- The race to develop a COVID-19 vaccine drove new healthcare innovations in logistics and delivery
- AI and other technologies are enabling more effective, less expensive treatments
- Improved data-sharing will lead to smoother rollouts of future vaccines
The success of several COVID-19 vaccines relied in part on advanced technology. The Moderna and Pfizer-BioNTech vaccines proved the value of messenger RNA to trigger immune system responses, and developers pioneered the use of AI to predict the efficacy of potential treatments. But that’s just the beginning.
The larger vaccine effort—which drove innovations in FDA trials, global cold-chain distribution, vaccine delivery, and patient-tracking—is producing dividends that experts say will pay off for years to come by speeding the discovery of new treatments, lowering the cost of delivering care, and fortifying healthcare systems against the impact of future pandemics.
Other benefits include a new leadership mindset, says Dr. Nimita Limaye, research vice president of life sciences, R&D strategy, and technology at IDC, a technology consulting firm. Pharma companies and healthcare organizations that were hesitant to embrace digital transformation before the pandemic have quickly added AI, blockchain, IoT, and other technologies to their portfolios.
“COVID has been a propulsive force in transforming a traditionally tech-averse pharmaceutical industry into a mode where it is open and ready to accept technology,” Limaye says. “The vaccine rollout has served as proof of concept for the implementation of AI in healthcare.”
Larger, faster drug trials
The need to develop a vaccine in record time, during a pandemic that required hundreds of millions of people to shelter in place, has already changed how pharmaceutical companies conduct clinical drug trials. New procedures have reduced the burden on participants and made it easier for them to participate, says Limaye. The shift to more “patient-centric” drug trials means that future vaccines are likely to gain regulatory approval much faster than in the past.
Just as the pandemic brought back the doctor’s house call via Zoom, it has also accelerated a shift toward remote management of drug trials. Patients don’t have to travel to get treatments. Instead, supplies are shipped to their homes. Consent forms went paperless. Researchers collected results data using electronic patient reported outcomes (ePROs).
[Read also: How COVID-19 offers a glimpse into the future of AI]
As a result, pharma companies can recruit larger and more diverse cohort of trial participants much more quickly. The more participants, the faster trials can advance through regulatory phases. This condenses the typical time frame for clinical trials from several years to just a few months in the case of COVID-19 vaccine trials.
Pfizer’s vaccine trial involved nearly 44,000 subjects over about eight months, and nine out of 10 participated from home. When there’s an urgent medical need, drug companies “don’t have the opportunity to run a clinical trial for 12 years,” Limaye says.
The vaccine rollout has served as proof of concept for the implementation of AI in healthcare.
Machine learning tools are also helping shorten trial calendars. Typically, companies need a month or so just to clean and organize the millions of data points collected in a clinical trial. Using a machine-learning platform, Smart Data Query, Pfizer was able to get its trial data ready for analysis in just 22 hours while ensuring its quality, Limaye says.
Smarter, more secure supply chains
Another dividend of the COVID-19 crisis for global healthcare will be a smarter, more secure supply chain for pharmaceutical products, says Deepak Jayakumar, industry analyst for healthcare and life sciences at business consulting firm Frost & Sullivan.
Last April Zuellig Pharma introduced its eZVax solution, a digital platform that helps government agencies and its partners in private industry to manage and track vaccine distribution. One key component is a feature based on blockchain technology that allows caregivers to verify the authenticity of vaccine units at every step in the supply chain. eZVax is currently in use in Hong Kong, Thailand, and the Philippines.
Similarly, FedEx tracked shipments of Pfizer’s BioNTech COVID-19 vaccine using SenseAware ID, bluetooth-based devices that transmit their location every two seconds to Wi-Fi access points within the company’s shipping network. That data, in turn, fed an AI-based system that predicts delays in shipment due to weather or other adverse conditions.
Because some vaccines must be stored at extremely low temperatures, any delays in shipping can result in spoilage. By anticipating potential barriers to timely delivery, FedEx can alert its personnel to intercept and reroute packages as needed.
“Better real-time visibility of the supply chain has become a top priority for pharma companies,” says Jayakumar. “It also creates a great base for the transportation of other vaccines and therapies, especially cell and gene therapies, in the future.”
Cheaper, better treatments
A number of AI-based drug-discovery companies are finding new uses for established drugs, says Jayakumar. That, in turn, could result in faster drug approvals, lower costs, and the ability to provide care to people in less-developed regions of the world. For example, a 40-year-old drug called Fluvoxamine, used for treating obsessive-compulsive disorders, has shown great promise in mitigating COVID-19 symptoms.
If a current national trial of Luvox confirms early results, a two-week course of the drug that costs just $10 could lower the number of COVID-related hospitalizations and deaths, according to a recent report in Kaiser Health News. Repurposing older drugs like Luvox in new ways would especially benefit developing countries, which may have to wait years before COVID-19 vaccines are widely distributed.
Confirming new treatments for old drugs requires due diligence. Making claims based on small, incomplete, or non-peer-reviewed studies—such as those used to support the use of the anti-malarial drug hydroxychloroquine during the pandemic—can prove disastrous.
To prevent similar scenarios, some companies, including Model Medicines and Pharma.ai, are using machine learning to predict the outcomes of clinical trials. If the results of a trial and the AI model disagree, that can prompt regulators and other researchers to take a closer look at the data.
“Data-driven healthcare delivery will allow physicians to better understand their patients, identify risks in advance, and become more patient-centric rather than disease-centric,” Jayakumar says.
Better virus surveillance
One key to preventing future pandemics is to develop an integrated information supply chain that travels faster than a virus, says Mike Luessi, general manager of healthcare and life sciences at Workflow publisher ServiceNow. By aggregating data from multiple sources, a smart information supply chain would enable caregivers and patients alike to know what treatments are available and where people can go to receive them, avoiding the confusion and misinformation that plagued the early days of the COVID-19 vaccine rollout.
Currently that’s an enormous challenge in the United States, made more difficult by the large number of federal and state agencies, pharmaceutical firms, and healthcare providers responsible for different elements of the crisis and different pools of data in the chain.
The ability to identify and track new COVID variants across the globe is an example of [the] nascent information supply chain in action.
To illustrate what’s possible, Luessi points to another infectious agent—computer viruses. Nearly two decades ago, viruses like MyDoom and Sobig spread to millions of machines virtually overnight, causing billions of dollars in damage. Over time, though, organizations got better at identifying vulnerabilities and protecting their networks. Cybersecurity companies collected malicious code and provided an early-warning system to spot attacks before they could spread.
COVID-19 is spurring the creation of a similar infrastructure for human viruses. Online portals that share data about vaccine availability from public and private sources; advanced contact tracing; and increased funding for agencies tasked with keeping the public better informed should ensure a smoother rollout of vaccines in the future.
The ability to identify and track new COVID variants across the globe is an example of this nascent information supply chain in action, Luessi says. Workflow engines make such orchestral interactions faster, predictable and efficient.
“We’ll see more areas for sharing information to monitor what’s happening and anticipate spread,” he says. “Given what we’ve learned with these vaccines, we know that if it’s only a slight variation and we catch it quickly enough, we may get really good at patching people.”
When the next pandemic hits, in other words, it won’t take a miracle to defeat it.