Zhao et al (Tianhe) Publish Successful Phase-I Results (in non-honeymoon diabetics)

In my opinion, this is big news.
It's the result of a phase-I study, published in BioMed Central's Medicine Journal (which is peer reviewed) by a group of researchers working at the University of Illinois at Chicago and in China.

What Did These Guys Do?

This trial was done on people with established type-1 diabetes.  Each person had a blood draw, and then a particular kind of immune cell was separated from the blood and specially processed.  The processing phase used umbilical cord stem cells, but not the patient's own umbilical cord.  (This was generic umbilical cord stem cells, not from the exact person being treated.)  The immune cells were then put back in the person.  The stem cells did not go into the person; they were only used for the external processing.

The goal was to teach the body's immune system to stop attacking beta cells.  The researchers refer to this as "education", and refer to the processing device used as an "Stem Cell Educator".

The treatment was done once and I think it took about 10 hours.  Patients were in the hospital for 2 days, but it's not clear to me if that was due to an over abundance of caution (this was a phase-I study, after all), or if it was really needed.  This trial was done at the General Hospital of Jinan Military Command (Jinan, Shandong, China).  The lead author of the paper is Yong Zhao, is an Assistant Professor at the University of Illinois at Chicago.

The patients averaged about 29 years old, and had had type-1 for an average of about 8.5 years.  Patients were followed for a total of 40 weeks, but most of the results data was gathered at 4, 12, and 24 weeks after the procedure.

The researchers divided their patients into three groups.  Group A (6 people) had some insulin production before treatment. Group B (6 people) had no measurable insulin production before treatment, and Group C (3 people) also had some insulin production  before treatment, but they got a "sham" (or placebo) treatment. 


What Results Did They Get?

There were no significant safety issues during the trial.

Average Daily Insulin Usage:
Group A's insulin requirement was down 38% at 12 weeks (from about 36 to 22 units/day).
Group B's insulin requirement was down 25% at 12 weeks (from about 48 to 36 units/day).
Group C's insulin requirements didn't change.

Note: earlier version of this blog had a typo in Group B's starting insulin units/day.  Fixed here.

Average A1C levels:
Group A started at 8.73 and dropped to 7.67 at 4 weeks and to 6.82 at 12 weeks (almost 2 points overall).
Group B started at 12.2 and dropped to about 10.5 at 12 weeks.
Group C started at 9 and was at 8.7 at 12 weeks.
Note: I consider a drop of 1 to be important, although some researchers consider even a drop of 0.5 as important.  Here we have drops of over 1.9 and 1.7.

Fasting C-peptide (ie. Body's ability to generate "basal" insulin at background levels)
All these numbers are average ng/ml, and are very approximate based on graphs in the paper.
Group A: Starts at about 0.35 and ends at about 0.8 at 24 weeks.
Group B: Starts at about 0 and ends at about 0.5 at 25 weeks.
Group C: Stays at about 0.4 at 4, 12, and 24 weeks.
Note: Non type-1 diabetics generally have a fasting C-peptide level between about 0.5 and 2.0.  The paper's authors felt that 0.6 was the lower bound of normal C-peptide in the population being studied.  So it is possible that Group A has moved into the bottom of the normal (for non-type-1 diabetics) range.

OGTT C-peptide (ie. Body's ability to generate "bolus" insulin in response to food)
All these numbers are average ng/ml, and are very approximate based on graphs in the paper.
Group A: Starts out generating 1 after a meal, at 4 weeks generates about 1.6, and at 12 weeks about 1.7.
Group B: Starts out generating about 0 after a meal, at 4 weeks generates about .05, at 12 weeks about 0.35, and at 40 weeks about 0.6.
Note: I don't know what "normal" is for this test, so can not compare either group to non type-1 diabetics.

The paper also contains data on immune system changes which the researchers felt showed significant improvement in autoimmunity.    I can not evaluate those results.

Also interesting, the researchers are very specific in saying that they think this study shows that beta cells do regrow, in people, if the immune system stops attacking them:

Notably, our clinical data provide powerful evidence that reversal of autoimmunity leads to regeneration of islet β cells and improvement of metabolic control in long-standing T1D subjects.

My take on all this data is that there is no question that the body is generating more of it's own insulin after this treatment than before.  And that happens in people who have had diabetes for a long time, and who are not generating any of their own insulin before treatment.  That's huge.  The results are large enough so that type-1 diabetics would see the improvement in their insulin usage and A1C numbers.

Discussion

Obviously, there is a lot of issues to discuss here:

Possible Conflict of Interest

I hate to start off with discussion of a conflict of interest, but in this case, the situation makes me nervous, so I'm discussing it first.  In the paper's pre-publication version, it says:

Competing interests 

The authors declare that they have no competing interests.

However, the unique and specialized equipment that they used in their clinical trial was manufactured by Tianhe Stem Cells Biotechnology.  This equipment was central to the clinical trial.  However, that company is connected to three of the authors (including both first and last authors).  Here is a quote from the University of Illinois web page:

Tianhe is a stem cell biotechnology company commercializing the inventions emanating from the labs of Drs. Zhao, Mazzone and Holterman within the Departments of Medicine and Surgery at the University of Illinois at Chicago. With operations in both Illinois and overseas, the Company is pursuing the application of stem cells for the treatment of autoimmune diseases.  Tianhe is initially pursuing the treatment of Type 1 diabetes through a clinical system which extracts the patient own stem cells and utilizes them to re-educate the patient’s faulty immune cells to prevent future immune response to the patient’s own insulin producing cell.    [from http://otm.illinois.edu/sites/all/files/files/otm-annual-reportseptember-1final.pdf]

Obviously, I don't know the exact financial details involving the University, Tianhe, and the three researchers, but I would like to see more details before I accepted the claim that the authors had no competing interests!  It certainly sounds like the researchers are testing equipment that they will make money off of, if it works.  On the other hand, if this relationship gets this research to market quicker, and the research helps people with type-1 diabetes, then I'm all in favor of it. :-)

Missing Data

There are a couple of obviously missing data points.  For example, Group A's 40 week post-meal C-peptide numbers are not reported.  And 40 week fasting C-peptide data is not reported for any group.  The trial design only collected insulin use and A1C data at 12 weeks, which is too bad.  I would have like to see it at 24 or even 40 weeks as well.

There are also a few other odditites.  For example: in Groups A and B (treated), about 2/3 of the patients are women, but Group C (placebo) is all male.  Also, the Group B A1C before treatment averaged 12.2, which is a lot higher than you would see in the US (I hope!)

None of this is particularly unusual for a phase-I trial; and none of it makes me nervous about the results.  It just makes me want to see data from a larger phase-II study.

Next Steps To Market

Getting a cure to market requires three things:
1. Scientific success.
2. Engineering and corporate success.
3. Regulatory approval.

This study is a solid, successful phase-I clinical trial.  No doubt (in my mind) about that.
Plus, it is the first time this has been used in people, as far as I know, and there are obvious ways to improve it.  More blood could be put through the machine.  The blood could spend more time in the machine.  It could be used repeatedly, etc.  The classic goal of a phase-II trial is to figure out the best dosing, and I think these guys are well positioned to do that.

There is a clear engineering path to market.  The "Educator" equipment is produced by Tianhe corporation, and I would expect that once the equipment is for sale, many doctors would be able to use it.  It sounds to me like there is a clear way to make money off this, and so I would expect that (if it works scientifically) Tianhe will have no problems getting funding and getting the corporate structure required to build a business around using their equipment to treat/cure type-1 diabetes.

As for regulatory approval, I think there are two paths forward.  The normal path to approval is a phase-II trial, and then two phase-III trials.  The already completed phase-I trial took about a year, I would expect the phase-II to be about that long, maybe a little longer, and phase-III to be still longer.  However, if they are successful and well funded, these guys could get to market in less than 10 years.  Of course, the next question is, what will they make it to market with?  Will the thing available be a treatment?  A cure?  A temporary cure?  Fewer long term complications?

There is a second, much faster, path to approval, called the "surgical procedure exemption".  I do not know all the details, but the FDA does not require that surgical procedures be proven "safe and effective" before a surgeon uses them.  Under this exception the FDA does allow some treatments to be sold without full approval, if those treatments involve taking something out of the body, processing it in some simple ways, and then putting it back in.  I don't know if this "educator" process would qualify or not.

I know some people are nervous about research done in China, and especially at a Chinese military hospital.  I think this is my first detailed blog post on research done in China.  I do want to point out that this trial has ethical approval from the review board at the University of Illinois at Chicago, and that it looks to me like the researchers followed US FDA standards in the trial, and international standards where applicable.  I don't have detailed knowledge of FDA requirements, but it does look to me like, even at this early stage, these guys are working to eventual US FDA approval.

News Coverage: http://www.medscape.com/viewarticle/756691?src=emailthis
Clinical Trials Record: http://clinicaltrials.gov/ct2/show/NCT01350219
Full Paper: http://www.biomedcentral.com/content/pdf/1741-7015-10-3.pdf

Some general information on C-peptide values:
http://www.diabeteshealth.com/read/2000/09/01/2020/interpreting-your-c-peptide-values/

A Personal Note
2011 was an awful year for following clinical trials aimed a curing type-1 diabetes.  We lost three phase-III clinical trials, and started zero new ones.  In phase-I and phase-II trials, the successes seemed small; the failures, great.   These results are a wonderful way to start 2012.

Joshua Levy
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My blog contains a more complete non-conflict of interest statement.
Blog: http://cureresearch4type1diabetes.blogspot.com/