The $10 billion unicorn in INmune Bio: on BioVie teachings and trustworthiness

Summary

BioVie always argued that it saw a correlation between effects on cognition and reduction of TNF levels. That TNF reduction proved to be all-in-all moderate. BioVie’s results, from a perspective of cognition and biomarkers, could be of guidance for INmune investors.

However, BioVie’s Phase 3 failed. Are there any teachings for INmune investors?

Finally, trustworthiness of management and major shareholders is of essence. I think INmune stands out in the Alzheimer’s space here.

BEZISTERIM/NE3107: Phase 3 fraud, cognition in AD, on-time in Parkinson’s

BioVie’s Phase 3 failure: where are we now, and is INmune’s trial at risk?

BioVie’s six-month Phase 3 trial in Alzheimer’s has failed. The company, backed by Pentara’s Suzanne Hendrix, pointed to massive fraud and data anomalies related to one specific geographical area (Florida), the trial having enrolled many healthy people from that area instead of patients suffering from Alzheimer’s disease. That led the study to be underpowered to detect statistical significance, as data of only 20% of the entire patient population could be assessed.

When announcing results, Annovis also claimed that some sites had enrolled people not suffering from Alzheimer’s. There was an overlap of Florida sites in BioVie’s Phase 3 trial and Annovis’ Phase 3 trial, so it is possible that such statement is correct.

However, it is still unclear to this day whether BioVie’s massive fraud allegations are correct. There have been no public statements from BioVie with regards to any or more of the trial sites that have been shutdown by the FDA as a result of the alleged fraud. Some investors have considered that the alleged fraud was ultimately the responsibility of BioVie, as the trial’s sponsor.

INmune considers that its trial supervision should exclude any risk of such fraud. The trial also did not enroll during Covid times, which may help. Ultimately, the FDA’s clinical hold may have prevented INmune Bio from enrolling any patients in the affected geographical area, which ultimately may have benefited the company.

Bezisterim/NE3107 MoA and results

Bezisterim/NE3107’s mechanism of action

Bezisterim is a chemical derivative of the natural mammalian sterol, βAET, which is a metabolite of DHEA. Bezisterim mechanism of action targeted the impairment in the insulin receptor signaling pathway that is causing chronic low-grade inflammation.

BioVie’s trial design and blinded results

BioVie did not select patients on the basis of their levels of inflammatory markers. As a consequence, there was a near-certainty that the trial may have non-responders.

However, a Phase 3 enrollment update shared by the company indicated that the trial may have enrolled patients with exceptionally high markers of obesity and inflammation. In fact, BioVie had informed investors that baseline data for these patients showed a mean CRP value of 4.1 mg/l, without even having selected patients on the basis of inflammatory markers. That should have been a red flag, with 20-20 hindsight. In comparison, INmune’s Phase 2 trial enrolls who have at least on inflammatory biomarker, among which the threshold for hsCRP is 1.5 mg/l.

BioVie had furthermore presented blinded data showing, among others, consistent population changes from baseline on a variety of cognitive rating scales.

Bezisterim results

I review, here, the entirety of data shared by BioVie from both its failed placebo-controlled Phase 3 trial in Alzheimer’s, its open label trial in Alzheimer’s, and its placebo-controlled Phase 2 trial in Parkinson’s disease.

To be clear: cognitive or motoric deterioration is the normal disease progression for these diseases.

- In the Phase 3 trial, in the small subset of patients whose data could be analyzed, Bezisterim numerically improved cognitive and functional scores of Alzheimer’s patients.

- In a phase 2, open-label, single arm trial of patients with mild cognitive impairment or mild Alzheimer’s disease, treatment with Bezisterim for 3 months resulted in enhanced cognition compared to baseline.

• Among MCI and mild Alzheimer’s disease patients, treatment for 3 months led to a 2.1-point improvement on the ADAS-Cog12 scale compared to baseline (p=0.0173; equivalent to a 21.1% change).

• Among only responders, the improvement was 3.7 points compared to baseline (p=0.0003; equivalent to a 36.2% change).

• Bezisterim treatment led to an improvement of 0.11 on the Clinical Dementia Rating scale (CDR) compared to baseline (p=0.0416; equivalent to a 19.4% change).

• Bezisterim treatment led to an improvement of 0.07 points on the Alzheimer’s Disease Composite Score (ADCOMS)(p=0.0094; equivalent to a 27.4% change).

• Patients improved in daily function as measured by an increase in the Global Rating of Change scale2 by +2.67 points.

- In a Phase 2a study in Parkinson’s disease, Bezisterim eased both motor and nonmotor symptoms when used in combination with a standard carbidopa/levodopa regimen in people with Parkinson’s disease. Bezisterim resulted in a 2.8-point greater reduction on part 3 of the MDS Unified Parkinson’s Disease Rating Scale (MDS-UPDRS), a measure of how severe motor symptoms are, compared with a placebo. In the MDS-UPDRS, lower scores mean less severe symptoms. Patients younger than 70, who made up about half of the study population, showed an even greater difference in motor symptoms, an advantage of 4.7 points in favor of Bezisterim. On day 28, nearly one-third (30%) of patients treated with Bezisterim reported better motor function in the morning, before taking their standard carbidopa/levodopa medication, whereas none of the patients on a placebo experienced this.

What could be in it for INmune investors?

Correlation of TNF reduction and cognition

BioVie in essence had an anti-inflammatory approach. I believe BioVie’s results so far do suggest that Bezisterium has some efficacy on cognition, which was in correlation with reduction of TNF-alpha. This is a slide from the open label Phase 2 study.

And this is a slide from the limited Phase 3 results.

TNF reduction through inhibition of the pro-inflammatory TNFR1 pathway may bear some similarity to sTNF inhibition

Bezisterim was not selective for soluble TNF, to be clear. So there is certainly no identity between the MoA of Bezisterim and that of XPro.

But Bezisterim may have inhibited some pro-inflammatory sTNF signaling through inhibition of NF-KB.

In its corporate presentation, BioVie hypothesized “that the modulation of TNFα levels and its inflammatory activation via TNFR1 lead to a multitude of changes among the many factors downstream from this master regulator, which collectively lead to improvements in neuronal health and cognition.”

My thesis, on the basis of this, was that inhibition of NF-κB as a central regulator of pro-inflammatory cytokines through TNFR-1 should lower inflammation. NF-kB signaling has been reported to have a critical role in the pathogenesis of AD by being activated through TLR2, TLR4 and CD14, as it facilitates the production of several factors that constitute AD pathology, such as amyloid plaques, neurofibrillary tangles and neuroinflammation. Its inhibition may therefore be useful for the treatment of neurodegenerative diseases. TNFR1 is the receptor to which soluble TNF binds. It mediates inflammation and apoptosis. Bezisterim inhibits JNK and ERK signaling in TNFR1, which may lead to (selective) inhibition of NF-kB and reduction of inflammatory factors. Again, that was my thesis.

Bezisterim’s lack of potency

However, Bezisterim was not very potent in reducing inflammation.

If BioVie’s drug indeed lowered inflammation, it should come as no surprise that it may have potential to affect cognition and function in patients with Alzheimer’s and Parkinson’s disease with biomarkers of inflammation. At an Alzheimer’s conference some years ago, it was mentioned to me that BioVie’s reduction of TNF was all in all minor.  Let’s look into that, because I believe it is essential.

In a slide of its older corporate presentation, BioVie had shared that the TNF-α reduction in all patients from its Phase 2 trial was -0.452 pg/mL after three months. The company had also shared this slide, showing baseline mean plasma TNF-α to be around 7–8 pg/mL.

From an absolute perspective, the TNF reduction is minor. From a relative point of view, it is too.

In fact, even Longeveron’s TNF reductions appeared to come in stronger.

However, the correlation with ADAS-Cog12 improvement (r=0.46–0.59, p<0.01–0.05) in the open label BioVie trial still implies that the TNF reduction, though small, may contribute to cognitive benefits which are clinically relevant in AD.

Insofar as I know, INmune Bio did not report changes in TNF. It did report this slide, showing MRI and CSF inflammatory markers being reduced substantially after three months of treatment on XPro. The MRI marker measured is white matter free water, an indirect measure of inflammation. The CSF markers are more direct markers of inflammation, with a mean reduction of 15%.

The -45%/-50% drop of CCL7 is important, because CCL7 is highly reactive to changes in TNF. In the relevant webinar, it was referred to as the TNF shadow. It goes up very quickly and robustly when TNF is increased and it reduces very quickly and robustly as it relates to TNF when it is attenuated.

Preclinical data on XPro1595 suggest near-complete sTNF neutralization. XPro1595 was designed to neutralize >99.9% of soluble TNF signaling. In a study in hemiparkinsonian rats, researchers administered XPro1595 subcutaneously to rats at a dose of 10 mg/kg every third day, which neutralized more than 99.9% of sTNF at concentrations typically observed in chronic inflammatory neurological disorders. “These levels of XPro®1595 within the CSF (1–6 ng/mL) are sufficient to neutralize >99.9% of soluble TNF at the pathological levels found in most chronic inflammatory neurological disorders, such as the 30–90 pg/mL reported for Parkinson's patients CSF“.

Another preclinical finding compared TNF-inhibitor Etanercept with XPro on different markers of inflammation, which showed that the TNF reduction of XPro was nearly identical to that of Etanercept, with many other pro-inflammatory markers decreasing stronger on treatment with XPro.

In terms of magnitude, I believe Bezisterim’s TNF reduction is a couple of order of magnitude smaller than that of XPro. That should eventually be reflected in INmune’s results on cognition in patients with biomarkers of inflammation. The biological plausibility of that, given XPro1595’s mechanism of action neutralizing sTNF, is also exceedingly higher.

On trustworthiness and investability

Apart from the praise I receive for these blogs ahead of INmune Bio’s Phase 2 readout,  I recently read a brief comment by an investor stating that I cannot be trusted. Though the headline reflects my own opinion on INmune’s potential, I try to make these blog posts fairly factual in order to avoid such remarks.

There are many other players in the stock market whose trustworthiness is, regrettably, doubtful. Some shorts are untrustworthy, others are not. Cassava Sciences’s stock promotor Joe Springer was untrustworthy in my eyes; it has taken me a while to figure that out, but once I had figured it out, I have tried to convey this to people, alas to no avail in most cases. Cassava Sciences’ former management was untrustworthy. I think that some investors in the stock market already knew ahead of Cassava Sciences’s first press release on cognition that the company would report that.

That is not the case for INmune Bio.

Some CEO’s are not trustworthy. Annovis’ CEO Maria Maccecchini could be a good example, in my view, and the current stock price reflects that. After announcing ‘good’ results for Alzheimer’s and Parkinson’s and having had a $700 million market cap in 2021, the market currently assigns Annovis a $22 million valuation. I believe quality of management reflects investability.

I believe management of Annovis and of INmune Bio are two worlds apart.

Terren Peizer, the ex-CEO of BioVie, may need to be considered untrustworthy. When an indictment had been made against him in March 2023 for abuse of an insider trading plan in Ontrak, another company he owns, he has been removed as chairman of the board. Some INmune investors have been close to the Ontrak story, and are wary of anything Terren Peizer is involved in. They have frequently warned me.  With regards to that indictment, I had been informed that the defense Mr. Peizer had was strong, and that chances of him getting convicted for insider trading were small. Something can be said for that, as the SEC had recently changed insider trading plan rules, and Mr. Peizer had been advised that his trading plan was legal. However, the people ruled different. In June 2024, it was reported that Mr. Peizer was found guilty of a multimillion-dollar insider trading scheme. The main allegations was that he allegedly avoided more than $12.5 million in losses by entering into two Rule 10b5-1 trading plans while in possession of material, non-public information concerning the serious risk that Ontrak’s then-largest customer would terminate its contract. Mr. Peizer knew about the expected customer loss when he created the plan. Mr. Peizer’s defense was that Mr. Peizer wanted to exercise warrants that were about to expire, not sell common stock, and that Mr. Peizer expressed his intentions to sell his expiring warrants long before the customer aired any problems over its deal with Ontrak. Mr. Peizer has appealed that judgment. His sentencing hearing should have taken place in February 2025. It is unclear to me at this time whether he has been sentenced. If he were to be sentenced to jail, he would follow junk bond salesman Michael Milken, whose protégé he used to be. That is not comforting, certainly knowing that Mr. Peizer had escaped convictions based on his testimony against Mr. Milken. I believe the above is related to Wall Street’s general appreciation of investability in BioVie.

Though the current CEO of BioVie, Cuong Do, appears to manages BioVie as it should be managed, one cannot judge or know what the relationship between management and Terren Peizer is. I believe BioVie’s low market cap reflects the involvement of Mr. Peizer as ex-CEO and chairman and continued large shareholder.

As for INmune, I believe management has an impeccable track record. That, including many other things presumably, may have led Harris Kupperman’s Praetorian Capital to take a 10% stake in the company.

Conclusion

There may be a correlation between efficacy in neurodegenerative diseases and reduction of inflammation. Meta analyses of patients treated with TNF inhibitors have shown it repeatedly, and I believe BioVie has too.

BioVie has shown interesting efficacy in three trials in two neurodegenerative indications. I believe that there may be some similarity between the mechanisms of action of BioVie’s Bezisterim and INmune Bio’s XPro. However, if that would be the case, its hypothesized signaling inhibition was less straightforward than XPro’s. Furthermore, BioVie’s drug may not be very potent. In fact, I assume it be far from as potent as XPro.

I believe both elements above may play in favor of a favorable readout of XPro’s Phase 2 trial.

As for fraud risk, both BioVie and Annovis have considered that their trial enrolled patients that did not have Alzheimer’s disease. One can continue to wonder whether BioVie’s trial would have been successful, had it not enrolled such patients. One could argue that fraud could never be accounted for. Others have stated that the trial sponsor is ultimately responsible, and that the alleged fraud could have been prevented with proper trial control. From what I understand, INmune Bio’s supervision of its Phase 2 trial is thorough. I consider the risk of running into a similar fraud allegation to be low.

Finally, on trustworthiness, I am happy not to see Terren Peizer or anybody with a similar profile in the management or on the board of INmune Bio. It wouldn’t portray well. I understand INmune Bio’s management team to be qualified and ethical. At least in terms of investability, I believe that to be of utmost relevance.

---

INMB investors discord: https://discord.gg/JEA8r7wCGY