Thursday, October 30, 2008

Faustman's Research (Part 1: History)

Before I describe the history of Faustman's research into curing type-1 diabetes, I wanted to say a little bit about my own history. In the past I have donated money to JDRF and also JLN; the first does not fund Dr. Faustman, the second does. I have never worked for either group. I'm not a diabetes researcher; I'm interested because my family is touched by type-1 diabetes. I've been tracking research into human trials to cure type-1 diabetes for years: both Dr. Faustman and others.

I have tried very hard to present facts and avoid opinions in this posting. However, that is sometimes hard to do. Especially with Dr. Faustman. When there are differences of opinion, I have tried to provide enough background so you can see why these differences exist. Different people evaluate research differently, so you may not agree with some of the things I say below.

This posting is still a work in progress. So if you see problems with it, or parts of it don't make sense, please tell me about them, so I can improve it.

As I write this one of the problems I have is how much detail to put in, and how many references. So I have tried to make it relatively short, but also added many end-notes (like [d1]) which contain more discussion of the points in the main text. You can just read the main text to get a nice flow, or read the end-notes to get a lot more information. References are shown like this: [r1]. So with that:

The Pre-History to Dr. Faustman's Work

In the 1990s immunology was the hot field in type-1 diabetes research. Several researchers were able to prevent diabetes in NOD mice [d1], by giving them a drug called CFA [d25] [r4,r5]. Others were able use CPA to stop the immune attack on insulin producing cells, even in mice with established type-1 diabetes [r26]. Obviously, a lot of excitement ensued, but also a problem: CFA is too toxic to give to people[r3]. However, BCG [d26] acts in people like CFA acts in mice. Also, BCG is known to be safe in people, and has been in some vaccines for decades, so it is cheap and well understood. Several studies were done on people given BCG specifically to try to prevent diabetes [r21,r24,r28], and other studies were done on people who had been given BCG for other reasons, such as part of a vaccine [r20,r22,r23]. Unfortunately, in both cases, no benefit was seen. CFA prevented type-1 diabetes in mice, but BCG did not prevent it in humans.

Dr. Faustman's Mice Research

At this time, Dr. Faustman set up an experiment to try to cure diabetes in mice who already had it [d4]. To do this she used a two treatment plan. The first treatment was CFA, which she hoped would stop the immune attack and the second was a transplant of spleen cells, which she hoped would help regrow insulin producing cells (previously destroyed by the immune attack). Spleen cells had previously [r4] been used to help prevent diabetes in NOD mice. This experiment on mice worked [r6,r2]: some were cured of type-1 diabetes! Again, a lot of excitement ensued.

The obvious next step was to run this CFA+spleen cell experiment in people. Faustman asked for research funding from the JDRF (Juvenile Diabetes Research Foundation) for exactly that human trial [d17, r27]. JDRF rejected this request. The exact reason why is not known [d5].

When scientists test two treatments together, there are several different designs they can use: Two groups of mice (one doesn't get either treatment, one gets both). Or three groups of mice (one gets no treatment, one gets one treatment, and the third gets both treatments). Or four groups of mice (no treatment, one treatment, other treatment, both treatments). The last design is the best, because it tells you exactly how each treatment interacts to give you the final results. The first design is the worst, because it doesn't tell you if one treatment is useless or not. Using a 3 or 4 group design is especially important if one of the treatments is known to affect the outcome by itself. Dr. Faustman used the first design in her mice study, which is the weakest design. Furthermore, it was an especially bad choice because it was known that CFA impacted type-1 diabetes in mice when used alone. So with her experimental design, it would be impossible to tell if the CFA alown impacted the mice's type-1 diabetes, or if the CFA and spleen cells together cause the impact. [d6]

Dr. Faustman was funded by JLN (Join Lee Now). However, it took them several years to raise the money and give it to Dr. Faustman. During that time no clinical trials were done.

Other Mice Research

However, with all the excitement (and the funding controversy) generated by Dr. Faustman's research, several other groups wanted to replicate it. In the end, three different groups attempted a similar experiment, but with a different design: three groups of NOD mice. One group got nothing, one group got CFA only, and the final group got CFA and the spleen cells. Furthermore, each group tested seperately if the transplanted spleen cells were contributing insulin to the mice, and each used a different scientific method to check this. JDRF did fund these experiments. All three of these researchers came up with the same basic results: many mice given CFA maintained normal blood glucose control, and many mice given CFA and spleen cells also maintained normal blood glucose, but at the same rate as the mice given CFA alone [r7], and the spleen cells did nothing to help maintain normal blood glucose levels [r8]. In all three cases they found that the transplanted spleen cells were not the source of the new insulin producing cells. So that is a second way to verify that the transplanted spleen cells were not functional to controlling the mice's BG levels. A study by Mezey [r25] is sometime sited as independant confirmation of the importance of spleen cells, but that is a mistake. This study was not independent, and did not show any insulin production from spleen cells [d7]

Understanding what these studies found, and what it means, is absolutely central to understanding why some people think Dr. Faustman is brilliant, while others think she has already failed.

The studies's conclusion can be described in two ways:
1. Dr. Faustman was right about CFA causing mice to have normal blood glucose control.
2. Dr. Faustman was wrong about the spleen cells being part of this process.

Dr. Faustman and her supporters describe what happened as "Three independent groups confirmed Dr. Faustman's results", emphasizing point 1. However it is equally true to say that all three independent groups, found that Faustman was wrong about spleen cells. However, from the point of view of curing type-1 diabetes, point 2 is much more important than point 1. CFA was already known to cure and prevent diabetes in mice, and this had not led to a human cure or prevention. Also, at the time of the original experiment, Faustman thought the spleen cells were the more important part of her experiment, and that was the part that later turned out to be wrong [d18, r2, r27]

The success of point 1 leads to a path of using CFA to cure diabetes, but that path had already been traveled down, and found not to work on people [r20,r21,r22,r23,r24]. But see also the discussion in [d19]. While point 2 closes off the new path that Faustman was hoping to follow to a human cure for diabetes. Basically, point 2 puts an end to the human trials that Faustman had asked JDRF (and later JLN) to fund.

Faustman has found that CFA cures type-1 diabetes in NOD mice, but that was already known [r26]. But since BCG has already been tested in humans, that is a less interesting finding. However, one guy used this information in a unique way.

Dr. Orban's Human Research

Another researcher, Dr. Orban, looked at all the mice research above, and tried a different approach from Dr. Faustman. Since CFA worked in mice, but BCG did not work in people, he tried curing type-1 diabetes with a different drug which was similar to CFA, but that you could give people and was not BCG [r10]. His work is based on Dr. Faustman's foundation, but he went in a different direction than she did. Orban's phase-I trial ended a year ago. I haven't seen the published results; last I heard they were still analysing the data. [d3]

Dr. Faustman's Human Research

Faustman looked at her own research and the other similar research, and tried this approach: she assumed that previous BCG human trials has failed because either they were not using the right dosing, or they did not have a sensitive enough machine to measure small improvements to the patient's immune response.

So, she stopped working with spleen cells and instead put the money from JLN into a super sensitive blood tester which could identify the difference between good immune cells and bad ones [d8]. She hoped that this machine would solve both problems with previous BCG experiments: it would detect tiny improvements in immune response, and this information could be used to find the right dosing.

Dr. Faustman's actions at this point show that JDRF was correct in not funding her original human trials. Once Faustman got the money to do trials, she herself did not do the trial that she had asked JDRF to fund. Instead she did a completely different one. The earlier one used BCG and spleen cells to cure diabetes in people. The current trial uses BCG alone to see if it affects the immune system (even a tiny amount) so that it can be used in future treatments. No cure is attempted in the current clinical trial. [d9]

This trial started in early 2008 and is expected to run to mid 2009. Any diabetic can take part: it is not limited to honeymoon diabetics. Basically, diabetics are treated with BCG, the same dose in certain vaccines, and then blood is taken over the next year to see how good and bad T-cells [d10] are affected.

The Future of Faustman's Work

The future of Dr. Faustman's work is an interesting question. Usually, getting funding for Phase-II clinical trials is easier than getting funding for Phase-I trials, even though you need more money. That's because, most of the time, you cured some people in your phase-I trial, so you can take that data to a funding source, and get more money.

However, Dr. Faustman's phase-I study isn't going to cure anyone [r15], and isn't even going to improve BG or A1C levels in anyone. The only evidence for improvement will be from Faustman's custom made, one of a kind, super sensitive machine [d12]. And no one will be able to double check her results. So even though the phase-I trial is still underway, it is clear that it will not provide strong information to help get funding for a phase-II trial. There is an alternate plan which would be to do a second phase-I trial, this one aimed at curing diabetes in people. If that panned out, then getting funding for a phase-II trial would be straight forward, but it would delay things.

If the current human trials are successful (meaning that BCG stops the immune attack), then Faustman might still have trouble getting funding for the next round of human trials (no matter if they are phase-I or phase-II). Remember, the last time she did a big experiment (the CFA+spleen test in mice) she was wrong about a major part of the results (the spleen part). This was discovered when other groups did the same experiment. However, no other group can replicate her current experiment, because no one else has her custom made, multi-million dollar equipment.

JLN has already announced that they will not fundraise for Faustman's next round of human trials. This is an important development. JLN knows more about Faustman's work than anybody except Faustman herself. They have been strong supporters since at least 2003, and have already sunk in millions of dollars. For them to cut her off at this point should be very troubling. Faustman and her supporters point out -- correctly -- that JLN said years ago that they would not fund a phase-II trial, so this is not new information. [d15] However it does seem unusual that Faustman's biggest supporters would cut off funding right at the end of a supposedly successful clinical trial. It does not send a good message to the rest of the funding community.

Level of Excitement

In a private email, someone asked me "Why aren't you excited about Dr. Faustman's research?" In a sense, this entire post answer that question. A more specific answer to this question is that I am excited about all research aimed at curing type-1 diabetes. I have faith that science will one day cure type-1 diabetes, and I am excited about all research progress in this area. The difference between me and most Dr. Faustman supporters, is that I look at her work as being much like any other research. She's at a very early stage of phase-I human trials, so I group her with all the other phase-Is out there. And behind the phase-IIs and phase-IIIs.

This is quite different than most of her supporters, who view her work as uniquely wonderful; much different and much better than most of the other work out there. To my mind, that is unwarranted. It is not that I'm unexcited about her work. It's that I have the same level of excitement for her work, as for any other research which is (a) in very early phase-I experiment, and (b) is testing a treatment which has previously failed -- in six previous experiments [r20,r21,r22,r23,r24,r28] -- when tested on people. Dr. Faustman supporters claim that these studies were quite different from Dr. Faustman's current work. This is further discussed in [d19].

Extra Discussion

Notes that start with a 'd' contain more discussion on an issue.

[d1] NOD mice are "non-obese diabetic mice". They are research mice specially bred to have diabetes, and widely used in type-1 diabetes experiments. All (or almost all) of these mice will get type-1 diabetes as they grow up, so the are good "animal models" to try to prevent or cure type-1 diabetes. However, it is also true that many things which have prevented or cured diabetes in NOD mice, have failed to work in people. NOD mice are generally considered the best animal model available, but it is still unclear how good they are.

[d] Any experiment involves risk, but transplants generally involve more risk than drugs. So a human trial at this point would have subjected the people to the extra risk of transplantation which could have been eliminated by doing a better mice experiment ahead of time. The better design used by the non-Faustman researchers proved conclusively that the spleen cell injection was worthless to the cure.

[d3] My experience is that if you need to spend a year analysing your data, the results are not going to be good. Good results are more obvious and are found faster.

[d4] Some people have claimed that Faustman was the first to stop the immune attack on insulin producing cells in mice with established type-1 diabetes, but this is incorrect. Faustman's own 2001 paper [r6] says that such cures were rare, but did exist, and she provided references to at least two previous such cures. These are listed here under [r26]

[d5] JDRF has a two phase system for evaluating requests for funding. One phase involves a review committee composed of researchers, scientists, and doctors expert in type-1 diabetes. The second phase involves a separate review committee composed of laypeople, mostly people who have type-1 diabetes, and the parents of children with type-1 diabetes. Approval of both committees is required to get money from JDRF.

[d6] I want to stress that I have no special insight into why Faustman's request for money was rejected. This discussion is based on my opinions about the research, not anything that I know happened inside of JDRF. I have no visibility inside JDRF or it's review process.

[d7] This study has never been published (or peer reviewed) on it's own. Rather, it was published as a "technical comment". The study was not independant because Denise Faustman was an author. In fact, she was the first listed author, and was listed as one of two authors to whom correspondence should be sent. As for what was found, consider their conclusion: "We show that islet regeneration predominately originates from endogenous cells but that introduced spleen cells can also contribute to islet recovery." They not claiming that the spleen cells did contribute to the cure, mearly that they could have contributed to the cure. They are also up front about the fact that any contribution from the spleen cells is -- at most -- a minor effect. However, if you read the entire report, they note they never looked for insulin produced by the spleen cells. Their optimism that the spleen cells are producing insulin is based on the fact that they exist at all. They claim to have measured the existance of spleen cells: not that they were active in any way.

[d8] The immune cells in question were T-cells that were mis-programmed to attack the beta cells in the pancrease which generate insulin. T-cells that attack beta cells are bad, and cause diabetes. Those that don't are good, and attack disease cells, as they should.

[d9] Symbolicly, the experiment Faustman asked JDRF to fund was this:
BCG + spleen cells = normal blood glucose control in patient
however the experiment that she actually ran with JLN's funding was this:
BCG = slight changes in patient's immunology
So there is a major difference in inputs and a major differences in results!

[d10] T-cells are part of the immune system. Bad T-cells cause the immune system to attack the body's own insulin producing cells, which starts type-1 diabetes.

[d12] A blogger who is very supportive of Faustman visted her lab and reported in his blog ( that:
They can't do trials in multiple centers because the equipment they've developed for testing results is not portable. In one case they moved a piece of equipment across the lab. It took 9 MONTHS to recalibrate it and get it working again.

We saw the equipment (no photos allowed). It's complex and large. The size of a full sized fridge on its side. To take a blood sample and extract the T cells takes an entire day.
(Bold and all-caps for "months" are in the original quote.)

[d15] At the time Faustman was planning a conventional phase-I clinical trial and everyone thought that she would have plenty of data showing lower BG numbers, lower A1C, and/or lower insulin requirements, so none of this would have been a problem. Faustman's phase-II research would have been funded based on her success with phase-I. However, when Faustman changed the type of experiment she was doing, it ment that her phase-I would not generate that kind of positive data.

[d16] Here is a quote from a Iacocca Foundation letter about stopping their funding of Faustman:
Now that we have reached this important milestone we have made the decision to stop actively fundraising through what has become The Iacocca Foundation/ for type 1 diabetes - a public charity. We will be accepting donations through December 31, 2008 when we will make our final committed payment to Dr. Faustman.

As we look to the future we remain committed as always to being part of a cure for type 1diabetes. The Iacocca family will continue to fund the brightest and best who are dedicated to ending this disease. And we hope you will turn your time and resources to continue to support Dr. Faustman directly as she and her team at Massachusetts General Hospital prepare to launch a campaign to raise funding for the Phase II BCG clinical trials.
[d17] Some people have claimed that Faustman never intended to do a CFA and spleen cell experiment. This is contradicted both by information available at the time, and by common sense. For example, the Mass General Hospital where Faustman worked (and still works) issues a press release saying specifically that "Mass. General's Diabetes Center has received approval from the US Food and Drug Administration to try the techniques pioneered by Faustman in humans. " [r27] So there is no doubt that she was planning on doing the same experiment in her human trial. From a common sense point of view, it makes no sense to run one test on mice, and then run a different test on people (unless the mice test is too dangerious to run on people). The whole point of doing mice studies to to prepare for human studies. So it only makes sense to run the same experiment in mice, that you are planning on people, if you possibly can.

[d18] For example, in the abstract to her own 2003 paper [r2] Faustman lists the spleen cells first, and CFA second: "Treatment of NOD mice with end-stage disease by injection of donor splenocytes and complete Freund's adjuvant..." In that same abstract, CFA is mentioned only once, while the spleen cells are mentioned three times. The official press release from MGH (where Faustman did the experiments) [r27] is even more clear: CFA is not mentioned at all, while the spleen cells are mentioned many times, and are the focus of the reporting: "Massachusetts General Hospital researchers have harnessed newly discovered cells from an unexpected source, the spleen, to cure juvenile diabetes in mice, a surprising breakthrough that could soon be tested in local patients and open a new chapter in diabetes research."

[d19] Naturually, Faustman's supporters object to comparisons between her current experiment [r15] which uses BCG, and the six previous experiments [r20,r21,r22,r23,r24] which used BCG and found that it did not effect type-1 diabetes. One paper [r28] found a slight improvement, but it was not statistically significant in a study of only 25 people. For example "kumquat79" says: "Those studies cited are not identical to the treatment Dr. Faustman is working on. Those were one-time vaccinations". While it is true that Faustman's experiment is slightly different than the six previous experiments (after all, experiments are almost never repeated exactly identically: what would be the point?) However, the experiments are very similar. Faustman's experiment uses 2 doses of BCG. The others either use 1 dose [r21,r22,r24,r28], or an unknown number, but probably just 1 dose [r20,r23]. Obviously the difference between 1 dose doing nothing, and 2 doses being a cure for type-1 seems a little farfetched! Furthermore, Faustman's has never previously claimed that the differences between her current 2 dose regimen, and the 1 dose previously given is important; her previous mice studies used only a single dose of adjuvant.

[d20] Honeymoon refers to the time period when a type-1 diabetic is first diagnosed, when they are are still generating some of their own insulin. This generally lasts for first few months after diagnosis. Researches use many different terms for "non-honeymoon", such as "established", "long standing", etc. A mouse with established diabetes is one that is not in the honeymoon phase.

[d25] CFA is Complete Freund's Adjuvant which is an adjuvant which can only be used in animals. In general, adjuvant increase immunological response. More information on adjuvants in general is here:

[d26] BCG is Bacille Calmette-Guerin which is an adjuvant which can be used in people. More information on adjuvants in general is here:
This adjuvant has been commonly used in TB immunizations, and others.

References and Sources

[r2] This is Faustman's 2003 publication:;302/5648/1223

[r3] For example, this book: contains this quote: "(CFA) is quite toxic and cannot be used in humans."
A general safety data sheet on CFA can be found here: Including this quote: "The undesirable side effects attributed to CFA use include increased pain and suffering and morbidity in inoculated test animals".

[r4] This study, published in 1992, researched both CFA and spleen cells in preventing type-1 diabetes (but not curing it):

[r5] This study was published in 1990:

[r6] This is Faustman's 2001 publication:

[r7] The best description of these studies and their results that I have found, is here:

[r8] Here are the three experiments that came to the same conclusion (no spleen cells required) in 2006:

[r10] Orban's experiment:

[r15] The Nathan/Faustman human trials are described here:


This study treated honeymoon diabetics with BCG and saw no improvement, compared with an untreated group. From the abstract: "Vaccination with BCG at the time of onset of type 1 diabetes does not increase the remission rate or preserve beta-cell". The full paper is here:

Neonatel vaccination with BCG has no good effect on type-1 diabetes rates: " The cumulative risks for developing islet autoantibodies by age 2 or 5 years were unaffected by BCG vaccination in the first 3 months of life" and "Progression to type 1 diabetes in BCG-vaccinated autoantibody-positive children was significantly faster than in nonvaccinated children" and " No evidence was found that BCG vaccination could prevent against ß-cell–damaging processes leading to type 1 diabetes in genetically at-risk children. The findings do not suggest that BCG vaccination will affect the overall incidence of type 1 diabetes"

Retrospective study on people who had been vaccinated with BCG.
"However, as a whole, results from these analyses fail to support a protective role of BCG vaccination against juvenile-onset IDDM."

This study was published in 1998:
"BCG vaccination in children who have been recently diagnosed with IDDM does not affect the progressive decline in C-peptide levels or alter the clinical course of the disease."

[r25] The Mezey/Faustman paper is here
See the discussion under [d7].

[r26] is a report from 1992 showing that CPA stopped the immune attack in mice with established type-1 diabetes. These same guys later went on to confirm their finding with BCG, previewing Faustman's research by five years:


[r28] Abstract is here:
Full paper: (in Korean)
Here is their conclusion from their abstract, in it's entirety:
BCG vaccine is safe and convenient to use, however, a large study is warranted for the use of BCG as a therapy of type 1 DM.
Note that there is no claim of any improvement to the patient. Here is the key sentence from their results section:
During follow-up, there was no significant difference in fasting and postprandial 2 hour C-peptides.
They then go on to list various good things they did see, in particular there were slight differences in C-peptide and insulin usages, and temporary remission in two patients, but these were not statistically significant. (And remember: this study was of honeymoon diabetics.)

Their paper is in Korean, which I can not read, however the tables are presented in English, and each table says very clearly "The differences between the two groups were not significant."

Sunday, October 19, 2008

JDRF Funding of Cure Research (phases II and III)

I was talking to someone today, and when I told him that I track cure research for type-1 diabetes that was being tested in humans (ie. "in clinical trails"). He asked me a question that really surprised me. He asked "Do you follow all cure research, or just the stuff funded by JDRF?" My answer was "I follow all research; however there's not much difference, because over their years of development, JDRF has funded almost all of the possible cures being tested in people today."

That's a strong statement, and I made it off the top of my head, without my notes in front of me. But in the evening, I double checked, and it is pretty impressive. JDRF has funded every cure currently in phase III human trials and every cure currently in phase II human trials. I haven't had time to check on all the phase I trials, yet. But still, 100% funding rate for the two phases of research closest to a cure is something to be proud of.

So lets look at the data.

Currently in phase III clinical trials are Diamyd (GAD65/DiamydT), ToleRx (TDX4), and Andromedia (P227). JDRF paid for early academic research on all three of these cures, and is directly funding ToleRx's phase III trial.

Currently in phase II clinical trials are Osiris (PROCHYMAL), Macrogenics (teplizumab), Transition Therapeutics (E1/G1), Pescovitz (Rituximab) and Gitelman (Thymoglobulin).
JDRF is helping to fund the clinical trials for every one of these, and funded some of the academic research on E1/G1. (They might have funded earlier research on Thymoglobulin and Teplizumab as well: I don't follow animal testing very closely, so am not sure where that funding came from.)

Currently there are about 10 phase I clinical trails, and I have not checked to see how many were funded by JDRF and how many were not. So that information will need to wait for a future posting. I can see that at least one was not funded (LCT), and at least two were (the Haller/Schatz work at Univ. of FL and Trucco's work at Pittsburg).

Some notes to remember:
1. All of the phase II and III cures are targeted at new-onset diabetics (we would call them "honeymoon" cures. They only work on people diagnosed a few weeks ago. There are a couple of non-honeymoon cures in phase I testing right now, but none farther along.
2. Some people think of Macrogenics's current trials as phase III rather than phase II. Officially, they call them "Phase II/III" so you can take your pick.
3. I consider JDRF to have helped fund the cure if they funded any part of the academic or commercial research that led directly to the current clinical trial. For example, I don't think that Diamyd the company has gotten any money from JDRF, but JDRF did fund some of the early GAD65 research which led directly to their human trial.
4. The Andromedia cure, which is called P227, has not made any forward progress in years, and does not look very promising right now, but it is in phase III trials, so I include it above.
5. Faustman considers her current work to be phase I, so I will discuss it when I cover all the phase I trails.

If you have any questions on this, or you think I've missed something, please don't hesitate to ask the question, or call out what you think I missed.

Joshua Levy