Transcript Integrating CHIC technologies into a clinical research application

Integrating CHIC technologies into a
clinical research application framework
(“CRAF”) for cancer modeling
Kostas Marias, Foundation for Research and TechnologyHellas, Greece
Stelios Sfakianakis, Foundation for Research and
Technology-Hellas, Greece
Aims of this talk
• Discuss the clinical relevance of predictive oncology
and in particular computational models.
• Discuss their role in assisting the clinician in disease
diagnosis and therapy decision making.
• Introduce CRAF as an end result of CHIC project
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
2
Is progress in cancer clinical trial slowed down?
• Increasing regulatory rigidity of clinical trials
• More complexity
• More bureaucracy
• Increasing costs
• Time from drug discovery to marketing
• 8 years in 1960  12 - 15 years in 2002
• Costs per life year gained: $2.700.000
Stewart DJ et al.: Equipoise Lost: Ethics, Costs, and the regulation of Cancer Clinical Research. J Clin
Oncol 17:2925-2935, 2010
Prof. Norbert Graf Kick-off Meeting p-medicine (270089) - February 14-15, 2011 - Homburg/Saar
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
3
Is progress in cancer clinical trial slowed down?
Without improved mathematical and
computer modeling (stochastic,
Markov process modeling among
others) we will fall short of progress!
100 %
Cure rate
Stephen M. Pribut, Professor of Surgery,
George Washington University School of Medicine
0%
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
timeline
4
Medicine is expanding
• Preventive
• Predictive
• Personalized
• Participative
• Psycho-cognitive
12/8/2016
In Silico Modeling provides opportunities
“Through modeling, researchers can now
predict the behavior of some of the
toughest cancers”
http://www.nature.com/nature/journal/v491/n7425_supp/full/491S66a.html
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
6
What about populations and clinical trials?
“Mathematical modelling is also helping to speed up the
clinical trials that test those drugs. Testing all the possible
permutations of drugs and doses to determine the best course
of therapy might take millions of clinical trials — and there are
nowhere near enough patients or resources for that”
http://www.nature.com/nature/journal/v491/n7425_supp/full/491S66a.html
In the period 2008-2011, the Office of Clinical Pharmacology
at the US Food and Drug Administration (FDA) received up to
25 submissions containing in silico clinical trials (through PBPK
analyses)
Zhao, P., L. Zhang, J. A. Grillo, Q. Liu, J. M. Bullock, Y. J. Moon, P. Song, S. S. Brar, R.
Madabushi, T. C. Wu,
B. P. Booth, N. A. Rahman, K. S. Reynolds, E. Gil Berglund, L. J.
Lesko and S. M. Huang (2011). "Applications of
physiologically based pharmacokinetic
(PBPK) modeling and simulation during regulatory review." Clin Pharmacol Ther
89(2):
259-67.
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
7
The EC Virtual Physiological Human effort
Virtual
•
Sophisticated computer models; usually complex workflows
Physiological
•
•
Function of living systems
Biology, biochemistry, biophysics…
Human
•
•
Focus is human healthcare, ‘business’ is medicine
…the modelling process is relevant to other species, disciplines
Initiative with significant funding from the EC
Community of over 2000 European researchers
Purpose: To improve diagnosis/treatment by physics simulations
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
8
VPH Integration strategy
Spatial
1m…
1mm…
1ns…
…1s…
1m…
1nm…
Temporal
…70 years
Organs &
Systems
http://bodybrowser.googlelabs.com
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
9
Microscopic to Macroscopic cancer modelling strategies
Organ
ECOSYSTEM LEVEL
Treatment
planning
POPULATION LEVEL
ORGANISM LEVEL
Tissue
Top-down
Bottom -up
Cell
SYSTEM LEVEL
ORGAN LEVEL
TISSUE LEVEL
CELLULAR LEVEL
Molecules
SUBCELLULAR LEVEL
MOLECULAR LEVEL
Drugs research
ATOMIC LEVEL
G.Stamatakos "In Silico Oncology Part I: Clinically Oriented Cancer Multilevel Modeling Based on Discrete Event Simulation" In T.Deisboeck and G. Stamatakos Eds 407-436 2011-01-01 CRC Press,
Print ISBN: 978-1-4398-1440-6 eBook ISBN: 978-1-4398-1442-0 DOI: 10.1201/b10407-19 Boca Raton, Florida, USA, 2010
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
10
What about predictions?
In Silico Predictive models
Predict the optimal
treatment/ therapy outcome
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
11
ContraCancrum project Clinical Workflows
http://contracancrum.eu/
Schema of the Glioblastoma workflow
Schematic for New GBM
Sensitivity Evaluation based on
TCGA glioblastoma subtypes.
Subtype sensitivity was
independently validated by
Ducray et al. 2010.
This was applied to classify the
repository of ContraCancrum
patients where microarray data
was available.
G.S.Stamatakos, In Silico Oncology Group Leader, ICCS-NTUA
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
12
Novel approach in multiscale Predictive Oncology Clinical Decision Support
Multi-scale predictive oncology start
Molecular level models
Drug-binding evaluation
Drug-sensitivity evaluation
Informs on drug ranking
Informs on patient sensitivity
molecular data
clinical data
Tissue level models
treatment data
Oncosimulator
imaging data
Clinical Decision Support Layer
Prediction:
Is treatment ( i )
beneficial?
Simulate ( i+1) line
treatment
no
yes
Predictive Oncology Optimization Layer
Optimize modeling components
no
Is response to treatment
as predicted?
yes
Predictive Oncology
Decision Confirmed
The envisaged use of CC environment, introducing a new paradigm in designing
multi-scale predictive oncology clinical decision support protocols
http://contracancrum.eu/
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
13
What about validation of cancer models?
Tumor
change
Tumor is in fact
regressing!
Tumor must be
shrinking!
Diagnosis and start of
treatment MRI
Follow up MRI
time
Illustrating the difficulties in validating predictive models with limited time points (which is normally the
case in the clinical setting). This limitation was discussed-introduced by Norbert Graf, Director of Pediatric
oncology at the University of Saarland.
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
14
What about validation of cancer models?
• If it is to be used in clinical practice In silico modelling
should be validated so as to prove the clinical relevance
• The use of animal models (mice) is an attractive
alternative for conducting initial validation experiments
of models
• Validation using retrospective hospital data is very
difficult (if not impossible) mainly due to the lack of
temporal cancer data needed to test the accuracy of
the model predictions
• CHIC project developed a dedicated environment for
demonstrating the clinical relevance of multiscale
cancer modeling.
Employing in-vivo Molecular Imaging in Simulating and Validating Tumor Growth
Eleftheria Tzamali, Rosy Favicchio, Alexandros Roniotis, Georgios Tzedakis, Giorgos Grekas, Jorge Ripoll, Kostas Marias, Giannis Zacharakis, and Vangelis Sakkalis , IEEE EMBC 2013.
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
15
Evolution of cancer modelling projects-priorities
Development of cancer
models and proof of
concept
12/8/2016
Demonstration of the
Clinically-driven cancer
modelling applications and clinical relevance of cancer
international collaboration hyper-modelling
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
16
The CHIC Clinical Research Application Framework
• Computational modelling can play a pivotal role in
the translation of research to clinical practice.
• It can link patient related data and biological and
disease specific information to provide key insights
for the benefit of cancer patients.
• The “Clinical Research Application Framework”
(CRAF) of the CHIC Project provides the bridging
between the Computational Modelling Research
and the Clinical practice.
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
17
The CHIC Clinical Research Application Framework
• CRAF is the entrance to the CHIC platform for
the clinicians
• Effectively hiding the complexity of the platform
• Allowing the clinical users to take advantage of
the CHIC outcomes
• A comprehensive suite of technical components
that bridge the gap between the computational
work in CHIC and the delivery of care and
research in the clinical setting.
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
18
The CHIC Clinical Research Application Framework
High Level view
Model Building
& Experimentation
Model
Repository
Hyper
modelling
Editor
Model
Execution
Computational
Modeller
Model
Model
Model
CHIC Research Domain
Model
Patient
Specific Data
Clinical Practice
CRAF
Answer
Clinical Domain
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
Clinician
19
The CHIC Clinical Research Application Framework
Conceptual model
• We consider 3 dimensions
when a Clinician interacts
with CRAF:
• Models: the computational
and simulation artefacts of
the CHIC research platform
• Data: patient specific
information coming from the
clinical domain
• ”Hypotheses”: Clinical
questions to be answered by
concrete models for a given
patient
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
20
The CHIC Clinical Research Application Framework
Login
• Required for authorizing patient data access to the
users
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
21
The CHIC Clinical Research Application Framework
Welcome page
• Select a type of Cancer or a specific patient
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
22
The CHIC Clinical Research Application Framework
Selection of Cancer Type
• Depending on the availability of Models, a few
select cancer types are shown:
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
23
The CHIC Clinical Research Application Framework
Selecting a Clinical Question
• A Clinical Question correspond to a specific hypothesis or
treatment plan the Clinician may present
• Example: the efficacy of pre-operative chemotherapy for a
given patient
12/8/2016
24
The CHIC Clinical Research Application Framework
Select a Patient
• Shows the list of Patients that the Clinician has
access to
• For each patient, the set of available data are also
shown
12/8/2016
26
The CHIC Clinical Research Application Framework
Selection of the Model to run
• The selection of Clinical Question and/or the
Patient specifies the set of applicable models:
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
27
The CHIC Clinical Research Application Framework
Set the Input parameters
• Patient data are automatically set
• Some parameters, e.g. the drug administration
plan, can be specified by the user
12/8/2016
28
The CHIC Clinical Research Application Framework
Monitor execution progress
• The currently running models as well as the
previous executions are shown in the History page:
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
29
The CHIC Clinical Research Application Framework
Results
• When the run completes, full details of the
execution are shown:
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
30
The CHIC Clinical Research Application Framework
Report generation
• A PDF report, to
be put in the
clinical record of
the patient, is
produced:
• Contains all the
details, graphs,
inputs, and
outputs of the
execution
12/8/2016
Integrating CHIC technologies into a clinical
research application framework (“CRAF”) for
cancer modeling
31