Since 1996, InBios International Inc. has worked on rapid and ELISA-based immunodiagnostic assays for infectious diseases. (ELISA, or enzyme-linked immunosorbent assay, is commonly used to measure antibodies, antigens, proteins, and glycoproteins in biological samples.) With a specialization in antibodies and antigens research—a proven method to detect Lyme disease—the team is applying its expertise to help transform Lyme disease diagnostics. Now, the team is one of 10 Phase 1 winners that have advanced to Phase 2 of the LymeX Diagnostics Prize, a prize competition to accelerate the development of Lyme disease diagnostics.
Through September 2023, the Phase 2 cohort is participating in a virtual accelerator designed to help them refine their concepts for detecting active Lyme disease infections in people. The goal of the multiphase LymeX Innovation Accelerator (LymeX) competition is to nurture the development of diagnostics toward Food and Drug Administration (FDA) review.
We spoke with InBios International’s Director of New Product Development James Needham to understand how the team is applying past diagnostics expertise to its concept for a Lyme disease test, improving upon existing testing, and considering how to increase patient accessibility.
InBios has extensive experience in infectious disease diagnostics, including Zika virus, West Nile virus, and most recently, COVID-19. How did the team come together to address Lyme disease and apply past learnings to your concept?
Needham: “We have a diverse group of individuals who have strong expertise in protein generation and protein purification, quality assurance and regulatory affairs, and development and design controls. We just have never done Lyme diagnostics formally; we’ve specialized primarily in flaviviruses and various other diseases. When COVID-19 came around, we all switched gears.
The LymeX Diagnostics Prize definitely gives an opportunity to switch gears and start to use the infrastructure we’ve already developed to apply to this disease. The prize competition helped us steer the resources we already had available.
We did have some previous researchers who had initiated work in Lyme disease years ago. We had developed some reagents in-house already and done some preliminary investigation work, so we knew we had some good reagents already available in-house. With some things in the freezer ready to go, we could start to finesse evaluation and testing in a more sophisticated assay than we’ve ever done in the past. So because we had some history already with developing these reagents, we had more confidence moving forward for attacking this challenge.
Our diagnostic approach is a little more sophisticated than what’s typically used in serological diagnostics. We’ve developed methods for actually screening serum or plasma samples in a faster way, which can provide a lot more information than is typically given by traditional test methods. So because we have so much more information, we’re able to use some machine learning tools to then interpret the additional data, which is usually beyond what would be typically done by any individual lab.
The approach that we’re using helps to automate and provide a huge dataset. So if there’s any kind of disease where a single biomarker is not sufficient to give you a competent diagnostic, we can now say, ’Let’s not just look at that one biomarker. Let’s look at five, 10, 50 different biomarkers in a format that is easily run by a technician who can run ELISA.’ That’s the key difference here. Running an ELISA is a very standard protocol in most any laboratory out there—so if they can run an ELISA, they can run this test. But instead of just having one data point, they get a huge set of data points that are then automatically interpreted for them.”
Similar to current FDA-approved testing methods, InBios’ concept is designed to detect antibodies. How does your diagnostic improve upon today’s two-tier testing method?
Needham: “InBios’ kind of test provides as much or more information than what’s currently provided for a two-tier testing algorithm for a draw from human serum or whole blood. So what that means is—with the single test that you run—you get the highest-quality information that you could possibly get right now in terms of your serological reactivity for both IgG and IgM antibodies with a wide library of potential targets.
All of this is quantified and then goes through machine learning algorithms to do the interpretation for the operator. We feel this gives a much more robust interpretation; it gives much more information to the physician and to the patient, and they’ll have a lot more confidence in the final result.
While we are still looking for antibodies, we’re looking at a large library of potential targets to look for early antibody generation. Compared to the current testing methodology, we think when people go into the position at day four or day five post-tick bite, they’re going to be in a good range where the test can detect the early seroconversion IgM antibody response.
The test is very simple to do in a standard format, and you can get data on the order of 26 different points right away with no issue. That provides a lot more information for early antibody response compared to what you typically get in the current ELISA test.”
Besides accuracy, the most important aspect of any diagnostic is accessibility. How has the team considered the provider and patient experience in design, and how could the test progress in the future?
Needham: “Currently, Lyme disease tests are fairly complicated. There are a lot of markers that people have to look at to have a good diagnostic for Lyme disease, and because of that, you need to have a test that’s prescribed by a physician.
We do think this test would be much faster than what’s currently done with a two-tier testing method. You could get a lot more information and have a lot more confidence in your results in terms of your sensitivity and specificity, even at the acute phase of Lyme disease. Those are the things that we’re aiming for. Trying to transition that to something that’s at the point of care or available over the counter is another leap forward, but the kind of platform we’re generating helps us take the stepping stone toward that point-of-care, over-the-counter test. But we have to get this kind of platform in place first before we can start to make something that’s even so simple that you could run it at the doctor’s office or offer it over the counter.”
In Phase 1 of the LymeX Diagnostics Prize, InBios was selected as one of 10 winners out of 52 submissions. Each winner was awarded $100,000 and received an exclusive invitation to participate in the nine month Phase 2 virtual accelerator, which provides access to digital resources and subject matter experts. What has been most useful to your team?
Needham: “What I find most exciting and useful are the one-on-one office hours. Having these office hours to speak with the advisors, and in particular to communicate with people who have access to these biorepositories. You have a good understanding and feel for them and where they’re coming from. And then you can talk about the data and improvements, and discuss what kind of samples they have. Really, the critical thing for developing any novel assay is having high-quality samples and highly characterized sample sets.
If you don’t have that, you can’t actually make a new diagnostic or improve anything. So having access to these biorepositories is very exciting for us. Having access to that expertise to just pick their brains and understand where they’re coming from is highly important, especially when we’re coming at this from a different diagnostic perspective.”
By utilizing machine learning and automation in interpretation, InBios’ test could potentially vastly increase the amount of data analyzed. If validated, what could this approach mean for other diseases?
Needham: “The test opens up a whole new window in terms of what can actually be used in the field, not just in publications. We have made it where it’s robust enough and simple enough to be used by an operator that it could actually go through that FDA approval process, and then open up that field for diagnosing diseases that are more complex in nature or that may represent co-infections, where you have multiple diseases that can be diagnosed simultaneously.
We’re not using the machine learning to actually determine what biomarkers to use. We’re using the machine learning to say, ’Here’s your signals on your biomarkers. Here’s the pattern of signals on your biomarkers. Now given this pattern that we have, how can you potentially identify the actual status of that person? Is it positive, is it negative?’ It’s not as simple as, biomarker A is above this level. You now have 52 different data points you’re looking at simultaneously, and then you can look at that and interpret it—and not have a person do this individually, but have the machine do it for you and give you the results.”
Looking ahead: Expert judging panel to convene in October 2023
Following the accelerator, the cohort will submit concept papers that detail solution refinement, clinical and patient input, and a roadmap from lab to market. The competition judging panel—composed of experts across biology, clinical and technology translation, patient experience and advocacy, diagnostic science and technology, exponential innovation, and ethics—will evaluate eligible submissions according to official Phase 2 evaluation criteria. Based on the judges’ evaluations, the panel will recommend up to five Phase 2 winners of the LymeX Diagnostics Prize.
To receive all Phase 2 updates, subscribe to the LymeX Diagnostics Prize newsletter and follow @Lyme_X on Twitter.