Tuesday, May 25, 2010

An exhaustive comparison of Exubera and Afrezza.


revised on july 25th 2010 for immunogenicity


Months back I wrote an article that compared these two, this is an expanded version.


This article is written for folks who fail to distinguish between Afrezza and Exubera. Other than the fact that both are inhaled, the similarities stop right there. Afrezza has benefited in some ways due to the failure of Exubera, the Mannkind management is now better prepared to present the value proposition (both to FDA and Diabetic population). Mannkind has also paid a price (still paying) as the whole inhaled insulin sector is unfairly maligned by the Exubera’s debacle.

Given all this contrasts, Afrezza is the only insulin in the class of “super rapid acting” insulin.



#

Category

Exubera

Afrezza

Winner
is?

1

Particle

Dry particle, recombinant regular human insulin +
Excipient

Dry particle Proprietary Technosphere particle that contains monomeric insulin (derived from recombinant DNA) that is electro statically linked.

Afrezza

2

Excipient

Mannitol, glycine
and sodium citrate

FDKP (fumaryl diketo piperazine). Proprietary
compound, inert, not metabolized and excreted.
Fumaryl diketopiperazine is not a penetration enhancer.

Draw

3

Delivery

Bong sized inhaler
Air-assisted mechanism disperses the powder from
single-dose blisters into a respirable cloud captured in a holding chamber.

This hand-held, compact, pocket-sized inhaler is easy to carry and use. The inhaler is a breath-powered, high impedance, low-flow device with a passive powder de-agglomeration mechanism.
Single-use cartridges containing TI are inserted into the dispersion chamber of the inhaler, and patients administer the insulin by taking a deep breath. Because the device is breath powered, the patient does not need to coordinate the timing of activation with inhalation.
 
Medtone inhaler C (used in trials), inhaler D (ruggedized
version). Newer one is dreamboat (see the pic at
bottom of the blog) which is the size of a whistle.
There’s also one disposable inhaler.

Afrezza

4

Pharmaco-
kinetics

Very similar to RAA in terms of PK profile,
T(max) for insulin – 45 mins
Duration of action – 6 hours
Total exposure AUC (area under curve) comparable to RAA.

Mimics first phase insulin spike
T(max) for insulin – 12-14 mins
Duration of action 2-3 hours
2/3rd of insulin action happens in first 2
hours.

Afrezza
You need prandial to better mimic natural secretion.

5

Pharmaco-
dynamics

Similar to RAA (say lispro) but
has more pronounced tail.
Exubera demonstrated a significantly faster onset of
action than both regular SC insulin and SC insulin lispro,
as indicated by the shorter mean time to half-maximal effect: 32 minutes vs 48 minutes (P < .001) and 41 minutes (P < .05),
respectively. The mean time to maximal effect of INH was comparable to that
of insulin lispro (143 minutes vs
137 minutes), but shorter than that of regular insulin (193 minutes; P <
.01). Finally, the mean duration of the metabolic activity (time to late
half-maximal effect) of INH was 387 minutes, significantly longer than that
of insulin lispro (313 minutes

The link
here compares all RAA with Afrezza and Exubera

Afrezza

6

Absorp-
tion

Following oral inhalation of a single dose of human
insulin approximately 60% of the emitted dose (about 40% of the blister
content) reaches the lung while 30% is deposited in the oropharynx and 10% in
the conducting airways. Insulin is a peptide drug with negligible
gastrointestinal absorption.
Consequently, the amount of drug deposited in the
oropharynx or swallowed is not expected to affect blood PK profiles.

Approximately 60% of the administered dose is deposited into the lungs.  Inhaled TI particles are evenly distributed throughout the deep lung. At the physiological pH of the deep lung, TI micro-particles rapidly dissolve, and insulin is absorbed across the pulmonary epithelium into the systemic circulation.
 
The FDKP molecules that reach the circulation are cleared and excreted. No metabolites of FDKP have been identified in urine or feces
Technosphere insulin did not disrupt the actin cytoskeleton of cells or increase cellular
permeability; nor did it promote cytotoxicity. In
animal models, there was no insulin accumulation in the lungs and no immunologic changes were observed

Draw

7

Company/
MGMT
behind it

Initially Aventis was a partner,
Aventis sold their rights to Pfizer later. Aventis now has 1.3 billion
reasons
to be happy.
Pfizer/Nektar therapeutics.
One important thing to note is,
Pfizer lacks expertise in Diabetes area.

Mannkind corporation/waiting for marketing partner
Al Mann was the inventor of the insulin pump. He also has
a long history in devices area.

Afrezza

8

Trials for
non-
inferiority

The clinical development program is extensive and consists
of 32 single-dose pharmacological studies, 1 multi-dose (6-month) exploratory
clinical pharmacology study (study 1026) and more than 20 Phase 2/3 studies.
Eight controlled Phase 3 studies (106, 107, 108, 109, 110, 1001, 1002 and
1009), and three controlled supportive Phase 2 studies (102, 103, 104) had
been completed at the time of the marketing authorization application and
represent the core of the INH Phase 2/3 clinical development program.
During the procedure, results on additional comparative
phase 3 studies (1022, 1027, 1029, preliminary results on studies 1028 and
1030) were submitted.
Overall, 821 subjects were exposed to INH during the 32
single-dose pharmacological studies and more than 1975 patients have received
INH for more than 6 months.

From clinicaltrials.gov

Phases

# patients

Other IDs

Phase II

228

MKC-TI-010

Phase II

110

MKC-TI-101

Phase II

16

MKC-TI-03B

Phase II

24

MKC-TI-016

Phase II

50

MKC-TI-112

Phase II

30

MKC-TI-118

Phase III

672

MKC-TI-126

Phase III

547

MKC-TI-103

Phase III

587

MKC-TI-009

Phase III

2343

MKC-TI-030

Phase III

676

MKC-TI-102
The AFREZZA clinical program involved 49 different studies
of AFREZZA and over 5,000 adult patients

Draw

9

Trials/
Studies for superiority

None
It is possible that Pfizer planned some as post marketing.
I’m not aware of any.

Study 117 & 142
There are also other studies being performed. I don’t have
the list currently.

Afrezza
(given the current info)

10

Dosage

1mg, 3mg blisters
1mg = 3 IU of RAA
3mg = 8 IU of RAA
The non linearity of dosage was a major pain.
Exubera lacks dose proportionality and also dose
equivalence
(Three 1mg blisters = 1.4 times one 3mg blister). This
makes the titration process more complex and unpredictable.

Cartridges come in 15, 30, 45 and 60 units.
Afrezza dosage will be Current RAA * 3 (based on Mann’s
interview)
6 TU = 1.56 U
12 TU = 3.12 U
24 TU = 6.24 U
Dosage is
linear.

Afrezza

11

Placebo formulation

Not possible

TI particle without insulin can be prepared and can be
used as placebo

Afrezza

12

Variability

The within-subject variability of glucose-lowering
activity of inhaled insulin is generally comparable to that of SC regular
insulin in subjects with Type 1 or Type 2 DM.
The inter- and intra-subject variability in PK of inhaled
insulin is generally high. From the entire NDA, the % CV, on average is
expected to be >50%. This is bad news.

In the six-period crossover isoglycemic
glucose clamp study of 13 subjects with type 2 diabetes, less within-subject variation in insulin absorption was observed. Less intra-patient variability should facilitate dose titration and minimize subsequent fluctuation in prandial glucose levels.
 
Check out the link
More details in link

Afrezza

13

Bio-
availability

5-11% for non-smokers
12-19% for smokers

25% of s.c (medtone)
Dream boat’s bioavailability could be around 30% of s.c.

Afrezza (less cost)

14

Risks: Lung function

The FDA concluded that people taking Exubera suffer a
decline in lung function – as
measured by forced expiratory volume or by diffusion capacity. However, it
was found that this decline is
not progressive and there is some evidence that it may be reversible.

Same as Exubera; articles say the declines are
statistically insignificant and are reversible
 
the newer dreamboat has less impact to lung compared to Medtone C.

Draw
(maybe Afrezza as there is no powder deposition)

15

Incidence of hypos

Similar to RAA

Way less compared to RAA

Afrezza

16

Toxicology, tumorigenic
risks

Inhalation toxicology studies in rats and monkeys for up
to 6-months there was no evidence for an accumulation of particles in the
lung.
The safety of the excipients has been demonstrated in the inhalation toxicology studies in rats at levels of
up to approximately 100-300 times the human level and in monkeys of up to approximately 10-30 times the human level. The safety is also supported by a literature review on the respiratory tract deposition and clearance of
insulin inhalation dry powder. The excipients are
therefore considered safe for the intended use
No data to suggest that insulin inhalation powder may promote pre-existing preneoplastic/neoplastic
changes in the lung.

The safety of FDKP was established in a comprehensive pharmacology/toxicology program including chronic inhalation toxicology in
rats (6 months) and dogs (9 months), , genetic toxicology, reproductive toxicology, safety pharmacology, and  Read my “All about TI” article.

Draw

17

Carcino-genicity

No carcinogenicity studies have been performed.

a 2-year inhalation carcinogenicity study in rats
a 6-month subcutaneous
carcinogenicity study in transgenic mice.

Afrezza as more tests were done

18

Costs

For a 70kg adult, using 0.5 units/kg/day
$8-9/day
Blister packs = $3042 +
1 chamber $19 + 28 release units/year = $3145.00/year

(guess) within 10-20% the current prandial insulin costs
~$3.5 - 5/day = $1300-$1800
Compare this with s.c
injections ~ $1000/year

Afrezza

20

Antibodies

Exubera showed dramatic increase in antibodies (30 fold
increase
), no one has identified the root cause.
FDA didn’t think of this as an issue
The antibodies did not have an effect on the action of insulin.
In the "Comparative Discontinuation Phase of Study 111" it was shown that among subjects with type 1 DM,
mean insulin antibody level fell by approximately 50% within the initial 3
months after inhaled insulin discontinuation. Among subjects with type 2 DM,
mean insulin antibody level also fell after  discontinuation, but the decline was slightly less and more protracted than in type 1 patients.

Type 2: Afrezza showed <3 fold increase, control arm 2 fold increase;
Type 1: Afrezza 8-fold vs 2-fold for sc insulins.

Afrezza

21

Manu-
facturing

No details; it has been widely reported that they botched
the marketing, manufacturing etc

Mannkind acquired cheap insulin from Pfizer,
Completed state of art manufacturing plant in Danbury to produce; mgmt
claims to have enough insulin for $10 billion sales; it looks like Mannkind
is better prepared.

N/A;

22

Moolah;
not
including opportunity
costs

$2.8 billion USD

Roughly $1.5+ Billion and counting;

Meaningless

23

Benefits – liver

Avoid injections for prandial insulin
Does a poor job of shutting off hepatic glucose production
(gluconeogenesis)

Avoid injections for prandial insulin
Does a far better job in shutting off hepatic glucose
production. Refer to “C-peptide Correction Method” articles in mannkind website.

Afrezza

24

Benefits –
PPG
Post prandial glucose
excursions

Same as RAA.

The best there is, PPG excursions cause long term
complications and can even cause beta cell toxicity and death.
Afrezza controls PPG excursions better than any other
product out there.

Afrezza

25

Benefits
continued..
Patient
satisfaction
surveys

Positive (85%)

Very positive (95%)

Afrezza

26

Effective glycemic
control – reduction in HbA1c
levels

Same as RAA; demonstrates non-inferiority over RAA
Link
for Type 1

This is a controversial topic, Afrezza is in
and out very fast and when compared with RAA, Afrezza looks slightly
inferior. Due to the big tail of RAA, the physicians maintain a high level of fasting plasma glucose and the effects of Afrezza are masked. To compare apples to apples, the fpg should be brought to near normal levels. In such a comparison, the patients RAA control arm would have fallen flat (due to Hypos) and Afrezza arm will be up and running great.
Who wants to sign up for the control arm for such a trial? :)

Can we have diabetics in Hapoalim securities sign up for the control arm?

Afrezza if proper
studies are done
(based on Pk/Pd
profiles)

27

Benefits continued..
Weight gain: Insulin and
weight gain
go like peas and carrots.
One major inhibition to initiating
insulin is the weight
factor.

No significant weight gain

Weight neutral to weight loss; Afrezza doesn’t cause the
weight loss per se, but the lack of snacking in between meals results in
losing some weight that was gained already. The snacking is obviously due to
the hypos (again the tail of RAA)

Afrezza

28

Benefits continued..
Improvement in quality of life

If you live in an open culture (Netherlands
anyone), areas of California,
then there is no inhibition to use a bong device.

Dramatic improvement in quality of life as the dreamboat
inhaler can be used in any public place without attracting attention.

Afrezza (as device is discreet)

29

Cough

The % is not known, but cough is an issue for Exubera also

1% of patients quit TI due to cough

Draw



In Diabetes mine interview, this is what Al Mann has to say about Exubera


“Exubera was very inconvenient, it had absolutely no benefits clinically, and it was not as good as the other treatments on the market. Comparing Afresa to Exubera is like saying that Rezulin — a bad drug that failed — is the same as Actos — a good drug with good outcomes.

What we have is a different form of insulin (from Exubera). It’s powder so we deliver it into the lungs. But that’s an advantage because it’s delivered in the arterial blood system instead of the capillary system. We’re actually delivering insulin monomers (molecules). Nobody ever did that before.

It behaves much like normal pancreatic insulin does. Normal people don’t get hypos, and people taking Afresa don’t either, even if they dose and don’t eat.”