Subscribe And Get My Articles Emailed Directly To You - It's Free!!

Sunday, March 10, 2013

Valuing Amarantus - Where Scientific Hope Meets Financial Reality

By Jason Napodano, CFA

What would you pay, right now, for a $10 billion drug in 2027?

In February 2013, I wrote an article outlining the significant preclinical data amassed over the past decade on Mesencephalic Astrocyte-Derived Neurotrophic Factor (MANF), a potential disease-modifying agent for Parkinson's disease and ischemic heart conditions at Amarantus Bioscience (AMBS). The article, entitled, "Scientific Literature Suggests Amarantus' MANF Has Big Potential" pulled conclusions from twenty scientific publications, including from prestigious journals such as Neuroscience, Science, Comparative Neurology, and Circulation.

Work done to date has been mostly conducted by universities, medical schools, and research institution like the Mayo Clinic. Amarantus has presented data showing superiority to Glial cell-Derived Neurotrophic Factor (GDNF) in a neurorestoration 6-hydroxydopamine (6-OHDA) rat model of Parkinson's disease. The company has also published data showing animal proof-of-concept in ischemia / reperfusion following myocardial infarction. 

But enough about the science! The entire article above is nothing but science. This article is about financials, and how I went about coming up with a fair value for Amarantus. It was a daunting task. Criticize away; but I hope that the models I've provided below give investors, at the very least, a good starting point to conduct their own analysis.

Let's Look At The Pipeline

This article is about Amarantus Bioscience, and more specifically the value for preclinical asset, MANF, in all its scientific glory. In February 2013, the company announced it plans to form a subsidiary and license the diagnostic intellectual property, which includes the LymPro Alzheimer's disease blood test and the NuroPro diagnostic platform and Parkinson's disease blood test. The new business will be called Amarantus Diagnostics. In time, we expect Amarantus Bioscience to spin-off Amarantus Diagnostics to shareholders; but that is the basis for yet another article.

For the purpose of this article, I will focus on four potential areas for which we suspect Amarantus will seek development with MANF. The lead indication is Parkinson's disease, where the preclinical data suggests MANF could have a disease modifying effect. Preclinical data is also suggestive of protection and restoration in animal models of cardiac ischemia. I start with these two indications because: 1) both are enormous opportunities, and 2) both utilize a localized delivery of the drug. Other indications such as traumatic brain injury (TBI) or stroke, while equally as large, would require systemic delivery of MANF. Thus we will push those indications to the back-burner until long-term toxicology data becomes available.


...Parkinson's Disease...

The Parkinson’s Disease Foundation estimates around one million, or around 30 in 10,000 (0.3%), Americans have Parkinson's disease (PD). Outside the U.S., another six to eight million have PD. The mechanism of action for MANF seems to limit utility to patients where viable dopaminergic neurons still remain (Petrova et al., 2004). I estimate that is around 65% of the active Parkinson’s patients in the U.S.

From a timing standpoint, I am expecting Amarantus to file an investigational new drug application (IND) in late 2014, with phase 1 single and multiple ascending dose studies to being in 2015 and offer data throughout the year. This should allow the company to sign a development partnership for the commencement of phase 2 proof-of-concept studies in 2016. If these data show the same type of encouraging signs that researchers have been able to demonstrate in Drosophila (Palgi et al, 2008) and rodent models (Dauer, 2007, PR-Oct.25,2012), I believe significant interest in Amarantus and MANF will materialize from both investors and potential licensing partners. Amarantus and its partner could be in position to commercialize MANF for PD in 2021.

I believe a drug like MANF could capture sizable market share within these patients. If MANF was on the market right now, it could conceivably be used in 100% of the eligible PD population. After all, no FDA approved disease modifying agents exist today for PD. But the competitive landscape in 2021 for PD therapies is nearly impossible to predict. The biggest risk to market share assumptions would be competing agents, including CDNF or a reformulated / improved delivery of GDNF, or a completely different class of molecule hidden somewhere deep in the pipeline of a big pharma company. As a result, I model only 15% market share of the eligible population at peak for MANF.

I believe MANF could be effectively priced at launch at $20,000 per course of treatment, with aggressive price increases to follow. This sounds expensive, but research shows that the annual direct and indirect cost to care for a PD patient is around $20,000 per year. A disease modifying agent like MANF that keeps patients out of nursing homes and long-term care facilities should be able to command such a price. I suspect the market opportunity outside the U.S. is, at a minimum, equal to that of the U.S, albeit pricing and reimbursement outside the U.S is always more difficult to predict. 

No doubt, Amarantus’ goal is to partner MANF for development and commercialization following the completion of phase 1 dosing and safety studies in 2016. Given the significant risk to such an early-stage collaboration, I believe Amarantus will struggle to pull-in a sizable upfront payment. Instead, the agreement will probably be heavily back-end loaded.

I looked several deals between small biotech companies and larger pharmaceutical companies to get a sense of how a deal for Amarantus' MANF could look. Below is a list of some deals I used as a roadmap for MANF, the players, the upfront payments, the total deal size, and the indications licensed.


For the sake of argument, the company can probably secure a deal all-in over $1 billion, with a milestone and royalty schedule as such:

 - Upfront = $25 million
 - Phase 2 Data = $50 million
 - Initiate Phase 3 = $100 million
 - File NDA = $100 million
 - U.S. FDA Approval = $200 million
 - Sales Milestones = $1+ billion
 - Royalties On Sales = “mid-teens”

I believe that Amarantus will require roughly $10 million in cash to get to a level where a partnership is obtainable; this includes finishing all the IND-enabling work, GMP manufacturing and GLP toxicology, and completing phase 1 single and multiple ascending dose studies alone.

Below is a representation of what the cash flows could look like to Amarantus for such a transaction, including modeling out sales of MANF to 2027 (peak), a breakdown of the milestone structure, corporate overhead, and appropriately discounting the cash flows at a rate of 20% over the next thirteen years (similar to the company's cost to secure capital in November 2012 and January 2013 through the offering of a convertible promissory notes to Dominion Capital).

According to data from PhMRA and Innovation.org, the odds of moving a drug from preclinical screening to FDA approval are 5,000:1. The odds of moving a drug from phase 1 to FDA approval are around 20:1. Amarantus is past the screening stage with MANF. The company has extensive preclinical data from animal studies to validate the molecule and targets. I suspect that the U.S. IND will be filed in late 2014. Thus, the odds of success for MANF lie between 0.02% (pre-clinical library screening) and 5% (odds for a phase 1 asset). For the purpose of my model, I am using a success rate of 2% to commercialization. After all, MANF is a naturally occurring biologic. This molecule was not built in a lab and screen from 5,000 formulations. MANF is not a shot in the dark. I also assume a 25% success rate to the 2015 upfront payment of $25 million.


All this financial rigmarole yields a value of $14 million today. That's what I think MANF is worth right now for Parkinson's disease. It's a $14 million call-option on a potential mega-blockbuster opportunity.

...Cardiac Ischemia...

Sitting right behind the opportunity in Parkinson's disease is an even larger opportunity in cardiac ischemia. Our review of the science behind MANF discovered several papers predicting utility in cardioprotection and reperfusion following an acute cardiovascular event, such as myocardial infarction (Tadimall et al, 2008; Glembotski CC, 2010; PR-Sept.26,2012).

The U.S. Centers for Disease Control and Prevention (CDC) estimates some 27 million Americans are living with ischemic heart disease. In any given year, approximately 20% of these individuals will suffer an acute major adverse cardiovascular event. Approximately 75% of these individuals will live to tell about it. I suspect that, similar to the disease progression paradigm of Parkinson's disease, only 75% of these individuals will qualify for treatment with MANF. For this indication, I think MANF could justify upwards of $10,000 per treatment.

Similar to my modeling in Parkinson's disease, I assume Amarantus partners MANF for cardiac ischemia. I model a similar milestone and royalty schedule to PD, only I push everything out one year because, hey, Amarantus is small and cash-strapped. Until a partner is on board, they will have to tackle these indications one at a time. I also drop the operating expense and overhead costs, as I've already accounted for the bulk of this in the PD model. I assume only 5% market share, as competition in this indication is already fierce. I model a similar 2% chance at success given my review of the preclinical data and the localized delivery of MANF directly into the damaged heart muscle. Below are the predicted cash flows, yielding a target valuation of around $18 million today.


...Stroke & TBI...

Beyond Parkinson's and cardiac ischemia, preclinical data is suggestive of utility in stroke and traumatic brain injury (TBI) (Airavaara et al, 2009; PR-Feb25, 2013). After all, the neuroprotective mechanism of action makes sense for both indications. The issue for Amarantus here is the required toxicology work for systemic delivery. This work should be completed by 2015, which may allow for IND filings in TBI and stroke in 2016.

Stroke is an enormous opportunity. Several regenerative medicine companies, including Athersys (ATHX), Cytomedix (CMXI), and Neuralstem (CUR) are pioneering cell therapy for the treatment of stroke. Investors know the minefield of drug development that has been phase 2 and phase 3 in stroke over the past decade for small molecules. Cell therapy, offering multiple mechanisms of action and multiple pathways to treat the disease yields hope. But MANF, a naturally occurring biologic, is equally as intriguing.

According to the CDC, there are an estimated 800,000 Americans that suffer a stroke each year. Approximately 90% of these are ischemic events, and 20% die from the event. The only approved treatment is Activase, a tissue plasminogen activator (tPA) with significant limitations for use. In fact, only around 2% of stroke patients qualify for tPA treatment. That leaves as estimated 575,000 Americans that could benefit from a new treatment paradigm.

I've made a number of adjustments to my financial projections for stroke compared to PD or cardiac ischemia. Firstly, I've pushed the potential IND filing to 2016, back two years given the need for additional preclinical and systemic toxicology work. I've pushed the potential approval back to 2023. I've lowered the probability to commercialization to 1%, and lowered the potential probability for an upfront payment here, which I also pushed back to 2017, to 10%.

Activase costs only around $2,000 per dose. I think MANF could cost ten times that given the potentially broader applicability and potential to dramatically improve outcomes. Data suggests that Amarantus could charge as much as $15,000 for MANF in stroke given the average cost of care for a patient post stroke. Below is my stroke model, which yields a value for MANF at around $6 million in this indication.


TBI is a potentially large market opportunity as well. The U.S. CDC estimates approximately 1.7 million Americans suffer a traumatic brain injury each year. Many of these are mild concussions occurring from motor vehicle crashes or during sporting events. Only approximately 25% of traumatic brain injuries are considered severe, and that statistic includes the 40% or so that die from the event. Therefore, the market opportunity for MANF I've chosen to model is in severe non-fatal TBI.

That got the number down to around 270,000 or so each year. Then it occurred to me, Amarantus has been speaking more and more over the past several months about the opportunity to pursue Orphan indications for MANF. One such potential may be in severe non-fatal TBI. I've looked at the TBI market extensively for this article. There are clearly sub-populations of TBI where MANF could qualify for Orphan Drug Status (ODS). ODS would not only provide tax credits and potentially Fast Track development, but seven years of market exclusivity as well.

For the purpose of my model, I focus on only severe non-fatal TBI with non-ODS pricing. However, ODS pricing or moving MANF into mild-to-moderate TBI could be an enormous opportunity for the company. Let's not lose sight of the fact that Amarantus CEO, Gerald Commissiong, was a professional football player, with friends around the NFL. You can bet Gerald is looking at potential funding from the NFL to further explore early-stage concussion work with MANF. Regardless, below is my TBI model, which shows MANF is worth $4 million right now.


Sum-It-Up

I believe biotech stocks should be valued on probability adjusted discounting of future cash flows. Valuing a company like Amarantus, a company with projected negative cash flows and one preclinical candidate in MANF, no matter how promising the scientific literature to date, is surely subject to variation. That means DCF models are only as good as the inputs used, and in the case of Amarantus these inputs include the timing for approval, the milestone schedule, pricing of the drug, penetration rates, peak sales, the economics to Amarantus, the interest rate used to discount the cash flows, and the probability assigned to the likelihoods of each model.

The models I've built tell me Amarantus is worth $42 million, as outlined by the chart below, summing up each of the four potential indications for which there is preclinical proof-of-concept. My overall probability rate for approval of MANF in this indications is roughly 1.5%.


...Sensitivity Analysis...

If I took a more optimistic stance on MANF and bumped my probability rates up to 3%, Amarantus would be worth $77 million. If I took a more skeptical position and lowered my probability rates to 0.5%, Amarantus would be worth around $20 million.

The biggest risk I see to my valuation is the clinical data and time to market. I'm predicting some pretty bullish forecasts for MANF based on the scientific literature. If one wants to know why I'm so optimistic on MANF, please read my scientific review article on Seeking-Alpha. However, what works in fruit flies and rodents does not always work in humans. The neuroregenerative and neuroprotective effects of MANF lead me to believe it has potential disease-modifying affects in Parkinson's disease or in ischemic states post stroke or TBI. If that thesis holds in human trials, MANF is a monster drug. If it does not hold, MANF could be a dud and the company is worth $0.

Finally, as noted above, this article does not touch on Amarantus Diagnostics. Amarantus Diagnostics houses the company's two most advanced products in the LymPro Alzheimer's disease blood test and the NuroPro Parkinson's disease blood test. LymPro is nearing a phase 2 CLIA validation study. Once this study is completed, which we estimate in 2014, Amarantus can file for U.S. approval as a medical diagnostic. NuroPro is about one year behind LymPro.

That means both products could be on the market by 2016 and generating revenues to the diagnostic division. Management believes both products are a $300 to $500 million revenue opportunity. Amarantus is planning to spin-off these assets to form a new company and dividend the shares to shareholders. We intend to focus on Amarantus Diagnostics with a third article. However, our preliminary "back-of-the-envelop" calculations for LymPro and NuroPro eclipse the valuation for MANF. That's presents interesting upside for current investors.

No comments:

Post a Comment