Global Blood Therapeutics: Short-Sellers Could Be In For A Ride

An image of multiple stacks of coins
Credit: Shutterstock photo

By Decision Analytics :

I have written about Global Blood Therapeutics ( GBT ) before here . In my prior article, I addressed a number of fairly legitimate concerns people have cited about the stock and its lead drug for sickle cell disease ("SCD") called GBT-440, and I explained how those risks are less significant than one might believe, or are being successfully mitigated by the company.

In this article, I will attempt to clear up new misconceptions recently brought about by short-selling skeptics, and explain how the moment of truth for this stock is sooner than many people think.

Shorts should be concerned about betting against this stock for a number of reasons:

1.GBT is extremely cheap by any measure of valuation

Analysts estimate peak sales for GBT-440 in SCD to be between $2.7 billion and $3.9 billion. Assuming a reasonable valuation of 3-4x peak sales means GBT deserves a valuation of between $7 billion and $16 billion upon approval of GBT-440. This compares to GBT's enterprise value of ~$400 million as I write this. Analysts have price targets of $26 (Goldman Sachs), $47 (Roth Capital), $62 (Morgan Stanley), $73 (Wedbush), $81 (Wells Fargo) and $83 (Cowen), compared to GBT's current market price of $16.

Are all of these analysts wrong by such a huge margin? Shorts are betting that all of these analysts are wrong, and that no company like AstraZeneca ( AZN ), Merck ( MRK ) or Amgen ( AMGN ) is going to swoop in and pick up a late-stage mega-blockbuster on the cheap.

Regardless of what follows in this article, I would never hold a short position in a stock with this kind of broad and significant price-to-target discrepancy. You would be betting against a stock that could go up several hundred percent overnight.

2. There is no evidence to support short-thesis arguments

I know this is a bold statement, but based on what I know about recent short arguments, I can find no evidence to support them. On the contrary, there is quite a bit of evidence to refute them.

Let us first take a look at the recent data release from ASH, showing excellent hemoglobin improvement results (46% above 1 g/dL improvement vs. 0 for placebo). A snapshot of the poster data illustrates the broad improvement vs. placebo below:

(click to enlarge)

One short-thesis argument that has been put forward is that these hemoglobin improvements will not translate into clinical outcome benefits for patients. For anyone who has done due diligence on this stock or understands hematology, this supposition is without basis.

Here's why:

  • The clinical community and FDA believe a 1 g/dL improvement in hemoglobin is clinically significant, hence the approval of the 1 g/dL primary endpoint of the Phase 3 trial for GBT-440 by FDA.
  • In a study of extracorporeal carbamylation with a 1g/dL increase in hemoglobin, there was an 80% reduction in Vaso-Occlusive Crises ("VOC") and other significant clinical improvements.
  • We know a 0.6 g/dL increase associated with Hydroxyurea results in a 50% reduction in VOC.
  • A 1g/dL increase is what people get transfused with a unit of blood, which dramatically improves SCD symptoms.
  • A curative bone marrow transplant is associated with a 1.5 g/dL improvement in hemoglobin.
  • The hemoglobin improvements with GBT-440, as shown above, are achieved in conjunction with drastic improvements in other parameters such as a 40% reduction in unconjugated bilirubin, a 77% reduction in irreversibly sickled cells and an 18% reduction in reticulocytes.

The chart below provides some points of comparison for hemoglobin increases, and the recent webcast from GBT explains much of the above:

(click to enlarge)

Another recent short-thesis argument that was put forward was that the patients should have had improved oxygen consumption after being on the drug, but the oxygen consumption, as measured by exercise tests, just stayed the same. Therefore, the hemoglobin rise must not be resulting in the drug working properly, and may actually impair oxygen release. This is preposterous because GBT is not designed to improve oxygen consumption. The analysis that GBT did is just a safety analysis that showed what it was supposed to - that there was no hypoxia.

This short argument is akin to saying that patients on Lipitor should run a faster mile than patients not on Lipitor. Otherwise, Lipitor isn't working properly. Lipitor is not designed to do that, and neither is GBT-440. There is no way you could demonstrate improved oxygen consumption with a few dozen patients anyways. This short argument is complete nonsense.

Another concern people have cited is that competitors will take away market share. Recently released SCD results from bluebird bio ( BLUE ), GlycoMimetics (GLYC) and now Novartis (NVS) (the latter through the acquisition of Selexys) show activity in SCD, and therefore, could reduce market share for GBT-440. Yet GBT-440 has the best efficacy and safety profile to date, has an easier and safer mode of administration (a pill), and addresses the underlying disease rather than specific complications like just VOC (and without the addition of some concerning safety issues ).

As a more specific example, the BLUE approach to treating SCD is a complex gene therapy procedure that has only one patient with good hemoglobin results to date out of seven that have been treated. Based on what we know, the BLUE approach appears much more risky than GBT-440. In the recent BLUE study , there were 10 Grade 3 bone marrow harvest-related adverse events reported in three patients, including one severe adverse event (pain/prolonged hospitalizations). Also, six patients experienced at least one severe adverse event post-infusion. What's more, the BLUE process requires a toxic drug to kill off bone marrow to make room for the genetically modified replacement bone marrow, which can cause infertility, infections and increases risk of death.

No patient on GBT-440 has experienced any severe adverse event, and there is no discernible adverse event profile from placebo.

Bottom line is, with the cost, complexity and risk of these other therapies, I just don't see why people would use or need them if they can take a daily pill with no side effects. And with a profound reduction in VOCs with GBT-440, there would be a profound reduction in the need for the other therapies. Therefore BLUE, GLYC and NVS should be very worried about GBT-440 taking share from them, not the other way around.

To finalize this section, I want to highlight what could be a driving force behind some of these short-thesis arguments. One of the most confounding and misleading skeptics of GBT is Elliott Favus . There is a good article by Adam Feuerstein that illustrates what I believe to be a strong inherent bias and desire to manipulate the stock.

Of course, the specific information from Favus notes is very difficult to obtain, and the notes are proprietary, so they cannot be posted, but everything I know about what he has said has thus far been proven wrong by GBT, or been entirely off base.

Perhaps more confounding than the Favus short-thesis notes themselves is that for some reason the stock reacts to them. Shorts should be very careful if they want to follow Favus's advice. This is someone who had a short thesis on Tesaro (TSRO), Sarepta (SRPT) and Relypsa (RLYP), all of which had very positive upside catalysts that have been calamitous for shorts. GBT could be a similar story.

3. The timing and potential impact of the next data release are greatly underappreciated by investors.

Another reason investors might have sold the stock recently (or be shorting), is that they don't believe there will be any stock-moving data until 2H 2019, so it will be "dead money" for a while. I believe this could be a significant miscalculation.

You see, the Phase 3 study has two parts. Part A has 150 patients for 90 days, and Part B has 250 patients for six months. Since Part B is bigger and longer, undoubtedly, GBT is assuming that Part B will take up most of the 2.5 years. Ergo, Part A should be done in about a year (my estimate), or year-end 2017. These results will be reviewed to determine changes in Part B, and there is no reason I can think of that they would not be presented when available.

Furthermore, the Part A study has a flexible design that may not require a full 150 patients. Importantly, although the 250-patient Part B would likely require a formal database lock at the end of the study for preservation of statistical alpha, Part A results could be presented as they are collected rather than after full completion. That means that we could get substantive results at ASH in 2017 or even some early data at EHA mid-year 2017.

Okay, but what will Part A show that is so important? Well, pretty much everything. It will show up to 50 placebo, up to 50 900 mg and up to 50 1,500 mg patients for 90 days, including VOCs and patient-reported outcomes for pain and fatigue. Based on the strength of the Phase 2 data and what we know of the strength of the hematological markers, these results could show dramatic and profound clinical impact. For example, an 80% reduction in VOCs is a reasonable expectation, in my opinion. Given the broad hematological improvement profile shown to date with GBT-440, there is every reason to believe the at-home pain crises and fatigue reductions would be equally stunning given the level of disease modification. An 80% reduction in symptoms, costs and disease burden would be irrefutable and highly publicized.

In other words, to put it bluntly, when the Phase 3A data comes out, it is my belief that GBT shorts will be eviscerated.

And by the way, no one is paying attention to the Phase 2 data of GBT-440 in idiopathic pulmonary fibrosis that will come out in 2H 2017. While I would not bank on this unproven indication, if the results are positive, this could add another multi-billion dollar revenue indication to this drug. No analysts have given any value to this indication, as far as I am aware.


I should note and acknowledge that with any drug development effort, there are risks. I believe the biggest risk with GBT-440 is that a safety signal could come up. To date, however, the safety and tolerability profile to date has been exemplary. Across the GBT-440 clinical development program, GBT-440 has now been dosed in 252 adults, including 128 subjects in multiple dosing cohorts up to 180 days. It has been shown to be well-tolerated with no drug-related serious adverse events. There has been no evidence of tissue hypoxia.

It is also possible (however, unlikely in my opinion) that GBT-440 does not reduce VOC by 80% or even 60%. Although even a 50% reduction in VOC with no safety or tolerability concerns would be a ground-breaking blockbuster treatment for patients with limited treatment options, and I believe it would still result in substantial stock price appreciation.


In conclusion, I believe GBT remains drastically undervalued and continues to be significantly misunderstood. As a result, I believe there is the potential for a 5-10 x appreciation of GBT in the next few years, and a significant portion of that move could come in the in the next 12-18 months. Shorts may benefit from daily swings in the stock but may want to avoid holding long term.

See also Brown-Forman Corporation 2017 Q2 - Results - Earnings Call Slides on

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

In This Story


Other Topics