Description

Summary

VTL has previously been written up by us in 2015.  However, there is now another chance to short the stock as their current Phase 3 trial is due to read out in late September.  Their key technology, an extracorporeal artificial liver called the ELAD system, is a 20 year old technology that they acquired in bankruptcy.  The ELAD System has careened from indication to indication - from FHF (Flare of Viral Hepatitis) to ACLF (Acute-on-Chronic Liver Failure) to the current sAH (Severe Alcoholic Hepatitis) - as they look for weak signals of efficacy in subsets of failed trials to raise capital to run the next next trial.  The current Phase 3 trial (VTL-308) of ELAD vs. standard of care treatment in alcoholic hepatitis is based on a retrospective analysis of the failed 2015 trial. They have carefully selected the ~30% of the population that showed a benefit of the treatment to enroll in their new Phase 3 study. With limited rationale for the treatment and for the selected population and given the high risk in retrospective analysis, we believe the current Phase 3 trial is very likely to fail as well.  As the current Phase 3 trial is the only value driver for the company and current cash expected to run out in 1Q19, we believe the stock has 75%+ downside in the event of trial failure.

Capital Structure

The company ended the 2Q 2018 with $31.1M of cash in the bank.  With a fully diluted share count of 50.9M they had $0.61/share of cash at the end of June.  

They do currently have a shelf outstanding and could execute an "at-the-market" sales agreement they have with Cantor Fitzgerald (another sign of a quality company) to raise $60M.  As of July 31st they have not exercise any of the "at-the-market" offering, but it is quite possible that they have done so since then.

We currently have an effective shelf registration statement on Form S-3 on file with the SEC, which expires in June 2021. The shelf registration statement currently permits the offering, issuance and sale by us of up to an aggregate offering price of $200.0 million of common stock, preferred stock, warrants, debt securities or units in one or more offerings and in any combination, of which $60.0 million may be offered, issued and sold under an “at-the-market” sales agreement with Cantor Fitzgerald & Co. In the event of positive topline results from VTL-308, we plan to raise additional capital.

VTL-308 Analysis

The ELAD system runs blood through cartridges of C3A cells in attempt to replicate certain liver functions

VTL has an artificial liver called the ELAD system that they are currently testing in a Phase 3 trial (the VTL-308 trial).  The previous write-up has more details on the system, but underlying idea is that you run people’s blood thru a system where it has the chance to pass thru 4 cartridges of C3A cells, which may help mimic certain liver functions.  

VTL-308 trial is likely to fail as it is based on a very elaborate and tortured retrospective subgroup analysis of the failed VTI-208 Phase 3 study

The VTL-308 Phase 3 trial is a 151 patient randomized, open-label trial testing ELAD vs. standard of care in patients with severe alcoholic hepatitis..  The primary endpoint of the study is overall survival. Enrollment began in Q2 2016 and the study finished enrollment in March of this year. The trial protocol mandates that all patients be followed for at least 91 days and there be at least 55 deaths before the overall survival is analyzed.  In early 2018 independent statisticians confirmed that the blinded event rate was sufficient with the targeted 150 patient enrollment. Average follow up period for the patients should be approximately 1 year given public disclosures of 67 patients enrolled by their 1Q17 call and 95 patients by the 2Q17 call.  

In March 2018 VTL completed enrollment in their Phase 3 clinical trial, VTL-308, with 151 severe alcoholic hepatitis patients randomized to receive ELAD or control.  This trial is based on a retrospective analysis of the failed VTI-208 trial.

The VTI-208 trial which failed was itself a retrospective analysis of a failed Phase 2 trial (VTI-206) - highlighting the inherent risk in this data mining and subgroup analysis.  The retrospective analysis are particularly dangerous when the overall trial is negative - as was the case with the VTI-206 study and the VTI-208 study which is the basis for the VTL-308 study.  

The results of the VTI-208 trial were completely negative.  The hazard ratio (HR) for survival was 1.027 (meaning that you were slightly more likely to die on the ELAD arm than on the control arm).  

Because the company has nothing else to exist for (nothing else in the pipeline and no real underlying technology), they decided to examine the failed VTI-208 trial to look for subset of patients they could run another trial in and raise some new money for.  Below you will see how they arrived at the inclusion criteria for the VTL-308 study. They decided to continue to examine a number of subsets and combine them all together in order to find a group of patients that appeared to experience benefit from ELAD. The inclusion criteria are interesting not just because of the amount of effort that they invested to find a subgroup that worked, but also because MELD score is already meant to incorporate a number of these factors, but they add on additional criteria to the factors to “hack” their hazard ratio to make it as low as possible.  MELD score is used to help determine treatment and mortality risk in patients.

Below is an excerpt on the calculation of the MELD score https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4509754/

The MELD score is calculated using bilirubin, creatinine and international normalized ratio (INR) levels. The objective parameters were originally used to predict early death following elective transjugular intrahepatic portosystemic shunts (TIPS) [20]. The use of objective and reproducible data was subsequently shown to be a reliable measure of short-term mortality risk in patients with end-stage liver disease independent of complications of portal hypertension, and is used as a disease severity index to determine organ allocation priorities [21]. Several studies have used the MELD score to assess disease severity in AH. In a study by Sheth et al [18], the MELD score had a similar performance as the DF in predicting mortality at 30 days. The sensitivity and specificity in predicting 30-day mortality was 86 % and 82 %, respectively, for MELD scores >11 compared to 86 % and 48 %, respectively, when DF was greater than 32. The utility of predicting mortality using MELD score represented as area under the curve (AUC) was 0.82 (95 % confidence intervals (CI): 0.65–0.98), and AUC of DF was 0.86 (95 % CI: 0.70–1.00). Sheth et al. therefore suggested that treatment for AH should be considered when MELD score is greater than 11.

However, all this effort data mining has now created a carefully selected subset of patients where they are able to show a very “impressive” HR of 0.283 (suggesting an 81% reduction in death from ELAD).  However, it is important to note that this subset of patients is only 30% of the patients of the original trial. And this means for the rest of the population it appears that ELAD would increase the risk of death.  If we believe that there is a subset of patients that benefit from the ELAD system, we must then also accept that there is a population that is harmed. The company’s argument is that sicker patients do not benefit from ELAD and may be harmed by the ELAD system.  The company’s explanation is that the patients they are excluding are the ones with worse kidney function and blood clotting - which ELAD either does not address and can exacerbate, respectively. The argument then would be that the system works to improve survival in patients who are less severe, but in more advanced patients, there is minimal benefit and it is completely outweighed by treatment harm thus leading to increased risk of death.  They are really trying to thread the needle here. This is exacerbated by the benefit they are attributing to the drug in the younger, healthier subset (see below).

The projected benefit of ELAD in the younger, healthier subset, if true, would make it among the most effective agents ever tested.  

The HR of 0.283 that VTL is suggested for ELAD in this subset of patients imply that it would be among the most amazing treatments ever studied.  An analysis of all the agents brought before the FDA Oncology Drug Advisory Committee from 2001 to 2012 found that of the 19 applications that reported HR (on either overall survival or progression free survival) the median HR was 0.67 with a range of 0.35-0.99.  The idea that VTL has a stronger treatment effect than every oncology drug brought before committee over 12 years is difficult to believe. This is not to say that it is impossible, just to highlight how unlikely it is that this effect we have seen in this retrospective analysis is true.  

It is important to also note that the trial is powered assuming the 0.30 HR is true.  If the ELAD system has a more reasonable effect - say 0.7 (similar to what you might expect from an effective cancer agent) then this trial will not show statistical significance.  Said another way, this trial will only work if the ELAD system shows among the most impressive treatment effects in a life threatening condition. This makes sense in the context of VTL's history and situation.  The company has limited resources, therefore they needed to "show" investors a high impressive effect in order to be able to raise money. Given the amount of money raised, and the difficulty in recruiting patients in this indication, they could not afford to power the trial for a more reasonable effective size.  The company has cash into early 2019 it seems likely that if they had decided to materially increase the trial size to allow for a smaller effect size they would have not had enough cash to get to trial readout.

Very difficult indication to show benefit in alcoholic hepatitis

There are no effective therapies for alcoholic hepatitis.  Mortality is very high in this disease and it is very hard to show a survival benefit.  As the editorial (Morgan, Liver Transplantation 2018) accompanying the VTI-208 study noted:

Although several formulas are used to assess the prognosis of AH, the absence of an unequivocally effective medical therapy limits their utility.

In fact, the STOPAH trial, a large study evaluating the two agents recommended in the treatment guidelines found no benefit for prednisone or pentoxifylline treatment for 28 days.  Though prednisone had a non-statistically significant effect at 28 days, that disappeared by 90 days and 1 year follow-up.

With likely close to a year follow-up on average for the VTL-308 study, even if ELAD were to show a short term benefit in survival like prednisone, there is no guarantee that is able to persist - remember ELAD treatment is only for 5 days.  

Increased treatment with ELAD did not correlate with improved survival

If increased treatment does not correlate with survival that is another red flag that ELAD treatment may not be very active.  From the VTI-208 publication:

Of the 82 subjects completing the minimum treatment of 72 hours for per protocol analysis, 29 (35.4%) subjects died.  Mortality differed minimally between subjects receiving between 72 and 120 hours (12/37) and completing 120 hours (17/45; 32.4% versus 37.8%).

Management and Board of Directors

Former Receptos CEO Faheem Hasnain was appointed as Chairman of the Board of Directors in September of 2017.  He was already on the Board, having joined in August 2016.

Russell Cox joined as the new CEO on January 3rd, 2018.  He was also named to the Board of Directors. He was previously the COO at JAZZ Pharmaceuticals.  The rationale here appears to be that they wanted someone with commercial experience if the VTL-308 trial succeeded.  Though you would expect that Russell Cox would have done his diligence on the trial before joining, there have been other examples of high profile CEO’s joining a company where the main value driver drug failed in pivotal trials - see David Hung with Axovant.

Satter Muneer a Director (and former Co-Chairman of the Board) purchased 4.25M shares in 2017.  Muneer and his entities are the largest shareholders in the company with 27.2%. The vast majority of the purchase 3.75M was a participation in a secondary offering done by VTL in March 2017.  

Muneer has certain rights regarding the directors he has the right to nominate if he holds certain share thresholds.  

The Senior Preferred IRA also provides that, for so long as certain stockholders affiliated with Muneer A. Satter, our Co-Chairman and Lead Director, referred to as the Satter Investors, hold at least 30% of our outstanding common stock, the Satter Investors have the right to nominate 40% of our directors (rounded up to the nearest whole number). If the Satter Investors hold less than 30% (but at least 20%) of our outstanding common stock, they have the right to nominate 30% of our directors (rounded up to the nearest whole number). If the Satter Investors hold less than 20% (but at least 10%) of our outstanding common stock, they have the right to nominate 20% of our directors (rounded up to the nearest whole number). If the Satter Investors hold less than 10% (but at least 2%) of our outstanding common stock, they have the right to nominate 10% of our directors (rounded up to the nearest whole number). For so long as the Satter Investors hold less than 2% of our outstanding common stock, they do not have the contractual right to nominate any representatives to our board of directors. To date the Satter Investors have not exercised their rights to nominate any directors, but they have reserved the right to do so in the future.

Muneer has a history of participating in the IPO, the secondaries, as well as buying on the open market.  He made purchases on the open market of common stock in 2014 following the IPO and before data readout in Aug 2015 for the previous Phase 3 study.  

Catalyst

The VTL-308 Phase 3 trial will read out in the second half of September.