Transcript Illinois Institute of Technology

Illinois Institute of Technology
Physics 561
Radiation Biophysics
Lecture 9:
Cancer, Genetics, and
High LET radiation
30 June 2014
Andrew Howard
7/7/2015
Stochastic Damage; High-LET
p. 1 of 75
Plan for this lecture
u
u
Cancer, concluded
–
–
–
–
Genetics, concluded
–
Mutagenesis
–
RNA vs. DNA
–
Animals & humans
–
–
Dose-response
–
–
Latency
–
The genetic code
Relative sensitivities
Organismal differences
u
Genetics
–
u
Chromosomal damage
Recombination
Frameshifts
Genes and Introns
Structural components
7/7/2015
High-LET radiation
–
–
–
–
–
–
Energy deposition
Impact on DNA
Resistance to repair
The Ullrich experiment
Damage to chromosomes
Cataracts
Stochastic Damage; High-LET
p. 2 of 75
Is an Ames Test a Good Substitute for
These Complex Systems?
u
u
No!
1,8-dinitropyrene is the
most mutagenic substance
known in the Ames test;
yet it is only weakly
tumorigenic in rats.
7/7/2015
Stochastic Damage; High-LET
p. 3 of 89
Why might we care about
dinitropyrene?
Most mutagenic substance known in Salmonella strain
TA98: 72900 revertants/nanomole
Nitroarenes like this one were found to be present in
used toner, i.e., combustion waste from Xerox toner
When this appeared, Xerox chemists reformulated
their toner to drastically reduce the nitroarene content
in the used toner.
Mermelstein (1981) Mutation Research 89:187-196.
Löfroth et al(1980) Science 209:1037-1039 and
Mermelstein et al (1980) Science 209:1039-1043.
So: all’s well that ends well!
7/7/2015
Stochastic Damage; High-LET
p. 4 of 89
This is also a story about enzyme
induction
Nitroarenes like dinitropyrene and other polynuclear
aromatic hydrocarbons, (e.g. benzo(a)pyrene) are known
to be inducers of enzyme activities
Some of these enzyme activities actually activate toxicants
rather than detoxifying them
Most of the activity of these enzymes will detoxify;
But if 1% makes things worse, we want to understand that
1% activation
So we found that pretreatment with these compounds
could induce subsequent binding of other compounds to
mouse DNA:
Howard et al (1986), Biochem. Pharm. 35: 2129-2134.
7/7/2015
Stochastic Damage; High-LET
p. 5 of 89
Animal Cell-Line Cancer Studies
How similar are these rodent cell systems
(CHO, mouse) to human cells?
Answer: Human cells:
–
–
–
7/7/2015
Are more resistant to spontaneous
immortalization
Tend to give more nearly linear responses to
dose
Radical scavengers and cold don’t protect as
much:
That suggests that direct mechanisms prevail
in humans and indirect mechanisms are more
important in rodents
Stochastic Damage; High-LET
p. 6 of 89
More on humans vs. rodents
High-LET studies indicate that
repair is less effective in human cell
systems than in animal cell systems
Is the timescale a factor in that?
Humans live a lot longer than
rodents.
Promotion can be studied in animal
cells, along with initiation
7/7/2015
Stochastic Damage; High-LET
p. 7 of 89
Radiation Carcinogenesis
in Human Populations
Occupational: radiologists, miners, dial painters
Medical exposures:
–
–
–
–
–
Ankylosing spondylitis
Nonmalignant disease in pelvis and breast
Multiple fluoroscopies in to chest (e.g. in TB patients)
Infants & children with enlarged thymus and ringworm
Children exposed in utero to diagnostic X-rays
Nuclear accidents and weapon detonations
Environmental background (see last chapter)
7/7/2015
Stochastic Damage; High-LET
p. 8 of 89
Dose-Incidence in Cancer Studies
We seek a relationship relating post-exposure
incidence ID to dose D and normal incidence In
Model might be:
Linear: ID = In + 1D
Quadratic: ID = In + 2D2
LQ: ID = In + 1D + 2D2
Corrected for loss of clonogenic potential:
ID = (In + 1D + 2D2)exp(-1D+2D2)
7/7/2015
Stochastic Damage; High-LET
p. 9 of 89
Linear, Quadratic, LQ Models
We try to devise low-dose models based on high-dose
data, where the three models are close together. It’s
often difficult:
Incidence Models
100
Cancer Incidence/100000
90
80
70
60
50
Linear
Quadratic
40
LQ
30
20
10
Dose, Gy
0
0
7/7/2015
2
4
6
8
Stochastic Damage; High-LET
10
12
14
p. 10 of 89
Latency
Definition (in the cancer context):
Time between the mutational events that began
cellular transformation and the appearance of a
medically observable malignancy
How long in humans?
–
–
–
–
A few years (blood or lymphatic cancers)
15-30 years for solid tumors
Animals: scale these numbers to animal’s lifespan
These numbers are minima:
leukemia can take > 15 yrs
7/7/2015
Stochastic Damage; High-LET
p. 11 of 89
Macroscopic Damage to
Chromosomes
First half of chapter 13
Structural changes in chromosomes
–
–
–
–
–
–
Inversions of fragments
Multiple hits
Isochromatid breaks
Dicentrics
Minutes
Cross over
7/7/2015
Stochastic Damage; High-LET
p. 12 of 75
Generalizations about Genetics
Main focus here is on ways that ionizing
radiation can damage the physical and
replicative properties of DNA
We’ll look at three length scales:
–
–
–
The individual base-pair (~0.5 nm)
The gene (1000 bp - many nm)
The entire chromosome (200 nm or more)
Focusing only on single-base mutations
(substitutions, deletions, insertions) does not
tell the whole story
7/7/2015
Rad Bio: Genetics
p. 13 of 75
Early studies
T.H. Morgan studied
chromosomes in Drosophila
Tradescantia (spiderwort)
microspores: Sax, 1938-1950
More Tradescantia: Lea et al, 1946:
Provided quantitative data on
chromosomal aberrations during first
mitotic cycle
7/7/2015
Rad Bio: Genetics
p. 14 of 75
1960’s-1970’s:
chromosomes
Metaphase chromosomes
readily studied
Human lymphocytes
extensively studied that way
“Premature chromosome
condensation” used to study
interphase chromosomal
organization
7/7/2015
Courtesy HelmholtzZentrum
München
Rad Bio: Genetics
p. 15 of 75
Chromosome Breakage
Quite common result of radiation exposure
Can happen before replication:
then the structural defect will be replicated
(if the cell survives)
Can happen afterward:
then one of the two chromatids
will differ from the other one.
Types of damage:
Subchromatid, Chromatid, Chromosome
Repair tends to be rapid
7/7/2015
Rad Bio: Genetics
Wheat chromosomes
Courtesy USDA
p. 16 of 75
Single-Hit Breaks
Cartoons show * as centromere and ^ as break
Single-Hit:
A B C * D E F^ G H I 
A B C* D E F + GHI
This can recombine in a number of ways:
ABC*DEFIHG
IHGABC*DEF
GHIABC*DEF
… or a circle plus a fragment without a
centromere, which won’t be able to use the
spindle machinery to get sorted.
7/7/2015
Rad Bio: Genetics
p. 17 of 75
Double Hits
A B C * D^ E F^ G H I 
A B C*D + EF + GHI
Numerous ugly combinations can
arise, e.g.
E F A B C *D I H G
Acentrics can readily arise
7/7/2015
Rad Bio: Genetics
p. 18 of 75
Multiple Hits in Replicated
Chromosomes
See figs. 13.1 and 13.2
Major losses of genetic
information possible
Balanced
translocations: no
harm done!
Dicentrics--two
centromeres in one
pair of chromosomes.
7/7/2015
Rad Bio: Genetics
p. 19 of 75
SCE: another form of multiple-hit
damage
Sister chromatid exchange: exchange of
DNA fragments from one chromatid to
another between the two chromatids of a
single chromosome
–
–
Important for chemical mutagens
Not common with radiation
Cartoon courtesy
madsci.org
7/7/2015
Rad Bio: Genetics
p. 20 of 75
Double-stranded breaks
7/7/2015
Rad Bio: Genetics
p. 21 of 75
Recombination
+
7/7/2015
Rad Bio: Genetics
p. 22 of 75
Breakage and Exchange Hypotheses
Breakage hypothesis (Sax, 1940’s onward):
Types of aberrations that require one break
follow linear dose-response; those requiring
two or three follow quadratic or cubic doseresponse.
Exchange hypothesis (Revell, 1950’s onward):
emphasizes reciprocal exchanges among
chromosomal materials: rate  D1.7
Alpen says the data don’t support this one
much.
7/7/2015
Rad Bio: Genetics
p. 23 of 75
Using FISH to
sort this out
Fluorescence in situ
hybridization: available
before modern genetic
tools
Probe segments bind to
DNA regions of high
homology
Used to sort out
differences between
breakage hypothesis and
exchange hypothesis
7/7/2015
From Wikipedia
Rad Bio: Genetics
p. 24 of 75
Gene Mutations
We’ve discussed these in detail previously
Types:
–
–
–
Deletions (frameshift)
Additions (slightly less likely) (frameshift)
Substitutions (e.g. C for T; no frameshift)
Remember that three bases code for an amino acid!
If we skip a single base, we can throw off every single
amino acid that is coded for downstream of the error.
Mutation frequencies are often linear with dose
7/7/2015
Rad Bio: Genetics
p. 25 of 75
Effect of Frameshift
Suppose the DNA duplex looks like this:
5’- A - T - C - G - C - A - A - 3’ (template strand)
3’- T - A - G - C - G - T - T - 5’ (coding strand)
Now we delete one base, so that the coding strand is
3’- T - A - G - G - T - T - 5’
Now the amino acids coded for will be different.
7/7/2015
Rad Bio: Genetics
p. 26 of 75
Frameshifts (continued)
• Most common form of (local) DNA damage
• Defined as any form of DNA damage that leads
to a loss of one base during replication or
transcription.
• Chemistry
• Deletion of bases
• Fragmentation of sugar-phosphate
backbone
• Distortion of base-base hydrogen bonds
and other 3-D elements
7/7/2015
Rad Bio: Genetics
p. 27 of 75
Consequences of frameshifts
• Result in complete misreading of remainder of
message
• Obviously they’re less dangerous near the
end of a gene!
• Frameshift in an intron near the end of a gene
might mean the mRNA will be correct
7/7/2015
Rad Bio: Genetics
p. 28 of 75
Genes
A gene is a unit within a chromosome (i.e., a
stretch of DNA) that gets transcribed and thereby
codes for a specific function.
Most genes ultimately code for proteins; but
Some code for transfer RNAs, rRNA, sRNA
Eukaryotic genes often have exons and introns:
–
–
Exons are segments of a gene that are ultimately
turned into proteins
Introns are segments of a gene that are removed
from the message before translation
7/7/2015
Rad Bio: Genetics
p. 29 of 75
How Introns are Removed
Suppose we begin with a
2000-base-pair gene.
It will be transcribed to
make a 2000-base
messenger RNA molecule
That gets chopped up to
make up a ~960 base
ultimate message
This will produce a 320amino acid protein
(960/3 = 320)
2000 base-pairs of DNA
transcription
2000 bases of RNA
exon
exon exon
exon
exon
intron intron
intron intron
splicing
960 bases
translation
320-amino-acid
protein
7/7/2015
Rad Bio: Genetics
p. 30 of 75
Structural components of genes
Bases
–
–
–
–
Adenine
Cytosine
Guanine
Thymine
A
C
G
T
Phosphate-deoxyribose backbone
Double helix enables the replication and transcription
processes and offers physical stabilization
7/7/2015
Rad Bio: Genetics
p. 31 of 75
Review of DNA backbone
O
Sugar-phosphate backbone on
each strand has a nitrogenous
acid base attached at the 1’
position of the ribose ring
The lack of an -OH group on the
2’ position prevents formation of a
reactive intermediate; this makes
DNA more stable than RNA
DNA’s double-helical structure
also helps its stability.
7/7/2015
Rad Bio: Genetics
O
(Base)
P ON
OCH O
2
2’
O
O
P OO
Next sugar
p. 32 of 75
Differences between DNA and RNA
RNA has a hydroxyl at the 2’ position on the ribose
ring
RNA is made up of A, C, G, U
rather than dA, dC, dG, dT
DNA is replicated using a molecular machine called
DNA polymerase
DNA codes for RNA (tRNA, mRNA, rRNA, or sRNA)
using a molecular machine called RNA polymerase
DNA is more chemically stable than RNA
7/7/2015
Stochastic Damage; High-LET
p. 33 of 75
Types of RNA
Our focus on central dogma sometimes makes us
forget about RNA types other than mRNA
Type
# bases
% of
synthesis
% of DS?
RNA
Role in translation
mRNA
150-104
25
3
no
Protein template
tRNA
55-91
21
15
part
Amino acid activation
rRNA
100-104
50
80
part
Scaffolding, catalysis
sRNA
15-1000 4
2
hybrid
various
7/7/2015
Stochastic Damage; High-LET
p. 34 of 75
RNA hairpins
DNA is almost always double helical;
Watson-Crick base pairing contributes to stability
RNA sometimes has complementary bases that
allow for bases to pair up around a (hairpin) loop
5’-A-U-G-G-C-C-G-A-A-U-G-C-C-A-U-3’
5’-A-U-G-G-C-C-G
A
3’-U-A-C-C-G
UA
7/7/2015
Stochastic Damage; High-LET
p. 35 of 75
Special features
Code is redundant
43 = 64 codons
Only 20 amino acids plus a few control codes;
most amino acids have more than one codon
each associated with them
(minimum=1, maximum=6).
Middle base is generally conserved; all the
codons for any given amino acid generally
have the same middle base.
First RNA base may be sloppy
7/7/2015
Stochastic Damage; High-LET
p. 36 of 75
Genetic code: mRNA version
Determined by
hard work in
~1958-1965
Slight variations
in a few
organisms
7/7/2015
Stochastic Damage; High-LET
p. 37 of 75
Errors in Fidelity and Their
Consequences
How do chemicals & radiation affect
fidelity
(rate of mutation)?
Increase likelihood of replication error
Radiation: bond disruption in bases
(or sugars)
–
–
7/7/2015
Direct (radiation breaks bonds in DNA)
Indirect (radiation causes free radicals;
free radicals break bonds in DNA)
Stochastic Damage; High-LET
p. 38 of 75
Chromosomal Instability
The observation is that cell death continues for
several cellular generations after the initial
exposure
The explanation is that the chromosome becomes
unstable, i.e. it loses some of its ability to remain
structurally sound.
This instability is heritable and therefore accounts
for some of these long time-scale effects
Little (1990): decreases in cloning efficiency lasts
for 20-30 population doubling times in mammalian
cells
7/7/2015
Stochastic Damage; High-LET
p. 39 of 75
Results in intact eukaryotes
Not as easy to study as in culture
Most work historically has been done in Drosophila
–
–
–
–
Rate of production of lethal mutations is linear with
dose
Dose rate is irrelevant
Male gamete’s sensitivity varies greatly with the stage
of development at which irradiation occurs
Rate of mutations ~
1.5*10-8 - 8*10-8 per locus per cGy.
n.b.: a locus is a testable genetic element; it’s
typically one gene, although it could be a group of
genes controlled by a single promoter
7/7/2015
Stochastic Damage; High-LET
p. 40 of 75
Results in Mammals
Concern was raised in the 1960’s that we
shouldn’t base mammalian exposure risks on
Drosophila data
Mouse specific locus test: suggests
–
–
2.2*10-7 mutations/locus/cGy (~5x higher).
Dose-rate matters: fewer observed mutations
at lower dose rate.
Difficult to extrapolate these mouse results to
humans
Rates around1 - 5*10-7 is typical.
7/7/2015
Stochastic Damage; High-LET
p. 41 of 75
Repair
Cell has mechanisms to recognize
& replace faulty bases before they
have a chance to be replicated
Some injury may disrupt a large
enough segment of DNA that repair
either fails or is error-prone
7/7/2015
Stochastic Damage; High-LET
p. 42 of 75
High LET Radiation
Review definition of LET:
Rate of change of energy with
distance along a track
Caveat I: Local Deposition, i.e.,
depositions far from track don’t
count
Caveat II: LET only associated
with charged particles
7/7/2015
Stochastic Damage; High-LET
p. 43 of 75
LET Equation: Bethe-Block
Formulation
If we let:
–
–
–
–
–
–
–
z* = Effective charge of projectile particle
Z = Atomic number of atoms in medium
A = Atomic weight of atoms in medium
C = sum of electron shell corrections
d/2 = condensed medium correction
 = v/c for particle
I = <ionization potential for absorbing medium>
-dE/dx = [0.307z*2Z/(A2)]•
{[ln((2m0c22)/(1-2)I)] - 2 - C/Z - d/2}
7/7/2015
Stochastic Damage; High-LET
p. 44 of 75
What does this mean?
It’s a messy equation, but we can derive several
pieces of wisdom from it.
-dE/dx is inversely proportional to 2, so slow particles
dump more energy per unit length
-dE/dx is proportional to z*2, so highly charged particles
deposit more
Chemistry of target doesn’t matter too much, because
Z/A is almost constant
The term inside the curly brackets is a correction
7/7/2015
Stochastic Damage; High-LET
p. 45 of 75
Where is Energy Deposited?
Most of the energy is
deposited just before the
particle stops moving
We’re interested in average
LET over a region or track
7/7/2015
Stochastic Damage; High-LET
p. 46 of 75
Relative ionization
Depositions with heavy particles
4
“Bragg peak”: most of the energy is
deposited over a narrow linear track.
W.H. Bragg (father of W.L. Bragg)
characterized this
425 MeVa neutrons
in water
1
0
7/7/2015
Range, cm
10
Stochastic Damage; High-LET
p. 47 of 75
Calculating Averages of Continuous
Variables
We wish to calculate the mean of a continuous function f(x) over a
range of x from a to b,
Then <f(x)>a,b = [∫ab f(x)dx ] / (b-a)
f(x)
f(x)
<f(x)> =
(f(a)+f(b))/2
Area under curve =
∫ab f(x)dx
a
7/7/2015
b
x
a
(a+b)/2
Stochastic Damage; High-LET
b
p. 48 of 75
Applying this to LET
Average LET over energies:
(LETav)S = { ∫E1E2 [LET(E)]dE } / (E2-E1)
So for a slowly varying LET function we
assume that it is close to linear, i.e.
(LETav)S = [(LET)1 + (LET)2] / 2
These are instances of track segment
averages, (LETav)TS;
they’re somewhat different from energy
averages, for which we average over energy
deposited.
7/7/2015
Stochastic Damage; High-LET
p. 49 of 75
Alternative averages
Energy-averaged LET
(see previous slide; energy range from E1 to E2):
<LET>E = { ∫E1E2 LET(E) dE } / (E2-E1)
where E is the energy that we’re averaging over.
Track-averaged LET
(track from position a to position b):
<LET>TS = { ∫ab LET(x)dx } / (b-a)
where x is the linear dimension through which the
energy is being deposited
a
b
7/7/2015
Stochastic Damage; High-LET
p. 50 of 75
Average LET, keV µm-1
Table 14.1:
Radiation type
(LETav)TS
(LETav)E
60Co g-rays
0.27
19.6
250 kVp x-rays
2.6
25.8
3 MeV neutrons
31
44
Radon  rays
118
83
14 MeV neutrons
11.8
125
Recoil protons
8.5
25
Heavy recoils
142
362
•Looks like the (LETav)TS is more in accord with our
intuitive notion of what constitutes LET!
7/7/2015
Stochastic Damage; High-LET
p. 51 of 75
Direct & Indirect Effects
High LET means that many ionizations occur in
a small neighborhood
LET
KeV/µm
60Co g
0.25
Radon  118
Spur energy Events/µm Spacing
eV
nm
60
4
250
60
2000
0.5
This fact by itself accounts for much of the
difference in biological consequence of high LET
radiation
7/7/2015
Stochastic Damage; High-LET
p. 52 of 75
Biological Effects of High-LET
Radiation
RBE = relative biological effectiveness
analogous to OER in its definitional form:
RBE = (dose for given end point for reference radiation) /
(dose for given end point for test radiation)
Problem with this formulation:
–
–
assumes that dose-response curves are described
by identical functions, i.e.
RBE is independent of the response level at which
it is estimated. Often not true!
7/7/2015
Stochastic Damage; High-LET
p. 53 of 75
MTSH comparisons
Cf. fig. 14.3: Xrays and fast neutrons
Fig.14.3: RBE
0
0
2
4
6
8
10
12
14
16
-2
Results
from
hamster
fibroblasts
log(S)
-4
-6
-8
-10
-12
Dose, Gy
7/7/2015
Stochastic Damage; High-LET
p. 54 of 75
Can we define a single RBE?
Not really, unless our survival curves are
really just rescalings of one another
(“dose-modifying effect”)
For MTSH data RBE will be close to
constant if the extrapolation numbers
aren’t too high
For LQ we’re definitely going to have to
pick a survival ratio
7/7/2015
Stochastic Damage; High-LET
p. 55 of 75
Survival Level Matters!
RBE vs Survival Fraction
8
7
6
RBE
5
4
3
2
1
0
-4
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
log(S)
7/7/2015
Stochastic Damage; High-LET
p. 56 of 75
LQ case
RBE varies significantly between S=0.1 and S=0.001:
Dose, Gy
1
0
2
4
6
8
10
12
14
16
Survival Fraction
0.1
LQ, Low LET
LQ, high LET
0.01
0.001
0.0001
LQ models: low and high LET
0.00001
7/7/2015
Stochastic Damage; High-LET
p. 57 of 75
Relationship between RBE and LET
RBE vs LET:
Generally higher LET radiations have higher RBE, up to a certain
point.
With some systems the curve turns over.
5
4
Two hamster cell
lines (from
fig.14.5)
3
2
1
LET, keV/µm
100
7/7/2015
101
102
103
Stochastic Damage; High-LET
p. 58 of 75
Cell-Cycle Dependence for RBE
Low LET radiations exert much more effect
in late G2 & M than elsewhere
High LET radiation exerts approximately
equal effects throughout the cell cycle
Reason: irreparable damage early in the
cycle will persist through mitosis, so it’s
just as bad!
7/7/2015
Stochastic Damage; High-LET
p. 59 of 75
OER vs LET
OER for High-LET Radiation is generally
smaller than for low-LET radiation because the
damage is less dependent on oxygen fixation of
radical species.
This plot shows OER going not to 1 but to ~1.5-because water-derived radicals are still
produced at high LET; cf. Fig. 14.6.
2
1.5
1
1
7/7/2015
10
100
Stochastic Damage; High-LET
LET, keV/µm
1000
p. 60 of 75
LET vs Fractionation
Recall that repair-competent systems
respond less to fractionated doses
than to single doses whereas repairdeficient systems are fractionationindependent
For high-LET radiation, fractionation
does not decrease biological effects
7/7/2015
Stochastic Damage; High-LET
p. 61 of 75
Contrarian results for fractionation
Sometimes in tumors fractionation increases the
undesirable biological effect, i.e. more healthy cells die
per 100 tumor cells killed!
– Why? - no explanation in Alpen
– 2-stage model for cancer: (Ullrich)
– Also: high dose rate (equivalent to fewer fractions)
is more likely to simply kill the cell rather than
producing clonogenically competent but mutated
cells
(remember the concept of correcting for mortality in
epidemiology)
7/7/2015
Stochastic Damage; High-LET
p. 62 of 75
Late Effects of High LET Radiation
You can induce cancer with neutrons
Ignores fractionation or (!) damage is
more likely with fractionation
Tumors also arise from heavy-ion
irradiation
7/7/2015
Stochastic Damage; High-LET
p. 63 of 75
LET vs. Cross Section
Cross Section, µm2
Alpen looked at particle fluence as a dose parameter-requires constant LET over volume
Result: power-law relationship between cross section for
tumor induction and LET (fig.14.7)
10
.1
0.001
0.1
7/7/2015
1
10
100
LET, keV/µm
Stochastic Damage; High-LET
p. 64 of 75
General note re linearity
Ordinary linear relationship:
–
–
–
Both axes on linear scales.
y = ax + b,
a = slope, b = Y intercept
y
b
Semi-log linear relationship:
–
–
–
–
–
X linear scale, Y logarithmic ln y
ln y = ax + b
exp(ln y) = exp(ax+b)
y = exp(b) exp(ax) = keax
b
for k = exp(b)
7/7/2015
Stochastic Damage; High-LET
x
x
p. 65 of 75
Third possibility: power law
Here we have linearity
on a log-log plot:
ln y
ln(y) = alnx + b
exp(ln(y)) = exp(alnx)exp(b)
y = k exp(a ln x) for k=exp(b)
But aln x = ln(xa), so
b
y = kexp(ln(xa)) = k xa
So this is a power law.
7/7/2015
Stochastic Damage; High-LET
ln x
p. 66 of 75
Probability of Traversals
Table 14.2 shows that multiple traversals of the nucleus are
very rare even at high doses
It also shows a close correspondence between cross section
for tumor production and cross section for traversal
Dose,
Gy
0.01
0.02
0.05
0.15
0.20
0.30
0.40
<fluence/
cell>
0.006
0.013
0.032
0.097
0.129
0.193
0.258
7/7/2015
Probability of traversals
0
1
2
0.993 0.006 2.1*10-5
0.987 0.013 6.8*10-5
0.968 0.031 5.0*10-4
0.907 0.088 4.0*10-3
0.878 0.114 7.0*10-3
0.823 0.159 1.5*10-2
0.772 0.199 2.5*10-2
>= 1
0.006
0.013
0.032
0.092
0.121
0.176
0.227
Stochastic Damage; High-LET
p. 67 of 75
Number of Traversals
0.3
Fluence & Traversals
0.25
0.2
0.15
<fluence/cell>
1 traversal
0.1
>= 1 trav.
0.05
0
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Dose, Gy
7/7/2015
Stochastic Damage; High-LET
p. 68 of 75
Cell Transformation in High-LET
Irradiation
Assertion: non-local effects on DNA predominate over point
mutation events
Certain assay systems suggest this assertion:
Kronenberg et al, human TK6 lymphoblasts-entire hprt gene is missing after irradiation
This gene codes for hypoxanthine phosphoribosyl transferase,
an important enzyme in DNA synthesis
“If large genomic changes are brought about by heavy ion
radiation, then a multistep process may be short- circuited to a
single event.”
7/7/2015
Stochastic Damage; High-LET
p. 69 of 75
Cataracts
The vertebrate eye is a remarkable
organ
All the living components must be
completely transparent for maximum
fidelity of light transmission.
Even the lens is made up of living
cells!
A loss of transparency is a cataract,
either evanescent or permanent.
A variety of insults can give rise to a
loss of transparency.
7/7/2015
Stochastic Damage; High-LET
p. 70 of 75
Cataracts
Cataracts are common with high-LET radiation
Reminder: cataracts involve loss of transparency in the
lens because problems with differentiation will lead to
failure of alignment of the fibers
Cataracts happened with workers in early accelerator
facilities
RBE values are high, and tend to be higher for lower
doses (i.e. low doses cause almost as many cataracts
as higher doses)
7/7/2015
Stochastic Damage; High-LET
p. 71 of 75
Cataracts from high-LET radiation
It’s long been known that high-LET
radiation can give rise to cataracts, even at
low doses.
Relative biological effectiveness of highLET radiation (neutrons) is 30 or more for
low doses
RBE goes down for higher doses
Beams of ions (Z > 20) produce similar
effects, again including lowered RBE for
higher doses.
7/7/2015
Stochastic Damage; High-LET
p. 72 of 75
Cataracts from specific types of
radiation
Neutrons:
–
–
In animals: cause cataracts with RBE~2 to 100
But with humans: cataracts become rare up to 2 Gy,
almost universal at D > 11 Gy.
Argon and iron ions: RBE ~12-40 for low doses (below
0.25 Gy), more like 2-5 for higher doses
Why does the RBE depend on dose?
–
–
Can’t kill a cell more than once?
Mechanistic explanations (see Ullrich discussion)?
7/7/2015
Stochastic Damage; High-LET
p. 73 of 75
The Ullrich Tumor Experiments
Ullrich found that with low doses of neutrons,
fractionation never diminished incidence of
tumors
With certain types (lung, mammary) there was
an enhancement in tumor rate with fractionation
Dose-response was nonlinear
Saturation and fall-off of incidence with high
doses
7/7/2015
Stochastic Damage; High-LET
p. 74 of 75
Ullrich Experiment: Why?
Mechanisms of Genetic Damage
Low vs High LET
–
–
Low-LET radiation exerts many of its effects at the level
of point mutations
(single-base substitutions, deletions, or additions)
High-LET exerts most of its effects on a more
macroscopic scale
7/7/2015
Stochastic Damage; High-LET
p. 75 of 75