Advances in the understanding of glioblastoma at a molecular level along with technological
progress have led to the identification of key genetic alterations, not only in scientific
projects but also in every-day clinical practice. These alterations increasingly refine the
sub-classification of glioblastoma and the introduction of molecular markers in this
classification, which ultimately may allow defining subset specific treatments.
The present umbrella concept for multiple biomarker-driven subtrials anchors at the
Heidelberg-based INFORM registry trial in recurrent pediatric malignancies, where in analogy
to NCT Neuro Master Match (N²M²) whole-exome, low-coverage whole-genome and transcriptome
sequencing is used to identify targeted agents, single or in combinations according to a
dedicated algorithm. It also shares conceptual similarities with international projects
currently developed for lung and breast cancer. Finally, approaches to use molecular
information in glioblastoma for the definition of a therapy at progression are also planned
by the "Defeat Glioma" Consortium in the US and a group of excellence centers also in the US.
The N²M² concept excels the aforementioned initiatives in the strict focus on newly diagnosed
patients, the option to cross-validate molecular biomarkers in an already analyzed
contemporary cohort of glioblastoma patients analyzed in the German Consortium for
Translational Cancer Research (DKTK) and the use of a parallel group treated with
standard-of-care (SOC). Further restriction is made by the inclusion of patients only with a
low likelihood to benefit from the SOC, temozolomide (TMZ) chemotherapy on the basis of
O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation, allowing to replace TMZ
with a molecularly targeted agent in combination with radiotherapy (RT) in each of the
experimental subtrials. Replacing TMZ by an experimental agent in the primary
chemo-radiotherapy has been done in at least four completed trials, albeit without
pre-selection of a targeted therapy but the more opportunistic use of an available drug with
no known MGMT interaction. Focusing on newly diagnosed patients not only harbors the greater
likelihood of impact on the disease, but also allows addressing questions on acquired
resistance in the more likely obtained tumor tissue at recurrence.
N²M² is formally divided into a DISCOVERY and a TREATMENT aspect. DISCOVERY starts with an
(epigenomewide) Illumina methylation array and a panel sequencing followed by an appropriate
and accepted standard method (Sanger Sequencing or Immunohistochemistry) within the scope of
these methods for target validation prior to any suggestion for patient allocation into one
of the subtrials. Suggestions for patients' allocation to one of the subtrials will be based
on results of accepted standard methods. These data will be generated from formaline-fixed
paraffine embedded (FFPE) tissue within in 2-3 weeks after surgery. The Heidelberg site has
already established the next generation gene panel sequencing (used for target discovery in
case of N²M²) and genome wide DNA methylation analyses for aiding daily routine. To meet the
criteria for a safe use of these data for decision-making the orthogonal, standard methods
performed in the Institute of Pathology are supplemented.
DISCOVERY also includes the use of whole exome, low-coverage whole-genome, and transcriptome
sequencing, the methylome analysis, and gene expression arrays to find new, unexpected
targets and get a more comprehensive view on affected pathways. Discovery is also the driving
force behind the work on resistant tumors. The latter may result in individual treatment
decisions at recurrence, knowing there are no relevantly active treatments in this setting.
TREATMENT is driven by a match/no match decision rendered in an algorithm that will be
subject to refinement in the process of the project, both by data generated in N²M², but also
external evidence; i.e. there may be some linear relations between an alteration, e.g. BRAF
V600E mutation and a distinct treatment or some others, but it is expected that these linear
relations will be replaced in a learning system by relations that take upstream and
downstream target alterations and also parallel signaling pathways into account and may
therefore already predict a certain likelihood of resistance development.
In detail, FFPE tissue (and blood) from patients diagnosed with a glioblastoma harboring an
unmethylated MGMT promoter after informed consent will be subjected to a (epigenomewide)
methylation array and panel sequencing as well as the appropriate methods to validate any of
the trial immanent targets, if they are present, with results available within a maximum of 3
weeks postoperatively. This allows for a timely decision at the molecular Neurooncology Tumor
Board and a timely initiation (within 4 to 6 weeks) of the postoperative treatment. Further
examinations on fresh tumor tissue (and blood) such as whole exome, low-coverage whole genome
and transcriptome sequencing as well as expression arrays will be done to enhance the
scientific background on the tumor tissue. These data will not be used for decision-making,
Runs already done within the INFORM project with glioblastoma samples and also dry runs
(n=43) for the N²M² project support the feasibility of the timelines and principal options
for discovery.
Matching will be defined as a molecular situation, which makes treatment with RT and a
matching targeted drug from a prespecified warehouse separated in subtrials meaningful.
Patients will be informed about the identified treatment option within the "matching"
open-label, parallel group Phase I/IIa trial. As for 2 of the experimental compounds (APG101
and Atezolizumab) no specific biomarker is validated at the moment, the nonmatching patients
will be equally allocated to receive either APG101, Atezolizumab or the current SOC
(radiochemotherapy with TMZ, TMZ-group). Patients allocated to the TMZ-group will serve as a
meaningful control group with basic efficacy parameters documented, if consent has been
obtained.
The objective of N²M² is the improvement of overall survival of patients with glioblastoma
with an unmethylated MGMT promoter based on molecular characterization and use of targeted
compounds in a modern trial design. The progression-free survival rate at six months (PFS-6)
will be used to make decisions.
Parallel and ongoing translational projects within the DKTK will examine prognostic
properties of the biomarkers identified to drive therapy decisions in this trial. Trial
accrual will be asymmetric into the different subtrials. It is expected that 75-100 patients
will be accrued into this trial per year at about 14 sites in Germany (mainly Deutsches
Konsortium für Translationale Krebsforschung, DKTK and Neuroonkologische Arbeitsgemeinschaft,
NOA). Importantly, the parallel SPECTAbrain initiative of the European Organization for the
Research and Treatment of Cancer (EORTC) is synergistic and not competitive to our study
proposal since it is focused on the treatment at recurrence, using paraffin-embedded tissues,
panels/arrays only and it would be desirable that data from these initiatives are looked at
in a joined manner.