Ensuring R&D investments deliver maximum potential: Notable ways to enhance clinical development strategy and design

laboratory drug development

As the drug development landscape becomes more complex, pharmaceutical and biotech companies understand their business strategies have to evolve. This need is putting a spotlight on how clinical trial sponsors are increasingly looking to fine-tune their trial programmes and related strategies as early as possible, to accelerate development efforts with efficiency, cost, and time in mind.

As innovative thinking meets advanced tech-enabled solutions and a breadth of relevant data that is exponentially growing, sponsors can gauge and integrate strategic approaches and scalable tools into research and development efforts like never before. Below are several noteworthy advancements helping sponsors to optimise trial strategy and design.

AI/ML for informed decision-making

Understanding an asset’s value in a highly competitive market is critical for trial sponsors, especially biotechs. However, given the breadth of data available to sponsors today, using traditional manual processes to extract insights that guide clinical planning is not possible. Industry stakeholders are now integrating artificial intelligence/machine learning-driven solutions and expertise into clinical strategy planning and design. With access to multiple trusted sources (e.g., trial registries, scientific literature, and toxicology testing databases), AI/ML is helping sponsors sift through and aggregate information and extract meaningful insights for analysis.

These tools generate deep insights and showcase hundreds of thousands of pathways helpful to programme strategy and design. This is why it is critical for solutions to be underpinned by experienced data scientists and AI/ML engineers who understand sponsors’ specific questions and can apply that knowledge when building the training models to pinpoint what insights are eventually extracted. With that awareness, sponsors can secure predictive insights for areas important to their programmes, such as:

  • Gauging the market landscape for a specific therapeutic area or mechanism of action.
  • Evaluating potential pricing and reimbursement outcomes country-by-country for treatments based on a specific disease of interest.
  • Better pinpointing estimated dates of a trial programme transitioning to the next stage and regulatory approval timing, which can improve projections for development timelines and related targets.
  • De-risking multi-year R&D investment forecasts through AI/ML-enriched data and predictions based on scientific rationale, clinical feasibility, and market attractiveness.

When based on sponsor and trial programme goals, the level of insight produced by AI/ML-driven solutions helps improve trial design and address trial complexities through novel design approaches and more. Sponsors can map out their programmes’ risks and rewards, allowing for smarter decision-making as to which pathways, if any, are to be taken for treatment development and related trial design and execution.

Early and deliberate focus on diversity

The industry’s ongoing commitment to improving diversity and inclusion in clinical trials, especially among traditionally underserved populations, has only grown stronger post-pandemic. Sponsors and clinical research organisations are creating solutions to increase trial awareness and access through any capacity within clinical trial activities, including approaching each trial with a deliberate focus on diversity goals during design stages. As such, it’s helpful to gather intelligence about varying motivations for trial participation by race and ethnicity and what design elements may affect willingness to enrol. These qualitative insights can help sponsors fine-tune trial protocols earlier in the process, potentially avoiding unnecessary protocol amendments and time delays.

Nuances in what drives patients to participate in trials can be critical to how well a trial’s design will work. For example, if a trial requires overnight stays or longer site visits, does that influence black/African American patients’ willingness to participate, compared to other communities (e.g., Asian, Hispanic, or Caucasian)? Recent survey findings have shown there are intricate differences, and that visit lengths or overnight stays did not influence black/African American respondents’ willingness to enrol as much as it did others’.

Taking the time to actively listen to patient perspectives early in the process helps sponsors understand patient burden better. Through advanced data-driven methodologies, sponsors can now analyse how burdensome their protocols may be for target populations and where adjustments can be made earlier, mitigating some challenges in recruitment and engagement and related protocol amendments.

Transforming data collection and flow

Data management plans are traditionally developed only after the study protocol is finalised, which limits the ability to connect and streamline data strategies. When study teams try to optimise data flow strategies and manage protocol execution in parallel, it creates inefficiencies, adding unnecessary time, effort, and frustration. Meanwhile, today’s trials collect approximately three times as much data as they did 10 years ago, and it comes from a growing number of sources (e.g., connected devices). It isn’t feasible to manually collect, monitor, clean, and analyse millions of data points.

Creation of comprehensive data strategy, from data collection to endpoint analysis, requires thoughtful and careful planning during protocol design. By using quality by design principles, teams should be able to recommend the optimal data collection, vendor selection, and end-to-end standardisation needed to support the needs of patients, sites, sponsors, and study teams. The result is a comprehensive end-to-end data management strategy providing continuous data flow, enabling automated data review and associated data insights throughout the study lifecycle.

When setting up effective data management strategies, sponsors need to consider multiple aspects, including:

  • Selecting patient-centric data collection methods (e.g., eSource or devices/sensors) to remove data transcription and reduce site queries, while increasing data quality.
  • Selecting data vendors who can provide continuous or daily data transfers to ensure study teams have the most current data at all times.
  • Identifying risks and mitigation strategy related to data collection and data flow during protocol design.
  • Ensuring data flow strategy does not add burden to patients or sites.

To build out comprehensive data management strategies, sponsors are increasing engagement with the necessary experts during trial design planning, including data strategists, digital health solutions specialists, standards engineers, therapeutic specialists, biostatisticians, etc.

Thinking through all aspects of study execution, sponsors who recognise the value of achieving quality data through quality by design principles and of data flow in trial activities will see smooth trial execution and improved decision-making capabilities to accelerate start-up, database lock, and submission timelines.

Always room for enhancement

We are in a time where sponsors are having to “do more with less” as they find ways to optimise trials and stay competitive, with a careful eye on timelines and cost. We are seeing in real time how earlier efforts to optimise trial strategy and design from the start can be beneficial to sponsors and the patients they aim to serve.

Knowing this, as we look ahead to the landscape in 2024, the industry will take a closer look at learnings from the spectrum of advanced capabilities, including those noted above, leveraged for improved trial strategy and design, and expand upon what is most beneficial for their business objectives.

About the authors

Greg LeverGreg Lever is director of AI solutions delivery at IQVIA. With more than 13 years of life sciences and technology experience, Lever currently helps clients discover innovative ways to bring therapies to patients faster within IQVIA’s Applied Data Science Center’s consulting sales team. Previously, he led a team of machine learning engineers within the Analytics Center of Excellence. Lever has worked with several technology start-up companies in London and helped see Genomics England’s 100,000 Genomes Project through project completion. He received his PhD at the University of Cambridge, combining quantum physics and ML to develop new approaches for small-molecule drug discovery and has worked as a postdoctoral associate at MIT.

Denise MesserDenise Messer, MA, is director of design Analytics at IQVIA’s Applied Data Science Center. With more than 25 years of clinical research and trial experience, Messer has an extensive background in clinical trial planning and design. In her varying roles, she has helped define methodologies to assess and score trial patient burden, ensure the voice of the patient is incorporated into design, and examine protocol complexity. At IQVIA, Messer has helped develop the IQVIA Data-Informed Patient Assessment, leveraging data to highlight areas for protocol optimisation before operationalisation, including the development of a patient burden algorithm and protocol scoring benchmarks.

Sabrina SteffenSabrina Steffen is head of data strategy and innovation for data management, biostatistics connected devices, medical safety services and lifecycle safety at IQVIA. Steffen has worked in clinical research for 18 years, within data management, risk-based monitoring, data strategy, process improvement, and innovation. For the last eight years, she has led the data strategy and innovation team, overseeing large-scale process and technology transformations from inception through delivery and change management to achieve fully embedded technology-enabled processes.

profile mask
10 January, 2024