NASH – an introduction
Fatty deposits and mild fibrosis in non-alcoholic fatty liver disease. Source
NASH – short for non-alcoholic steatohepatitis – has been a magnet for biotech investors over the past few years. NASH designates the most severe form of non-alcoholic fatty liver disease (‘NAFLD’). In addition to showing accumulation of fat in the liver, NASH also comes with additional complications such as fibrosis and portal hypertension. Up until now, NASH is primarily diagnosed via biopsy. NAFLD and NASH are strongly associated with ‘metabolic syndrome’; i.e. insulin resistance, obesity, sub-optimal nutrition and lack of exercise. Unfortunately, NAFLD and NASH are on the rise and they are on track to account for the majority of liver transplants in the U.S. by 2020.
From a health policy perspective, such a lifestyle-related epidemic calls for all sorts of interventions aimed at improving people’s nutrition and lifestyle. However, this is easier said than done as lifestyle is such a deeply personal issue and government intervention would have to be quite sweeping to result in meaningful change, which in turn would be opposed by libertarian concerns.
In the absence of such sweeping lifestyle adaptation, there’s a big pie (a big foie gras?) to be shared by the pharmaceutical industry, and given the chronic nature of this affliction, the possibility of a statins 2.0 scenario is alluring: easy-to-manufacture, small molecule drugs with fat margins that will create a steady revenue stream.
NASH is a crowded & highly competitive space
As a result, investor interest in potential NASH therapies has been significant for a number of years, but one needs to contrast enthusiasm for a potentially lucrative, large indication with the competition that exists and the probability of success (‘POS’) for a given drug & company.
Players in the space include industry staples Pfizer (PFE), Novartis (NVS) and Gilead (GILD), but a lot has been made on SeekingAlpha and on social media of the smaller, independent players such as Conatus (CNAT) or Viking Therapeutics (VKTX). In fact, the space continues to become more and more crowded with small players vying for a future share in this budding market.
Of the above, Viking shareholders have benefited the most from the excitement – some may say hype – around NASH as a significant emerging market for small molecule drugs in the U.S., with VKTX shares up from $1 in 2017 to $11 currently:
This meteoric rise is largely premised on significant expectations around VK2809 given the drug’s structural similarity to Madrigal’s (MDGL) MGL-3196 and the very promising mid-stage results of the latter in NASH. Excitement has been stoked by reputable analysts, as this tweet illustrates:
What I will take issue with is that VK2809 is “uniquely positioned for treating NASH” as William Blair appear to claim.
A bit of history: Viking is a small biotech shop headed by a research scientist-turned biotech analyst who joined Viking from SunTrust. The company’s current drug candidates are in-licensed from biotech powerhouse Ligand (LGND):
We have exclusive worldwide rights to a portfolio of five drug candidates in clinical trials or preclinical studies, which are based on small molecules licensed from Ligand Pharmaceuticals Incorporated, or Ligand.
Source: Viking SEC filings
There is nothing fundamentally wrong with developing in-licensed compounds, and Ligand have a reputation of excellence with regards to the molecules they provide or help develop. However, investors need to realize that VKTX don’t have any meaningful internal R&D capabilities and the value proposition of this company is premised on the performance of their in-licensed assets in clinical trials and their future value relative to competing drugs. In fact, Madrigal have also in-licensed MGL-3196, from Roche (OTCQX:RHHBY), but unlike Madrigal, Viking have as of yet been unable to complete the required GLP toxicology studies to advance VK2809 into a NASH trial. We will revert to this important point further down.
Understanding the competitive environment is paramount for VKTX shareholders. First, let’s examine the comparison between VK2809 and MGL-3196 and see if the enthusiasm holds up to the data presented by both companies to date.
VK-2809’s tiny dataset in NAFLD pales in comparison to the MGL-3196 data in NASH
In May 2018, Madrigal presented PhII data on MGL-3196 in NASH confirmed by liver biopsies at baseline and end of study. 125 patients were enrolled in this study. At 36 weeks, 27% of MGL-3196-treated patients showed resolution of NASH, 50% of which also showed resolution of liver fibrosis with statistically significant reductions in the other 50% vs placebo. These long-term data, cemented by liver biopsies, are what you want to see from a drug that is intended to treat NASH in a real-world setting, since resolution of NASH is a straightforward, FDA-approvable endpoint.
What have VKTX presented us with thus far? The company announced 12-week data from a 45-patient trial in NAFLD – not NASH – that showed median liver fat reduction of 57-60% of VK2809-treated patients (n=31). These values were higher than those achieved by higher-dose MGL-3196, -42%, in 44 patients. Observers were also enthused by VK2809 showing a higher % of “super-responders” whose liver fat reduction exceeds 30%:
The caveat with all this, besides the smaller number of patients treated in the Viking trial, is of course that Madrigal were testing their drug in patients with biopsy-confirmed NASH whereas Viking were going after NAFLD patients. NAFLD is a more common, less severe condition, as explained by the NIH. The higher reduction in liver fat observed in the Viking trial in NAFLD than what was observed in the Madrigal trial in NASH cannot be extrapolated to mean that VK-2809 will display superior performance over MGl-3196 in actual NASH patients.
Viking’s very own PR sounded a cautionary tone with regards to the applicability of these NAFLD data to potential outcomes in NASH:
Investors, however, either didn’t pay attention or didn’t care that this very limited dataset in NAFLD still needs to be duplicated in NASH, as the stock jumped 100% following the small NAFLD readout:
Chart generated by the author.
While NAFLD is a very common affliction that is clearly linked to lifestyle, NASH represents the real unmet medical need, as medical intervention becomes urgent before progression to liver cirrhosis and possibly liver cancer:
Source: Madrigal corporate presentation
Given the more advanced & complex pathology of NASH as compared to NAFLD, it stands to reason that resolution of NASH is less straightforward than resolution of NAFLD.
Madrigal are getting ready to begin PhIII testing of MGL-3196 whereas Viking, if they in fact want to pursue NASH, are almost certainly going to be required to run an additional PhII trial – with liver biopsies – in actual NASH patients. At the company’s most recent conference call, in response to an analyst question, Viking’s CEO Brian Lian explained that long-term toxicology studies in rodents and non-human primates should be concluded by the end of 2018 / early 2019 respectively, which appears to be a prerequisite before moving into a PhIIb trial in NASH.
Madrigal completed long-term toxicology studies ahead of its PhII trial in NASH. Source
Madrigal’s trial took about 20 months from initiation to top-line data. If we are willing to give Viking the benefit of the doubt and assert that:
- Long-term GLP tox studies would be completed by mid-2019 and will raise no concerns around VK-2809
- Following favorable conclusion of GLP tox studies, Viking would be able to file an IND in biopsy-confirmed NASH for VK-2809 in mid-2019 / H2 2019
- Viking would be able to initiate this PhIIb trial in NASH following IND clearance in 2H 2019
- Viking would be able to enroll patients as efficiently as Madrigal did throughout 2016 & 2017 despite today’s much more crowded clinical trial environment in NASH,
We still could not expect top-line data from a 36-week PhIIb NASH trial with liver biopsies before H2 2021 / H2 2022. In other words, Madrigal have at least a 2 ½, possibly a 3-year lead over Viking in NASH.
VK-2809’s positioning in the crowded NASH space is even worse than the MDGL first-mover advantage suggests.
In a token 2-drug market, one would expect the first-mover to capture 60% and the second-mover to capture 40% of total addressable market (‘TAM’) over time (Source: Pharmagellan data).
However, NASH is not a 2-drug market, and neither is it a monolithic indication. A plethora of drug candidates are currently being tested, of which >20 in mid- and late-stage trials, for NASH and associated conditions:
Note that VK-2809 is misleadingly listed as being in ‘PhII’ in the above chart, as that only applies to its development status in NAFLD but not in NASH (where an IND has yet to be filed).
As you can tell, many of these investigational therapies for NASH and associated conditions are being advanced by Big Pharma and Big Biotech. And some notable investigational therapies associated with small-cap biotech in fact also benefit from Big Pharma backing.
Case in point: Emricasan, an inhibitor of enzymes involved in cell death processes with a convoluted development history. Invented in 1998 by Idun Pharma, the molecule was acquired by Pfizer (PFE) in 2005, before Conatus (CNAT) acquired it from Pfizer and proceeded, in 2017, to out-license / co-develop the drug with Novartis (NVS). Testing of emricasan is now well underway in 3 PhII trials which together will enroll nearly 800 patients. Unlike Madrigal and Viking, Conatus & Novartis are going after liver fibrosis and portal hypertension more specifically; both of which are conditions associated with NASH. As a fellow SA contributor has explained, emricasan is targeting patients with the most advanced form of NASH whose portal hypertension is refractory (treatment resistant) to standard-of-care non-selective beta-blockers. As a last-line therapy, emricasan is expected to command premium pricing.
Besides buying into emricasan, Novartis have been busy building additional alliances to tackle different aspects of NASH. In somewhat of a surprise move, the Swiss pharma giant recently announced a collaboration with Pfizer, aimed at evaluating a combination therapy approach to NASH while acknowledging that NASH is a complex disease:
“Novartis has a leading development portfolio in non-viral liver diseases and I believe especially in our combination therapies. Liver diseases, including NASH, are multifaceted with various factors that contribute to the progression of the disease. This makes them difficult to treat with a single compound,” said Eric Hughes, Global Development Unit Head, Immunology, Hepatology and Dermatology. “We want to collaborate with multiple partners to drive the science and understanding of how to treat non-viral liver diseases. Targeting different pathways in NASH with a broad array of therapies is an essential strategy to bring the best treatments to patients.
In a similar move, Novartis also announced a collaboration with Allergan (AGN) in 2017, in order to evaluate a combination approach targeting at once chemokine receptors to reduce inflammation and fibrosis and the bile acid receptor.
And it isn’t just Novartis and Pfizer who are stepping up their R&D and partnership game. Earlier this year, Regeneron (REGN) and Alnylam (ALNY) announced a collaboration to emulate a genetic variant which appears to be associated with lowered risk of NASH by using RNAi therapy. How this plays out remains to be seen, but this line of inquiry seems promising based on the fact that RNAi therapeutics are particularly well-suited for afflictions of the liver due to the PK/PD of their vectors.
Gilead (GILD) are running a number of compounds in NASH & alcoholic hepatitis. The most advanced being Selonsertib, for which a 800-patient, long-term PhIII trial was initiated in early 2017. A PhII trial showed encouraging, dose-dependent activity and enrolment of the 800-patient PhIII trial was, remarkably, completed a short year following the trial’s initiation:
“Gilead is focused on addressing the greatest unmet need in NASH, which is in patients with advanced fibrosis. Reflective of this unmet need, the STELLAR-3 and STELLAR-4 studies of selonsertib in patients with F3 and F4 fibrosis have completed enrollment ahead of schedule. We expect data from these Phase 3 studies in the first half of 2019,” said Norbert Bischofberger, PhD, Executive Vice President of Research and Development and Chief Scientific Officer at Gilead. “We are now exploring combination therapy approaches with compounds with distinct and potentially complementary mechanisms of action. The initial data presented today are important advances toward our goal of improving outcomes for patients with advanced fibrosis due to NASH.”
At best, small-cap NASH investors might take heart in knowing that GILD and other big players are keen on combination therapy approaches but given their own late-stage assets and partnerships between themselves, these large players appear more likely to in-license assets via partnerships as was the case with regards to NVS / CNAT than outright acquire a company like VKTX.
If we presume that the endgame for any small- and mid-cap NASH player is an acquisition – which it should be, seeing how the commercial effort to address such a disease ‘in the community’ is daunting – then these Big Pharma & Big Biotech alliances don’t bode well for investors in these smaller companies, and the odds seem especially daunting when investing in VKTX with its unusually high Mcap in the absence of data in NASH.
If one of the pharma majors decided they absolutely need a thyroid receptor molecule in their repertoire, possibly as an add-on to their existing programs, it appears more likely they would partner with, or acquire, Madrigal.
A partnership with Viking to explore VK-2809 in NASH on terms favorable to VKTX shareholders appears altogether implausible in the absence of an IND & data in NASH.
VKTX is significantly overvalued and share price is likely to decline as the company faces a data desert and uncertainty around GLP tox for NASH
As I’ve laid out, VK-2809:
- Has only generated 11-week data in NAFLD
- Has not completed GLP tox studies that would enable an IND in NASH
- Is not expected to produce PhIIb data in NASH before 2H 2021 even in a best-case scenario
- Does not only suffer from MGL-3196’s first-mover advantage but also from various differentiated, competing drug candidates in mid- and late-stage development, many of which are backed by large and reputable players
- The recent Big Pharma alliance between NVS & PFE, and additional collaborations, do not bode well for M&A prospects
VKTX’s fair value is, IMO, best derived via comparative valuation with MDGL.
If we presume that whatever TAM – in this case a subset of NASH patients – MGL-3196 & VK-2809 will eventually capture in NASH would be split 60/40 between the two drugs and if we presume that MDGL’s valuation reflects MGL-3196’s risk-adjusted NPV with regards to 60% of said TAM, VKTX’s Mcap should reflect VK-2809’s rNPV of the remaining 40%.
The simplest way to proceed here is to establish an appropriate multiple between MDGL and VKTX on the basis of the POS of their respective drugs, followed by a NPV discounting of VKTX vs MDGL.
With a successful PhII in NASH completed for MGL-3196, the drug’s probability of success (‘POS’) for that indication can be estimated to clock in around 50%.
- The absence of a full safety dataset that is relevant to NASH: GLP tox studies have not been completed
- The difficulty of extrapolating POS in NASH from a small, 11-week dataset in NAFLD
… make it hard to establish a POS ‘scientifically’, but it seems only logical to attribute a POS corresponding to PhI stage development to VK-2809. Such a POS would be in the single digit range, and if we generously accept the limited NAFLD dataset as supportive of higher POS, we may settle for 15% POS in NASH for VK-2809.
|MGL-3196 POS||VK-2809 POS||Multiple|
MDGL currently trades around $2.2bn Mcap, which yields:
- $2.2bn / 3.3 = $666m Mcap for VKTX.
Next, we need to further discount VKTX’ Mcap on the basis of the 2.5 – 3 year delay in the development of VK-2809 over MGL-3196. Discounting $666 for 2 1/2 years at a discount rate of 20%, we obtain:
- $666m / 1.2 / 1.2 / 1.1 = $420m Mcap for VKTX
Discounting for a 3 year delay at 20% we obtain:
- $666m / 1.2 / 1.2 / 1.2 = $385m Mcap for VKTX
The Mcap of VKTX currently clocks in around $800m, suggesting that the stock is overvalued by approximately 50%.
Viking enthusiasts will point to another of the company’s in-licensed molecules, VK-5211, for which the company reported a small dataset in hip fractures, as a saving grace. The study evaluated increase in lean body mass and VK-5211, as was expected due to its MOA, delivered.
There is only one problem – there is no agreement with FDA on an approvable endpoint yet and it seems all but certain that ‘increase in lean body mass’ will not be acceptable, per the most recent CC:
One contributor on SeekingAlpha has been so euphoric as to suggest that agreement on a particular endpoint or set of endpoints could “make an instant blockbuster out of” VK-5211. That is to completely ignore clinical development and commercialization challenges, and as laid out by the company in their most recent earnings call (cf. above), they would be more than happy to leave the bulk of that work, including regulatory discussions, up to a potential partner.
Seeing how the company appear to be unable to manage these essential aspects of regulatory work & clinical development themselves, I would not hold my breath for a lucrative partnership agreement, either. If a partnership for VK-5211 materializes, I would expect that a deal is structured such that the vast majority of potential future revenues would be absorbed by Viking’s partner.
As a result, I believe that VK-5211’s intrinsic value is hard to assess and can be safely ignored for the purposes of establishing the fair value of VKTX.
Viking is materially overvalued, no meaningful data will be generated in either NASH or hip fractures for years to come, and it seems unlikely that any potential partnership would be structured in a way that provides investors with a meaningful payoff.
According to interactivebrokers.com, approximately 750k shares of VKTX are currently available to borrow.
Risks to my thesis include, but are not limited to:
- a meaningful partnership, i.e. one that lets VKTX shareholders participate materially in future revenues, for either of the company’s molecules
- advancement of VK-2809 straight into PhIII in NASH (highly unlikely)
- an acquisition of the company
Members of Second-Level Investing, a service dedicated to in-depth research & analysis of the biotech sector, provides members with deep discussion of sectorial dynamics, expert interviews, close follow-up on the stocks we monitor, and an actively managed model portfolio.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
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.
Editor’s Note: This article covers one or more microcap stocks. Please be aware of the risks associated with these stocks.