A Risk Assessment For Acasti

Summary

ACST has been riding on the success of AMRN in the past year; however, several risk factors have cropped up.

A review of past data suggests that there is a high possibility ACST's CaPre will have more side effects than Vascepa.

The HbA1c reduction story, while interesting, has a few points of caution to consider.

A likely need for an outcomes trial, long timeline, and uncertain market prospects will probably provide additional points of entry in the future.

Amarin (AMRN) has been re-defining the omega-3 market with last year's REDUCE-IT results, wherein they demonstrated that addition of Vascepa led to a 25% relative risk reduction in onset of first major cardiovascular event vs statin monotherapy. This was followed by a recent FDA label expansion for Vascepa, which significantly increased its addressable market. Meanwhile, Acasti (ACST) has been riding on the coattail of AMRN's results and has been witnessing significant share price appreciation over the past year.

 

ChartData by YCharts

 

Figure 1. Recent closing prices for ACST.

For unclear reasons, a spontaneous crash and recovery in ACST's share price occurred on Dec 23, which was followed by a press release announcing the delay of top line results for TRILOGY 1, a phase 3 trial, from Q4 2019 to January 2020. Given the opportunity, we decided to take a closer look at ACST to see if it would be worth a position leading up to the data release for their TRILOGY 1 and TRILOGY 2 trials in January.

Quick Intro

Comparison of different options for marine omega-3Figure 2. Comparison of the different prescription grade omega-3 fatty acids from ACST's 2017 National Lipid Association presentation

 

The cardiovascular benefits of omega-3 fatty acids have been debated over the years with clinical outcomes varying widely between trials. However, in 2018, AMRN demonstrated clear and significant benefits for Vascepa in cardiovascular disease. This has generated a lot of interest in similar companies in the space, one of which is ACST.

The two main forms of marine-derived omega-3 fatty acids are docosahexaenoic acid ("DHA") and eicosapentaenoic acid ("EPA"). Different prescription-grade formulations have varying amounts of each component (Figure 2). Vascepa, the market leader, contains only pure EPA. Part of the reasoning is that formulations containing DHA has been associated with an increase in low-density lipoproteins, which have traditionally been associated with increased cardiovascular risk. Interestingly, ACST's Vascepa challenger, CaPre, offers a unique take by using both DHA and EPA derived from krill oil, as opposed to fish oil, and including phospholipids into the mix. Interestingly enough, no significant increases in LDL has been observed with CaPre (Figure 3).

49787334-15776913590490177.pngFigure 3. Lipid effects for CaPre from ACST's October 2019 Investor Presentation.

Shifting Focus

The primary endpoint for the TRILOGY trials is to determine whether 4g/day of CaPre is effective in lowering fasting triglycerides after 12 weeks vs placebo in patients with severe hypertriglyceridemia. This has a decent probability of success, given the amount of data accrued for this class of compounds. The only question is how effective will it be? Instead of focusing on the primary end point, let us shift the focus to adverse events.

Vascepa has an innocuous safety profile, as seen in the REDUCE-IT study. More specifically, daily intake of 4g of Vascepa did not result in diarrhea (9% vs 11% placebo), a common side effect, although there was constipation (5.4% vs 3.6% placebo). There was also a significant increase in atrial fibrillation (5.3% vs 3.9% placebo) and peripheral edema (6.5% vs 5% placebo). Overall, adverse events and event-related withdrawals were essentially the same as placebo in one of the most successful omega-3 trials to date.However, adverse effects can vary depending on the formulation.

 

A look at the label for Epanova suggests a few more complications. The main events were diarrhea (15% vs 2% in placebo), nausea (6% vs 1% in placebo), abdominal discomfort/pain (5% vs 2% in placebo), and eructation (3% vs <1%). Other reported side effects include abdominal distension, constipation, vomiting, fatigue, nasopharyngitis, arthralgia, and dysguesia.

It's arguable if Lovaza fared any better. The most common were eructation (4% vs 1% placebo), dyspepsia (3% vs 2% placebo), and taste perversion (4% vs <1% placebo); however, they had additional skin and gastrointestinal problems, along with elevated levels of alanine aminotransferase and aspartate aminotransferase, which are typically associated with liver problems. As such, there is a warning for patients with hepatic impairment to be regularly monitored.

The similarities and differences in adverse effects for Epanova and Lovaza are quite interesting, given the biggest difference is how they are formulated (e.g. ethyl esters vs free fatty acids and different doses of DHA/EPA).

Where CaPre stands

The question now is what kind of profile would CaPre end up with? The only data we have to work with comes from the COLT trial, an open-labeled (no placebo), dose ranging study. Patients were divided into standard of care (SOC) or one of 4 treatment arms, which were: 0.5g CaPre for 4 weeks, followed by 1.0g CaPre for 4 weeks (0.5-1.0g), 1.0-2.0g, 2.0-4.0g, or just 4.0g for 8 weeks.TAEA for CaPre from COLT

Figure 4. Trial results for COLT, a CaPre dose ranging study

As can be seen in figure 4, as the dose increased, there was a marked increase in adverse events experienced (13.5% to ~40%) in treatment groups. Unfortunately, the distribution of adverse events between cohorts is masked by collation of the results (Figure 4). As such, some of the adverse events may potentially represent 5-10% of the high dose population, as opposed to 3-4% across all groups. In other words, if a placebo were to be present, then there would be good probability that CaPre's side effect profile would not be largely placebo-like, as was in the case of Vascepa. Instead, there may be a number of mild side effects, as is the case for Lovaza and Epanova.

 

A Closer Look at HbA1c Reduction

One of the more interesting aspects of CaPre from the COLT trial was the observation that it could modify serum hemoglobin A1C (HbA1c) levels, a marker for average blood glucose levels over the past 3 months.

COLT trial resultsFigure 5. Trial results for COLT, a CaPre dose ranging study.

If true, it could help offset a potentially (relatively) less appealing adverse effect profile. That being the case, it is difficult to make much from the data in COLT due to a combination of factors.

To further break it down, 0.5-1.0g and 1.0-2.0g saw an increase in HbA1c, but less than 10% of the patients were diabetic (5/50+ for both). In contrast, 2.0-4.0g and 4.0g had reduced HbA1c (-6.8% and -3.5%), but had 26% and 18% diabetics, respectively. Moreover, 2.0-4.0g was 4 weeks at 2.0g CaPre followed by 4 weeks at 4.0g. As such, the reduction effect may be dose-related, and only appears at higher doses, or it may be dependent on the underlying demographics (e.g. diabetics). Likewise, it could be associated with 4.0g (and not 2.0g) in the 2.0-4.0g cohort, but is a transient effect that reduces over time. It could also be that it occurs only in a specific sub-population in the study. Finally, it could be combination of all of the above. There is also the odd observation that a reduction in HbA1c was not paralleled by a reduction in glucose levels. These, and other potential factors, muddle the interpretation of the data and may make reproducing the reduction more difficult.

If it is dependent on a specific sub-population, then the TRILOGY trial, which runs 4 weeks longer than COLT, introduces a new risk in the form of more sickly patients. The COLT trial focused on mild to high hypertriglyceridemia (fasting levels between 200 and 877 mg/dL). In contrast, TRILOGY specifies that triglycerides should be ≥500 mg/dL and <1500 mg/dL, which will result in a greater disease burden.

As a point of reference, cohorts in the EVOLVE trial, which assessed Epanova in patients with serum triglycerides ≥500 mg/dL and <2000 mg/dL, had a mean HbA1c between 6.0-6.2% (median of 5.7-5.9%), with diabetic rates between 30-44%. A further breakdown can be seen in the figure below.

49787334-15775993588739097.png

Figure 6. Hb1Ac data from the EVOLVE trial as seen in the clinical efficacy review for Epanova.

Even in the absence of diabetes, mean and median baseline HbA1c in the EVOLVE trial was in the mid 5% range, which is higher than what was in the COLT study (4.14-5.21%, diabetics inclusive).

HbA1c Reduction Pitfalls?

Besides an increased disease burden, recent pre-clinical data also gives food for thought. ACST recently issued a press release that disclosed preliminary data suggesting that CaPre increased β cell insulin production in a type 2 diabetic mouse model. While the data is awaiting compilation and publication, let us tentatively assume that it translates into humans.

Severe hypertriglyceridemia tends to be associated with metabolic syndrome, which includes insulin resistance, so increased insulin may help reduce Hb1Ac, which is already naturally happening. However, the majority of non-diabetic patients in the EVOLVE trial tended to be in the normal range to begin with. This also begs an additional question of whether hypoglycemia risk will become a problem over an extended period of time, as is the case for sulfonylureas, a popular class of medications used to increase insulin secretion in diabetics.

The situation is slightly different in diabetics. As diabetes worsen, pancreatic β cells continue trying to increase insulin production up until a certain point where they can no longer compensate; moreover, this is paired with subsequent loss of β cell mass. As such, it's unclear how effective stimulated insulin production would be in more advanced diabetics.

 

Moreover, extended stimulation of β cells may not be good. There is a theory that β cell burnout may occur, although the evidence is still inconclusive. However, sulfonylureas have a tendency to lose effectiveness quicker than other anti-diabetic drugs [1,2] and has been associated with increased β cell dysfunction. As such, it is unclear if this would result in closer scrutiny for CaPre.

As an aside, a large study in type 2 diabetics found that aggressive glycemic control (median 8.1% HbA1c with a target < 6%) actually resulted in significantly more deaths. Likewise, there were other trials that suggested glycemic control medications could elevate cardiovascular risk, which has resulted in the FDA requiring cardiovascular outcome trials for new type 2 diabetic medications since 2008. This also leads to another point - that targeting biomarkers can sometimes create unexpected outcomes.

A Long Road Ahead

Data risk, potentially expensive trials, and market navigation make ACST a higher risk hold. Unless CaPre performs unexpectedly well in TRILOGY 1 and 2, it may be met with mitigated levels of enthusiasm. And even if it does perform well, it may be met with some market challenges in the hypertriglyceridemia market. The point of lowering triglyerides and improving lipid readouts is to reduce the risk of cardiovascular events, which only Vascepa has clearly demonstrated to date.

While an outcome trial will unlikely be needed for a hypertriglyceridemia label, subsequent uptake in the marketplace likely will. During the 2020 Q2 conference call, ACST management had indicated that the likelihood of an outcomes trial was high. The question is more of what type of trial will it be, as they wait on results from both their own phase 3 trials and the STRENGTH trial. However, it may be difficult for ACST to rely on results from the Epanova STRENGTH trial beyond light guidance, as CaPre has an added variable in the form of biologically active phospholipids, which may require additional validation.

 

If they were to perform a trial similar to AMRN, then that would take ~7 years and 0M, and this would be on top of any SG&A and other expenses for trying to market CaPre. With a market cap of ~0M, this could equate to significant dilution over the next few years, depending on how share price and finances play out.

Concluding Remarks

ACST has been buoyed by the success of Vascepa, but it is unclear how long this can continue. While CaPre has a decent likelihood of meeting its primary endpoint for the TRILOGY trials, potential side effects may introduce turbulence along the way. Likewise, while the HbA1c reduction aspect adds a unique dimension, it does have some potential pitfalls that should be noted. If the trials work out, ACST will still have to resolve how they plan to address a likely need for an outcomes trial and gain market share. Overall, we will follow the trial results, and may make a short-term play when the time for approval comes along. Beyond that, given the potentially long timeline ahead and biotech's natural volatility, there will likely be more opportunities to re-evaluate the situation.

Author's note: For more articles, please visit my profile here. If you want to receive notifications for new publications, then click the "Follow" button in my profile. Thanks for reading and good luck.

Disclosure: I am/we are long AMRN. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

0
0
0
0
0
0
0 0 254
Submit comment
    No comments yet

Acasti's Value Hidden In Plain Sight

Sept 12 2019 - Shopify: A Reality Check

Oct 19 2020 - AMD: A Gambler's Bet

Aug 13 2020 - YRC Worldwide Inc.: Shares Still Heavily Discounted - Part1B

Feb 20 2020 - Zoom-ing Into The Stratosphere? Not So Fast

Jan 31 2020 - An Ode To Luck: Revisiting My Tesla Valuation

Jan 4 2021 - CrowdStrike Continues Its Huge Run, We Continue To Rate The Stock At Buy

Jan 26 2020 - Sorrento Therapeutics: Growing On Its Own Terms

Submit media
Enter your nickname

Show

Show

Enter your email address and we will send you an email explaining how to change your password or activate your account.

Back to login form

Close