7
Nurturing Innovations Through Novel Ecosystems to Accelerate Research and Development
The second panel of the workshop’s fourth session explored strategies for incubating research and development through novel ecosystems; it was moderated by Rick Bright, director, Biomedical Advanced Research and Development Authority (BARDA). Panelists discussed the key environmental features of novel ecosystems that enable innovations to tackle microbial threats. Sabrina Welsh, director of programs and operations, Human Vaccines Project, discussed the model of collaboration used by the Human Vaccines Project to broaden the understanding of the immune system in order to accelerate innovation.
Maurizio Vecchione, executive vice president of Global Good and Research, Intellectual Ventures, explored the concept of reverse innovation as a radical approach to forging public–private partnerships and developing ecosystems of innovation. Sally Allain, head, JLABS, described Johnson & Johnson Innovation’s model of nurturing innovation to accelerate research and development. Ranga Sampath, chief scientific officer, Foundation for Innovative New Diagnostics (FIND), focused on the need to spur innovation in diagnostic tools.
BIOMEDICAL ADVANCED RESEARCH AND DEVELOPMENT AUTHORITY
Rick Bright commenced the panel by remarking that in many cases, “we are still using yesterday’s technology to fight yesterday’s challenges, but we are also using yesterday’s technology to try to fight today’s challenges.” He noted that a pandemic scenario underscores the potential impact of failing
to implement innovations that have been developed. Needles and syringes have been used to deliver vaccines since the first mass vaccination campaign against smallpox in the 1800s. Although investment has supported innovation in vaccination administration that could reduce cost, improve efficiency, and allow for a more rapid and accessible response, those innovations have not been implemented. Consequently, in a scenario of pandemic influenza, it is unlikely that the United States would be able to acquire and deploy hundreds of millions of vaccine doses quickly enough to stay in front of the pandemic, said Bright. Furthermore, he noted that it would probably take about 3 years for the United States to manufacture enough needles and syringes to administer the vaccine nationwide in a pandemic scenario.
Bright also reflected on his experience working with BARDA, which was established in 2006 to adopt a novel approach to fulfilling the promise of bridging government and industry within an intentionally designed organization. BARDA’s mission is to form public–private partnerships to develop and accelerate the development of medical countermeasures to protect people from the greatest threats faced today. BARDA’s collaboration with the National Institute of Allergy and Infectious Diseases, the U.S. Food and Drug Administration (FDA), the U.S. Centers for Disease Control and Prevention, and industry partners has contributed to an increase in the number of FDA-approved products over the past several years, he said. However, approved products such as vaccines that just sit in a vial or warehouse are completely ineffective.
Large amounts of investment are channeled into developing drugs or vaccines without commensurate investment in methods to accelerate and improve drug development or in systems to support the administration of products—particularly in the last mile of care. He observed that the road in that last mile and the vehicles used to get there have also changed over the years, and they will continue to evolve. New consortiums and partnerships will be necessary to create a future-oriented ecosystem and culture that can address the last mile in the future, he said. Without a forward-looking culture and ecosystem, there may be an erosion of the progress that has been made in the past.
To help BARDA change the way it does business beyond simply creating public–private partnerships, Bright travels extensively to experience innovation and entrepreneurship within the industry firsthand. This motivated him to establish a new division of BARDA called the Division of Research, Innovation, and Ventures,1 along with a venture capital fund to stimulate investment of government funding in these areas.
___________________
1 For more information on BARDA’s Division of Research, Innovation, and Ventures (DRIVe), see https://drive.hhs.gov (accessed March 3, 2020).
DECODING THE HUMAN IMMUNE SYSTEM TO TRANSFORM HUMAN HEALTH
Sabrina Welsh’s presentation explored how decoding the human immune system has the potential to transform the future of human health. She discussed the model of collaboration used by the Human Vaccines Project, a nonprofit public–private partnership with a large network of global collaborators. The Human Vaccines Project’s mission is to broaden the understanding of the immune system and to accelerate the development of therapeutics, diagnostics, and vaccines for major global diseases.
A Paradigm Shift in Research and Development
The established research and development system is proving poorly suited to today’s problems, said Welsh. The diseases being battled are more complex and evasive than in the past, especially with the rise of resistant microbes, and the traditional research and development paradigm does not move quickly enough to address these needs. Welsh highlighted a number of late-stage vaccine and immunotherapy failures for infectious diseases (e.g., HIV, tuberculosis, malaria, dengue) and noted that many noncommunicable diseases, such as cancers and autoimmune diseases, are lacking broadly effective treatment options.
Limitations in effectiveness are another challenge, she said. Many currently licensed vaccines are not effective in the most vulnerable populations, such as infants and older adults. Vaccines often require multiple immunizations and have limited durability. Immunotherapy works only for a small subset of cancer patients, leaving a large gap in treatment options. Research and development is a hugely expensive and lengthy process, yet the probability of success is low, she added. Animal models are not always good proxies for predicting how a candidate will respond in humans. Additionally, the resources required for vaccine research and development are in the range of $1 billion over decades.
Welsh considered why some people respond better than others to vaccines and therapeutics. Vaccine response is variable across vaccines and populations. Longitudinal data on measles, rubella, diphtheria, tetanus, and vaccinia vaccines demonstrate that the magnitude of antibody responses in individuals can vary by 10–200-fold. Some people respond immediately and achieve protection with a single dose; others may achieve partial protection with many doses or never achieve protection at all. Vaccination outcomes are also variable, she stated. For example, some people exposed to Ebola become infected, develop Ebola disease right away, and become symptomatic. Others exposed to Ebola may be asymptomatic or experience less severe disease symptoms. This variability has also been observed in HIV outcomes,
in which some people’s disease progression is rapid and others—called elite controllers—are able to control their viral load and stay asymptomatic for long periods of time.
HIV is a good example of the challenges faced in vaccine development that illustrates why a new approach is necessary, she noted. The virus and its weaknesses are well understood, and how a neutralizing antibody could block HIV from replicating or infecting has already been established. Even with this knowledge, an effective vaccine that elicits broadly neutralizing antibodies has not yet been developed.
New Approach to Understanding the Human Immune System
Lack of understanding of the human immune system is impeding development of new and improved vaccines, diagnostics, and therapies for major diseases, said Welsh. The key to a new approach, Welsh remarked, lies in unlocking the inner workings of the immune system. A convergence of technological advancements has created an unprecedented opportunity to harness the power of the immune system in the fight against disease. The Human Vaccines Project is taking an interdisciplinary approach by coupling advanced tools, systems biology, computational biology, bioinformatics, artificial intelligence (AI), and machine learning to gain a more comprehensive understanding of how the human immune system functions (Human Vaccines Project, 2020).
In the past, science has largely been organized into silos focused on individual diseases or specific components of disease. The Human Vaccines Project is using a different model that involves working across sectors and diseases and sharing data with the whole field. Its approach works within a network of scientific leaders from top universities, nonprofits, government, and industry to tackle multiple diseases and to examine immunology as a system.
By examining immunology as a system, the Human Vaccines Project intends to innovate and accelerate the development of products across the board, said Welsh. No single institution has the capacity to conduct this work independently, so the Human Vaccines Project has adopted a collaborative model to drive transformational leaps instead of incremental steps. The Human Vaccines Project’s model includes transparent data sharing through a bioinformatics hub, guided by a stepwise data-sharing policy that includes specific timings for when data are to be released, uploaded, and available for access. The policy varies depending on the type of data being shared, she added.
The Human Vaccines Project has also developed a set of data standards and templates to enable data to be integrated and analyzed with the tools it has developed and made available, Welsh continued. The Human Vaccines Project has developed its data policies with input from individuals who have
experience with data hosting and sharing in order to create comprehensive data standards. The ethos is not to reinvent the wheel, but “to make sure all the wheels get on to one car so the car can move,” she said.
Overview of the Model’s Scientific Approach
The Human Vaccines Project’s main strategic initiative is the creation of an environment that facilitates the planning and implementation of iterative clinical trials to speed up and expand the scale of what those clinical trials can do, said Welsh. The Human Vaccines Project’s scientific approach focuses on key populations who are most at risk of developing disease. The project uses licensed and experimental vaccines as probes in clinical trials to address specific scientific questions, such as how the immune system develops specific responses, the durability of desired responses, how to identify the correlates of protection, and why some people respond while others do not.
The Human Vaccines Project’s systems analyses include imprinting, transcriptomics, metabolomics, genomics, epigenomics, microbiomes, immunome, cytokine assays, antibody repertoire, flow cytometry, and controlled human challenge studies. Within clinical trials, the Human Vaccines Project performs immune profiling that is among the most comprehensive and in-depth ever performed, she said. Its network has the expertise to conduct antibody repertoire sequencing and immune system imprinting. It is working with lymph node fine-needle aspirates and bone marrow biopsies to look at germinal center responses, as well as multi-omic profiling to provide a holistic view of what is happening in the immune system before and after vaccination. Once the data are collected, the Human Vaccines Project’s bioinformaticists work with specialists to integrate the data and present it in a clear and effective way.
In addition to conducting clinical trials, the Human Vaccines Project aims to leverage existing datasets and stored sample collections. She added that they have received approval from the Clinical Study Data Request Repository, which is a collection of sponsors who deposit clinical data in a repository that can be used for analyses.
Welsh explained that the Human Vaccines Project’s ultimate goal is an AI-driven model of the immune system. By coupling AI with bioinformatics, the Human Vaccines Project hopes to make sense of the exponential leaps in the scale of data being generated—it is estimated that 1 trillion terabytes (1 yottabyte) of data would provide a complete picture of human biology per individual, she said. AI and machine learning will be central for the analysis of “big data” and will transform the future of vaccination, diagnostic, and therapeutic development, she added.
Another long-term goal for the Human Vaccines Project is to be able to run a clinical trial simulation based on real immunoresponse data. They
intend to collect large amounts of immunoresponse data and run a simulation to see how a clinical trial candidate would do in the real world. This kind of simulation would enable initial testing of candidates that would allow developers to modify their candidates before clinical trials. She noted that in the early 2000s, the Human Genome Project engaged multiple sectors and countries to develop fundamental knowledge of genetics and the genetic blueprint. The Human Vaccines Project aims to do for immunology what the Human Genome Project did for genetics, by providing better understanding of the complex interactions that govern immune responses. Partnership across the board is needed in order to tackle these complex problems by collecting, analyzing, and presenting data in a way that is useful for the field, she said.
REVERSE INNOVATION
Maurizio Vecchione discussed the concept of reverse innovation as a radical approach to forging public–private partnerships and developing ecosystems of innovation to counter microbial threats. He explained that Global Good is a sister organization to the Bill & Melinda Gates Foundation, established to address global health priorities by conducting work through the Gates ecosystem of laboratories, scientists, and institutes. Given the opportunity costs and risks inherent in research and development, this work could not be conducted simply by granting funds to an institute.
He explained that because Global Good has virtually unlimited funding from a generous private donor, they are able to take risks that other entities cannot take. To leverage this advantage, Global Good’s infrastructure is designed to absorb its investment and direct it to the riskiest parts of the world. Once the risk of a project is reduced, conventional partners take over the execution. In this way, Global Good’s engagement model is to take on the greatest risks and support scale-up components, while leaving its partners with the lower risk.
Enabling Convergent Technologies to Address the Research and Development Gap
Vecchione discussed the implications of Global Good’s work from a global ecosystem perspective. According to data from the 2017 Global Burden of Disease Study,2 noncommunicable diseases account for 61 percent of deaths worldwide, with communicable diseases accounting for 28 percent and injuries accounting for 11 percent. Global datasets like these
___________________
2 Data from the Global Burden of Disease Study are available from http://www.healthdata.org/gbd (accessed February 9, 2020).
may give the impression that the worldwide burden of communicable diseases is diminishing, he noted. However, this picture is misleading in that the Global Burden of Disease Study data are presented as a blended average for the entire world. Consequently, data from two countries—China and India—dominate the Global Burden of Disease Study. For the richest billion and poorest billion people, the burdens of disease are very different than the blended average burdens.
For the richest billion, noncommunicable diseases represent the majority of the burden of disease; for the poorest billion, communicable diseases account for more than half of the burden of disease. If divided into quartiles by daily income, the two lower quartiles of the global population have a far greater burden of communicable disease than the two higher quartiles. The human population is expected to reach 11 billion in the near future, he noted, at which point the majority of the population growth will occur in the two lower quartiles by daily income, meaning that communicable diseases will be substantial global threats.
Vecchione said that cancer and other noncommunicable diseases are important research priorities. However, in terms of global investment in research and development, a large gap exists in spending to address the burden of communicable diseases that affects the majority of the global population. Data from Booz & Co. in 2011 estimated annual investment in scientific research and development to be $600 billion, with health care research and development accounting for $130 billion. Only $3 billion is spent each year on research and development for G-FINDER’s3 34 neglected diseases, including $2.1 billion on HIV, tuberculosis, and malaria (Moran et al., 2012). To address this gap in funding, the Bill & Melinda Gates Foundation and Global Good are addressing the “Great R&D Gap” by focusing on the diseases with the greatest impact on the bottom two quartiles of the global population by daily income.
Vecchione pointed out that there are multiple scientific revolutions under way that are enabling a variety of multidisciplinary and converging approaches to solving global health challenges. The methodologies for developing drugs, vaccines, and diagnostic tools are being transformed by new tools, such as AI. The development of next-generation AI is making strides toward making this true cognitive intelligence a reality that can be leveraged to address these challenges in the future. He explained that much of Global Good’s work is designed to take a systemic approach to these threats, because it is not sufficient to focus on individual diseases, such as polio, in a siloed way that is centered on surveillance. Rather, it is necessary to strengthen health systems as a whole to make transitions into
___________________
3 More information on the G-FINDER project can be found at https://www.policycuresresearch.org/g-finder (accessed March 30, 2020).
novel approaches to patient-centered treatment. Health care needs to be reinvented, because half of the 5 million children who will die in 2020 will die in the first 28 days of their lives (Goalkeepers, 2016).
Leapfrog Innovations and Transforming Data-Driven Technologies
Global Good focuses on reverse innovation because countries with nonexistent or poorly functioning health systems are prime groups for leapfrogging and innovating, said Vecchione. To illustrate, he described a leapfrog innovation in Kenya that uses mobile technology. Kenya has the highest use of mobile phones and the highest use of advanced data systems for mobile payment systems in the world. Because the country has no wire-line infrastructure or banking system, they leapfrogged and invented a banking system on mobile platforms, he noted.
Global Good’s vision is to transform data-driven technologies to allow the technologies of the future to be designed and optimized in silico using computer models before they are ever manufactured, said Vecchione. Existing technologies already allow for modeling to predict the effect of an intervention on a disease before it is implemented. However, he noted that the real power of these models is not just in visualizing the situation as it exists today; the greater value comes from looking at potential future scenarios to understand what mix of resources is needed to improve the chances of eradication. For instance, Global Good is working on predicting the effectiveness of a particular set of parameter optimizations on a diagnostic for tuberculosis. Modeling suggests that the new test will have a significant effect on the number of new tuberculosis infections. This modeling is being done before ever building the diagnostic tools, he noted. This type of technology allows for the creation of the targets for intervention that will deliver patient-centered transformations in health care.
In Global Good’s laboratory environment, teams are also creating the capacity to allow for the rapid development of new technology. For example, they have designed an AI-based robotic system that is tied to the epidemiological simulations and allows for the direct, automatic identification and optimization of new assays in record time. It takes Global Good approximately 6 months to develop a concept into a finished diagnostic product in the field with clinical trials, which collapses the normal development cycle for these types of technologies.
Vecchione presented an example of a project that used a multidisciplinary approach that leverages the traditional modality of ultrasound. By combining ultrasound with other technologies such as bioinformatics, Global Good developed the first portable automatic pneumonia assessment tool, which is being implemented into primary care in most countries in Africa. This new tool is 93 percent sensitive, 93 percent specific, and had 93
percent specificity in clinical trials, exceeding the sensitivity and specificity of X-rays with expert interpretation (Liu et al., 2013).
A similar breakthrough recently occurred in cervical cancer, he added. Because of the availability of a cervical cancer vaccine, cervical cancer is considered a solved problem—but this is only the case for those who can access the vaccine. Even if global access to the human papillomavirus (HPV) vaccine were achieved, an estimated 20 million more cases of cervical cancer will occur in the next 20 years before complete prevention of cervical cancer is realized (Gage and Castle, 2010). Global Good discovered that approximately 85 percent of cervical cancer deaths occur in low-resource settings because it has been difficult to establish the clinical laboratory infrastructure needed to conduct adequate screening via Pap smears. Global Good discovered that by using a blended approach, a simple camera phone picture of a woman’s cervix could be used to predict cervical cancer more accurately than molecular histopathology or Pap smears through a specific kind of cognitive intelligence and machine learning (Hu et al., 2019). This approach has unlocked a new standard of care for women in 111 countries, he said. The technology was recently adopted by the World Health Assembly as the standard of care for cervical cancer screening in most of the world. Vecchione emphasized that this technology is an example of what can be achieved when a multidisciplinary approach is used to tie diagnostics to treatment and develop a system-level intervention that is patient centered and not siloed.
NURTURING INNOVATION TO ACCELERATE RESEARCH AND DEVELOPMENT
Sally Allain presented on nurturing innovation to accelerate research and development, using the example of Johnson & Johnson Innovation’s model for driving external partnership, venture capital investment, and working across the entrepreneurial ecosystem. She explained that Johnson & Johnson is the largest and most diverse health care company in the world, working across three sectors: pharmaceuticals, consumer products, and medical devices and medical device innovation. The company has an internal mandate to innovate and an obligation to deliver innovative products to patients and consumers along with medicines that increase years of life, quality of life, and overall well-being.
Addressing Unique Needs of the Partnering Equation
The innovation level necessary for success has dramatically increased, said Allain. A product ought to be differentiated in order to be brought to market and surpass market expectations, so the company cannot rely on internal innovation alone. She said that Johnson & Johnson Innovation is
agnostic in the way it innovates, with no preference for internal innovation over innovation through external partnership. Johnson & Johnson Innovation has a mutually beneficial relationship with external entrepreneurs, she explained. The company benefits from the innovation, small nimble teams, and cost savings that entrepreneurs bring, while entrepreneurs benefit from the capital, infrastructure, and expertise in development and commercialization that Johnson & Johnson Innovation can provide.
Allain described how Johnson & Johnson Innovation has developed comprehensive solutions to address unique needs of the partnering equation. When Johnson & Johnson Innovation began building a model for external innovation, it decided to embed itself into existing ecosystems using its network of innovation centers, its corporate venture arm Johnson & Johnson Development Corporation (JJDC), the JLABS Life Science Incubator model, and business development support. Together, these solutions make up Johnson & Johnson Innovation’s comprehensive global solution for engagement with innovators across the consumer, health technology, medical devices, vision care, and pharmaceutical sectors. It offers entrepreneurs a “partner for every stage” by providing opportunities all along the research and development pipeline, from start-up and innovation to proof of concept, sector onboarding, and going to market. The JLABS model provides incubation and mentoring, while the innovation centers offer advice and mentorship, innovation acceleration, and research and development collaboration. Through their business development and strategic investment arms, it can provide equity investment and venture funding, strategic collaborations, licensing, acquisitions and divestments, and new company creations.
Innovation Centers
Four innovation centers are located in life science hotspots on three continents, with broad networks across regions connecting innovation ecosystems to the central innovation center hubs to create flexible collaborations, said Allain. The innovation centers focus on early-stage acceleration in the pipeline from discovery through early clinical studies; they house experts in pharmaceuticals, medical device innovation, and consumer products. She reported that the centers have facilitated more than 400 deals and deployed more than $1 billion since 2013.
Strategic Investment and Business Development Support
Strategic investors from JJDC offer extensive health care investing experience and work in partnership with the innovation centers to form new companies, said Allain. Mid- and late-stage business development deals are supported by business development teams from Janssen, Johnson & Johnson
Medical Devices and Johnson & Johnson Consumer, which provide help in licensing, mergers and acquisitions, and alliance management. Janssen Business Development works with established biotech and pharmaceutical companies at all stages of licensing and mergers and acquisitions.
JLABS Life Science Incubators
JLABS life science incubators create an enabling platform and an ecosystem that brings many groups together to drive innovations in science and technology in collaboration with entrepreneurs, said Allain. JLABS has 13 locations across the globe, including JPODS in North America, with 600 portfolio companies and more than 135 collaborations with Johnson & Johnson. JLABS engages across all sectors, including the consumer, health tech, medical device, and pharmaceutical sectors. She explained that JLABS offers the benefits of a large company’s infrastructure to small companies.
Johnson & Johnson Innovation wants entrepreneurs to use its capital to drive their science and technology, which is why it offers access to centralized infrastructure, capital, equipment, benchtop facilities, and offices at low cost. It also provides educational programming to build skills, knowledge, and networks that empower and enable local innovation communities, along with funding series support to connect capital with innovators to increase the volume and velocity of deal flow. It creates cross-sector opportunities to build an environment for solutions and not only products, she added. Recognizing that companies are at various stages of development and may need different types of support, JLABS also offers mentorship and support ranging from preparation for clinical trials to medical device development to help move companies’ products and drugs forward.
To promote open innovation and to help entrepreneurs build equity and value, JLABS uses a no-strings-attached model and does not ask for ownership of intellectual property. She noted that of the 23 JLABS portfolio companies that have gone public, 13 have been acquired, including a recent large acquisition of 1 portfolio company by Astellas for $3 billion. Around 88 percent of the JLABS portfolio companies are still in business or have been acquired, she added.
Collaboration with BARDA
Allain said that Johnson & Johnson Innovation partnered with BARDA in 2018. Through a specialized innovation zone, it provides residency for companies and entrepreneurs focused on solutions with the potential to improve the country’s response, capacity, and capabilities to address evolving 21st-century health security threats. BARDA and JLABS will leverage their mutual expertise and resources to develop programs and initiatives that cata-
lyze a new community of entrepreneurs, investors, and thought leaders committed to meeting the national medical defense needs by mounting a rapid and effective response against threats with innovative, end-to-end solutions.
One of the collaborations between BARDA and JLABS is the Invisible Shield QuickFire Challenge. For example, the current program invites action against airborne viruses, both in repelling and protecting against them, that could be easily integrated into daily life. Awardees receive grant funding, access to the JLABS ecosystem, and mentorship from expertise in the Johnson & Johnson family of companies. The first QuickFire challenge was awarded in October 2019 to Air99, which created a product that reimagines respiratory protection and air mask filters.
Allain described the progress of two companies that emerged from the JLAB portfolio. Inflammatix, Inc., is developing rapid diagnostics to distinguish between bacterial and viral infections and judge their severity, through gene expression patterns. BARDA will support the advanced development of the novel testing technology developed by Inflammatix. She also highlighted the collaboration between Janssen Pharmaceuticals and Locus Biosciences, who have signed an exclusive collaboration and license agreement for CRISPR4 products intended to treat bacterial infections. The partners will develop and commercialize a CRISPR-Cas3-enhance bacteriophage candidate that targets two bacterial pathogens.
SPURRING INNOVATION IN DIAGNOSTIC TOOLS
Ranga Sampath emphasized that without appropriate diagnostic tools, the world’s most pressing public health needs cannot be sustainably addressed because “without diagnosis, medicine is blind.” For patients, diagnostics enable correct treatment and universal health coverage; for communities, diagnostics help halt the spread of antimicrobial resistance and disease outbreaks. For governments, diagnostics accelerate disease elimination, provide data for health interventions, and reduce spending. He remarked that despite the progress made over the past decades, the global community remains unprepared to grapple with an unknown virus or pathogen. His presentation focused on challenges and opportunities in improving diagnostics.
Foundation for Innovative New Diagnostics
Sampath explained that FIND was established in 2003 as a global nonprofit driving diagnostic innovation to combat major diseases affecting the world’s poorest populations. FIND’s business model is intended to address
___________________
4 CRISPR is clustered regularly interspaced short palindromic repeats, a segment of DNA found in the genomes of prokaryotes.
areas of market failure by partnering to develop and deliver diagnostic solutions for low- and middle-income countries (LMICs). FIND is headquartered in Geneva with offices throughout Southeast Asia and Africa; it is interested in partnering with industries in Europe, Asia, North America, and elsewhere. For instance, China, India, and African nations have increasingly been moving toward models of in-country innovation, research, and design, so FIND intends to facilitate this trend by situating in-country efforts at the center of its work in advancing diagnostics.
FIND’s areas of focus include antimicrobial resistance, hepatitis C, HIV, malaria, fever, neglected tropical diseases, pandemic preparedness, and tuberculosis. He noted that many of the challenges facing LMICs are driven by practices of the Global North, such as antibiotic dumping. FIND is a member of WHO’s Strategic Advisory Group of Experts on In Vitro Diagnostics. This group offers a quality management system for in vitro diagnostics clinical trials certified by the International Organization for Standardization. Sampath said that FIND’s business model has been successful in transforming tuberculosis diagnostics in LMICs using modern molecular technologies, and it is now seeking to expand those efforts across multiple diseases. FIND works openly and transparently across the industry with multiple partners, serving as a bridge between industry and WHO. Additionally, FIND is a diagnostic collaborating center and a laboratory strengthening partner at WHO. Diagnostics often require constant advocacy, which is also a routine part of FIND’s activities.
Strategies for Delivering Effective Innovations
Sampath described the strategic pillars that FIND uses to tackle three “valleys of death” that need to be overcome in order to deliver effective new diagnostic tools to those who need them. High rates of attrition occur at each step in the evolution from concept to product development to commercialization to rollout of a diagnostic product. FIND works to address scientific, market, and policy failures and bridge the valleys of death between each of those stages in the process. To move from conceptualization and product development—the first valley—tools must be fit for the purpose and meet both countries’ and patients’ needs. FIND works in this space by catalyzing development.
The next valley—between product development and commercialization—is caused by the need for large data packages for regulatory and policy change. FIND addresses this valley by guiding use and policy. The third valley lies between commercialization and rollout, because multistakeholder engagement is required for financing, procurement, and workforce training. Sampath remarked that FIND works in this area by accelerating access. In addition to those three areas of activity, FIND works to shape the agenda through advocacy and publications to engage with funders and donors to support innovations.
To catalyze development by spurring diagnostic innovation across the value chain, FIND’s partnership model is to work with developers of any size to ensure they are developing products that are the right solution for the right setting, said Sampath. In addition to bringing donor funding to developers, it also offers technical expertise, guidance, sample banking, data sharing, matchmaking, and other tools to enable its partners’ success. FIND works across many technologies, from paper-based rapid diagnostic tests to complex molecular diagnostics.
Sampath provided examples of some of the projects that FIND is currently supporting, noting that FIND works agnostically across different technologies. Multiplex rapid diagnostic tests are being developed that can differentiate multiple causes of fever. Lab-in-a-box innovations can facilitate portable rapid assessment of pathogens with pandemic potential, while point-of-care molecular platforms can support confirmatory diagnosis at patients’ bedsides. A next-generation sequencing solution for rapid assessment of drug-resistant tuberculosis pathogens across Brazil, China, India, Russia, and South Africa is also being developed, he said.
To demonstrate how FIND’s co-investment model addresses market failure in LMICs, Sampath described some of their achievements. Between 2015 and 2018, FIND brought forward 16 new tools through its collaborative partnership model, in which companies bring in technology along with some type of in-kind or financial support. To guide use and policy, FIND has been instrumental in the development of 11 WHO recommendations, 71 clinical trials, and 32,500 patient enrollments. To accelerate access, FIND has helped train more than 6,000 health workers, strengthened more than 3,000 laboratories and sites, and provided more than 50 million FIND-supported products to recipients in 150 LMICs. FIND has shaped the agenda through the publication of 241 scientific articles.
The global health context is challenging and complex, said Sampath, so development must address country needs across the ecosystem to promote uptake. FIND partners with countries to address their downstream needs. He noted that it is important to ensure that pull from the country exists for a product, because sustainability is contingent on the country’s demand, which can be quashed by donor funding. Considerations in these efforts include
- Policy and regulatory guidance,
- Training and advocacy,
- Capacity assessment and workforce preparation,
- Supporting local research and development,
- Establishing research and developing partnerships,
- Ensuring relevant local test menus,
- Promoting sustainable business models, and
- Conducting analytical and clinical validation.
Examples of Innovations Implemented in Countries
Sampath provided several examples of innovations that have been implemented in countries with support from FIND.
Paving the Way for Rapid Uptake
FIND helped to pave the way for rapid uptake of molecular tuberculosis point-of-care diagnostics in India, said Sampath. The country had been reliant on GeneXpert, which had to be imported and added costs that were a burden to local economies. Through partnership with FIND, India transitioned from GeneXpert testing to Truenat, a battery-operated, point-of-care testing device that is cheaper than GeneXpert and feeds data directly into Nikshay, India’s tuberculosis elimination program. He added that India’s tuberculosis program tendered almost 6 million of these tests in just 1 year, which illustrates how locally built technology can facilitate uptake of innovations.
Informing Antimicrobial Resistance Policy and Practice
FIND has embarked on a large-scale project to inform antimicrobial resistance policy and practice through an antimicrobial resistance (AMR) Diagnostics Use Accelerator, an initiative designed to gather crucial evidence for an AMR policy and practice guidelines globally and locally through a study of 22,000 patients with acute febrile illness in 6 countries: Burkina Faso, Ghana, India, Myanmar, Nepal, and Uganda. Sampath explained that the program is intended to gather harmonized global and local evidence to develop a package of interventions at the primary health care level. This package will include point-of-care tests, clinical algorithms, patient flows, and training and communication tools. The program will use clinical outcomes and antibiotic prescriptions for impact assessment, he added. The unique study design with harmonized protocols enables evaluation at country and global levels. He explained that data gathered from this project are intended to inform WHO and in-country policy makers in developing new patient care algorithms. Ultimately, this project is about democratizing diagnostics and engaging closely with patients, he said.
Platform Approaches to Improve Surveillance and Preparedness
FIND uses innovative platform approaches to improve surveillance and preparedness, said Sampath. Diagnostics laboratories often have to deal with a large number of tools developed by different diagnostics developers—Sam-path likened this to each mobile phone app requiring a separate phone. FIND
is working to leverage existing technologies while also incorporating new assays by brokering partnerships between players who can develop assays on existing platforms. This creates a semi-open platform approach that allows for the incubation of ideas across companies to encourage synergizing across existing platforms and technology to reduce cost and increase efficiency, he said. The approach maximizes value for countries by offering a platform for sentinel and other testing during nonoutbreak periods; it drives economies of scale in manufacturing, supply, and regulation—thus mitigating manufacturer resistance to making assays for small markets—and it enables innovative business partnerships.
This platform helps to encourage sustainable investment in product development during peace time to reduce the reliance on panic investment in technologies during an outbreak, said Sampath. For example, more than 70 companies submitted product development plans for a near-patient Ebola diagnostic test over the course of the West Africa outbreak, but only 7 companies received WHO emergency use approval and 11 received emergency use authorization from FDA. None of these companies received approval through nonemergency FDA or WHO mechanisms. Most of these companies have since left the market, with repeat outbreaks highlighting the national and international manufacturing gaps for those Ebola diagnostic tests. Similarly, the majority of companies developing Zika products dropped out of the market as the epidemic waned alongside funding.
Co-Creation of Digital Tools to Expand Diagnostic Impact
FIND has also supported the co-creation of digital tools to expand diagnostic impact, including the use of network optimization and other models to deploy technological solutions. It takes advantage of connectivity and other means to measure how these technologies are being used. For instance, FIND has co-created digital tools to reduce the delay in tuberculosis diagnosis, said Sampath. These efforts helped empower health care staff to improve tuberculosis care in Myanmar and ensured that more patients entered into the treatment pathway. Digital innovation can also address issues of poor data availability and the limited use of data.
Toward Sustainable Ecosystem Evolution
Sampath closed with a discussion of sustainability and opportunities for ecosystem evolution. He described a scenario where tuberculosis and HIV care are delivered in separate settings; in such a scenario, an individual with HIV and tuberculosis has to go to two locations to receive care. Synergy needs to be developed to eliminate this inefficiency, he said. Although many simple solutions have been developed, implementing these solutions is a
complex task, in part because of stakeholders who are entrenched in existing systems. Silos must be broken down and funding mechanisms should be redesigned to enable multiple diseases to be addressed with little or no additional resources, he said.
True partnerships should be country driven and use demand-driven technologies and broaden the donor base, Sampath continued. Because so much of this work has traditionally been donor driven, it calls into question whether innovative funding models will be able to drive technology uptake sustainably. However, he suggested that potential funding mechanisms can make use of structured, innovative financing, including priority review vouchers, advance market commitments, and volume guarantees. He added that global partnerships and innovation are the waves of the future for pushing this agenda forward to ensure that the diagnostic needs of all countries are met, not just LMICs.
DISCUSSION
Matt Zahn asked about opportunities to spread the burden from public laboratories to private laboratories with respect to diagnostic testing during an outbreak response. Sampath remarked that the challenge of shifting testing into private laboratories lies in obtaining the requisite accreditations and approvals. However, he suggested that this could be addressed through forging public–private partnerships and collaborating prior to an outbreak.
Eva Harris asked how Global Good’s cervical cancer innovation would figure into countries’ broader protocols for cervical cancer care, which are complicated by challenges related to vaccination timing, linkages to care after diagnosis, and conducting follow-up. Vecchione explained that work is under way by international agencies and funders to integrate this novel diagnostic with high predictive value into the continuum of cervical cancer prevention and treatment, which is particularly challenging in low-resource settings. However, if the disease can be detected before it becomes cancer, there are well-established treatment protocols and related technologies that have been developed and would allow for diagnosis and treatment during the same point-of-care session.
The aim of this new technology is to transition cervical cancer treatment from a surgical procedure conducted by a specialist toward a prevention-like treatment. He added that this new technology is not intended to be an alternative to population-level vaccination; rather, it is being developed as a solution to contribute to cervical cancer elimination before population-level immunity is reached. Given the lack of national-level efforts to expand HPV vaccination, Global Good is working to link diagnosis to treatment and establish a new protocol that collapses access to services to the base of the pyramid of health care.
Rajeev Venkayya asked what Global Good has learned about engaging communities around user-centered design, uptake, and adoption. Vecchione replied that Global Good began working on health technology by providing huge budgets to laboratories that developed products through an “accidental pipeline”—that is, a pipeline of solutions in search of a problem. In public health, there have been numerous cases where a new technological solution has been deployed that cannot be scaled up or does not address the realities of a problem on the ground. One reason for this is that these technologies tend to be invented in laboratories without a proper understanding of the problem at hand or of the context in which the new technology will be deployed. However, most public health problems are not technology problems; they are systemic and multifaceted problems. He explained that Global Good adjusted by adopting the practice of understanding the problem first, before developing solutions.
Global Good has also made large investments in bioinformatics and data, which allow for the prediction of pathogen outbreaks at a district level anywhere in the world. They can also account for the effects of climate change on pathogenicity of some targets owing to genetic pressures of those pathogens. By combining data and evidence with human-centered design components, Global Good works to curate a list of problems and determine which problems require a technical solution that the organization can support.
Keiji Fukuda remarked that innovation is evolving at a rate that countries and communities cannot maintain without support from stakeholders; bringing in AI and machine learning will likely lead to even more rapid changes. Vecchione replied that the Bill & Melinda Gates Foundation tries to address gaps by reimagining problems through a lens of new technologies that may not be immediately endorsed by experts in the field. For example, when Global Good began working to address cervical cancer, leaders in the field said that machine learning could not work. He said that innovators must not let past failures inhibit them when striving to develop and iterate disruptive technologies. He added that governments, corporations, and donor-funded organizations have different levels of appropriate risk.
From a business perspective, innovators assume great risk because the odds of failure are high, and a certain degree of failure is to be expected. In business and government ecosystems, there is an aversion to this risk, which is why partnership with large-scale donors can foster innovation. It is donors’ tolerance for risk that enables many innovations, including the cervical cancer innovation, which would not have been possible without donor-funded risk reduction of the technology. Vecchione encouraged innovators to keep this stratification of risk in mind as they pursue innovation; even within the Gates ecosystem, there are different tolerances for risk among different programs.
Sampath remarked that many of the technologies that are driving innovation were not developed based on a blueprint for how technology should solve health care problems; instead, they were successfully developed for other purposes and are now serving as a platform for innovative health care technology—such as mobile phones and communication technology. These technologies advance in leaps and bounds and thus will sometimes fail, but if the right questions are asked, then the right applications will usually be brought to bear. Often the challenge lies in asking the right questions and developing systemic answers to these questions, he added.
Allain commented that there are examples of simple innovations that can be developed relatively inexpensively by small companies with modest seed funding. For example, before Air99 innovated respiratory masks, the technology had not been changed for 50 years. Using a seed grant from Johnson & Johnson Innovation, Air99 developed a respiratory mask that would fit any type of face from infants to adults, which was an effective innovation that addressed a real need. Another company in the Johnson & Johnson Innovation portfolio is Certa Dose. It developed a color-coding system for syringes to reduce dosing errors, which has the potential to substantially improve the delivery of care, particularly in low-resource settings.
Welsh commented that investment firms can also mitigate risk. When the Human Vaccines Project began, it had financial partners who shared the risk of the project. The Human Vaccines Project’s promise to these partners was that it would work in a different, more efficient way that could accommodate the needs of the research network, for example, by conducting flexible iterative clinical trials that begin with small groups then scale up depending on the needs of the product. These strategies allowed the Human Vaccines Project to change its risk profile for investors. The Human Vaccines Project provides smaller grants and other kinds of benefits—such as access to large datasets, immunology tools, and bioinformatics—that can support its partners in developing vaccines in a different way, which can offset the risk of investment.
Nitika Pant Pai asked about barriers and challenges that the Human Vaccines Project has faced in proposing a new method of immunology research. Welsh explained that it is challenging to sell the idea of the Human Vaccines Project, because donors perceive the project as similar to a think tank or innovation engine. Because the Human Vaccines Project is not a laboratory and is not developing a disease-specific vaccine, it is difficult for donors to understand the project. However, they are working to scale up and move from the pilot phase toward larger consortium-style projects. She suggested that once the Human Vaccines Project’s infrastructure, data systems, and processes are in place, it will be easier to see the power of applying those data beyond research on infectious diseases and cancers.
Turkan Gardinier expressed concern about calling the outputs of research “intelligence.” Systematic analysis and operations research have
been widely used to find linkages between occurrences and the spread of diseases, but those analyses do not constitute intelligence. She cautioned against the use of language that suggests that these processes are developing intelligent systems that surpass human efforts. Vecchione replied that AI could more accurately be called “statistical intelligence,” because machine learning reveals statistical correlations. He suggested that the exuberance about machine learning and the sentiment that it is the solution to all problems may be running its course.
Machine learning techniques work well in certain applications. However, they need to be blended with statistical stochastic models in many predictive models, especially when large populations and uncertainty across numerous factors are involved. He calls this blend “cognitive intelligence”—meaning, allowing a computer to become a thinking machine—and predicted that it will be the successor to AI. It is currently available in the average machine learning toolkit, he continued, but mechanistically driven forms of intelligence are being developed on the cutting edge of drug discovery. Although this technology is not yet practical to use, he suggested that it will revolutionize immunology and genome projects by extracting meaning from the correlations and distinguishing between causation and correlations.
George Haringhuizen asked about the potential for a global, well-curated sequence database of viruses and bacteria that could be used for day-to-day diagnostics. Sampath said that sequencing has great potential, but it is not a technological advancement that will solve the problem of diagnostics overall. Sequencing may be a piece of the solution in terms of content and processing ability, but technology cannot be a sole solution.
Vecchione commented that more technology is not necessarily better in terms of clinical care. He noted that there are more magnetic resonance imaging machines on the west side of Los Angeles than in all of western Europe combined; this does not mean that people in Los Angeles are healthier than people in western Europe. Sequencing is a tremendous revolutionary technology that will advance many areas of diagnostics, he said, but next-generation sequencing should be used to strengthen the patient-centered continuum of care. He also mentioned there are certain diagnostic innovations that cannot be relied on as a sole diagnostic tool because they do not address host–response mechanisms, comorbidity, or other factors that are addressed in the continuum of care. He suggested improving decision-making support along the clinical continuum of care so that the best diagnostic tools are being integrated into a system that is designed to facilitate the best treatment decisions.
Jyoti Joshi commented that technological innovation occurs so rapidly that it is difficult for clinicians, academics, or students to keep up with the evolving paradigm. She asked how all stakeholders can be better supported
to avoid the fear of new technologies that was seen during the Ebola and Zika outbreaks, when vaccines were available, yet people were hesitant to get vaccinated.
Allain replied that when JLABS moves into an ecosystem where innovation is ongoing, it offers programming that addresses science, technology, venture capital, and business development for anyone who may wish to become involved. This is intended to foster open dialogue across stakeholders and reduce silos among them. The new JLABS site in Washington, DC, is located at the Children’s National Hospital so innovators can work closely with the stakeholders whose problems they are innovating to address.
Vecchione remarked that in the near future, the bulk of the market for innovation will be outside of the United States, as countries like China and India accrue increasing spending power and capacity. He suggested that these emerging markets need to be empowered to enter into the ecosystem and innovate for themselves. This can be supported by building local capacity to incubate innovation and reverse innovation, he added. As an example, he described the “Silicon Savannah,” an ecosystem of venture capital and entrepreneurs outside of Nairobi, Kenya, that is centered around mobile financial services. He suggested that existing, suboptimal practices that have been in use in Western countries are poised to be leapfrogged.
Welsh commented that there are different ways to disseminate information to catalyze young investigators to engage with these problems. For instance, the Human Vaccines Project oversees the Michelson Prize for young investigators who are interested in immunology. Such efforts not only spread the Human Vaccines Project’s message and publicize its work, but they also encourage the next generation of scientists to innovate.
Marcos Espinal remarked that the most successful innovations seem to be shaped within partnerships that are focused on empowering and engaging with communities where the problems lie. For example, the Amazon Basin in Brazil has large burdens of malaria and dengue, but new technological innovations would not be appropriate for indigenous communities. In contrast, it would be beneficial for the country if the vaccines currently under development were manufactured in Brazil for its own people.
Vecchione pointed out that Global Good’s work in cervical cancer encountered cultural issues that were not addressed by their technological innovations. For example, communities tended to mistrust government representatives, which undermined any implementation that relied on a government mandate to screen every woman in each community. To address this barrier, Global Good engaged with community health workers—typically older women—who evangelized for cervical cancer screening within their communities on behalf of the implementers of this program. This approach has not been scaled up, but it demonstrates the importance of stakeholder engagement, he noted.
This page intentionally left blank.
