In:
Cancer Research, American Association for Cancer Research (AACR), Vol. 74, No. 19_Supplement ( 2014-10-01), p. 1889-1889
Abstract:
Background The current World Health Organisation (WHO) classification of central nervous system tumors comprises over 100 entities. Most of these are defined by purely histological criteria, with varying and often overlapping spectra. Histological diagnosis is often challenging, however, especially in cases with limited or non-representative biopsy material. Thus, molecular technologies that can complement standard pathology testing have the potential to greatly enhance diagnostic precision and improve clinical decision-making. DNA methylation profiling, acting as a ‘fingerprint’ of cellular origins and molecular alterations, is one such promising technology. Methods We have assembled a reference dataset of more than 2,000 methylation profiles using the Illumina HumanMethylation450 (450k) array, currently representing over 50 brain tumor entities or subgroups. The array platform is suitable for both frozen and paraffin-embedded material, with minimal DNA input required. This reference set is iteratively updated, and a ‘random forest’ algorithm used to produce a classification scheme. Each new diagnostic case receives an entity prediction with an associated probability score. Genome-wide copy number profiles and target gene methylation data (e.g. MGMT) generated from the array provide important additional information. Results In addition to the reference cohort, more than 300 diagnostic samples from Heidelberg University Hospital and external institutions have been processed. Many cases displayed a discrepancy between histological and molecular diagnoses. Careful re-examination of these often resulted in refinement of the original diagnosis, and improved patient care. Furthermore, samples collected for the reference cohort have led to significant improvements in our understanding of the biology of several tumor types, including the identification of further subgroups for several entities and associations with recurrent copy number changes. Conclusion Our understanding of the molecular alterations underlying brain tumors has grown enormously in recent years, and it is crucial that this is translated into the clinic promptly. DNA methylation profiling is one tool with the potential to become an important part of the diagnostic armoury of neuropathologists. This relatively inexpensive and robust method is well suited to complement standard histopathologic techniques and improve diagnostic accuracy, thereby optimising patient management. It offers a diagnostic assessment completely independent of histopathological evaluation, and may thus be especially valuable in histologically uncertain cases. We are currently expanding our pipeline to include additional diagnostic sites, allowing for further refinement and validation as well as broader access across the globe. Citation Format: David T. W. Jones, David Capper, Martin Sill, Volker Hovestadt, Leonille Schweizer, Roger Fischer, Matthias Schick, Melanie Bewerunge-Hudler, Axel Benner, David Zagzag, Peter Lichter, Matthias A. Karajannis, Kenneth D. Aldape, Andrey Korshunov, Andreas von Deimling, Stefan M. Pfister. Next-generation neuropathology - Improving diagnostic accuracy for brain tumors using DNA methylation array-based molecular profiling. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1889. doi:10.1158/1538-7445.AM2014-1889
Type of Medium:
Online Resource
ISSN:
0008-5472
,
1538-7445
DOI:
10.1158/1538-7445.AM2014-1889
Language:
English
Publisher:
American Association for Cancer Research (AACR)
Publication Date:
2014
detail.hit.zdb_id:
2036785-5
detail.hit.zdb_id:
1432-1
Permalink
