Mesoblast (MBLTY) (MESO) is in the process of raising $70 million through an IPO in the U.S. for ADR shares (link here). Potential investors should be aware that Mesoblast is not a new company but instead a disappointing and failing Australian biotech. It is already listed in the Australian stock exchange, and its share price has been hitting 4-year lows despite a historic bull market for biotech stocks.
Mesoblast's share price has declined, as Australian investors have come to realize that the pipeline has missed major endpoints in clinical studies, the touted results from clinical trials have been data-mined, and the company is rapidly burning cash. With three straight strikes against it, we believe that Mesoblast is clearly bringing a turkey to American investors. We are short shares of Mesoblast with a price target of $1.25 per ADR. This is approximately 90% below current levels for the ADR (link here).
Mesoblast's MPC failed its Phase 2 trial and data-mined to show success
Mesoblast's lead product is MPC-150-IM (formerly Revascor), which is made of mesenchymal precursor stem cells. It is currently in a Phase 3 for advanced chronic heart failure (CHF), which started in late-2013 (link here). Mesoblast is partnered with Teva Pharmaceutical Industries (NYSE:TEVA), which is conducting the Phase 3 trial.
The premise of MPC-150-IM is based on findings from a Phase 2 study that were presented in 2011 at the American Heart Association conference (link here). The Phase 2 study was a single-blinded trial that enrolled 60 patients across 4 arms: control, low dose of MPCs, medium dose, and high dose. Mesoblast also combined the three doses into a single pooled group.
Per the presentation, the Phase 2 trial had 5 pre-specified endpoints over 3 time periods (3, 6, and 12 months) (see slide 7, "Study Design").
- NYHA assessment
- 2D echocardiogram / left ventricular ("LV") end systolic volume
- 6-minute walk test
- Angiography versus baseline
- Left ventricular ejection fraction (LVEF)
The 2011 AHA presentation clearly shows the study missed its LVEF endpoint, left ventricular end-systolic volume endpoint, and 6-minute walk test. None of the p-values were close.
A paper published in 2015 in the journal Circulation Research on the Phase 2 trial results revealed there was a secondary endpoint based on major adverse cardiac events, or MACE (link here). Mesoblast also missed that endpoint. The paper states:
"We found no significant difference between MPC-treated and control patients for survival probability, MACE-free survival probability, MACE-free probability, and rates of all-cause mortality."
We believe that having missed its primary and secondary endpoints in its Phase 2 trial, Mesoblast resorted to data-mining and post-hoc analysis
In the 2011 AHA presentation, Mesoblast highlighted two endpoints that did reach statistical significance: changes in LVEF at 3 months for the low-dose group, and LV end systolic volume at 6 months for the high-dose group. However, it's clear that these results were data-mined and simply by chance and not a meaningful clinical result. The LVEF endpoint hit for the low-dose group at 3 months, but did not show a benefit at 6 or 12 months. The LV end systolic volume endpoint hit for the high-dose group at 6 months, but failed at 3 months and 12 months. There's no consistency to either the therapeutic timeframe or the therapeutic dosage.
In addition, Mesoblast claimed the 150M dosing group was successful in reducing heart failure-related MACE, or HF-MACE (see 2015 Circulation Research paper link above).
However, the paper clearly states that HF-MACE was a post-hoc endpoint that the investigators created after the original MACE endpoint failed. We have called out the use of post-hoc analysis by biotech companies in the past (link here) and find this one especially egregious. It's clear from the graph above that this is just another exercise by the company in data-mining. For example, in the HF-MACE endpoint, the mid-dose 75M MPC group did even worse than the control group, which suggests that this is just data mining and there is no real benefit in HF-MACE. Again, there's no consistency in the data. This is what happens when companies data-mine to find any positive results.
Recently, Mesoblast has been touting a left ventricular remodeling endpoint (link here). However, like every other supposedly successful endpoint from the Phase 2 study, this one too was data-mined through post-hoc analysis. The company clearly states:
"A post-hoc analysis was performed in 30 patients from the Phase 2 trial who had been randomized to receive either placebo or a single administration of 150 million MPCs (MPC-150-IM)."
Teva has an early termination clause in the Phase 3 trial in CHF
Given the true results of the Phase 2 trial, we believe Teva is conducting the Phase 3 trial but has very low expectations. First, Teva is only conducting the trial because Cephalon, which Teva acquired in 2011, entered into an agreement with Mesoblast in 2010 that obligated it to fund a Phase 3 trial (link here). The agreement states, "Cephalon will be responsible for the conduct and expenses of all Phase IIb and III clinical trials." It is important to note that the agreement was struck in 2010 before the final results of the Phase 2 were released in 2011.
Second, Teva has an early termination option for the Phase 3 trial once 125 patients are treated. In its 3Q 2013 conference call (link here), Teva's Global R&D Chief Scientific Officer, Michael Hayden, stated:
"The trial is a Phase III trial with, importantly, an interim analysis. The interim analysis will be done after about 125 patients and will help determine whether there is a reason to continue the trial. In other words, the signal is such that efficacy is likely to be achieved in the broader Phase III trial. Teva is committed to the first phase of the trial."
This doesn't sound like a ringing endorsement by Teva, and Teva has never identified cardiovascular therapies as one of its focus areas. We believe it's clear that Teva inherited the Mesoblast program from its Cephalon acquisition and, having seen the full Phase 2 results, structured the Phase 3 trial so that it can terminate the trial as soon as possible.
Mesoblast's other highly touted programs have also failed their clinical trials
In addition to MPC-150-IM, Mesoblast has been touting MPC-06-ID in Chronic Low Back Pain due to Degenerative Disc Disease, and MSC-100-IV / TEMCELL HS Inj. in Acute Graft vs Host Disease. We found several issues with both of these product candidates.
Mesoblast's MPC-06-ID failed its Phase 2 study, and Mesoblast resorted to post-hoc analysis and data-mining
Another lead candidate of Mesoblast is MPC-06-ID, which entered a Phase 3 trial earlier this year (link here). Like Mesoblast's MPC-150-IM, MPC-06-ID failed to hit its original primary endpoints in its Phase 2 trial in patients with lumbar back pain, and Mesoblast resorted to data-mining and post-hoc analysis in order to present positive results.
Mesoblast has been highlighting pain and functional response rates at 6 and 12 months, and a composite endpoint for both pain and function over 24 months (link here, see slides 25 and 26). It's important to note that none of these are pre-specified endpoints of the original clinical trial (link here). The original endpoints were safety, "change of treated lumbar intervertebral discs using Magnetic Resonance Imaging (MRI) at 6 months post injection of MPCs," and "effectiveness of MPCs in reducing chronic lumbar back pain using the visual analog scale (VAS)."
Mesoblast revealed in a press release in April 2013 that the trial had missed the MRI endpoint (link here). It states:
"MRI status of disc morphology remained unchanged at 6 months relative to baseline in all groups."
As for change in VAS score, we note that Mesoblast had to slice and dice the data in order to find anything statistically meaningful. For example, Mesoblast highlights the percentage of patients who are VAS responders as being statistically meaningful (see link here, slide 25). A detailed analysis reveals that Mesoblast defines "responders" as having a "50% VAS back pain reduction AND no intervention at the treated level." The Phase 2 clinical trials page does not state that the VAS endpoint must meet a 50% reduction threshold nor that it must be paired with no intervention at the treated level (link here). We believe Mesoblast simply engaged in aggressive post-hoc analysis and data-mined to find statistically significant results. The same can be said for its functional responder findings in that same slide.
As for the composite endpoint on slide 26 that Mesoblast highlights, it's important to note that Mesoblast changed the fields within the composite endpoint. The statistically significant findings were for a composite endpoint consisting of "50% VAS back-pain reduction and 15-point ODI improvement and no intervention at the treated level." This is a very different definition from the original pre-specified composite endpoint, which Mesoblast revealed in its April 2013 press release (link here).
In that press release, it clearly states that the pre-specified composite endpoint is defined as,
"requiring a patient to meet all of the following criteria: clinically significant levels of improvement in back pain and function at 6 months compared to baseline, maintenance or improvement in neurological status from baseline, maintenance of disc morphology compared to baseline as evaluated by MRI, and no serious adverse event or secondary intervention."
The new composite endpoint that Mesoblast is touting is completely different and is missing the MRI component, which we know the trial missed. We believe this clearly indicates that Mesoblast has again engaged in post-hoc analysis to data-mine for positive results.
We believe all the supposedly positive findings from Mesoblast's Phase 2 trial in chronic low back pain due to Degenerative Disc Disease are meaningless. It's quite clear to us that the trial failed.
Mesoblast's MSC-100-IV is Osiris Therapeutics' old, failed stem cell product called Prochymal
Mesoblast's MSC-100-IV is actually Osiris' (NASDAQ:OSIR) Prochymal product that Mesoblast acquired and renamed back in October 2013 (link here). Mesoblast has been touting MSC-100-IV's potential in acute Graft Versus Host Disease (GVHD); however, we think any enthusiasm for MSC-100-IV is woefully misplaced. First, Mesoblast acquired Osiris' entire stem cell therapeutic business for $20 million, which is not indicative of blockbuster potential. Such a low price likely was possible because Prochymal had failed not one but two Phase 3 trials in GVHD in 2009 (link here). When these trials were reported, Osiris' stock tumbled 40% (link here).
Despite failing its two Phase 3 trials, Osiris was able to obtain conditional approvals for Prochymal in Canada in 2012 (link here) and New Zealand (link here). However, it hasn't generated any meaningful revenue since approval. In addition, there's been little development under Mesoblast's ownership other than approval in Japan earlier this year (link here). While this sounds promising, we don't believe there is much value here. First, Osiris originally entered into the agreement with JCR Pharmaceuticals, which is the partner for MSC-100-IV in Japan, in 2003. JCR Pharmaceuticals paid Osiris a total of $3.0 million in upfront payments (link here, see "JCR Pharmaceuticals Co., Ltd.-License Agreement"). This hardly indicates that MSC-100-IV has big potential in Japan. Furthermore, Mesoblast only collects royalties from sales in Japan, which are expected to be small.
We don't see any value in Mesoblast's program for MSC-100-IV in Graft versus Host Disease. We warn investors against assigning any real value to this program.
Mesoblast has less than 1 year of cash remaining
Despite failing its clinical trials, Mesoblast has been very good at two things: burning cash and raising money from investors. Over the past 4 years, Mesoblast has burned over $300 million in cash and has raised over $240 million in cash from share issuances.
As of June 2015, Mesoblast had $144 million in cash (link here). Based on its trailing-twelve-month cash burn, Mesoblast now has only about 1 year of cash remaining.
Mesoblast's low cash balance and continued high cash burn is the reason it's trying to sell shares. We believe the company is trying to do an IPO in the U.S. marketplace now because Australian investors are starting to wise up to Mesoblast's actions and are fed up.
Conclusion: Short Mesoblast shares. Price target $1.25
For the reasons stated above, we believe Mesoblast's pipeline is worthless. It has had to data-mine for positive findings from both its chronic heart failure and back pain Phase 2 trials. For its MSC-100-IV product, we note that it has failed two trials in GVHD, and we don't believe that a Japanese approval is financially meaningful. We believe it has zero value, given its history but will assign it its acquisition value of $20 million.
Our price target for Mesoblast is calculated as follows.
Disclosure: I am/we are short MEOBF, MBLTY.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
Editor's Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.