% | $
Quotes you view appear here for quick access.

MannKind Corp. Message Board

  • rapp78 rapp78 Jun 25, 2013 5:39 PM Flag

    Affinity 1. Lower Your Expectations. Less is More.

    In Affinity 1 Afrezza must have less hypos than the RAA arm or the FDA will have serious questions regarding its clinical utility. 10% less is good, but 20% less is much better. As we know, the A1c reduction only needs to be non-inferior. Anywhere from a .1 to a .5 reduction will suffice as long as the hypos are also considerably less. In 117, Afrezza didn’t quite reach the .1 mark. It was .04 better, but had 25% less overall hypos than the RAA arm (6 vs. 8 per subject-mth). However, Afrezza had more SEVERE hypos (.19 vs. .18 subject-mth) than the RAA arm after 16 weeks.

    In trial 117, A1c baseline was 7.75 and ended at 7.69 after 16 weeks for the Afrezza group. Fasting bg levels were a little over 140 mg/dl. If bg levels drop below 60-70 mg/dl, a “hypo event” can be triggered.

    The FDA wants to see fasting bg levels under 120 mg/dl for Affinity 1. If MNKD accomplishes this, there will be even more severe hypos than in 117. Unfortunately any advances in A1c reductions come at the expense of more hypos.

    Hypo events in this trial are more important than the non-inferiority percentage. Safety is more important than efficacy for this particular trial. A .1 reduction with 15% less hypos is much better than a .2 reduction with more hypos than RAA's. In 117, A1c’s were virtually the same after 16 weeks, yet Afrezza had more severe hypos. This needs to improve, but it’s difficult to achieve better fasting bg levels with even fewer hypos. Herein lies the challenge of Affinity 1. Less is more for this trial. Don’t expect anywhere near .5 superiority.

    Hope for a .1 reduction with at least 10% less overall hypos. I believe the FDA will be very pleased with this result. Couple that with excellent superiority results in Affinity 2 and we get a significant upside in the stock . . . but only if Wall St's expectations are considerably lower than they currently are for Affinity 1.

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • Hey OPC and PFG. I bump this post for both of you. In fact, I'm dedicating this post to the both of you.

      Since you both seem a little shocked and surprised that the FDA is "finally" admitting that they've had a problem with Afrezza efficacy, I wanted to bump this old JUNE 2013 to let you know that I have warned you of this on several occasions before.

      Not only is this JUNE 25, 2013 post of mine smart because I'm telling you 171 will not be as good as many people think, but it's also an intelligent post because I reference Study 009 several times, which PFG today seems surprised to find out the FDA had a problem with its A1c numbers.

      Of course MNKD announced back in 2008 that Study 009 met the primary endpoints, but I tried warning people a YEAR AGO that the FDA wasn't happy with the A1c numbers, and that was the main reason for MNKD's first CRL. In fact, even MNKD admitted in response to their first CRL that the FDA had an "A1c issue" with 009 and were conducting 117 to fix those problems. 117 ended up with better A1c's than aspart, so that trial in 2009, 2010 had much better results than 009. Then you get 171, which did much worse than 117.

      Even more important is that the A1c results in 009 are VERY SIMILAR to the A1c results in 171. So you can form you own opinions on what the FDA will do next.

      Whatever you do, don't listen to me. I'll just guide you wrong. Right PFG? Right OPC?

      • 1 Reply to rapp78
      • and what your ignorance fails to allow you to understand is that 171/175 INCREASED the equivalent dosing ratio 30% resulting in far less Afrezza dosing.

        please tell me you got something more.....i'm bored with your misrepresentations. you're really going to have to step it up if you seek a reply from me. it's no longer a challenge.

        ps-FDA states the reason for the first CRL was not as you wrote. oops.

    • Rapp78, I think you are overemphasizing a statistically insignificant number. I am referring to the Severe Hypo of 0.19 vs 0.18 (subject-month on a sample size of 65 patients). Compared to overall hypo of 6.0 vs. 8.0 a 25% improvement. If you compare that to the 52-week study with 587 patients I quoted, sever hypo was 32.76 vs 37.60 ( a 5% improvement. this is already in %). I am not sure what you are trying to emphasize but it is a bit misleading. Correction welcome.

      • 1 Reply to bbarra1
      • Point taken, but severe hypos are taken seriously by the FDA because those are the ones that can cause instant death. I'd like to see that number go down in 171, but it's hard to reduce that number and improve the A1c differential at the same time. Going from 5% better to worse, even if only slightly worse, isn't a positive trend.

        cThe market and FDA should realize that there is a only a very small acceptable window available for MNKD to show both good safety and efficacy numbers in Affinity 1. Their goal of 120 fasting is overly ambitious, if for no reason other than inciting an increased chance for severe hypos.

        I'm trying to emphasize how a .1% reduction with 10% fewer overall hypos and LESS severe hypos than RAA arm would be a really, really good result in Affinity 1.

        Only 2 Afrezza patients out of 52 in 117 achieved a1c's less than 7% without a severe hypo event. The RAA arm had 5 out of 58 achieve this. Yet .5 superiority is expected to be safe? Are you kidding me?

        This is extremely important b/c MNKD and Joe Springer's biggest argument for better results in 171 compared to 117 is more AGGRESSIVE basal titration. This means MORE insulin. How is MNKD going to give MORE insulin and experience LESS hypos?

        You don't see this as a concern? You are not concerned that many people expect superiority without understanding the direct inverse relationship between more titration and increased hypos?

        You don't see how increased hypos, severe or otherwise, could be a concern for the FDA? What am I missing? What is misleading about me trying to inform people why expectations are too high for Affinity 1?

        I do not want to see a huge "sell on the news" scenario as soon as Affinity 1 is announced with .1 improvement and more hypos than expected. I'd rather see a big uptick upon .1 news. If investors were more informed on how difficult it is to safely titrate insulin for Type 1's, the market and FDA would give MNKD more leeway in August.

    • rapp, I have the 2009 ADA Compendium (Abstracts and Articles) for the Afresa trials (For the others, Yes, the spelling is correct 'Afresa' is the original name).. in the 52-week study, Afresa had:

      (n=301 for T1; n=288 RAA)
      -Lower Total Hypo (in %) - 86.01 (T1+LA1) vs 92.65 (for RAA+LAI)
      -Lower MildModerate Hypo (%) - 85.67 vs 92.66 for RAA
      -Lower severe hypo (%) - 32.76 vs 37.50

      What am I missing? Why isn't this the same as what you were saying?

      • 1 Reply to bbarra1
      • This is for Trial 009 I believe and actually solidifies my point. In 009, the RAA A1c was .23 BETTER than Afresa after 52 weeks. In 117, Afrezza was .04 better. If you google "Is there a need for ultra rapid insulin" and look at the 117 results published by MNKD, you'll see how much Afrezza severe hypos increased from 009 to 117.

        This proves my thesis: The lower you reduce A1c's, the more severe hypos you get in the trial (Afresa had 4.75 % less severe hypos in 009, but more total severe hypos than RAA arm in 117). The improvements in A1c from 009 to 117 explain the difference.

    • Rapp,

      How do you reconcile mgmt's repeated comments regarding next to zero incidents of hypoglycemia vs the findings in 117?

      Also, why would mgmt set up a superiority target of .5 if they didn't think they could meet it?


      • 1 Reply to opc11
      • Mgmt never said anything regarding next to zero incidents of hypoglycemia in 117. Go back and look at the reports. Al Mann and mgmt has said several times that you almost have to try to get a hypo in regards to Type 2 patients and Affinity 2, but never in regard to Type 1's. People need to understand how different Type 1 and Type 2 is.

        I agree that Type 2's and hypos are virtually a non-issue. But in Affinity 1, hypos are a HUGE issue.

        Google "Is there a need for ultra fast insulin" and review MNKD's presentation in Denver where it published the 117 results. You will see for yourself how hard it is to bring down A1c levels to under 7% without severe hypos in Type 1 patients.

        For example, if a Type 1 has a fasting bg level at 100 and takes the lowest possible dose of Afrezza at mealtime, but only eats 10 carbs instead of 40, he or she will get a hypo when Afrezza peaks in 30 minutes. Very few people understand this, but this is the challenge in Affinity 1.

3.061+0.051(+1.69%)12:49 PMEDT