BIVI - BioVie: Well-Powered Phase 3 Trial In Alzheimer's Could Greatly Benefit Investors

2023-09-08 09:00:00 ET

Summary

• In 2022, BioVie had reported impressive data from a small Phase 2 trial in Alzheimer’s.
• In October-November 2023, the company will report topline data from its ongoing Phase 3 trial in Alzheimer's.
• The reported baseline data of the patients in the Phase 3 trial appears to match the drug's mechanism of action.
• Several recent events including a recent SEC filing give me the impression, similar to what I saw in 2022, that BioVie is setting up for success again.

Thesis

It has been about five months since I last covered BioVie. This should be the last article on the company prior to topline data from its Phase 3 trial in Alzheimer’s disease [AD] in October-November 2023.

I believe there are good chances of success for that trial. I try to analyze how results from the Phase 2 trial could be compared to the Phase 3 trial, and how to look at BioVie’s biomarker data.

I try to define what I consider to be a ‘success’ for Wall Street. BioVie has given back most of its gains after reporting results from its Phase 2 trial, and currently has a market cap of about $110 million. In case of success, the gains could be enormous.

About a year ago, I had written an Editor’s Pick in which I considered that BioVie was possibly setting up for success with Alzheimer's results imminent. A recent SEC filing has me consider that BioVie may again be setting up for success.

Company

BioVie ( BIVI ) has two treatment candidates and is targeting three indications. NE3107 is a drug candidate for Alzheimer’s and Parkinson’s, BIV201 for refractory ascites. These were the company’s timeline and expected catalysts as announced during BioVie Day in March 2023.

The commercial benefit in case of success in these indications is as follows according to the company.

A Phase 2 trial in Alzheimer’s had ended successfully . Three further trials were planned in Alzheimer’s over the course of 2023 and 2024, entitled Radiance-MCI 1, Radiance-MCI 2 and Radiance-AD. The company has yet to start these.

A Phase 2 trial in refractory ascites had been ended preliminarily after clear success . The company is awaiting feedback from the FDA.

Finding a cure for Alzheimer’s is one of the largest medical needs in the world today. Hence, good Phase 3 results could spark unseen investor interest.

With a low of about $2 and a high of about $14, one could say BioVie’s stock has been very volatile over the past year. Currently, the company’s stock is trading close to its year-low.

BioVie has recently issued baseline patient data for its Phase 3 trial. The focus of this article will be on the threshold for investor success in Alzheimer’s right now, why I interpret the Phase 3 baseline data as bullish, and which data from the Phase 2 trial can be used when assessing chances of success in the Phase 3 trial.

The threshold for investor success in Alzheimer’s

The years 2021 to 2023 have been a renewed interest of investors in potential treatments for Alzheimer’s, with Phase 3 results coming among others from Eisai ( ESALF ) ( ESAIY ) / Biogen ( BIIB ) and Eli Lilly ( LLY ). These results were results from blinded placebo-controlled Phase 3 trials, and with BioVie’s upcoming Phase 3 trial topline data, are best suited for reasons of comparison. Successes in AD also come with substantial market cap gains. In the case of Eli Lilly, one-day market cap gains after announcing topline results for donanemab were $28 billion, and the ride up did not stop there.

Although even rather limited success could be worthwhile for investors given BioVie’s market cap and the potential to use NE3107 as an add-on therapy without adverse effects, I believe investors will only start looking at BioVie when topline Phase 3 data reaches or outperforms thresholds now set by anti-amyloid antibodies. Adverse event-related issues such as ARIA that coincide with anti-amyloid antibodies seem to be given less importance.

Eisai’s Leqembi had set that threshold at a 27% slowing of cognitive decline, and Eli Lilly’s donanemab raised that bar to 35%. Both results were reported on the CDR-SB rating scale. They came from 18-month trials in an MCI/mild AD population. Though results have been interpreted by many investors as being a solution to Alzheimer’s overall, they are not.

For reasons of comparison, the Adas-Cog data for Leqembi is shown below.

That is a 1.44 point difference to placebo after 18 months. The divergence started only at six months.

BioVie’s Phase 2 trial was designed to see an efficacy signal

BioVie reported the following results in its open-label Phase 2 trial, after three months.

The green color indicates an improvement. BioVie flatlined or improved cognition at the 3-month timepoint, in a population that was identical to the one in which the afore-mentioned anti-amyloid antibodies had been tested. That data came in particularly strong, even with 8 non-responders to treatment, most of which - but not all - were at the moderate end of dementia.

BioVie itself noted 61% of patients improved, and 82% in the MCI/mild population. BioVie also reporting that 60% of patients saw reductions in TNF levels.

As impressive as I find that data is, I think that one would be wrong to expect the same for the upcoming Phase 3 data. The Phase 2 trial was not designed to lead to improvement in cognition. The trial was designed to see whether NE3107 could show an efficacy signal in the entire MCI/mild AD population. It had few inclusion and exclusion criteria.

For me, the efficacy signal taken together with multiple biomarkers over the course of 3 months of treatment, which coincided with the expectation that one would see an effect, is what to take from that data. Across different articles on different companies, including on BioVie, I have always cautioned for non-responders in trials with all-comers.

As mentioned above, BioVie’s Phase 2 trial came with 8 non-responders, which I had discussed here (title: ‘The Cognition Data’).

If any trial may come with non-responders, one may want to ensure there are as little as possible. In my opinion, it would be inappropriate to focus on a subset of 5 non-responders in the moderate population to try to assess how this subgroup would perform in the Phase 3 trial. If one does, one can be bearish on the upcoming readout, as is Seeking Alpha analyst Biotech Pharma Investor. The merit in that approach could be the idea that a moderate population is harder to treat, and that is legitimate.

As it appears, at least two patients in the MCI/mild AD subgroup had similar trajectories. I explain below why I called these patients non-responders, why I believe BioVie’s Phase 3 trial may have a very limited amount of those, why I believe the totality of biomarker data needs to be taken into account, and why I am bullish. But before explaining that, I would like to address the group of patients that could potentially benefit from NE3107, and whether it is risky to target another group of patients.

The biomarker and epigenetics data

Cognitive efficacy is the result of changing biomarkers, and a good interpretation thereof may give an indication of potential cognitive effects. I have covered biomarkers reported by BioVie before.

Specifically as to the biology of aging, DNA methylation of its Phase 2 trial patients was significantly reduced by -3.3 years on the Horvath SkinBlood clock over the course of 3 months. Insofar as I am aware, that is the most impressive data on any biological clock to date. Furthermore, 19 out of the 22 patients experienced a reduction in their SkinBlood clock score. On July 18, 2023, BioVie reported that treatment with NE3107 led to >50% reductions in DNA methylation of 400 Sites on DNA strands where DNA can be methylated, some in a manner that was significantly correlated to observed clinical improvements in cognition and various biomarkers. Over 3,000 significant correlations were found linking reductions in DNA methylation of various CpGs and cognitive, biomarker and neuroimaging endpoints. At the time this article was submitted, BioVie had a webinar with professor of neurology Steven Arnold scheduled to discuss the data so far, ahead of its upcoming Phase 3 trial data.

NE3107’s mechanism of action

Given the multifactorial nature of AD, I have cautioned against an imperfect Phase 3 trial design. Trial designs would be better suited if they focused on the biology of patients the drug intends to treat. The Alzheimer’s Drug Discovery Foundation [ADDF] recently stated :

But this is just a start, and we must continue advancing the drug pipeline to develop the next class of drugs centered around the biology of aging to ultimately stop Alzheimer’s in its tracks. Like in cancer, the goal is to address the many underlying pathologies of the disease through a precision medicine approach.

NE3107 is an anti-diabetic drug with anti-inflammatory properties, or an anti-inflammatory insulin-sensitizer. Insulin is the master energy regulator and TNF is the master regulator of inflammation. Both are well-known targets for the treatment of AD. A scientific article set out the rationale for the Phase 3 trial, and another document on NE3107 from the ADDF states (own highlighting):

NE3107 may exert neuroprotective effects through its anti-inflammatory and insulin-sensitizing activities (Reading et al., 2021). There is also ample evidence that type 2 diabetes and Alzheimer’s disease share certain characteristics, including impaired insulin signaling and oxidative stress (Sebastiao et al., 2014). Thus, addressing brain insulin resistance, theoretically, is likely to show neuroprotection and cognitive benefits in Alzheimer’s disease as well as in age-related cognitive decline. [...]

Overall, in metformin-treated type 2 diabetes patients, NE3107-responsive population was found in the upper two tertiles of the inflammation marker MCP-1 (>40 pmol/L). In treatment-naïve type 2 diabetes patients, the NE3107-responsive population was found in people with higher than median BMI (BMI over 31 kg/m 2 ). Thus, NE3107 appears to be targeting the impairment in the insulin receptor signaling pathway that is causing the chronic low-grade inflammation. People with type 2 diabetes without the chronic low-grade inflammation are likely to have glucose intolerance due to disruptions in pathways other than the insulin receptor signaling, and therefore, are not responsive to NE3107 treatment .

For NE3107 to be efficacious, patients with higher than median BMI and with markers of inflammation are the perfect target. Administering a drug that does not match patients’ biology could end up making such patients worse. TNF inhibitors in depression in patients with a high level of inflammation saw benefit from a traditional TNF inhibitor, but patients without such high levels of inflammation performed worse than placebo. If something similar were to reoccur in AD trials, then the trial is at risk of not reaching statistical significance. Interestingly, whereas the responders in BioVie’s Phase 2 trial saw maximal cognitive benefit, the non-responders which were about 39% of the entire population declined faster than normal progression over the same period.

There is, therefore, an interest in excluding a potentially non-responding patient population.

The characteristics of the Phase 3 trial

The Phase 3 trial is both in mild and moderate patients

BioVie’s Phase 3 trial enrolled patients that had an MMSE score of ?14 and ?24. An MMSE score of 21-26 results in mild AD, a score of 10-20 in moderate AD and a score of less than 10 in severe AD. Mild patients were also included in the Phase 2 trial.

The Phase 3 trial is in probable AD

This is a similarity with the Phase 2 trial which could allow BioVie to address a larger patient population than only an amyloid-positive one. Probable AD according to the NIA-AA 2011 criteria indicates that a patient displays symptoms consistent with AD, and may have undergone biomarker testing that shows amyloid positivity.

Different to amyloid, neuroinflammation is a hallmark across several neurodegenerative diseases/ dementia , so it makes sense to consider that NE3107 may be efficacious there - cfr. BioVie’s positive results in Parkinson’s.

BioVie’s baseline Phase 3 data mentioned that both A?+ and A?? patients with AD were enrolled in the study.

The baseline Phase 3 data may exclude or minimize non-responders

BioVie’s Phase 3 trial has enrolled 85% of patients with a high waist-to-hip ratio, at the threshold of obesity.

BioVie’s showed the following CRP values for the enrolled patient population.

A CRP value of 3 is a threshold often considered to represent acute inflammation. For reference, INmune Bio’s ( INMB ) hsCRP-threshold for inclusion in the Phase 2 trial is 1.5 mg/l. With 67% of patients having a CRP value above 3, BioVie has enrolled a patient group with markers of inflammation.

Contrary to its Phase 2 trial, BioVie’s Phase 3 baseline data appear to exclude or minimize the risk of having enrolled non-responders. That should allow raising the responding population from 61% to a higher number.

Potentially higher efficacy in reducing TNF

The high CRP values also give me confidence that TNF values may be reduced more than those seen in the Phase 2 trial. In the Phase 2 trial, BioVie had reported a positive correlation between cognition and the reduction of TNF, but TNF values were only reductions in 60% of patients, and the overall reduction was all-in-all moderate.

High baseline inflammatory markers may see higher reductions of TNF, and correlated cognitive effects.

Primary Phase 3 trial endpoint aligned with that of Eisai / Biogen and Eli Lilly

Recent successes from anti-amyloid antibodies were reported on the CDR-SB rating scale. BioVie has changed its primary endpoint to this scale. Keeping the Adas-Cog rating scale as primary endpoint would not have allowed such a clear comparison. I believe that, by changing that primary endpoint, BioVie has wanted to provide a comparison to such investors. I see that as correct, and bullish. BioVie’s intent-to-treat population is nonetheless harder to treat.

On trial size, duration and expected efficacy

Some have criticized BioVie’s Phase 3 trial may not sufficiently powered as it could be too short. Anti-amyloid antibodies reach the efficacy threshold very slowly, but are not the gold standard. Anti-inflammatory treatments may have a faster impact. How large and long a trial needs to be depends on 3 things: the magnitude and variability of impact and the time to efficacy. NE3107 has shown impressive efficacy in responders, and preclinical studies, the Phase 2 data and Parkinson’s data show that the drug acts fast. If placebo patients decline about 1.5 Adas-Cog points over 7 months, then there are statistical approaches to estimate sample size needed to show the difference between a 1.5 point worsening and expected result. I assume BioVie has applied these approaches.

A 6-month Phase 3 trial in 378 patients, enlarged by BioVie’s own decision from an initial patient population of 316 patients without a pre-specified interim analysis due to fast enrollment, is not small and could be sufficiently powered to show the difference between the treated and placebo groups.

The relevant threshold for investors: 27-35% slowing of cognitive decline

BioVie’s Phase 3 trial is not a rerun of its Phase 2 trial, and I do not expect BioVie to flatline or improve cognition. At the end of a Phase 3 trial, the threshold for Wall Street is whether NE3107 can perform similarly to Leqembi or donanemab. Anything similar to that could be world news, anything above may discard Leqembi from its current top place and may have all eyes on BioVie, including those of big pharma.

Looking at the baseline Phase 3 data, chances are reasonable that the totality or at least the majority of patients will respond to therapy. That may de-risk the trial from the perspective of statistical significance.

In summary, I consider chances of seeing an efficacy signal that is much higher than that of the Phase 2 trial, and if so, reaching statistical significance, to be very realistic. Whether or not NE3107 will reach the above threshold of slowing down cognitive decline cannot be stated in advance, but I believe BioVie has good cards.

BioVie may again be setting up for success

A year ago, I had noted that BioVie was possibly setting up for success. BioVie was rather unknown for its AD trial at that stage. Its stock had more than tripled three months later. As it appeared, BioVie had been setting up for success. Three SEC filings had me consider this:

- an updated corporate presentation with the inclusion of the Phase 2 trial;

- a SEC filing allowing BioVie to raise $20,000,000;

- a SEC filing reporting a private placement of 3,636,364 shares at $1.65 per share, and a warrant to purchase 7,272,728 shares for Acuitas Group Holding, BioVie’s major shareholder’s investment vehicle.

I have highlighted baseline Phase 3 data above.

At this time, I want to further draw investors’ attention to:

- BioVie Day, covered here, which was an obviously bullish event;

- a recent SEC filing one of the order of magnitude rarely seen in smaller biotech companies.

On August 18, 2023, BioVie made an S-3 filing for a primary offering of not less than $300,000,000 in different securities and a secondary offering of up to 311,002 shares of class A common stock. That offering is now effective . The company’s market cap is currently $113 million. The only way I can imagine BioVie picking up that kind of money in the current market is by bringing news that would create enormous investor interest. Hence, I see that filing as at least a vote of confidence in BioVie’s future.

Terren Peizer appears to be in it for the long haul

Investors have questioned whether BioVie’s major shareholder would be inclined to sell his stake in BioVie. As it happens, he does not appear to have participated in any sales, neither on the way up or on the way down. He appears to be in it for the long haul. The vast majority of his shares and warrants are unregistered with the SEC. These cannot be sold until the company registers them.

Financials

As of June 30, 2023 the Company had cash of approximately $33.9 million. Yearly net loss as reported for the year ending on June 30, 2023 was $50.3 million, $24 million more than the year before. BioVie has used an ATM facility to generate cash, and I assume it will continue to do so. I have mentioned the recent S-3 filing for an amount of $300 million above.

Risks

I have covered several risks related to investing in BioVie in my previous coverage. At this time the most prominent risks I would note are:

• Trial failure, given the potentially binary nature of the upcoming event;
• FDA delays or response letters with regard to trials BioVie intends to start;
• Further administrative or other hurdles to approval;
• The need or wish to pick up additional funding, as seen from the recent S-3 filing;
• Competitive risk, such as Cassava Sciences Phase 3 trial – although risk is limited if one makes a 15% market penetration assumption, like BioVie.

Conclusion

The last patient visit of BioVie’s Phase 3 trial should occur in September 2023. BioVie has guided topline data to be reported in October-November 2023.

The company’s news over the past year has been as remarkable as the volatility of its stock price. BioVie has reported successes from a Phase 2 open label study in Alzheimer’s patients, a placebo-controlled trial in Parkinson’s disease, and a Phase 2 trial in refractory ascites. The company’s stock has given back most of its gains and is currently around $100 million.

The upcoming topline readout has the potential to be a binary event. A recent S-3 filing that would allow BioVie to raise $300 million and sell up to 311,002 shares would indicate such potential.

I have set out above why I consider BioVie’s Phase 3 trial as rather de-risked, insofar as that’s possible in a historically difficult-to-treat disease such as Alzheimer’s disease. My considerations revolve mostly around the enrolled patients’ biology, which seem to be a match for NE3107’s potential for efficacy. I believe that, where the Phase 2 trial allowed all-comers and had a 61% response rate, that response rate could be much higher in the Phase 3 trial, which should benefit efficacy consistency, confidence intervals between treated and placebo patients, and the potential to reach statistical significance. I do not expect BioVie to repeat flatlining or improving cognition as it did in its Phase 2 trial. If it does so, all the better. I believe the threshold for investors to be about two thirds. If NE3107 would be able to generate such result in a harder-to-treat mild to moderate AD population, then I would look at that as an outright success.

If the trial fails to reach statistical significance, then I believe BioVie may have to revisit its announced trials. I consider chances for that to happen low.

For all the reasons above, I have a Bullish rating on BioVie at this time. Though I have not become less bullish since the last time I covered the company, a Very Bullish rating right before topline data in a notoriously hard-to-treat disease seemed not appropriate.

For further details see: