Memorandum submitted by Mr J Urquhart
RADIATION DOSE
AND GENETIC
RISK
Abstract
The current radiation paradigm for genetic risk
is not secure. Present estimates for the purposes of radiation
protection are derived mainly from studies of spontaneous mutation
rates of human genes, and radiation-induced mutation rates of
mouse genes. This has led to the conclusion that genetic risk
is considerably lower than that for cancer. An examination of
research into stillbirth rates in the offspring of male Sellafield
radiation workers showed a dose response relationship between
stillbirth rates and preconceptional paternal irradiation (PPI).
The Excess Relative Risk (ERR) for stillbirths was estimated to
be 21.5 per Sievert. For stillbirths with congenital anomaly,
the estimated ERR was 35.5 per Sievert. There is some evidence
that genetic effects via preconceptional maternal irradiation
(PMI) may be considerably higher. Belarus studies after the Chernobyl
nuclear accident indicated high ERR for congenital anomalies:
40 per Sievert due to PPI, and 400 per Sievert for preconceptional
maternal irradiation (PMI). Animal studies provide evidence of
persistent transgenerational effects, possibly involving genomic
instability after the initial radiation dose from Chernobyl fallout.
Saturation dose effects were observed, even at low levels of radiation
exposure. If the animal model is applied to human populations,
this suggests that genetic effects may persist for more than 600 years
after initial radiation exposure of the first generation. The
existing radiobiology paradigm, which is fundamental to the estimation
of environmental radiation risk, cannot explain the phenomena
of radiation-induced genomic instability and the bystander effect.
Current radiation protection standards for genetic risk appear
to be optimistic. This has important implications for the use
of nuclear power.
1. CURRENT RADIATION
STANDARDS
Ionising radiation is relatively easy to quantify
and our knowledge about its effects have accumulated over 100 years.
An extensive literature on radiation health effects has been based
mainly on analysis of the health consequences of Japanese A-bomb
survivors who were exposed in 1945, uranium miners and hospital
patients. This provides a considerable weight of evidence to support
current international radiation standards. These standards have
not been explored in detail in this review because the models
they employ are well known and reflect established scientific
thinking. A recent paper by Mobbs et al (2009) provides a useful
short summary of the current approach of the Health Protection
Agency (HPA) in the UK, which also reflects international standards
recommended by the International Commission on Radiation Protection
(ICRP).
— Increase in fatal cancers is 5% per Sievert
(Sv). A Sievert is a unit of radiation exposure to an individual;
the estimated annual dose to individual in the UK in the UK from
all radiation sources is 2.4 milliSieverts and the average
chance of an individual getting cancer in a lifetime is about
35%.
— The HPA accepts that the cancer risk
estimate could vary by a factor of three either way for external
radiation and some radionuclides.
— The HPA suggest that there are a further
group of radionuclides which the true effect could range by a
factor of 10 either way.
Mobbs et al (2009) do not point out that
at certain periods in human development the individual may be
much more sensitive to radiation, for example the excess relative
risk (ERR) for in utero exposure was estimated from the
Oxford Survey of Childhood Cancer (OSCC) to be 51 per Sv
(Doll and Wakeford 1997). However, a later paper (Little
and Wakeford 2003), suggested that figure might be an over estimate
and the true ERR might be one order of magnitude lower.
2. GENETIC RISK
The HPA paper (Mobbs et al 2009) considers
cancer risks from low level radiation but dismisses possible genetic
risks in the following terms:
"As well as the possibility of causing
cancer in the exposed individual, it is biologically feasible
that mutations to genetic material could be passed on to future
generations (this is called a heritable effect). However there
is no direct evidence of radiation-induced heritable effects in
humans (ICRP 2007) and this genetic risk is judged to be considerable
lower than that of cancer."
UNSCEAR (2001) estimates that the excess
relative risk is only 0.4-0.6% per Gray compared with the ERR
for cancer of 5% per gray but this calculation is not based on
epidemiological studies and does not equate with recent scientific
results.
Parker et al (1999) published a key paper
in The Lancet on stillbirths among offspring of male radiation
workers at the Sellafield nuclear reprocessing plant. They found
a significant positive association between the risk of a baby
being stillborn and the fathers' exposure to external ionising
radiation before conception. The adjusted odds ratio was 1.24 per
100mSv (95% CL 1.04-1.05, p=0.009). The risk was higher for stillbirths
with congenital anomaly and they estimated that 30 of the
130 stillbirths to the workforce may have been attributable
to the fathers radiation exposure. These findings were disputed
by Doyle et al (2000), who reported that male radiation workers
in the nuclear industry had not observed an increase in stillbirths
or miscarriages in their partners over the expected. On the other
hand, female radiation workers reported stillbirth rates that
were more than double the expected, and an increase in miscarriages
over the expected rate.
The Parker et al (1999) study was far
more comprehensive and based on actual records as opposed to male
recall of their partners' pregnancy condition. The study reported
an ERR of 0.24 for 100mSv, based on total dose before conception.
It is likely only a proportion of the total dose was involved
in producing adverse effects in the offspring. The study reports
an ERR of 0.86 per 10mSv during the three month period before
conception. This originally assumed that the effective PPI dose
could be confined to this three month period, but it was discovered
that higher stillbirth rates were associated with fathers who
had been irradiated up to twelve months before conception. However,
it was subsequently discovered that some fathers who produced
stillbirths had received no recorded radiation in that three-month
period, but had received high doses within twelve months of conception.
At this stage, the exact mechanism of how the sperm is affected
by ionising radiation is not known. For example, it may be via
damage to the spermatozoa, which would extend the period of vulnerability
before conception. A more conservative estimate for ERR assumes
0.86 per 40mSv—equivalent to 21.5 per Sievert.
The odds ratio for stillbirths with congenital
anomaly was higher than that just for stillbirths: 1.46 vs.
1.26, suggesting a 65% increase in ERR equivalent to 35.5 per
Sievert.
3. FEMALE VS
MALE GENETIC
RISK
Preconceptional maternal radiation (PMI) in
women may be a more potent factor than paternal preconceptional
irradiation (PPI). The higher level of stillbirth and miscarriage
rates reported by female radiation workers in Doyle et al (2000)
is indicative of this possibility. Draper et al (1997)
carried out a study of over 100,000 male and over 10,000 female
radiation workers in the UK. They found that in the offspring
of the male radiation workers, the odds ratio for leukaemia was
1.83, but they did not find an increase in other cancers. On the
other hand, for female workers, they found four cases of leukaemia
against an expected of one, but eleven cases of other cancers
compared with an expected of two. It can be concluded that the
relative increase for all cancers for male radiation workers was
3.83 vs. 3.0 but for female workers was 15.0 vs.
3.0; a relative increase of 14.5 for children born to female
radiation workers. Some of this increase may have been due to
in utero exposure, but nevertheless genetic risk may have
been a key factor.
4. PRECONCEPTIONAL
MATERNAL IRRADIATION
(PMI)
As early as the 1970's, Alberman (1972)
reported an increase in Down syndrome in children born to mothers
who had been exposed to X-rays, but only after a delay of at least
six years. This little-noticed finding may now have important
implications.
5. BIRTH DEFECTS
IN BELARUS
AFTER CHERNOBYL
Moller (2005) has drawn attention to
the fact that the disaster at the Chernobyl nuclear power plant
in April 1986 released 80 petabecquerele of radioactive
caesium, strontium, plutonium and other radioactive isotopes into
the atmosphere. A becquerele is one radioactive disintegration
per second; a petabecquerele is one thousand million million becquereles.
Six months after the accident the radioactive rate in more contaminated
areas had fallen by a factor of ten (Ryabokon 2006). However research
into the biological and genetic consequences has been carried
out by only a few individuals rather than a concerted research
effort by the international community despite the fact that the
effects of the disaster were continent wide. Hoffman (2002)
has pointed out that the radiological consequences in Europe are
greater than those from all 420 above ground nuclear tests,
but the current radiation risk dose models suggest the additional
number of cancers would be very small.
There is however no explanation within the existing
radiation paradigm for the significant increases in birth defects
that have occurred in Belarus since the nuclear accident. Lazjuk
(2003) reported at least an 80% rise in 10 important congenital
anomalies after Chernobyl in Belarus. Using the estimated dose
to the population of 2mSv per individual, this would suggest an
excess relative risk for birth defects of 400 per Sv. This
is much higher than indicated by the Sellafield study. A closer
view of the congenital malformation time trend in Belarus is shown
in Figure 5.1.
Figure 5.1
BELARUS CHILDHOOD HEREDITARY DISEASES PER
100 BIRTH 1981-2004 (BELARUSSION 2006)
It can be seen that a peak occurred in the most
contaminated areas for 1987/88/89. After that, rates throughout
Belarus were at pre-Chernobyl levels, but in 1993 started
to rise again in all areas, so that by the year 2004 the
national rate was two and half times the pre-Chernobyl level.
It should be noted that after the accident there was considerable
migration out of contaminated areas by prospective mothers.
The return to pre-Chernobyl levels and then
the subsequent rise in defects may reflect a delayed response
to PMI. The initial peak in contaminated areas in 1987/88/89 may
correspond to PPI effects. The relative contribution of this peak
is estimated to be only 10% of the total number of extra birth
defects observed, suggesting that the ERR from PPI in Belarus
after the Chernobyl accident is only 40 per Sievert.
A report by the Chernobyl Forum (2005)
has played down the Belarus results, suggesting that only a modest
rise in birth defects has occurred since the Chernobyl accident—even
though this is more than double—and that any such rise could
be attributable to increased registration. The report claimed
that no observed increases were supported by the existing radiation
paradigm (UNSCEAR 2001). It has been pointed out earlier that
the radiation paradigm for genetic risk has no data on radiation-induced
mutations in humans. The Chernobyl Forum (2005) rejected the human
data on the grounds that it lies outside the current radiation
paradigm.
6. ANIMAL STUDIES
AFTER CHERNOBYL
Animal studies in the area near Chernobyl do
not support the claim that increases in human birth defect rates
after the accident were an artefact of social, psychological,
or registration influences.
6.1 Barn swallow studies
Moller et al. (2007) carried out an examination
of barn swallows from Chernobyl. A long-term study demonstrated
the presence of eleven morphological abnormalities in populations
around Chernobyl, but much less frequently in uncontaminated Ukrainian
control populations, and three more distant control groups. An
earlier paper had shown a long-term reduction in fertility in
the Chernobyl sample: 23% of adults were infertile, compared with
virtually 0% in the control groups.
6.2 Belarus bank vole studies
Ryabokon (2006) examined the possible
transgenerational effects of radiation fallout from the Chernobyl
nuclear accident on colonies of bank voles in Belarus. Murine
rodents are an important proxy for establishing genetic and health
paradigms in human populations. The study was carried out over
ten years, tracing health and genetic effects in 22 generations
of bank voles in five separate sites in Belarus, exposed to different
levels of radiation.
The authors of the study found that in spite
of exponentially decreasing external irradiation, the proportion
of chromosome aberrations in successive generations did not decline,
and in fact the percentage of foetal deaths rose, particularly
towards the end of the study period. The results from site three
with medium exposure are depicted in Figure 6.1.
Figure 6.1
COMPARISON OF EXTERNAL RADIATION EXPOSURE*,
CHROMOSOME DAMAGE AND % FOETAL DEATHS OVER 22 GENERATIONS OF A
BANK VOLE COLONY IN BELARUS (RYABOKON 2006)
It can be seen that estimated external dose
declined steeply over the period of study, but the percentage
of voles with chromosome damage remained fairly constant, while
the percentage of foetal deaths rose markedly.
Table 6.1
BELARUS BANK VOLES—PRE CHERNOBYL TO
1996—BACKGROUND DOSE, CHROMOSOME DAMAGE AND FOETAL DEATHS,
22 GENERATIONS OVER 10 YEARS (SITES 1, 2, 3 AND
4) (RYABOKON 2006)
|
| | % Chromosome Cell Aberrations
| % Foetal Deaths
|
| Site | Initial dose
µGy/day
| Pre-
Chernobyl
| 1986 | 1988-89
| 1991 | 1996
| Pre
Chernobyl
| 1986 | 1988-89
| 1991 | 1996
|
|
| 1 | 6 | 0.4
| 0.4 | 0.4
| 0.4 | 0.4
| - | -
| - | -
| - |
| 2 | 70 | 0.4
| - | -
| 1.2 | 1.2
| 7.2 | -
| - | 5.4
| 11.4 |
| 3 | 875 |
0.4 | 2
| 1.5 | 2.5
| 2 | 7.2
| - | 6.5
| 9.9 | 26.7*
|
| 4 | 6,055 |
0.4 | 1.2
| 1.9 | 1.9
| 2.1 | 7.2
| - | 4.8
| 15.5 | 21.6**
|
|
* significant at p=0.05; ** significant at p=0.01
Table 6.1 compares results for sites one, two, three
and four. This table compares chromosome cell aberration levels
and percentages of foetal deaths in four sites of widely differing
radiation background from the pre-Chernobyl period through to
1996. Site one had an estimated initial dose of six µGy per
day, equivalent to normal background radiation. On site two, the
initial radiation levels were about 12 times background.
Site three had an initial estimated external dose 145 times
background. Site four was 1,000 times background. Nevertheless,
site four had only 60% more chromosome aberrations than site two,
and only double the percentage of foetal deaths, even though the
initial radiation level was nearly 100 times greater. This
suggests a saturation dose effect. In other words once the initial
external dose rate exceeded a value between 875-6,055 micro
Grays per day, there was very little increase in biological and
genetic effects.
Not shown here are the results for a sample of bank voles
bred in the laboratory. Even though the radiation levels were
much lower here succeeding generations continued to show adverse
chromosome and health effects.
The authors concluded that their results were evidence for
the transference of genomic instability through successive generations.
7. GENOMIC INSTABILITY
A new phenomenon discovered within the last 20 years
is that when cells are subjected to ionising radiation and exhibit
a cell response, this cell response can also be found in neighbouring,
non-targeted cells. This is known as the bystander effect. This
cell response can then be propagated into new generations of cells
which is known as genomic instability (Kadheim 2007; Mothersill
2001; Mothersill 2006; Coates et al 2004).
If the cells are germ line cells then genomic instability
can be transferred from one generation of individuals to the next.
The transfer is non-mendalian, in other words, the next generation
does not inherit the exact mutation of the previous generation,
but instead the instability of the inherited genome in itself
creates new and unpredictable mutations in the second generation.
This genomic instability effect is transgenerational and can last
through many generations
Baverstock (2000) has pointed out that "The existing
paradigm governing radiobiology, which is fundamental to the estimation
of environmental radiation risk cannot explain the phenomena of
radiation induced genomic instability and the bystander effect.
Animal studies undertaken after the Chernobyl accident show evidence
of long-term transgenerational effects which may apply to human
populations. So far, the evidence from Sellafield and Belarus
is only available for first-generation effects. It is not known
to what extent even these are related to bystander effects and
genomic instability.
Two important aspects of genomic instability, which to not
appear to be addressed in the current radiation paradigm for genetic
risk are: the long term effects in human populations from an initial
dose to the first generation—the bank vole study suggests
that in human terms, this might last at least 600 years;
and whether the original genetic response to radiation follows
a linear dose relationship at low levels of radiation.
8. CONCLUSIONS
The following conclusions are drawn from this report:
— The current radiation paradigm for genetic risk is
based on spontaneous mutation rates in human genes and radiation-induced
mutation rates of mouse genes, and not on human or animal epidemiological
studies.
— The current radiation paradigm for genetic risk estimates
an excess relative risk (ERR) of 0.4%-0.6% per Gray(Sievert) compared
with an ERR for fatal cancer of 5%.
— Epidemiological studies of stillbirths on offspring
of male Sellafield radiation workers suggest an ERR of 35.5 per
Sievert for stillbirths with congenital anomaly.
— There is some evidence that genetic risk for women
may be ten times greater than for that of men. The ERR for congenital
malformations in Belarus after the Chernobyl accident are estimated
to be 40 per Sievert for men and 400 for women.
— Animal studies in areas near the Chernobyl accident
show increased levels of chromosome abnormalities and foetal deaths
even after 22 generations. This has been attributed to transgenerational
genomic instability effects.
— In human terms these animal results implies effects
for at least 600 years.
Finally, it should be noted that radiation protection limits
for somatic (cancer) risk from ionising radiation have been lowered
five times, but little attention has been given to genetic risks,
particularly associated with epidemiological studies. The new
epidemiological data referred to in this paper for both animals
and humans suggests the current radiation paradigm is not secure
enough to provide a stable basis for decision making purposes.
REFERENCES
Alberman E 1972 Parental exposure to X-irradiation and
Downs Syndrome. Ann Hum Genet 36(2): 195-208.
Baverstock K 2000 Radiation-induced genomic instability:
a paradigm-breaking phenomenon and its relevance to environmentally
induced cancer. Mutation Research 454: 89-109.
Coates P et al 2004 Induced genomic instability: An epigenetic
consequence of radiation exposure. Mutation research V568 5-20.
Doll R and Wakeford R 1997 Risk of childhood cancer from
foetal irradiation. British Journal of Radiology 70: 130-139.
Doyle P et al. 2000 Fetal death and congenital
malformation in babies born to nuclear industry employees: report
from the nuclear industry family study. Lancet 356 (9238):
1293-1299.
Draper GJ et al. 1997 Cancer in the offspring of
radiation workers—a recorded linkage study. National
Radiological Protection Board NRPB-R298.
Dubrova YE et al. 2002 Nuclear weapons tests and
human germline mutation rate. Science 295(5557): 1037-1037
Hoffmann W 2002 Has fallout from the Chernobyl accident
caused childhood leukaemia in Europe? A commentary on the
epidemiologic evidence. Eur J Public Health, 12, 72-6.
Kadhim M 2007 Environmental causes of genomic instability.
Mutagenesis 2434-434.
Lazyuk G et al. 2003 The congenital anomalies registry
in Belarus: a tool for assessing the public health impact of the
Chernobyl accident. Reproductive Toxicology 17(6): 659-666.
Mobbs et al. 2009 An introduction to the estimation
of risks arising from exposure to low doses of ionising radiation.
Health Protection Agency HPA-RPD-055.
Moller AP et al. 2007 Elevated frequency of abnormalities
in barn swallows from Chernobyl. Biol Lett 3: 414-417.
Moller AP et al. 2005 Condition, reproduction and
survival of barn swallows from Chernobyl. Journal of Animal
Biology 74(6): 1102-1111.
Mothersill C 2001 Radiation induced bystander effects:
past history and future directions. Radiation Research 155,
759-767.
Mothersill C 2006 Radiation-induced genomic instability
is associated with DNA ethylation changes in cultured human keratinocytes.
Mutation Research-Fundamental and Molecular Mechanisms of Mutagenesis,
597(1-2):87-97.
Parker L et al. (1999) Stillbirths among the offspring
of male radiation workers at Sellafield nuclear reprocessing plant.
Lancet 3054: 1407-1414.
Pearce MS et al. 2002 Stillbirths among the offspring
of male radiation workers at the Sellafield nuclear reprocessing
plant: detailed results and statistical aspects. J R Statist
Soc A 165(3): 523-548.
Ryabokon NI and Goncharova RI 2006 Transgenerational accumulation
of radiation damage in small mammals chronically exposed to Chernobyl
fallout. Radiat Environ Biophys 45: 167-177.
The Chernobyl Forum 2005. Chernobyl's Legacy:Health, environmental
and Socio-Economic Impacts and Recommendations to the Governments
of Belarus, the Russian Federation and Ukraine (http://www.iaea.org/Publications/Booklets/Chernobyl/chernobyl.pdf)
UNSCEAR (United Nations Scientific Committee on the Effects of
Atomic Radiation) 2001 Hereditary Effects of Radiation
(http://www.unscear.org/unscear/en/publications/2001.html)
January 2010
| 
