Today, neurological and psychiatric disorders are diagnosed, prognosed, and monitored with subjective clinical exams and scales, imaging, and sometimes liquid biomarkers. These allow physicians to capture a snapshot of a patient’s health during a scheduled office visit or acute event, but they don’t capture disease activity between visits. Also, bias and differences in how clinicians conduct and score the tests affect subjective exams and scales.
Digital and hybrid biomarkers collect objective physiological and behavioral data via mobile and remote methods and devices, including wearables, sensors, and smartphones. Collecting data about how a patient walks, talks, sees, and feels in daily life—by measuring their gait, balance, activity, sleep, speech patterns, vision, smell, and mood—could allow neuroscience sponsors to augment (or replace) subjective rating scales or intermittent measures with more objective and continuous data. Digital and hybrid biomarkers have the potential to capture changes in disease progression earlier and more precisely using passive and active monitoring, remotely or at home. This is less burdensome for patients, lowers the cost of trials due to fewer site visits, allows smaller trials of shorter duration, and empowers sponsors to make faster adaptive trial decisions. Well-designed and executed studies that utilize digital biomarkers and measures thus create an opportunity for significantly positive ROI.
Digital biomarkers (DBx) could benefit a wide range of neurological and psychiatric disorders. That said, their application in neurodegenerative disorders, such as Parkinson’s or Alzheimer’s disease, is especially exciting since they may be able to detect early signs of disease that are not possible (or not practically possible) with today’s methods. Early signs and symptoms of these diseases are typically not recognized, and the disease is only diagnosed once the patient has suffered significant and irreversible declines. DBx could also measure symptoms of mood disorders over more extended periods, which would be beneficial because mood disorder symptoms vary considerably from day to day.
Most digital biomarkers are nascent
To identify and accelerate neuroscience treatments that deliver value to patients, DBx and measures must advance “from ambition to impact.”1 While only a few of these solutions have been clinically and analytically validated, many more are in process. Sponsors are understandably cautious about collecting novel digital data; they must manage regulatory risks, development costs, and product timelines while achieving a reasonable return on investment (ROI).
That requires a strategic plan for developing, evaluating, and integrating DBx into clinical trials, considering post-approval commercial opportunities, and anticipating how the DBx validation process may evolve. At Parexel and Health Advances, we have advised large pharmaceutical and emerging biopharma companies on realizing DBx's potential while mitigating the risks and controlling costs.
Here are three strategies that work:
Leverage digital biomarkers and measures where they add the most value
Many sponsors assume that their investment in DBx and measures will accelerate clinical development and that the solutions used in clinical trials can be deployed commercially to accelerate adoption. That is not always the case.
Some will add the most value during development—to make clinical trials more efficient and inform faster go-no-go decisions—but it will not yet be practical to implement them in the real world due to current clinical practice, access to technology, and other confounding factors. To understand whether a DBx could add value in a commercial context, sponsors must examine a technology’s current level of validation, the quality of evidence generated, product-market fit, the likelihood of real-world adoption and reimbursement, business model feasibility, and go-to-market strategy.
Recently, a pharma sponsor and their digital therapeutics partner used an AI/ML approach to develop an algorithm that objectively assessed pain in a chronic inflammatory disorder with wearable sensor data. The partners conducted an observational pilot study to collect training and test data to validate the tool and patient feedback on the device's usability. This yielded a novel measure that generates more valuable data than periodic site visits. It provided continuous feedback on participants' perceptions of pain, which could be incorporated as an exploratory endpoint in clinical trials and inform decision-making during clinical development.
The partners also wondered if the technology could help providers and patients monitor disease and improve care, particularly by more rapidly identifying patients for whom the standard of care was inadequate. However, they encountered several challenges: First, payers require significant evidence to reimburse DBx, meaning the partners needed to conduct additional extensive studies. Second, providers don’t necessarily embrace novel technologies that could introduce new workflows or processes and where reimbursement is uncertain or inadequate; indeed, they might actively stifle such innovations. Finally, marketing DBx means competing for clinicians’ and payers’ limited attention and engagement.
The novel digital data collection tool delivered a significant ROI in clinical development. However, commercial applications were more challenging and required defining patients', providers’, and payers’ unmet needs, motivations, and financial constraints.
Select technology partners that fit your development strategy
The market landscape for digital biomarkers and measures is evolving rapidly, so companies seeking to invest or find a partner have many choices of innovative technologies and business models. In making their choices, sponsors should consider the level of innovation, validation, scalability, technical capability, hardware availability and suitability, and user-friendliness of digital technology. Emerging companies with limited budgets should not assume DBx will replace traditional subjective scales and assessments, most of which will continue to be required for at least the next five years.
A large biopharmaceutical company recently asked us to evaluate the landscape of novel DBx solutions for several neuroscience indications. The company wanted to know how to accelerate their use, especially in R&D, and how specific technologies and partners might advance its business interests. The sponsor did not want to acquire and independently develop DBx but to facilitate their development where these technologies aligned with their business interests.
We began by assessing the landscape of DBx startups and identifying those that might streamline trials and expand our understanding of disease progression. We evaluated the maturity of each DBx and advised our client to avoid proof-of-concept technologies and focus on those with less risk and more significant near-term business potential. We then stratified solutions based on their potential applications, such as screening and diagnosis, disease activity, and prognosis. We also provided a list of solutions and partners that could help our client accelerate R&D for the various assets in their neuroscience portfolio. We discovered that a variety of potential partners are working on a diverse array of sensors and measures with varying levels of validation. Validation is critical for near-term application in clinical trials. Thus, partners with better-established sensors and measures (such as gait) had a clear advantage. Companies exploring newer measures—especially those with multi-modal measures—tended to have less validation, limiting their applications for sponsors in the near term.
Incorporating digital biomarkers or measures into a clinical development program may initially increase costs since these measures are typically unable to replace traditional measures today. However, when measured against the information gained by sponsors, it can be a worthwhile investment. As these endpoints move from exploratory to secondary to eventually primary, they produce a positive ROI on trial costs.
Also, it is critical to consider the additional impact on patients. Passively collected DBx and measures create little to no burden for patients, while some actively collected measures may require patients to be more methodical and consistent in their data collection. For example, it’s one thing to ask patients to repeat three sentences into their smartphone once a day to track their speech patterns during a 12-week trial, but it’s another to ask them to wear a head cap that measures brainwaves to bed each night. Sponsors must weigh the burdens, costs, and benefits of collecting data digitally against those of traditional methods and biomarkers.
Understand commercial deployment realities
Successful commercial application of DBx and measures is challenging and requires meaningful evidence for payer coverage and reimbursement, physician uptake, and patient engagement. Deploying digital biomarkers to generate evidence is more complex than adding smartwatches to a clinical trial. It requires operational expertise, specialized patient and site training, sustained tech support and troubleshooting, and sophisticated data analytics.
Companies that identify a DBx and measures capable of detecting disease progression, medication side effects, or adverse medication events earlier or more accurately can brand that tool and own the possibility of delivering effective treatment to more patients. At Parexel and Health Advances, we provide strategic guidance and recommend approaches to evidence generation, paths to revenue generation (including traditional market access), and the selection of realistic business models.
We recently advised a sponsor on a strategy for a digital biomarker that identified a common sequela of a disease that would prevent patients from continuing treatment. We developed strategies to support payer and physician adoption by identifying the incentives and value created for all stakeholders, including patients. Another client asked for advice on an AI algorithm that interpreted an existing clinical measure to help identify patients with advanced cardiovascular disease at an earlier point. We worked on coding, coverage, and contracting strategies to support commercialization within health systems.
Digital biomarkers could confer a competitive edge
According to the Digital Medicine Society (DiME), there are currently more than 500 digital sensor-derived endpoints, many of them primary, being tested in clinical studies.2 One report estimates that, from 2010 to 2020, there was a 39% compound annual growth rate in the use of digital health technologies in neurological clinical trials. Parkinson’s, Alzheimer’s, multiple sclerosis, and epilepsy trials had the highest use rates.3
For sponsors confronting risky and expensive neuroscience clinical development pathways, DBx could boost development efficiency and commercial viability. Although incorporating novel DBx in clinical development poses some risks, companies that invest early in these fast-evolving technologies could gain a competitive advantage over those that don’t.
Contributing Expert
