Aaron J, Pavlova I, Malpica A, Lotan R and Richards-Kortum R 2003 Real-time vital opticalimaging of precancer using anti-epidermal growth factor receptor antibodies conjugated to gold nanoparticlesCancer Res. 63 1999–2004

Stepanek J 2000 Methods to determine the fluorescence and Auger spectra due to decay of radionuclides or due to asingle atomic-subshell ionization and comparisons with experiments Med. Phys. 27 1544–54

Stoletov K, Montel V, Lester R D, Gonias S L and Klemke R 2007 High-resolution imaging of the dynamic tumorcell vascular interface in transparent zebrafish Proc. Natl Acad. Sci. USA 104 17406–11

Unezaki S, Maruyama K, Hosoda J, Nagae I, Koyanagi Y, Nakata M, Ishida O, Iwatsuru M and Tsuchiya S 1996Direct measurement of the extravasation of polyethyleneglycol-coated liposomes into solid tumor tissue by invivo fluorescence microscopy Int. J. Pharm. 144 11–7

van Heeckeren W J, Bhakta S, Ortiz J, Duerk J, Cooney M M, Dowlati A, McCrae K and Remick S C 2006 Promiseof new vascular-disrupting agents balanced with cardiac toxicity: is it time for oncologists to get to know theircardiologists? J. Clin. Oncol. 24 1485–8

Warrell D A, Cox T M, Firth J D and Benz E J 2005 Oxford Textbook of Medicine (London: Oxford University Press)Zhang S X, Gao J, Buchholz T A, Wang Z, Salehpour M R, Drezek R A and Yu T-K 2009 Quantifying tumor-

selective radiation dose enhancements using gold nanoparticles: a Monte Carlo simulation study Biomed.Microdevice 11 925–33

Zhu X, Palmer M R, Makrigiorgos G M and Kassis A I 2010 Solid-tumor radionuclide therapy dosimetry: newparadigms in view of tumor microenvironment and angiogenesis Med. Phys. 37 2974–84