Physical Perspective on Radiation
Response Modulation of Tumors
with Gold Nanoparticles
Sang Hyun Cho, Ph.D.
Medical Physics Program
Woodruff School of Mechanical Engineering
Georgia Institute of Technology
Atlanta, Georgia
Acknowledgment
• Georgia Tech:
S.-K. Cheong, B. Jones, A. Siddiqi, F. Liu,
N. Manohar, K. Dextraze
• M. D. Anderson Cancer Center:
S. Krishnan, P. Diagaradjane
• Georgia Research Alliance
Perfecting Radiation therapy …
Prologue
Variation of
radiation quality
h, e-, n, p, …
Manipulation of
radiation intensity / fluence
Control of organ motion
Image guidance / tumor tracking
Radiation
Response
Modulation of
Tumor
1
Radiation Response Modulation of
Tumors using high Z media
● Almost universally achievable in theory
with enough accumulation (< ~ 10 wt. %)
Rationale for Use of GNPs
● Efficacy varying dependent on the ability
to control the two following parameters:
- Amount of physical dose enhancement
- Location of high Z materials: sites of
interactions with radiation
Why GNPs ?
Why GNPs ?
• Mediators for increased secondary electron
production in tissue irradiated with h, e-, p, etc.
• Possibility of uniform delivery of gold to tumor
site due to particle size in nm range: enhanced
permeability and retention (EPR) due to passive
extravasation of nanoparticles through leaky
- Example: Interaction probability for
photoelectric absorption ~ Z4 to Z4.8
tumor vasculature  “Passive Targeting”
water
- iodine (Z=53)
- gadolinium (Z=64)
- platinum (Z=78)
- gold (Z=79)
Adapted from Brigger et al, Adv. Drug Deliv. Rev. 54, 2002
2
Why GNPs ?
Why GNPs ?
• Possibility of achieving high tumor specificity of
gold by conjugating GNPs with antibodies for
tumor markers (e.g., EGFR) or angiogenesis
markers (e.g., VEGFR)  “Active Targeting”
• Possibility of achieving significant cellular
uptake of GNPs via active targeting 
“Internalization”: high Z gold in close proximity
to the cell nucleus and DNA, the main target for
radiation damage
Adapted from Katz et al, Nanoparticles, Ed. G. Schmid 2004
Why GNPs ?
Why GNPs ?
• Possibility to generate therapeutic dose of heat
through photothermal effect / plasmon resonance
• acceptable toxicity for high concentration (up to
3% of body weight) of gold in the body
- potential toxicity of GNPs: inconclusive at this
time and needs to be investigated further
• possibility to perform in-vivo molecular imaging
before/during/after the treatment through x-ray
fluorescence imaging (e.g., XFCT)
1.2
Abs max
Source: Nanospectra Bioscience, Inc.
Normalized OD
1.0
0.8
0.6
0.4
0.2
0.0
400
500
600
700
800
900
Wavelength (nm)
3
Physical Dose Enhancement around GNPs
GNP-mediated Physical
Dose Enhancement
Physical Dose Enhancement around GNPs
Macroscopic Estimation of
Dose Enhancement
• Average dose enhancement across tumor
• Applicable to passive targeting scenario or
so-called contrast-enhanced radiation therapy
• Dose enhancement as a function of GNP
concentration and photon beam quality
- Uniform distribution of GNPs within the tumor
- Uniform mixture of gold and tissue
- MC calculations or Analysis of energy absorption
coefficients
4
MC Simulation Geometry:
Macroscopic Dose Enhancement Factor (MDEF)
External beam cases
External beam cases
photon/x-ray beam
SSD
tumor + GNPs
Concentration
( per gram of tumor )
140 kVp
7 mg Au
2.114
1.007
1.014
1.009
1.019
18 mg Au
3.811
1.015
1.032
1.019
1.044
30 mg Au
5.601
1.025
1.053
1.032
1.074
6 MV FF 6 MV NFF 4 MV FF 4 MV NFF
FF: flattening filter, NFF: no flattening filter
tissue phantom
MDEF = average tumor dose with Au / average tumor dose without Au
(30 x 30 x 30 cm 3)
Macroscopic Dose Enhancement Factor (MDEF)
250 kVp in-vivo experiment
MC Simulation Geometry:
Brachytherapy cases
tumor + GNPs
250 kVp MDEF(50 cm SSD, 1.5x1.5 FS, 1 cm^3 tumor)
3.0
no gold
2.5
0.07% gold
0.2% gold
0.7% gold
2.0
MDEF
1.8% gold
1.5
1.0
0.5
0.0
0.0
0.5
1.0
1.5
Depth (cm)
2.0
2.5
3.0
30 cm radius tissue phantom
brachytherapy source
at the center
5
MDEF: Ir-192 & Yb-169
MDEF: 50 kVp & I-125
3.0
3.0
Yb-169 & 7 mg Au
Yb-169 & 18 mg Au
Yb-169 & 7 mg Au + 2 mg Au
Ir-192 & 7 mg Au
Ir-192 & 18 mg Au
Ir-192 & 30 mg Au
2.5
50 kVp & 7 mg Au
50 kVp & 18 mg Au
50 kVp & 7 mg Au + 2 mg Au
I-125 & 7 mg Au
I-125 & 18 mg
I-125 & 7 mg + 2 mg Au
2.5
2.0
MDEF
MDEF
2.0
1.5
1.5
1.0
1.0
0.5
tumor
tumor
tissue
0.5
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
0.0
1.0
Radial Distance (cm)
3.0
4.0
5.0
6.0
7.0
Water
1.E-02
1.E-03
1.E-04
1.E-05
1.E-06
Energy (MeV)
0.03
0.04
Electrons per Source Photon
1.E-03
Water and 0.7 wt% Au
0.02
10.0
6 MV photon irradiation
1.E+00
0.01
9.0
Secondary Electron Spectra for 6 MV X-rays
Secondary Electron Spectra for 125-I
1.E-01
8.0
Secondary electron spectra
I-125 gamma ray irradiation
1.E-07
0.00
2.0
Radial Distance (cm)
Secondary electron spectra
Electrons per Source Photon
tissue
0.0
0.0
Water and 0.7 wt% Au
Water
1.E-04
1.E-05
1.E-06
1.E-07
1.E-08
0.01
0.10
1.00
10.00
Energy (MeV)
6
Experimental Verification of Dose Enhancement
GNP-loaded Radio-sensitive Gels irradiated by 110 kVp
Microscopic Estimation of Dose Enhancement
• Dose enhancement around GNP on nanoscale
• Applicable to active/passive targeting scenarios
with heterogeneous distribution of GNPs
MAGIC+Formaldehyde Gel Calibration Curve
5.50
5.40
• Basis for GNP-mediated DNA damage calculations
1% Au+12 Gy
5.30
5.20
R2 (1/s)
5.10
y = 0.0242x + 4.6565
R2 = 0.9037
5.00
4.90
4.80
4.70
1% Au + No
radiation
4.60
4.50
0
5
10
15
20
25
30
Dose (Gy)
• Computational Approaches
- Detailed history MC calc with hypothetical
dimensionless GNPs
- Mixed history MC calc with various sizes of GNPs
- Semi-analytic calc using microscopic e- energy
deposition kernel/pattern
Microscopic Estimation of Dose Enhancement
Microscopic Estimation of Dose Enhancement
microscopic dose enhancement factor (mDEF)
Dose point kernels
GNP
Hypothetical Water NP
7
Microscopic Estimation of Dose Enhancement
Microscopic Estimation of Dose Enhancement
microscopic dose enhancement factor (mDEF)
microscopic dose enhancement factor (mDEF)
Microscopic Estimation of Dose Enhancement
Microscopic Estimation of Dose Enhancement
microscopic dose enhancement factor (mDEF)
alternative approach using mixed history MC simulations
250 kVp
21 MeV e-
GNP d=20 nm
H2O d=20 nm
250 MeV p
GNP d=20 nm
H2O d=20 nm
8
Modeling of GNP-mediated
Radiation Response Modulation
Correlation between physical
model and biological outcome
● Quantification of physical dose
enhancement on micro-/nano-scale
(cellular & DNA level estimation)
● Identification of GNP locations within
tumor (cell)
● Identification of proper biological
pathways (known & unknown)
Cellular & DNA level quantification of
microscopic dose enhancement
Microscopic dose enhancement in tissue
2 nm
on the order of 10 m
e-
e-
e-
ee-
ee-
micro-/cellular-dosimetry regime
nanodosimetry regime
6 MV
Left image re-created from Herold et al, Int. J. Radiat. Biol. Vol. 76, 2000
9
Microscopic dose enhancement in tissue
Microscopic dose enhancement in tissue
Microscopic dose enhancement in tissue
GNP-mediated radiosensitization with 6 MV?
• In-vitro setting: ~ 20% per cell survival assays
h
10s of m
Predictable using calculated mDEF for 6 MV !
10
GNP-mediated radiosensitization with 6 MV?
• In-vivo setting: Inconclusive
Gold Nanoparticle-aided
Radiation Therapy (GNRT)
Physically, no significant dose enhancement again,
but the location of GNP is probably the key ….
Biological Consequences of GNRT
● Dual action
Various GNRT scenarios
h
h
h
Active targeting
Active targeting
- tumor blood vessel disruption
- enhanced cell-killing: increase in
relative biological effectiveness (RBE)
for radiation
● Further improvement of therapeutic
ratio for radiotherapy
Passive targeting
Or
GNP contrast agent
Internalized GNPs
t
11
Radiation sources for GNRT
● Brachytherapy sources
- Yb-169
Epilogue
- 50 kVp miniature x-ray source
● External beam sources
- low-energy enhanced MV beam
- synchrotron/monochromatic or
orthovoltage x-ray beams:
tomotherapy/rotational delivery
- electron and proton
Radiation Response Modulation of
Tumors using GNPs
● The concept appears to be applicable to
human cancer treatment
Thank you for your attention !
● Physical model alone may not be
adequate to explain all biological
outcomes
● Even so, physical model based on
realistic outlook may still play an
important role for clinical translation
12