Sunday, January 30, 2011

Glucagon / Leptin type-1 "Cure" in the News

This posting discusses research which has mostly been done on mice.  Please remember that anything that cures type-1 diabetes in mice is always years away from general availability in people.  I'm posting about this because it is making a big splash in the news, so I expect to get questions about this from friends and relations. I've already gotten questions from BBs.  Here is an example of one the better news articles:

This research is quite different from other research aimed at curing type-1.

Drs. Unger and Lee Publish a Mouse Study Explaining Why Leptin Might Cure Type-1 Diabetes (or make it asymptomatic)


Several months ago, these researchers published a paper showing that giving Leptin to NOD mice was in some ways better than giving them Insulin.  Mice given Leptin had much better BG control than mice given Insulin, and that many mice given Leptin did not need to take Insulin at all.
Previous blog:
Related research:

Just recently, these researchers also started a clinical trial of Leptin, to see if people would have the same benefit seen in NOD mice, which I've covered a little bit.  But I did not cover it aggressively, because it was not clear if it was a cure for type-1 diabetes, or just a better treatment:

Now, they have published another mouse study, which describes why they think Leptin helps type-1 mice so much.   And it is clear that they think they are on to a cure mechanism for type-1 diabetes.
Extract of commentary:

What The Current Paper Means

This paper suggests a much different mechanism for type-1 diabetes than previously believed.
First, it suggests that mice can utilize the carbs in food without producing their own insulin.  That even animals with no beta-cells to produce insulin and no injected insulin, will still live long, happy lives if they are injected with Leptin.  Insulin is not the "key that lets sugars into cells" that we have been told that it is.
Second, The high blood glucose levels that are seen in type-1 diabetics who do not take insulin are actually caused by glucagon.  And these researchers believe that glucagon (in type-1 diabetics) is incorrectly over generated by the body, triggered by a lack of insulin.

Obviously, this suggests that type-1 diabetes can be treated by preventing (or lowering) the generation of glucagon.  In fact, from a functional point of view, it may be that preventing (or restricting) the body from making glucagon results in a "cure" where the person does not need to measure carbs, inject insulin, or suffer the long term side effects of type-1 diabetes.  This could "kick off" a long discussion of what is a cure, what is not a cure, and so on.  Different people have fundamentally different definitions of a cure, and this can lead to huge arguments.  My definition of a cure (which is on my blog) is "functional".  Others think you must stop the autoimmune attack for a real cure. I don't think it is a worthwhile argument to have now, but do remember: even in the absolute best case, this research is aimed at a functional cure.

Leptin is an example of a treatment which is known to restrict glucagon generation, but there may be others.  Some research showing Leptin's impact on glucagon:

As this research progresses, it will be interesting to see if Leptin used in this way effects the use of emergency glucagon injections.  My guess is that it would not, there would be so much glucagon in the injection that a Leptin dose earlier in the day would make little to no impact.  But testing would be needed to make sure of this.

Some General Discussion

One comment I got about this research is this: "They cured mice.  So what.  Happens all the time, why care about these guys."  And that's true.  When someone announces a mouse cure, they are usually 2 years away from the start of a human trial, and often never get even that far.  But these researchers are in a different situation. They are NOT announcing a mouse cure.  They announced the mouse cure a year ago.  They have already started the human trial (3 out of 15 people already have started the treatment).  What they are announcing now is research into the mechanism that is curing these mice.

But notice how many times I used the word "mouse" in the previous paragraph.  One thing to keep in the front of your mind throughout this discussion, is that people are not mice, and mice are not people.  We know that type-1 diabetes is not the same in mice as in people.  We get mice cures at a rate of several per year.  We get people cures at a rate of zero so far.  Over 100 mouse cures have NOT worked in people.  One of my serious worries about this research is that it won't work on people at all.  That it is based on something in mice that is not there in people.  The only way to allay that fear is human trials, and that is why I'm excited that these guys have already started such a trial.

Also, it is important to remember that there were two studies, with two types of mice.  The first study, which used Leptin to cure type-1 diabetes used NOD mice.  The mice in the second ("mechanism") study were not NOD mice, and did not have autoimmunity. They were given a drug that killed their beta-cells, which induces diabetes, but not the autoimmunity part. In theory, for this research, that should not matter, but I'm always nervous about researching type-1 diabetes in mice that don't have autoimmunity.

One obvious issue is that this theory assumes that insulin is NOT actually needed to get the energy from carbs.  It assumes that if glucagon is removed, a type-1 diabetic will still be able to use all the energy in the carbs eaten.  However, this goes against mainstream medical theory that has lasted for decades.  It seems a little unlikely that no one has noticed -- even after decades of research -- that type-1 diabetics can process carbs without insulin.

Another issue is that some people have had their entire pancreases removed, and those people would not generate any insulin or glucagon.  However, they do have the classic symptoms of type-1 diabetes.  That doesn't fit in with this theory.  (Thanks to CWD's LantusFiend for pointing this out.)

I sometimes get asked "do you think it will work" (about a lot of research, not just this research).  I know that question is in the back of many people's minds.  But the question is meaningless.  It doesn't matter if I think it will work, for me or anyone else.  It only matters if it actually works, and that means data from clinical trials.  Not opinions.

How do we get from here to a cure?

I don't think we can replicate the recently announced mouse experiments in people.  These experiments used specially bred "glucagon knockout" mice.   These are mice which are a specific genetic defect that turns off their glucagon system.  I don't think there are people like that walking around.  Even if there were, we would need to find people who didn't have a glucagon system and did have type-1 diabetes.  And even if such people existed, according to these researchers, their type-1 diabetes might be asymptomatic.  They wouldn't even know they had it!  So they couldn't be recruited for a study.

So back in reality, the only way to test this in people is to find a safe way to turn down the glucagon system of someone who already has type-1 diabetes.  If this theory is right, their type-1 diabetes would become asymptomatic (or maybe they would require far less insulin than they do now).  But turning off a major hormone is not something done lightly.  Hormones are very important, and a little tricky: we often don't know all the functions that one hormone has.  Researchers are often surprised that a hormone that they thought just did X, also does Y, or stops Z from happening.  So shutting it down to help with X might cause problems in Y or Z.

So, with all that as background, these researchers know that Leptin lowers glucagon generation, so giving Leptin to type-1 diabetics is an obvious clinical trial to run, and they have already started such a trial.  However, there may be other ways to limit, control, or stop glucagon generation, and it might be interesting to run other clinical trials that use other method to turn down a type-1 diabetic's glucagon system (carefully).

An obvious question is: how safe is Leptin?  The current phase-I study in type-1 diabetics is not the first trial for Leptin.  It has already completed some phase-I studies in people who have type-2 diabetes and in people who are over weight.  However, the real question is, how safe is it when used on type-1 diabetics in the dose needed for them to get the good effects we want?  And also, do any safety issues surface in the larger phase-II and phase-III studies (for any indication)?  And those issues can only be resolved by going through the clinical trial process.

Finally, Amylin is a commercial company that is researching Leptin, and they are a collaborator in the clinical trial.  Amylin is a big name in the treatment of type-2 diabetes, but as far as I know, they do not have a drug aimed at type-1 (so no "conflict of revenue").  They are interested in Leptin as a combination weight loss treatment.  But obviously, good results in treating type-1 could open up a whole new market for them.

One Last Point

If these guys are right about glucagon and Leptin in people, and their study uses the right dose, then they are going to find out quickly.  Their current study is NOT blinded, and they believe the effect is not honeymoon dependent.  So even though their experiment is supposed to last until early-2013 (and may last longer), the researchers involved (and the patients involved) may know sooner.

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. 
To Get as Email Join here:

Wednesday, January 19, 2011

Possible Cures for Type-1 in the News (Jan)

Polyclonal Tregs Starts a Phase-I Clinical Trail

This one is a little complex.  The body's immune system includes T-regulatory cells, which help control the "killer" T-cells which (in type-1 diabetes) mistakenly attack the beta cells.  Some researchers have attempted to reduce the number of "killer" T-cells, while other researchers are trying to raise the number of T-regulatory cells.  This research is trying to raise the number of T-regulatory cells.  The basic technique is as follows: remove some of a patient's own T-regulatory cells, which is the same processes as giving blood.  Then separate and purify those T-reg cells, and grow them for about 2 weeks.  You can end up with 1000 times as many as you started with. Finally, put them back into the patient.  Since these cells naturally regulate (or control) the body's immune system, having more of them may result in the body stopping it's own autoimmune attack.   They've had good results in NOD mice.  Finally, some researchers believe that the previously seen good results in some other treatments (such as anti-CD3 and ATG) might be caused by these treatments stimulating T-reg production, rather than, or in addition to, lowering "killer" T-cell production.

This study involves 14 people, who have had type-1 diabetes for more than 3 months, but less than 2 years.  It is primarily a safety trial (to make sure the procedure is safe), but has secondary measures to also see if the treatment improves type-1 diabetes (for example: higher C-peptides, lower A1Cs, etc.).  It is a single site study, being done at UCSF (San Francisco, California, USA).  Dr. Gitelman  ("Dr. Steve", if you attend Bearskin Meadows camp) is running it.

Unfortunately, this study will not be quick.  It is expected to last until 2016.

Clinical trial record:
UCSF's web page:

Canakinumab Starts a Phase-II Clinical Trial

Canakinumab is a monoclonal antibody, which is designed to lower inflammation.  It was approved in 2009 (both US FDA and EU EMEA) for a collection of rare autoimmune based inflammatory diseases.  Good results have been seen in people with type-2 diabetes, and it has been used in children as young as 3. Therefore, it is known safe, and can start a phase-II trial for type-1 diabetes.  The trial is for 66 people who will be followed for 4 years.  It is honeymoon only (100 days from diagnosis).  The treatment is monthly injections for a year: 2/3 get treatment, 1/3 get placebo.  My general comments on inflammation based cures apply here:

This is a large trial, recruiting in many locations in the US, which are listed in the clinical trial record below.  For locals: both UCSF and Stanford are participating.

Clinical trial record:
Wikipedia entry:

Data Published from extended follow up to Diamyd Phase-II Trial

Diamyd as published some extended follow up up data from their phase-II trials, which I have not had time to review, but you can see it here:

Full Paper:
Clinical Trial:

Not Yet In Human Trials

There were several new types of artificial pancreas which I heard about in 2010, and this is the latest.  It is called a "Bionic Pancreas" and is basically a dual-hormone (insulin and glucagon) AP, but designed as a single, custom computer chip.  They hope to start human trials "this year [2011]".  These guys know about the dual-hormone work being done at Harvard, and the two groups are using some of the same ideas, but pushing different parts of the technology.  The Harvard guys are focused more on exact algorithms, and these guys (at Imperial Collage, in the UK) are focused more on a single chip package.

News report:

Not Type-1 Diabetes Related

The following article describes the ethical issues of a chemist who works with brain chemicals in rats, and who's work is often used by black marketeers to create street drugs, which sometimes kill people.  It is an interesting read and an interesting moral dilemma.;_ylt=Apj8w6un3xq9J8J_t4MMSEAPLBIF;_ylu=X3oDMTJwM2I5NnNlBGFzc2V0A2FwLzIwMTEwMTA1L3VzX3NjaV9oYXVudGVkX3NjaWVudGlzdARjcG9zAzEEcG9zAzIEc2VjA3luX3RvcF9zdG9yeQRzbGsDc2NpZW50aXN0aGF1

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. 

Sunday, January 16, 2011

Symposium on Therapeutic and Preventive Vaccines for Autoimmune Diseases

I recently attended a Symposium on Therapeutic and Preventive Vaccines for Autoimmune Diseases as a guest of the JDRF.  This posting contains some information that I got there, and also some thoughts and opinions that bubbled up while I was there.  This is not a summary of the meeting, it's more "bits and pieces" motivated by the meeting.

But before I do that, I want to discuss the word "vaccine".  When I talk to parents or people who have type-1 diabetes, and ask them what the word "vaccine" means, they always say something like "something you give to a healthy person to prevent a disease".  And they wonder why anyone who already has type-1 diabetes would care about a vaccine.  But in the scientific world, the term "vaccine" means something that changes how the body responds to a specific substance (called an "antigen").  As researchers use the term, a vaccine can cause the body to respond more strongly or less strongly.  So from this point of view, a vaccine which causes the immune system to stop attacking it's own beta cells would be very important to cure research.

This symposium was jointly sponsored by JDRF and FastForward which is focused on finding a cure for multiple sclerosis.  Both of these are autoimmune diseases (ie. diseases caused by the body's immune system mistakenly attacking a working part of the body).  So the idea is to jointly research and promote cures for autoimmunity in general.

There was one piece of bad news about why companies do not invest in type-1 diabetes, and two pieces of good news about why they do:

The bad news:  If you've got 50 or 100 million dollars to invest in research, are you going to spend it on type-1 diabetes, or type-2 diabetes (which has about 10 times a big a market)?  You are going to spend that money to develop a type-2 drug.  Bigger market.

The first good news:  Curing type-1 is an "unmet need".  That means there is no competition.  You make more money if there is no competition, so that is motivating people to research type-1 even in the face of the larger type-2 market.  If you enter the type-2 market you have lots of competition (and some of it is really cheap, so it is hard for you to charge a lot for a new drug, if an old one is cheap).  However, if you enter the type-1 market, you own it.

The second good news: Type-1 is related to other autoimmune diseases (especially: rheumatoid arthritis, multiple sclerosis, and possibly lupus and others.)  So that means that money you put into curing type-1 diabetes might also lead to a cure for these other diseases, and that could be very profitable.  A sort of three for one deal on cures.  This also motivates companies to put more money into type-1 research.

This leads into discussion of the symposium's magic word: "platform".  You might think that a platform is a wooden box that you stand on.  I'm a software engineer, and we use "platform" to mean a bunch of software that helps develop new software and can be used over and over again for that purpose.  The pharma guys use the word in much the same way that software guys do.  A pharma "platform" is a way to speed the development of multiple drugs.  Everyone who is working on a drug, talks about their platform.  They hope that their development can be used again and again to develop multiple different cures for different diseases.   The word "platform" represents the unbridled optimism common to researchers; and the funding opportunity that every venture capitalist, pharma company, and non-profit is looking for.  They haven't even started a clinical trial, and are already talking about how they will cure multiple diseases in the future. "Platform" is the pot of gold at the end of the rainbow.

To be a little more specific, two types of platforms were discussed.  The first was a common collection of ingredients that you can customize to cure different diseases.  Consider this bread analogy: you have a recipe for pecan bread.  You try replacing the pecans with almonds, now you have almond bread.  You replace them with blueberries and now you have blueberry bread.  Your bread is what the pharma guys would call a "platform".  You drop in one new ingredient to create a new bread.  If your bread recipe is good at this sort of flexibility, then a commercial bakery would be very interested in it, as well as all the "specialty" bread recipes.

Bayhills has exactly this kind of platform:  it is a ring structure of several chemicals, one of which is specific to type-1 diabetes.  That drug is called BHT-3021 and is targeted at type-1 diabetes.  But if you take the same basic ring, and replace the type-1 chemical with a different one aimed at multiple sclerosis, then the drug is called BHT-3009 and is aimed at MS.  And so on....

Another kind of platform is a method to find drugs, which you can use again and again on different diseases.  Again, to use a cooking analogy, let's say you are looking for a dinner recipe and so you grab a can of chicken soup and pour it over chicken meat and bake it.  A week later you grab mushroom soup, pour it over beef and bake that.  You now have a "platform" for making recipes.  The platform is this: pour a can of soup over a meat and bake it.  None of the ingredients are reused (as they are above), but the basic technique is reused.    There were a couple of different researchers with this kind of platform at the meeting.  Including Apitope and Dr. Mannie (neither in clinical trials as yet).

The second most important word at the symposium was "bio-marker".  A bio-marker is a way to do an experiment for cheap.  For example, lets say you have a drug and you think it cures type-1 diabetes.  Running an experiment to see if it does will take years: you need to make sure type-1 doesn't come back.  However, if you had some blood test that told you the person no longer had type-1, then you would not need to follow them for years.  You would just need to do the blood test.  That blood test would be for a "bio-marker".  Something that showed you the drug had worked, but was cheaper, quicker, and easier than seeing if it had really worked.  Finding a bio-marker for a disease speeds up ALL research aimed at curing that disease, and it attracts research money to that disease, since research there is less risky.

C-peptide is such a bio-marker for type-1 diabetes, but the pharma guys are always wishing that there were more.  It makes research, cheaper, quicker, less risky, and less unknown.  It is especially important that the government regulators agree to the use of the bio-marker.  If they do, then you can get government approval that the drug is useful, based on the bio-marker: quicker, cheaper, less risky, and you know approval will be granted.

Type-1 diabetes is diagnosed in about 78 thousand people per year in the US and the EU.  That is your core market for any new, honeymoon treatment.  There are about 2.1 million people who have type-1 diabetes in the US and the EU, so that is your market for any new treatment.  At the time of diagnosis about 10%-20% of the beta cells are still working.

Since I focus on clinical trials, for me, the most important presenters were those who had clinical trials underway:

Diamyd (makers of GAD-Alum) expects to have results from the Euopean phase-III trial "this Spring, before the Summer".  This study has three branches (people who did not get Diamyd, people who got two injections and people who got four injections).  It only includes people within three months of diagnosis.  Their phase-II study included people from longer after diagnosis, but the results were not so good for those people, so the phase-III study cut off eligibility at 3 months post diagnosis.

There is also have a small prevention trial in Sweden: 50 kids aged 4-7 who already tested positive to 2 or more different antibodies associated with type-1 diabetes, including the GAD antibody.  This study will take years, since they need to wait and see how many of the patients actually come down with type-1 diabetes.  This is an example of a the type of trial that doesn't have a bio-marker to speed it along, so it takes a long time to do even a small study.

Bayhill is expecting to be ready to start their phase-II study on BHT-3021 "very soon".  They expect to publish the full data from their phase-I study "later this year".  They have already published interim phase-I results last Summer, and I'm trying to get a copy of that presentation.  Although the FDA did not allow them to enroll children in their phase-I trial, they do expect their phase-II trial to be open to patients under 18 years old.  Obviously, this make it easier to complete enrollment.  It is also a general sign of safety that the FDA will let you include children in your trial.

One of the interesting things about their phase-I trial is that everyone gets treated eventually.  When you enroll, half the people get the drug, and half get a placebo.  Each group is then followed for a year, and the research is based on this treated vs. untreated comparison.  But then, the people who previously got the placebo are given the drug.  The advantage to this (I think) is easier recruitment.   Because, in the end, everyone gets the drug.  In classic trials, some people don't what to participate, because they are afraid that they will be in the placebo group, and go through all the work, and not get the treatment.  This design ensures that: everyone gets the treatment.

Finally, ToleRx (makers of Otelixizumab) were there, but my understanding is that their drug is not antigen specific and therefore would not be considered a vaccine.  In any case, I did ask, and was told, that they expected to publish results mid this year from their first phase-III trial.  I know their second phase-III trial is about a year behind their first, and both need to be done before they can submit to the FDA for market approval.

Most of the research discussed at this meeting had been done on animals, not people, and I discuss some of it (not all of it), below:

I was reminded (yet again) how enticing animal research is.  There were at least three or four researchers at this one symposium who had cured type-1 diabetes in NOD mice.  Some of them had cured it in several different animals models, and had performed all sources of confirmatory tests to make sure the animals were permanently cured of type-1 diabetes.  It would have been so easy to latch one to one of these guys as the one-true-cure that will lead us all to the promised land.  Again and again I reminded myself: don't get too excited about animal cures: they are always a long time and an uncertain result away from a human cure.

The most interesting to me was Parvus.  They have licensed Dr. Santamaria's research, which is based on attaching proteins to nanomolecules.  He published mice research in April 2010, which made quite a splash.  The Parvus guys hope to start human trials in 18-24 months.  If the FDA requires them to do monkey trials before human trials, then it will be on the longer side of the time frame.  If not, then closer to the shorter time. Previous news:

I particularly liked this paragraph:
According to Teodora Staeva, Ph.D., JDRF Program Director of Immune Therapies, a key finding from [this research] is that only the immune cells specifically focused on aggressively destroying beta cells (or, alternatively, regulating these cells) responded to the antigen-specific nanoparticle vaccine.  That means the treatment did not compromise the rest of the immune system -- a key consideration for the treatment to be safe and effective in an otherwise healthy person with Type 1 diabetes. 
Another interesting piece of animal research was Dr. Daniell, who I've blogged on before:
Dr. Daniell is still looking for funding to start a human trial, and his presentation sounded like a funding pitch. Basically, he has genetically modified lettuce to contain proinsulin, and because proinsulin is wrapped in cellulose, it is not digested, but enters the system of whoever ate it.  Why is this important?  Because giving proinsulin to NOD mice prevents the development of type-1 diabetes.  However, previous researchers have tried to slip proinsulin past the digestive system in people, but this has not succeeded in preventing or curing type-1 diabetes.  (As I discuss in more detail in my previous blog entry.)   I asked Dr. Daniell about this specifically, and his reply was simple.  In previous proinsulin clinical trials the methods used to get the proinsulin past the digestive system were not very good, and he believes the experiments failed because they were not good enough.  However, generating proinsulin, naturally wrapped in lettuce cellulose is a much better system.  This proinsluiln will get past the digestive system, and therefore will be successful even in the face of the previous failures.

A third interesting research area was Dr. Mannie.  He has a platform which he used to create an MS drug, which he has successfully tested in animals.  He hopes to get IND paperwork (required to start a human trial) done by 2015 for that drug.  So far, he has spent $700,000 on the project.  He is confidant that his platform will work for type-1 diabetes as well, if someone gave him a few $100,000s for that project.  So, think about this: would developing another cure for type-1 diabetes in NOD mice, be worth $500,000?  Should JDRF (or anyone else) put in that money now?  Or wait about 6-7 years and see what happens to the MS drug in people?

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.