Albireo (ALBO)

[Katmandu2]

Sep 19, 2013 (MarketLine via COMTEX) -- Biodel Inc., a biopharmaceutical
company, has announced preliminary results from Study 3-201, a Phase II clinical
study of BIOD-123, an investigational ultra-rapid-acting mealtime insulin, in
patients with type 1 diabetes.

BIOD-123 achieved the primary endpoint of non-inferiority for HbA1c relative to
insulin lispro, a widely prescribed rapid acting mealtime insulin analog
marketed as Humalog.

Dr Alan Krasner, chief medical officer of Biodel, stated: "BIOD-123 has passed
the first and most critical test-showing that HbA1c non-inferiority can be
achieved in a multi-dose outpatient setting. This study provides a rich database
which will be used to explore how two insulins with distinct pharmacodynamic
profiles can influence a number of important clinical outcomes. Continued
analysis of this database will be invaluable in our development of BIOD-123 and
other meal-time insulin candidates."

Dr Errol De Souza, president and CEO of Biodel, stated: "The establishment of
non-inferiority in HbA1c of BIOD-123 versus Humalog in this study gives us a
high level of confidence that BIOD-123 could achieve this primary endpoint for
FDA approval in larger pivotal studies. We look forward to completing the full
analysis of the results, sharing the data with experts and potential partners to
obtain feedback and preparing for an end of Phase II meeting with the FDA to
define the path forward."

Study 3-201 is a Phase II, open-label, parallel group study conducted at 32
centers in the US In the trial, 132 patients with type 1 diabetes and HbA1c
levels between 6.5-8.5% were randomized to receive either BIOD-123 or Humalog to
use as their mealtime insulin during an 18-week treatment period. Both arms of
the study used insulin glargine, sold as Lantus, as the basal insulin. Following
randomization, subjects entered a 6-week dose titration period during which
basal insulin and then prandial insulin doses were to be titrated in order to
reach standard American Diabetes Association (ADA) recommended pre-prandial
glucose targets. Upon completion of the titration period, subjects entered a
"relative stable dosing period" for an additional 12 weeks.

Secondary endpoints include hypoglycemic events measured and analyzed according
to ADA Workshop recommendations, weight recorded at multiple time points and
postprandial glucose excursions measured in multiple ways including a liquid
meal challenge test, 10-point glucose profiles, continuous glucose monitoring
profiles and routine self-monitoring of blood glucose (SMBG) levels throughout
the study. The study was powered to demonstrate non-inferiority of HbA1c. As is
typical with a Phase II Study, the goal with respect to the secondary measures
was to look for initial trends that would assist in the design of future
studies. Endpoints were analyzed primarily using the day of randomization as
baseline. These analyses involve data from the entire 18-week treatment period,
including the initial 6-week titration phase. Secondary analyses were also
performed using Week 6 (the end of the dose titration period) values as
baseline. These results assess changes during the 3 month stable dosing period.

Sixty six (66) subjects were randomized into each arm of the study. A central
randomization process was utilized. At baseline, mean HbA1c was balanced between
treatment groups: 7.36% in the BIOD-123 group and 7.33% in the Humalog group.
Several between-group imbalances were noted in the baseline characteristics of
the randomized subjects. The mean age in the BIOD-123 group was 48.8 years
compared to 41.3 years in the Humalog group. The mean duration of diabetes was
25.8 years in the BIOD-123 group compared to 20.5 years in the Humalog group.
50.8% of the subjects in the BIOD-123 group were female compared to 28.8% in the
Humalog group. The mean weight in the BIOD-123 group was 78.6 kg compared to
84.3 kg in the Humalog group, likely related to the difference in gender make-up
between the two groups. Body mass index (BMI), however, was more closely
balanced at baseline: 26.8 kg/m2 in the BIOD-123 group and 27.0 kg/m2 in the
Humalog group. Subjects randomized to the BIOD-123 arm were treated at the
beginning of the study with an average of 52.4 units (0.64 units/kg) of insulin,
and those randomized to Humalog were treated with 60.2 units (0.71 units/kg) of
insulin.

The randomization was performed properly and the observed baseline imbalances
occurred by chance in this Phase II study with limited sample size. The baseline
imbalances do not affect the conclusion of non-inferiority of HbA1c. Some
gender-based differences in weight change were observed. The effect of gender
and possible dose imbalances on other secondary variables are under
investigation.

The primary objective of the study was to demonstrate non-inferiority in change
from baseline HbA1c, defined as the upper bound of the 95% confidence interval
around change from baseline HbA1c < 0.40%. The mean HbA1c change from baseline
in the BIOD-123 group was -0.08 0.064% and -0.25 0.063% in the Humalog group.
The 95% confidence interval (-0.01, 0.35) of the between group difference in
change from baseline HbA1c did not exceed the pre-determined threshold of 0.40%,
thereby establishing non-inferiority. HbA1c change during the stable dosing
period was similar in both treatment arms. During this period, the mean change
in HbA1c in the BIOD-123 group was -0.01% and in the Humalog group was +0.02%.

Hypoglycemia frequencies and mean event rates were not statistically
significantly different between groups. However, in the most frequent forms of
hypoglycemia reported, median event rates appear to be lower in the BIOD-123
group compared to Humalog. This observation requires further investigation.

Baseline weights were substantially different between groups, however, the
change from baseline in weight was not significantly different; both groups
gained on average 1.0 kg over the course of the study. During the 6-week
titration period, patients in the BIOD-123 group gained an average of 0.94 0.31
kg, and patients in the Humalog group gained an average of 0.43 0.35 kg. In
contrast, during the subsequent stable dosing period of 12 weeks, patients in
the BIOD-123 group gained an average of 0.10 kg compared to 0.60 kg in the
Humalog group. This beneficial weight trend in favor of BIOD-123 during the
stable dosing period was observed in both genders, yet was more pronounced in
females in which a small weight loss was observed.

Postprandial glucose was measured using multiple methods, generating a dataset
of more than 1.5 million data points, many of which are still being analyzed.
Postprandial glucose excursions are defined as the change in glucose
concentration from before to after a meal. In the liquid meal challenge test,
the average baseline glucose values of 177.3 mg/dl were higher in the BIOD-123
group compared to 148.3 mg/dl in the Humalog group. The maximal postprandial
glucose excursion of 71.8 mg/dl was significantly lower in the BIOD-123 group
compared the 92.6 mg/dl maximal glucose excursion in the Humalog group; this
difference was significant at p=0.034. Additional statistical analysis indicates
that this significant difference in excursion is not due to the differences in
baseline glucose or gender.

Initial overall analyses show mostly no differences in 10-point SMBG or
continuous glucose monitoring profiles. Some time points show lower postprandial
glucose values in the Humalog arm. Additional analyses of within-patient glucose
excursions are pending.

Overall, the adverse event profile between treatment groups appears to be
balanced with the exception of injection site pain. Drop-out rates in the two
arms of the study were similar. No patients in either the BIOD-123 or Humalog
arm reported an incidence of severe pain or dropped out of the study because of
injection site pain. One patient (1.5%) in the BIOD-123 arm reported a single
incidence of moderate pain, compared to none in the Humalog arm. Nine patients
(13.8%) in the BIOD-123 group reported at least one episode of mild injection
site pain during the study compared to 1 patient (1.5%) in the Humalog group.
Six out of 10 patients reporting discomfort with BIOD-123 had complete
resolution during the course of the study while continuing study drug. It was
noted that about half of the patients in the study reporting injection site pain
were from 2 out of 32 investigative sites, with patients from 25 out of 32 sites
reporting no injection site pain. Injection site pain associated with BIOD-123
was not a medically important safety issue and was greatly improved relative to
that associated with VIAject.



URL: http://www.datamonitor.com

Republication or redistribution, including by framing or similar means,
is expressly prohibited without prior written consent. Datamonitor shall
not be liable for errors or delays in the content, or for any actions
taken in reliance thereon


Copyright (C) 2013 Datamonitor. All rights reserved

-0-




SUBJECT CODE: Diabetes

Pharmaceuticals

Healthcare biotechnology

Healthcare R&D

Product