About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Aileron Reports Some Stapled Peptide Results | Main | An HIV Structure Breakthrough? Or "Complete Rubbish"? »

August 15, 2013

Mannkind's Latest Data

Email This Entry

Posted by Derek

I haven't written much about Mannkind recently. This has been a long, long, expensive saga to develop an inhaled-insulin delivery system (Afrezza), which is an idea that all by itself has seems to have swallowed several billion dollars and never given anything back yet. (That link above will send you to some of the story, and this one will tell you something about the disastrous failure of the only inhaled insulin to reach the market so far).

In 2011, Mannkind looked as if they were circling the drain. But (as has been the case many times before), more money was heaved into what might still turn out to be an incinerator, and they kept going. Just in the last few days, they've released another batch of Phase III data, which looked positive. You can see from the year-to-date stock chart that people have been anticipating this, which might account for the way that MNKD hasn't exactly taken off on the news. The stocked jumped at the open yesterday, then spent the rest of the day wandering down, and opened today right back where it was before the news came out.

People might be worried about possible effects on lung function, which show up in the data (FEV1 as well as a side effect of coughing). But there are potentially even bigger concerns in the number for HbA1c and fasting glucose. A closer look at the data shows that Mannkind's product may not have clearly established itself versus the injected-insulin competition. As that FiercePharma story says, this might not keep the product from being approved, but it could give it a rough time in the marketplace (and give Mannkind a rough time finding a big partner).

I wonder if there are any investors - other than Al Mann - who have stuck with this company all the way? If so, I wonder what effect that's had on their well-being? It has been a long, bizarre ride, and no one knows how many more curves and washed-out bridges might still be out there.

Comments (25) + TrackBacks (0) | Category: Clinical Trials | Diabetes and Obesity


1. Bob Sacamano on August 15, 2013 12:08 PM writes... many more curves, mannholes, and washed-out bridges might still be out there.

Permalink to Comment

2. Philip on August 15, 2013 12:34 PM writes...

I am a computer geek that has invested in Mannkind. I maybe in over my head talking about the science of Afrezza and why it could be a good investment, but I will give it a try.

I bought into Mannkind shortly after hedge fund manager Martin Shkreli got the FDA to require a more extensive test of the new inhaler used to deliver Afrezza. You may remember that a bit over two years ago, FDA chemist Cheng Yi Liang was arrested for using confidential information about upcoming announcements, including Afrezza's upcoming approval and the FDA's change of mind. Part of the news was Martin Shkreli's undue (IMHO) influence with the FDA to change the approval to a CRL. I thought that the new inhaler would work as well as the old one, and that would lead to FDA approval. Yesterday's data release shows I was right about the inhaler, we will see if I am correct about FDA approval.

Lung function:
About 6% of Afrezza users dropped out of the studies because of a persistent cough. There is also a statistically significant, but clinically insignificant reduction in lung function (see older studies). So take 6% of potential patients out of Afrezza's market. Not to bad.

Type 1 diabetes patients:
Fasting blood glucose levels improved compared with injected insulin (-25.3 mg/dL versus +10.2 mg/dL, P=0.0027). Significantly less hypoglycemia events (9.80 versus 13.97 events per 100 subject-months, P

Type 2 diabetes patients:
This is a much harder comparison. First where does Afrezza fit into the DM2 treatment time line? Does it go in between oral medications and injected insulin? Is it a replacement for injected insulin? Is it an adjunct for injected insulin? To me it looks like it fits in before injected insulin and as an adjunct to injected insulin, but not a replacement. To be honest, I am in way over my head for this one.

Afrezza is not Exubera, because it is faster acting and easier to use. Mannkind is not Pfizer, because Mannkind does not have the marketing power to sell Afrezza.

I am happy with my investment to this point. Mannkind needs a partner or better a buy out. I am willing to wait for that part of the game to play out. Afrezza may not be the block buster drug that some MNKD longs expect, but I think it will do well and some big pharma will buy MNKD within two years.

Permalink to Comment

3. BuyersStrike! on August 15, 2013 1:12 PM writes...

Perhaps the biggest issue is that Study 175 actually failed.

Throughout 2012* MannKind repeatedly stated that the 175 study must reduce A1c levels by 0.5% more than placebo.

The study results reported yesterday showed a reduction in A1c levels of 0.82% from baseline, but only 0.4% better than placebo.

And, simple math tells us that (0.82-0.42)

Now the p value of the 175 results was excellent at

Notice the weasel wording of MKND's press release and their statements in the conference call when describing the results of 175. They are careful to say that 175 established the superiority of Afrezza to placebo (which it did) but it did NOT meet the required endpoint.

*Find an example of last years statements here:

Permalink to Comment

4. BuyersStrike! on August 15, 2013 2:10 PM writes...

Looks like part of my prior comment was cut off:

And, simple math tells us that (0.82-0.42)

Now the p value of the 175 results was excellent at

Permalink to Comment

5. BuyersStrike! on August 15, 2013 2:13 PM writes...

Wow happened again.

What it should say is that the reduction in the placebo group was 0.42, giving a difference over placebo of only 0.4, which is less than 0.5

The reported P value was excellent, but it proves the statistical significance of a different proposition.

Permalink to Comment

6. Hap on August 15, 2013 3:12 PM writes...

To write a less-than sign, use (ampersand)lt.
To write a greater-than sign, use (ampersand)gt.

The less-than and greater-than signs are HTML lead commands and so text following them (or at least following a less-than sign) may not be shown.

Permalink to Comment

7. Philip on August 15, 2013 3:36 PM writes...

@Hap, sorry, I did not realize that I was posting in HTML.

I wonder if the placebo group for Afrezza is the same as for lorcaserin? Just joking. Both groups had a reduction in HbA1c of 0.40%. The lorcaserin placebo was diet and exercise counseling, not keep doing what you have been doing. Orlistat placebo patients had a 0.29 decrease, but this placebo group had their T2DM medications adjusted if HbA1c increased by 2% or went above 12%. Other studies where the placebo group did not have intervention have not shown that placebo works very well for T2DM patients for reducing HbA1c. The placebo group for linagliptin gained 0.04%. For dulaglutid the placebo group had no change in HbA1c. Empagliflozin's placebo group had a 0.09% increase. Dapagliflozin's placebo group did really bad at a 0.37% gain. LC15-0444's placebo group did a bet better at a reduction 0.06%. And finally, sitagliptin's placebo group had no change in HbA1c.

I know a quick meta data analysis is not worth a whole lot, but it does show why Mannkind thought they had a walk in the park for a 0.5% reduction in HbA1c. I only looked at the first studies I found for each drug. Other studies could have shown different results. What can you expect from a 30 minute meta analysis? The question for Mannkind and its investors is does the T1DM study alone get them FDA approval? I think the answer is yes. The second question is what will the label say about T2DM patients? Harder to say, but I think that Mannkind will point out that Afrezza's placebo worked way better than any other placebo has for HbA1c reduction. If this gets them T2DM usage on the label, that will make a huge difference in sales. Remember I am looking at this through the eyes of an investor.

Permalink to Comment

8. Philip on August 15, 2013 3:42 PM writes...

Sorry for the saying that Aferzza's placebo worked better than any other placebo. I did not check them all out, but it did work better than any others that I found.

Permalink to Comment

9. andybaron on August 15, 2013 4:22 PM writes...

The FDA never set a requirement of .5% improvement over placebo. The requirement was for statistically significant superiority.

Mannkind calculated the number of subjects that would be needed for the trial based on an assumption of .5% improvement -- if they had recruited the minimum number calculated, then they would have needed a .5% improvement to achieve statistical significance. But in fact they recruited additional patients, and therefore they were able to achieve statistical significance even though the improvement was only .4%.

They successfully met the primary endpoint defined for the 175 trial, and based on this I expect that the label will include type 2 diabetes as an approved indication.

But the last CRL was based mainly on doubts about the safety and efficacy equivalence of the gen1 and gen2 inhaler devices, and that was clearly established in the 171 trial. Based on these results, I think that FDA approval is assured.

Permalink to Comment

10. johnnyboy on August 15, 2013 5:45 PM writes...

Beyond the parsing of the trial results, to me the biggest strike against Affreza or other inhaled insulins is that in the end, diabetics don't seem that interested in having them. They're used to injections, know how they work; inhalers are an unknown, more complicated, and if you add to that potential adverse effects on lung function, the benefits of inhaled just don't stack up nearly enough to justify a change. If Affrezza is approved, I think it will face the same crash and burn as Exubera.

Permalink to Comment

11. Hap on August 15, 2013 5:54 PM writes...

I wasn't trying to be irritated - the person before me kept having comments cut off, and that's the reason I commented.

Permalink to Comment

12. BuyersStrike! on August 15, 2013 7:40 PM writes...

@Hap, thanks for the HTML tip. It was driving me crazy.

Permalink to Comment

13. BuyersStrike! on August 15, 2013 8:00 PM writes...

@Philip - Both groups in Study 175 did have concomitant oral drugs (Metformin, etc). I think that explains a good chunk of the A1c reduction.

@JohnnyBoy - Agreed. Anyone investing in, or researching, this area should take the trouble to test their blood sugar with a finger prick and then give themselves a subq saline injection with a 31 gauge needle. The injection is painless. The blood test...ouch. I think inhaled insulin is a solution without a pressing problem.

Permalink to Comment

14. Philip on August 15, 2013 9:18 PM writes...

I think people concentrating on the inhaled part of Aferzza are missing a very important point. Aferzza is a very fast acting short tail insulin. It allows for tighter glucose control than injected insulin.

Permalink to Comment

15. lol on August 16, 2013 12:37 AM writes...


The profile is better only in theory. It does not really mimic the pacreas relsese only looks a little like it. This is essentially proven by the poor HBA1c data-the true test of how well you are treating the condition

Variable doseing of insulin is potentialy deadly.

Based on ther own publicaitons this technology can be used for other drugs as well and is not specific to insulin- they should use it on someting less deadly and more tolerant to day to day variability

Permalink to Comment

16. Electrochemist on August 16, 2013 8:00 AM writes...

The sad fact is that the final chapter in the inhaled insulin story will not be about science. It will be about whether insurance companies will reimburse patients for a product that has little (if any) incremental benefit over the current 50-year old standard of care (injectable insulins in their many varieties).

At least 2 major pharma companies already predicted how this final chapter in the story will read. And again, science does not play into the script at all.

Permalink to Comment

17. Philip on August 16, 2013 8:07 AM writes...

@15 lol

My understanding of treatment of insulin dependent DM, is that it is a delicate balancing act. If the DM patient does not take enough insulin at the correct times, blood glucose goes up and you have the well documented long term effects of DM. If the DM patient takes too much insulin you have the short term and sometimes fatal hypoglycemic event. With current insulin treatments you have both short acting and long acting. The long acting basal insulin is dosed so that the patient has good glucose control, without worry about hypoglycemia. During meals a short acting insulin is given to counter the increase in glucose. This is where Aferzza has an advantage. As Aferzza is faster acting and shorter half life (tail) than current fast insulin treatments. This allows the patient and doctor to raise basal insulin and keep the number of major hypoglycemic events the same, or leave glucose control the same, but lower major hypoglycemic events. I assume different patients and doctors will use Aferzza in different ways. My reading of the data supports the above. You can cut major hypoglycemic events or improve glucose control by using Aferzza as a replacement for current fast insulin treatments.

As for the delevery system, I agree that its use with other treatments is a big deal.

Permalink to Comment

18. p on August 16, 2013 8:37 AM writes...

@17: "With current insulin treatments you have both short acting and long acting."

This was true when people started working on inhaled insulin. Today most Type 1 diabetics use an insulin pump which supplies a steady basal rate of fast acting insulin.

Insertion of the pump inlet is, relative to BG testing, painless. Pumps are a proven and easy to use technology. What is needed for big improvements in diabetes treatment is an easier way to test BG so that it can be done often and insulin doses adjusted on the fly.

Without having done the reading, a faster acting, shorter tail insulin would be valuable; can it be injected? Fiddling with inhalation is silly when an effective delivery system is already worked out and millions of diabetics are used to working with it.

Permalink to Comment

19. Sideline Chemist on August 16, 2013 8:42 AM writes...

I'm going to leave the clinical results aside and let the FDA decide the approval fate of Aferzza since I think ultimately the fate of Aferzza will be decided by non-scientific means. Assuming FDA approval is granted, will different packaging make a significant difference on patient acceptance and market penetration? Is there a better game plan for widespread distribution to pharmacies, doctor education, and device use training? Will insurers provide coverage similar to existing treatments? Lots of questions and the only way to find out is to get onto the market. Even then likely to be a long slow slog.

I'm personally rooting for Mannkind just because of their underdog status--but not enough to put my money on the line. Perhaps I'll rue that decision later...

Permalink to Comment

20. mad on August 16, 2013 3:48 PM writes...

Lack of significant A1c improvement means it doesnt work in the long run. Whatever gain they might belive can come from the ultra-rapid PK is not realized. Probably due to the lack of tight control in the overall dose. e.g. dosing half of what you need (or twice) with an ultra-rapid profile is just not as good as the correct dose by current rapid acting and standard inuslins

Permalink to Comment

21. B on August 16, 2013 4:45 PM writes...

@17 and 18:

18 pretty much has it right. There is no significant problem with insulin injections that is being solved with inhalable insulin. Most T1D patients are now on a pump type system that is incredibly straight forward to use and allows the user to have tight control of their diabetes, especially in combination with a CGM (Continuous Glucose Meter).

The major advancement needed in the field for T1D's is a combination CGM and inuslin pump, a la artificial pancrease, that can monitor glucose levels on the fly and adjust insulin dosing as needed without user input. There is no significant advantage in inhalable insulin over the fast acting insulin (e.g. novolog) that is currently used in subcutaneous pump systems.

Permalink to Comment

22. chemistType1 on August 16, 2013 9:35 PM writes...


Based on the stats I found, only 20% of Type 1 diabetics in US use pump, the rest are still under multiple daily injections.

Also, injections do hurt sometimes but most diabetics deal with them and do not complain. Why? Maybe because it keeps them alive! Injecting a good 30 min before a meal, waiting a good 3 hours to make your numbers go down if you mess up and eating to cover that damn insulin tail. That's the fast insulin that's available right now. People should read more about diabetes before making assumptions.

Permalink to Comment

23. p on August 18, 2013 12:09 PM writes...

chemistTypeI: You're right, I haven't looked up stats. I'm surprised to read it is at 20%. Aside from myself, I know about a dozen Type 1s, ranging in age from 15-60. All have a pump.

I'm not saying I wouldn't ditch the pump. If you can show a significant increase in control and ease of lifestyle, I'm all for it. I don't see from any of the lit how inhalable insulin achieves that.

I was willing to ditch shots because the pump is better. I have nothing against inhalable insulin. It's just that getting rid of shots isn't a big deal. Improving control is. My question is (and I haven't seen this anywhere) is if the new forumlation has to be inhaled to have the better PK numbers. Could it not be injected.

Today's rapid acting insulin is so much better than what was available 25-30 years ago. Another big improvement would be welcome, injection or no.

Permalink to Comment

24. devicesRus on August 18, 2013 8:15 PM writes...

I think that Afrezza might be a great way to make feedback control with an insulin pump a reality. The issue is that even fast (Humalog/Novolog/Apidra) insulins are not very fast i.e. they take a long time (45 to 75 minutes) to peak and they last for a long time (4-6 hours). Afrezza is much faster and goes away much more quickly (no subcutaneous depot) so adding it to pump therapy might be a great clinical tool, but with only 250K patients on pumps it won't be a huge blockbuster.

Permalink to Comment

25. pacard on August 21, 2013 7:11 PM writes...

Thank you for the balanced discussion. There were a lot of quality comments as well.

If any of you are interested in discussing Afrezza and Mannkind more. We are in the process of setting up a new forum to avoid the bashers, pumpers, spammers and id clones that tend to populate a number of forums.

Please join us at

Last, Derek, would you mind if I re-posted your article on our new forum?

Thanks again.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry