Accelerating the Availability of an Artificial Pancreas

1
Accelerating the Availability of an Artificial
Pancreas Aaron J. Kowalski Ph.D. Eighth Annual
Diabetes Technology Meeting November 14, 2008
Bethesda MD
2
Our Mission

• JDRFs mission has been constant since we were
  founded 37 years ago find a cure for diabetes
  and its complications.

3

• Where Weve Been

3
4
Some Diabetes Technologies Milestones

• Early days Ants and Urine Tasters
• 1921 - Insulin
• Late 70s BG and CSII
• 1977 - Biostator
• Early 80s Recombinant Human Insulin
• Late 90s - CGMS
• 2001 1st Annual DTS Meeting
• 2005 First Real-Time CGM FDA Approved
• 2005 JDRF APP Launch
• 2006 FDA Adds Artificial Pancreas to Critical
  Path
• 20? First System to Automatically Push Insulin

4
5

• Why We Need an
• Artificial Pancreas

5
6
The Current State JDRF CGM Trial
7
Hemoglobin A1c Levels are Elevated in the United
States and appear to have Plateaued

• Reported at Diabetes Mellitus Interagency
  Coordinating Committee (DMICC)
• CDC NHANES III Mean A1c 7.7, NHANES II Mean
  A1c 7.6, 60 gt7.0
• Kaiser TRIAD A1cs have stayed the same or
  declined slightly over the past 10 years
• VA 59 of people with diabetes in VA system are
  above A1c 7.0
• Summary Many factors, but tools may be
  suboptimal for reducing HbA1c below 7.0

8

• The JDRF Artificial Pancreas Project

9
The First Generation of an Artificial Pancreas
10
APP Project Goals

• Accelerate the availability of a first generation
  artificial pancreas
• Ensure the artificial pancreas and its components
  are widely available
• Ensure devices from multiple companies are
  approved and reimbursed
• Create a robust and thriving market that
  stimulates significant industry investment in
  next generation technologies

11
Multiple Steps to Improved Patient Outcomes
UPDATE!!

Improved Outcomes
Product Development
Regulatory Approval
Health Insurance Coverage
Clinician Adoption
Widespread Patient Use
12
JDRF Funded Research

• Track One CGM Trial
• Track Two Artificial Pancreas Consortium
• Track Three JDRF Company Partnerships

13
JDRF Continuous Glucose Sensor Trial
Joslin x 2
U. Washington
U. Iowa
Yale
Stanford
Atlanta Diabetes Assoc.
U. Colorado
Kaiser
Nemours
14
JDRF Continuous Glucose Monitor Trial
15
JDRF Funded Artificial Pancreas Consortium
Jaeb Center for Health Research
16
Hybrid open/closed system 1st subject
17
Results Increased Time in Target Range from 58
to 82
Home CSII
Closed-Loop
p lt 0.002
18
Industry Must Commercialize Artificial Pancreas
Feasibility Studies in Outpatient Setting

Feasibility Studies in Hospital Setting
Pivotal Trials Leading to FDA Submission

Will require novel mobile system
incorporating glucose sensor, algorithm, and
insulin delivery
Non-mobile system combines 1) modified versions
of glucose sensor and insulin delivery
devices 2) Algorithms running laptop computers
Will require novel mobile system
incorporating glucose sensor, algorithm, and
insulin delivery
Industry
Now
Artificial Pancreas Consortium
18
18
19
Potential Pathway to an Artificial Pancreas
1
2
Very Low Glucose a Insulin Off Pump Shuts off
due to user not responding to low-glucose alarm
Hypoglycemia Minimizer Predictive hypo causes
alarms followed by reduction in or cessation
of insulin delivery below LOW THRESHOLD
START
END
19
20
Target Product Pathway Today w/ CGM
BG mg/dL
High and low alarms that are preset by use
features such as predictive alarms and very low
alarm vary by manufacturer
180
Threshold alarms Abbott, DexCom, Medtronic
Predictive Hypo/Hyper Alarm Abbott
70
55
Very low alarm - DexCom
Time
20
21
Target Product 1 Very Low Glucose Insulin-off
Pump
BG mg/dL
Alarm warns that glucose has crossed pre-set
threshold if no response, insulin delivery
suspended for 2 hours
180
Alarm 1
Alarm 2 no response
70
Insulin Off - two hours
55
Time
21
22
Target Product 2 Hypo Minimizer
BG mg/dL
Alarm of impending hypoglycemia if not
response, reduction or suspension of insulin
delivery to minimize time below hypo threshold
180
Alarm impending hypo
No response alarm plus insulin reduction or
insulin shut off
70
55
Time
22
23
Target Product 3 Hypo Hyper Minimizer
BG mg/dL
Same as product 2 plus mirror functionality to
minimize time spent hyperglycemia
Resume preset basal rate
180
Alarm impending hypo
No response alarm plus automated insulin push
to bring level below threshold
No response alarm plus insulin reduction or off
Alarm impending hyper
Minimize time in Red zones
70
55
Time
23
24
Target Product 4 Hybrid Closed Loop
BG mg/dL
Automated basal/hybrid closed-loop User-assist
meal bolus with algorithm that aims to restore
glucose to pre-set target
Alarm impending hyper
180
Algorithm-driven insulin increase to restore
target glucose level
No response alarm plus insulin reduction or off
120
70
Alarm impending hypo
Modification of basal to restore glucose to target
55
Time
24
25
Target Product 5 Fully Automated Closed Loop
BG mg/dL
Algorithm-driven insulin increase to restore
target glucose level with minimal user
interactions
180
Minimize post meal peaks and minimal hypoglycemia
Maximize time in target range
120
70
55
Time
25
26
Target Product 6 Multi-hormone Closed Loop
BG mg/dL
Multi-hormone fully automated closed-loop with
minimal user interactions with system
140
Approaching non-diabetes glucose patterns
Maximize time more euglycemic target range
90
70
55
Time
26
27
If you build it, will they come?










































27
28
A thought in conclusion
29
You Can Help

• Sign up for JDRFs CGM advocacy campaign
  (www.jdrf.org/cgm)
• Keep up with the latest
• Visit www.jdrf.org/artificialpancreas, sign up
  for newsletter

30
For more information

• See the JDRF Artificial Pancreas Project
  Website
• www.jdrf.org/artificialpancreas
• Or
• Email me akowalski_at_jdrf.org