This Thursday, Aastrom Biosciences (ASTM) is scheduled to present interim phase II data on its tissue repair cell (TRC) therapy for dilated cardiomyopathy (DCM). In preparation of this event, I wanted to bring together the known data for this trial and provide a prediction of what you can expect the data to look like.
For those of you unfamiliar with Aastrom’s technology, I would suggest checking out an animation of the TRC process here.
Basically, it’s an autologous therapy in which cells are harvested from the bone marrow of the patient, multiplied in a lab, and then returned to the patient. The expanded population contains many cells that are quite good at repairing damaged tissue at a higher concentration than would be found naturally. And because the therapy is simply an increased number of cells normally found in the body, adverse events should be kept to a minimum. In fact, Aastrom boasts no treatment related adverse events in over 400 patients thus far.
The current DCM trial involved harvesting and multiplying the aforementioned cells, then injecting the therapy directly into the heart muscle. Scary stuff. But keep in mind these are very sick patients in the most severe classifications of heart failure. Aastrom only enrolled those patients with heart failure stages of class III or IV (description of NYHA classes can be found here).
The trial was initiated in November 2008 with enrollment completed by January 2010. The trial consisted of 40 total patients, 20 with ischemic DCM and 20 with non-ischemic DCM. The patients were followed followed for up to twelve months. The primary objective of the study was to assess the safety of the therapy, but efficacy measures including cardiac dimensions and tissue mass, cardiac function, cardiac perfusion and viability were also investigated. The interim data released on Thursday will contain all patients but just at the six month mark.
We are not completely flying blind on this phase II data release as a number of press releases have given us some real data. I’ll summarize those below.
* Three treatment patients have completed the 3-month follow-up visit.
All of these treatment patients improved from New York Heart
Association (NYHA) class III to class II. This indicates
clinically meaningful improvement in these patients. In contrast,
NYHA class did not improve in 2 of 3 control patients.* Overall quality of life scores (Minnesota Living with Heart Failure
Questionnaire) improved in all treatment patients. Physical and
emotional well-being of all treatment patients also improved based
on patient responses to this questionnaire. There was no
consistent trend in the control patients.* No CRC-related serious adverse events were reported in any of the 4
treatment patients who have completed at least their 1-month
follow-up visit.
* Of the 7 treatment patients who have completed the 1-month
follow-up visit, 1 patient had improved to NYHA Class I and 3
patients had improved to Class II. In contrast, at the 1-month
follow-up visit, the NYHA Class did not improve in 3 of the 5
control patients.* Of the 6 treatment patients who have completed the 3-month
follow-up visit, 1 patient improved to NYHA Class I and 2
patients improved to Class II. In contrast, at the 3-month
follow-up visit, the NYHA Class did not improve in 4 of the 5
control patients.* Of the 5 treatment patients who have completed the 6-month
follow-up visit, 2 patients improved to NYHA Class I, 2
patients improved to Class II and 1 patient deteriorated to
NYHA Class IV. In contrast, at the 6-month follow-up visit,
the NYHA Class did not improve in 3 of the 5 control patients,
including 1 patient who deteriorated to NYHA Class IV.
The presentations have been mostly useless as Aastrom has opted for an anecdotal approach. That presentation can be seen here.
On the efficacy side, the data is absolutely promising. It would appear, just looking at the above numbers, that there most certainly is a trend towards improvement in the treatment group. Although the data is presented in a particularly misleading way (i.e. for those patients on placebo, how much did those that improved improve? Aastrom doesn’t say), it would appear that at least twice as many treatment patients as placebo patients show improvement in NYHF classification.
On the safety side, we haven’t been given much, but there likely wasn’t much to say. I highly doubt Aastrom’s first treatment-related side effect will come in this trial.
I should also note that the trial was placed on clinical hold twice (here and here) – once for what appears to be a bad reaction to the anesthesia and once for a patient death. However, both of these holds were later removed by the FDA and determined to not be treatment related.
So the treatment appears efficacious and safety is top-notch. This is a good investment, right? Not so fast. As a general rule of thumb, I never play phase II data releases; and I’ll tell you why:
- The efficacy is likely to not be statistically significant. The market will have to extrapolate the data to what it thinks results will look like in a broader phase III study. I find this to be wildly unpredictable and is usually based on a general impression given by the press release.
- The safety data is likely to be just noise. What if there’s a death imbalance? What if there happens to be more of a certain adverse event in the treatment populations? This will also be extrapolated, although unfairly, and lead to a crash in share price.
- Phase II is the first demonstration of proof-of-concept. Although not as applicable in this case because we have some interim data, no one typically knows if the treatment works in humans. It’s usually cell and animal models up to this point. As such, the probability of transition from phase 2-to-3 is roughly 50% – lower than both the transitions from 1-to-2 and 3-to-approval.
So in summary:
The data is likely to be quite positive; demonstrating a positive trend in efficacy. The safety data will also likely be flawless, as it has been in the past for Aastrom’s TRC therapy. These are known with a relatively high probability. However, the market’s reaction to this data will be a HUGE unknown. I would say that of phase II data releases you’re likely to see this year, this may be the best one to play. However, I will not be putting my money in any of the phase II trial data releases.
Good luck to those holding a position though the rest of the week!
Disclosure: No positions.
Note: Data on phase transitions can be found in Biotechnology Valuation, by Karl D. Keegan.
