Aastrom’s critical limb ischemia trial – failure analysis

by Alexey Bersenev on March 31, 2013 · 0 comments

in failure

This week, US-based company Aastrom Biosciences announced that they are terminating Phase 3 trial for critical limb ischemia (CLI) – REVIVE. The company will lay off half of its workers and shift focus to other indications. This was rather sudden news than expected. Nine days earlier, the company reported financial results of 2012 and didn’t give any hint to strategic changes for investors. Most professionals called it as a big setback or “big bummer“/ “huge disappointment” for cell therapy industry. Obviously, this is very sad news for the field. But we have to analyze the root causes for this failure and learn some lessons. Analysis is very important, because the failure in the final phase of trial is a big deal and we don’t have many examples of such failures in cell therapy.

The major reason for REVIVE failure is a poor trial design. The secondary reason is financial and strategic business considerations. Apparently, the failure has nothing to do with safety and efficacy of Aastrom’s cell product Ixmyelocel-T. Let’s look at
these reasons more closely.

Poor trial design
Slow enrollment
The Phase 2 was completed in 2011 and in May 2012 enrollment in Phase 3 study has been started. With enrollment goal of 594 patients, by March 2013 only 40 patients were enrolled. If the enrollment rate will be the same, it will take about 14 years to reach the enrollment goal. Even though trial could be finished earlier in case of “excellent results”, it still very very slow.

Inclusion criteria are too stringent
This is a possible reason for slow enrollment. You can look at list of eligibility criteria on a trial description site. I think, the mistake was made in estimation of number of patients, who would perfectly fit in eligibility criteria.

On the 4th quarter of 2012 report, the company proposed a plan for enrollment acceleration, which included (i) increase of trials sites from 70 to 100+, (ii) change (loose it by removing “no option” definition) the enrollment criteria. It took only one week from company management to realize that they “have no muscles” to implement these strategic changes rapidly, effectively and without big financial loses.

Analyst Henry McCusker wrote in his blog:

My issue is that ASTM had “prior” knowledge and a “glimpse” of the recruitment problems based on the facts and evidence of the P2 recruitment experience – who and where was the “internal” voice on the trial issues.

Primary end point and number game
Yet another thing to consider is that if they will make inclusion criteria wider, they can face a risk of “therapeutic power dilution” among heterogeneous population of patients. The targeted enrollment number (=594) was defined based on results of Phase 2 and estimated to reach significant statistical power between “placebo” and “cells” groups. Because ~600 is a big number for cell therapy, I’d suspect that the efficacy of Ixmyelocel-T is not as dramatic as one may think. Thus, Phase 2 included 77 patients and no difference in amputation free survival (AFS) has been reached. The company still put AFS as a primary end point in Phase 3 design, hoping to reach a significant difference by shooting a big number for enrollment.

Strategic business considerations
Even though company has enough cash to run trials in 2013, it is getting very low. The company consistently reported net loss from quarter to quarter, without any news about significant investment. Since REVIVE trial will required a lot of cash and, apparently, very long duration, one may ask if Aastrom can really afford it. Unfortunately a potential deal between Takeda and Aastrom about commercialization of Ixmyelocel-T in CLI didn’t work out. Seem like Aastrom was really in need of such partners. Yet another interesting fact is that company has changed 4 CEO’s in the last 5-6 years. You can read more about Aatrom financial shape here and here.

Analyst Bill Busa wrote:

Lacking proprietary technologies of interest to other companies, Aastrom generated zero revenues from licenses and royalties for the 2010 through 2012 period (though it should be noted that other SCSI companies did little better, averaging just under $500K in license and royalty revenues in 2012). More puzzling still was the company’s unwillingness, or inability, to pursue grant and contract revenues…
Perhaps the worst threat to Aastrom’s well-being of all has been its chronic inability to lure a strong strategic partner — an obvious necessity when all other access to capital is barred.

Hot competition
The field of commercial gene- and cell therapeutic products for CLI is very hot and competitive. You can read a recent review about potential players on a market (available in open access .pdf). Just to mention a few – Harvest Technologies and Biomet with poin-of-care processing devices, Pluristem, Baxter, Medistem, Juventas and few gene therapy companies. Harvest Technologies is the only one (other than Aastrom) with Phase 3 ongoing trial. Harvest is enrolling up to 210 patients in 25 US centers. The company is running clinical trials in Asia and marketed their device for CLI in Europe. In case of the relatively similar efficacy, the companies with point-of-care devices will win the competition, because of low cost and convenience to use.

Ixmyelocel-T and the cost issues
The leading Aastrom therapeutic product Ixmyelocel-T is quite unique in a terms of composition. It is an autologous bone marrow mononuclear cells, expanded ex vivo for 2 weeks and contained lymphocytes (5x reduced), monocytes + macrophages (expanded 200x), granulocytes (5x reduced), hematopoietic progenitors and stromal cells (50x expanded CD90+ cells). It look like just a “cell product” to me, but not “stem cell product”. I’d speculate that expanded cells of monocytic lineage plus hematopoietic progenitors could have even greater therapeutic value than CD90+ stromal cells. But does unique composition of Ixmyelocel-T correlates with its unique therapeutic value? I don’t think so. Taking in account good results from many CLI cellular trials, Ixmyelocel-T is not very special.

You can learn more about Ixmyelocel-T manufacturing from this video:

Business analyst Jason Napodano has estimated a potential cost of Ixmyelocel-T as $40,000 USD. Taking in acoount 2 weeks of ex vivo expansion in bioreactor, it sounds reasonable to me. Here it comes to competition. Obviously, point-of-care cell processing devices (by Harvest and Biomet) will make cost of the therapeutic product/ procedure much cheaper. I’d expect something less than $10,000 USD. Let’s look at another example – a gene therapeutic product Neovasculgen (VEGF), which was approved on Russian market in 2011 and priced ~ $6,600 USD. I’d speculate that most of gene therapeutic products for CLI will be priced in the range <$10,000 USD. So, in order to compete with such lower priced drugs on the market, Ixmyelocel-T must have much much better efficacy profile.

Amit Patel wrote about the cost of cell therapies for CLI:

Irrespective of the expected low pricing for allogeneic cell dosing, autologous cell manufacturing, with plastic disposables, may drive production costs below $100.

New therapies costing more than $60,000 are probably prohibitive, but may be a possibility in the $20,000–$40,000 range, while therapies costing under $20,000 have the highest likelihood of success.

To conclude: It’s very sad that Phase 3 REVIVE trial, assessing cell therapy for CLI, has been halted by Aastrom. For the company it was probably wise business strategic decision, unless they had some buyers. This case shows that could be a number of reasons for clinical trial failure, other than lack of safety or/ and efficacy. Unfortunately commercial cell therapy trials can become a very long and and very expensive exercise. We can also learn that it is extremely important to design a trial in wise and smart way. The design, which allows to predict enrollment rate and power the efficacy with lower patients number.

Read 3 more analysis of failure by:
Bill Busa
Henry McCusker
Jason Napodano

{ 0 comments… add one now }

Leave a Comment

Previous post:

Next post: