COVID-19 disrupts the standard trial transparency process
As the death toll from the novel coronavirus continues to rise across North America and globally, the pressure is on the broader scientific community to generate useful, interpretable data and analyses as efficiently and transparently as possible.
Why? It is highly unlikely that any one drug or treatment will be effective to address the high-risk conditions that can result from coronavirus infection. Instead, different combinations of drugs and treatments will be necessary. Only by sharing, pooling, and comparing data across clinical trials can the broader research community meet the urgent need to develop effective treatment options and come to understand which ones will work best for which patients.
Despite the urgency of the pandemic, the imperative to balance trial transparency with individual patient privacy has not changed. For data to be shared safely and responsibly, it must first be anonymized. The process to assess and address privacy protection in the dataset must begin at the same time as the trial, or even earlier, with its design. This will ensure data sharing can proceed as safely and quickly as possible once a trial is complete.
Swiss pharmaceutical company Novartis is taking this proactive approach by considering how trial data will be shared and in what context it may be used as part of how it designs a trial. Novartis is a key player in the global industry, with products reaching 799 million patients in 155 countries. In 2019, it manufactured 72 billion treatments.
Meaningful research data needed now
An effective vaccine for the coronavirus is only a matter of time but there is always a need for new therapies to treat conditions related to a coronavirus infection. A vaccine is unlikely to be 100 percent effective. We can look at the example of the influenza virus – even in good years, flu vaccines only reduce the risk of flu illness by between 40 and 60 percent.
One coronavirus-related condition is cytokine release syndrome which occurs during the hyper-inflammatory stage of the virus if the condition of the patient worsens without successful treatment. In patients with COVID-19 pneumonia, an overreaction of their immune systems can spark a massive release of an inflammatory protein called cytokine into the bloodstream. The result is that the body begins to attack its own cells and tissues, instead of just the virus, often with fatal results.
Researchers around the world are investigating whether existing drugs developed to treat autoimmune diseases may also be effective in suppressing this cytokine response in COVID-19 pneumonia patients. One of these drugs is Novartis’ Ilaris® (compound name canakinumab).
Novartis recently concluded a Phase III trial (known as the CAN-COVID trial), to test canakinumab’s efficacy for treating cytokine release syndrome. This trial involved 454 patients in the U.S., Europe, and Russia. Novartis started work on this trial in late March 2020, began enrolling patients by May 1st and all patients reached the Day 29 visit for the primary efficacy analysis by late September.
Novartis is making this trial data available to non-profit TransCelerate BioPharma, which fosters collaboration across the global biopharma research and development community. From here, the data will be accessible to other participating pharmaceutical companies around the world via the DataCelerate® data sharing platform.
“It’s all about helping people who are passionate about finding effective treatments for a virus that has now killed well over one million people worldwide,” said Peter Mesenbrink, Executive Director of Biostatistics at Novartis. “If we can help save lives and improve recovery times, that is the ultimate reward.”
Accelerating safe and responsible data sharing, even in a time of crisis
With time of the essence, Novartis wanted to share its insight-rich trial data safely and quickly. To ensure the anonymization process unfolded as efficiently as possible, Novartis relied on its standing relationship with Privacy Analytics.
Over the past four years, Novartis has worked with Privacy Analytics on multiple projects for clinical trial data. The drug maker has come to rely on Privacy Analytics’ expertise in statistical risk-based anonymization – an approach increasingly favored by regulators and the broader research community. For the CAN-COVID trial, Novartis employed Privacy Analytics’ anonymization-as-a-service, which is fast, reliable and delivers the highest data utility while protecting privacy. With this service, Privacy Analytics’ experts leverage proprietary software to automate much of the anonymization process.
Novartis has made every effort with the CAN-COVID trial to simplify data collection for busy frontline healthcare workers. Even so, the dataset does pose unique challenges from an anonymization standpoint.
Patients enrolled in the trial were in hospital for a relatively short period of time and likely shared their experiences with friends and family on social media. There may also have been intense media coverage of their local area during the height of the pandemic’s first wave. All these factors combine to increase the re-identification risk for any one patient versus a typical trial that unfolds over a longer period of time and with less public attention. This increased risk of identifiability is offset only in part by the fact that Novartis’ trial data will be shared under secure and controlled circumstances through TransCelerate’s DataCelerate® platform.
“Protecting the privacy of people represented in clinical trial data is our priority,” said Aziz Khanchi, CTT Analyst at Privacy Analytics. “That’s why we always apply the same level of rigor when considering and addressing patient identifiability, regardless of the context under which the data will be shared and accessed.”
As the trial commenced, Privacy Analytics was already laying the groundwork for the anonymization process that would follow. One example of this was getting a head start on privacy protection, by completing assessment forms that detail the context under which the trial data will be accessed. This provided the Privacy Analytics team with the insight to determine how privacy should be addressed, while preserving as much of the data’s utility for third-party researchers as possible.
Driving advancements in patient health boosts brand image
For Mesenbrink, the pandemic has emphasized the fact that developing effective treatments for other viral events or diseases such as cancer will demand a greater willingness to share data across the entire drug development ecosystem. Global non-profit consortia like TransCelerate demonstrate that this level of cooperation is not only necessary, but eminently feasible.
Underpinning it all is the need for trial sponsors to have in place the kind of robust and efficient anonymization program that Novartis has developed in partnership with Privacy Analytics. This ensures that useful data can be shared, safely and responsibly, to drive greater advancements in patient health generally, and specifically for COVID-19.
“We are hopeful that what will come out of the pandemic response (in addition to new treatments and vaccines) is a set of best practices for greater efficiency and transparency for sharing data in the future,” Mesenbrink said.
Enabling more efficient and transparent data-sharing across the drug development ecosystem is just good business. It can help to enhance the brand reputation of a trial sponsor and support its critical business outcomes, such as patient recruitment for future drug trials, in addition to serving a greater good.
“No one knows what the future holds, but our industry can unite to better wage a global fight against other diseases and viruses, in a manner that will benefit both trial sponsors and the broader scientific community,” Mesenbrink added. “This is the direction in which the science is heading.”