Agile in Life Sciences R&D


Applied Clinical Trials

Agile Product Development Methodology holds great promise for use in Adaptive Clinical Study Development.

Borrowed from the Software Industry, Agile Product Development Methodology holds great promise for use in Adaptive Clinical Study Development. Agile, is best described as having two basic properties-stability and dynamism, that allows organizations to respond and rapidly adapt to changes in the environment. As such, Agile seeks to institutionalize three critical transformative principles, the 3 E’s that applies to organizations and project management organization within them:

Envision. Create actionable, transparent, and strategic guidance focused on stakeholder satisfaction, enabled by strong sponsorship at the critical levels in the organization.

Engage. Create teams with a shared purpose and vision, end to end accountability and a sense of ownership; Create a culture that enables team interdependency.

Empower. Define clear, flat structures/roles, allow for robust communities of practice, encourage active partnership, provide next generation technology support, and unleash the innovative spirit.

These three critical principles enable agile readiness within organizations, and in turn, help set the foundational framework for project management offices to support agile projects.

Let’s look a little deeper into how this translates to Adaptive Clinical Studies.

2016’s 21st Century Cures Act requires the FDA to assist sponsors in incorporating complex adaptive and other novel trial designs into regulatory submissions. Although the Act does not explicitly state whether it intends for the FDA to finalize the 2010 adaptive design draft guidance it does suggest that Congress expects the FDA to provide greater support and encouragement for adaptive trials. Scott Gottlieb, M.D., the Commissioner of the FDA, has repeatedly exhorted and referenced the use of adaptive trial design as a lynchpin of an FDA regulatory strategy aimed at faster drug development and approvals.

While the promise of adaptive trials has been around for over a decade now, and some elements, such as futility analysis, are already widely used, adaptive design is still the exception rather than the rule. One recent estimate showed only 10% of device studies used adaptive designs (Yang et al., 2016).

By contrast, one of the early adopters of Agile transformation has been the software development industry. This industry not only embraced Agile, but also successfully defined a tool for implementing adaptive transformation, called Scrum. Scrum is used as a framework to rapidly develop and deploy products, ensure higher customer satisfaction while maintaining profits, reducing waste, and increasing employee engagement.

With so many hallmark similarities between Agile software development and adaptive clinical study execution, how can these same principles be better implemented to add value? 

Let’s first look at the traditional approach to product development-be it in software or biopharmaceuticals. This approach will certainly include sequential, segregated, pre-define phases with fixed user requirements or Target Product Profiles (TPP). In this environment, all development requirements are stipulated beforehand, with no mechanism for changing requirements once project development begins. This traditional approach is termed Waterfall methodology. When applied to conventional clinical trials, one accepts or rejects the null hypothesis in a well-defined population. There is no modification in trial design, statistical method, or patient population once the trial has started without documenting protocol amendments and seeking the permission of the IRB/Ethics Committee. In the Waterfall model, development is completed as single, multi-phase project, where each phase appears only once during the development lifecycle.

By contrast, Agile methodology flexibly allows project development requirements to be changed even after the initial planning has been completed through pre-specified modifications based on interim analysis. This methodology can be considered a collection of many different projects-all focused on improving quality by leveraging interim data analysis and feedback from both stakeholders and users. Unlike the Waterfall methodology, project phases such as designing, development, and testing, are completed more than once over the development lifecycle.

Agile and Waterfall methodology also take very different approaches toward quality and testing. In the Waterfall model, the Testing/Clinical Study phase follows the Build/Protocol Design phase, but with Agile methodology, testing is typically performed concurrently or at least in the same iteration.

However successful Agile methodology has been in the software industry, the methodology does not lend itself to a simple lift-and-shift to the Clinical Trial Operations environment. Installing a Certified Scrum Master Project Manager and employing Scrum frameworks is truly not enough to fully prepare a Clinical Operations organization for Agile readiness.

Agile methodology is best geared towards product development where requirements are expected to change and evolve, and requires a highly collaborative development process with fully engaged and accountable participants from customers-end users, regulators, payors, providers, partners, vendors, etc.

Empowering the teams through Holistic design and operational processes to support efficient and effective adaptive study execution is crucial. While the logistics of managing traditional, fixed format clinical trials are already quite complex, adapting trial design as results arrive adds exponentially to the complexity of design, monitoring, drug supply, data capture, and randomization. Some of the related components to this principle are looked into detail below:

Are decision rights and governance set up as to whom approves the changes to the trial? Are they sufficiently isolated from the conduct of the study, e.g. a data monitoring committee (DMC), independent statistician, or CRO? Are decision criteria, such as the interim analysis plan and DMC charter, kept confidential and described in documents only available to DMC members and the unblinded statistical team supporting the DMC (and not the sponsors).

Data Security
How often will the unblinded analyses be practical? How will the unblinding be protected by process and technology to prevent bias in treatment assessment and operational bias, where if information leaks out to study investigators even knowledge of adaptive choices and adaptation decision rules can introduce operational bias, as it may influence the way investigators treat, manage, or evaluate patients?

While monitoring is of course always critical are there robust monitoring processes in place to ensure protocol deviations are identified, investigated, and resolved before interim data analysis as to not possibly invalidate any decisions made?

Successfully standing up the necessary tools, methodologies, and frameworks associated with a future-state Agile environment is not for the faint of heart-especially with the added complexities of today’s Clinical Development environment and how well rooted the traditional Waterfall methodology is within the industry. 

Agile methodology in an organization, extends beyond methodologies and framework, to envisioning a culture change and a mindset shift that needs to cultivated across the organization by ensuring strong stewardship, actionable strategic guidance, and a strong change management framework.

Characterized by rapid communication across trial sites and with the data monitoring committee, flexibility in drug supply, customer feedback tightly coupled with rapid observable drug responses and interim analysis, and rapid bursts of work, or scrums for efficient design and fast computing in statistical work, the challenges are many-but so are the benefits. 

Today’s evolving Clinical Development environment requires a much more dynamic system Using the Agile approach originally developed for rapidly evolving user requirements in the software development lifecycle (SDLC), in Clinical Development allows for dynamic response, planning, and execution in response to interim data analyses.

This change in mindset starts at the leadership level and will need to be propagated to every employee within the organization with the three critical guiding principles-the 3 E’s.

If you would like to continue the conversation or learn how TayganPoint can help your clinical development organization navigate the challenging world of Agile you can reach us at the email addresses below or connect with us on LinkedIn.


Christina Corera, B.E. Computer Science Engineer, PMP | Consultant | TayganPoint Consulting Group |  |  LinkedIn

Jeffrey S. Handen, Ph.D. | Senior Consultant | TayganPoint Consulting Group |  |  LinkedIn

© 2024 MJH Life Sciences

All rights reserved.