Annals of Oncology Advance Access originally published online on August 20, 2007
Annals of Oncology 2007 18(10):1734-1742; doi:10.1093/annonc/mdm188
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 European Society for Medical Oncology
pediatric malignancies |
Up-to-date monitoring of childhood cancer long-term survival in Europe: central nervous system tumours
1 Division of Clinical Epidemiology and Aging Research, German Cancer Research Center, Heidelberg, Germany
2 German Childhood Cancer Registry, University of Johannes Gutenberg, Mainz, Germany
3 Data Analysis and Interpretation Group, International Agency for Research on Cancer, Lyon, France
4 National Childhood Cancer Registry Spain (RNTI-SEOP) and Instituto López Piñero (CSIC Universitat de València), Valencia, Spain
* Correspondence to: Dr V. Arndt, Division of Clinical Epidemiology and Aging Research, German Cancer Research Center, Bergheimer Strasse 20, D-69115 Heidelberg, Germany. Tel: +49-6221-548144; Fax: +49-6221-548142; E-mail: v.arndt{at}dkfz-heidelberg.de
 |
Abstract |
|---|
 Top Abstract introductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
Background: Tumours of the central nervous system (CNS) account for 15–20% of all malignant childhood tumours in developed countries. Steady improvement of survival of children with CNS tumours has been reported for the past decades. However, these results, obtained by cohort analysis of survival, do not reflect the full extent of recent improvement.
Methods: Using selected registries from the database of the Automated Childhood Cancer Information System (ACCIS), we calculated period survival estimates for the years 1995–99 for children diagnosed with a malignant CNS tumour.
Results: The overall 10-year period survival estimate for the years 1995–99 was 59% for children with all CNS tumours combined, 73% for children with astrocytoma, 53% for children with ependymoma and 45% for children with primitive neuroectodermal tumours. On average, estimates derived by cohort analysis (pertaining to children diagnosed in 1985–89) were around 4% units lower. Region-specific analysis revealed that recent progress was largest in Eastern Europe, where prognosis nevertheless remained lower than in other European regions. In Northern and Southern Europe, 10-year survival remained essentially unchanged.
Conclusion: Although period survival of children with CNS tumours is higher than previously reported cohort survival, their long-term prognosis remains modest compared to other childhood malignancies.
Key words: cancer registries, childhood cancer, CNS tumours, Europe, population-based, prognosis, survival
|
introduction |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
Tumours of the central nervous system (CNS) represent the largest group of solid tumours during childhood in developed countries [1]. They represent approximately 15–20% of all paediatric malignancies and are characterized by a large diversity of morphologic entities. Astrocytomas account for about 40–45% of all CNS malignancies, primitive neuroectodermal tumours (PNET) comprise 20–25%, other gliomas 10–15% and ependymomas an additional 10–12% [1]. The heterogeneity of these tumour groups is reflected in variable prognosis, which emphasizes the need for tailored diagnostic and therapeutic strategies. Five-year population-based survival of European children diagnosed in the 1990s was around 75% for astrocytoma, around 60% for ependymoma, and around 50% for primitive neuroectodermal tumours [2, 3].
Despite the relatively low survival of children with CNS tumours, as compared to the results achieved for some other common childhood cancers, such as lymphomas and leukaemias, steady improvements in prognosis have been observed during the past decades. A significant increase in 5-year survival of children with CNS tumours has been reported for Europe over the last 30 years, from 45% for patients diagnosed in the 1970s to 66% in the 1990s [4]. Improvements have been reported from individual European countries [5–10] but also from the United States of America [11, 12], New Zealand [13] and Japan [14].
In almost all of the cited studies, the 5-year cumulative cohort survival was arbitrarily used for comparison, although a substantial proportion of deaths is known to occur also after the fifth year following the diagnosis with a CNS tumour [15–17]. Long-term survival estimates based on cohort survival analysis require that a cohort of patients is followed over a long period, with a large proportion diagnosed a long time ago. To ensure complete follow-up of the cohort, the results are inevitably produced with a delay. To produce the results faster, the complete method is used, whereby the experience of the most recently diagnosed patients is also considered, even if their follow-up is incomplete. Because of the incomplete follow-up of most recently diagnosed patients, however, their survival experience is not fully reflected in survival of the cohort. If recent survival is on the rise, as it is the case for CNS tumours in children, the cohort- or complete-based estimates of survival will under-estimate the prognosis of patients diagnosed currently. The recently introduced period-analysis methodology was shown to provide more reliable estimates of long-term survival, by giving more weight to the experience of the most recently diagnosed patients with cancer [18–20]. It therefore depicts recent pattern in survival in a more timely way than the cohort approach [21].
In this paper, we present period estimates of survival for up to 10 years following diagnosis among children diagnosed with a malignant CNS tumour in Europe for the 1995–99 period, using the database of the Automated Childhood Cancer Information System (ACCIS) [4]. In addition, we compare the period estimates with the most up-to-date survival estimates derived by cohort-based methods of survival analysis, using the same database.
|
methods |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
Data for the analysis were extracted from the database of the Automated Childhood Cancer Information System (ACCIS), September 2003 edition. Children aged 0–14 years at first diagnosis of a tumour of the central nervous system during 1985–99 in selected cancer registries were included. The criteria for selection of the registries in this analysis were: availability and high completeness of follow-up for vital status throughout a time window from at least 1987 to 1997 or later, and fulfilment of high standards of quality and comparability of the ACCIS data [22]. Data from the contributing cancer registries were grouped into four European regions as follows:
- East: Estonia, Hungary, Slovakia;
- South: Italy, Slovenia, Spain;
- West: France, Germany, Netherlands, Switzerland, United Kingdom;
- North: Denmark, Finland, Iceland, Norway.
Period estimates of 10-year survival for the 1995–99 period (exclusively reflecting the survival experience in 1995–99 of children diagnosed in 1985–99) were estimated according to the method of Brenner and Gefeller [18] for children with CNS tumours resident in the four European regions. The most recent estimates that might have been obtained by cohort analysis (pertaining to children diagnosed in 1985–89 with full 10-year follow-up) or complete analysis (pertaining to the survival experience of children diagnosed in 1985–99 with up to 10-year follow-up by the end of 1999) are given for comparison [25]. The methodology used has been described in full detail elsewhere [26]. For comparison purposes we also report five-year survival for selected groups of patients.
|
results |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
A total of 7747 malignant CNS tumours were included in the analysis after exclusion of 1198 CNS tumours with either benign or uncertain behaviour. Almost half of all cases came from the German registry with nationwide coverage. Other countries with nationwide registration include Denmark, Estonia, Finland, Hungary, Iceland, Norway, Slovakia and Slovenia (see Table 1). Overall, 3433 children with astrocytoma, 2027 children with PNET, 838 children with ependymoma and 1449 children with other or unspecified CNS tumours were included for subtype specific analysis.
|
Table 2 provides a comprehensive comparison of the most recent estimates of 10-year survival derived by the cohort, complete and period analyses. For all regions combined, the period survival estimate for the 1995–99 period was 59% for children with any type of malignant CNS tumour. This estimate was 4% units higher than the cohort survival estimate of 55%, which refers to children diagnosed in 1985–89. The corresponding 5-year survival estimate, obtained with the period method, was 65% (data not tabulated). Differences between the 1995–99 period estimates and the 1985–89 cohort estimates were in the range of 3–11% units for all major CNS tumour subtypes.
|
Prognosis differed markedly between the CNS tumour subtypes, irrespective of the method of survival analysis used. The 1995–99 period estimate of 10-year survival of children with astrocytoma was 73%, while that for children with ependymoma was 53% and that for PNET was 45%. The corresponding 5-year survival estimates obtained with the period method were 77, 60 and 54% for children with astrocytoma, ependymoma, and PNET, respectively (data not tabulated).
Complementary information regarding the survival experience of children with malignant CNS tumours during the 10-year period after diagnosis is provided by the survival curves (Figures 1–3). Figure 1 shows the survival curves for children from all European regions by tumour type. Probability of survival decreases rapidly during the first years after diagnosis. The figure also indicates that a substantial proportion of children with CNS tumours still die after the fifth year, especially for PNET and ependymoma.
|
|
Age-specific 10-year survival estimates for the children with CNS tumours from all regions combined are given in Table 2. The corresponding age-specific survival curves are shown in Figure 2. For all tumours combined, but also within each subtype (data not shown), long-term survival was much lower for children who developed CNS tumours during the first year of life in comparison with older children. Similar differences between the age groups were obtained using the cohort or complete estimates. The results of the period analysis, as compared to those obtained by cohort or complete method, suggest that prognosis strongly improved in the 1990s for the age group 1–4 years, whereas such improvement for all other age groups was less distinct. As can be seen from Figure 2, the 5-year period survival estimate for children under 1 year was about 41% versus 63–66% for older children.
|
With respect to gender, 10-year survival estimates obtained with the cohort analysis were slightly higher for boys than for girls but the differences between the sexes were smaller according to the estimates derived by period and complete analyses. The comparison of the different methods indicates that the sex difference has diminished during the 1990s (Table 2).
In the region-specific comparison (Table 2, Figure 3) the highest survival for children with all forms of malignant CNS tumours was observed in Northern Europe, whereas children from Eastern Europe experienced the lowest survival. According to period analysis, 10-year (5-year) survival for all forms of malignant CNS tumours combined was 50% (56%) in Eastern Europe, 56% (63%) in Southern Europe, 61% (67%) in Western Europe and 63% (66%) in Northern Europe. This pattern was observed regardless of the analysis method used. However, the differences in survival between Eastern Europe and the other European regions was smallest for the results obtained by the period method, which indicates an increasing rate of improvement in the Eastern Europe towards the end of the study period. The difference between the curves produced by the different methods of survival analysis indicate further that the prognosis for children with malignant CNS tumours has also improved in the 1990s in Western Europe, although to a lesser extent. In contrast, lack of difference between the curves for Northern Europe point to no further improvement. The reverse position of the three curves seen for Southern Europe may suggest a stabilization of the prognosis, or even its slight deterioration. Actually, for children with CNS tumours in Northern and Southern Europe, 10-year survival estimates obtained with the cohort analysis were 1–4% units higher than those obtained by the period method, respectively, but these differences have to be interpreted with caution given the size of the respective standard errors (Table 2). Notwithstanding the precision of the estimates, the higher cohort estimate, as compared to the period estimate, would predict a stabilization of survival in the cancer patients diagnosed most recently. Having said that, the survival estimates remain highest in Northern Europe.
|
discussion |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
This is the first population-based survival analysis of children with tumours of the central nervous system for various European regions using the period methodology. Survival of children has steadily improved during the past and our study indicates further progress in the prognosis of children with CNS tumours in the 1990s. However, the long-term survival remains modest, specifically for children with ependymoma and PNET. About 50% of them die during the first decade after diagnosis, the probability of death being strongest in the first 5 years after diagnosis. In contrast, children with astrocytoma have a much better prognosis and at present over 70% are expected to survive 10 years or more.
This study is complementary to the results of other European studies of population-based survival, using cohort-based methods, such as the EUROCARE studies [2, 27, 28] and the recent ACCIS report on incidence and survival of childhood CNS tumours in Europe (children diagnosed in 1978–97) [3]. In the ACCIS study, 5-year survival for children diagnosed in 1988–97 was 75% for astrocytoma, 53% for ependymoma and 49% for PNET, while the overall survival for all CNS tumours combined was 61%, when only the malignant tumours were included [3]. Survival probabilities derived using our period analysis of survival for the period 1995–99 are higher, revealing thus further improvement during the second half of the 1990s, especially for children with ependymoma and PNET. Despite slightly different design and inclusion criteria between the cited studies, the 5-year survival observed in our study is concordant with the previous findings [2, 3, 27, 28]. Thus, period analysis provides the most updated available information on long-term survival without altering the general pattern of survival by diagnostic group, geographical area or demographic details.
The observed improvement of survival in children with cancer is attributed to a combination of better diagnostic procedures, more effective drugs, greater intensity of treatment, more recruitment to standardized collaborative protocols and more effective supportive therapy [2, 3, 11, 29, 30]. However, other factors are also important for interpretation of population-based data, such as establishment of paediatric-oncology units, improved access to specialized care and improvement of socio-economic conditions. Finally, differences in classification of tumours or in registration practices may be responsible for a part of the observed variation [3].
Our study indicates that the recent progress was seen in the countries in Eastern Europe, much less so in the West and not at all in Northern and Southern Europe. In consequence, differences in prognosis between Eastern Europe and the other European regions diminished to some degree. However, Eastern Europe has remained the region with the poorest survival estimates for all types of CNS tumours throughout the study period, and Northern Europe remained the region with the highest survival. Low survival rates for children in Eastern Europe have also been reported elsewhere [2–4, 9, 27, 28]. The reasons for lower survival are likely to be linked to the socio-economic conditions including, in some countries, less efficient access to health care, delayed introduction of modern imaging procedures and delayed adherence to effective treatment protocols [27, 31]. The recent political changes in this part of Europe will undoubtedly contribute to changes in the pattern of survival, as can be judged from the differences in survival already seen between the countries grouped within this region [26].
Discrepancies in survival by age among children with CNS tumours have also been described [3, 5, 16, 27, 32, 33]. The lower survival of infants with brain tumours may be due to diagnostic difficulties, followed by lack of appropriate treatment [16]. The use of protocols avoiding or delaying radiotherapy to prevent late effects might have contributed to the low survival in infants, especially with PNET or ependymoma [34]. The better prognosis for older children has been linked to a more favourable histology and tumour grade and, for the oldest children, also with a more common supratentorial location of the tumour, which facilitates its total removal [5, 35, 36].
There are some limitations that should be addressed when interpreting the results of our study. First, the number of patients was limited for conducting region-specific analysis of survival for the various tumour subgroups, as the estimates turned out to be very imprecise. We could not therefore evaluate whether differences between the regions were similar for all subgroups. In the cohort analyses of the same database, the ranking of the regions varied slightly between the subgroups [3]. Second limiting factor was the imbalanced coverage of the childhood European population. The overall results were therefore strongly influenced by survival of children observed in the German Childhood Cancer Registry, which provided almost half of all cases for this analysis. Other limits were set by the lack of data on clinical prognostic factors, such as tumour location, extent of tumour resection and application of adjuvant radio- and chemotherapy [37], which are not available within the ACCIS database [22]. Beside treatment protocols, the population-based survival estimates may also reflect, for example, ease and rapidity of access to diagnosis or well-designed referral systems. It would be desirable in the future to include prognostic factors in the data-collection protocol, such as diagnostic delay, inclusion in clinical trials, stage at diagnosis and possibly the initial treatment, even though this information may be difficult to obtain systematically. However, a large proportion of children with childhood cancer in the contributing European regions was reportedly included in clinical trials [3, 5]. Finally, our results rely solely on malignant CNS tumours, which represent over 85% of all CNS tumours, and this may slightly reduce the comparability with other studies including all CNS tumours. In an earlier cohort-based survival analysis, exclusion of non-malignant tumours reduced the 5-year survival by about 2–3% points in Europe and its regions [3]. Our period survival estimates are thus also slightly lower, overall and for some diagnostic groups (namely ependymoma and unspecified CNS tumours), than if non-malignant tumours were not excluded. Nevertheless, our results are consistent with those in other studies using the cohort-based method [2, 3, 28] and may be therefore equally well applicable to the European childhood population in general as those from cohort analyses.
At present, our study provides the most up-to date and presumably most reliable survival estimates for children diagnosed with tumours of the central nervous system in Europe. Some of our results may indicate a stabilization of the survival improvement in the most advanced European regions in the recent period. This represents a challenge for further clinical research, since there are still 40% of children with CNS tumours, who are dying within a decade of their diagnosis.
|
Acknowledgements |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
This project was funded by the European Commission from the Europe Against Cancer programme (contract SPC.2002303), jointly with the International Agency for Research on Cancer. ESF was partly supported by the Cancéropôle Lyon Auvergne Rhône Alpes and the Federal Ministry of Health of the Federal German Government. RPB was partly supported by ISCIII Network G03/089, Villavecchia Foundation and AECC Scientific Foundation, Spain. We thank Dr Risto Sankila from the Finnish Cancer Registry, Helsinki, for helpful discussions and comments on the manuscript. The authors are grateful to the following cancer registries participating in ACCIS for their contribution to this study: Denmark: Danish Cancer Society; Estonia: Estonian Cancer Registry; Finland: Finnish Cancer Registry; France: Childhood Cancer Registry of Lorraine and Association of Childhood Cancer Registries of the Rhone-Alps Region; Germany: German Childhood Cancer Registry; Hungary: Hungarian Paediatric Cancer Registry; Iceland: Icelandic Cancer Registry; Italy: Cancer Registry of Latina Province, Lombardy Cancer Register, Childhood Cancer Registry of Piedmont, Ragusa Cancer Registry and Tuscany Cancer Registry; Netherlands: Eindhoven Cancer Registry of Comprehensive Cancer Centre South; Norway: Cancer Registry of Norway; Slovakia: National Cancer Registry of Slovakia; Slovenia: Cancer Registry of Slovenia; Spain: Asturias Cancer Registry, Granada Cancer Registry and Tarragona Cancer Registry; Switzerland: Basel Cancer Registry and Geneva Cancer Registry; United Kingdom: Scottish Cancer Registry.
Received for publication November 17, 2006. Revision received April 4, 2007. Accepted for publication April 11, 2007.
|
References |
|---|
|
TopAbstractintroductionmethodsresultsdiscussionAcknowledgementsReferences |
|---|
1. Parkin DM, Kramarova E, Draper GJ, et al. International Incidence of Childhood Cancer. In: IARC Scientific Publication No. 144 (1999) Lyon: International Agency for Research on Cancer.
2. Gatta G, Corazziari I, Magnani C, et al. Childhood cancer survival in Europe. Ann Oncol (2003) 14(Suppl 5):V119–V127.[CrossRef][Medline]
3. Peris-Bonet R, Martínez-García C, Lacour B, et al. Childhood central nervous system tumours—incidence and survival in Europe (1978 –1997): report from the Automated Childhood Cancer Information System project. Eur J Cancer (2006) 42:2064–2080.[CrossRef][Web of Science][Medline]
4. Steliarova-Foucher E, Stiller C, Kaatsch P, et al. Geographical patterns and time trends of cancer incidence and survival among children and adolescents in Europe since the 1970s (the ACCIS project): an epidemiological study. Lancet (2004) 364:2097–2105.[CrossRef][Web of Science][Medline]
5. Kaatsch P, Rickert CH, Kühl J, et al. Population-based epidemiologic data on brain tumors in German children. Cancer (2001) 92:3155–3164.[CrossRef][Web of Science][Medline]
6. Robertson CM, Hawkins MM, Kingston JE. Late deaths and survival after childhood cancer: implications for cure. BMJ (1994) 309:162–166.
7. Stiller CA. Population based survival rates for childhood cancer in Britain, 1980 –91. BMJ (1994) 309:1612–1616.
8. Dama E, Pastore G, Mosso ML, et al. Time trends and prognostic factors for survival from childhood cancer: a report from the Childhood Cancer Registry of Piedmont (Italy). Eur J Pediatr (2006) 165:240–249.[CrossRef][Web of Science][Medline]
9. Kramarova E, Plesko I, Black RJ, Obsitnikova A. Improving survival for childhood cancer in Slovakia. Int J Cancer (1996) 65:594–600.[CrossRef][Web of Science][Medline]
10. de Nully Brown P, Olsen JH, Hertz H, et al. Trends in survival after childhood cancer in Denmark, 1943 -87: a population-based study. Acta Paediatr (1995) 84:316–324.[Web of Science][Medline]
11. Ries LAG, Eisner MP, Kosary CL, et al. SEER Cancer Statistics Review, 1975–2000 (2003) Bethesda, MD: National Cancer Institute.
12. Brenner H. Up-to-date survival curves of children with cancer by period analysis. Br J Cancer (2003) 88:1693–1697.[CrossRef][Web of Science][Medline]
13. Douglas NM, Dockerty JD. Population-based survival of children in New Zealand diagnosed with cancer during 1990 –1993. Eur J Cancer (2005) 41:1604–1609.[CrossRef][Web of Science][Medline]
14. Ajiki W, Tsukuma H, Oshima A. Survival rates of childhood cancer patients in Osaka, Japan. Jpn J Clin Oncol (2004) 34:50–54.
15. Pastore G, Mosso ML, Carnevale F, et al. Survival trends of childhood cancer diagnosed during 1970 –1994 in Piedmont, Italy: a report from the Childhood Cancer Registry. Med Pediatr Oncol (2001) 36:481–488.[CrossRef][Web of Science][Medline]
16. Stiller CA, Bunch KJ. Brain and spinal tumours in children aged under two years: incidence and survival in Britain, 1971 –85. Br J Cancer (1992) 18(Suppl):50–53.
17. Mertens AC, Yasui Y, Neglia JP, et al. Late mortality experience in five-year survivors of childhood and adolescent cancer: the Childhood Cancer Survivor Study. J Clin Oncol (2001) 19:3163–3172.
18. Brenner H, Gefeller O. An alternative approach to monitoring cancer patient survival. Cancer (1996) 78:2004–2010.[CrossRef][Web of Science][Medline]
19. Brenner H, Gefeller O. Deriving more up-to-date estimates of long term patient survival. J Clin Epidemiol (1997) 50:211–216.[CrossRef][Web of Science][Medline]
20. Steliarova-Foucher E, Arndt V, Parkin DM, et al. Up-to-date monitoring of childhood cancer long-term survival in Europe: Empirical evaluation of the period analysis approach Ann Oncol 2007; In Press.
21. Brenner H, Hakulinen T. Advanced detection of time trends in long-term cancer patient survival: experience from 50 years of cancer registration in Finland. Am J Epidemiol (2002) 156:566–577.
22. Steliarova-Foucher E, Kaatsch P, Lacour B, et al. Quality, comparability and methods of analysis of data on childhood cancer in Europe (1978 –1997): report from the Automated Childhood Cancer Information System project. Eur J Cancer (2006) 42:1915–1951.[CrossRef][Web of Science][Medline]
23. Kramarova E, Stiller C. The international classification of childhood cancer. Int J Cancer (1996) 68:759–765.[CrossRef][Web of Science][Medline]
24. Percy C, Van Holten V, Muir CS, eds. International classification of diseases for oncology (1992) 2nd edition. Geneva: World Health Organization.
25. Cutler SJ, Ederer F. Maximum utilization of the life table method in analyzing survival. J Chron Dis (1958) 8:699–712.[CrossRef][Medline]
26. Brenner H, Steliarova-Foucher E, Arndt V. Up-to-date monitoring of childhood cancer long-term survival in Europe: methodology and application to all forms of cancer combined Ann Oncol 2007; In Press.
27. Magnani C, Aareleid T, Viscomi S, et al. Variation in survival of children with central nervous system (CNS) malignancies diagnosed in Europe between 1978 and 1992: the EUROCARE study. Eur J Cancer (2001) 37:711–721.[CrossRef][Web of Science][Medline]
28. Gatta G, Capocaccia R, Stiller C, et al. Childhood cancer survival trends in Europe: a EUROCARE Working Group study. J Clin Oncol (2005) 23:3742–3751.
29. Kühl J. Modern treatment strategies in medulloblastoma. Childs Nerv Syst (1998) 14:2–5.[CrossRef][Web of Science][Medline]
30. Parker RJ. Childhood CNS tumours. Curr Opin Pediatr (1997) 9:551–557.[Medline]
31. Liigant A, Kulla A, Linnamagi U, et al. Survival of patients with primary CNS tumours in Estonia. Eur J Cancer (2001) 37:1895–1903.[CrossRef][Web of Science][Medline]
32. Duffner PK, Cohen ME. Recent developments in pediatric neuro-oncology. Cancer (1986) 58:561–568.[CrossRef][Medline]
33. Farinotti M, Ferrarini M, Solari A, Filippini G. Incidence and survival of childhood CNS tumours in the Region of Lombardy, Italy. Brain (1998) 121:1429–1436.
34. Blaney SM, Kun LE, Hunter J, et al. Tumors of the central nervous system. In: Principles and practice of pediatric oncology—Pizzo PA, Poplack DG, eds. (2006) Philadelphia: Lippincott Williams & Wilkins. 786–864.
35. Rickert CH, Paulus W. Epidemiology of central nervous system tumors in childhood and adolescence based on the new WHO classification. Childs Nerv Syst (2001) 17:503–511.[CrossRef][Web of Science][Medline]
36. Rickert CH, Probst-Cousin S, Gullotta F. Primary intracranial neoplasms of infancy and early childhood. Childs Nerv Syst (1997) 13:507–513.[CrossRef][Web of Science][Medline]
37. Rickert CH, Paulus W. Prognosis-related histomorphological and immunohistochemical markers in central nervous system tumors of childhood and adolescence. Acta Neuropathol (Berl) (2005) 109:69–92.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||