# Virtual Clinical Trials: Testing new drugs from afar

by Isabella Grabski
figures by Jovana Andrejevic

Clinical trials are critical to study the safety and effectiveness of new drugs, but they are no small endeavor. The cost of a clinical trial can range anywhere from $7 million to over$50 million, and they can take up to a decade to fully complete. Not only can this represent a substantial strain on a pharmaceutical company, it also contributes to the long wait times between the discovery of a drug and its deployment in the market. Every extra year of a trial means an extra year that patients are without access to a drug that might save or dramatically improve their quality of life.

In the last decade, a potential new solution has emerged to improve this problem: virtual clinical trials. Although there are several variants of this idea, what virtual clinical trials all have in common is that much of the study is conducted remotely, rather than at a physical research center. These studies rely heavily on new technology, using anything from an app on a smartphone to wearable gadgets collecting important health measurements.

Virtual clinical trials can’t eliminate all of the costs and time involved in the trials, but they can certainly put a dent in these figures. As pharmaceutical companies grow increasingly interested in this idea, regulatory bodies like the FDA have to figure out the implications. Conducting trials remotely and with new technology raises questions about their effectiveness, as well as their ability to keep data private – a hot-button issue in the medical field. Could virtual clinical trials replace traditional trials, or are the risks too great to pursue?

## Traditional vs. Virtual Clinical Trials

The specific form of a clinical trial can vary quite a bit, and the number and type of participants enrolled will depend on which phase the trial is currently in. If the drug seems sufficiently safe in laboratory studies, trials generally proceed through stages known as phase I, phase II, and phase III. At any point, if the drug appears to be too dangerous, the trial can be terminated without completing all of the phases.

Phase I is the smallest, and involves between 20 and 100 healthy volunteers to slowly increase dosage and see how well the drug is tolerated. In phase II, several hundred participants are recruited, now with the condition that the drug is aimed at treating. The goal is to determine how well the drug works, as well as to identify side effects and any further safety concerns. Finally, in phase III, the participant pool increases even more, and the trial takes place over a much longer time period. Participants are usually assigned at random to one of two groups, where one group takes the drug being tested and the other takes a placebo. The pharmaceutical company can then learn how effective the drug is, compared to a placebo.

Because of the longer time scales and larger number of participants, phase II and phase III trials are generally the most expensive. One of the largest expenses is the central clinic, often located in a large research institution or a hospital, to which the participants must report on a regular basis. This is done to record health measurements, which can be anything from blood pressure to brain scans, both to understand the effectiveness of the new drug and to monitor the participants’ overall health and safety.

On top of the expense of maintaining and staffing this clinic, recruiting participants can be challenging because many people are unwilling or unable to commit to making so many visits, which can prolong the recruitment stage (Figure 1). Moreover, frequent travel is among the top reasons cited by patients for dropping out of a trial. If too many participants exit the trial, it can extend the trial time as researchers recruit more people, since too few participants can hurt the reliability of the results.

This dependence on travel leads to another, perhaps even more serious problem: it reduces the diversity of trial participants. If the only participants are those who can make frequent trips to a clinic, then socioeconomic status, familial responsibilities, and occupation can all affect someone’s ability to join. Furthermore, large hospitals and research institutions are often located in urban areas, so people in rural regions might be excluded from the get-go. This can have a serious impact on the findings of the trial, since the safety and effectiveness of a drug can vary from one subset of the population to another. The FDA, which oversees and regulates clinical trials, has even recently made clear that improving the diversity of trial participants is a priority.

Virtual clinical trials offer one solution to this travel problem by allowing the trial process to be completed at home. Every step, from recruiting to screening to data collection, is performed remotely, thanks to the internet. Staff may still have to make occasional home visits for certain types of health measurements, but as wearable devices become increasingly sophisticated, there is less and less reliance on in-person data collection (Figure 2). Smartphone apps can track daily movements, watches can monitor heart patterns, and biosensor patches can record respiratory rate, to name a few examples. This means more frequent measurements, more convenience for patients, and the ability to collect some types of data that aren’t even possible in the clinic, such as the number of steps taken each day, or glucose levels after every meal for diabetics. Not only does this reduce travel burden for the participants, but it also means reduced cost for trials that no longer need to support frequent visits to the clinic.

## The Downsides of Virtual Clinical Trials

However, virtual clinical trials are not appropriate for every kind of setting. If a study requires a much more complicated type of health measurement, such as a brain scan, that certainly requires a trip to a clinic. In addition, even though virtual clinical trials are considered a means to expand accessibility for participants, they aren’t always entirely inclusive. In many cases, for instance, participation requires a smartphone with some sort of data plan or WiFi access, which can limit lower income populations’ access. Thus, virtual clinical trials do not represent a perfect solution to the lack of diversity in trial participants, but the hope is that they will still make trial participation more accessible than traditional trials to a larger, more diverse subset of the patient population.

Another significant concern is data privacy. Continuous data collection through smartphones or wearables means there is a constant stream of highly sensitive data being communicated to researchers through an app. Any kind of data breach in this setting could represent a tremendous privacy and security risk.

The FDA is currently piloting a new risk-based approach to regulating these kinds of third-party health data apps. Under this approach, some organizations, including Apple and FitBit, are pre-certified, which means they can move to market more quickly through an expedited approval process. Other organizations’ products will require greater scrutiny if they are deemed higher risk. The FDA evaluates cybersecurity, among other attributes, as it examines these software products, and the agency has released extensive guidelines on ensuring that sufficient measures are taken to protect participants.

However, these guidelines are only recommendations to industry, and they’re not perfect. In the last twelve months, the FDA has released safety communications warning of cybersecurity vulnerabilities in a data communication network, an insulin pump, and an implantable cardiac device. In the case of the insulin pump, the FDA warned that an unauthorized person could use these vulnerabilities to connect to a pump and change the settings, which could have potentially life-threatening consequences to patients.

It should be emphasized that this cybersecurity problem is not isolated to the realm of virtual clinical trials. These kinds of software are part of many patients’ treatment plans, even beyond trials, so their issues cannot purely be taken as an indictment of virtual clinical trials. Nevertheless, this is still an important problem that future policy will need to be developed around. One potential solution is using blockchain technology, which is already used in electronic currency to ensure privacy and security.

It is clear that virtual trials offer a compelling improvement over their traditional counterparts, especially for those involving geographically spread out patient populations. However, more research is needed to see if there are further ways to reduce the cost and time of trials, and participant diversity and data security are both ongoing issues that virtual trials will still need to grapple with.

Isabella Grabski is a second-year PhD student in Biostatistics at Harvard University.

Jovana Andrejevic is an Applied Physics Ph.D. student in the School of Engineering and Applied Sciences at Harvard University.