Twenty years’ experience with post-Chernobyl
Dillwyn Williams*MD, F Med Sci
Strangeways Research Laboratory, Worts Causeway, Cambridge CB1 8RN, UK
Chernobyl, the largest ever nuclear accident, caused a huge release of radioactive isotopes, in-
cluding nearly 2 ? 1018Bq of iodine-131. Four years later an increase in thyroid cancer inci-
dence, virtually all papillary carcinomas in children, occurred in the highly exposed areas. The
increase has continued, and with increasing latency the tumour molecular and morphological pa-
thology has changed; further changes may occur in the future. Children under the age of 1 at
exposure show the highest susceptibility, and carry this risk with them into adult life; 4000 cases
have been attributed to the accident, but so far very few have died. The risk falls rapidly with
increasing age at exposure; it is doubtful if there is any risk for adults at exposure. Other factors
linked to susceptibility to thyroid carcinogenesis after Chernobyl include dose, iodine deficiency,
and genetic factors. Other consequences are briefly covered.
Key words: thyroid cancer; radiation; Chernobyl; latency; genotype–phenotype correlation.
Itisoverahundredyearssince radioactivitywas discovered,and over60 yearssincethe
energy contained within the atom was recognized and the techniques to access that en-
ergy developed. Two eventssincethen havehad a major impacton the public attitude to
the use of atomic energy: the atomic bombing of Japan, with the devastation of the cities
bombing, the atomic bombs led to death and serious injury; unlike any previous events
they led to lifelong effects on the health of the exposed population due to radiation.
These effects have been studied in detail; they still are being studied, and have provided
most of our current information about the effect of radiation on human health.
The type of radiation exposure after Chernobyl was quite different from the expo-
sures after the atomic bombs. In Hiroshima and Nagasaki well over 100,000 people
were exposed to external whole-body radiation from gamma rays and neutrons.
* Tel.: þ44 1223 740180; Fax: þ44 1223 741147.
E-mail address: email@example.com
1521-690X/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved.
Best Practice & Research Clinical Endocrinology & Metabolism
Vol. 22, No. 6, pp. 1061–1073, 2008
available online at http://www.sciencedirect.com
Exposure to fallout from the radioactive isotopes released from the explosion was
trivial. After Chernobyl, millions of people were exposed to significant levels of radio-
activity from fallout, and if low levels of exposure are included that figure rises to hun-
dreds of millions. External whole-body radiation was relevant only to those working in
or close to the reactor, particularly during and in the few days after the accident. The
radiation from the released radioactive isotopes was largely beta and gamma radiation.
While the whole-body radiation from the bombs affected all organs fairly evenly, and
the dose was dependent on distance from the hypocentre, the isotopic radiation from
Chernobyl affected organs differently depending on the nature of the isotope, and the
dose from fallout was influenced by many factors: time, atmospheric, dietary, and
environmental. The health consequences of Chernobyl will therefore differ greatly
from those of exposure to the atomic bombs, and studies of the lifetime effects
of Chernobyl are equally as important as those of exposure in Hiroshima and
The accident at Chernobyl happened on April 26th 1986, it was the result of a combi-
nation of an ill-judged experiment, poor reactor design, and human error. The reactor
overheated, the graphite core caught fire, there was a steam explosion that blew off
the reactor lid, scattered fragments of radioactive fuel and burning graphite in the im-
mediate vicinity, and ‘boiled off’ the volatile isotopes present, releasing in total about
1019Becquerel into the atmosphere. The main isotopes released were xenon-133, io-
dine-131, tellurium-132, and neptunium-239, with smaller amounts of caesium-134 and
-137, strontium-89 and -90, and a range of others.1Iodine-133 and -135 were also re-
leased, but the half-lives of these isotopes are extremely short and they are unlikely to
be important for health consequences, apart from to those living close to the reactor.
Chernobyl is located in the far north of Ukraine, close to the border with Belarus. The
radioactive cloud was at first carried to the north, and the heaviest fallout occurred in
the south of Belarus, particularly in the district (Oblast) of Gomel.
The immediate aftermath of the accident was marked by secrecy and confusion.
There were no general health precautions, no sheltering, and no effective distribution
of stable iodide. Evacuation of the nuclear workers’ town of Pripyat started about 36
hours after the explosion, but evacuation from the designated 30-kilometre exclusion
zone continued for weeks. For the first few critical days consumption of locally pro-
duced milk and other food continued as usual. Severe short-term health effects were
seen in those working on the reactor site who had been exposed to high doses of
whole-body radiation. About 150 workers suffered from acute radiation syndrome,
28 died within the first 4 months from skin, bone marrow and intestinal effects.
The death rate in this group has risen to over 50, including deaths from diseases
such as myeloproliferative disorders. In contrast, those exposed to fallout suffered
no serious direct radiation effects in the first few years, although there were consider-
able psychosocial effects. The consequences of evacuation and relocation, the fear of
potential health effects from exposure to unknown amounts of radiation, and partic-
ularly concern for children and the destruction of a way of life all contributed to this.
The International Atomic Energy Authority (IAEA) arranged a study of the conse-
quences. The delegation visited some of the most affected areas, examined those
exposed, held meetings with the inhabitants, and issued a reassuring report that 4
years after the accident there appeared to be no significant health consequences.2
THYROID CARCINOMA INCIDENCE
The first indications of a rise in the incidence of childhood thyroid carcinoma were re-
ported to the International Atomic Energy Agency (IAEA) in 1990, but were not fol-
lowed up. Over the next 2 years increasing numbers of children with thyroid
carcinoma were seen in Minsk, the capital of Belarus, and Kiev, the capital of Ukraine.
Reports of this increase reaching countries outside those most affected were largely
discounted, because while it was known that external radiation would lead to an in-
crease in thyroid carcinoma, the latent period was thought to be 10 years, and isotopes
of iodine were thought to be of very low carcinogenicity or even non-carcinogenic. In
1992 a delegation from WHO Europe visited Minsk, and was shown 12 children with
thyroid carcinomas, together with histological sections from 25 cases that had been
seen in Minsk in about 6 months. Belarus is a country with about 2 million children,
and thyroid carcinoma in children is normally a rare disease, with a UK incidence of
the order of 1 case per million children per year, although some tumour registries re-
cord a rate several-fold higher. The recorded rate in Belarus before Chernobyl was less
than 1 per million. These findings clearly indicated a major increase, and the visit led to
two letters to Nature, setting out the number seen, the clinical and pathology confir-
mation, and making the case for wider studies.3,4Reviewing the Belarus records the
first rise probably started in 1989, only 3 years after the accident; thereafter the num-
ber of cases of childhood thyroid cancer rose steadily, reaching about 100/million/year
in Gomel Oblast by 1995. The incidence correlated with the level of fallout, highest in
Gomel Oblast in the south of Belarus, just over the border from Chernobyl, very low
in Vitebsk Oblast in the north of Belarus where fallout was negligible. The incidence of
childhood thyroid cancer in Northern Ukraine was also increased, but to a lesser de-
gree than in Gomel. Radiation from administered iodine-131 gives a dose to the thyroid
that is up to a thousand times higher than the average dose to other tissues, and there
is no doubt that the increased incidence of thyroid cancer is largely if not entirely due
to ingestion and inhalation of isotopes of iodine, particularly iodine-131. Analysis of the
incidence figures showed that the sensitivity to radiation-induced thyroid carcinoma
was critically dependent on age at exposure, with the youngest being most at risk.
The rate of increase of childhood thyroid cancer slowed after 1994, the incidence
reached a peak and then declined as those who were children at the time of the acci-
dent passed into adolescence. As the increase in children declined, so the incidence in
adolescents increased; by 2002 the increase in adolescents stopped as those exposed
in later childhood became adults(Figure 1).5While there is no doubt that those ex-
posed as children are carrying their increased risk for thyroid carcinogenesis with
them as they become adults, it is still uncertain whether those who were adults at
the time of exposure have an increased risk. The incidence of recorded thyroid cancer
in adults in the exposed areas has increased since 1990. At least some of this increase is
likely to be a result of increased ascertainment, and thyroid carcinoma rates have been
increasing globally. Studies of thyroid carcinogenesis after external radiation found no
evidence of an increase in adults over 406, but the numbers studied were small com-
pared to the size of the exposed population after Chernobyl. For these, and for the-
oretical reasons, it seems that if there is an increase in radiation-induced thyroid
carcinoma in those exposed as adults to Chernobyl fallout, it will be very much lower
than the increase seen in those exposed as children.
By 2006, 20 years after the accident, the WHO and IAEA reported that over 4000
thyroid carcinomas had occurred in those under 19 at exposure in Belarus, Ukraine,
Post-Chernobyl thyroid cancer1063
and the contaminated areas of the Russian Federation.7These figures must be consid-
ered carefully. There is no doubt that ascertainment increased, because of both
screening and increased awareness of the risk. This would lead to more cancers being
discovered at a younger age. In addition there is no reason to suppose that the in-
crease in incidence stopped at the borders of the countries or the designated contam-
inated areas. Virtually all of Europe was exposed to fallout containing isotopes of
iodine, with some areas – often mountainous areas which experienced rain as the ra-
dioactive cloud passed overhead – exposed to relatively high levels, but still much less
than the areas close to the reactor. There is at present no irrefutable evidence of
a Chernobyl-related increase in thyroid carcinoma in areas remote from the reactor.
In trying to estimate the possible risk there is a considerable debate as to whether it is
appropriate to apply a linear-no-threshold (LNT) model. The LNT model is the ac-
cepted approach, and if this is applied to most of Western Europe the expected in-
crease in thyroid carcinoma due to Chernobyl has been estimated as about 16
thousand cases8, but these would be spread over a large population over many years,
and the change in incidence in most areas would be too small to be detectable with
standard epidemiological approaches.
Uncertainties about the dose, the population studied, the LNT model, and the
DDREF (dose and dose rate effectiveness factor) all contribute to the great varia-
tion in the estimates for the total number of deaths expected to occur as a result
of the Chernobyl accident. The press release from the 2006 WHO/IAEA 20 year
anniversary conference gave a figure of 4000, without making it clear that this re-
ferred to the designated contaminated areas only. Greenpeace has quoted figures
of over 100,000, while Cardis estimated 15,700 deaths across the whole of Europe
up to 2065.8Cardis also pointed out that this figure represented less than 0.01% of
all cancer deaths expected in the same population over the same period. Only
a small proportion of these deaths are likely to be from thyroid carcinoma. Twenty
years after the accident only 15 deaths had been reported in those exposed under
the age of 19 in the contaminated areas, and that figure did not separately identify
patients with medullary carcinoma, which is a significant cause of death in unex-
posed children in that age group. Because of the slow growth of differentiated thy-
roid carcinoma, death from the disease may occur decades after the initial diagnosis.
Further deaths will undoubtedly occur, but the survival rate is likely to be well over
90%, and probably over 95%.
8889 90 9192 93 9495 96 97 9899012
Year of Operation
Figure 1. Changes in thyroid carcinoma incidence rate (per 100,000) with time from 1988 to 2002. The
initial rise in children diminishes as those exposed become adolescents, and in turn as the older children
exposed become young adults.5
PATHOLOGY AND MOLECULAR BIOLOGY
The earliest cases in the dramatic rise in incidence of thyroid carcinoma in children
were all papillary carcinomas. In the first few years 98% of Belarussian and 94% of Uk-
ranian cases were PTCs.9As the exposed population aged the proportion of follicular
carcinomas has increased slightly, but there is as yet no proof that this is due to radi-
ation, and the proportion of papillary carcinomas in those who were under 18 at ex-
posure in Belarus has remained above 95% in most years. The incidence of thyroid
carcinoma in unexposed young children is so low that virtually all the early exposed
cases in young children can be assumed to be radiation induced. However as those
exposed as children reach adulthood, their increased risk continues, but spontaneous
tumours form an increasing proportion of the total.
While the great majority of the tumours are PTCs, there have been interesting
changes in the frequency of different subtypes of PTC. The earliest reports com-
mented that nearly all showed a solid subtype, and speculated that this could be
a marker of radiation induced tumours. Later studies showed a decline with time in
the proportion of the solid subtype and an increase in the proportion of the classic
subtype. Quantification of the morphological changes showed that these changes
were significant; the less mature solid tumours also showed more direct invasion
than the more mature classic papillary carcinomas. Because the great majority of tu-
mours occurred in children who were very young at exposure, the changes correlated
both with increasing age and increasing latency. Study of a group of tumours in children
who were older at exposure showed that latency was the key factor.10
Molecular studies of the early Chernobyl-related tumours found that a very high
proportion showed a RET rearrangement, and almost all were RET-PTC3. Again it
was speculated that this rearrangement might be a marker for radiation-induced tu-
mours. Over time, the proportion of tumours with a RETrearrangement has declined,
but in the RET-positive tumours the proportion with RET-PTC1 has increased and the
proportion with RET-PTC3 has decreased.11TRK and RET-PTC2 rearrangements have
been found in only a small proportion of tumours; various other RET-PTC rearrange-
ments have been described, usually in single cases.
In adult non-radiated cases about half show RET-PTC rearrangements and about half
BRAF point mutations, with little or no overlap. In the post-Chernobyl cases BRAF
point mutations are uncommon12, but the studies were carried out in young patients.
In unirradiated patients BRAF mutations are less frequent in childhood than adult PTCs.
It remains possible that BRAF mutations will become more common in ‘Chernobyl’ tu-
mours as the exposed population ages, but one study of thyroid carcinomas from
adults who had received external radiation found a low frequency of BRAF mutations.13
It is also interesting that a small number of cases with BRAF rearrangements have been
found in PTCs in children exposed to Chernobyl fallout.14
Several groups have investigated the relationship between morphological subtypes
of PTC and the molecular pathology. All have found that the solid type of PTC is linked
to RET-PTC3, and the classic type to RET-PTC1.15,16
Two important conclusions can be drawn from these studies. The first is the impor-
tance of latency; we have seen two successive waves of tumours in those exposed to
high levels of fallout as children, each with different molecular, morphological and clin-
ical findings. It is impossible to predict with certainty what further waves of tumours
may occur. A diagrammatic representation of the genotype–phenotype-latency–clinical
behaviour interaction is shown in Figure 2.
Post-Chernobyl thyroid cancer1065
These studies also allow the construction of a plausible scenario to explain the
pathobiology of the development of these tumours. While most spontaneous and
chemical carcinogen-induced mutations are point mutations, radiation preferentially
induces double-strand DNA breaks, which can lead to deletions and rearrangements.
Although a variety of rearrangements may occur, some will be lethal, and only those
that lead to increased growth in the follicular cell are likely to increase the chance
of carcinogenesis. RET-PTC3 has been shown in vitro to induce a higher growth rate
than RET-PTC117, and it may well be that this explains the shorter latency and greater
aggressiveness of RET-PTC3 tumours. The range of rearrangements induced by radia-
tion is also likely to be influenced by the interphase arrangement of DNA; in the fol-
licular cell nucleus the break points for RET-PTC1 have been shown to lie very close
together.18They lie at the crossover point where the DNA strand forms a loop, so
that a single ‘hit’ could break both strands. If they are then joined wrongly, the inver-
sion that is seen in RET-PTC1 is formed. Attempts have been made to identify a radia-
tion-specific pattern of expression using microarrays, but there is at present no
FACTORS INFLUENCING THE RISK OF THYROID
The main risk factor is the radiation dose to the thyroid. The Chernobyl accident is
estimated to have released about 1.7 ? 1018Bq of iodine-131, and although large
amounts of much shorter-lived isotopes of iodine as well as tellurium-132, decaying
to iodine-132, were released, these are of importance only for the population living
close to the reactor and exposed within a very short time of the accident. Thyroid
radioactivity was directly measured in thousands of those exposed, but measurements
were not made immediately after exposure, release continued for about a week, and
Mutation RETPTC3RETPTC1 ? BRAF, RAS, PAX8
Tumour Pap Ca
Morphology ? Encapsulated
ClinicalTypical ? ‘benign’/typical
Latency 7–17 years? 15–
Pap Ca ? Pap Ca ?Foll Ca
Figure 2. Diagrammatic presentation of successive changes in molecular, morphologic and clinical findings in
papillary thyroid carcinomas (PAP Ca) with latency after Chernobyl.
1066 D. Williams
doses have had to be reconstructed from the available data. Despite this, a strong re-
lationship was found between the dose received in childhood and the subsequent risk
of thyroid cancer, with in one study an odds ratio at 1 Gy of 5.5–8.4, depending on the
model used.19This estimate is in the same range as that for the risk of thyroid carci-
nogenesis following external radiation.
The risk of developing thyroid carcinoma after Chernobyl is also strongly corre-
lated with a young age at exposure (Figure 3).9Twelve years after the accident the
number of thyroid carcinomas in Belarus in those who were under 1 at the time of
the accident was 220, and the number fell rapidly with age at exposure; 22 cases oc-
curred in those who were aged 10 at the accident.20Some at least of the age corre-
lation is due to the higher dose received by the thyroid of young children living in an
area of fallout without taking any precautionary measures. The main route by which
radioactive isotopes of iodine reach the body is through milk. Fallout on grass is in-
gested by cows, their mammary epithelium concentrates the iodine, and children drink
relatively more milk than adults. Nursing mothers also drink a lot of milk, and the lac-
tating human mammary epithelium also concentrates iodine. In addition, the uptake of
radioactive iodine is relatively greater in the young child, dropping with age. It is likely
that there is also a difference in biological susceptibility to radiation carcinogenesis in
the thyroid, as an age-related risk was also found in children exposed to external
A further important susceptibility factor is the iodine status. The areas around
Chernobyl are relatively iodine-deficient, and in Belarus children from areas with a sta-
ble iodine intake in the lower tertile were found to have approximately 3 times the risk
of those in the upper tertile.19Iodine deficiency affects the uptake of radioactive io-
dine, and therefore the dose to the gland, but most of the effect depends on the in-
creased size of the gland that follows long-term iodine deficiency. Iodine deficiency is
likely to affect the consequences of exposure of a population to iodine-131 in fallout in
a number of ways. A quantitative morphological study found that there was no differ-
ence between thyroid carcinomas from children exposed to fallout after Chernobyl
and age-matched children from the same areas born after the accident. There were,
however, marked differences between childhood PTCs from the Chernobyl area
and those from Japan, with PTCs from England and Wales being generally inter-
mediate. The Japanese tumours showed much greater differentiation and less aggres-
siveness than the Chernobyl area tumours. While ethnic causes could not be
<12468 10 1214
Age at exposure
No. of cases
Figure 3. Change in sensitivity to thyroid carcinogenesis after Chernobyl with age at exposure.22
Post-Chernobyl thyroid cancer 1067
excluded, the very large differences in iodine status were thought to be the most likely
cause.21A molecular comparison could not be carried out, so it was not possible to
determine whether the less rapidly growing thyroids in Japan were less susceptible to
the RET-PTC3 rearrangement, or the effect of the rearrangement was modified by the
lesser growth stimulation resulting from high levels of dietary iodine. Iodine deficiency
is a possible factor in the unexpectedly short latent period for the first increase in thy-
roid cancers after Chernobyl, only 4 years after the accident.
Another potentially important factor influencing the risk of developing thyroid can-
cer after Chernobyl is genetically determined susceptibility. Non-medullary thyroid
cancer has a relatively high familial element in non-radiated cases, but while genes
have been identified that are associated with familial follicular and oxyphil tumours,
and with the special type of thyroid cancer associated with familial adenomatous poly-
posis, no gene has yet been linked with ‘ordinary’ papillary carcinomas. PTCs were
found in exposed siblings after Chernobyl more often than would have been expected
by chance22, but as yet no specific genetic link has been demonstrated. Some genes are
known to be associated with the repair of DNA double-strand breaks, among them
the BRACA genes. In view of the link between thyroid and breast cancer the possibility
of germline defects in BRACA as well as other dsb repair genes needs investigation.
CLINICAL OUTCOME OF THYROID CARCINOMAS
Despite the large numbers of thyroid carcinomas that have occurred in those who
were exposed to high levels of fallout from Chernobyl as children, the number of
deaths from thyroid carcinoma has been very low; 15 deaths were reported by the
WHO/IAEA 20-year review7, but deaths from causes other than thyroid disease
were not separately identified, nor were deaths from medullary carcinoma. The ag-
gressiveness of the early cases has been well documented, and in these cases direct
extrathyroid invasion and lung metastases were relatively frequent. The results of sur-
gical treatment of 740 cases of childhood thyroid cancer from Belarus have been re-
ported in a recent important paper23; 92% had been exposed to Chernobyl fallout,
95% were PTCs, and nine of these had a family history of thyroid malignancy.
Lymph-node spread was found in 69% of cases, lung metastasis by conventional
x-ray in 2.3% (17 cases), but the first post-therapeutic radioiodine scan showed lung
spread in 76 cases. Lung metastasis was strongly associated with a young age at pre-
sentation. Total thyroidectomy was recommended for childhood thyroid carcinoma,
and was carried out in the majority of patients. Radioiodine therapy was used for ab-
lation of thyroid remnants and/or lung metastases in 63% of cases. The outcome was
excellent, with 5- and 10-year survival for the whole series of 99.5 and 98.8 respec-
tively. Of the eight patients who died, three died of causes other than thyroid cancer,
three from medullary carcinoma, and one from follicular carcinoma. Only one person
died from papillary carcinoma; this patient, with widespread pulmonary metastases,
was one of the early cases, and was not treated with radioiodine. These results under-
line the generally good prognosis of appropriately treated childhood papillary carci-
noma, but of the 128 patients with lung metastases complete remission was
achieved in 37 of 128 cases, and those with incomplete remission are still maintained
on radioiodine therapy.
There is of course a risk of complications from treatment. In the series just dis-
cussed23, 6.2% suffered permanent recurrent laryngeal nerve damage, and 12.3% per-
manent hypoparathyroidism. These figures are higher than in adult series, but at least
1068 D. Williams
in part reflect the difficulty of surgery in small children often with local tumour spread.
Second malignancies have been reported after radioiodine therapy in adults, and one
report refers to the occurrence of one salivary gland tumour and one syringoepithe-
lioma in a follow-up of 245 radioiodine-treated thyroid carcinomas from Belarus.24Sal-
ivary-gland tumours are known to be radiation-induced, and an increased risk of their
occurrence has been reported after radioiodine therapy. It is important to balance the
risks and benefits of high-dose radioiodine therapy for children with thyroid carci-
noma, and the risks include pulmonary fibrosis as well as second malignancies. Studies
of the very large cohort of childhood thyroid carcinomas after Chernobyl should pro-
vide very valuable information to help in these decisions.
With increasing latency the patient are of course older, and the tumours have in
general become less aggressive. It currently seems likely that the eventual cause-
specific death rate for the tumours that have occurred in the first 20 years after
the accident will be less than 5%, possibly much less. Hopefully this trend will continue
for cases that will arise in the future, but it remains possible that different mutations
may lead to tumours with a different morphology and different clinical behaviour.
OTHER THYROID EFFECTS
An increase in thyroid nodularity in more exposed areas was an early finding, although
interpretation was complicated by the variable levels of iodine deficiency. A Ukranian
study has shown an increase in the incidence of follicular adenomas in exposed chil-
dren and adolescents, with a linear dose–response relationship.25The risk was smaller
than the risk for carcinoma, but it must be remembered that the mean latency for fol-
licular adenomas after external radiation was longer than that for carcinomas, so the
risk ratios may change with time.
Autoimmune thyroid disease has been linked to radiation; it also occurs in associ-
ation with PTC. In a large cohort study26, no radiation-related increase in autoimmune
thyroiditis was found. However, there was a modest link between radiation dose and
the levels of thyroid peroxidase antibodies, which was present in cancer-free
Because of the early and dramatic increase in thyroid carcinoma incidence, most of the
attention on the health consequences of the Chernobyl accident has been focused on
the thyroid. The psychosocial consequences have been briefly discussed, and are im-
portant. The nature of the radiation exposure suggests that non-thyroid consequences
could follow exposure to isotopes of iodine or to other isotopes, including strontium
which is bone-seeking and caesium which is generally distributed throughout the body.
Cs-137 with a half life of 30 years is still present at above generally accepted safe levels
in areas surrounding the exclusion zone. The very high dose to the thyroid from io-
dine-131 is dependent upon the ability of the gland to concentrate, bind and store
the isotope. A variety of epithelia possess the iodide symporter, allowing them to con-
centrate iodide, but the absence of binding and storage means that the tissue dose is
very much less than that of the thyroid. Mammary epithelium, particularly during lac-
tation, concentrates iodine-131, and an increase of breast cancer in young women in
Gomel Oblast has been reported.27It is not clear yet whether the increase is related
to lactation at the time of Chernobyl. The most susceptible time for radiation-induced
Post-Chernobyl thyroid cancer1069
breast cancer in the atomic bomb studies was at puberty, but it is uncertain whether
pubertal breast epithelium has a functional iodide transporter. Exposure to iodine-131
does involve an element of whole-body radiation, as does exposure to other radioac-
tive isotopes, especially caesium. The possibility that breast cancer may occur in the
future in post-Chernobyl thyroid cancer patients who were treated at a young age
with high doses of radioiodine also needs consideration.
A well-documented radiation-induced malignancy is leukaemia, and radiation from
isotopes of strontium bound to bone could have contributed to the marrow dose. In-
creases in leukaemia incidence in exposed populations have been both claimed and de-
nied; a recent assessment concludes that, apart from clean-up workers, there is no
proof of a link to exposure.28There have been reports, sometimes anecdotal, of in-
creases in a variety of other tumours in those exposed to Chernobyl fallout, including
for example brain and kidney. In the absence of thorough epidemiological studies,
based on verified diagnoses and taking into account the problems of ascertainment,
it is difficult to be certain that these are Chernobyl-related increases. What they do
show is the continuing need for well-supported long-term studies similar to those still
being carried out after the atomic bombs.
The possibility that radiation can lead to germ-cell mutations that can be transmitted
to subsequent generations has been known for many years, but has been associated
with relatively high doses of radiation. The finding of minisatellite instability in the un-
exposed children of fathers exposed to Chernobyl fallout29was unexpected and con-
cerning. The technique is very sensitive, and there is as yet no evidence of any physical
disease in those carrying the instability. The dose to the testis from fallout would be
orders of magnitude lower than the thyroid dose, unless there was an unexpected
concentration in the testis of one of the radioactive isotopes released. Several authors
have failed to find significant elevation of mini- or micro-satellite mutations in children
of Chernobyl clean-up workers.30
The accident at Chernobyl in 1986 led to the world’s largest release of radioactivity,
and to the largest number of tumours of one type due to one cause on one date that
has ever occurred. While it is true to say that the overall effects have been less disas-
trous than many feared, and that the projected Europe-wide health effects pale into
insignificance besides those of tobacco, the effect of the accident on the lives of hun-
dreds of thousands of people living around the reactor have been catastrophic.
The health consequences of radiation exposure after Chernobyl differ greatly from
those after the atomic bombs. The major consequences that have followed from ex-
posure to fallout have been discussed, and reviewing the situation after over 20 years
a number of principles emerge. Radiation induces a range of mutations; those that con-
fer a growth advantage or an increased liability to further mutations will be selected.
Different mutations or mutational pathways will lead to tumours with different molec-
ular, morphological and clinical characteristics, and especially with differing latent pe-
riods. It follows that observations at any one period must be interpreted with care:
a good example was the belief that radiation-induced thyroid tumours were more ag-
gressive than sporadic tumours, and that solid PTCs and RET-PTC3 rearrangements
1070 D. Williams
were typical of radiation-induced thyroid tumours. We now recognize that these were
all features of short-latency tumours, and the findings changed with increasing latency.
The search for a radiation-specific molecular marker has not produced results; it may
be that the main molecular characteristic of radiation-induced tumours is a high fre-
quency of rearrangements or deletions compared to point mutations. Another lesson
comes from the attitude in the West to the early reports of an increased incidence in
childhood thyroid carcinomas only 4 years after the accident. The scepticism was
based on the dogma that the latent period for radiation-induced thyroid carcinoma
was 10 years, and that iodine-131 carried low or no carcinogenic hazard. Both dogmas
were wrong. The first was based on external radiation in an iodine-sufficient popula-
tion with very many fewer children studied than were exposed after Chernobyl; the
second was based on experiences with iodine-131 administered to adults, mostly using
high doses as therapy. In retrospect, it was not appropriate to apply this experience to
the exposure of millions of children living in iodine-deficient areas to sub-ablative
doses of iodine-131.
The final point arises from a comparison with the experience after the atomic
bombs. Comprehensive international studies were set up to monitor the long-term
health consequences. The great majority of these effects did not become apparent un-
til more than 20 years after exposure. It may well be that because of the different pat-
tern of radiation exposure the same will not be true after Chernobyl. However, if the
international community does not support studies of the effects both in the high-ex-
posure areas around the reactor and in more distant low-dose areas for the life-time
of those exposed, the full consequences of exposure to fallout may never be known.
Studies of the consequences of the massive release of radioisotopes in the first 2 de-
cades after Chernobyl have increased our understanding of radiation and thyroid car-
cinogenesis. While radiation from iodine-131 has little or no carcinogenic risk to
adults, children are at particular risk, and the risk is greatest for infants. This has con-
siderable significance for protection in the event of a similar accident. The latent pe-
riod may be no more than 3–4 years, and those exposed in childhood carry the risk
with them into adulthood, certainly for 20 years; without life-time studies the full risk
cannot be assessed. The nature of the exposure to iodine-131 with its short half-life,
the huge size of the exposed population, and the large number of thyroid carcinomas
that have occurred allow a correlation with latency. To date virtually all the tumours
are PTCs, and latency correlates with different RET-PTC variants, morphology and clin-
ical behaviour. Tumours with longer latency, yet to occur, may show other different
patterns. Despite the initial conclusions that radiation-induced tumours were unusu-
ally aggressive, it is now evident that the aggressiveness is linked to short latency
rather than radiation, and the overall death rate to date is extremely low; only one
Belarussian patient from over 600 Chernobyl-related PTCs has so far died from the
disease. There are many unanswered questions: the degree to which screening has
contributed to the size of the outbreak, the need for aggressive treatment has been
shown for young patients, but will it be needed for small tumours as those at risk
age and are subject to intensive screening? The risks of radiation therapy for young
patients who have already been exposed to radiation need to be evaluated. Above
all, diagnostic, molecular morphological and clinical studies need to be continued
for the life-time of those at risk.
Post-Chernobyl thyroid cancer 1071
*1. UNSCEAR. Sources and effects of ionising radiation (Report to the General Assembly with Scientific
Annexes); Volume II Effects; Annex J. New York: United Nations, 2000.
2. Mettler FA, Williamson MR, Royal HD et al. Thyroid nodules in the population living around Chernobyl.
JAMA: The Journal of the American Medical Association 1992; 268: 616–619.
3. Kazakov VS, Demidchik EP & Astakhova LN. Thyroid cancer after Chernobyl. Nature 1992;
4. Baverstock K, Egloff B, Pinchera A et al. Thyroid cancer after Chernobyl. Nature 1992; 359: 21–22.
5. Cardis E, Howe G, Ron E et al. Cancer consequences of the Chernobyl accident: 20 years after. Journal
of Radiological Protection 2006; 26: 127–140.
*6. Ron E, Lubin JH, Shore RE et al. Thyroid cancer after exposure to external radiation, a pooled analysis
of 7 studies. Radiation Research 1995; 141: 259–277.
7. WHO. Health effects of the Chernobyl accident and special health care programmes. Report of the UN
Chernobyl Forum Expert Group ‘Health’ (EHG). Geneva: World Health Organisation, 2005.
*8. Cardis E, Krewski D, Boniol M et al. Estimates of the cancer burden in Europe from radioactive fallout
from the Chernobyl accident. International Journal of Cancer 2006; 119: 1224–1235.
*9. Williams ED. Effects on the thyroid in populations exposed to radiation as a result of the Chernobyl
accident. In: One decade after Chernobyl. Vienna: International Atomic Energy Authority, 1996, pp.
*10. Williams ED, Abrosimov A, Bogdanova Tet al. Thyroid carcinoma after Chernobyl, latent period, mor-
phology and aggressiveness. British Journal of Cancer 2004; 90: 2219–2224.
*11. Rabes HM, Demidchik EP, Siderow JD et al. Pattern of radiation induced RET and NTRK1 rearrange-
ments in 191 post Chernobyl papillary carcinomas: biologic, phenotypic and clinical implications. Clinical
Cancer Research 2000; 6: 1093–1103.
12. Lima J, Trovisco V, Soares P et al. BRAF mutations are not a major event in post-Chernobyl childhood
thyroid carcinomas. The Journal of Clinical Endocrinology and Metabolism 2004; 89: 4267–4271.
13. Collins BJ, Schneider AB, Prinz RA et al. Low frequency of BRAF mutations in adult patients with
papillary thyroid cancers following childhood radiation exposure. Thyroid 2006; 16: 61–66.
14. Ciampi R, Knauf RA, Kerler R et al. Oncogenic AKAP9-BRAF fusion is a novel mechanism of MAPK
pathway activation in thyroid cancer. The Journal of Clinical Investigation 2005; 115: 94–101.
*15. Nikiforov YE, Rowland JM, Bove KE et al. Distinct pattern of ret oncogene rearrangements in
morphological variants of radiation-induced and sporadic thyroid papillary carcinomas in children. Can-
cer Research 1997; 157: 1690–1694.
*16. Thomas GA, Bunnell H, Cook HA et al. High prevalence of RET-PTC rearrangements in Ukranian and
Belarussian post-Chernobyl thyroid papillary carcinomas. The Journal of Clinical Endocrinology and
Metabolism 1999; 84: 4232–4238.
17. Soares P, Trovisco V, Rocha AS et al. BRAF mutations and RET/PTC rearrangements are alternative
events in the etiopathogenesis of PTC. Oncogene 2003; 22: 4578–4580.
18. Nikiforova MN, Stringer JR, Blough R et al. Proximity of chromosomal loci that participate in radiation-
induced rearrangements in human cells. Science 2000; 290: 138–141.
*19. Cardis E, Kesminiene A, Ivanov V et al. Risk of thyroid carcinoma after exposure to 131I in childhood.
Journal of the National Cancer Institute 2005; 97: 1–9.
20. Demidchik EP, Mrochek A, Demidchik Yu et al. Thyroid cancer promoted by radiation in young people
of Belarus. In Thomas G, Karaoglou A & Williams ED (eds.). Radiation and thyroid cancer. Singapore:
World Scientific, 1999, pp. 51–60.
21. Williams ED, Abrosimov A, Bogdanova T et al. Morphological characteristics of Chernobyl-related
childhood papillary thyroid carcinomas are independent of radiation exposure but vary with iodine in-
take. Thyroid 2008; 18: 847–852.
22. Cardis E, Amoros E, Kesminiene A et al. Observed and predicted thyroid cancer incidence following
the Chernobyl accident. In Thomas G, Karaoglou A & Williams ED (eds.). Radiation and thyroid cancer.
Singapore: World Scientific, 1999, pp. 395–405.
*23. Demidchik YE, Demidchik EP, Reiners C et al. Comprehensive clinical assessment of 740 cases of
surgically treated thyroid cancer in children of Belarus. Annals of Surgery 2006; 243: 525–532.
24. Kumagai A, Reiners C, Drozd V et al. Childhood thyroid cancers and radioactive iodine therapy. Endo-
crine Journal 2007; 54: 839–847.
25. Zablotska LB, Bogdanova TI, Ron E et al. A cohort study of thyroid cancer and other thyroid diseases
after the Chernobyl accident. American Journal of Epidemiology 2008; 167: 305.
26. Tronko MD, Brenner AV, Oliljnyk VA et al. Autoimmune thyroiditis and exposure to iodine 131 in the
Ukranian cohort study. The Journal of Clinical Endocrinology and Metabolism 2006; 91: 4344–4351.
27. Pukkala E, Kesmininene A, Poliakov S et al. Breast cancer in Belarus and Ukraine after the Chernobyl
accident. International Journal of Cancer 2006; 119: 651–658.
28. Howe GR. Leukemia following the Chernobyl accident. Health Physics 2007; 93: 512–515.
29. Dubrova YE, Nesterov VN, Krouchinsky NG et al. Further evidence for elevated human minisatel-
lite mutation rate in Belarus eight years after the Chernobyl accident. Mutation Research 1997; 381:
30. Slebos RJ, Little RE, Umbach DM et al. Mini- and microsatellite mutations in children from Chernobyl
accident cleanup workers. Mutation Research 2004; 559: 143–151.
Post-Chernobyl thyroid cancer1073