17. NOA Tagung 2015 in Heidelberg

17. NOA Tagung 2015 in Heidelberg
17. NOA Tagung, Heidelberg 2015
EINLADUNG ZUR NOA NACH HEIDELBERG
Liebe Kolleginnen, liebe Kollegen,
wir laden Sie im Namen des NOA-Vorstands sehr herzlich zur 17. NOA-Tagung ein.
Am 18. und 19. Juni 2015 werden wir erneut den aktuellen Stand der klinisch
relevanten Grundlagenforschung und der klinischen Forschung bei primären
Hirntumoren diskutieren.
Schwerpunkte der diesjährigen Tagung sind ein Überblick über die molekularen
Biomarker, Gliomtherapien, patientenzentrierte Studienendpunkte und Vorträge mit
Tumorboardelementen. Am 18. Juni werden wir abends in der Kulturbrauerei neben
der Verleihung des Assmus-Preises für Neuroonkologie Gelegenheit zu
weiterführenden Diskussionen haben.
Aufgrund der geänderten Finanzierungsoptionen und aus Compliancegründen
müssen wir einen Kostenbeitrag von 50 € von Ihnen erheben, den Sie bitte mit der
Anmeldung auf das unten angegebene Konto überweisen.
Wir laden zur Einreichung von Abstracts für Posterbeiträge, die während der
gesamten Tagung ausgestellt werden, ein. Abstractautoren müssen keinen
Kostenbeitrag leisten. Außerdem wählen wir aus den Abstracts im NOA-Vorstand
einige freie Vorträge aus.
Zimmer können Sie aus dem reichhaltigen Angebot der Stadt buchen. Wir haben
Zimmerkontingente zu Sonderpreisen im HIP-Hotel (http://www.hip-hotel.de/
Buchungsoption bis 18.04.2015) und im Parkhotel Atlantic (http://parkhotelatlantic.de/
Buchungsoptionen bis 24.04.2015) für Sie reserviert. Bitte geben Sie bei der
Buchung das Stichwort „NOA-Tagung 2015“ an. Des Weiteren stehen in der
Umgebung Unterkünfte in verschiedenen Preiskategorien zur Verfügung. Wir freuen
uns darauf, Sie in Heidelberg begrüßen zu dürfen.
Herzlichst
Ihr
Prof. Dr. Wolfgang Wick
Sprecher der NOA
2
17. NOA Tagung, Heidelberg 2015
17. NOA Tagung 2015
Ort: Molkenkur, Heidelberg, Klingenteichstrasse 31, 69117 Heidelberg
Programm (Version 3, 1.03.2015)
Donnerstag, 18. Juni 2015
09:00 h
Vorstandssitzung (bis 11:00 h)
ab 11:00 h
Akkreditierung und Imbiss
12:00 h
Begrüßung
12.15 h
Aktueller Stand und Neuigkeiten der NOA Studien und der NOAassoziierten Studien
Vorsitz: J.P. Steinbach, Frankfurt & S.E. Combs, München
NOA-04
NOA-07
NOA-09/CeTeG
NOA-10
NOA-11/PDT
NOA-12/NONK-3
NOA-14/Hipporad
NOA-16/NONK-6
W. Wick (Heidelberg)
W. Wick (Heidelberg)
P. Hau (Regensburg)
M. Glas (Bonn)
A. Grosu (Freiburg)
W. Stummer (Münster)
M. Platten (Heidelberg)
A. Grosu (Freiburg)
M. Platten (Heidelberg)
14.00 h
Kaffeepause
14:30 h
Neues aus der Grundlagenwissenschaft
Vorsitz: G. Tabatabai (Tübingen) & P. Vajkoczy (Berlin)
Gliome: eine Netzwerkerkrankung
15 min
F. Winkler
(Heidelberg)
Die hypoxische Nische und „hallmarks of cancer“
15 min
T. Acker
(Giessen)
Heterogenität bei Gliomen
15 min
M. Glas
(Bonn)
Stammzellen bei Gliomen
15 min
G.
Niedermann
(Freiburg)
Vier freie Vorträge à 8 min
ErbB2/HER2-targeted NK cells display potent activity against glioblastoma
and enhance survival
M. Burger (Frankfurt)
3
17. NOA Tagung, Heidelberg 2015
Synergie von Strahlentherapie und Antigen-spezifischer Vakzinierung in der
Behandlung maligner Gliome
K. Ochs (Heidelberg)
ZEB1 is ubiquitously expressed across subtypes in human glioblastoma and a
surrogate marker of tumor purity
P. Euskirchen (Berlin)
Über die parakrinen immunmodulatorischen Effekte des Onkometaboliten 2Hydroxyglutarat im Gliommikromilieu
T. Schumacher (Heidelberg)
16:00 h
Kaffeepause
16:15 h
Fokus: Kontroverse Themen der Neuroonkologie
Vorsitz: J.-C. Tonn & F. Schmidt (München)
17:45 h
Ependymome: Standardtherapie
15 min im TuBo Stil
J. Boström
(Bonn)
Ependymome: Entwicklungen
15 min im TuBo Stil
J.-C. Tonn
(München)
Primärtherapie bei PZNSL
15 min
A. Korfel
(Berlin)
Rezidivtherapie bei PZNSL
15 min
U. Schlegel
(Bochum)
Zerebrale Oligomestatasierung in das ZNS: Relevanz
15 min
W. Stummer
(Münster)
Therapiekonzepte aus Sicht der Radioonkologie
15 min
S. Combs
(München)
Entwicklungen in der Rezidivtherapie von Gliomen
Vorsitz: W. Stummer (Münster) & C. Herold-Mende (Heidelberg)
Rezidivoperation: wer, wann, warum? (Re-Surge)
A. Raabe
(Bern)
Temozolomid, Lomustin, Rebestrahlung, und nun?
Im TuBo Stil
J.P.
Steinbach
(Frankfurt)
18:15 h
Keynote Lecture: Immuntherapie für Gliome
Vorsitz: Wolfgang Wick (Heidelberg)
M. Platten
(Heidelberg)
19:00 h
Mitgliederversammlung der NOA
ab 20:00
Gesellschaftsabend mit Verleihung des Assmus-Preises für
Neuroonkologie
4
17. NOA Tagung, Heidelberg 2015
Freitag, 19. Juni 2015
08:15 h
Studien der EORTC Hirntumorgruppe
08:30 h
Molekulare Neuroonkologie
Vorsitz: C. Hartmann (Hannover) & D. Capper (Heidelberg)
M. Weller
(Zürich)
Grading in der WHO Klassifikation: fact or fiction
15 min
A. von
Deimling
(Heidelberg)
Methylierungsplattform: Neuropathologe 2.0
15 min
D. Jones
(Heidelberg)
Biomarker für Therapieentscheidungen
15 min Im TuBo Stil
M. Weller
(Zürich)
INFORM Register für kindliche Tumoren
10 min
O. Witt
(Heidelberg)
NCT Neuro Master Match (N2M2)
10 min
W. Wick
(Heidelberg)
Drei freie Vorträge à 8 min
ATF4 as a mediator of resistance to targeted therapy in high-grade gliomas
S. Moeckel (Regensburg)
Prognostic value of the extent of resection and IDH1 mutation status in WHO
grade II astrocytomas
C. Jungk (Heidelberg)
Next-Generation-Sequencing in routine neuropathology diagnostics
F. Sahm (Heidelberg)
10:00 h
Kaffeepause
10:30 h
Fokus: Patientenzentrierung – alternative Endpunkte
Vorsitz: A. Wick & A. Unterberg (Heidelberg)
Psychoonkologie
15 min
D. Wiewrodt
(Münster)
Cognitive Function
15 min
M. Klein
(Amsterdam)
Supportivtherapie – Palliation
15 min
J. Rieger
(Frankfurt)
Immunmonitoring für Gliomstudien
10 min
M. Platten
(Heidelberg)
5
17. NOA Tagung, Heidelberg 2015
Alternative Studienendpunkte
10 min
W. Wick
(Heidelberg)
3 freie Vorträge
MRI tumor progression patterns in the GLARIUS trial
S. Kebir (Bonn)
Resection of prerolandic motor areas defined by nTMS motor evoked
potentials correlates with postoperative motor function S.M. Krieg (München)
Supportive care in patients with high-grade gliomas – how to integrate clinical
assessment in daily routine? Results of a prospective multicentric trial
“ERASMUS” in 175 patients.
M. Renonvanz (Mainz)
12:00 h
Bildgebung: Wissenschaft, Surrogat, Klinik
Vorsitz: E. Hattingen (Bonn) & M. Bendszus (Heidelberg)
Internationales MRT Standardprotokoll
M. Bendszus
(Heidelberg)
Advanced Imaging mittels MRT: Bonner Sicht
E. Hattingen
(Bonn)
Advanced Imaging mittels MRT: Heidelberger Sicht
A. Radbruch
(Heidelberg)
PET zur RT Planung und Verlaufskontrolle
A. Grosu
(Freiburg)
PET zur Differenzierung von strahlenbedingten Veränderungen
N. Galldiks,
Köln
13:15 h
Mittagspause
14:00 h
Entwicklungen in der Primärtherapie von Gliomen
Vorsitz: P. Hau (Regensburg) & A. Weyerbrock (Freiburg)
Frühe aggressive Therapie von WHO Grad II Gliomen – pro
15 min
R.
Goldbrunner
(Köln)
Frühe aggressive Therapie von WHO Grad II Gliomen – contra
15 min
U. Schlegel
(Bochum)
Standard und neue Fragen für 1p/19q kodeletierte Gliome
15 min Im TuBo Stil
U. Herrlinger
(Bonn)
NOVO TTF: Ergebnisse der Primärtherapiestudie bei Patienten A. Hottinger
mit neu diagnostiziertem Glioblastom
(Genf)
20 min
Immuntherapie:
6
17. NOA Tagung, Heidelberg 2015
15:30 h
fact?
10 min
O. Grauer
(Münster)
fiction?
10 min
P. Roth
(Zürich)
Keynote Lecture: WHO Classification of Brain Tumors
Vorsitz: G. Reifenberger (Düsseldorf)
WHO Classification of Brain Tumors
16:15 h
David Louis
(Boston)
Kaffepause
7
17. NOA Tagung, Heidelberg 2015
Informationen
Anmeldung:
über die NOA-Homepage
Kosten:
50 €
Baden-Württembergische Bank Stuttgart
BLZ 600 501 01
Kto.Nr. 7421 500 429
IBAN DE64600501017421500429
Stichwort: NOA-Tagung, D 10083480
Webseite:
www.neuroonkologie.de
CME Punkte:
18 (beantragt)
Tagungsort:
Molkenkur, Klingenteichstrasse 31, 69117 Heidelberg
Sponsoren:
Gold:
Silber:
Bronze:
8
17. NOA Tagung, Heidelberg 2015
Abstracts der 17. NOA-Tagung
*Kurzvorträge sind markiert
Stereotactic fractionated radiotherapy of the resection cavity in patients with
one to three brain metastases
A. Bilger, Milanovic, Lorenz, A.-L. Grosu
Dept. Of Radiation Oncology, University Clinic Freiburg
In patients undergoing surgical resection of brain metastasis local recurrence is about
60%. Whole brain Radiation Therapy (WBRT) can significantly reduce the risk of
local relapse but fails to improve overall survival. The most important side effects of
WBRT are neurocognitive deficits, which can reduce quality of life. Stereotactic
fractionated radiotherapy (SFRT) of the resection cavity could be an alternative to
WBRT in patients with a limited number of brain metastasis.
The goal of this study is to evaluate the role of SFRT of the resection cavity in
patients with 1 to 3 brain metastases after surgical resection.
From November 2009 to August 2013 62 resection cavities of 60 patients were
treated by SFRT. All patients had ≤3 brain metastasis. The total irradiation dose was
30 Gy in 5 Gy/d, 5x/week after complete macroscopical resection and 35 Gy (5Gy/d)
in patients with macroscopic residual tumor after surgery. The gross tumor volume
(GTV) encompasses the residual tumor delineated on the contrast-enhanced T1MRI, the clinical target volume (CTV) encompasses the surgical cavity plus 1 mm
and the planning target volume (PTV) the CTV plus 2 mm.
In 52 patients there was a follow-up with MRI to evaluate local and distant intracranial
tumor control. There were 6/52 (11.54%) local failures and 29/52 (55.77%) distant
failures. Local control was correlated with age (p=0.046). 37/60 (61.6%) patients died
during follow-up. Median follow-up was 8 months. Median Overall survival was 15
months. Survival was correlated with KPS score ≤ 70% and PTV (ccm). No severe
side effects were seen.
SFRT of the resection cavity could be an alternative to WBRT after surgical resection
of ≤3 brain metastasis. Prospective studies are warranted.
*ErbB2/HER2-targeted NK cells display potent activity against glioblastoma
and enhance survival
Michael Burger1,2, Congcong Zhang2,3, Iris Mildenberger1,2, Lukas Jennewein4,
Sabrina Genßler3, Kurt Schönfeld3, Pia Zeiner4, Elke Hattingen5, Patrick N. Harter4,
Michel Mittelbronn4, Torsten Tonn6, Winfried S. Wels2,3, Joachim P. Steinbach1,2
1
Institute for Neurooncology, Goethe University, Frankfurt am Main, Germany.
German Cancer Consortium (DKTK) and German Cancer Research Center (DKFZ),
Heidelberg, Germany. 3Georg-Speyer-Haus, Institute for Tumor Biology and
Experimental Therapy, Frankfurt am Main, Germany. 4Edinger Institute, Goethe
University, Frankfurt am Main, Germany. 5Institute of Neuroradiology, Goethe
University, Frankfurt am Main, Germany. 6Institute for Transfusion Medicine, German
2
9
17. NOA Tagung, Heidelberg 2015
Red Cross Blood Donation Service North-East and Medical Faculty Carl Gustav
Carus, TU Dresden, Dresden, Germany.
Glioblastoma (GBM) is the most common and malignant intracranial tumor in adults
and is currently incurable. Recent studies suggest that natural killer (NK) cells have
potential for adoptive immunotherapy of GBM if consistent activation of their cytolytic
activity is maintained in the tumor microenvironment. To specifically target NK cell
activity to GBM, we employed NK-92/5.28.z cells which are continuously expanding
human NK cells that carry an ErbB2-specific second generation chimeric antigen
receptor (CAR). The cells were previously derived from the clinically applicable NK92 cell line under GMP conditions by transduction with a lentiviral CAR vector. In
immunohistochemical analysis we found elevated ErbB2 protein expression in >20%
of primary GBM samples and in the majority of GBM cell lines analyzed. In in vitro
cell killing assays, NK-92/5.28.z in contrast to untargeted NK-92 cells lysed all ErbB2positive established and primary GBM cells tested, with cell killing kinetics correlating
with the degree of NK cell activation upon contact with tumor cells. Potent in vivo
antitumor activity of NK-92/5.28.z cells was observed in different orthotopic GBM
xenograft models in NSG mice, resulting in inhibition of tumor growth and marked
extension of symptom-free survival upon repeated stereotactic injection of the CAR
NK cells into the tumor area. Our data demonstrate the potential of ErbB2-specific
NK-92/5.28.z cells for adoptive immunotherapy of glioblastoma, justifying evaluation
of this approach for the treatment of ErbB2-positive GBM in clinical studies.
*ZEB1 is ubiquitously expressed across subtypes in human glioblastoma and a
surrogate marker of tumor purity
Philipp Euskirchen1, Josefine Radke2, Marc Sören Schmidt1, Eva Schulze Heuling1,
Eric Kadikowski1, Meron Maricos1, Felix Knab1, Jun Cheng3, Peter Hufnagl4, Marcus
Czabanka5, Christoph Dieterich3, Hrvoje Miletic6,7, Sverre Mørk7, Arend Koch2,
Matthias Endres1 and Christoph Harms1
1
Dept. of Neurology, Charité – Universitätsmedizin Berlin
Dept. of Neuropathology, Charité – Universitätsmedizin Berlin
3
Max-Planck-Institute for the Biology of Ageing, Cologne
4
Dept. of Pathology, Charité – Universitätsmedizin Berlin
5
Dept. of Neurosurgery, Charité – Universitätsmedizin Berlin
6
Dept. of Biomedicine, University of Bergen, Norway
7
Department of Pathology, Haukeland University Hospital, Norway
2
The transcription factor ZEB1 has gained attention in tumor biology of epithelial
cancers because of its suggested roles in epithelial-mesenchymal transition, DNA
repair, mediation of stem cell properties and tumor-induced immunosuppresion, but
its role in CNS tumors, especially glioblastoma multiforme is not well understood.
We comprehensively characterized ZEB1 expression at single cell level in biopsy
samples from 273 human astroglial tumors. Interaction of ZEB1 and EGFR signalling
was interrogated in vitro by realtime quantitative PCR and immunocytochemistry in
glioma stem cell lines from subtype-stratified parental tumors based on RNAseq
transcriptomic gene expression profiles.
10
17. NOA Tagung, Heidelberg 2015
ZEB1 is expressed in nearly all examined tumors but in highly variable fraction of
cells. Co-staining of reactive CNS cell types in non-tumor tissue show that besides
bona fide tumor cells, only reactive astrocytes but not leucocytes, microglia and
macrophages contribute to the ZEB1 positive population. In GBM, ZEB1 labelling
index correlates with EGFR amplification and IDH1 mutation, but EGFR modulation
by cetuximab treatment did not alter ZEB1 expression in vitro.
ZEB1 is expressed ubiquitously and across molecularily defined subtypes in
glioblastoma multiforme as well as in non-GBM astroglial tumors. ZEB1 labelling
varies across GBM subtypes, but a causal link remains questionable. We speculate
that ZEB1 is a pan-astroglial tumor marker that reflects tumor purity.
*Prognostic value of the extent of resection and IDH1 mutation status in WHO
grade II astrocytomas
Christine Jungk1, Moritz Scherer1, Andreas Mock1, David Capper2, Alexander
Radbruch3, Andreas v. Deimling2, Christel Herold-Mende1, Andreas Unterberg1
1
Department of Neurosurgery, Heidelberg University Hospital, Heidelberg, Germany
Department of Neuropathology, Heidelberg University Hospital, Heidelberg,
Germany
3
Division of Neuroradiology, Department of Neurology, Heidelberg University
Hospital, Heidelberg, Germany
2
Current evidence is in favour of a maximized extent of resection (EOR) in low-grade
gliomas (LGG). However, conclusions were drawn from pooled cohorts of astrocytic
and oligodendroglial histology, underrating the fact that histologic subtypes and
molecular markers like mutations in the isocitrate dehydrogenase (IDH)1 entail
divergent outcomes. We therefore evaluated the prognostic impact of extensive
surgery in pure astrocytic LGGs stratified for IDH1 mutation status.
Forty-six consecutive cases of 1st resections for WHO grade II astrocytoma with
known IDH1 mutation status were analysed. Pre-, postoperative and follow-up
tumour volumes were calculated on FLAIR images. Cases with EOR ≥40% (n=39)
were analysed separately. Analysis for overall (OS), progression-free (PFS),
malignant progression-free (MPFS) survival and time to reintervention (TTR) was
conducted with uni- (uv) and multivariate (mv) regression models.
Analysing the complete cohort, a median final EOR of 90.4% (range 17.5 – 100%)
was achieved at follow-up MRI, corresponding to a median residual tumour volume of
4.09 cm3. EOR did not impact on OS but was significantly associated with PFS (HR
0.23; p=0.04) and TTR (HR=0.23; p=0.04; uv). IDH1 was the only independent
prognosticator for OS (HR 0.04; p=0.002; mv). When restricting survival analysis to
cases with EOR ≥40%, both initial and residual tumour volumes were predictive of
OS (HR >1, p<0.01 and HR>1, p<0.05, respectively; uv), even after adjustment for
IDH1. No significant correlation between EOR and neurologic morbidity was
observed (p=0.73).
In this histology-adjusted cohort, a greater EOR significantly delayed tumour
progression, malignisation and TTR without causing additional neurologic morbidity.
OS was predominantly determined by the presence of IDH1 mutations. Noteworthy,
in IDH1-mutated, EOR ≥40% cases, extensive surgery was predictive of OS as well.
11
17. NOA Tagung, Heidelberg 2015
Our approach illustrates the diverse impact of biological and surgical factors upon
outcome, underscoring the need to strive for maximized and safe resections.
*MRI tumor progression patterns in the GLARIUS trial
Sied Kebir1,2, Nina Junold1, Elke Hattingen3, Christina Schaub1, Niklas Schäfer1,2,
Joachim P. Steinbach,4 Astrid Weyerbrock,5 Peter Hau,6 Roland Goldbrunner,7
Michael Niessen1, Frederic Mack1, Moritz Stuplich1, Theophilos Tzaridis1, Oliver Bähr,
Martin Proescholdt,8 Martin Glas1,2,9, Ulrich Herrlinger1
1
Division of Clinical Neurooncology, Dept. of Neurology; University of Bonn, 2Stem
Cell Pathologies, Institute of Reconstructive Neurobiology, University of Bonn,
3
Divison of Neuroradiology; Department of Radiology, University of Bonn, 4Dr.
Senckenberg Institute of Neurooncology, University of Frankfurt, 5Department of
Neurosurgery, University of Freiburg, 6Department of Neurology and Wilhelm Sander
NeuroOncology Unit, University of Regensburg, 7Department of Neurosurgery,
University of Cologne, 8Department of Neurosurgery, University of Regensburg,
9
Clinical Cooperation Unit Neurooncology, MediClin Robert Janker Klinik, Bonn,
Germany
We evaluated patterns of tumor progression in patients with primary glioblastoma
who were assigned to undergo treatment with bevacizumab/irinotecan (BEV/IRI)
versus standard temozolomide (TMZ) within the randomized phase II GLARIUS trial.
In 160 patients (106 BEV/IRI, 54 TMZ), we reviewed MRI scans at baseline and
tumor progression. Based on contrast-enhanced T1-weighted and FLAIR images we
assessed tumor patterns and invasiveness. Tumor patterns were classified as either
multifocal or local at baseline and progression; at progression, we additionally
assessed whether distant lesions appeared. Invasiveness was determined as either
diffuse or non-diffuse. Associations were calculated using Fisher’s exact test.
At baseline, 118 of 160 evaluable patients (73,8%) had a locally confined tumor. In
the BEV/IRI arm 10% and in the TMZ arm 8% of patients with an initially local tumor
growth changed to a multifocal pattern (p=0.78); distant lesions were found in 24% in
the BEV/IRI arm and 19% in the TMZ arm (p=0.55). Eight percent of patients in the
BEV/IRI arm and 11% in the TMZ arm developed a diffuse growth pattern from an
initially non-diffuse pattern (p=0.58).
The tumor progression and invasiveness patterns do not differ between BEV/IRI and
TMZ-treated patients in the GLARIUS trial. BEV/IRI did not increase the rate of
multifocal, distant or highly invasive tumors at the time of progression.
Cerebrale Ischämie unter Glioblastom-Rezidiv-Therapie mit Bevacizumab
Koeppen S1, Hense J2
1
Klinik für Neurologie, 2Westdeutsches Tumorzentrum, Universitätsklinikum Essen
Bevacizumab, ein humanisierter monoklonaler Antikörper gegen zirkulierenden
Vascular Endothelial Growth Factor (VEGF), findet insbesondere in Kombination mit
12
17. NOA Tagung, Heidelberg 2015
Lomustin breite Anwendung in der Rezidivbehandlung des Glioblastoms (GBM).
Gliom-Patienten haben bereits therapieunabhängig ein erhöhtes Risiko, einen
ischämischen Hirninfarkt oder eine intrakranielle Blutung zu erleiden. Unter antiangiogener Therapie lag die Inzidenz für ischämische Schlaganfälle und
intrakranielle Hämatome in einer auf Patientendaten aus 4 Phase I- bzw. Phase IIStudien im Zeitraum 2005 – 2010 basierenden retrospektiven Analyse des National
Cancer Institute bei jeweils 1.9%.
Eine 50-jährige Patientin, bei der 2/2014 ein links frontales GBM mit negativem
Methylierungsstatus teilreseziert und anschließend eine standardmäßige RadioChemotherapie mit konkomitanter und adjuvanter Temozolomid-Gabe durchgeführt
worden war, unterzog sich 6/2014 wegen Tumorprogress einer Navigations- und 5Aminolävulinsäure (ALA)-gestützten Wachkraniotomie, erhielt 7/2014 eine
Lumbaldrainage aufgrund eines Liquorkissens links frontal, wurde dann mit
Irinotecan ± Temsirolimus im Rahmen einer unizentrischen Phase I-Studie (TEMIR)
behandelt und ab 11/2014 mit Bevacizumab wegen erneuten Tumorwachstums,
welches sich im weiteren Verlauf regredient zeigte. 3/2015 traten subakut
holocephale Kopfschmerzen, eine fluktuierende expressive Aphasie mit
phonematischen und semantischen Paraphasien, eine Dysphagie, Dyslexie,
leichtgradige zentrale Facialisparese rechts und ein Babinski-Phänomen links auf. Im
MRT stellten sich hämorrhagische, gering Kontrastmittel (KM)-anreichernde Areale in
der Umgebung der Resektionshöhle, ventral in den Stammganglien, links insulär und
rechts frontal dar neben Tumoranteilen ohne KM-Enhancement links, zusätzlich
akute ischämische Areale links temporo-parietal und bds. ventral in den
Stammganglien. Doppler-/duplexsonographisch fand sich extra- und transcraniell
keine Stenose / Occlusion der hirnversorgenden Arterien. Unter Ischämieprophylaxe
mit Acetylsalicylsäure (ASS) 100mg/d und Enoxaparin 60mg/d war die neurologische
Symptomatik innerhalb von 4 Tagen partiell rückläufig bei gleichzeitiger
antiödematöser Therapie mit Dexamethason 8mg/d. 9 Tage später erfolgte allerdings
eine notfallmäßige stationäre Aufnahme der Patientin wegen Diplopie, Schwindel und
ausgeprägter Fatigue.
Während es sich bei den bisher beschriebenen mit anti-angiogener Therapie
assoziierten cerebralen Ischämien in 50% der Fälle um lakunäre
Durchblutungsstörungen handelte, die sich mit einer medianen Latenz von 16.2
Monaten nach Therapiebeginn entwickelten, kam es im vorliegenden Fall unter
Monotherapie mit Bevacizumab bereits nach 4 Monaten zu einer territorialen
Ischämie
im
Arteria
cerebri
media
Areal
links,
die
unter
Thrombozytenaggregationshemmung mit ASS partiell reversibel war. Die rasch
nachfolgende Verschlechterung des Allgemeinzustandes und Zunahme der
neurologischen Symptomatik ließ sich einer Tumorprogression nach Absetzen von
Bevacizumab zuordnen.
Venous thromboembolism in human glioblastoma:
The journey from patients’ symptoms to tumor biology and back to a clinical
trial – from Trousseau Syndrome to CoaGlio IV
Susanne A. Kuhn1, Tobias Kratzsch2, Linn Handel3, Rolf Kalff3, Uwe-Karsten
Hanisch(†)4,5, Peter Vajkoczy2
1
Hospital Ernst von Bergmann, Department of Neurosurgery, Potsdam, Germany
13
17. NOA Tagung, Heidelberg 2015
2
Charité-Universitätsmedizin Berlin, Department of Neurosurgery, Berlin, Germany
Medical Center of Jena University, Department of Neurosurgery, Jena, Germany
4
Universitätsmedizin Göttingen, Institute for Neuropathology, Göttingen, Germany
5
Paul-Flechsing-Institut, Universität Leipzig, Germany
3
Venous thromboembolism (VTE) is a largely preventable, often fatal complication in
cancer patients. Multiple anticoagulants are widely and affordably available, but
unfortunately VTE is often not prevented and represents the second most frequent
death cause in cancer patients. Further, the annual average total costs for cancer
patients with VTE were found to be almost 50% higher than that of cancer patients
without VTE. Additionally, growing evidence shows VTE as paraneoplastic syndrome,
acting as external monitor of highly coagulation-active tumors that use the
coagulation system for supporting their own aggressive malignancy. VTE events in
malignant tumors will harm patients and recur, unless the tumors’ coagulative activity
isn’t blocked. Currently, no guidelines can recommend long-term prophylactic
anticoagulation, as there exists not any officially approved anticoagulant drug, neither
subcutaneously nor orally to administer, that was tested for its VTE prophylactic
power in a patient population of solely cancer patients. This study aims at pathway,
target, and drug identification and the generation of a clinical trial for providing a
rationale for consequent VTE prophylaxis.
Clinical file review, prospective cohort study (n=41), glioblastoma (GBM) sample
immunochemistry (n=112), cell line assays (n>450) and animal models (n=20) were
performed. Target for VTE prevention and experimental drug was identified. Clinical
study for VTE prophylaxis was developed as randomized, controlled, triple-blind,
multinational clinical phase III trial – CoaGlio IV.
GBM patients experience clinically evident VTE in one third of all cases; silent events
are even more common. As they show pathologically activated coagulation factors
(FII-FXII) in their systemic coagulation, those with VTE display significant correlation
with coagulation factor activation (p=0.018). Further, patients with preoperative,
protective co-activation of antithrombin III experience longer survival times, whereas
patients with preoperative, normal or low antithrombin III survive shorter. Not enough
the coagulation activation in the periphery, glioblastoma tissue pathologically
expresses all coagulation factors of the intrinsic pathway as well as the common final
pathway (FII, FIX, FX, FXI, FXII, p<0.05), and the coagulation factor receptors PAR-1
and PAR-4 (p<0.05). The activation of receptors with coagulation factors increases
cellular DNA synthesis (p<0.01), proliferation (p<0.01), and migration (p<0.05).
Consequently, VTE prophylaxis by low molecular weight heparins reduces size of
established human glioblastomas in xenografts (p<0.05), mitotic activity within
xenografts and tumor angiogenesis. Taken together and reflected against the
background of other human cancers, clinical trials for VTE prophylaxis are urgently
needed. Therefore, the controlled, triple-blind, multinational phase III study (CoaGlio
IV) was created to randomize 530 patients in a 1:1 fashion to receive either
international first line standard therapy after surgery (radiochemotherapy with
temozolomide + adjuvant temozolomide) and placebo or consequent postoperative
prophylactic anticoagulation with oral factor Xa blocker apixaban over a 12-monthperiod. The power calculation is a 10% difference in overall survival, power at 80%,
and two-sided alpha level of 0.05. The primary endpoint is overall survival. Trial will
activate at more than 20 centers in Germany, Austria, Switzerland, and the United
States. Trial is registered at European Clinical Trials Database: 2015-000425-37.
Finally, factor Xa inhibition not only should prevent VTE frequency, severity, as well
as related deaths in the short time, but also should improve the long-term prognosis
14
17. NOA Tagung, Heidelberg 2015
by active inhibition of tumor biology. Transfer of clinical problems into the lab as well
as translation of results into a clinical study protocol is necessary, as VTE is a
dangerous complication, and glioblastoma is still a deadly disease.
Ibuprofen and diclofenac inhibit migration and proliferation of human glioma
cell lines in vitro
Verena Leidgens1, Corinna Seliger1, Petra Leukel1, Arabel Vollmann-Zwerenz1, Lisa
Rauer1, Ulrich Bogdahn1, Marina Kreutz2, Paul Sander3, Oliver M. Grauer1,4, Peter
Hau1
1
Wilhelm Sander-NeuroOncology Unit and Department of Neurology, University of
Regensburg, Regensburg, Germany, 2Department of Internal Medicine III, University
Hospital Regensburg, Regensburg, Germany, 3Institute of Biophysics and Physical
Biochemistry, University of Regensburg, Regensburg, Germany, 4Department of
Neurology, University Hospital Münster, Münster, Germany
Hallmarked by infiltrating tumor cells, Glioblastoma (GBM) are characterized as
highly malignant brain tumors. Nonsteroidal anti-inflammatory drugs (NSAIDs) have
been described to inhibit tumor cell proliferation as well as to induce tumor cell
apoptosis in a Cox-independent way. The current study was designed to evaluate the
effects of the NSAIDs diclofenac and ibuprofen (in concentrations achievable in
patients) on proliferation, migration and lactate formation in several human glioma
cells. Significantly decreased migration and proliferation was observed with
diclofenac or ibuprofen, whereas ASA had negligible effects. Additionally, diclofenac
and ibuprofen proved to affect cell proliferation as ibuprofen caused cell cycle arrest
in G1 whereas cells treated with higher diclofenac concentrations arrest at the G2/M
checkpoint. Both drugs lead to a decrease of lactate and pSTAT3 levels. To further
investigate STAT3 effects, the specific inhibitor STATTIC preventing STAT3
phosphorylation was used, revealing a comparable effect on cell proliferation and
migration supporting the substantial role of STAT3 in this regulation cascade. In
summary, this study shows an in vitro treatment advantage of diclofenac or ibuprofen
over ASA. Thus both NSAIDs may be suited to augment GBM treatment in patients,
due to their migration and proliferation inhibiting effects. Consequential further
comparative studies with different NSAIDs and in vivo studies are necessary to affirm
the positive effects of diclofenac and ibuprofen in glioma patient treatment.
*ATF4 as a mediator of resistance to targeted therapy in high-grade gliomas
Sylvia Moeckel1, Bart Neyns2, Edward Pan3, Markus J. Riemenschneider4, AnjaKatrin Bosserhoff5, Arabel Vollmann-Zwerenz1, Katharina Meyer6, Rainer Spang6,
Peter Hau1
1
Wilhelm Sander-NeuroOncology Unit and Department of Neurology, 4Department of
Neuropathology, Regensburg University Hospital, 6Institute of Functional Genomics,
University of Regensburg, Regensburg, Germany
5
Institute of Biochemistry, Emil-FischerCenter, Friedrich-Alexander-University
Erlangen-Nuernberg, Erlangen, Germany
15
17. NOA Tagung, Heidelberg 2015
2
Medical Oncology, UZ Brussel, Brussel, Belgium,
Department of Neurology and Neurotherapeutics, UT Southwestern Medical Center,
Dallas, TX, USA
3
Small molecule inhibitors have been investigated in a large set of clinical trials in
high-grade gliomas (HGG). Despite promising preclinical studies, results of pilot trials
have been generally disappointing. A deeper understanding of the complex biology of
malignant glioma cells, and their adaptation to targeted agents is therefore critical for
further therapy development.
We performed microarray analysis in 18 short-term serum-free cultures of high-grade
gliomas enhanced for brain tumor initiating cells (BTIC) before and after in vitro
treatment with the tyrosine kinase inhibitor Sunitinib. Based on the observation of
gene network analysis we hypothesize that the central mediator of the integrated
stress response ATF4 (activating transcription factor 4) plays an important role in the
regulation of the Sunitinib induced expression profiles and probably in the adaptation
to treatment conditions. ATF4 is involved in metabolism and nutrient uptake,
antioxidation, and regulation of apoptosis and autophagy. Of note, ATF4 has been
associated with multidrug resistance in different cancer models.
We analyzed the expression of ATF4 in paraffin embedded tissue blocks from HGG
patients treated with Sunitinib by immunohistochemistry. Interestingly, ATF4
significantly correlated with shorter overall survival from the beginning of Sunitinib
treatment.
In vitro studies confirmed a dose dependent induction of ATF4 protein expression in
Sunitinib treated BTICs. Furthermore we observed an intracellular accumulation of
autophagosomes. Co-administration of the autophagy inhibitor Chloroquine could
enhance the sensitivity to Sunitinib treatment.
In summary our data suggest that ATF4 expression may be predictive for response to
Sunitinib and might also be involved in general resistance mechanisms in HGGs.
Further studies are ongoing to elucidate the induction of apoptosis and autophagy in
dependence of ATF4 expression and its contribution to therapy response.
*Resection of prerolandic motor areas defined by nTMS motor evoked
potentials correlates with postoperative motor function
Tobias Moser, Lucia Bulubasova, Jamil Sabih, Florian Ringel, Bernhard Meyer,
Sandro M. Krieg
Department of Neurosurgery, Klinikum rechts der Isar, Technische Universität
München, Munich, Germany
Navigated transcranial magnetic stimulation (nTMS) is used to determine the
distribution of motor eloquent areas. As shown in previous studies there are primary
motor areas frontal to the precentral gyrus. However there is no clear evidence about
the importance and eloquence of these areas for surgical considerations. We
therefore investigated the correlation surgically resected nTMS-positive prerolandic
motor areas and postsurgical loss of motor function.
Forty patients with gliomas in the precentral or prerolandic cortex where examined
with nTMS prior to surgery to determine the localization of the primary motor areas.
Based on the fusion of the mapping points with the postsurgical MRI, we identified
16
17. NOA Tagung, Heidelberg 2015
nTMS-positive points which were resected in the infiltration zone oft he tumor. The
resected points where then classified in localization and latency of the motor evoked
potentials.
The 40 patients with an average age of 57.1 years presented tumor entities
consisting of 25 glioblastomas, 8 diffuse and 5 anaplastic astrocytomas, one
anaplastic oligodendroglioma and one oligoastrocytoma. Within the 40 patients 27
showed nTMS-positive motor eloquent points in the prerolandic gyri during nTMS
mapping. The resection of nTMS-positive cortical areas lead to a considerable
number of transient pareses, which we observed in 8 patients with direct postsurgical
motor function deficits. In 5 of these patients nTMS points were resected in the
prerolandic gyri, in two patients points within the precentral gyrus and one patient
was showing a pareses without resection of motor eloquent spots.
Despite nTMS identifies a considerably high number of prerolandic motor areas,
these should also be considered as highly eloquent and spared during resection.
*Synergie von Strahlentherapie und Antigen-spezifischer Vakzinierung in der
Behandlung maligner Gliome
Katharina Ochs1,2, Martina Ott1,2, Sara Chiblak3,4,5, Michael O. Breckwoldt6, Amir
Abdollahi3,4,5, Wolfgang Wick1,7, Michael Platten1,2
1
Neurologische Klinik, Universitätsklinikum Heidelberg und Nationales Zentrum für
Tumorerkrankungen (NCT), Heidelberg, Deutschland;
2
Klinische Kooperationseinheit Neuroimmunologie und Hirntumorimmunologie,
Deutsches Konsortium für Translationale Krebsforschung (DKTK), Deutsches
Krebsforschungszentrum (DKFZ), Heidelberg, Deutschland;
3
Abteilung für Radioonkologie und Strahlentherapie, Universitätsklinikum Heidelberg,
Heidelberg, Deutschland
4
Deutsches Konsortium für Translationale Krebsforschung (DKTK), Zentrum für
Medizinische Strahlenforschung in der Onkologie, Heidelberger Institut für
Radioonkologie (HIRO), Heidelberg, Deutschland
5
Molekulare und Translationale Radioonkologie, Heidelberger Ionen-Therapie
Zentrum (HIT), Medizinische Fakultät der Universität Heidelberg und Nationales
Zentrum für Tumorerkrankungen (NCT), Deutsches Krebsforschungszentrum
(DKFZ), Heidelberg, Deutschland
6
Abteilung für Neuroradiologie, Universitätsklinikum Heidelberg, Heidelberg,
Deutschland
7
Klinische Kooperationseinheit Neuroonkologie, Deutsches Konsortium für
Translationale Krebsforschung (DKTK), Deutsches Krebsforschungszentrum (DKFZ),
Heidelberg, Deutschland
In der Behandlung maligner Gliome gewinnen Immuntherapien wie beispielsweise
die Vakzinierung gegen Tumor-spezifische Antigene zunehmend an Bedeutung,
wobei immunsuppressive Faktoren des Tumormikromilieus das Entstehen einer
effektiven Anti-Gliom-Immunantwort erschweren. Zudem ist bisher weitgehend
unklar, wie eine Immunisierung in die Standardtherapie, die eine kombinierte
Radiochemotherapie umfasst, integriert werden soll. Ziel dieser Studie ist, mit Hilfe
eines syngenen orthotopen Mausmodells Effektivität und molekulare Mechanismen
17
17. NOA Tagung, Heidelberg 2015
einer kombinierten Radioimmuntherapie bestehend aus einer Peptidvakzinierung und
einer niedrig dosierten Bestrahlung zu untersuchen.
Während eine alleinige Vakzinierung von Mäusen mit das hoch immunogene ModellTumorantigen gp100 überexprimierenden GL261-Gliomen keinen therapeutischen
Nutzen zeigte, führte eine fraktionierte Bestrahlung mit 5 x 2 Gy zu einer relevanten
Verzögerung des Tumorwachstums. Eine Kombination aus anti-gp100-Vakzinierung
und Bestrahlung dagegen hatte bei ca. 70 % der Mäuse eine komplette Regression
zuvor etablierter Gliome zur Folge. Durchfusszytometrische Analysen der Tumorinfiltrierenden Lymphozyten zeigten einen signifikanten Anstieg der gp100spezifischen zytotoxischen T-Zellen in der kombiniert behandelten Gruppe im
Vergleich zu den jeweiligen monotherapierten Kontrollgruppen. Auch die
Kombination einer Photonenbestrahlung mit einem adoptiven Transfer gp100spezifischer CD8-positiver T-Zellen aus T-Zell-Rezeptor-transgenen Mäusen führte
zu einer signifikanten Akkumulation der transferierten T-Zellen im Tumor
interessanterweise allerdings ohne Effekt auf das Tumorwachstum.
Eine niedrig dosierte, fraktionierte Bestrahlung scheint also das Gliom-Mikromilieu so
zu verändern, dass die Infiltration Antigen-spezifischer zytotoxischer T-Zellen
gefördert und so das Entstehen einer Vakzinierungs-induzierten effektiven
Immunantwort erleichtert wird.
Hippocampal-avoiding WBRT selectively prevents hippocampal atrophy as
determined by automated volumetry
Oliver Oehlke1, Deborah Sturm1, Karl Egger2, Anca-Ligia Grosu1,3,4, Lars Frings1,5,6
1
Department of Radiation Oncology, University Medical Center Freiburg, Germany
Department of Neuroradiology, University Medical Center Freiburg, Germany
3
German Consortium for Translational Cancer Research (DKTK), Freiburg, Germany
4
German Cancer Research Center (DKFZ), Heidelberg, Germany
5
Department of Nuclear Medicine, University Medical Center Freiburg, Germany
6
Center for Geriatrics and Gerontology, University Medical Center Freiburg, Germany
2
Hippocampal-avoiding whole brain radiotherapy (HA-WBRT) for multiple brain
metastases may prevent treatment-related cognitive decline, compared to standard
WBRT. Reduction in hippocampal volume over time has been shown to be
significantly related to decline in memory and learning. This study aims at exploring
brain volume changes after whole brain radiotherapy with hippocampal avoidance
(HA-WBRT).
Twenty-two patients who had been assigned to HA-WBRT following clinical indication
received MRI before and up to 19 months after treatment (Mean = 5 months).
Observer-independent, automated volumetry of the hippocampi, the lateral ventricles,
and the whole brain was performed on 77 MRI data sets using Matlab and statistical
parametric mapping (SPM8). Longitudinal volumes have been tested for significant
brain changes after HA-WBRT. Linear mixed models were computed with regional
volumes as dependent variables and predictor variable time with random intercepts
and slopes for time across subjects and variance components as covariance
structure (controlling for age). Volume change rates after RT have been calculated
for each region.
18
17. NOA Tagung, Heidelberg 2015
At the group level, hippocampal [-0.02 ml (-0.19%) per month] and whole brain
volumes [-2.03 ml (-0.19%) per month] showed decreases which did not reach
significance. By contrast, the lateral ventricles significantly expanded [p <
0.0001; +1.2 ml (3.6%) per month (43% per year)].
In this pilot study we observed no significant hippocampal atrophy after HA-WBRT.
The lateral ventricles, however, expanded after treatment, indicating cerebral atrophy
at a higher rate than reported in the literature on healthy subjects. While brain tissue
degeneration seems to occur in this group of patients, hippocampal tissue is spared
from this process.
*Supportive care in patients with high-grade gliomas – how to integrate clinical
assessment in daily routine? Results of a prospective multicentric trial
“ERASMUS” in 175 patients.
Den Bedarf an supportiven Maßnahmen bei Patienten mit malignen Gliomen im
ambulanten Bereich frühzeitig erkennen - Ergebnisse aus einer prospektiven
teilrandomisierten multizentrischen Studie “ERhaltung von Lebensqualität, Aktivität
und psychischer Stabilität in Mainz, Ulm/Günzburg und Stuttgart.”
Mirjam Renovanz1, Marlene Hechtner2, Anne-Katrin Hickmann3, Minou Nadji-Ohl3,
Mareile Janko1, Karoline Kohlmann1, Jochem König2, Jan Coburger4, Alf Giese1
1
Klinik für Neurochirurgie, Universitätsmedizin Mainz, 55131 Mainz
Institut
für
medizinische
Biometrie,
Epidemiologie
und
Informatik,
3
Universitätsmedizin Mainz Neurochirurgische Klinik, Klinikum Stuttgart, 70174
Stuttgart
4
Klinik für Neurochirurgie, Universität Ulm, Standort Günzburg, 89312 Günzburg
2
Improving quality, effectiveness, and efficiency of therapy and care in patients with
malignant gliomas involves assessment of psychosocial and supportive care needs.
The aim of this study was 1) to test and optimize an assessment of psychosocial and
supportive care needs in patients with high-grade gliomas in an outpatient setting
and 2) to assess possible predictors for the patients’ need of psychosocial help
during the further course of a glioma disease aiming at a more patient-centered
health care.
In neuro-oncological outpatient departments of three study centers 175 glioma
patients were assessed using EORTC QLQ-C30 + BN20, Distress Thermometer
(DT) and SCNS-SF34-G. The questionnaires were completed either with personal aid
(group A) or by patients alone (group B). Socio-demographic data, tumor stage, KPS,
and adjuvant therapies were documented. Feasibility to implement the
questionnaires in clinical routine was tested by evaluating time needed to complete
the questionnaire, rate of missing values, errors in completion by comparing Groups
A and B. Socio-demographic data, perioperative data (e.g. (eloquent) localization of
the tumor, and clinical data (ECOG/KPS, adjuvant therapies) of the patients were
assessed and their probable influence on the mental status analyzed by multivariable
linear and logistic regression models.
The study included 175 patients, m:f ratio was 1.1:1. The 72 (41%) patients in group
A needed an average of 25 min (range 10 to >50) for completion of all 3 surveys.
19
17. NOA Tagung, Heidelberg 2015
Most of the patients (>90%) had no problems in answering QLQ-C30+BN20 and DT,
but 33% misunderstood questions within SCNS. There was no evidence that
assistance of personal aid influenced the results of DT (mean 4.6±2.7 vs. 4.6±2.8,
p=0.88) and the EORTC-scores (e.g. C30, global health scale 62.3±22.1 vs.
58.9±23.3, pt=0.33). However, the proportion of incorrectly filled items in the SCNS
questionnaires was slightly lower in group A than in group B (7 (10%) vs. 19 (20%),
p=0.07). In several regression analysis models the following issues: “study center”,
“gender”, “profession”, “age”, “entity and WHO°”, and “ongoing chemotherapy” were
not found to influence the probability of patients need or desire for psychosocial
support according to DT>6, whereas we found “KPS” highly significant influencing the
patients’ wish for psychosocial support (p>0,0001, CI: 0,881-0,963). Linear
regression for global health status (EORTC, C30) showed that cognitive functioning
was influenced by localization of the tumor.
Early identification of psychosocial and supportive care needs in brain tumor patients
is of high importance for treatment success. However, the SCNS-SF34 turned out to
be not a suitable tool in glioma patients, whereas the DT and EORTC were well
implemented and useful in clinical routine. Supportive care surveys specifically
designed for needs and abilities of glioma patients are needed and tailored
assessment should be integrated in outpatient routine.
*Next-Generation-Sequencing in routine neuropathology diagnostics
Felix Sahm1,2, Daniel Schrimpf1,2, Jochen Meyer2, David Jones3, David Reuss1,2,
David Capper1,2, Christian Koelsche1,2, Annekathrin Kratz1,2, Andrey Korshunov1,2,
Mark Zapatka4, Stephan Wolf5, Stefan Pfister3,6, Andreas Unterberg7, Wolfgang
Wick8,9, Andreas von Deimling1,2
1
Department of Neuropathology, Institute of Pathology, Ruprecht-Karls-University
Heidelberg, INF 224, 69120 Heidelberg, Germany
2
Clinical Cooperation Unit Neuropathology, German Consortium for Translational
Cancer Research (DKTK), German Cancer Research Center (DKFZ), INF 224
Heidelberg, Germany
3
Division of Pediatric Neurooncology, German Consortium for Translational Cancer
Research (DKTK), German Cancer Research Center (DKFZ), INF 224 Heidelberg,
Germany
4
Division of Molecular Genetics, German Cancer Research Center (DKFZ), INF 280,
Heidelberg, Germany
5
Genomics and Proteomics Core Facility, German Cancer Research Center (DKFZ),
Im Neuenheimer Feld 280, Heidelberg 69120, Germany.
6
Department of Pediatric Oncology, Hematology & Immunology, Heidelberg
University Hospital, Im Neuenheimer Feld 430, Heidelberg 69120, Germany
7
Department of Neurosurgery, University Hospital Heidelberg, INF 400, 69120
Heidelberg, Germany
8
Neurology Clinic, University Hospital Heidelberg and National Center for Tumor
Diseases, INF 400, 69120 Heidelberg, Germany
9
Clinical Cooperation Unit Neurooncology, German Consortium for Translational
Cancer Research (DKTK), German Cancer Research Center (DKFZ), INF 224
Heidelberg, Germany
20
17. NOA Tagung, Heidelberg 2015
With the numbers of prognostic and predictive genetic markers in neuro-oncology
steadily growing, the need for comprehensive molecular work-up of neuropathology
samples has tremendously increased. Although several aberrations can be detected
by immunohistochemistry (mutant IDH1R132H and BRAFV600E protein, loss of
ATRX expression), the traditional detection of mutations, intragenic deletions and
gene fusions requires time consuming methods such as Sanger sequencing and
fluorescence in-situ hybridization. Covering all potentially clinically relevant genes in
diagnostic routine is virtually impossible by these means. Thus, reliable highthroughput methods allowing parallel analysis of multiple targets will be required in
diagnostic neuropathology.
We developed an enrichment-based gene panel comprising the entire coding and
selected promoter regions of 130 genes recurrently mutated in brain tumors.
Optimization of probe design, extraction, library generation and sequencing
conditions on 150 samples yielded a 5-workday routine workflow from FFPE sample
to neuropathological report: Library generation is based on the Agilent SureSelect
enrichment technology and sequencing is performed on an Illumina NextSeq 500.
Raw data are processed with BWA-MEM for alignment, SAMtools mpileup for single
nucleotide variant calling, Platypus for indel calling and De-Fuse for fusion detection.
Variants in the germline, where available, are subtracted from the variants in the
tumor. Variants in the tumor are subsequently annotated with dbSNP, 1000-genome
and COSMIC entries as well as with SIFT and PolyPhen2 scores to inform on their
biological relevance.
An average coverage of 500x enables detection of mutations also in samples with
low tumor cell content. Presence of entity-specific signatures (e.g. rare IDH1 variant,
ATRX, TP53; IDH2, TERT, CIC) added valuable information in cases where only
small samples, representing the infiltration zone of the tumor, is provided. Further,
inclusion of intronic regions of genes involved in fusion events enables the detection
of 7 pivotal fusion events in parenchymal and meningeal brain tumors. Finally, high
density of the designed probes, combined with the high coverage, permits the
detection of copy number variations within the target regions with high resolution.
In conclusion, we present the settings for high-throughput next-generationsequencing in routine neuropathology diagnostics. Such approach will likely become
indispensable as more therapeutic targets emerge and genetic information enters the
classification of brain tumors.
Evaluation of Postoperative Imaging Protocols in Low-Grade WHO°II
Astrocytomas: Preference for Late Postoperative MRI for Precise Assessment
of Resection Extent
Moritz Scherer1, Christine Jungk1, Andreas Mock1, Alexander Radbruch2, Christel
Herold-Mende1, Andreas Unterberg1
1
2
Neurochirurgische Klinik, Universitätsklinikum Heidelberg
Abteilung für Neuroradiologie, Neurologische Klinik, Universitätsklinikum Heidelberg
The extent of resection (EOR) after surgery has frequently been identified as an
independent positive prognosticator for survival in low-grade gliomas. However, times
when to determine the EOR after surgery have been chosen very heterogeneously in
past studies limiting the comparability of results. Particularly in non contrast21
17. NOA Tagung, Heidelberg 2015
enhancing tumors, surgical trauma and formation of edema are sources of bias for
the interpretation of early postoperative imaging. This study sought to evaluate
differences of early vs. late postoperative imaging to determine the most appropriate
time for the assessment of EOR.
In total, 29 cases with primary resections for WHO°II astrocytomas and simultaneous
early (24-48h) and late (median 3,6m, range 1,9-7,5m) postoperative imaging were
retrospectively identified since 2004. Cases were adjusted for histology and previous
radiation therapy to exclude confounders of image interpretation. Volumetric analysis
and calculation of EOR was performed on FLAIR sequences (Fluid-Attenuated
Inversion Recovery) early and late postoperatively. Influence of volumetric values on
progression-free survival (PFS) was calculated in univariate models.
Median postoperative EOR significantly increased by 23% from 67% early, to 90%
late postoperatively (p<0,01, Wilcoxon matched pairs test). Moreover, correlation
analysis confirmed this difference to be systematic (Spearman r=0,71, p<0,0001).
Evidence of surgical trauma (ischemia, edema) was found on 66% (19/29) of early
postoperative MRI diffusion-weighted-imaging (DWI). In survival analysis, only late
EOR was significantly associated with PFS (HR=0,23, p=0,04), early EOR was not a
significant confounder of survival.
After resection of WHO°II astrocytomas, EOR depends strongly upon the time when
it is defined after surgery. EOR was systematically underestimated early after surgery
and only late EOR was a significant prognosticator for PFS in this series. Postsurgical trauma can mimic FLAIR hyperintensity that bias early postoperative image
interpretation and might account for differences in EOR calculation. Results from this
analysis advocate the use of late instead of early postoperative imaging for the
evaluation of residual tumor and definition of EOR in non contrast-enhancing
astrocytomas.
*Über die parakrinen immunmodulatorischen Effekte des Onkometaboliten 2HG im Gliommikromilieu
Theresa Schumacher1,2, Lukas Bunse1,2, Stefan Pusch3,4, Andreas von Deimling3,4,
Wolfgang Wick2,5, und Michael Platten1,2
1,
KKE
Neuroimmunologie
und
Hirntumorimmunologie,
Deutsches
Krebsforschungszentrum, Heidelberg; 2, Neurologische Klinik, Universitätsklinik
Heidelberg; 3, KKE Neuropathologie, Deutsches Krebsforschungszentrum,
Heidelberg; 4, Abteilung Neuropathologie, Universitätsklinik Heidelberg; 5, KKE
Neuroonkologie, Deutsches Krebsforschungszentrum
Mutationen in den Genen der Isozitratdehydrogenasen (IDH) betreffen ca. 80 % aller
niedriggradigen und anaplastischen Gliome sowie sekundärer Glioblastome und
resultieren fast ausschließlich im Aminosäureaustausch von Arginin zu Histidin an
Position 132 (R132H). Allen IDH Mutationen ist die neomorphe Produktion des
Onkometaboliten 2-Hydroxyglutarat (2-HG) gemein, der durch Inhibition von DNAund Histonmethylierung zu genomweiter Hypermethylierung (G-CIMP), genetischer
Instabilität und letztendlich zur malignen Transformation führt. In zahlreichen
präklinischen Studien wurden tumormetabolische Konsequenzen durch die
intrazelluläre Akkumulation von 2-HG gezeigt, wie z.B. die Entdifferenzierung durch
DNA-Hypermethylierung, aber auch Veränderungen zellulärer Signalwege wie die
22
17. NOA Tagung, Heidelberg 2015
des Hypoxie induzierbaren Faktors (HIF) und TGF-β. Unsere bisherigen Arbeiten
haben gezeigt, dass a) die IDH1R132H-Mutation als solche vom Immunsystem
erkannt wird, b) die IDH1R132H-Mutation in Patienten mit IDH1R132H positiven
Gliomen spontane Immunantworten auslöst und c) eine spezifische
Peptidvakzinierung das Tumorwachstum IDH1R132H-exprimierender Tumoren in
Mäusen hemmt. Ob und wie (IDH1R132H-spezifische) T-Zellen und
antigenpräsentierende Zellen (APZ) parakrin von der 2-HG-Produktion IDH1(R132H)mutierter Gliomzellen beeinflusst werden ist bislang ungeklärt.
Wir zeigen in TCGA-Expressionsanalysen anaplastischer Astrozytome eine
verminderte Expression spezifischer T-Zellmarker in IDH1R132H-positiven im
Vergleich zu IDH1-wildtypischen Tumoren. Die verminderte T-Zellinfiltration wurde
anhand der CD4- und CD8-Expression immunhistochemisch verifiziert. Aufgrund
dieser Ergebnisse wurden direkte parakrine Effekte von 2-HG auf T-Zellen in vitro
untersucht. Wir zeigen, dass exogenes 2-HG moderat zellpermeabel ist und sowohl
in T-Zellen als auch in APZ transloziert. Nach 2-HG-Behandlung waren die
Proliferation und die Th1-Zytokinproduktion nach antigenspezifischer Aktivierung von
T-Zellen reduziert. Gleichzeitig regulierten APZ ko-stimulatorische Moleküle herunter.
Die Differenzierung naiver CD4+ T Zellen in Th1 und Th17 Zellen wurde durch die 2HG-Behandlung begünstigt, während die Differenzierung in Th2 Zellen beeinträchtigt
war. Unsere Ergebnisse zeigen, dass exogenes 2-HG die T-Zellimmunität beeinflusst
und die für Antitumorimmuntherapie relevante Th1-Aktivität beeinträchtigt. Es
resultiert die Rationale einer Kombinationstherapie aus IDH1-Inhibitoren und einer
IDH1R132H-spezifischen Vakzinierung.
Use of cardiac glycosides and risk of glioma
C. Seliger1, C. R. Meier2,3,4, S. S. Jick3, P. Hau1 and M. F. Leitzmann5
1
Department of Neurology and Wilhelm Sander-NeuroOncology Unit, Regensburg
University Hospital, Regensburg, Germany
2
Basel Pharmacoepidemiology Unit, Division of CIinical Pharmacy and
Epidemiology, Department of Pharmaceutical Sciences, University of Basel,
Switzerland
3
Boston Collaborative Drug Surveillance Program, Boston University School of
Public Health, Boston University, USA
4
Hospital Pharmacy, University Hospital Basel, Switzerland
5
Department of Epidemiology and Preventive Medicine, University of Regensburg,
Regensburg, Germany
Cardiac glycosides induce apoptotic effects on glioma cells, but whether cardiac
glycosides protect against risk for glioma is unknown.
We performed a case-control analysis using the Clinical Practice Research Datalink
(CPRD) involving 2,005 glioma cases diagnosed between 1995 and 2012 who were
individually matched to 20,050 controls on age, gender, general practice, and
number of years of active history in the database. Conditional logistic regression
analysis was used to evaluate the association between cardiac glycosides and the
risk of glioma adjusting for body mass index and smoking. We also examined use of
common heart failure and arrhythmia medications to differentiate between a specific
23
17. NOA Tagung, Heidelberg 2015
glycoside effect and a generic effect of treatment for congestive heart failure or
arrhythmia.
Cardiac glycoside use was inversely related to glioma. After adjustment for
congestive heart failure, arrhythmia, diabetes, and common medications used to treat
those conditions, the OR of glioma was 0.47 (95% CI=0.27-0.81) for use versus nonuse of cardiac glycosides. In contrast, no associations were noted for other
medications used to treat congestive heart failure or arrhythmias. The corresponding
ORs were 1.09 (95% CI=0.92-1.29) for use of ACE-inhibitors, 0.94 (95% CI=0.811.09) for use of beta-blockers, 1.03 (95% CI=0.89-1.19) for use of diuretics, and 1.00
(95% CI=0.84-1.19) for use of anti-arrhythmics other than glycosides. The OR of
glioma in people with congestive heart failure was 0.65 (95% CI=0.40-1.04) and for
arrhythmia it was 1.01 (95% CI=0.78-1.31).
These data indicate that cardiac glycoside use is independently associated with
reduced glioma risk. Confirmation of our findings in future studies is required before
determining whether glycosides protect against glioma.
Changing the clinical course of glioma patients by preoperative motor mapping
with navigated transcranial magnetic brain stimulation
Nico Sollmann, Thomas Obermueller, Jamil Sabih, Lucia Bulubasova, Chiara
Negwer, Tobias Moser, Florian Ringel, Bernhard Meyer, Sandro M. Krieg
Department of Neurosurgery, Klinikum rechts der Isar, Technische Universität
München, Munich, Germany
Mapping of the motor cortex by navigated transcranial magnetic stimulation (nTMS)
can be used for preoperative planning in brain tumor patients. Just recently, it has
been proven to actually change outcomes by increasing the rate of gross total
resection and by reducing the surgery-related rate of paresis. Yet, we also need data
that shows if these changes also lead to a changed clinical course.
We prospectively enrolled 70 patients with supratentorial motor eloquently located
high-grade gliomas (HGG) undergoing preoperative nTMS (2010-2014) and matched
these patients with 70 HGG patients who did not undergo preoperative nTMS (20072010).
The overall size of the craniotomy was 25.3 ± 9.7 cm2 (median 22.5 cm2, range 12.0
– 61.6 cm2) for nTMS and 30.8 ± 13.2 cm2 (median 28.0 cm2, range 4.6 – 65.7 cm2)
for non-nTMS patients (p=0.0058). Median inpatient stay was 12 days for the nTMS
and 14 days for the non-nTMS group (nTMS: CI 10.5 – 13.5 days; non-nTMS: CI
11.6 – 16.4 days; p= 0.0446). 60.0% of patients of the nTMS group and 54.3% of
patients of the non-nTMS group were eligible for postoperative chemotherapy (OR
1.2630, CI 0.6458 – 2.4710, p=0.4945), while 67.1% of nTMS patients and 48.6% of
non-nTMS patients received radiotherapy (OR 2.1640, CI 1.0910 – 4.2910,
p=0.0261). There was a trend toward higher survival in the nTMS group (nTMS: 15.7
± 10.9 months; non-nTMS: 11.9 ± 10.3 months; p=0.1310). Moreover, 3, 6, and 9
month survival was significantly better in the nTMS group (p=0.0298, p=0.0015, and
p=0.0167), while there was a trend in 12 month survival (p=0.0544).
The data illustrate that HGG patients benefit from preoperative nTMS motor mapping
with regard to craniotomy size, duration of inpatient stay, and eligibility for adjuvant
24
17. NOA Tagung, Heidelberg 2015
therapy. Even more important, a trend towards higher survival rates for the nTMS
group compared to the non-nTMS group was revealed.
Preoperative repetitive navigated transcranial magnetic stimulation for
language mapping and outcome improvement in
brain tumor patients
Nico Sollmann, Theresa Hauck, Sebastian Ille, Stefanie Maurer, Chiara Negwer,
Bernhard Meyer, Florian Ringel, Sandro M. Krieg
Department of Neurosurgery, Klinikum rechts der Isar, Technische Universität
München, Munich, Germany
Language mapping by repetitive navigated transcranial magnetic stimulation (rTMS)
is used for resection planning in patients suffering from brain lesions within regions
known to be involved in language function. Yet, we also need data that show whether
preoperative rTMS for language mapping also has an impact on the patients’ clinical
course and functional outcomes as it was proved for motor mapping.
We enrolled 25 patients with language eloquently located brain lesions undergoing
preoperative rTMS language mapping (GROUP 1, 2011-2013) with mapping results
not being available for the surgeon, and matched these patients with 25 subjects who
also underwent preoperative rTMS (GROUP 2, 2013-2014), but mapping results
were taken into account during awake mapping and tumor resection via
implementation into the neuronavigation system. In addition, cortical language maps
were generated by analyzing preoperative rTMS data.
Mean craniotomy sizes were significantly smaller in patients of GROUP 2 (GROUP 1
vs. GROUP 2: 50.1 ± 16.8 cm2 vs. 41.6 ± 10.4 cm2, p=0.0373), and postoperative
language deficits were found significantly more often in patients of GROUP 1
(p=0.0153), although preoperative language status did not differ between groups
(p=0.7576). Additionally, there was a trend towards less unexpected tumor residuals
(GROUP 1 vs. GROUP 2: 32.0% vs. 16.0%, p=0.1853), shorter surgery duration
(240.3 ± 53.2 min. vs. 215.5 ± 48.5 min., p=0.0914), and higher postoperative KPS
scores (80 vs. 90, p=0.2102) in GROUP 2 patients.
In general, the clinical course and functional outcomes of patients suffering from
brain tumors might be improved by preoperative rTMS language mapping.
Furthermore, reliable and precise language maps can be generated by rTMS in brain
tumor patients, and this approach has once again proven to be safe, effective, and
well-tolerable for the individual subject. However, more patients have to be enrolled
for more definite data.
Interhemispheric connectivity as a sign of language function at risk in brain
tumor patients
Nico Sollmann, MD1,2*; Chiara Negwer, MD1,2*; Lorena Tussis, BSc1,2; Theresa
Hauck1,2; Sebastian Ille, MD1,2; Stefanie Maurer1,2; Katrin Giglhuber1,2; Jan S. Bauer,
MD3; Florian Ringel, MD1; Bernhard Meyer, MD1; Sandro M. Krieg, MD1,2
* these authors contributed equally
25
17. NOA Tagung, Heidelberg 2015
1
Department of Neurosurgery; 2TUM-Neuroimaging Center; 3Section of
Neuroradiology, Department of Radiology; Klinikum rechts der Isar, TU München,
Germany
Resection of brain tumors in language-eloquent areas harbors the risk of
postoperative aphasia. Thus, various neuroimaging techniques have been introduced
for preoperative localization of cortical and subcortical language-related areas, but
there is no parameter known that sufficiently allows for reliable aphasia risk
assessment based on individual functional data prior to surgery. Since it has been
demonstrated via navigated transcranial magnetic stimulation (nTMS) that language
function can partially shift to the unaffected hemisphere due to tumor-induced
plasticity, the present study was designed to evaluate whether interhemispheric
connectivity (IC) detected by nTMS-based diffusion tensor imaging fiber tracking
(DTI-FT) can be used to predict surgery-related aphasia in brain tumor patients.
Thirty-eight patients with left-sided perisylvian brain lesions underwent cortical
language mapping by nTMS of both hemispheres prior to awake surgery. Then,
nTMS-based DTI-FT was conducted with a fractional anisotropy (FA) of 0.01 and 0.2
to visualize IC. Receiver operating characteristics (ROC) were calculated for the
prediction of a postoperative (irrespective of the preoperative state) and a new
surgery-related aphasia by the presence of detectable IC.
nTMS-based DTI-FT was successful in all patients. Regarding the correlation of
aphasia to IC, statistically significant differences were revealed for both evaluated FA
values. However, better results were observed for tractography with an FA of 0.2,
which led to a specificity of 93% (postoperative aphasia) and 90% (surgery-related
aphasia). For postoperative aphasia, the corresponding odds ratio (OR) was 0.1282
(95% CI: 0.0143 – 1.1520) and 0.1184 (95% CI: 0.0208 – 0.6754) for surgery-related
aphasia, respectively.
In general, IC detected by preoperative nTMS-based DTI-FT might be regarded as a
risk factor for surgery-related aphasia in brain tumor patients with a specificity of
90%. Despite the limitations of the DTI technique, this novel approach allows for
individual patient consultation prior to tumor resection in clinical practice.
26

Download Report