Improving Pharma R&D Efficiency

January 15, 2019
Icon plc

Applied Clinical Trials

Declining research and development (R&D) efficiency is one of the biggest challenges pharmaceutical manufacturers face today. Total sponsor cost per new drug approved in the US jumped 145 percent in just 15 years to more than $2.5 billion in 2014, according to the Tufts Center for Drug Development. At the same time, just seven percent of first-in-human drugs gained FDA approval.[1]

These inefficiencies largely result from an outdated clinical trial model. Nearly $1.5 billion per approved new drug is attributed to clinical development, the majority of which is for clinical studies. 

Designed primarily to meet regulatory requirements for therapies targeting large patient populations, the existing development model of three discrete, fixed trial phases lacks the flexibility, analytical power, and efficiency to address today’s development demands.

These include managing product pipelines of complex new therapies targeting smaller and more heterogeneous patient populations[2], meeting rising standards of evidence from payers that are moving towards value-based reimbursement models, and addressing the patient’s rising role in care decisions.

With development cycles becoming too long, trial complexity increasing, and greater scrutiny of the economic value of new treatments, pharma R&D business models are under significant pressures to improve R&D efficiency.

Our industry must adopt an entirely new framework for development infrastructure and processes. Focus must go beyond cost containment alone, and instead unlock value through fundamental reforms that allow sponsors to move more treatments to market faster.

In an ICON industry survey of pharmaceutical executives and professionals on Pharma R&D Efficiency, the challenges most frequently cited are:

  • patient enrolment-56% of survey respondents 

  • site start-up-43% of respondents

  • regulatory approval delays and changes-43% of respondents

These operational issues reflect the difficulty of designing studies that address critical patient and investigator needs, as well as evolving regulations.

Whilst our survey shows that companies realise the need for a holistic effort to transform trials, only one in five survey respondents stated their organization currently has a holistic/integrated approach to drive clinical trials transformation. In too many organizations, efforts to drive large-scale operational efficiencies remain siloed and therefore potentially not delivering their full potential for full process efficiencies and economic value. Efforts need to be integrated if they are to be effective.

Elevating Efficiency & Enhancing Trial Savings

Patient identification and recruitment and risk-based approaches to study monitoring are expected to have the most impact in transforming the efficiency, speed, and productivity of clinical development.

In our survey, the top five key areas identified by industry experts as having the most potential for generating savings and improving trial efficiency were:

  • Improving Protocol Development-38% of respondents

  • Study Start Up Activities-37% of respondents

  • Patient Recruitment & Retention-37% of respondents

  • Vendor Selection & Management-32% of respondents

  • Study Monitoring-25% of respondents

Declining pharmaceutical R&D efficiency and the resulting deterioration in return on investment is largely driven by lengthening development cycles. These, in turn, typically involve increasing trial complexity and regulatory approval delays. These complexities and delays are symptomatic of deep structural changes in therapeutic markets that conventional clinical trials are simply not designed to address.

Market changes

These market changes include:

Smaller Targets

Driven by scientific advances in areas including biochemistry, genomics, and biomarkers, the market for new therapies has moved toward targeted therapies and orphan indications, with big gains in approvals for neoplastic therapies and declines in cardiovascular and broad spectrum anti-infectives since 1980.[3]

The smaller potential markets mean the R&D enterprise-and clinical trial designs and procedures-must be tightly focused on patient needs, relevant clinical and research expertise, and maximizing efficiency in demonstrating safety and efficacy. The traditional three-stage randomized clinical trial structure was built to study drugs with mass market potential to treat large populations. Today’s drug targets have fundamentally different economics.

Personalized Medicine

Taking the smaller target trend to its logical conclusion, the market for personalized medicine is growing exponentially. New product offerings target specific biomarkers, such as biologic chemotherapy agents, or even individual patients, such as CAR-T immunotherapy. Similarly, therapies that combine mobile sensors and devices with drugs and delivery devices, such as apps assessing the daily effects of Parkinson’s or other mobility-restricting conditions, require evidence of real-world efficacy and safety that cannot be generated in a controlled environment.

This is also being driven by patient demands: patients are increasingly active, informed consumers entering clinics armed with research and opinions about their healthcare options. They desire data on similar patients while calculating the benefits and risks of their choices. "Precision” or personalized medicine represents the ideal standard of care: the individualized, evidence-based treatment that provides the best opportunity for a positive outcome. This desire for evidence of a treatment’s relevance, in conjunction with evolving pressures for clinicians and payers, has fuelled the meteoric success of genomic tests and targeted oncologics.

Value-Based Care

Rising healthcare costs as a percentage of GDP is driving greater scrutiny of the economic value of new treatments by government and private payers. In addition to efficacy and safety, clinical trials increasingly must demonstrate a meaningful impact on patients’ lives. This is particularly true for high-cost therapies targeting smaller patient groups, many of which struggle to be covered by national health systems and private insurers. Screening patients to identify potentially better responders and linking payments to individual patient outcomes are among the measures payers are negotiating with sponsors to ensure they are getting value for the money they spend.

As industry shifts from a fee-for-service healthcare delivery model to one with a greater focus on the value a treatment brings to patient health and system outcomes, drug and medical device developers are taking a more integrated approach in developing their product’s regulatory and reimbursement strategies.

By considering health technology assessment, value frameworks and market access and reimbursement challenges earlier, often as early as Phase I, developers are producing more optimized evidence generation plans, and looking carefully at how each piece of evidence impacts not only product authorization, but earlier and broader market access, and greater product adoption by health care systems, providers, and patients.

The impact of digital disruption

New technologies are enhancing the efficiency and scope of clinical trials through:

  • Patient-focused technologies, such as mobile sensors and smartphone apps

  • Big data and predictive analytics which enable quick identification of promising study subjects and sites

  • Artificial intelligence (AI) processing large amounts of data to help guide patient management and protocol design

  • Electronic Medical Records (EMR / EHR) increasing data collection reach and efficiency, and help better integrate trials into clinical practice

Wearables & Mobile Devices

Over the last number of years the use of mHealth technologies is being embraced by patients, healthcare providers, and payers.  It has been estimated that over 7.1M patients’ worldwide benefit from remote monitoring and the use of connected medical devices as part of their care regimen[4].  

mHealth device technology has evolved to the point where it is now possible to collect a vast array of physiological data including vital-signs such as heart rate, respiration rate, oxygen saturation, continuous glucose monitoring, sleep and activity data, and using advanced analytics to monitor patients in their own home outside of the hospital environment.

mHealth patient-focused technologies, including mobile sensors, smartphone apps, and telemedicine, are seen as ways to collect richer patient data, develop new endpoints and help design novel kinds of trials that may better demonstrate real-world clinical and functional value.

Within trials, whilst mobile apps and sensors can measure such symptoms and signs, diagnostic apps, reminders, and telemedicine can help keep patients engaged and on protocol. 

Currently the penetration and use of wearables and devices in the Pharmaceutical Industry is still somewhat limited-although in our survey respondents cited Smartphones, Wearables & Sensors as well as Big Data and Predictive Analytics as amongst the top disruptive technology trends which will have the greatest impact on clinical trial operations.

Industry concerns focus on a number of key areas: Patient Acceptance, Device Suitability, Data Complexity and Insight Generation, Operationalization, Privacy and Security Issues, and Regulatory Acceptance. For a more detailed consideration of these issues than this article space allows, please review our blog ‘How to implement successful digital clinical trials’ available from

Sensors also may support development and measurement of new endpoints that are more relevant to patient needs. However, as with big data, though, new outcome measures and endpoints using mobile devices must be rigorously validated.


Big Data

The world’s capacity for producing data is expanding exponentially, and that data, from medical and non-medical sources, has the potential to greatly increase the efficiency of clinical R&D. Leveraging big data and predictive analytics can enable the efficient identification of promising study subjects and sites, as well as risk-based monitoring of trial performance in real time.

However, establishing and structuring data sourced from diverse systems is required, and this can be technically daunting. Outcomes based on such data also must be modelled and validated, and this, too, requires significant expertise.

Artificial Intelligence (AI) & Statistical Analysis

AI is evolving alongside the data explosion-AI-enabled measures include data integration, data management, and interpretation. These can improve trial performance at every level, from enabling risk-based trial monitoring to modelling investment return at the portfolio level.

Data-driven protocols and strategies powered by advanced AI algorithms processing data collected from mobile sensors and apps, electronic medical and administrative records, and other sources have the potential to reduce trial costs. They achieve this by improving data quality, increasing patient compliance & retention, and identifying treatment efficacy more efficiently and reliably than ever before.

As a result, fewer patients are needed to generate statistically significant study data, and fewer patients drop out.

AI analysis of RWD-generated by mHealth technologies has the potential to shape insights from masses of real-world data (RWD) into protocol designs. Objective, high-quality, real-time data from devices and sensors collected as patients live their normal lives has the potential to capture more meaningful clinically relevant insights among masses of data not possible using current analytic techniques. With advanced analytics, researchers can gain deeper insights into how a treatment affects symptom progression or quality of life as well as to assess and develop trial objectives, endpoints and procedures.

Moreover, expertise in machine learning can help to develop novel endpoints.

Powerful statistical approaches, including Bayesian statistics for guiding trial design based on accumulating evidence and MCP-Mod for dose-finding, can greatly increase the efficiency of trials, making smaller trials possible by achieving adequate statistical power with fewer subjects. Regulators increasingly are embracing these advanced features, as evidenced by the FDA’s designation of MCP-Mod as ‘fit for purpose’ to improve dose finding efficiency, and are being incorporated into powerful trial design software packages such as ADDPLAN.

However, applying AI and advanced statistical methods requires effort, often including extensive process modelling, and a high degree of specialised skill to achieve results useful for development and acceptable to regulators. 

Electronic Medical Records

Integrating study and electronic medical records (EMR) may increase data collection reach and efficiency, and help better integrate trials into clinical practice.

Basing trial inclusion criteria on actual patient data, automatically identified from electronic medical records (EMR), reduces the risk of extra cost and delays when unrealistic recruitment protocols need to be revised. Accessing EMR data can also cut recruitment costs, while automated site support and monitoring greatly reduce start-up and site management costs while ensuring that data are properly collected and validated. Remote data links enable data collection directly from patients at home, reducing the number of costly site visits required for a trial. EMR data allow automated post-market surveillance in Phase IV trials.

While the technical challenges of applying these new technologies and data sources to clinical trials are significant, their value already has been confirmed in many studies, saving millions in development costs. They make possible innovations that are fundamental for transforming clinical trials, such as seamlessly combining phase I and II of clinical trials, developing novel patient-centered endpoints, and collecting and analyzing real world data.

Approaches to adopt

Adaptive clinical trials 

Broader use of trials that modify study protocols in predetermined ways based on interim patient data have the potential to eliminate many unanticipated risks that undermine efficacious drugs and unnecessarily extend development timelines. 

For example, adaptive approaches often can deliver in a single two-year period combined Phase II/III trial information that otherwise might require three or more consecutive conventional trials over three or more years. These seamless trials reduce the total sample size needed by combining data from patients studied in both phases of the trial. We estimate that optimal use of adaptive trials across a portfolio, which is encouraged by regulatory agencies in Europe and the US, could reduce trial costs, by 25 percent.

Real World Data

The traditional three-stage randomized control trial (RCT) model is not designed to collect information that meets emerging market needs. In some cases, the populations involved are too small to conduct randomized trials. While RCTs are likely to remain the gold standard for validating the safety and efficacy of new compounds for initial registration, innovative trials using real-world data are likely to play an increasing role in defining new, patient-centred endpoints and expanding and refining indications.

Collecting real world data (RWD) to expand label indications, or to truly personalize therapies, cannot be done in a strictly controlled trial structure. Demonstrating value also requires collecting real-world clinical information, as well as non-clinical data on costs.

Patient centricity

Improving patients’ lives is the ultimate goal of drug development. It means everything from defining outcomes that make the most difference in patient’s lives, to offering trials to patients identified through EMRs in their physicians’ office, to minimizing control arms using advanced statistical methods and providing study results as soon as they are available.

Patients want to have more input into research and treatment of their condition and its impact on their lives. It is estimated that 1 in 20 Google searches is healthcare related[5]. This can benefit clinical research in many ways. Focused patient advocacy groups can help define new therapeutic targets, plan and recruit patients to trials, and demonstrate the value that new therapies bring to their lives.

Making clinical trials available as a treatment option when patients present for primary care, secondary care, and at the consultant level could greatly expand participation. Organizations such as PMG Research address this gap by partnering with physician networks to provide sophisticated trial support in community settings.

A holistic, integrated approach to transformation

The cost pressures on drug development are driving the search for savings. Whilst large-scale operational efficiencies are being instituted in many pharmaceutical organizations, efforts need to be integrated if they are to be effective.

There is a growing understanding that improving R&D efficiency and return on investment will take more than gradually adopting a number of new technologies. It will require a holistic approach to transforming trials, rethinking and redesigning the trial product itself and the enterprise that supports trials from the ground up. Adopting individual initiatives and tactics can improve clinical trial efficiency, but the potential is even greater when they are applied in a coordinated fashion to reimagine and reinvent the R&D enterprise.

ICON is addressing this need through its Transforming Trials initiative. This comprehensive rethinking of the entire clinical trials process uses new approaches coupled with existing, tested technologies to substantially reduce the risk and cost of clinical drug development.

Learn more at

[1]DiMasi et al. Cost of Developing a New Drug. Tufts Center for the Study of Drug Development, 2014.

[2]Jones DS et al. The Burden of Disease and the Changing Task of Medicine. 

N Engl J Med 2012; 366:2333-2338 June 21 2012, DOI: 10.1056/NEJMp1113569.

[3]Tufts Center for the Study of Drug Development – Outlook 2016.


[5]Ramaswami, P. 2015. A remedy for your health-related questions: health info in the Knowledge Grap

Related Content: