Clinical Trial Technology: New Kids on the Block are Changing Industry Dynamic

Since 2017 many clinical trial vendors have developed products aimed at improving patient recruitment for clinical trials, tracking patient responses to potential therapies, or integrating all data produced in the trial technology ecosystem. The COVID-19 pandemic hastened these changes, especially related to the technologies that support decentralized clinical trials, although some observers say that many technologies now coming into view existed pre-pandemic.

“The pandemic triggered not a digital revolution, but an attitudinal revolution in the pharmaceutical and life sciences industry. Decentralized clinical trials (DCTs) were a fashion statement two years ago, but now everyone has a strategy,” says Nimita Limaye, research vice president for tech-focused market research firm IDC Health Insight.

“There’s been huge growth in things like wearables and sensors, which are really mapping how patients are responding to interventions in a very subtle way,” observes Limaye. Thanks to these sensors, it’s now easier to tell if patients are taking prescribed medications, or trial participants are maintaining experimental treatments, she says. Limaye adds that the practice of e-consenting patients has become much more routine in the last five years, along with using machine learning to detect safety signals that might not be captured in a structured adverse event report but are hidden in the more idiosyncratic narrative records of the trial.

“Extracting intelligence from structured and unstructured data is an important use case” for machine learning, says Limaye. Another is to use machine learning to mine real-world data about how different populations fared on FDA approved drugs post-approval, to build synthetic control arms for future clinical trials in which real-world experience on the already approved drug is compared head-to-head against the new intervention.

Machine learning depends on training algorithms on large sets of data, Limaye notes, but many pharmaceutical companies securely protect their own clinical trial data in a way that’s impenetrable to machine learning algorithms. Limaye adds that federated learning technologies, in which algorithms access data that never leaves its secure location, are emerging solutions to this problem that pharma companies are beginning to embrace.

Overarching these technological developments is a broad sense among clinical trial observers that there’s a once-in-a-lifetime opportunity to infuse equity and inclusion into the design and deployment of clinical trials that use digital tools. Clinical trial participants, today, are not representative, which means that the results of clinical trials are not always generalizable.

“When we pick up these tools—whether that’s the use of real-world data, whether it’s running [artificial intelligence] and machine learning techniques on these data, whether it’s the use of telehealth for virtual visits, whether it’s the use of digital technologies and remote patient monitoring—it’s not enough to just say we’ll do it and we’ll figure out the equity bit afterwards,” says Jennifer Goldsack, CEO of the nonprofit Digital Medicine Society (DiMe), which aims to promote digital-first clinical trials that could not have been done in the analog world.

“We could be encouraged by how the current process works, when we talk about the history of the informed consent process,” adds Jude Ngang, executive director of representation in clinical research at Amgen and a DiMe partner in a new initiative to develop more inclusive clinical trials. According to Ngang, Amgen takes great pains to ensure that everyone, regardless of language or literacy level, understands what they are agreeing to as they begin a clinical trial, and says the same care can be taken with ensuring that all clinical trial participants know how to use a wearable sensor or other digital device.

Part of solving the trial recruitment puzzle is simply to alert more people about relevant trials. Amgen has invested in CancerIQ, a genetic risk assessment software program linked to patient medical records (EMRs). Many people go to community oncologists, not academic medical centers where frequent clinical trials occur, notes Ngang. With CancerIQ, a community oncologist can alert patients about enrolling studies for new cancer treatments they might not have learned about otherwise.

CancerIQ is just one of many clinical trial-related technologies. This article will detail other emerging tools that enhance clinical trial recruitment, improve participant monitoring once a study is underway, or aim to integrate clinical trial data in more impactful ways. Read on to learn about the brand new kids on the block, or the old kids who’ve learned how to pivot.

Linking patients, improving engagement, and novel measurements

“There’s this idea that doctors are afraid of losing patients to clinical trials, and that’s why they don’t refer them, but that’s not true. The main reason physicians don’t tell patients about trials is because it takes extra time,” says Mukul Mehra, a gastroenterologist and co-founder of IllumiCare. The company aims to make clinical trial referral a “prescribable event,” in the same way that a physician might prescribe a statin and then rely on existing infrastructure to get that drug to the patient.

Mehra says that trial notification systems do already exist within electronic health records (EHRs), but as email inbox messages that physicians might not read until hours or days after seeing the patient. At that point the physician needs to call up the patient record and triage whether they are eligible for the study, which is actually a better task for clinical trial coordinators.

IllumiCare’s “Smart Ribbon” technology surfaces matched trials within the EMR workflow, as the doctor is seeing the patient, and works with all major EHR platforms. A doctor can either discuss the trial with the patient on the spot, or refer them to a clinical trial coordinator to help them decide whether they’d like to enroll. Just launched in May for this purpose, after 24 months of development, Mehra expects that Smart Ribbon will make clinical trial recruitment and discovery easier.

Enrolling patients into a trial is just one challenge. Preventing attrition is another.

“How can we improve the participant experience in a meaningful way?” asks Andrea Valente, CEO of the clinical trial technology vendor ClinOne, whose prime client base are biopharma sponsors. Valente says that the average cost per patient of an oncology trial is $100,000, and up to $300,000 when involving a rare disease patient. Replacing patients who drop out mid-trial is expensive, Valente notes, and could lead to delays in availability of new treatments.

Much of ClinOne’s innovation in recent years is due to COVID-19, although the company pre-dates the pandemic. During the pandemic, e-consenting and video consults bloomed, as many ClinOne clients began to offer DCTs. Interactive SMS messages to remind patients to take medications also became more common, and a partnership between ClinOne and Uber Health offered transportation to patients who were unable to drive as a condition of enrolling in a study for a major depressive disorder treatment.

ClinOne found that, at some trial sites, 98% of elderly patients in a cardiovascular drug trial were successfully taking medications 18 months after the trial began, due in part to the medication reminders; and that offering free transportation (up to two round trips a day within a 200 mile radius, for personal and not only trial-related reasons) significantly enhanced enrollment in the depressive disorder study. ClinOne and the trial sponsor expect that retention in this study will be enhanced as well, thanks to the transportation solution.

Medical sensors and wearables, which people wear regularly as they move about their daily lives, open up new possibilities for clinical trials that the five-year-old company ObvioHealth aims to explore. Start with the humble cough.

“It sounds funny, but everyone’s cough is different,” says ObvioHealth’s CEO Ivan Jarry, who is working on a project that uses AI to detect hidden health risks within someone’s coughing patterns. This could ultimately inform new clinical trials, once clinicians and researchers have a better understanding of the health implications of different coughing patterns.

Meanwhile, ObvioHealth has developed a platform that continuously collects data gathered from wearables such as blood pressure levels, oxygenation levels, or heart rates of patients in clinical trials for different treatments—with the data fed to the research site in real time (or as soon as possible if the patient loses connectivity). ObvioHealth doesn’t just transmit the wearable data wholesale to the researcher, but also uses algorithms to detect if there are safety signals, and sends the results of this analysis, too. Its clients include pharma and biotech companies, contract research organizations (CROs), and academic medical centers.

For example, if a patient’s blood pressure always plunges right after receiving an experimental treatment, either that treatment is a total non-starter or can only work if the patient takes precautions. In the analog world, such a relationship between the medication and blood pressure might take weeks or sometimes months to detect.

“Our aim is to detect actionable signals from data,” says Jarry. ObvioHealth also wants to be a leader in women’s health studies, which Jarry notes have historically been understudied or studied with invasive procedures. This year, ObvioHealth was the research vendor for a fully virtual study that concludes that digital therapeutic devices improve the effectiveness of home pelvic floor muscle training (kegels) for reducing urinary incontinence symptoms in women.

Integrating data from every possible source

Several clinical trial technology companies focus on the so-called “back end” of digital clinical trials, or the data management and integration needed to make these studies possible.

KORE has existed since 2002, but just this year the organization acquired the companies Business Mobility Partners and SIMON IoT (internet of things) to enhance its connected health offerings. The acquisition helps KORE to serve high volume, Phase III clinical trials that can involve up to 3,000 participants, according to KORE’s Chief Marketing Officer Landon Garner.

IoT means that a physical object, like an internet-enabled blood pressure monitor, is linked to other connected objects such as a handheld device that records and communicates results (in this case, the blood pressure readings). IoT is why a programmable thermostat, say, can also turn on a HVAC system. And IoT is the reason why scientists and regulators moved from a COVID-19 vaccine concept to placing vaccines in arms within 12 months, Garner says.

“The pandemic became a forcing function for virtual or decentralized trials, using IoT technology,” says Garner, because the connectivity afforded by IoT technologies enabled real-time data capture that showed the efficacy of now-approved COVID-19 vaccines. The same flexibility is available to any company that wishes to run a global drug trial.

“We have access to 600 mobile networks around the world,” says Garner. According to the executive, KORE can provide tools such as digital weight scales or continuous glucose monitors that will work in any trial location. KORE’s technology does not analyze the data generated by a trial directly—that’s the work of algorithms and the clinical trial research team—but instead it federates the various data streams so that they all go to the right place.

The healthcare technology company Seqster created SeqsterOS to capture patient-matched EHRs, genomic DNA, wearables, and pharmacy and lab test data collected during a clinical trial, via its dashboard. Trial participants authorize with whom they share this data, in a HIPAA and FDA-compliant manner that also complies with European privacy regulations, according to Seqster CEO Ardy Arianpour. Once the trial participant authorizes sharing any of their data with vetted Seqster users, it is available.

“We’re at the tip of an exploding patient-centric and digital clinical trials industry,” says Arianpour, who co-founded Seqster in 2016 and launched an early version of its research portal in 2018. Seqster has continuously iterated its operating system since then, with the aim to “break down health data silos at scale,” in Arianpour’s words. Its most recent software integration, announced in April, is with Salesforce Health Cloud.

For health outcomes researchers interested in how economics affects participation and retention in clinical trials, Seqster includes socioeconomic determinants of health information, such as a trial participant’s zip code or annual income.

“Your zip code is one of the most important determinants of health, and most people don’t know that,” adds Arianpour. Seqster’s goal is to integrate every possible source of data relevant to someone’s health while in a clinical trial or as they are receiving care, and to enable algorithmic and human analysis of these various data streams.

Looking toward the future

Data curation and management will only become more critical to clinical trial deployments and healthcare delivery, contends Limaye, who notes “we’re a data-hungry world,” particularly when it comes to harnessing personalized data. For example, the ability, she says, to capture not only how people’s heart rate fluctuates or glucose levels vary, but also to sequence individual genomes or proteomes or lipid levels.

One challenge today, however, is that much of this data resides in silos, either in the purview of a clinical trial sponsor or EHR or a third-party provider. Limaye argues that developing holistic and longitudinal views of a patient’s health requires developing interoperable protocols that enable data flow between systems while respecting data security, privacy, and sovereignty.

At the same time as data volume is increasing, the proportion of hard-to-interpret medical data is rising as well. IDC predicts that less than 5% of all data generated electronically will be “analyzable” by 2025, meaning that many data-driven health insights will be hard to derive with current tools.

One exciting data analysis tool Limaye foresees is “codified listening,” in which machine learning and speech emotion-recognition algorithms provide new insights into the emotions or sensitivities that drive the behavior of patients or clinical trial participants. A patient might leave a voice memo that such software decodes. This technology is in its early days, Limaye cautions, but could open new doors to understanding the patient or trial participant experience.

Meanwhile, the distance between the clinical trial site and point-of-care delivery is shrinking. “There is a growing confluence between healthcare and life sciences,” says Limaye, noting that pharmacy giants such as CVS and Walgreens are becoming clinical trial sites, not just places for dispensing medications that were approved through an entirely separate clinical trial process. Limaye foresees an eventual collapse of the distance between healthcare delivery and clinical trial sites. Goldsack agrees.

“When you have to lock the data in a filing cabinet and only the [principal investigator] or the office manager has the key, yeah—you can see why that happened,” says Goldsack, in describing the historical separation between the clinical trial and healthcare delivery spaces. “But we are so far from that reality now.”

New Vendors in the Clinical Trials Industry Discuss Their Products