European Journal of Cancer (2012) 48, 1532– 1542

Available at www.sciencedirect.com

journal homepage: www.ejconline.com

Epidemiology of glial and non-glial brain tumours in Europe

Emanuele Crocetti a,⇑, Annalisa Trama b, Charles Stiller c, Adele Caldarella a,

Riccardo Soffietti d, Jana Jaal e, Damien C. Weber f, Umberto Ricardi g, Jerzy Slowinski h,
Alba Brandes i, RARECARE working group

a
Clinical and descriptive epidemiology unit, ISPO – Palazzina 28/A Via delle Oblate 2, 50141 Florence, Italy
b
Evaluative Epidemiology, Department of Preventive and Predictive Medicine Fondazione IRCCS “Istituto Nazionale dei Tumori” Via Venezian
1, 20133 Milano, Italy
c
Childhood Cancer Research Group, University of Oxford, Richards Building, Old Road Campus, Headington, Oxford OX3 7LG, UK
d
Department of Neuroscience, University and San Giovanni Battista Hospital Via Cherasco 15, Turin, Italy
e
Department of Radiotherapy and Oncological Therapy, Haematology and Oncology Clinic, Tartu University Hospital, Vallikraavi str. 10, 51003
Tartu, Estonia
f
Radiation Oncology Department, Hôpitaux Universitaires de Genève, 4 rue Gabrielle Perret-Gentil, CH-1211 Geneva 14, Switzerland
g
Radiation Oncology, University of Turin, Azienda Ospedaliero-Universitaria San Giovanni Battista di Torino, Via Genova 3, 10126, Italy
h
Jerzy Slowinski Dept. of Epidemiology in Bytom,School of Public Health, Medical University of Silesia, Piekarska 18, 41-902 Bytom, Poland
i
Department of Medical Oncology, Ospedale Bellaria-Maggiore, Azienda ASL Via Altura 3, 40139 Bologna, Italy

Available online 7 January 2012

KEYWORDS Abstract To the central nervous system (CNS) belong a heterogeneous group of glial and
Central nervous system non glial rare cancers.
Glial tumours The aim of the present study was to estimate the burden (incidence, prevalence, survival and
Non-glial brain tumours proportion of cured) for the principal CNS cancers in Europe (EU27) and in European
Rare cancer regions using population-based data from cancer registries participating in the RARECARE
Epidemiology project.
Estimates We analysed 44,947 rare CNS cancers diagnosed from 1995 to 2002 (with follow up at 31st
Incidence December 2003): 86.0% astrocytic (24% low grade, 63% high grade and 13% glioma NOS),
Prevalence 6.4% oligodendroglial (74% low grade), 3.6% ependymal (85% low grade), 4.1% Embryonal
Survival tumours and 0.1% choroid plexus carcinoma. Incidence rates vary widely across European
regions especially for astrocytic tumours ranging from 3/100,000 in Eastern Europe to
5/100,000 in United Kingdom and Ireland. Overall, about 27,700 new rare CNS cancers were
estimated every year in EU27, for an annual incidence rate of 4.8 per 100,000 for astrocytic,
0.4 for oligodendroglial, 0.2 for ependymal and embryonal tumours and less than 0.1 for
choroid plexus carcinoma.
More than 154,000 persons with rare CNS were estimated alive (prevalent cases) in the EU at
the beginning of 2008.

⇑ Corresponding author: Tel.: +39 0557972508; fax: +39 0557972535.

E-mail address: e.crocetti@ispo.toscana.it (E. Crocetti).

0959-8049/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ejca.2011.12.013
 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542 1533

Five-year relative survival was 14.5% for astrocytic tumours (42.6% for low grade, 4.9% for
high grade and 17.5% for glioma NOS), 54.5% for oligodendroglial (64.9% high grade and
29.6% low grade), 74.2% for ependymal (80.4% low grade and 36.6% high grade), 62.8%
for choroid plexus carcinomas and 56.8% for embryonal tumours. Survival rates for astrocytic
tumours were relatively higher in Northern and Central Europe than in Eastern Europe and in
UK and Ireland. The different availability of diagnostic imaging techniques and/or radiation
therapy equipment across Europe may contribute to explain the reported survival differences.
The estimated proportion of cured patients was 7.9% for the ‘glial’ group to which belong
astrocytic tumours.
Overall results are strongly influenced by astrocytic tumours that are the most common type.
This is the first study to delineate the rare CNS cancer burden in Europe by age, sex and Euro-
pean region.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction tumours are included. However, despite the rarity of enti-

Central Nervous System (CNS) cancers are a group
of different tumour entities anatomically close to each
other but diverse in terms of morphology, site, molecu-
lar biology and clinical behaviour and presumably
aetiology.1
In Europe, the standardised (World) incidence of pri-
mary CNS cancers ranges from 4.5 to 11.2 cases per
100,000 men and from 1.6 to 8.5 per 100,000 women.2
The two most common CNS cancers, high-grade glioma
and brain metastasis occur more frequently during adult-
hood and especially among the elderly. In Europe, the
peak of incidence is 18.5/100,000 in people aged
P65 years.2 The relative frequency of CNS tumours is
however highest during childhood, when they account
for 23% of all the cancers diagnosed.3
In adults the 5-year survival rate for the primary CNS
cancers in Europe was 17% for males and 19% for
females (1995–2002),4 with differences across European
regions.5 Survivorship is higher for young European
patients – 63% – than for the elderly ones.6
Statistics on CNS tumours are estimated by grouping
all malignancies arising in all the CNS anatomic sites:
meninges, brain, spinal cord, cranial nerves and other
localisation of CNS (ICD-10 topography codes C70-
C72).7
However, rare tumours are more appropriately
defined as a combination of topographical and morpho-
logical characteristics, as defined by the International
Classification of Diseases for Oncology (ICD-O).8 The
Surveillance of Rare Cancers in Europe (RARECARE;
www.rarecare.eu) project, is a large collaboration of pop-
ulation-based cancer registries (CRs) across Europe
which provides a list of rare cancers on the basis of topog-
raphy and morphology. Clinical factors, such as difficul-
ties in achieving diagnosis, in clinical decision making
and in conducting clinical studies and the lack of stand-
ardised treatment mainly affected the definition of the
type of cancers of the RARECARE list. Under the
threshold proposed by the RARECARE project, inci-
dence lower than 6 per 100,000 per year, CNS malignant
ties selected by using this cut-off, there are many sub-enti-
ties for each group (i.e. different histologies and WHO
grade of glial tumours) with different prognosis and dif-
ferent treatment approaches that complicate the picture.
This means that management of these rare brain tumour
subgroups has to be considered hyper-specialistic, as well
as treatment options centralised in large volumes hub
centres with recognised expertise in this field. Further-
more, central pathology review in CNS tumours is partic-
ularly important. The experience of modern clinical trials
(i.e. European Organisation for Research and Treatment
of Cancer (EORTC) and Radiation Therapy Oncology
Group (RTOG) studies on grade III gliomas) showed
very high (up to 50%) inter-observer variability among
neuropathologists in the differential diagnosis between
astrocytic and oligodendroglial tumours.9
The objective of this study was to establish a picture
of incidence, prevalence and survival of rare CNS can-
cers in Europe based on the new RARECARE list of
tumours.
2. Materials and methods/cancer cases
Rare cancers of the CNS presented in this paper are
based on the new list of cancer types provided by
RARECARE. The list is organised into three tiers.
The bottom tier (tier 3) corresponds to the World
Health Organisation (WHO) name of individual cancer
entities [http://www.iarc.fr/en/publications/pdfs-online/
pat-gen/] and their corresponding ICD-O-3 morphology
and topography codes. Tier 3 entities were grouped into
categories of cancers (tier 2) considered similar from the
point of view of clinical management and research.
These categories were further grouped into more general
categories (tier 1), considered to involve the same clini-
cal expertise and patient referral structure. Accordingly
the following cancer entities were identified and will be
described in this paper:
Glial tumours (tier 1 entity) which include astrocytic,
oligodendroglial, oligoastrocytic and ependymal
1534 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542

tumours with malignant behaviour only (tier 2 enti-
ties) and
Non glial tumours of CNS and pineal gland (tier 1
entity) which includes embryonal tumours and cho-
roid plexus carcinomas (tier 2 entities).
Glial tumours were also divided and analysed in low-
and high-grade gliomas according to the WHO classifi-
cation.10 Low grade astrocytic included the ICD-O3
morphology (M) 9382, 9400, 9410, 9411, 9420, 9423
and M 9424; high grade astrocytic included the ICD-
O3 M 9381, 9401, 9430, 9440-9442. Low grade oligoden-
droglial tumours were M9450 and M 9460; the high
grade group included M 9451 only. Finally the low
grade ependymal tumours included M9391 and M
9393; the high grade group was made by M 9392.
Table 1 shows the morphology and topography codes
of the tumours described in this article.
It is worth stressing that pilocytic astrocytoma (grade
WHO I), the most frequent brain tumour of children is
excluded, following its downgrading to non-malignant
behaviour in ICDO-3. Other primary and secondary
tumour entities usually included in the CNS (metastasis,
sarcomas, germ cell tumours, meningiomas, gliomas of
optic nerve) are also excluded from this analysis because
they are described in other dedicated papers. Unspeci-
fied malignant neoplasms (8000/3 and 8001/3) are also
excluded from the analysis.
RARECARE data were extracted from the EURO-
CARE-4 study dataset covering the period of diagnosis
1978–2002, with status information available up to 31st
December 2003. For 11 countries CRs covered the entire
national population (Austria, Iceland, Ireland, Malta,
Norway, Slovakia, Slovenia, Sweden, Northern Ireland,
Scotland and Wales). Other 10 countries were repre-
sented by regional CRs, covering variable proportions
of their respective national populations. In the present
paper, data from 76 CRs out of the 89 accepting to par-
ticipate in the RARECARE project were considered.
CRs which did not classify cancers according to the
third edition of the ICD-O3 and also those which col-
lected data on childhood cancers only were excluded
from the analysis.
The incidence analysis considered case incidents from
1995 to 2002 and excluded specialised registries. Thus,
64 CRs were included in the incidence analyses. Age-
standardised incidence rates per 100,000 were computed
using the European standard population.
The prevalence per 100,000 was estimated at the
index date of 1st January 2003. Only data from 22
CRs with data covering the period 1988–2002 were used
for prevalence estimates. The counting method,11 based
on CR incidence and follow-up data, was applied to CR
data from 1988 to 2002. The completeness index
method,12 was used to estimate complete prevalence,
and involved adding the estimated surviving cases diag-
nosed prior to 1988 to those counted in 1988–2002.
The expected number of new cases per year and of
prevalent cases in Europe (EU27) was estimated multi-
plying the crude incidence and prevalence estimates
(obtained as described above) to the 2008 European pop-
ulation (EU27 = 497,455,033) provided by EUROSTAT

Table 1
Data quality indicators of rare central nervous system cancers diagnosed 1995–2002 and archived in 76 RARECARE cancer registries.
Tier Entity Cases Data quality indicator ICD-O3a codes
(N)
Death Autopsy Microscopic Cases 1995–1998 Topography Morphology
certificate (%) verification censored before five
only (%) (%) years (%)
Rare central nervous
system (CNS)
tumours
1 Glial tumour Of CNS 43,125 1.2 0.5 87.0 0.5 C71, C72.0, 9380-9384, 9391-9460
C72.8-C72.9
2 Astrocytic tumours 38,653 1.3 0.5 86.1 0.4 C71, C72.0, 9380-9382, 9384,
of CNS C72.8-C72.9 9400-9442
2 Oligodendroglial 2866 0.5 0.4 94.9 1.0 C71, C72.0, 9450-9451,9460
tumours of CNS C72.8-C72.9
2 Ependymal tumours 1606 0.4 0.7 95.2 1.6 C71, C72.0, 9383, 9391-9394
of CNS C72.8-C72.9
1 Non-glial tumour of 1872 0.4 0.2 96.8 2.0 C71, C72.0, 9362, 9390, 9470-
CNS and pineal C75.3 9474, 9490, 9500-
gland 9505, 9508
2 Embryonal tumours 1822 0.3 0.2 97.0 2.0 C71, C72.0, 9362, 9470-9474,
of CNS C75.3 9490, 9500-9505, 9508
2 Choroid plexus 50 4.0 0.0 92.0 0.0 C71, C72.0, 9390
carcinoma of CNS C75.3
a
The following tumours are excluded: metastasis, unspecified malignant neoplasms (8000/3 and 8001/3), sarcomas, germ cell tumours,
meningiomas, gliomas of optic nerve and pilocytic astrocytoma (grade WHO I).
 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542 1535

(ec.europe.eu/eurostat). In providing glial and non glial
tumours burden estimates, we assumed that the popula-
tion covered by our CRs was representative of the popu-
lation of the EU27 as a whole. We assumed that
incidence, survival and the population for the period
2003–2008 did not change. A total of 4,302,067 cancer
cases were used to produce the prevalence estimates.
Period survival indicators for the years 2000–2002
were estimated using the Brenner algorithm.13 Period
analysis provides more up-to-date survival experience
by considering survival experience in 2000-2002 (sur-
vival data were collected from 46 European CRs partic-
ipating in the RARECARE study). Survival has been
computed as relative survival that is the ratio of absolute
survival to expected age-specific and sex-specific survival
of the general population. It is a measure of the excess
mortality for patients with cancer compared with that
for the general population. The proportion of cured
patients for rare CNS tumours was estimated using a
mixture or cure model.14 This type of survival model
assumes that cured cases have the same mortality as
the general population, while the complementary frac-
tion (the fatal cases) has an excess death rate attributed
to the CNS cancer.14
This analysis was carried out on 44,947 cases of rare
CNS cancers diagnosed in 1995–2002.
The main data quality indicators for the cancer diag-
noses in 1995–2002 are presented in Table 1. Among glial
tumours 1.2% of the cases were known from death certif-
icate only (DCO); among non glial tumours DCO were
0.4% ranging from 0.3% (embryonal tumours) to 4.0%
(choroids plexus carcinomas). About 87% of the glial
tumour cases included in the analysis had histological
documentation, although the proportion varied among
cancer entities from 86.1% for astrocytic tumours to
95% to ependymal tumours. The 97.0% of non-glial
tumours of CNS and pineal gland was microscopically
verified. The proportion of cases known from autopsy
was very low, with a maximum of 0.7% for ependymal
tumours.
Quality indicators for CNS tumours are computed on
all participating CR (44,947 cases from 76 cancer regis-
tries) while incidence rates are reported for general CR
only (44,842 cases from 64 cancer registries). The sur-
vival analysis was undertaken on 46 CR because only
those had data available to perform period survival.
3. Results
3.1. Incidence
Table 2 shows crude, sex and age-specific incidence
rates recorded in Europe during the period 1995–2002.
Among CNS cancers, astrocytic tumours were the
most common tumours, with a standardised incidence
rate of 4.8 (3.0 high grade, 1.2 low grade and 0.6 glioma
NOS) per 100,000 per year in Europe, followed by oligo-
dendroglial tumours (0.4 overall; 0.3 low and 0.1 high
grade), embryonal (0.2) and ependymal tumours (0.2
overall; 0.17 low grade and 0.03 high grade). The inci-
dence rate of choroid plexus carcinomas was under 0.1
per 100,000/year.
Rare CNS tumours were slightly more common in
men than in women. The male to female crude rates
ratio ranged between 1.5 (embryonal), to 1.4 (astrocytic)
and to 1.3 (oligodendroglial), Table 2.

Table 2
Observed cases with crude incidence (rate per 100,000/year) and standard errors (SE) in Europe. Rates and SE by sex and age, with estimated
incident cases in Europe (EU27). Cases diagnosed 1995–2002 in 64 European cancer registries (CRs).
Rare central nervous EU overall Sex Age Estimated cases
system (CNS) cancers in EU27 per year
Male Female 0–19 20–39 40–59 60+
Observed Rate SE Rate SE Rate SE Rate SE Rate SE Rate SE Rate SE N
Cases 1995-
2002
Glial tumour of CNS 43,037 5.4 <0.1 6.3 <0.1 4.5 <0.1 1.2 <0.1 2.5 <0.1 7.0 <0.1 12.1 0.1 26,610
and pineal gland
Astrocytic tumours of 38,588 4.8 <0.1 5.7 <0.1 4.0 <0.1 0.9 <0.1 2.0 <0.1 6.2 <0.1 11.6 0.1 23,859
CNS
Oligodendroglial 2,845 0.4 <0.1 0.4 <0.1 0.3 <0.1 0.1 <0.1 0.3 <0.1 0.6 <0.1 0.4 <0.1 1759
tumours of CNS
Ependymal tumours 1604 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 992
of CNS
Non-glial tumour of 1805 0.2 <0.1 0.3 <0.1 0.2 <0.1 0.6 <0.1 0.2 <0.1 0.1 <0.1 <0.1 <0.1 1116
CNS and pineal
gland
Embryonal tumours 1755 0.2 <0.1 0.3 <0.1 0.2 <0.1 0.6 <0.1 0.2 <0.1 0.1 <0.1 <0.1 <0.1 1085
of CNS
Choroid plexus 50 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.0 <0.1 0.0 <0.1 0.0 <0.1 31
carcinoma of CNS
1536 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542

As regards age-specific incidence rate (per 100,000),
incidence rates for glial tumours overall (11.9/100,000/
year) and for astrocytic tumours (11.6/100,000/year)
were highest in the oldest age group (60 years and
more). Incidence peaked in the age group 40–59 years
for oligodendroglial tumours (0.6). The incidence of
ependymal tumours was quite stable across the age-
groups while embryonal tumours had their highest inci-
dence among younger patients (0–19 years; 0.6/100,000/
year) and then incidence fell throughout the age range.
Table 2 shows the number of new cases of CNS
tumours estimated in the European Union (EU27) per
year. There are about 27,700 new diagnosis of CNS can-
cers per year of which, 26,610 are ‘glial tumours of CNS’
(23,859 astrocytic tumours with unspecified gliomas;
1,759 oligodendroglial tumours and 992 ependymal
tumours) and 1,116 are ‘non glial tumours of CNS
and pineal gland’ (99% of which are embryonal
tumours).
Fig. 1 shows incidence trend for the period 1995–2002
for glioma, as overall and for age-classes. The overall
trend is flat and the same is true for all the analysed
Fig. 1. Glioma incidence trend by age groups.
age-groups with the exception of the oldest one (60+
years) that shows an increasing-incidence trend.
Table 3 describes the standardised incidence rates
(1995–2002) for EU in the whole and each European
Region. For the purposes of the analyses five regions
were identified as follow: Northern Europe (Iceland,
Norway, Sweden), United Kingdom and Ireland (Eng-
land, Scotland, Wales, Northern Ireland, Republic of
Ireland), Central Europe (Belgium, Austria, France,
Germany, The Netherlands, Switzerland), Eastern Eur-
ope (Poland, Slovakia) and Southern Europe (Italy,
Malta, Portugal, Slovenia, Spain).
There was marked geographical variation in inci-
dence for astrocytic tumours, with the highest rate in
UK and Ireland (5.1 cases per 100.000) and the lowest
(3.1) in eastern Europe. The differences in geographical
rates for all the other types of rare CNS tumours were
negligible, Table 3.
3.2. Prevalence
Table 4 shows the observed prevalence proportions at
2, 5, 15-year and the estimated complete prevalence in
Europe (index date 1st January 2003).
More than 154,000 persons were alive in EU at the
beginning of the year 2008 with a past diagnosis of rare
CNS cancers, of whom 130,000 had had in their past
clinical history a ‘glial tumour of CNS’ and about
23,000 a ‘non glial tumour of CNS and pineal gland’.
As regards ‘glial tumours of the CNS’, 19% and 32%
were diagnosed within 2 and 5 years before the preva-
lence date, respectively. The difference (13%) between
these two proportions represents the proportion of cases
in the 3rd–5th years after diagnosis, presumably still
undergoing clinical follow-up. The remaining fraction
of 68% represents long-term survivors (surviving more
than 5 years), and 47,000 of these (47% of the total) were

Table 3
Age-standardised incidence rates (per 100,000) for central nervous system (CNS) tumours in 1995–2002, with standard errors (SE) by European
region.
Rare central nervous system (CNS) Regiona EU overall
cancers
Northern Central Eastern Southern United
Europe Europe Europe Europe kingdom and
Ireland
Adj. SE Adj. SE Adj. SE Adj. SE Adj. SE Adj. SE
rate rate rate rate rate rate
Glial tumour of CNS and pineal gland 5.3 0.1 4.8 0.1 3.5 0.1 4.4 0.1 5.7 <0.1 5.0 <0.1
Astrocytic tumours of CNS 4.6 0.1 4.2 0.1 3.1 0.1 3.8 0.1 5.1 <0.1 4.4 <0.1
Oligodendroglial tumours of CNS 0.4 <0.1 0.3 <0.1 0.3 <0.1 0.3 <0.1 0.4 <0.1 0.4 <0.1
Ependymal tumours of CNS 0.3 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1 0.2 <0.1
Non-glial tumour of CNS and pineal 0.3 <0.1 0.3 <0.1 0.2 <0.1 0.3 <0.1 0.2 <0.1 0.3 <0.1
gland
Embryonal tumours of CNS 0.3 <0.1 0.3 <0.1 0.2 <0.1 0.3 <0.1 0.2 <0.1 0.3 <0.1
Choroid plexus carcinoma of CNS <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
a
Northern Europe (Iceland, Norway, Sweden), Central Europe (Austria, Belgium, France, Germany, the Netherlands, Switzerland), Eastern
Europe (Poland, Slovakia), Southern Europe (Italy, Malta, Slovenia, Portugal, Spain).
 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542 1537

Table 4
Observed prevalence per 100,000 (Prev) with standard errors (SE) by time (2, 5 and 15 years) from diagnosis, with estimated complete prevalence
per 100,000 and estimated total prevalent cases in EU27 of central nervous system (CNS) tumours. Index date 31 December 2003.
Rare central nervous system (CNS) cancers Observed prevalence Estimated prevalence
2 years after 5 years after 15 years after Complete EU27
diagnosis diagnosis diagnosis
Prev. SE Prev. SE Prev. SE Prev. SE No. of cases
Glial tumour of CNS and pineal gland 5.0 0.1 8.3 0.1 14.5 0.2 26.3 0.4 130,764
Astrocytic tumours of CNS 4.1 0.1 6.3 0.1 10.9 0.1 20.4 0.4 101,593
Oligodendroglial tumours of CNS 0.6 <0.1 1.2 <0.1 2.0 0.1 2.7 0.1 13,187
Ependymal tumours of CNS 0.3 <0.1 0.8 <0.1 1.7 0.1 3.8 0.1 19,125
Non-glial tumour of CNS and pineal gland 0.4 <0.1 0.8 <0.1 1.6 0.1 4.7 0.2 23,569
Embryonal tumours of CNS 0.4 <0.1 0.7 <0.1 1.5 <0.1 4.3 0.2 21,470
Choroid plexus carcinoma of CNS <0.1 <0.1 <0.1 <0.1 0.1 <0.1 0.3 0.1 1735

those surviving more than 15 years after diagnosis.
Therefore, the majority of patients alive with a diagnosis
of glial tumours of the CNS are cases diagnosed several
years before (long survivors). A possible explanation for
this is the high frequency of low grade tumours which
have a good prognosis.
As regards ‘Non-glial tumours of the CNS and pineal
gland’, 9%, 17% and 34% of the prevalent cases were
diagnosed within 2, 5 and 15 years before the prevalence
date, respectively. The very long term survivors, those
who survived more than 15 years after diagnosis, repre-
sented the majority of these cases (66% of the total,
about 15,500 subjects).
Astrocytic tumours were the most common among
prevalent rare CNS cancers (about 100,000 cases), fol-
lowed by embryonal tumours (about 21,000), ependy-
mal tumours (about 19,000), oligodendroglial tumours
(about 13,000) and choroid plexus carcinomas (1,700).
Choroid plexus carcinomas, embryonal and astrocytic
tumours were those with the highest proportion of very
long term (more than 15 years) survivors (67%, 63% and
47%, respectively), Table 4.
3.3. Survival
Table 5 shows 1 and 5-year survival of first and sec-
ond tier entities of rare CNS tumours. Survival was bet-
ter for ‘non glial tumours of CNS and pineal gland’ (601
cases analysed), than for ‘glial tumours of CNS’ (13,667
cases). One- and five year survival were 81.0% and
43.8%, and 56.9% and 19.6%, respectively.
Within glial CNS tumours, ependymal tumours had
the best 5-year prognosis (74% overall, 80% low grade
and 37% high grade). Intermediate prognosis was

Table 5
One-year and 5-year relative survival (%) for central nervous system (CNS) tumours with standard errors (SE) by European region for 2000–2002.
Rare central nervous system (CNS) EU overall Regiona
cancers
EU EU Northern Central Eastern Southern UK and
Ireland
1 year 5 years 1 year 5 years 1 year 5 years 1 year 5 years 1 year 5 years 1 year 5 years
(SE) (SE) (SE) (SE) (SE) (SE) (SE) (SE) (SE) (SE) (SE) (SE)
Glial tumour of CNS and pineal gland 44 20 51 23 46 23 46 17 51 19 36 17
(0.5) (0.4) (1.1) (0.9) (0.9) (0.8) (1.7) (1.2) (1.2) (0.9) (0.7) (0.6)
Ependymal tumours of CNS 90 74 94 79 89 69 82 66 89 75 91 78
(1.4) (2.1) (2.5) (4.2) (2.7) (4.2) (6.9) (8.3) (3.7) (5.1) (2.7) (4.1)
Oligodendroglial tumours of CNS 83 55 87 63 82 60 69 34 86 54 81 51
(1.4) (1.8) (2.7) (0.4) (2.9) (3.6) (6.6) (5.8) (3.2) (4.3) (2.3) (3.2)
Astrocytic tumours of CNS 39 15 46 16 42 17 43 14 46 13 31 12
(0.5) (0.3) (1.2) (0.9) (1.0) (0.8) (1.8) (1.2) (1.3) (0.9) (0.7) (0.6)
Non-glial tumour of CNS and pineal 81 57 80 48 87 68 78 50 79 55 79 56
gland (1.7) (2.2) (3.7) (4.7) (2.8) (3.8) (6.3) (7.9) (4.5) (5.5) (3.3) (4.4)

Choroid plexus carcinoma of CNS 78 63 100 50 86 50 50 100 100 57 38
(8.7) (10.7) (25.1) (13.2) (35.4) (35.4) (18.2) (20.0)
Embryonal tumours of CNS 81 57 79 48 87 67 79 49 78 54 80 56
(1.7) (2.2) (3.8) (4.8) (2.9) (3.8) (6.3) (8.2) (4.6) (5.5) (3.3) (4.5)

Statistic could not be calculated.
a
Northern Europe (Iceland, Norway, Sweden), Central Europe (Austria, Belgium, France, Germany, the Netherlands, Switzerland), Eastern
Europe (Poland, Slovakia), Southern Europe (Italy, Malta, Slovenia, Portugal, Spain).
1538 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542
reported for oligodendroglial tumours (55% overall;
65% low grade and 30% high grade) and low survival
was estimated for astrocytic tumours (15%). Astrocytic
tumours include aggressive-phenotype tumours as glio-
blastoma (ICD-O 9440) but also lower grade astrocyto-
mas (ICD-O 9400) and gliomas with a relatively good
prognosis. According, the 5-year survival varied from
4.9% for high grade to 43% for low grade tumours.
Within ‘non glial tumours’, a good prognosis was
estimated for choroid plexus carcinomas (63%) and an
intermediate survival was observed for embryonal
tumours (57%).
For glial tumours five-year survival was slightly
higher for women (20.7%; 95% confidence interval
19.6–21.9) than for men (18.7%; 95% CI 17.8–19.7)
(data not shown). Overall, 5-year survival was quite
good for children and adolescents (0–19 years; 58.1%),
and for young adults (20–39 years; 52.8.0%), and
decreased steadily for adults (40–59 years; 19.1%), and
especially for older subjects (60+ years; 4.4%) (data
not shown). The decreasing survival with increasing
age was a common pattern for all rare CNS types with
the exception of ependymal tumours for which the
decrease was less pronounced than for other tumours.
The overall EU 5-year survival for glial tumours was
19.6%, with the highest values in northern (22.8%) and
central Europe (22.6%), intermediate value in southern
Europe (19.2%) and the lowest survival rates in eastern
Europe (17.2%) and in UK and Ireland (16.6%). The
overall geographical difference was mainly driven by dif-
ferences for astrocytic tumours with less evident gradi-
ent for the other rare CNS types.
There was geographic variation in 1-year survival
from astrocytic tumours, ranging from 31% in UK
and Ireland, to 46% in northern and southern Europe
(data not shown).
3.4. Estimated proportion of cured patients
The proportion of patients that did not show an
excess mortality compared with that for the general pop-
ulation (‘cured patients’) varied widely among the differ-
ent groups of rare CNS tumours. The highest
proportion of cured, 40.8%, was reported for ‘non glial
tumour of CNS and pineal gland’; the lowest (7.9%) for
‘glial tumours of the CNS’, mainly due to the very poor
survival of astrocytic tumours that are the most com-
mon glial tumours (and with high grade astrocytic which
represent 62% of all astrocytic tumours).
4. Discussion
The present study shows that rare CNS cancers rep-
resent a vast array of tumours with different biological
characteristics and different epidemiological features.
Although all rare CNS tumours belong to rare tumours
according to their relative frequency (overall < 6/
100,000), astrocytic tumours represent the majority of
them accounting for about 86% of all the 27,726 CNS
tumours annually estimated in EU and for about 90%
of the 23,859 ‘glial tumours of CNS’. Therefore, the
overall epidemiological figures of rare CNS tumours in
Europe are greatly influenced by astrocytic tumours,15
especially by high grade astrocytic that are the most fre-
quent (62% of astrocytic tumours) and have the worse
survival. Astrocytic tumours influence also the overall
prevalence of rare CNS tumours. In fact, due to the
poor survival of astrocytic (high grade) tumours, their
relevance is slightly lower for complete prevalence than
for incidence. Astrocytic tumours represent 78% of the
154,000 EU complete prevalent cases, and ‘glial
tumours’ in total represent 85%.
Incidence of astrocytic tumours is higher among men
than women. The 1.4 male excess observed in Europe
has also been documented in the US, both in the
SEER16 and in the Central Brain Tumour Registry of
the United States (http://www.cbtrus.org/).
Overall, the median age estimated at diagnosis for
CNS tumours is 56 (55 for males and 57 for females)
years in the United States16 and 53 (53 for males and
54 for females) years in Europe.
The overall incidence increases with age with the
highest values among subjects of 60+ years. This overall
result is mainly driven by astrocytic tumours, while
embryonal tumours have the highest incidence rate in
the age-group 0–19, oligodendroglial tumours in the
age-group 40–59 and ependymal tumours have stable
rates. This confirms that, although rare CNS incidence
increases with age, some of the CNS tumours are more
common in younger adults. During childhood CNS
tumours are the most common type of solid tumour
accounting for one-fifth to one-fourth of all the
cancers.3
The incidence of CNS cancers varies widely across
the world with the highest rates in developed Western
(Europe and North America) and Western-type (Aus-
tralia/New Zealand) countries.2 Also across Europe
there is certain variability in age-standardised incidence
rates—the highest values were evidenced in UK, Ire-
land and northern Europe and the lowest ones in East-
ern Europe. The geographical differences worldwide
may be due, at least in part, to the availability of
highly advanced imaging technology,17 and this may
be true, to some extent, also across Europe. The use
of magnetic resonance imaging (MRI) may also lead
to the diagnosis of unexpected lesions. In a recent sys-
tematic review and meta-analysis of 16 studies, the
prevalence of neoplastic incidental brain findings was
0.70%.18 The prevalence of neoplastic lesions
significantly increased with age and it was more likely
using high resolution MRI sequences than standard
resolution sequences.18 However, survival data do not
 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542
support a relevant amount of indolent lesions to
explain the differences in Europe, and especially the
high incidence rates in UK and Ireland.4
As regards incidence of CNS tumours although the
aetiology of such tumours is largely unknown there is
a major concern for the possible role of cellular phone.
The studies published up to the beginning of 2011 pro-
vided conflicting results thus are not useful to interpret
geographical differences in incidence of CNS across
EU regions. Several studies, including the recent Inter-
phone study, did not demonstrate an increased risk
within approximately 10 years of ‘reasonable’ use of
phones for any tumour of the brain or any other head
tumour.19,20 However, other studies suggested higher
risk for malignant brain tumours in people with ipsilat-
eral mobile phone use and >10 years latency period.21
The issue has been recently addressed by a working
group coordinated by the International Agency for
Research on Cancer specifically aimed at a IARC
Monograph on the potential carcinogenic hazards from
the exposure to radiofrequency electromagnetic fields
frequency (range 30kHz-300GHz). According to the
critical revision of the available evidences IARC classi-
fied radiofrequency electromagnetic fields as possibly
carcinogenic to humans (Group 2B). The Monograph
is in press,22 a concise summary with the main conclu-
sions has been recently published.23
As regards time trends, data from the present study
show stable (or even decreasing) incidence trends in all
the age-classes analysed with the exception of the age-
group 60+ years which shows an increasing trend.
The increased use of technology imaging devices, such
as computed tomography (CT) and MRI, may have con-
tributed to the more precise diagnosis of such tumours.
Trends for CNS with and without microscopic confirma-
tion showed an increase in incidence for microscopic ver-
ified cases especially among the elderly. The improved
ability to diagnose brain tumours by stereotactic and
frameless biopsy procedures may have contributed to
the increase in microscopically verified cases.24,25
The present study showed very different survival
results for different rare CNS tumours with the worst
prognosis for astrocytic tumours (14.5% survival at
5-years, 5% for high grade astrocytic tumours),
moderately good values for embryonal (56.8%), oligo-
dendroglial (54.5%) and choroid plexus carcinomas
(42.6%) and the best survival for ependymal tumours
(74.2%). The overall CNS value is strongly influenced
by the frequency of astrocytic tumours. Recent
European estimates – that included all the CNS histolo-
gies – showed an overall 5-year survival of 19.7% for
brain tumours.4
This datum was quite similar across Europe with val-
ues higher than the European average for Finland
(26.8%), Portugal (24.5%) UK-Wales (23.8%) and
Norway (23.5%) and values below the average for
1539
UK-England (17.6%) and UK-Northern Ireland
(15.3%).4 Low survival values were measured in UK
and Ireland also in the present study. In a recent paper,
the survival differences between European regions were
mainly attributed to the case-mix (morphological type
of tumours). However, also after the adjustment for
age and morphology there were an excess risk of death
of 10% in Southern Europe, 20% in Central Europe,
30% in Eastern Europe and 40% in UK and Ireland in
comparison with Northern Countries.5
The overall geographical difference was mainly driven
by differences for astrocytic tumours with less evident
gradient for the other rare CNS types. Astrocytic
tumours include aggressive-phenotype tumours as glio-
blastoma but also tumours with a relatively good prog-
nosis. In our series glioblastomas were less frequent in
Eastern (38%) than in the other European regions
(around 50%). This suggests that other prognostic fac-
tors than case mix of the neoplasm contributed to the
survival variation across Europe.
Similar geographical variation in survival was found
for adults with any type of brain tumour in EURO-
CARE-4.4 Survival was quite similar across Europe
for children and young adults, though with lower values
in UK.6
Survival was slightly better for women than for men,
as elsewhere documented2,26 and the strong tendency for
survival rates to worsen with age observed in the present
study is consistent with previous findings from survival
time-trend analyses in Nordic countries2
The geographical differences in 1-year survival may
be due to a variation in the timeliness of diagnosis and
therapy which may exert its effect in the short term.
Brain cancer 1-year survival in 23 European countries
ranged between 34.2% and 48.3%, with high survival
estimates in Switzerland, France, Sweden, Belgium and
Italy and lower estimates in Poland, Czech Republic,
Ireland, Denmark and United Kingdom–Northern Ire-
land.4 Five-year survival conditional on having survived
the first year after diagnosis varied less between coun-
tries.4 Also in the present study there were values lower
than the EU average for UK and Ireland for 1-year sur-
vival for astrocytic tumours.
In the treatment of brain cancer, radiotherapy is
widely used. Therefore, the access to radiotherapy might
influence the treatment outcome and consequently sur-
vival of CNS tumours. The number of linear accelera-
tors per million population is far lower in UK and
Ireland, as well as in Eastern Europe compared to Nor-
dic and Central-Europe countries. ESTRO QUARTS
project has analysed the ratio of actual number of
megavoltage therapy units to the evidence-based
required numbers in 13 European countries.27 The
largest gap was seen for Eastern-European countries
(Slovenia and Poland) followed by the Czech Republic
and UK. In contrast, in Northern and Central European
1540 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542
countries (Sweden, France and Belgium), the availability
of megavoltage therapy units exceeded 90% of the
QUARTS estimate.
The lower survival rates reported in this study for
some EU regions could be explained by less accessibility
to Magnetic Resonance Imaging (MRI). EUROSTAT
data on medical technology (http://epp.euro-
stat.ec.europa.eu/) showed for the year 2000 a huge var-
iability of the number of MRI units among EU
countries: from 0.0 per 100,000 persons in Lithuania
and 0.1 in Slovakia to 0.8 in Italy. Different availability
of surgical, radio-therapeutical and chemotherapeutical
facilities across Europe can also contribute to explain
the lower survival observed in some countries. Further
data from EUROSTAT showed that in the year 2000,
the number of operations on the nervous system (not
only for cancer) varied from 69 in Bulgaria and 74 in
Romania to 1268 per 100,000 inhabitants in Belgium.
Evidence based medicine is, by definition, weaker in rare
tumours: this results, in CNS tumours, in a reduced util-
isation of more aggressive, still not of high-evidence lev-
els, treatments such as total resection in low grade
gliomas achievable by more modern techniques (i.e.
awake surgery) or less intense chemotherapeutic
approaches in malignant tumours, such as adjuvant
temozolomide longer than the 6 conventional cycles,
more than one salvage treatment or the use of Gliadel
wafer at recurrence.28,29
In the present study pilocytic astrocytoma were
excluded from the analyses. We do not know whether
these tumours are represented in the same proportions
in the glioma NOS and astrocytoma NOS tiers across
Countries. This is especially important for the younger
age groups. In EUROCARE-4, pilocytic astrocytoma
accounted for 49% of all astrocytomas in children aged
0–14 and 24% of astrocytomas in adolescents and young
adults aged 15–24 years; inclusion of pilocytic astrocy-
toma increased the 5-year period survival estimates for
astrocytoma from 63% to 78% in children and from
56% to 64% in adolescents and young adults.4
Prevalence estimates are a useful tool for public
health strategies and evaluations, as they measure the
burden on the health care system.30 Prevalence of CNS
tumours was about 30 per 100,000, below the threshold
of the European definition for rare diseases (50 per
100,000) therefore these tumours are rare according also
to the European definition on rare diseases31 and may
profit to the EU Directive on orphan drugs. Prevalence
estimates showed that there are about 154,000 citizens
living with rare CNS tumours in EU, 27,000 of them
diagnosed within 2 years and therefore in a period with
very intensive follow-up. As with most cancers, mortal-
ity is highest for CNS tumours in the short term. The
ratio between complete prevalence and estimated annual
number of diagnosed cases is around 4 for astrocytic
tumours, 8 for olygodendroglial tumours, 20 for
ependymal and embryonal tumours and almost 60 for
choroid plexus carcinomas.
Rare CNS tumours represent a heterogeneous group
with great differences in terms of frequency, prognosis
and treatment approaches. Despite remarkable
improvement in the ability to diagnose and treat brain
tumours, the prognosis for most of rare CNS cancer
patients remains poor. The varying availability of highly
advanced imaging technology and of radiotherapy
equipment as well as the lack of evidence based proto-
cols reduces the possibility of getting timely diagnosis
and treatment. This means that management of these
rare brain tumours has to be considered hyper-specialis-
tic, and the treatment has to be centralised in large vol-
umes hub centres with recognised expertise in this field.
Funding
This research was supported by the European Com-
mission through the Executive Agency for Health and
Consumers (Grant No. 2006113), and the Programma
Italia-USA Malattie Rare (Grant No. 526D/42).
Conflict of interest statement
None declared.
Appendix A. The RARECARE Working Group for this
paper consists of:
Austria: N. Zielonk (Austrian National Cancer Reg-
istry); Belgium: E. Van Eycken (Belgian Cancer Regis-
try), H. Sundseth, (European Cancer Patient
Coalition), S. Marreaud (European Organisation for
Research and Treatment of Cancer), R. Audisio (Euro-
pean Society of Surgical Oncology); France: G. Hedelin,
(Bas-Rhin Cancer Registry); G. Launoy (Calvados
Digestive Cancer Registry); A.V. Guizard (Calvados
General Cancer Registry); A.M. Bouvier (Côte d’Or
Digestive Cancer Registry); A.S. Woronoff (Doubs Can-
cer Registry); A. Buemi (Haut-Rhin Cancer Registry);
B. Tretarre (Hérault Cancer Registry); M. Colonna
(Isère Cancer Registry); S. Bara (Manche Cancer Regis-
try); O. Ganry (Somme Cancer Registry); P. Grosclaude
(Tarn Cancer Registry); Germany: B. Holleczek (Saar-
land Cancer Registry); J. Geissler (C.M.L. Advocates
Network; Iceland: L. Tryggvadottir (Icelandic Cancer
Registry); Ireland: H. Comber (National Cancer Regis-
try of Ireland); Italy: F. Bellù (Alto Adige Cancer Reg-
istry); S. Ferretti (Ferrara Cancer Registry); D. Serraino
(Friuli Venezia Giulia Cancer Registry); M. Vercelli
(Liguria Cancer Registry c/o IST/UNIGE, Genoa); S.
Vitarelli (Macerata Province Cancer Registry); M. Fede-
rico (Modena Cancer Registry); M. Fusco (Napoli Can-
cer Registry); A. Traina (Palermo Breast Cancer
Registry); M. Michiara (Parma Cancer Registry);
 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542
A. Giacomin (Piedmont Cancer Registry, Province of
Biella); R. Tumino (Cancer Registry and Histopathol-
ogy Unit, “M.P. Arezzo” Civic Hospital, Ragusa); L.
Mangone (Department of Research Azienda Ospedali-
era Arcispedale Santa Maria Nuova – IRCCS, Reggio
Emilia); F. Falcini (Romagna Cancer Registry); A. Ian-
nelli (Salerno Cancer Registry); M. Budroni (Sassari
Cancer Registry); S. Piffer (Trento Cancer Registry);
T. Intrieri (Tuscan Cancer Registry); F. La Rosa
(Umbria Cancer Registry); P. Contiero (Varese Cancer
Registry); P. Zambon (Veneto Cancer Registry); P.G.
Casali, G. Gatta, A. Gronchi, L. Licitra, M. Ruzza, S.
Sowe, (Fondazione IRCCS Istituto Nazionale dei
Tumori); R. Capocaccia, R. De Angelis, S. Mallone,
A. Tavilla (Centro Nazionale di Epidemiologia, Istituto
Superiore di Sanità); A.P. Dei Tos (Local Health Unit
No. 9, Region of Veneto); A.A. Brandes (Medical
Oncology Department, Local Health Unit, Bologna);
Malta: K. England (Malta National Cancer Registry);
Norway: G. Ursin (Cancer Registry of Norway); Nether-
lands: O. Visser, R. Otter, S. Siesling, J.M. van der Zwan
(Comprehensive Cancer Centre the Netherlands);
J.W.W. Coebergh (Eindhoven Cancer Registry), H.
Schouten (University of Maastricht); Poland: J. Rach-
tan (Cracow Cancer Registry); S. Gozdz (Kielce Cancer
Registry); M. Zwierko (Warsaw Cancer Registry); M.
Bielska-Lasota (National Institute of Public Health –
National Institute of Hygiene, Warsaw); J. Slowinski
(Department of Neurosurgery in Sosnowiec, Medical
University of Silesia); Portugal: A. Miranda (Southern
Portugal Cancer Registry); Slovakia: Ch. Safaei Diba
(National Cancer Registry of Slovakia); Slovenia: M.
Primic-Zakelj (Cancer Registry of Slovenia); Spain: A.
Mateos (Albacete Cancer Registry); I. Izarzugaza (Bas-
que Country Cancer Registry); A. Torrella-Ramos
(Castillon Cancer Registry); R. Marcos-Gragera (Epide-
miology Unit and Girona Cancer Registry, Oncology
Coordination Plan, Department of Health and Catalan
Institute of Oncology, Girona, Spain); M.J. Sánchez
(Granada Cancer Registry); C. Navarro (Department
of Epidemiology, Murcia Regional Health Authority,
Murcia, CIBER Epidemiologı́a y Salud Pública (CIB-
ERESP)); Eva Ardanaz (Navarra Cancer Registry); J.
Galceran (Tarragona Cancer Registry); C. Martinez-
Garcia, J.M. Melchor (Escuela Andaluza de Salud Púb-
lica); Sweden: J. Adolfsson (Stockholm-Gotland Cancer
Registry); M. Lambe (Uppsala Regional Cancer Regis-
try), T.R. Möller (Lund University Hospital); U. Ring-
borg (Karolinska Institute); Switzerland: G. Jundt
(Basel Cancer Registry); M. Usel (Geneva Cancer Reg-
istry); S.M. Ess (St. Gallen Cancer Registry); A. Bor-
doni (Ticino Cancer Registry); I. Konzelmann (Valais
Cancer Registry); J.M. Lutz (National Institute for Can-
cer Epidemiology and Registration); UK-England: D.C.
Greenberg (Eastern Cancer Registration and Informa-
tion Centre); J. Wilkinson (Northern and Yorkshire
1541
Cancer Registry); M. Roche (Oxford Cancer Intelli-
gence Unit); J. Verne (South-West Public Health Obser-
vatory); D. Meechan (Trent Cancer Registry); G.
Lawrence (West-Midlands Cancer Intelligence Unit);
M.P. Coleman (London School of Hygiene and Tropical
Medicine), J. Mackay (University College of London);
UK-Northern Ireland: A. Gavin (Northern Ireland Can-
cer Registry); UK-Scotland: D.H. Brewster (Scottish
Cancer Registry); I. Kunkler (University of Edinburgh);
UK-Wales: J. Steward (Welsh Cancer Intelligence &
Surveillance Unit).
References
1. Wrensch M, Minn Y, Chew T, Bondy M, Berger MS. Epidemi-
ology of primary brain tumors: current concepts and review of the
literature. Neuro Oncol 2002;4:278–99.
2. Ferlay J, Shin HR, Bray F, et al. GLOBOCAN 2008, Cancer
Incidence and Mortality Worldwide: IARC CancerBase No. 10
[Internet]. Lyon, France: International Agency for Research on
Cancer; 2010. Available from: http://globocan.iarc.fr.
3. Kaatsch P. Epidemiology of childhood cancer. Cancer Treat Rev
2010;36:277–85.
4. Sant M, Allemanni C, Santaquilani M, et al. EUROCARE-4.
Survival of cancer patients diagnosed in 1995–1999. Results and
commentary. Eur J Cancer 2009;45:931–91.
5. Sant M, Minicozzi P, Lagorio S, et al. Survival of European
patients with central nervous system tumours. Int J Cancer 2011;
doi:10.1002/ijc.26335. [Epub ahead of print].
6. Gatta G, Zigon G, Capocaccia R, et al. Survival of European
children and young adults with cancer diagnosed 1995–2002. Eur J
Cancer 2009;45:992–1005.
7. World Health Organization. International Classification of Dis-
eases. 10th revision, Geneva; 1992.
8. Percy C, Shanmugaratnam K, Whelan S, et al. International
Classification of Diseases for Oncology (ICD-O), 3rd ed. Geneva,
Switzerland: World Health Organization; 2000.
9. Kros JM, Gorlia T, Kouwenhoven MC, et al. Panel review of
anaplastic oligodendroglioma from European Organization For
Research and Treatment of Cancer Trial 26951: assessment of
consensus in diagnosis, influence of 1p/19q loss, and correlations
with outcome. J Neuropathol Exp Neurol 2007;66:545–51.
10. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO
classification of tumours of the central nervous system. Acta
Neuropathol 2007;114:97–109.
11. Capocaccia R, Colonna M, Corazziari I, et al. Measuring cancer
prevalence in Europe: the EUROPREVAL Project. Ann Oncol
2002;13:831–9.
12. Capocaccia R, De Angelis R. Estimating the completeness of
prevalence based on cancer registry data. Stat Med
1997;16:425–40.
13. Brenner H, Söderman B, Hakulinen T. Use of period analysis for
providing more up-to-date estimates of long-term survival rates:
empirical evaluation among 370,000 cancers patients in Finland.
Int J Epidemiol 2002;31:456–62.
14. Francisci S, Capocaccia R, Grande E, et al. The cure of cancer: a
European perspective. Eur J Cancer 2009;45:1067–79.
15. Schwartzbaum JA, Fisher JL, Aldape KD, Wrensch M. Epidemi-
ology and molecular pathology of glioma. Nat Clin Pract Neurol
2006;2:494–503.
16. Altekruse SF, Kosary CL, Krapcho M, et al., editors. SEER
Cancer Statistics Review, 1975–2007, Bethesda, MD: National
Cancer Institute. Available from: http://seer.cancer.gov/csr/
1975_2007/, based on November 2009 SEER data submission,
posted to the SEER web site, 2010.
1542 E. Crocetti et al. / European Journal of Cancer 48 (2012) 1532–1542
17. Parkin DM, Bray F, Ferlay J, Pisani P. Global Cancer Statistics,
2002. CA Cancer. J Clin 2005;55:74–108.
18. Morris Z, Whiteley WN, Longstreth Jr WT, et al. Incidental
findings on brain magnetic resonance imaging: systematic review
and meta-analysis. BMJ 2009;339:b3016.
19. Ahlbom A, Feychting M, Green A, et al. Standing
committee on epidemiology. epidemiologic evidence on
mobile phones and tumor risk. A review. Epidemiology
2009;20:639–52.
20. INTERPHONE Study Group. Brain tumour risk in relation to
mobile telephone use: results of the INTERPHONE international
case-control study.Int J Epidemiol 2010;39:675–94.
21. Khurana VG, Teo C, Kundi M, Hardell L, Carlberg M. Cell
phones and brain tumors: a review including the long-term
epidemiologic data. Surgical Neurol 2009;72:205–15.
22. IARC. IARC monograph on the evaluation of carcinogenic risks
to humans. In: Non-Ionizing radiation, radiofrequency electro-
magnetic fields, vol. 102. Lyon: International Agency for Research
on cancer, in press.
23. Baana R, Grosse Y, Lauby-Secretan B, et al. Carcinogenicity of
radiofrequency electromagnetic fields. The Lancet Oncol
2011;12:624–6 [published Online: 22 June 2011].
24. Dammers R, Schouten JW, Haitsma IK, et al. Towards improving
the safety and diagnostic yield of stereotactic biopsy in a single
centre. Acta Neurochir . doi: 10.1007/s00701-010-0752-0.
25. Hoffman S, Propp JM, McCarthy BJ. Temporal trends in
incidence of primary brain tumors in the United States, 1985–
1991. Neuro-Oncol 2006;8:27–37.
26. Micheli A, Ciampichini R, Oberaigner W, et al. The advantage of
women in cancer survival: An analysis of EUROCARE-4 data.
Eur J Cancer 2009;45:1017–27.
27. Bentzen SM, Heeren G, Cottier B, et al. Towards evidence-based
guidelines for radiotherapy infrastructure and staffing needs in
Europe: the ESTRO QUARTS project. Radiother Oncol
2005;75:355–65.
28. Brem H, Piantadosi S, et al. Placebo-controlled trial of safety and
efficacy of intraoperative controlled delivery by biodegradable
polymers of chemotherapy for recurrent gliomas. The Polymer-
brain Tumor Treatment Group.. Lancet 1995;345:1008–12.
29. Mason WP, Maestro RD, Eisenstat D, et al. Canadian recom-
mendations for the treatment of glioblastoma multiforme. Curr
Oncol 2007;14:110–7.
30. Porter KR, McCarthy BJ, Freels S, Kim Y, Davis FG. Prevalence
estimates for primary brain tumors in the United States by age,
gender, behavior, and histology. Neuro Oncol 2010;12:520–7.
31. European Parliament and Council of the European Communities
(1999). Decision No 1295/1999/EC of the European Parliament
and of the Council of 29 April 1999 adopting a programme of
Community action on rare diseases within the framework for
action in the field of public health (1999 to 2003).