Modernizing Clinical Trials: Digital Technologies and the Cloud
2
Published by FiercePharma
D
eveloping and bringing a new drug to
market is costly, with estimates ranging
from $650 million to $2.9 billion, a 145
percent increase over the last ten years.
1
E xecution of clinical trials is historically the most
expensive part of the process, accounting for around
two-thirds of total R&D costs
2
, and taking upwards
of 10 to 15 years to complete. Factors contributing
to these high costs include enrollment of patient
cohorts that are not optimal for the trial, unproductive
trial sites, low patient engagement and dif ficulty in
sharing and analyzing resulting data.
The problems begin with the limitations of sponsors'
ability to identify patients in the active disease state
and determine the best sites to enroll them. By one
calculation, around 30% of clinical trial sites never
enroll a subject.
3
These unproductive trial sites delay
data generation and cause a large amount of wasted
ef for t, which contributes to the high cost of drug
development.
Similarly, once a trial is underway, sponsors need agile
tools to incorporate the use of mobile technologies,
such as wearables and mobile devices, which could
generate a wealth of data while promoting bet ter
patient engagement. To compound mat ter s, the
dif ficulty associated with sharing, aggregating and
analyzing trial data can limit the speed and efficiency
of studies at each step in the process.
The ef ficient generation of high-quality clinical trial
data is critical to success for new therapeutics. Yet,
1 ht tps://w w w.polic ymed.com/2014/12/a-tough-road- cost-to -
develop - one -new- drug-is-26 -billion-approval-rate -for- drugs-
entering- clinical- de.html
2 ME ASURING THE GLOBAL BIOMEDICAL PULSE The
Biopharmaceutical Investment & Competitiveness (BCI)
Sur vey – 2015. Available at: ht tps://w w w.pugatch- consilium.
com/repor t s/BCI%202015%20 -%20Measuring%20 the%20
Biomedical%20Pulse.pdf. (Accessed: 5th September 2018)
3 Non-Enrolling Sites Come at a Price | Geeks Talk Clinical.
Available at: ht tps://blog.mdsol.com/non- enrolling-sites- come-
at-a-price. (Accessed: 5th September 2018)
companies continue to struggle with the challenges
of enrollment, engagement and data sharing. Against
this backdrop, pioneering companies are turning to the
AWS Cloud to modernize their clinical trials with the
analytic capability to optimize studies by predicting
the right patients to enroll, securely coordinating
and sharing data and incorporating regulated digital
technologies.
HOW SCALABLE CAPACITY CAN
STREAMLINE TRIAL DEVELOPMENT
Companies such as Bristol-Myers Squibb are performing
in silico clinical trial simulations using cloud-based high-
performance computing (HPC) to help them optimize
the design of early-phase trials. Bristol-Myers was
able to reduce total analysis time by 98% by running
simulations in support of a pediatric phase 1 study on
the AWS Cloud instead of their on-premises system.
4
It
can be challenging to scale the computational power
necessary to run trial simulations on-site, as it ties up
resources and causes bottlenecks for other researchers.
However, vir tually unlimited HPC resources can be
accessed on-demand in the AWS Cloud.
"Because of the compute capacity that we're able to
derive from AWS, we can now hand out dedicated
4 AWS Case Study: Bristol-Myers Squibb. Amazon Web Ser vices,
Inc. Available at: ht tps://aws.amazon.com/solutions/case -
studies/bristol-myers-squibb/. (Accessed: 11th December 2018)