04 Sep 2018
Written by Shae Harmon, Digital Editor, Longitude Prize
We interviewed Dr Cassandra Kelly-Cirino, Head of AMR and Outbreaks at the Foundation for Innovative New Diagnostics (FIND) – an international non-profit organisation that accelerates the development, evaluation and delivery of high-quality, affordable diagnostic tests for poverty-related diseases – about the challenges faced in this domain.
Our focus is specifically on diagnostic tests that are well-suited for low-resource settings. FIND’s portfolio of projects spans six disease programmes, and recently incorporated antibiotic resistance, a key priority for the global health community. Within these individual programmes, we work with country-level experts, the people actually implementing tests, to understand the gaps and challenges they face, and then we talk to donors about the health service needs in these countries. Once the funding is secured, we provide industry partners with seed funding and work with them to develop (or sometimes just improve) their diagnostic technology and platforms in order to be suitable in low-resource settings.
Many industry partners don’t have good foundational knowledge of the challenges in lower-resource settings, as they are used to designing technologies for high-income countries where the higher price tags, ready markets and return on investments are clear-cut and measurable.
Some industry partners have the impression they can easily adapt an existing technology and simply work on designing a new or modified test for a low-resource setting, but if only it were that simple. If you start from a high resource setting perspective, it is not likely such diagnostics and test systems will work globally. There are a lot of variables which need to be considered, such as their cost, ease of use, robustness, etc. If you take these and other factors into account, you can make a product that is universally applicable.
Our work at FIND includes getting teams to design a product that is fit-for-purpose and then do early stage research and feasibility work. We also design clinical trials, collect data, and apply for the clinical and regulatory qualifications for any of the three regulatory bodies that control uptake of new diagnostics in various countries – the FDA, IBDR, WHO pre-qualification, or CE marking.
Tuberculosis, HIV and malaria are housed under the Global Fund, but with AMR there is no clear pathway for the introduction or procurement of a diagnostic. So at FIND, we also have an access programme, which specifically is meant to close that gap. If you develop a new diagnostic test, there is no guarantee it will be successfully implemented and scaled up. There are many distribution and procurement challenges, and the markets in low-resource settings tend to be smaller, with lower price points. That’s why our Access programme also plays an important role in the early stage R&D portfolios.
Firstly, the design requirements, as researchers really need to understand the market in which they will be placing this device. If that is not taken into consideration from the beginning, scaling up and sustaining access to AMR diagnosis will be onerous.
As I mentioned earlier, making things globally applicable is a major challenge. Resistance within pathogens varies with geography, so making a highly specific and accurate diagnostic test that is globally relevant – will be difficult. What may work in the UK will likely be very different to what works in south east Asia, for example. The pathogens that circulate are different, as are their resistance profiles. So, you can make a general technology that is applicable, but you need different tests if your technology is to diagnose a specific pathogen.
Another complexity is that when we talk about AMR, we talk about a whole range of different syndromes, by a multitude of different pathogens, and they are delivered to patients and treated with different tools across health care systems. You need diagnostics that are applicable to sexually transmitted infections, respiratory infections, urinary tract infections, or sepsis, to take just a few examples, as these are all diagnosed with different tests and treated at different places in high resource settings – so having one test for a multitude of pathogens is almost impossible, especially one that is accurate, cost-effective, easy to use, etc.
If you are developing a technology that is broad and flexible that is one thing, but if you are developing something that is a single assay, that is very specific.
We have programmes at FIND looking at biomarkers so that we can do more triaging to clearly identify the infectious agent. For AMR, you must either be very specific or very high level – in other words, is it a bacteria or is it a virus, should we give antibiotics or should we not? These types of tests have a lot of potential and applicability in low-resource settings if they can meet all the criteria the Longitude Prize has set out. But the concern we have is that the biomarkers we are evaluating are from cohorts of patients that are not representative of low-resource settings.
If you have a biomarker that you think might be indicative of differentiating between bacterial and viral causes for fever, and you go and test it in the USA or the UK, the cut-off where it would be relevant is going to be very, very different compared with sub-Saharan Africa, where we have malnutrition, high rates of malaria, dengue, yellow fever and other co-infections, and where the health system and general state of health differs greatly. Because of these differences, the variability could be quite great and so the companies that are developing these targets for diagnostic tests must do thorough evaluations of applicability, with cut-offs that correspond to these populations, and in the correct settings specifically.
One of the challenges is that they these teams don’t often see the downstream, or end-user, perspective. They will see things from a technological perspective and while they are very bright, driven and dedicated to their technologies and designs, there are a lot of downstream challenges that need to be addressed, from manufacturing capacity to forecasting volumes, to understanding what price points mean in your business model, and how to distribute the products. Many of these smaller teams don’t have distribution channels, and going through distributors adds a markup price to their diagnostic tests. This has a massive impact on your actual price and it’s where most early stage and smaller industry partners falter.
We hear a lot of people say “We can make a diagnostic test for a dollar,” but have they considered the complete manufacturing process, and how the regulatory processes work – as these can be very expensive and time consuming. In other words, what is the regulatory path for that diagnostic test going to be? What are the distribution channels? What are the pay models? Who is buying the test and how will you get uptake, scale and return on investment?
Since AMR remains, for the time being, outside global funding mechanisms, countries have to take it upon themselves to allocate dollars from their health budgets to purchase these tests. Alternatively, the cost will fall to the patient, and since antibiotics are quite easy to come by and very inexpensive in a lot of low-resource settings, the notion that a patient should incur extra costs to pay for diagnostic testing when they can just go and buy antibiotics, would require the more complicated path of behavioural change.
We have an open competitive process, similar to the Longitude prize, for selecting who we work with, where we put out a call for proposals describing specific requirements that we want to address. We also work through what we call “target product profiles”, so we develop a TPP which is really a product requirement document or a preferred characteristics document of the technology we are looking for. The TPP will describe the target population, whether we are looking for a pathogen-specific test, what the sensitivity and specificity should be both from an optimal and minimal perspective, the cost, usability and the setting in which we expect to implement the tool. We use TPPs to let industry partners know we have available development funding, and then we vet proposals in an internal review process, followed by an external review as well, and finally it goes to our scientific advisory board for a final decision.
When we make the awards, we sit together with the industry partners and develop a close collaborative relationship that we build on over time. We are usually working in a one to one capacity. Although we have about 80 projects going within FIND as a whole, we are committed to help shape each project plan, scope, milestones, deliverables, data review, and the way the technology is being designed, if we are in the early stages. In the later stages we find clinical sites and train them in protocols and using the test. We then compile data for them, acting as the arm’s length supervisor.
With the data we can then also do analyses, for which we have a very strong team. The FIND biobank of samples from various infectious disease patients (TB, neglected tropical diseases, malaria) are made available to other researchers looking to develop diagnostic tests. And as we work closely with our various partners, we see very quickly how to proceed to the next steps. It is very milestones-based and a very “go versus no-go” process.
Instead of having one winner, we encourage multiple winners that can address many problems, as we look at everything from a lateral flow test that might be applicable in one area to molecular platforms that have a better fit in others.
The industry partners we have are predominantly in North America, Europe and Asia but we also have partners in Africa, South-East Asia and India.
We really welcome the Longitude Prize as it a good opportunity to bring in new innovation to this space. We look forward to seeing who will stand out and potentially to work with those industry players that show promise for low-resource setting uptake of new tools.
