Acasti Pharma Inc. (NASDAQ:ACST) Q4 2020 Earnings Conference Call June 29, 2020 10:30 AM ET
Alexandra Botie – Crescendo Communications, LLC
Jan D’Alvise – President and CEO
Pierre Lemieux – COO, Chief Scientific Officer and Co-Founder
Brian Groch – Chief Commercial Officer
Jean-François Boily – VP, Finance
Conference Call Participants
Leland Gershell – Oppenheimer & Co.
Mayank Mamtani – B. Riley FBR, Inc.
Andre Uddin – Mackie Research Capital Corporation
Nathan Weinstein – Aegis Capital
Good day, ladies and gentlemen, and welcome to Acasti Pharma Fiscal Year 2020 Business Update Conference Call. [Operator Instructions] At this time, it is my pleasure to turn the floor over to your host Alexandra Botie, Investor Relations. Ma’am, the floor is yours.
Thank you. Good afternoon, everyone, and welcome to Acasti Pharma’s fourth quarter and year-end fiscal 2020 conference call. On the call with us today are, Jan D’ Alvise, President and CEO; Pierre Lemieux, Chief Operating Officer, Chief Scientific Officer and Co-Founder; Brian Groch, Chief Commercial Officer; and Jean-François Boily, Vice President, Finance. If you have questions after the call or would like any additional information about the company, please contact Crescendo Communications at 212-671-1020.
I’d also like to remind everyone that statements on this conference call that are not statements of historical or current fact constitute forward-looking information within the meaning of Canadian Securities Laws, forward-looking statements with — and within forward-looking statements within the meaning of U.S. federal security laws. Such forward-looking statements involve known and unknown risks and uncertainties and other unknown factors that could cause the actual results of Acasti to be materially different from historical results or from any future results expressed or implied by such forward-looking statements.
In addition to statements, which explicitly describes such risks and uncertainties, readers are urged to consider statements labeled with the terms believes, belief, expects, intends, anticipates, potential, should, may, will, plans, continue or other similar expressions to be uncertain and forward-looking. Listeners are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date of this conference call.
Forward-looking statements in this conference call include, but are not limited to, information statements about Acasti’s strategy, future operations, prospects and the plans of management, Acasti’s ability to conduct all required clinical and nonclinical trials for CaPre, including the timing and results of those trials, CaPre’s potential to become the best-in-class cardiovascular drug for treating severe hypertriglyceridemia, Acasti’s ability to commercially launch CaPre and to fund its continued operations, the timing and outcome of the unblinding of TRILOGY 2, Acasti’s ability to report top line results for TRILOGY 2 within the contemplated timing as well as Acasti’s ability to report key secondary and exploratory endpoints from both TRILOGY studies within the contemplated timing, and Acasti’s ability to file an NDA based on the TRILOGY studies.
The forward-looking statements contained in this conference call are expressly qualified in their entirety by this cautionary statement, the “Special Note Regarding forward-looking statements” section contained in Acasti’s latest annual report on Form 10-K and most recent management discussion and analysis, which will be filed — which will be available on SEDAR at www.sedar.com, on EDGAR at www.sec.gov, and on the Investors section of Acasti’s website at www.acastipharma.com.
All forward-looking statements in this conference call are made as of the date of this conference call. Acasti does not undertake to update any such forward-looking statements whether as a result of new information, future events or otherwise, except as required by law. The forward-looking statements contained herein are also subject generally to assumptions and risks and uncertainties that are described from time to time in Acasti’s public securities filings with the Securities and Exchange Commission and the Canadian Securities Commission, including Acasti’s latest annual report on Form 10-K and most recent MD&A.
I’d now like to turn the call over to Jan D’Alvise. Please go ahead, Jan.
Thank you very much, Allie, and I want to welcome everyone participating on the call today. Our primary focus for this call will be to share the results of our post-hoc data analysis and audit findings for TRILOGY 1, as well as our plans going forward to unblind TRILOGY 2. Now that we’ve received feedback from the FDA, it’s really a pleasure to finally be able to share a TRILOGY 1 findings and data with you. Jean-François will also briefly review our Q4 and fiscal year end results.
Now as reported in our press release this morning, we have identified a phenomenon in the TRILOGY 1 data that we refer to as pre-randomization triglyceride normalization. Specifically, this phenomenon occurred during the qualification period of the study. That is between the last screening Visit or Visit 1 and the pre-randomization VISIT or VISIT 4, which was just prior to patients starting on drug or placebo.
So while our initial top line results from TRILOGY 1 were disappointing due to the unusually large and unprecedented placebo effect, we now have a much better understanding of this pre-randomization triglyceride normalizations phenomenon that we believe contributed significantly to the TRILOGY 1 outcome. We appreciated your patience as we worked through our way through this investigation over the last several months. And we’re pleased to finally be able to share our findings with you now.
As previously disclosed, at the end of April, we provided all of the background information on TRILOGY 1 and the accompanying data to the FDA, and we recently received their response. As you may have seen in our June 19 press release, and as stated in the current statistical analysis plan for TRILOGY 1, the FDA confirmed that they will still require that pivotal or primary end point efficacy analysis for TRILOGY 2 be performed on the full intent to treat population that is inclusive of every patient who was randomized, regardless of whether or not he or she actually complied and took the drug or placebo or whether they actually even completed the trial.
It’s also important to note that the FDA supported our conduct of additional post-hoc analysis on the TRILOGY 1 data for exploratory purposes. We’re carefully considering the FDA’s comments on the TRILOGY 1 data, and we will continue to conduct further post-hoc analysis based on their feedback. We are also now working to finalize the statistical analysis plan for TRILOGY 2, which we plan to submit to the FDA by the end of July.
And as noted in our public disclosures, we continue to re remain blinded to the TRILOGY 2 data and believe that we will be ready to unblind and report top line data by end of August. We also expect to report key secondary and exploratory end points from both TRILOGY 1 and TRILOGY 2, sometime after the unblinding of the TRILOGY 2 top line results.
Now based on our analysis and the review of all of the data thus far, we remain optimistic that we may still have a viable path forward to file an NDA if TRILOGY 2 successfully meets its primary endpoint, and we can get a significant P value after combining the results from the intent to treat populations for both of our trials.
With that said, I’d like to now share a bit more information about the interesting journey that we’ve been on since reporting our top line results back in January. And I’d like to provide you with a brief summary and explanation of our TRILOGY 1 findings.
Now when we reported our top line TRILOGY 1 results, we indicated that we had achieved a 30.5% median reduction in triglyceride levels among all patients receiving CaPre at 12 weeks, and a 42.2% median reduction in triglyceride levels and a subset of patients receiving CaPre while on background statin therapy at — also at 12 weeks.
Additionally, we reported a 36.7% median reduction in triglyceride levels among all patients receiving CaPre 26 weeks. Despite the positive results in the CaPre arm due to an unusually large placebo response that resulted in median reductions in triglyceride levels of 27.5% and 28% at 12 and 26 weeks, respectively, TRILOGY 1 did not reach statistical significance.
And as we stated previously, the observed reductions in triglyceride levels in the placebo group were far greater than anything seen in previous prescription omega-3 trials in hypertriglyceridemia. In fact, they were at least 2x to 3x greater than what would have been expected. Not only were we caught by surprise, but our expert clinical advisors, key investigators and other partners in the trial were all equally confounded by these seemingly inexplicable results.
So the obvious question was why? Why were placebo results so strong in TRILOGY 1, and what could have caused this outcome? We immediately ruled out any misallocation of treatment or a mix up of CaPre and placebo capsules, as well as any implication that the placebo that was actually used in the TRILOGY studies, which is simple cornstarch could have done the cost. Cornstarch is generally regarded as safe and is a commonly used placebo in the pharmaceutical industry. It’s well known to be an inert and inactive excipient with low nutritional value.
Now the protocol for TRILOGY 1 and 2 had input from and was approved by the FDA, and followed essentially the same standard design as had been used by all other companies who had previously run trials in severe hypertriglyceridemia. The eligibility criteria and the design of the screening and qualification phases of the study were written to ensure comparability with these prior trials and to minimize difference.
Having said that, there actually were a few slight differences in the TRILOGY 1 patient population, as compared to previous studies. For example, we had a slightly higher percentage of diabetics and patients on statins, and also as required by the FDA we included patients who were stabilized on fibrates. But all of our post-hoc analyses confirmed that these differences did not contribute to the significant placebo effect.
We thoroughly investigated the effect of various other categorical demographic and baseline characteristics, such as alcohol use, race [ph], diabetes, concomitant use of diabetic medications and or statins, we looked at all of these things and the effect on triglyceride normalization in an attempt to further identify or eliminate possible contributors to this phenomenon.
Specifically, various factors were compared between treatment arms, such as washout or discontinuation of lipid lowering medications at screening, use of lipid lowering medications at randomization and subsequent change in these medications during the study, use of anti-diabetic medications at randomization and any subsequent changes in those meds during the study. Bottom line, we did not find any imbalance in any case between the treatment groups that was likely to explain the unusually large placebo effects seen in TRILOGY 1.
So that left us with two possible hypotheses for further investigation. For something else related to how the patients were screened and qualified for the study or how the trial was conducted may have contributed to this unusually high placebo response; or second, and this outcome — or the second possibility was that this outcome was simply due to chance or what is referred to as regression to the mean. However, because of the magnitude of the placebo response, we felt that chance was unlikely to be at least the sole contributor to this phenomenon.
Routine monitoring activities were conducted throughout the TRILOGY 1 trial by our CRO and by us to ensure adherence to the protocol and to detect potential protocol violations. However, as previously reported, as we began to analyze the data, we identified some unexpected and inconsistent findings that we suspected may have contributed to this unusual placebo effect.
Given these observations, we then focused our efforts on a comprehensive and rigorous post-hoc review of the data and in parallel the conduct of thorough independent audits of patient data and records at the five clinical sites who had been among the highest enrollers, and who had also experienced the largest placebo responses.
Some of the best experts in the field were intrigued by our data and graciously offered their help, including of course, our principal investigator, Dr. Dariush Mozaffarian. These experts provided us with their thoughts, ideas, and guidance as our investigations moved forward, and as we ultimately compiled all of the data into our briefing package for the FDA.
And as you hopefully saw in our press release today, the post-hoc analyses of the TRILOGY 1 data revealed a rapid significant and sustained reduction in triglyceride levels at the two VISITS during the patient qualification period, which took place between the last screening VISIT or VISIT 1 and the time of the patient’s randomization at VISIT 4, which we also refer to as week zero.
Very importantly, a significant percentage of the patients had triglyceride levels that went up dramatically during this qualification period, and then drop significantly prior to being put on drug or placebo. We’re referring to this phenomenon as pre-randomization triglyceride normalization. And Acasti believes that it may have contributed to an artifactual overestimation of triglyceride reduction and consequently a significant underestimation of the post randomization treatment effective CaPre.
It’s also very important to note that this pre-randomization triglyceride normalization effect was much greater in the placebo group as compared to the CaPre treatment group, further compromising the ability of the study to detect a clinically significant drug treatment effect. As a result, our post-hoc analyses indicated that this phenomenon was a major contributor to the unusually large placebo effect that we saw in TRILOGY 1.
Now we’ve not found a single root cause that explains this normalization phenomenon, we have several theories as to what may have caused this, but unfortunately there’s no way to corroborate our hypothesis based on the data and the audit findings. The most likely possibility is that the effective patients in the trial may not have been adequately fasted during the qualification Visits. In spite of those patients claiming to the clinical site administrators that they had been fully fasted. This could have artificially elevated their triglyceride levels during the qualification phase, which would have increased the average of the three triglyceride values used to calculate their baseline value and thereby allowing them to be accepted into the trial.
It’s important to keep in mind that triglycerides can be highly variable and significantly affected by fat in the diet. This is why it’s important that a patient is fully fasted when they get their blood drawn for triglyceride analysis. For example, we know that following a meal triglycerides will dramatically increase up 2x to 2.5x.
They tend to peak after four hours and then go back to baseline only after more than nine hours of fasting. For this reason, we believe factors such as misreporting or underreporting of the patient’s fasted condition may have contributed to this normalization effect. Again, we cannot prove this at this time, but it still remains our primary working hypothesis.
Now based on the National Cholesterol Education program working group on lipoprotein measurement, fasting triglyceride levels of healthy individuals can typically vary by about 10% or so within a 1 to 2 week time period. And that variation tends to be random going in either direction up and down. However, this variation can be a bit greater in certain physiological and disease States. And we know, for example, that subjects with very high triglycerides as was the case in TRILOGY 1 maybe somewhat more susceptible to experiencing wider week-to-week fluctuations.
However, the extent and magnitude of the triglyceride variations seen in TRILOGY 1 was much larger than the expected 10%. And what tipped us off that’s something else was going on was that the variation only went in one direction. It only went down and the reduction was quite significant. The overall pre-randomization reduction in triglycerides across all subjects in TRILOGY 1 was 20%.
And very importantly, a full one quarter of all subjects experienced a triglyceride reduction equal to or greater than 38%. And remember, this is before they were randomized on drug or placebo. Finally, the median triglyceride normalization reached 30% or more in 12 of the highest enrolling sites. Therefore we believe that this pre-randomization triglyceride normalization effect was not likely due only to the typical individual day-to-day triglyceride variation.
And again, while we’re hypothesizing that additional factors such as misreporting or underreporting of the patient’s fasted state may have contributed to this normalization phenomenon in TRILOGY 1, the audits of the affected patient’s information data we’re not able to confirm this. And I’d like to add again, that while fasted glucose levels could have potentially provided some additional information regarding patient compliance during the qualification phase of the study.
It was not required by the protocol for the reason that as I mentioned earlier, it’s use is unreliable in metabolically impaired subjects such as those with type 2 diabetes or those with metabolic syndrome, as they tend to have much higher levels of glucose as compared to normal individuals. So keep in mind that about 50% of the patients in TRILOGY 1 had diabetes.
Now I want to mention that the other major finding that came out of our post-hoc analyses was that about 40% of all randomized and eligible subjects had triglyceride levels at randomization, again, defined as Visit 4 or week zero, and before they started on drug or placebo that were outside of the protocol specified average qualification threshold for patients with severe hypertriglyceridemia. And again those limits are greater than 500 milligrams per deciliter and less than 1.500 milligrams per deciliter.
So it had become quite clear to us that the extent and magnitude of the pre-randomization triglyceride normalization had a substantial impact on the ability of TRILOGY 1 to accurately determine the therapeutic effect of CaPre on the primary endpoint in true patients with severe hypertriglyceridemia.
Now, let me take a moment to further explain the actual impact of this normalization effect on our results. As per our statistical analysis plan for TRILOGY 1 and I should add consistent with previous studies conducted in patients with severe hypertriglyceridemia, we used an average of three triglyceride values for the calculation of the baseline corresponding to time points during the two qualification measurements at Week minus 1 and Week minus 2. And again at Week 0, which is the Visit just prior to randomization.
So once we recognize the significant effect of this artefactual pre-randomization reduction of triglycerides had on this study outcome, we conducted a new post-hoc analyses using a revised single point triglyceride baseline value collected just at Week 0, just prior to when the patient was randomized and started on drug or placebo. We now refer to this as Week 0 — this Week 0 triglyceride level as the revised baseline.
Only those subjects meeting the protocol specified triglyceride threshold of having triglyceride levels greater than 500, but less than 1,500 milligrams per deciliter at Week 0 were included in this post-hoc analyses. This revised approach corrected for a significant amount of the artefactual pre-randomization triglyceride reduction that was seen in the subjects that were most affected by the large placebo effect.
After these patients were removed, a total of 42 out of the 69 subjects remained in the placebo group and 101 out of the 173 subjects remained in the CaPre group. And were included in the post-hoc analyses representing 61% and 58% of all randomized subjects, respectively.
Not surprisingly the significantly smaller sample size resulted in reduced power to detect a treatment difference of 20% as specified in the statistical analysis plan and as compared to the plan sample sizes in the original analysis plan. Nevertheless, nevertheless, the results do suggest clinical relevance, even if statistical significance was not demonstrated.
Furthermore, based on the post-hoc trends implied by these summarized results, it is plausible that a larger number of patients such as those in TRILOGY 2 could enable us to achieve statistical significance. Upon removing subjects with triglyceride levels below 500 and above 1,500 milligrams per deciliter at Week 0, subjects receiving CaPre showed a 28.1% median reduction in triglycerides compared to a 15.4% median reduction among subjects receiving placebo for an unadjusted difference of 12.7% after 12 weeks and a P value of 0.29.
Compared to the original analysis, the revised baseline reduced the placebo response by approximately 12 percentage points from a minus 27.5% to a minus 15.4%. This actually represented a 44% reduction in the placebo response. While the response in the CaPre arm remained mostly unaffected and was reduced only slightly from a minus 30% versus a minus 28.1%. So that was only about a 6% difference.
After 26 weeks of double-blind treatment, the efficacy of CaPre showed good persistency of effect with a 32.6% median reduction of triglycerides compared with a 14.6% median reduction in the placebo group, reaching an unadjusted difference of a minus 18%, which trended towards statistical significance with a P value of 0.08. Compared to the original analysis, the revised baseline reduced the placebo response at 26 weeks by approximately 13 percentage points from minus 28% to a minus 14.6%. So roughly a 46% reduction. While the response in the CaPre arm remained mostly unaffected, reduced from a minus 36.7% to a minus 32.6% or roughly a 10% reduction.
An important sub — an important subgroup of patients are those with triglyceride levels above 750 milligrams per deciliter. We know from the previous hypertriglyceridemia studies that the higher the baseline triglyceride value, typically the greater the reduction that is seen when these patients are treated with the therapeutic omega-3. Interestingly, that was not the case with the original top line analysis of TRILOGY 1 reported back in January.
We were troubled by the fact that this subgroup of sicker patients actually seem to show a smaller reduction in the CaPre group than those who had baseline triglyceride levels much lower at between 500 and 750 milligrams per deciliter. When we conducted the subgroup analysis on these patients with triglyceride levels above 750 using the revised baseline at Visit 4, Week 0, it corrected this discrepancy from the original top line analysis. These patients represented 41% of the subjects retained in the post-hoc analyses.
And within this group, the median triglyceride reduction in the subjects receiving CaPre increased as would be expected reaching a minus 36.3% at week 12 and a minus 43% at week 26. In comparison, the median triglyceride reduction for the placebo group was a minus 11.8% at week 12 and a minus 14.4% at week 26, resulting in unadjusted differences of a minus 24.5% and a minus 28.6%, respectively in favor of CaPre with P values of 0.22, and 0.15, respectively.
So you can see that the post-hoc analyses of TRILOGY 1 data using a revised single-point baseline showed a meaningful trend towards correcting for the unexpectedly large placebo response observed in the original analysis and allowed for a clearer understanding of the impact on the triglyceride primary endpoint and the potential therapeutic effect of CaPre over the full 26 weeks of treatment.
Overall, these results based on the revised baseline indicate a trend towards improved efficacy of CaPre to reduce triglycerides as compared to placebo. However, as discussed, the medium difference in triglyceride levels between CaPre and placebo from the TRILOGY 1 post-hoc analyses still fell short of reaching statistical significance at week 12, which is the primary endpoint due to the reduced number of patients and the resulting reduced statistical power of the analysis.
For this reason and based on the results of our Phase 2 data, we’re moving quickly now to unblind the TRILOGY 2 results. We eagerly anticipate the outcome of those top line results in the important secondary and exploratory end points that will follow shortly thereafter.
While the triglyceride reduction achieved by CaPre based on the revised baseline analysis compares favorably to that of other omega-3 therapies most recently studied in pivotal Phase 3 trials in severe hypertriglyceridemia. These data will remain as post-hoc analyses, and for now at least the FDA considers them to be exploratory. Therefore we did not believe that we will be able to use the study purely on the basis of these post-hoc analyses. However, we do believe that they can be used to show a trend towards a positive treatment effect in favor of CaPre.
Furthermore, as TRILOGY 2 can achieve a positive P value and it’s the pooled integrated efficacy results with TRILOGY 1, using the intent to treat population are also favorable, we plan to discuss with the FDA, whether they would accept this data in support of an NDA. We would plan to discuss this with them at our pre-NDA meeting later this year,
Now consistent with the company’s product disclosures, it’s important to keep in mind that the FDA may still require us to conduct another confirmatory Phase 3 trial in severe hypertriglyceridemia patients to support the filing of an NDA, in which case it’s possible that the design, the sample size and the duration of such a third trial could be different from the previous two TRILOGY studies.
Finally, I’m sure you’re wondering what the odds are that TRILOGY 2 will be affected by the same pre-randomization triglyceride normalization effect that we saw in TRILOGY 1. It’s honestly hard for us to say, as we remain blinded to the TRILOGY 2 data, and therefore we have no idea of the same phenomenon also occurred. Having said that, however, there’s a chance that we could see some degree of this same placebo effect in TRILOGY 2.
However, we believe that if we do see it, it could potentially be to a lesser degree and here’s why. We know that the contribution from the five highest enrolling sites is much lower in TRILOGY 2 as they contributed only 12% of the patients in TRILOGY 2 versus 36% of the patients in TRILOGY 1. And we had more sites randomizing fewer subjects in TRILOGY 2, as compared to TRILOGY 1. We also saw more uniform enrollment across all of the sites in TRILOGY 2.
In other words, we had fewer of these high enrolling sites. So the odds are that any given site would have less influence on the overall outcome of the TRILOGY 2 trial. Also in the event that localized practices may have contributed in some way to the pre-randomization triglyceride normalization phenomenon, then we would expect this effect to be much more diluted in TRILOGY 2 as well.
Furthermore, while the study show the same design they were conducted in different geographies, although a few of the same sites in the U.S did randomize subjects in both studies. Specifically, as I mentioned before, only 12 sites that randomized subjects in TRILOGY 1 also randomized subjects in TRILOGY 2. And keep in mind, there was a total of 71 sites in TRILOGY 2. These 12 sites accounted for 62% of all randomized subjects in TRILOGY 1, but only 19% of the patients in TRILOGY 2.
And again, while all of the five highest enrolling sites in TRILOGY 1 also enrolled patients in TRILOGY 2, they contributed to only about 12% of the total patients. So for these reasons, we believe the two studies can be regarded as being completely independent from each other.
So with that, I’d like to now turn it over to Jean-François Boily, our VP of Finance to present our Q4 and year-end results. Jean-François?
Yes. Thank you, Jan. Before I begin, I’d like to point out that our consolidated financial statements have been prepared in accordance to us GAAP as opposed to IFRS in the past — as we have reported in the past. We are also now reporting our results in U.S dollars versus Canadian dollar.
So turning to our results for the quarter, R&D expenses before depreciation, amortization and stock-based compensation expenses were $13.2 million for the year ended March 31, 2020 compared to $26.9 million for the year ended March 31, 2019. The $13.6 million decrease was mainly attributable to a $14.4 million decrease in the research contracts, partially offset by an increase in salary and benefits due to increased headcounts and related benefits. This lower research contract expense is primarily attributed to the Phase 3 clinical trial program getting closer to completion.
Our general and admin — administrative expenses, again before stock-based compensation expenses were $4.6 million for the year ended March 31, 2020, an increase of $1.3 million from $3.3 million for the year ended March 31, 2019. This increase was mainly attributable to $450,000 increase associated with the company’s insurance policy, as well as an increase of $830,000 in corporate accounting and legal fees associated with the implementation of our new ERP system and the conversion of our financial reporting from IFRSs to US GAAP.
Our loss from operation, again for the year ended March 31, 2020 was $24.4 million compared to a loss from operation of $34.4 million for the year ended March 31, 2019. The decrease was due mainly to a reduction in research contract expenses as the Phase 3 clinical program for CaPre is getting closer to completion.
Our net loss for the year ended March 31, 2020 was $25.5 million or $0.30 per share compared to a net loss of $39.4 million or $0.73 per share for the year ended March 31, 2019. The decrease in net loss is primarily due to the reduction of the research development expenses, again as a Phase 3 clinical program for CaPre was getting closer to completion, lower net financial expenses, and to the change in fair value of the warrant derivative liability.
Now turning to cash and cash equivalent and our short-term investments, these totaled $14.2 million as of March 31, 2020, compared to $25.8 million for the year ended March 31, 2019. We believe that existing cash will fully fund the company’s operation into the first quarter of 2021, the first calendar quarter of 2021.
Now, in addition to our annual filing to be filed this afternoon, we will also file a shelf registration statement with an expended at the market facility. As many of you are aware, we have had an ATM facility in place for some time and has been very careful and prudent in our use of the ATM. Not that we have to become a U.S filer, we were required to convert our existing F-3 and file a new F-3 with a corresponding ATM in order to keep this program in place.
Keep in mind that when we convert our S3, it will start a new 3-year clock so that is a reason that we will take the opportunity to increase the size of the facility to $200 million. Generally speaking, ATMs provide an attractive and flexible alternative and lower cost of capital compared to traditional financings, which often come with big discounts and warrants.
I’d like now to turn it back to you, Jan. Thank you.
Okay. Thanks so much Jean-François. So wrap — to wrap up our prepared remarks, we made steady progress throughout fiscal 2020. Although we were disappointed by the outcome of TRILOGY 1, we now have a better understanding of what contributed to the unprecedented placebo effect, and we remain cautiously optimistic about the outlook for TRILOGY 2.
We’re now carefully considering the FDA’s comments on TRILOGY 1 and we’ll consider or continue to conduct further post-hoc analyses based on their feedback. We’re working to finalize the statistical analysis plan for TRILOGY 2, which we plan to submit to the FDA by the end of July and we currently expect to report top line data by the end of August, 2020.
We also plan to report the key secondary and exploratory end points from both TRILOGY 1 and TRILOGY 2 trials, as soon as possible after the unblinding of TRILOGY 2 results. One final note, we’re not slowing down and we remain focused on the end goal, which is delivering value for our shareholders. In addition to completing our Phase 3 trials in fiscal 2020, we also began important pre-commercialization activities, including advanced planning for the scale up of our manufacturing, preparation of our NDA, as well as advancing partnering and other market development objectives.
We’re also pleased to have been awarded additional composition of matter and method of use patents in Canada, United States, Mexico, China, Hong Kong, Chile, and Israel since the start of fiscal 2020 alone. These are very important and valuable patents and add to the growing portfolio of intellectual property for CaPre. Overall, we continue to believe in CaPre and despite the recent bumps in the road, we’re committed to doing everything in our power to seeing this through and achieving a successful outcome.
With that, operator, we can now open the call to any questions.
Thank you. [Operator Instructions] We’ll take our first question today from Leland Gershell with Oppenheimer. Please go ahead, sir.
Hi. Good afternoon, Jan. Thank you for the very comprehensive update on what sounds like a very thorough analysis. A few questions for me. Just to clarify, so would you be able to analyze TRILOGY 2 using a Week 0 baseline single point or because of the FDA’s commentary about using the prior SAP, would you not be able to analyze TRILOGY 2 according to the single point baseline, therefore have to use the three measurements between screening and randomization?
Yes. So what the FDA clarified is that we could not eliminate patients. And so, we have to do the analysis with the full intent to treat population. We do have some flexibility around the analysis, however, we likely will not be able to use a single point endpoint. However, having said that, I think the FDA has allowed us to use the single point baseline as the further — in-further post-hoc analyses.
Okay. And would you be able to — when you analyzed post-hoc the TRILOGY 1 data to get the patients at Week 0 who fit between the 500 and the 1,500, obviously a 40% reduction in the number, would you be able to tell us what the numeric mean baseline at Week 0 was for the placebo and drug groups and what they were at the end of the study? Thanks.
I don’t have that information in my head, but Pierre, do you by chance have that information?
Well, there was a — I mean the baseline value, as you can imagine, losing 5 — almost 40% of the population when we decided to remove — to use on the one baseline. So obviously the baseline value went from — it was around 670 milligram per deciliter just went down to 550. So there was a major decrease of the baseline overall. So — which is not helping by the way to show a stronger efficacy. And for the rest of all the weeks of the week, 12 and 26, I cannot tell you right now, I don’t have the numbers in front of me to be precise. But there was a significant baseline by removing these patients that were on the — using the baseline at Week 0.
Okay. Thanks. And should we expect when you do report the TRILOGY 2 data, even though you are using the ITT analysis that you will also potentially if indicated report the Week 0 only baseline analysis similar to what you did with TRILOGY 1?
Pierre, you want to take that?
Yes, definitely. I mean, we’re going to be planning to publish these information, of course, because there’s a story to be told and both analysis should be carried in parallel. So for regulatory purposes, of course, the primary end point was to use the three values as a baseline. But the fact that we’ve conducted post-hoc analyses, it makes a lot of sense for us to tell the world that behaves and what happened and how this placebo effect is impacting the results. Because at the end of the day, the question remains, does CaPre work or not. And so, I think there’s a very strong sentiment that when you get rid of this placebo effect, and when you eliminate some of the patients that were below 500 before starting the treatment that the product is telling us something, that there’s a — a treatment effect is taking place over time. So that’s encouraging. And we need to say that and we need to report this. So with [indiscernible] to BI is the — we’ve agreed that we will publish these information at some point. Now we still need to go through the — some of the, the post-hoc analyses. We don’t have the complete clinical study report as of yet. It’s going to come from our CRO, but when we have all those two sets of data, we will. We will certainly share that information, which is critical and key for us.
Okay. Thank you. And then my last question relates to kind of the overlap between what you had discussed previously on previous calls about the five sites that were the higher enrollers, but also may have had irregularities in how they had handled patients in the trial and the kind of unprecedented randomization correction that you saw between the screening and the randomization. Is it fair to say that there was a fairly tight association between those sites and the degree of this correction over those three Visits? Just want to be clear on that versus the other sites in TRILOGY 1, and perhaps not having as much of an effect observed in these — in this correction of the — sorry, at the end of the trial. Thanks.
Pierre, you want to take that?
The normalization is across all the sites, but I’d just say those five sites were selected for an investigation as you know because they were high enrolling. So for us, it was easy to refocus our attention in those five sites. So definitely those five sites really demonstrated a higher normalization effect, placebo effect. But overall, when you — the thing for us was to say, okay, if we get rid of those five sites, for instance, what would be the outcome, and it’s not translating, if you will, into something useful or it’s not — it’s really not helping in the outcome of the treatment effect of CaPre. So it’s — and that’s because we’ve realized, I think the five sites allowed us really to dig into this story. But when you start analyzing across the board all the sites, you realize that a lot of patient, all of them in average are losing at least 18%, 20% of their triglycerides. So that’s — so, overall it would have been an very quick and easy way to fix this, but it did not lead to that outcome, unfortunately.
But for TRILOGY 2, if I may, Leland on TRILOGY 2, so Jan said it in the — and what she reported to you just a minute ago, TRILOGY 2 has less of those sites, less patients from the — coming from those sites. So, we should expect a much less placebo effect in TRILOGY 2. So it’s hard for us to say still, but I think we’re — I think we’re in a much better place with TRILOGY 2. And again, keep in mind that this is signs when we spoke with our experts sometimes, we hear words like well, you’ve been unlucky, this is by chance. I know it doesn’t sound too scientific, but I guess it’s part of the — we’re running Phase 3 trials. It’s part of the picture, unfortunately.
Okay, thanks. I’ll come back in the queue.
We will take our next question from Mayank Mamtani with B. Riley. Please go ahead.
Thanks for taking my question. I appreciate the detailed update here. Definitely been a fascinating 6 months going through this. Just quickly taking a step back, is there, like, what can the feel like take away from this, like has there been another study, there’s something like this has happened or how do you think this informs as some of these severe hypertriglyceridemia trials run. Just curious, your thoughts on that. And then I have specific questions.
Yes. Well, it’s hard for us to pin down the — what the actual cause was. Again, we believe that, perhaps many of these patients were not fully fasted as I stated earlier. I think, we’ve discussed this exhaustively in-house. I mean, are there ways that in the future, in future protocols, you can somehow ensure that the patients were fasted. And you’re relying to a great extent on them, telling you the truth whether or not they’ve eaten in the last nine hours, right? Because that can most definitely throw off the triglyceride results. And we — as I mentioned earlier, you can look at doing glucose — looking at glucose levels, but in these patients that have diabetes and metabolic syndrome, it’s not always indicative. I mean, I think it’s one measure we probably would add in the future. I think also — I’ll comment on this and let Pierre jump in, but I think to the extent we could have stuck with more, I would say routine clinical sites that are actually treating the patients, these high enroller sites were more of these clinical trial factories, if you will. And they take people in off the street and they don’t have a history with those patients. So there’s not that connection with the patient. On the flip side, it’s not easy to find these patients with severe hypertriglyceridemia. So we do believe if we could have ensured that we had true severe hypertriglyceridemia patients in the trial, it might’ve reduced this effect to some degree. Pierre, I don’t know if you want to add anything.
Well, thank you. You’ve said it all, Jan. Yes, I mean, the difference is between our product goals and the — our competitors or the previous protocols, I mean, there are very small differences. And so, to explain this — this difference in placebo that this was unprecedented for sure. And maybe over time, there’s also either better management of those patients or they’re really in demand. So again, we can prove it, but we could have thought that some of the sites were now geared to redefine those patients and tweak in a way some of the ways to make them eligible. We can prove it. But maybe over time there’s a gain of intelligence from those sites in order to find those patients. So — but again, it’s not something that we can prove and just pure speculation from our part. It might be a mixture, a mix of all these things too, lack of fasting. The fact that, like Jan said, I think this is — we would have been more, I would say comfortable knowing that all the patients had a medical history long-term severe hypertriglyceridemia, but it’s not the case and it was not part of the protocol as well. So, in terms of differences in the future, we might impose and put into protocol in writing because it was not, and none of the other protocols as well. It did not impose on the site, a long-term medical I would say a dossier on the patient themselves, but this is something that can be probably added. At the same time in the future. I mean, in the same time, it would have had a lot of time to these — to those trials. So that’s also to keep in mind. So the more you’re imposing restrictions and exclusion criteria, it’s already difficult to find these patients [indiscernible] put a long list of the same criteria, it becomes — it would have taken us maybe 3, 4 years before we could have completed. So, overall, if I may, I think we don’t think the protocol was wrong or bad or miss, I think it was according to what was done in the past by others and even improved in some cases. So the fact that we have fibrates in our trial as well was a plus. But overall, this is what I can add on what Jan have covered.
Okay. Thanks for that. And I understand it’s a cash [indiscernible]. So two quick specific questions. Can you maybe comment on the error bars that you may have seen with the post-hoc versus ITT? Just trying to understand like the super responder phenomena, if there is. And then the background fish oil supplement use, I know you said lipid medication in general, but just curious if there was any background fisher supplement news in pre-randomization or afterwards that you noted could be something worth highlighting. Just curious about that.
Yes. Well, I take the second question, to be honest. The first one, I hadn’t quite understood, but to answer your question, of course, supplements, fish oil, or other concomitant medications even lifestyle, lifestyle change as well was looked at to determine whether or not this — these elements could have disturbed the trial and we didn’t find anything. So honestly there was no unbalance or anything to be reported here that would have changed the outcome of the trials. So it’s — it was looked at. Good question. But we can — so we ruled it out. The first question, I’m not sure I understood what you were asking.
Yes. The error bars of when you look — when you do your [indiscernible] just kind of the super responders or under responders like just curious if that was different in ITT versus post-hoc analyses?
No. I mean — yes, exactly. So that’s a good question. So in our protocol, we had planned for all the statistical power is based on potential standard deviation, for instance. In our protocol it was set at 40%. And again, it’s based on what we’ve seen in the — in previous trials. And when we did the original analysis versus the, let’s say, the post-hoc analyses, there was not much difference in variability. So the standard deviation was not that different. You could have expected because you’re only using one value that you have a lot of variation, using only one value. And surprisingly, we were in the same ballpark comparing the original analysis versus the post-hoc analyses. So …
Okay, great. And just final one, if you could comment on the FDA request that you had for a [indiscernible] like what specific outcome you were looking for and as you think about going forward from TRILOGY 2, just kind of highlight what is the desired outcome from TRILOGY 2 that will help you get to that request that you had for the FDA in type C?
You know, I’m not sure I caught the first part of that Mayank, sorry. You said — could you repeat it?
Yes, sure. So as you obviously highlighted when you have the FDA meeting request, I’ve seen some request, there was a number of items you wanted to discuss with them live …
… and there was a desired outcome from that, and it seems like they responded to you with a written response. So I’m just curious, what was your desire like your blue sky [ph] scenario there in that meeting and how do you think TRILOGY 2 could help you get there to that desired outcome?
Okay. Yes. So if I understand — and keep in mind that during COVID, the FDA is not conducting meetings. So this was handled we had to submit questions in writing and all of our background materials in writing. So we sent them an extensive package and a list of questions and they reviewed the package and provided some answers back. So it’s really based on that as we said that we will continue to do some additional actual post-hoc analyses based on the FDA’s response, and then proceed with modifying the — and finalizing the TRILOGY 2 statistical analysis plan, which will include the ability to do some of these post-hoc analyses now that we understand what happened in TRILOGY 1 and submit that to the FDA, and then we’ll proceed to unblind. And once we unblind TRILOGY 2, the plan would then be to pull the data from both trials to see if we could get to a favorable P value. Now this would be combining the ITT populations from both trials. And so if — we believe this, if we can get a strong positive P value with TRILOGY 2, and if the pooled efficacy results with TRILOGY 1 using the ITT population from TRILOGY 1 are also favorable, that we would then plan to discuss with the FDA whether they would allow us to use that data to proceed to file an NDA. And again, there’s precedence for that. And I think we believe that the post-hoc analyses has definitely shown a strong trend in favor of CaPre. That’s what we were hoping to understand what happened, how do we explain this placebo effect. And when you correct for that placebo effect, do you have resulting data that supports the effectiveness of CaPre. We believe we have that now. So the next step is to unblind TRILOGY 2, pull the data and see if we can get to a positive P value. And if we do, then I think it’s a matter of sitting down with the FDA in a pre-NDA meeting, which would be held sometime later this year, and to go through all the data, discuss it with them and get their guidance on whether or not we can proceed to file an NDA. I don’t know if that answered your question, but that would be the plan.
Yes, that’s very helpful. Thank you, Jan. I appreciate it and look forward to more updates.
Thanks. Thanks, Mayank.
We’ll take our next question from Andre Uddin with Mackie Research Capital. Please go ahead.
Hi, Jan. Hi, Pierre. Just a …
Just a couple of questions. In terms of the TRILOGY post-hoc analyses, how many patients would have needed — would you have needed to show statistical significance?
I don’t know that. Pierre, do you have that number in your head?
I don’t actually. Yes, I don’t. We can get back to you Andre on this.
Okay. Sure. And just in terms, I know you sort of answered this before, but in a future based on that, you don’t quite understand what happened in the pre-randomization in a future trial design would you be able to prevent the same issue if you ran another Phase 3 trial?
Yes, so again, as I think, Leland kind of touched on this as well. I think — and I guess Mayank did that we probably would want to do fasted blood glucose levels, even though they’re not as meaningful in that population. I think we would want to get those levels just to see what they tell us. And see if there’s, significant change any surprises there, throughout the qualification period. And I don’t know, Pierre, if you want to comment. I mean, I know you guys have been kicking around some additional things that in terms of monitoring that you might do.
Yes. I mean as I said to Mayank as well, so the medical history becomes also very important. I think it gives you more confidence about the eligibility of the patient as well. So that’s one thing. There’s also this — potential compliance issue that might still — again, we [indiscernible] could prove that either, but that too would have been easy for us to rule out the sites or patients to say, they didn’t take the drug. So — and we have the data, but I think we could do more to monitor, if you will, levels of EPA DHA, we’ve done it. But maybe we could do more. And again, it really — previous trials in other fields and lipid management, for instance sometimes, some of the trials have failed because the patients were not taking their drugs. So in our case, it doesn’t seem to be the issue, but I wouldn’t take any chance in the future. So that’s something we could probably improve. Not that it was an issue here, but I think it could be useful, yes.
Okay. That’s great. Thanks. Thanks, guys.
We’ll take our next question from Nathan Weinstein with Aegis Capital. Please go ahead.
Hi, guys. Thanks for taking my question. And it’s clear you’ve been literally working on going through TRILOGY 1 and thanks for sharing your findings with us. And so my first question relates to the fact that the FDA is going to require the analysis to be performed on the [indiscernible]. I just was curious whether that was a surprise compared to your expectations going in?
No, not at all. We honestly expected that. I think we’re encouraged by the fact that they allowed us to do the — and will continue to allow us to do these additional post-hoc analyses. The other thing I should mention is that we, again, we don’t have the answer yet, Pierre and the statisticians are working on this now, but we’re taking a look at modeling what kind of results we would need to get in TRILOGY 2 to be able to get an overall favorable P value in a pooled analysis. So that’s an important analysis that’s underway right now. So I don’t know if that answers your question, but …
Yes, thank you. And so obviously I’ll be watching to see what TRILOGY 2 looks like, but to sort of take a little weight off of a binary situation if we think about a scenario where you do kind of have to run another Phase 3 trial, prospectively speaking, would that be separate in terms of the analysis and data [indiscernible] that went into it separate from the first two TRILOGY trials?
Pierre, do you want to answer? I think it would be considered part of the TRILOGY program. So it would be in effect a TRILOGY 3, if you will. The question is — and Pierre can jump in here, but the question is how big that study would need to be, how long the study would need to take, because we already just in TRILOGY 2 alone, have more patients than for example [indiscernible] had with their Marine [ph] trial. Remember they just had one trial to get approval in severe hypertriglyceridemia. So, the question is how big will that trial need to be? And could it go for 12 weeks instead of 26 weeks? Pierre, I don’t know if you want to add anything.
Yes. I mean, it’s part of the story. So if we have TRILOGY 3, it will be used as — all the trials, even our Phase 2 trials will be part of the story. So you cannot just take a trial and hide it, it’s out there and if they’re going to be looking at this. And so, I think, and — on the efficacy as well. So not only safety, but — not only figures about safety as well, it will be combined. So definitely, but like Jan said, I think it’s an important point. We have the biggest program and if we have TRILOGY 3, it’s going to be even bigger than the others. And those numbers now might be very, very useful for us. So and this is what Jan was alluding earlier. The fact that we have this is the biggest program if TRILOGY 2 turns out to be positive, there’s ways to, if you will, mitigate the impact of TRILOGY 1. So this is the magic of statistical analysis, but we’ll know more in the future about this, but it’s [indiscernible] that the more patients are better. And it allows really to mitigate some of these fluke or bad luck that we’ve seen. So I would imagine that the TRILOGY 2 is have to be done. It will be, you’ll need to confirm some of the doubts the FDA will have. So — and that’s why those post-hoc analyses become important because you’re — you can show [indiscernible] to the FDA that if you — if you’re really limiting your population to this that you have in a very interesting outcome, and this is how you can define the future design of phase of the TRILOGY 3, if need to be. So, yes — so overall, everything will be taken together for sure.
Thank you. And if I could just ask one more question. After you guys reported all the secondary and exploratory end points from the two TRILOGY 1 and TRILOGY 2, and if you just sort of a confirmation of the [indiscernible] effect, and maybe do you think you could build in something to look more at a diabetic population in a future trial? Thank you.
Yes, I’m not sure I caught what the question was, Nathan, but we absolutely are looking at all of that in the secondary and exploratory endpoints. None of that has been unblinded yet. So we’ll look at all of that after we unblind TRILOGY 2. Remember, it was really important to make sure we were clear on how the FDA wanted us to handle the TRILOGY 1 data before we fully did the analysis on all the secondary and exploratory endpoints. So, of course, that’s really, really important because we believe that CaPre has some unique advantages. And we would hope to see what we saw in Phase 2, that we have at least a neutral to favorable effects on LDL, neutral to positive effect increasing HDL. And then it’s going to be very, very interesting the fact that we had 50% of our patients were diabetics in TRILOGY 1, we’ve got the population there to really look at what’s going on with their hemoglobin A1C. And we don’t know what the mix is yet in TRILOGY 2, but we should have enough data there to really look at those key lipid end points and be able to show hopefully some advantage with CaPre.
Thanks and I appreciate the color.
Thank you, ladies and gentlemen. This concludes our question-and-answer session. We’ll turn the floor back over to our speakers.
Okay. Well, I don’t have much more to add. I want to thank everyone for listening in today and we absolutely look forward to continuing to provide you with additional updates as soon as we have them and as soon as possible. So with that, I’d just say, everyone take care. Stay safe and we’ll speak to you all soon. Bye-bye.
Ladies and gentlemen, this does conclude today’s teleconference. We thank you for your participation. You may disconnect your lines at this time and have a great day.