BioVie, Inc. Common Stock Study result

Please go ahead, Cuong. Good morning, everyone.

Thank you, Tara. This, we believe, is a great day for the Long COVID patient community with some of the results that we'll be discussing today. I'm very excited to be sharing this with you. I'm joined today by two experts, which will come on camera later on. Michael Peluso is one of our clinical investigators who's been part of many Long COVID trials, as well as Esra Cakir, who is a Long COVID patient advocate who's also been participating in many trials. Over the course of this presentation, we will be making certain forward-looking statements. I will let you read this at your leisure. This presentation could also be found at our website. We are here because of bezisterim. Bezisterim is a very novel first-in-class molecule that is orally bioavailable, and it crosses freely into the blood-brain barrier.

It is believed to act to block ERK and NF-κB activation, and thus it blocks the production of TNF-α, which is considered to be the master regulator of inflammation. We are studying bezisterim in Long COVID because it's believed, if you look at the graphic on the right, it's believed that the spike protein, the envelope protein on the COVID virus, if you have the infection or if in Long COVID, those fragments of those proteins are believed to activate something called the Toll-like receptor 2 or the Toll-like receptor 4, which essentially activates NF-κB and ERK to lead to TNF-α and the production of other inflammatory mediators. Of course, since bezisterim blocks the activation of NF-κB and ERK, it is believed that it can have all these downstream impact and thus hopefully help with Long COVID symptoms.

So we conducted this Long COVID, this ADDRESS-LC phase II trial. It is designed as a phase II signal-finding proof of concept trial. As you know, drug development typically goes through three different phases. Phase I first establishes that a drug is safe. Phase II is a signal finding. So try to figure out, does the drug have a signal? Does it have a chance of working in some patient population, and how do you show that it works? Then once you get the information from phase II that shows that, well, if we use this endpoint with this group of patients, you then move on to phase III confirmatory trial, which basically says, "This is our endpoint, and we're trying to get a statistical significance." That's the purpose of phase III. It's confirmatory. Phase II is purely signal finding.

What we're trying to do is to identify the efficacy signal. Namely, what endpoint can we use to show that the drug has a favorable treatment benefit? We're trying to measure patient populations that can benefit from that. We need all this information so that we can design and size the phase III trial. In this trial, since it's the first time that we're looking in Long COVID, we evaluated 22 different clinical outcomes to try to figure out which measure works best. We also collected blood samples to assess a series of biomarkers. We're conducting a proteomic study evaluating over 380 different proteins. It should be noted that this work was supported by a grant from the U.S. Department of Defense. Let me first jump to give you the key findings before we go through the details.

The first thing to recognize is that Long COVID patients are not one homogenous group, and in fact, it is a very diverse and heterogeneous population. When we think of Long COVID, we tend to think of fatigue, malaise, and brain fog, and we often assume that a patient has all three of those, and in fact, they do not. Patients can have any combination of one, two, or three of those symptoms. The key finding from our trial is that we believe bezisterim is very effective for these different patients who have the severe symptoms when they start the trial. But the critical thing to recognize is that you need to have the symptom to begin with in order to improve. If you have high symptoms of fatigue, you improve statistically significantly versus placebo on five different fatigue endpoints, along with trending in sleep and post-exertional malaise.

If you have high malaise levels, you improve on malaise as well as cognition. If you have high cognitive impairment at baseline or brain fog, you improve on your brain fog. Lastly, the safety and tolerability profile of bezisterim was very similar to placebo. This is a very critical finding, especially in this patient population, where virtually every patient is on multiple different medications. One thing to recognize is that Long COVID is not COVID. It is not the same, and it shares an unfortunate part of the name. COVID, as you know, is the active infection that comes from the virus. Long COVID is a description of the aftermath of the infection. These are symptoms that are quite diverse.

It could be fatigue, it could be brain fog, it could be very heterogeneous, and it could last for months and years after the actual COVID infection itself. It's a very large population out there. 17 million to 20 million U.S. adults are believed to have Long COVID symptoms now, and 3.8 million have it so severely that they're essentially considered to be disabled. They can no longer work or keep up with the physical demands of their job, and there are no therapies out there. There is a huge, great medical need for a therapy to address Long COVID symptoms. The challenge of developing a drug for Long COVID is actually not only to show that it works, but it's actually finding how. Because one, you see it's a very heterogeneous population. There are no validated biomarkers.

There is no biomarker that you can point to that says, "Ah, this person has Long COVID." It is purely assessed symptomatically and has large background variability, and everybody's on multiple medications. It's actually very difficult to find out how it works, how a drug works, who can benefit from it, and that's exactly what we set out to do, and we believe we have achieved that. I mentioned we measured 22 different endpoints, so this gives you a list of what those endpoints are. And a sense of how to read the numbers. When we're looking at clinical trial results, we're looking at two different things. One is just how large is the effect. So how much of a benefit does treatment provide compared to placebo?

Statistics gives us a tool to do that, something called the Cohen's d, which measures the magnitude of the impact, the difference between two groups. The larger the number, the better. The second, it gets to a question of how sure are we that this result is significant? Excuse me. We look at the P value. The smaller the P value, the better, and typically something is considered to be statistically significant if the P is less than 0.05. At baseline, we compared the bezisterim treatment group versus the placebo group, and we see that virtually everything is very well-balanced, with the one possible difference being slightly more women in the bezisterim treated group compared to those that are on placebo. But everything else is pretty much balanced across the board. The other thing we find at baseline is there is huge variability in the symptoms.

The way to explain this chart is that if you look at the left blue circle, you see that there are 112 patients out of a total of 203 that have high levels of fatigue at the start of the trial. If you look at the green, there were 98 patients who have brain fog, high levels of brain fog, and at the red, there are 72 patients with high post-exertional malaise. Equally important, if you look at the yellow, there are 44 patients, or 22%, with low symptoms in all three. As you can see, there is huge mixtures. Right in the middle, there's only-- If you look around, there are about 35% of the patients who have some combination of two. Likewise, there are about 35% of patients who have just one symptom.

This is really, really critical to understand, because this is where everything starts. If you look at the 112 patients who have high fatigue, that's about half of 55% of the trial. But that also means that there are 45% of the patients in this population that does not have fatigue symptoms. One thing that we have always known about bezisterim, and frankly, many, many other drugs, is that you need to have the symptom to begin with before you can improve on that symptom. If you don't have the symptom, you can't improve, or if you have low levels of that symptom, you cannot improve. So understanding what symptom you actually have, what symptom a patient subgroup actually belongs in fact, is critical to understanding the results. You will see that on the next page.

Because if you throw all 203 patients together and look at the 22 different endpoints, you will see something that's very interesting pattern. You will see that on all except one, all of the patients, all the Cohen's d favor bezisterim treatment. It basically giving us a signal that the drug appears to be working for patients who are taking the drug. Yet, when you look at the p-value for any one of these endpoints, you see that they do not reach the 0.05 level for statistical significance. So if you were to just stop there, you would say, "Oh, the drug did not work because we did not have statistical significance." But drawing that conclusion would be a huge mistake because of the point I made here about the different subgroups. You have to have a symptom to improve upon that endpoint.