|
1 Department of Public Health Sciences, University
of Edinburgh Medical School; and
2 Leukaemia Research Fund Centre at the Institute of Cancer Research/
Chester Beatty Laboratories, London, UK
Correspondence: Dr FE Alexander, Dept of Health
Science, University of Edinburgh Medical School/ Teviot Place,
Edinburgh, EH8 9AG, UK Received 14 April 1998, accepted 13 May 1998
Unexplained clusters of childhood leukaemia have generated concern
that they may be causally related to environmental exposure to ionising
radiation. The workshop provides in-depth examination of the aetiology
of childhood leukaemia, patterns of clustering exhibited by cases
and the influence of exposure to ionising radiation. Special attention
has been focussed on the EUROCLUS study of clustering of childhood
leukaemia and monitoring of populations exposed to contamination
following the Chernobyl accident. There is insufficient evidence
to conclude that environmental ionising radiation exposure is a
causative agent for small clusters such as that reported in the
vicinity of the Krümmel nuclear facility
Keywords: childhood leukaemia; clusters; ionising radiation; common
infections
Introduction
The workshop was conducted as an adjunct to a symposium on 'Molecular
Biology of Hematopoiesis and Treatnlent of Leukaemias and Lymphomas/
in Hamburg, 2-6 July 1997. A key motivational factor was the intense
level of scientifict public health and community concerns regarding
the 'Elbmarsch clustert of childhood leukaemias.l Although the area
concerned is geographically very close to a nuclear facility (the
Krummel nuclear power plant) the workshop started from the premise
that a broadly based scientific approach was appropriate and that
this should consider other causes of leukaemiast other studies of
clustering of childhood leukaemiat as well as other evidence of
association between ionising radiation and leukaemia. This view
was endorsed by Dr Fritz Vahrenholtt State Minister of the Environment
whot in his welcomet stressed the need to consider the evidence
for involvement of infectious agents in the aetiology oft specificallyt
childhood leukaemia and the relevance that this might have to the
Elbmarsch cases.
Aetiology: current knowledge
Leukaenlias constitute about 5% of all malignancies in most populations.
There are four main subtypes of leukaemia (acute lymphoblastic leukaemia
(ALL); acute myeloid leukaemia (AML); chronic lymphocytic leukaemia
(CLL); and chronic myeloid leukaemia (CML)) that may not only differ
in biological termst but also in causal mechanisms. Incidence and
mortal ity rates generally increase wvith age with the exception
of acute lymphoblastic leukaemia which peaks in early childhood
and then rises slowly with age from a trough in late adolescence.
Rates have remained relatively stable in the recent past but show
some population-specific variations. The distribution of subtypes
differs markedly between adults and childrent with ALL representing
a small minority of adult cases but the majority of childhood cases.
More subtle variations between adults and children occur within
the four broad groups including ALL. Thus the aetiology of ALL in
children and adults may differ.2 Leukaemia aetiology was considered
by Dr Alexander (speaking in place of Professor Greaves), Professor
Gassman and Dr Zeeb.
Leukaemias can be induced experimentally in aninlals by ionising
radiationt chemicals and viruses. In donlesticated animals (catst
cattlet chickens) leukaemia/lymphoma can occur at a high rate due
to interplay between retroviruses, social conditions and genetic
background. In humanst the same group of causative factors are known
to apply to particular subtypes of tile disease. Experimental leukaenlial/lytlmpllOlllam
can be induced indirectly in aninlals by proliferative stress in
the immune system and at least one human lymphoid neoplasia (gastric
lymphoma in adults) has a similar aetiology. At present no more
than a small proportion of all cases (perhaps about 15%) can currently
be attributed to known risk factors.
In a lecture focussing on adult leukaemiast Dr Zeeb reminded the
workshop that the factor most conlmonly linked with leukaemia is
ionising radiation. The elevated incidence compared to the baseline
is much higher in leukaemias than in other cancerst as evident from
the study of atomic bomb survivors. Studies involving persons exposed
to low level ionising radiation suffer from a number of methodological
problems. For workers in nuclear power plantst a slightly elevated
risk for leukaemia has been reported. Studies among populations
in proximity to nuclear power plants remain inconclusive. Studies
of populations exposed to relatively high levels of ionising radiation
have usually been interpreted as providing firm evidence that ionising
radiation can cause all leukaemia cell-types apart from CLL; this
view wast Ilowever, challenged by Dr Kellerer whot in a provocative
lecturet questioned the strength of the evidence for ALL. In her
talk on tile aetiology of childhood leukaemia, Dr Alexander acknowledged
an established causative role for exposure to ionisitlg radiation
(eg from exposure to the atomic bomb, in utero xray) but suggested
that the proportion of cases explainled ill this way is likely to
be small.
A large number of studies have confirmed an increaased risk of (adult)
leukaemiat particularly AML, anlong workers exposed to benzene.
Occupational exposure to a number of other solvents and other chemicals
may also increase risk. There is limited evidence that similar exposures
(eg to betlzene via parental smoking) may influence childhood AML.
Chemotherapy with alkyiating agents for malignant tumours
carries increased risk for AML as a second tumour, with the risk
ratio elevated up to 300-fold. Chemotherapy appears to induce different
subtypes of myeloid leukaemia, sometimes with short latent periods.
There is strong evidence for a viral causation of the adult T cell
leukaemia/lymphoma (A TL). Infection with HTL V-1 early in childhood
poses a 1-4% lifetime risk of ATL. Other viruses are currently being
investigated with regard to their role in haematopoietic tumours.
Dr Alexander described the current research focus on patterns of
exposure and response to common infectious agents in childhood leukaemia.
The work of Kinlen and colleagues 3-5 has repeatedly tested and
confirmed the hypothesis that rates of total childhood leukaemia
would be elevated in populations which, though once isolated, had
experienced substantial population growth or mixing through other
mechanisms. These demographic features would lead to dysregulation
of herd immunity and contribute to microepidemics of common infectious
agents. Several reported clusters of childhood leukaemia including
four in the proximity of nuclear facilities can be interpreted in
these terms. 3-6 The childhood peak of ALL has developed in diverse
populations as they experience socio-economic development; this
phenomenon is attributable to the common sub-type of ALL (cALL)
and leads to a hypothesis that this specific subtype may arise as
a result of patterns of exposure to common infections which are
correlates of 'development'.² It is predicted that these patterns
include absence of exposures during infancy followed, perhaps, by
relatively late first exposure or high titre exposure. Epidemiology
provides considerable indirect support for this hypothesis (reviewed
in Ref. 7). The evidence includes both descriptive epidemiologY
and studies of individuals; an example of the latter is lower risk
in children who are not first-born in their family and/or who attend
early daycare.8 Biological mechanisms could involve direct cellular
transformation by viruses and almost certainly do for HTL V1, 2.
Other indirect mechanisms may apply for childhood leukaemia, especially
cALL. There is now evidence that inherited factors such as HLA haplotype
may influence risk; this is supportive of a role for infectious
agents.
From an epidemiological point of view, leukaemias pose an ongoing
challenge, not least because currently established risk factors
only explain to such a limited extent why these diseases occur.
Clustering of childhood leukaemia (The EUROCLUS study)
Clusters of childhood leukaemia are frequently reported but extremely
difficult to evaluate. They have been tentatively linked to environmental
hazards (eg polluted water supplies, ionizing radiation in the proximity
of a nuclear facility). However, the possible role of chance should
not be under-estimated. This can apply in several ways; firstly,
random allocation of rare events will naturally contain a number
of apparent clusters, secondly, cases caused by several different
factors may happen to reside close together and thirdly, a gen u
i ne cluster , caused by shared exposu re to another factor (eg
an infectious agent) may, by chance, occur near to an obvious industrial
or other factor which is readily suspected, though unrelated. No
cause of a cluster has ever been identified with certainty. Two
problems are ignorance of the causes of childhood leukaemia and
of whether cases are likely to occur in clusters (ie more likely
to do so than 'chance', or Poisson variability, would predict).
The EUROCLUS project was established with two objectives: firstly,
to determine whether childhood leukaemia and/or specific subgroups
show a general tendency to cluster within small areas of Europe,
and secondly, to explain any clustering which is observed. The project
involved the collection of leukaemia incidence data for 13551 cases
of childhood leukaemia (CL) diagnosed between 1980 and 1989 in defined
geographical areas in 17 countries. The cases were geograph ically
referenced to small census areas of wh ich there were over 26425
in the study regions. The PotthoffWhittinghill method 9 was applied
to test for evidence of generalised clustering within the small
areas and to estimate the magnitude of clustering. Results indicate
statistically significant evidence of generalised clustering of
CL but this is of small magnitude. 10 It follows that intense clusters
are unusual events and deserve serious attention. There is more
evidence of clustering of ALL in the childhood peak ages than of
other subgroups but this did not attain statistical significance.
The clustering in the total data set was attributable both to proximity
of these ALL cases and to proximity of cases of different cell-type
and age group at diagnosis. It may, therefore, be influenced by
some shared aetiological factor. Clustering is most evident in areas
of intermediate population density (150499 persons/km²).
For the second objective, up to 25 small areas were selected in
each of the 17 regions as the most 'clustered' ; these were matched
to 'control' areas of similar size but without evidence of excess
CL. The clustered areas were examined for evidence of temporal overlap
of cases at times of predicted susceptibility to exposure (eg for
the childhood peak of ALL, the 18 months preceding diagnosis). Clustered
and control areas were compared for two sets of factors: demographic
and environmental. Data for these comparisons were extracted from
routine sources or by observers blind to area status. The temporal
patterns in the clustered areas confirmed predictions based on the
aetiological hypothesis of shared exposures at critical times and
these results were highly statistically significant.1 10 When clustered
areas were compared with control areas, the clustered areas showed
statistically significant evidence of demographic features indicative
of initial isolation and subsequent population mixing; this was
evident using both objective criteria (population density) and more
subjective criteria (eg housing influx, construction camps and others
used by Kinlen and colleagues). Very few clustered areas had identifiable
environmental factors such as have been noted for single reported
clusters (eg proximity to nuclear facilities, which applied to just
four of the 240 clustered areas) and no comparisons of clustered/control
areas were statistically significant.
These results from EUROCLUS (see Table 1) strongly suggest that
clustering of cases of CL occurs and is related to the aetiology
and biology of the disease but involves distribution of infectious
agents rather than environmental hazards.
The Elbmarsch childhood leukaemia cluster near the nuclear power
plant of Krümmel
In the early 1990s, a childhood leukaemia cluster was recognised
in the 1500 children of a small community, Samtegemeinde Elbmarsch,
south of the nuclear power plant of Krümmel near Hamburg. 10
This phenomenon prompted a retrospective incidence study encompassing
the years 19841993. Leukaemias and related disorders were ascertained
in 4800000 inhabitants (up to age 65 years) of three counties surrounding
the nuclear plant (Lüneburgh, Hersogtum Lauenburg, Hamburg-Harburg).
The investigators12 calculated the leukaemia incidence rate by distance
to the nuclear power plant discriminating the following areas: <5
km, 5-10 km, 10-15 km, 15-20 km, > 20 km and reported an increased
risk of leukaemia for adults living in the 5-km circle surrounding
the nuclear plant (78% increase of total leukaemias).
Table.1
1 Childhood leukaemia (CL) does not display strong
spatial clustering in small areas In Europe over a 10 year period.
.Intense clusters should be taken seriously
2 CL shows evidence of spatial clustering in small areas of
Europe which is statistically significant but of small magnitude
(2% addition to Poisson variation).
3 Clustering of CL is focussed in areas of intermediate population
density ( 150-500 persons/km²) where addition to Poisson variation
is 5%
.For areas with density 250-500 persons/km², this is 15%
.For areas with density 500-750 persons/km² incidence is (uniformly)
elevated
.For more densely populated areas, incidence is reduced and clustering
is generally absent
4 Cluster areas compared with control areas have demographic
characteristics indicating
.isolation (initially)
.population mixing
Their (final) population density is likely to be moderately dense
(500-750 persons/km²) rather than sparse or very dense Their
separation from towns is most likely to be moderate (ie journey
times ½ -1 h).
5 Cluster areas compared with control areas fail to demonstrate
any unequivocal association with environmental hazards.
6 Within cluster areas, cases show statistically significant
evidence of temporal overlap involving putative critical times identified
in advance:
.The childhood peak of ALL, during 18 months preceding diagnosis
.CL diagnosed later (5-14 years), during 2 years around birth
.Infant cases during in utero period
7 Aetiology these results support hypotheses involving association
of CL with micro-epidemics of as yet unknown infectious agent(s).
Populations of density 250-500/km² and 500-750/km² are
most likely to support epidemics For the childhood peak of ALL,
delayed first exposure may increase the risk but the agent may be
causal for some other cases, including some infants
From Alexander et al.23
A critique of the interpretation of these results was provided to
the workshop by Professor Gassmann. The most important point he
raised was the geographic distinction between the areas where incidence
was elevated in children (south of the nuclear plant in the community
samtegemeinde Elbmarsch), and in adults (north of the plant); these
areas were at opposite sides of the river Elbe. In the 3000 children
of the northern half of the 5-km area (city of Geesthacht) just
Orne more case of leukaemia could be found in 1984 (1.55 had been
expected). In the adults of the samtegemeinde Elbmarsch (7900 inhabitants),
the leukaemia rate observed was that expected. This is true, too,
for the first 5-year period, as well as for the years 1989-1993
(five cases observed, 4.3 cases expected). In the adults of the
northern area of the 5-km area (city of Geesthacht) (24000 inhabitants),
surrounding the nuclear power plant, an increased number of leukaemias
was found in the 10-year period. Excesses were evident for CLL (19
cases observed, 13.7 expected) and CML (eight cases observed, 4.2
expected) but not for acute (lymphoid/myeloid) leukaemias for which
observed numbers were less than expected. It should be remembered
that CLL is known not to be caused by exposure to ionising radiation.
Professor Gassmann concluded that the two separate phenomena (high
rate of childhood leukaemia south of the power plant vs high leukaemia
rate for adults in the northern region with normal rates for the
complementary populations) must, if real, have separate causes.
There are, however, also clear indications of better case ascertainment
in the 5-km area compared to the other areas: 42% of the cases in
the 5-km area were reported by three or more institutions compared
to 20% in the 5-10 km area and 29%, 25(% and 31% ill distant regions.
The health impact of environmental pollution in Belarus after
the Chernobyl disaster
The question of health effects from the severest nuclear accident
in history has been a matter of considerable controversy and debate.
Reports to the workshop were made by Professor AM Kellerer, Professor
E Konoplya and Dr R Hille. In the wake of the catastrophic breakdown
of the reactor in Chernobyl, there was limited information from
the former Soviet Union for at least 2 years after the accident.
Professor Kellerer argued that this 'second disaster' enhanced the
prevailing uncertainties and led to a lasting loss of credibility
and that resulting confusion and fears have made it extremely difficult
to estabI ish a rei iable picture of the health effects among the
population in the contaminated zones.
The initial exposures due to the short-Iived radioiodine have led
to a dramatic increase in thyroid tunl°urs, especially in the
young.13-16 This has created a grave health problem that requires
increased efforts for early diagnoses of thyroid diseases and for
effective treatment. The continued radiation exposure due to the
10ng-Iived radioactivity is a different and difficult problenl.
It has involved decisions on the evacuation or resettlement of certain
regions and the imposition of va1rious constraints that affect the
living conditions of the population. It is the long-term health
impact of this exposure pattern that is subject to the greatest
uncertainty and controversy. As leukaemia, especially in children,
is the first and most sensitive indicator of late effects due to
whole body radiation exposures, a major effort has been made to
documelt any alteration in childhood leukaenlia rates in regions
most affected by the disaster. In an earlier project within the
franlework of the European Commission's Program on Radiation Research,
cooperation with the Institute of Radiationl Medicine, Kiev, has
led to the publication of cancer trends in the highly contaminated
regions of the Ukraine.17 No increase ill leukaemia or cancers (excluding
thyroid) were found. This result is in line with the magnitude of
exposure of the population, but it differs greatly from the almost
universal public perception .
The highest contaminations have occurred in Belarus. Childhood leukaemia
rates have been assessed in the seven major regions of Belarus for
the period 1982 to 1994.19 This is the first detailed documentation
of the childhood leukaemia rates in Belarus before and after the
accident. There is no increase in childhood leukaemia rates in anyone
of the six regions of Belarus. These findings are, however, based
on small numbers, and are consistent with increased risk of up to
20% in individual regions. Further careful monitoring of trends
in leukaemia and cancer incidence are therefore set to continue
with the support of the German and French Ministers of the Environment.
The Chernobyl accident also provides a unique opportunity to confirm
or refute risk estimates derived from the atomic bomb exposure in
Japan. However, this requires reliable dosimetric measurements of
exposure. For this purpose, measurements have been carried out in
the highly contaminated areas of Belarus by Belarussian and German
scientists. The German measurements in the years 1992 and 1993 involved
contaminated food and environmental samples and 42000 measurements
of individuals. Averaged incorporated radioactivity of the population
was significantly higher than in control areas, but only 2% of exposed
cases showed body activities of more than 25 kBq per person. This
low level represents only a slightly enhanced cancer risk, but justifies
careful monitoring as with radiation workers in the nuclear industry.
The individuals with the higher body burdens were mostly elderly
persons consuming contaminated food, including milk or forest products,
and the fact they are older reduces their life-time cancer risk.
The average accumulated dose for the most highly contaminated settlements
was below 100 mSv in 10 years. This dose is high (10 x average background)
though similar natural or background levels do exist as in the Kerala
region of south west India. No significant health impact has been
recorded there.
From these studies, the speakers concluded that no significant increase
of long-term health effects other than thyroid cancer are to be
expected in Belarus. However, the scientists involved emphasize
that this prediction does not exclude individual radiation-related
problems and for this reason, humanitarian aid and medical support
for the people in the highly contaminated areas is still necessary
and fully justified. Recent studies18 indicating that children from
Gomel, including those born after the Chernobyl accident, may have
higher incidence of non-random chromosome translocations compared
with a control series of Italian children suggest that continued
monitoring of cancer rates is warranted.
In April 1997, the Korma project 'Children and Future of Korma'
was initiated to make a more detailed evaluation of the radiological
and health situation for two special settlements in the highly contaminated
region. The aim was to evaluate the interdependence between the
soil and food contamination, the radiation exposure and the social
status and health consequences for individuals. The two villages
involved in the project, Voronovka and Vassokaja, are situated in
the Korma region with an average ground contamination of about 15
Ci/km² Cs-137 .Soil and food samples have been taken and evaluated
in the Institute of Radiobiology at Minsk. A measuring vehicle equipped
with body counters and a dosimetric and medical laboratory has been
placed at Voronovka. Blood samples have been taken and analysed
in the nearby Korma district hospital. Dosimetric monitors are also
distributed to the population and gathered after 2-3 weeks for the
evaluation.
Preliminary results show that internal exposure of the population
varies considerably with a mean value of about 100 Bq. In two cases
there was an association between contaminated food and high internal
doses. The study is continuing.
Alongside the scientific studies reported above it should be noted
that the ECLIS study²O,²l and others have monitored childhood
leukaemia and cancer in areas contaminated to lesser extents by
the Chernobyl disaster .No evidence of increases have been reported
except for thyroid cancer and infant leukaemia2² but the causal
nature of the latter association is unclear.
Concluding remarks
The workshop succeeded, as hoped, in considering the issues in
a very broad context. Presently known risk factors for leukaemias
certainly include exposure to ionising radiation but explain only
a minority of cases. Amongst risk factors under investigation, especially
for childhood leukaemia, patterns of exposure to common infectious
agents are likely to be important. EUROCLUS, the largest study ever
to have been conducted of clustering of childhood leukaemia points
to limited generalised clustering of which further scrutiny points
to explanations in terms of demographic factors suggestive of a
role for infectious agents. Examination of the Elbmarsch area and
of other noteworthy clusters fails to link them firmly to environmental
exposure to ionising radiation. Most are, however, capable of explanation
in terms of population demography. The intense monitoring that has
been in progress since the Chernobyl disaster has found little evidence
of excess leukaemia in populations having 'environmental level'
exposures. In summary, there is at present insufficient evidence
on which to base the conclusion that radiation exposure is causative
agent for the Elbmarsch childhood leukaemia cluster.
Acknowledgements
The Editor wishes to thank Professor Rolf Neth for his incredible
commitment to dissemination of information on the leukaemias now,
as well as in the past, in organizing the first brain storming sessions
in the barn of Wilsede, in the Lüneburger Heide in Northern
Germany. Professor MF Greaves and Dr FE Alexander are acknowledged
for their input in making available to us their thoughts and clear
thinking on this intricate, at times emotional and often controversial
topic. Supported by Walter Gastreich-Stiftung im Stifterverband
für die Deutsche Wissenschaft.
References
1 Hoffman W, Dieckmann H, Schmitz Feuerhake I. A cluster of childhood
leukemia near a nuclear reactor in northern Germany. Arch Environ
Health 1997; 52: 275-280.
2 Greaves MF. Aetiology of acute leukaemia. Lancet 1997; 349; 344-349.
3 Kinlen L]. Evidence for an infective cause of childhood leukaemia:
comparison of a Scottish new town with nuclear reprocessing sites
in Britain. Lancet 1988; 2: 1323-1327.
4 Kinlen L], O'Brien F, Clarke K, Balkwill A, Matthews F. Rural
population mixing and childhood leukemia -effects of the North Sea
oil industry in Scotland, including the area near Dounreay nuclear
site. Br Med J 1993; 306: 743-748.
5 Kinlen LJ, Hudson C, Stiller LA. Contacts between adults as evidence
for an infectious origin of childhood leukaemia: an explanation
for the excess near nuclear establishments in West Berkshire? Br
J Cancer 1991; 64: 549-554.
6 Viel J-F, Pobel D. Incidence of leukaemia in young people around
La Hague nuclear waste re-processing plant: a sensitivity analysis.
Statist Med 1995; 14: 2459-2472.
7 Greaves MF, Alexander FE. An infectious etiology for common acute
lymphoblastic leukemia in childhood. Leukemia 1993; 7: 349-360.
8 Petridou E, Kassimos D, Kalmanti M, Kosmidis H, Haidas S, Flytzani
V, Tong D, Trichopoulos D. Age of exposure to infections and risk
of childhood leukaemia. Br Med J 1993; 307: 774.
9 Muirhead CR, Butland K. The Potthoff-Whittinghill method. In:
Alexander FE, Boyle P (eds). Statistical Methods of Investigating
Localised Clustering of Disease. IARC: Lyon, 1996, pp 28-39.
10 Alexander FE, Boyle P, Carli PM, Coebergh JW, Draper GJ, Ekbom
A, Levi F, McKinney PA, McWhirter W, Michaelis J, Peris-Bonet R,
Petridou E, Pompe-Kirn V, Plesko I, Pukkala E, Rahu M, Storm H,
Vatten L, Wray N on behalf of the EUROCLUS project. Spatial clustering
of childhood leukaemia: summary results from the EUROCLUS project.
Br J Cancer 1998; 77: 818-824.
11 Alexander FE, Boyle P, Carl i PM, Coebergh JW, Draper GJ, Ekbom
A, Levi F, McKinney PA, McWhirter W, Michaelis J, Peris-Bonet R,
Petridou E, Pompe-Kirn V, Plesko I, Pukkala E, Rahu M, Storm H,
Vatten L, Wray N on behalf of the EUROCLUS project. Spatial temporal
patterns in childhood leukaemia: further evidence for an infectious
origin. BrJ Cancer 1998; 77: 812-817.
12 Hoffman W, Greiser E. Increased incidence of leukaemias in the
vicinity of the Krümmel nuclear power plant in northern Germany.
In: Frenzel-Beyne R, Ackermann-Liebrich U, Bertazzi PA et al (eds).
Environmental Epidemiology in Europe. Proc Int Symp, Bremen, Germany:
European Commission Directorate General V, 1996, pp185-206.
13 Becker DV, Robbins J, Beene GW, Bouville AC, Wachholz BW. Childhood
thyroid cancer following the Chernobyl accident a status report.
Endoc Meta Clin N Am 1996; 25: 197-211.
14 Sobolev B, Heidenreich WF, Kaior I, Jacob KP, Goulko G, Likhtarev
I. Thyroid cancer incidence in the Ukraine after the Chernobyl accident
comparison with spontaneous incidences. Radiat Environ Biophys 1997;
36: 195-199.
15 Danese D, Gardini A, Farsetti A, Sciacchitano S, Andreoli M,
Pontecorvi A. Thyroid carcinoma in children and adolescents. Eur
J Pediatr 1997; 15: 190-194.
16 Henshaw DL. Chernobyl 10 years on -thyroid cancer may be the
only measurable health effect. Br MedJ 1996; 312: 1052-1053,
17 Prisyazhniuk A, Gristchenko V, Zakordonets V, Fouzik N, Slipeniuk
Y, Ryzhak I. The time trends of cancer incidence in the most contaminated
regions of the Ukraine before and after the Chernobyl accident.
Radiat Environ Biophys 1995; 34: 3-6.
18 Scarpato R, Lori A, Panasiuk G, Barale R. FISH analysis of translocations
in lymphocytes of children exposed to the Chernobyl fallout: preferential
involvement of chromosome 10. Cytogenet Cell Genet 1997; 79: 153-156.
19 Ivanov EP, Tolchoko GV, Shuvaeva LP, Becker S, Nekolla E, Kellerer
AM. Childhood leukaemia in Belarus before and after the Chernobyl
accident. Radiat Environ Biophys 1996; 35: 75-80.
20 Parkin DM, Cardi SE, Masuyer E et al. Childhood leukemia following
the Chernobyl accident -the European Childhood Leukemia Lymphoma
Incidence Study (ECLIS). Eur J Cancer 1993; 29A: 87-95.
21 Parkin DM, Clayton D, Black RJ et al. Childhood leukemia in Europe
after Chernobyl -5-year follow-up. Br J Cancer 1996; 73: 1006-1012.
22 Petridou E, Trichopoulos D, Dessypris N, Flytzani V, Haidas S,
Kalmanti M, Koliouskas D, Kosmidis H, Piperopoulou F, Tzortzantau
F. Infant leukaemia after in utero exposure to radiation from Chernobyl.
Nature 1996; 382: 352-353.
Appendix 1
Contributors
Professor MF Greaves
Leukaemia Research Fund Centre at the Institute of Cancer Research
Chester Beatty Laboratories
Fulham Road, London, SW3 6JB
UK
Professor Dr E Konoplya Institute of Radiobiology Zhodinskya str.
2
22600 Minsk
Republic of Belarus
Professor Dr AM Kellerer Stahlenbiologisches Institut fer Universitat
München
Schillerstr. 42
80336 Munchen
Germany
Dr R Hille
Department of Radiation Protection Research Centre Julich GmbH Postfach
1 91 3
524525 Julich
Germany
Professor Dr W Gassmann St Marienkrankenhaus Kampenstr 51
5707 Siegen
Germany
Dr FE Alexander
Department of Public Health Sciences University of Edinburgh, Medical
School Teviot Place,
Edinburgh, EH8 9AG
UK
Dr H Zeeb
Abt Epidemiologie
Deutsches Krebsforschungszentrum
Im Neuenheimer Feld 280
69120 Heidelberg
Germany
Professor Dr RD Neth
Universitäts-Krankenhaus Eppendorf Einrichtung für Knochenmark
Transplantation Martinistraße 52
20246 Hamburg
Germany
|