Sunday, July 19, 2015

Artificial Pancreas Updates From ADA (July 2015)

In my opinion, artificial pancreas (AP) technology was the single most important technology of the American Diabetes Association Scientific Sessions this year.  There were many papers, posters, and presentations on artificial pancreases and related technologies (such as continuous glucose monitoring, stable glucagon, etc.)  Nothing generated more Twitter buzz.

This blog posting attempts to summarize a little of the research presented.  There was way too much for me to cover it all, and so this posting gives light coverage to the most important news.

My overall summary is that we are no longer talking about if there will be an AP.  The question is: when there will be an AP, and how many APs, and how many different types of technology they will use.  I continue to be heartened by:
(a) the progress made by AP research projects over the last few months (and years)
(b) the large number of new companies entering the field
(c) the product roadmap announced by Medtronic as they move their existing partial AP to a full AP.

The AP world is more and more driven by commercial considerations, rather than research considerations, and that is a strong signal that we are close to general availability.  (Close meaning years, but not too many of them.)

This tweet summarizes my thoughts, as well:
“Closing the loop is no longer a mystery; it’s not a puzzle. We just have to do it.”

Second Bi-Hormonal Artificial Pancreas In Development

I recently found out about a second bihormonal AP  (in addition to Dr.  Ed Damiano's).  These researchers are in The Netherlands, and you can read more about it here:

It's a little hard for me to figure out exactly where they are, and I have not found any articles in scientific journals.  However, it appears that earlier this year they ran a two day test on one person, and got quite good results.  The average BG was 125.  Based on that they are hoping to run two clinical trials (called APPEL 4 and APPEL 5) later this year.  They hope those trials will used as the basis for European "CE" approval.

Maybe a Third?

They haven't started recruiting yet, but this looks like a phase-II trial for yet a third bi-hormonal artificial pancreas.  I'll blog more fully if they start recruiting:

Back To Ed Damiano's Bi-Hormonal Artificial Pancreas

Not from ADA, but from another conference, comes this summary of Ed Damino's current status:

But if you want a summary in 140 characters, here it is, from
Damiano says #BionicPancreas showing ave BGs of 135-142, which would equate to 6.5% A1C.
[What's not to like!]

Plus there was this good news tweet:
Ed Damiano says there is room temperature stable glucagon that stays good for up to a year. Needs FDA approval.

Medtronic's 670G

The Medtronic 670G, as announced, will be an "all but meals" style, single hormone artificial pancreas, available in the United States in April 2017.  As far as I know it is the most powerful AP with a clear commercial delivery date.  For comparison, the 640G is already available in Europe, but does not prevent "highs," only "lows".

diaTribe interviewed a patient who has used the 670G as part of a clinical trial:

This is Medtronic's press release summary of results from a 640G study, a 670G study, and some other research they presented at ADA:
The current plan is for the 640G to be available next year, and the 670G the year after.

This news resulted in the following tweets (and many more, of course):

More General AP News

The following, more cautionary tweets report on other aspects of AP research:

During artificial pancreas study, the AP couldn't be used 1 in 3 nights because of tech issues (sensors) or family/health issues.[But I don't know which AP this tweet is about!]

CDE Laurel Messer: Still enough tech issues exist with CGM sensors to impact hoe [sic, should be "how"] closed loop is used in kids and young adults.
Predictive pump technology significantly reduced nighttime lows in kids in almost 1000 night study. [But again: I don't know what AP this refers to!]

The following link goes to a summary of one of the ADA sessions:

Personal testimonial (with graph) of how well an AP works, but I don't know which one!  Anyone recognize the screen?

Bigfoot Explodes Big  (In a Good Way!)

Bigfoot is a one year old company.   They had a very successful ADA and generated a lot of "buzz", which this tweet tries to summarize.
@BigfootBiomed acquires Asante pumps! Goal to enter pivotal trial of full #artificialpancreas system late 2016

A "pivotal" trial is industry-speak for a phase-III clinical trial.  What they are saying is that they hope to start phase-III trials of an AP next year.  That would put them shoulder to shoulder with Medtronic and front running bihormonal research projects.  Even if they are overly optimistic, and they actually start those tests in 2017, they are still not that far away from a commercial AP.

But even more promising (to me) is that Bigfoot feels like a tech start up, not a medical device company. It feels like Silicon Valley rather than Washington D.C.  I can't tell if that's a marketing strategy or the truth, but if you look at the founders, look at the tactics (reuse the failed carcass of another project), and the strategy: it really does look like the way things happen "out here".

When a company like that can successfully create an AP, it means that APs are technology gizmos, and not medical devices, and technology products can improve much more quickly than medical products.  I'm sure the FDA will have something to say about it, but even if Bigfoot is just a 20% step in the direction of Silicon Valley, that's going to "light a fire underneath" some medical device companies.

The following link goes to one diaTribe story, but that story has links to other stories, and you can get the whole saga of how big they've grown, how quickly, and how important that might be:

TypeZero: Yet Another New Company Focused On A Single Hormone AP 

And if that were not enough, TypeZero Technologies is yet another new company trying to create a single hormone AP "from scratch".  DiabetesMine has a summary article:

This is based on AP research done at the University of Virginia which I have reported on before.


There was some debate at ADA about which was better: single hormone artificial pancreases, or bihormonal artificial pancreases.  My first reaction to this debate was this: who cares?  Either one is so much better than what we have now, let the people who like insulin only APs develop those, and let the people who like insulin and glucagon APs develop those, and let the market decide, or maybe let the market decide that there is room for more than one AP technology.  I mean some people have cable while others have satellite, why not the same with APs?  Have we learned nothing from the death of communism?  Let the market decide, as fed by the strongest proponents of each system.

My second reaction to this debate was that it can never be resolved by research. The data we have now is pretty clear: single hormone APs result in average BG numbers about 10 points higher than dual hormone APs  (for example mid 130s vs mid 140s).  However, dual hormone APs are more complex to manufacture, more complex to operate, and involve the tiny risk of long term, very low dose glucagon (and that risk is unknown, and will remain unknown for years).  So some people are going to say dual hormone is 10 points better than single hormone, end of story.  But others will say, mucking about with refilling two reservoirs is just not worth 10 points.  And neither of these opinions is right or wrong, they are just different.  And always will be.

However, even if research cannot answer a question, it can still inform the discussion, and there have been some papers directly comparing the two systems. The following tweet is one summary, but the links to the abstracts below contain more data.

Haidar: Both single and dual hormone artificial pancreases better than pump for BG control at night, dual better for reducing hypos

Joshua Levy
publicjoshualevy at gmail dot com
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My daughter has type-1 diabetes and participates in clinical trials, which might be discussed here. My blog contains a more complete non-conflict of interest statement. Thanks to everyone who helps with the blog.

Saturday, July 4, 2015

News from ADA 2015

The American Diabetes Association's annual conference is the largest scientific gathering covering diabetes in the United States.  Even though most of the conference is aimed at type-2 diabetes, the part covering type-1 diabetes is still overwhelming.  About 18,500 people attended this year.

I was not able to go this year (it was in Boston), so I tried an experiment.  I closely followed tweets and news coverage of the convention.  I thought of myself as a wire service and the tweeters/bloggers/reporters at the conference as my freelancers.  I was reading about 1000 tweets a day from the following sources: #2015ADA, #ADA2015, @diaTribeNews, @DiabetesMine, @sarhoward, @AmyDBMine, and @kellyclose.

I'm going to summarize these in two postings.  This posting has four sections: an overview, research in people aimed at a cure, research which has cured mice, and non-cure related research. The next posting will cover artificial pancreas research. There was a lot.

ADA includes a sea of expert opinions, but also the results of many experiments.  So it is especially important to remember these words by Robert H. Mathies  (Often incorrectly attributed to von Braun):

"One experiment is worth a thousand expert opinions"


You can read the abstract to every poster session at the link below, and most of them have a PDF of the poster itself:
Although the posters are easy to search, I cannot set up a link to one once I find it.  So I have put in the poster number (like 1234-P  or 42-LB).  Go to the link above and search for the number to find the abstract and the poster itself, when it's on line.

The following search page can get you to abstracts for many presentations, but I found it very hard to use:!/3699

The link below is a summary of the conference by Insulin Nation.  Some of it is reporter-insider snark, but other points are quite good.  The useful and the useless are mixed together:
P.S. The answer to 41 is "brown fat".

The link below is someone else's summary of the conference.  I completely disagree with the headline, but the section on type-1 research is a good summary.  You'll notice that only one of the possible cures they mention are in human trials now.  (Unless you consider AP or transplantation to be cures.)

Below is a link to Sarah Howard's summary.  She follows possible environmental causes for both types of diabetes, so she focuses on different parts of the conference than I would:

A big theme of the conference seemed to be treating patients rather than numbers, and generally emphasizing holistic care, and looking at more complex (and realistic) measures of health, than single numbers.  For me, this was typified by the following tweet which I saw repeated several times:  "A three-month average cannot tell the full story."  I think the tweet originated with @kellyclose (diaTribe), but I'm not sure.

Research On Potential Cures In Human Trials

I expect to blog on each of these trials in the coming months.

Fastuman Starts a Phase-II Trial Off BCG
Here's the press release:

Obviously, this is big news, and I'll blog about it in the next few months. But in the meantime, you can read my last blog on her BCG research:
The new news is that the trial is actually recruiting patients now.  Any updated details will be in my new blog.

ViaCyte's Phase-I Trial is 1/10th Enrolled
JDRF sent out this update on ViaCyte's Phase-I clinical trial:
There are as many as 4 people in SoCal currently testing @ViaCyte's experimental devices. That's progress.
The trial plans to enroll 40 people, so it sounds like they are 1/10th of the way there (depending on exactly what "as many as" really means).

1102-P: Immune Modulation of Stem Cell Educator Therapy in Caucasian Type 1 Diabetic Subjects
This poster contained information from the Phase-II trial of the Stem Cell Educator currently running in Spain.  I will blog more on this later, but for me, this poster had a "split personality".  On the one hand it was reporting on good changes to the immune system in people who got the treatment.  More good immune cells; less bad ones.  It looks like good news: the Educator is changing the immune system for the better.   On the other hand, there is none of the patient-focused data that would show progress to curing type-1 diabetes.  No data on C-peptide production; no improvements to A1c; no changes reported in insulin use.  (And the phase-I trial did have this kind of data.  So I'm not sure if this poster just focused on the immunology details, and a future paper will include C-peptide data, or if the C-peptide data in this second trial was not worth reporting.)

1099-P and 102-LB: Effect of Cell-Free Mesenchymal Stem Cells Microvesicles (MVs) and Exosomes Therapy on ß-Cell Mass in Type 1 Diabetes Mellitus (T1DM)
As far as I can tell, these two posters were identical, except that the LB poster only listed one author, and included the clinical trial registration number.
Patients were given stem cells (source not described), and compared to a control group.  Results were good:  HbA1c (6.67 ± 0.321 at 12 weeks vs. 8.245 ± 0.72 at baseline), fasting C-peptide levels (1.095 ± 0.215 at 12 weeks vs. 0.245 ± 0.069 at baseline) and C-peptide response following a 75-g oral glucose tolerance test (1.817 ± 0.27 at 12 weeks vs. 0.504 ± 0.065 at baseline).   To me, this looks like it is well worth a follow up.

Transplantation Work
So far with Edmonton protocol of islet Tx, 58% insulin free (need immunosupression), 92% have C-peptide after 7 years.  But those people will need to take immunosuppressives for the rest of their life.

Cured in Mice

Random quote from Richard Asher: "despair is best treated with hope not dope"

Poster 1812-P:
A new peptide (small protein like chemical), called KGYY15 is a honeymoon cure of type-1 diabetes in NOD mice.

Poster 1804-P:
Amylin Induces CD4+CD25+Foxp3+ Regulatory T Cells
Amalyin is a hormone given to type-2 diabetics, but there is some research to show that it would also help type-1 diabetics maintain better BG and A1C numbers.  This poster reports that it also increases the number of a specific type of helpful T cell.  This work is related to human trials, because at least two groups are "growing out" CD4+CD25+ T cells in the hopes that they will be a honeymoon cure to type-1 diabetes. Obviously, if Amylin has the same effect in people as these researchers found in mice, and if the researchers "growing out" these cells in people get a good result, that's good all around.

Using Th17 to prevent type-1 diabetes in mice:

Poster 1818-P: Local Expression of CCL21 in Pancreatic Islets Prevents Autoimmune Diabetes in Mice and Is Associated with Beta Cell Antigen-Expressing Lymphoid Stromal Cells
The poster was not on line, but the abstract was.  CCL21 (another small protein) prevented type-1 diabetes in NOD mice.

Poster 1826-P:
Prevention or Early Cure of Type 1 Diabetes by Intranasal Administration of Gliadin in NOD Mice
The poster was not on line, but the abstract was.

Poster 218-LB:
Nasal Administration of Novel Insulin Degrading Enzyme Inhibitor Ameliorates Autoimmunity in [NOD mice]
The poster was not on line, but the abstract was.

For those keeping track, that is six mouse cures presented at one scientific meeting.  For me, the interesting point will be: how many eventually get into human trials.  One?  None?  Only time will tell.

Non-Cure Research, But Interesting To Me

FDA Bullshit
This tweet really angered me:
Stayce Beck of @US_FDA calls out companies for not including kids in pediatric studies, says it's opposite of what agency wants.
This is completely two-faced, as far as I'm concerned.  Every researcher I interact with wants to have more kids involved in their studies.  Especially for honeymoon research, there are just so many more kids around.  But they can't because the FDA has specific regulations which prevent it!  It's the FDA who is creating and enforcing unreasonable limitations, and to turn around and "blame the victim" is scummy.  And the FDA's policies on kids in research does victimize researchers (as well as children) because it really slows down research and at the same time limits approved options for treating kids.

(I don't think I've ever sworn on this blog before, but this quote -- if true, and if really said by an FDA employee -- really deserves it.   I'm proud to say:  I call bullshit on that!)

Type-1 vs. LADA
There were several posters and papers which reported on differences between people diagnosed with type-1 as kids, and those diagnosed with it as adults.  Some researchers consider these to be the same disease with slightly different natural histories (sort of like chickenpox vs. shingles) while others consider these two to be two different diseases (with different causes or "etiologies"). All of the following posters described differences in early onset type-1 diabetes and late onset type-1 diabetes:
Poster 1801-P: Heterogeneity in Type 1 Diabetics Is Defined by Contrasting C-Peptide Declines, Autoreactive T Cell Burdens, and Metabolomic Differences
Poster 1819-P: IGRP-specific CD4+ T Cell Response Is Distinct between Adult-Onset and Juvenile-Onset Type 1 Diabetes Patients
Poster 212-LB: Anti-paralemmin 2 Antibody as a Novel Antibody for Latent Autoimmune Diabetes in Adults
Poster 1723-P: Determinants and Prognosis of Early- vs. Late-Onset Islet Autoimmunity

Poster 1737-P:
Ecological Study between the Incidence of Type 1 Diabetes and Geochemical Data in Sardinia: Negative Correlation with Zinc and Copper
Sardina has a high rate of type-1 diabetes, and the rate is not the same throughout the island, so these researchers tried to correlate various heavy metals in the environment to the higher type-1 rates in the same areas.  Higher levels of Zinc and Copper were found to protect people from type-1 (ie. higher levels of these metals correlated with lower levels of type-1), and no correlation was found for the rest of the metals they researched.  Here are the metals they checked: As, Be, Cd, Co, Cr, Cu, Mn, Ni, Pb, Sb, Se, Sn, Th, Tl, U, V, Zn.

Unexpected Stability of Type 1 Diabetes Incidence in a U.S. Cohort, 1994-2010
Basically, they found a type-1 rate of about 1 in a 1000 (lower than expected) and found that this number was going up before 2002, but going down after that.  That's not the common wisdom.  Most researchers think the type-1 rate is steadily going up.

For the opposite viewpoint: 1735-P:
The Incidence of Type 1 Diabetes Mellitus in Romanian Children Aged 0-14 Years Increased Constantly
(No poster, but a very nice graph in the abstract.)  This is the opposite finding as the previous poster, but this is the conventional wisdom about what is happening.  The two posters covered different geographical areas.

Dance-501 Inhaled Human Insulin
Research results from another inhaled insulin.  Phase-I trial, 24 people, type-2 diabetics.  Affrezza might have competition someday, and that can only be good, in terms of more choices for type-1s.

Big claim by Dr. Dandona: “Predict w/ this 3x therapy (insulin, GLP-1, SGLT-2) we can get at least 50% of #T1D patients [below] 6% A1c"
If he is right (and that is a big "if"!) this will be a big change in the treatment of type-1 patients. Right now, insulin-only treatment is normal, and A1Cs are rarely this low.  GLP-1 is widely used in type-2, but not widely used in type-1.  In my opinion, the results of the GLP-1 in type-1 studies that we do have, show some improvement, but not a lot. SGLT-2 is even farther away from type-1 use, as there is even less research.  But again, in my opinion, the improvements that have been seen are real (in early testing) but not dramatic.  However, this doctor thinks that by combining these two treatments, we can lower A1c between 1 and 2 points;  that would be great, if it were true, and if type-1s were willing and it was safe to take both drugs for the rest of their lives.

In large Dexcom-funded analysis, CGM users had a 42% lower hospital admission rate and 17% lower ER ad. rate over non-CGM users.

“Preprandial Oral Insulin (ORMD-0801) Reduces Rapid-Acting Insulin Requirements and Fasting Glucose Levels in T1DM Patients”

Some things are more scary than others:
Studies of diabetes blogs don't show discussion of complications but more near term issues like nighttime hypoglycemia

If you got all the way down here, you deserve a laugh:
which came from this tweet:

Joshua Levy
publicjoshualevy at gmail dot com
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My daughter has type-1 diabetes and participates in clinical trials, which might be discussed here. My blog contains a more complete non-conflict of interest statement. Thanks to everyone who helps with the blog.