Take the innovation leap: Four things pharma companies can do now for a synthetic data-driven approach to clinical trial design. https://accntu.re/3vjVjVs
2. With drawn-out timelines and billions
in investments, traditional clinical trial
methods are increasingly a barrier to
cost-efficient and timely drug
development.
3. A role for
synthetic data
What is it? Synthetic data
is generated by applying a
sampling technique to
historic data or by creating
simulation scenarios where
models and processes
interact to create completely
new data not directly taken
from the real world.*
How will it improve clinical
development?
Minimize or replace the control arm of a trial
Reduces the demand for patient
recruitment which saves time and
money, and mitigates ethical issues
around patients receiving placebos
Model target patient population and define
the boundaries of a trial
Optimizes design and feasibility which
positively impacts operational success
*Adapted from Gartner
4. Overcoming obstacles
Pharma companies need to exploit the dynamics of data along these three dimensions
Velocity
Pharma companies are often
utilizing historical and distinct
data sets without incorporating
new data as it is generated and
have limited capacity to interpret
incoming data.
Variety
Pharma companies typically do not
have broad external data and have
been reluctant to look beyond their
own clinical data to other sources,
such as industry data from other
clinical trials and real-world data
from EMRs.
Volume
Pharma companies have their own
separate data, but this is just a
fraction of the potential data
available for use at an indication
level and are usually not sufficient
for rigorous statistical analysis.
6. Where to start
01 02 03 04
Draft your big
plan while getting
started with pilots
Create a bold global
vision for leveraging
synthetic data in clinical
trials; prioritize assets
with which to pilot the
new approach.
Take stock of your
data and look
outside for more
Supplement internal data
with external data
sources to achieve a
robust validated data set.
Deploy smart
algorithms for
“what-if” analysis
Integrate and automate
algorithms into analytics
platforms to evaluate and
predict potential
outcomes.
Evolve your
operating model
New governance, skills
and processes are needed
so that predictive data
analytics can inform
clinical deployment plans.