3/20/2014

1

Type 1 Diabetes, treatment, prediction and prevention

Going to school

Carla Greenbaum MD

Director, Diabetes Program

• Math

• History

• English

• Philosophy

• Civics

MATH

3/20/2014

2

MATH: Absolute and Relative Risk

Type 1 diabetes runs in families

TRUE FALSEor

Raise your hand if you are a health care provider for someone with type 1 diabetes

3/20/2014

3

Keep your hand up if you are a health care provider to those with more than one family member with type 1 diabetes

Paradox?

How is it that most people do not have multiple family members with diabetes and yet, type 1 diabetes runs in families?

100 Newly diagnosed patients with type 1

diabetes

Families without diabetes

90

0.3%

10

Families with diabetes

~5%

15X

3/20/2014

4

300 people without a family member with diabetes

X      0.3% = 1 with diabetes

300 people with a family member with diabetes X      5 % = 15 with diabetes

and (300‐15) = 285 without diabetes

Risk summary

• About 0.3‐0.5% or 3‐5/1000 people have T1D. – This is the same as saying that the absolute risk of having T1D in the general population is 3‐5/1000

• About 5% of those with a family member have T1D– This is the same as saying that the absolute risk of having T1D in families is 5/100 or 50/1000

• The Relative Risk of a family member is thus 15X greater than someone in the general population.

HISTORY

3/20/2014

5

HISTORY: Type 1 Diabetes and the immune system

Type 1 diabetes is a immune disease

“insulitis”

Normal islet Islet missing insulin producing beta cellsLots of immune cells

1970’s

2014’s Insulitis not seen in every islet

Type 1 diabetes is a immune disease

1980’s

Islet Cell Antibodies (ICA)

Insulin autoantibodies (IAA)

Jerry Palmer, University of Washington

ICA512 (IA‐2); ZnT8 ab; GAD ab

3/20/2014

6

Type 1 diabetes is a immune disease

1980’s

Risk Group Type 1 diabetes

Population 1:300

Family members 1:20

Association with class II HLA type and HLA involves the immune system

Jerry Nepom, MD, PhDBenaroya Research Institute

Genetic Predisposition

Bet

a ce

ll m

ass

(?Precipitating Event)

OvertImmunologicabnormalities

Normal insulinrelease

Progressiveloss insulinrelease

Glucosenormal

Overtdiabetes

C-peptidepresent

NoC-peptide

Age (years+

Eisenbarth Model-T1D Natural History

1986

The modified model of disease: 2014

Its really complicated - when you get diabetes depends upon:

• Number of beta cells you start with• When and how fast the immune system starts

destroying beta cells• Whether you slow or turn off the disease• How much insulin you need to control glucose

3/20/2014

7

1. Each person who eventually develops T1D starts out with a variable amount of beta cell mass/function

Joe John

Jim

Number of beta cells you start with

Bet

a ce

ll f

un

ctio

n

Time

1. Each person who eventually develops T1D starts out with a variable amount of beta cell mass/function

2. If the beta cells are destroyed at the same rate, they will get diabetes at different times

Joe

John

Jim

Number of beta cells you start with

Bet

a ce

ll f

un

ctio

n

Time

1. The rate at which the immune system attacks the cells also determines when you get diabetes

When and how fast the immune system attacks the beta cells

Joe

John

Jim

3/20/2014

8

Bet

a ce

ll f

un

ctio

n

Time

1. If you control or turn off the immune attack, you may delay or never get T1D

Joe

John

Jim

Whether you can turn off the immune attack

1. Clinical Diabetes occurs when there is not enough insulin secretion to keep up with demand

How much insulin you need to control glucose

Diabetes No Diabetes

Genetic Predisposition

Bet

a ce

ll m

ass

(?Precipitating Event)

OvertImmunologicabnormalities

Normal insulinrelease

Progressiveloss insulinrelease

Glucosenormal

Overtdiabetes

C-peptidepresent

NoC-peptide

Age (years+

Even though it is complicated; we can predict who will get disease

3/20/2014

9

ENGLISH

ENGLISH: Determining Risk for T1D

Genetic Risk: HLA testing or Family History

Sensitivity:  Proportion of people with disease who test positive:Number of people who have genetic risk who will get diabetesNumber of people with diabetes

Specificity:  Proportion of people without diabetes who test negative:Number of people without genetic risk and who do not get diabetesNumber of people without diabetes

About 75% of people with Type 1 diabetes have high risk HLA, but 25% do not.About 10% of people with type 1 diabetes have a family member with disease, but 90% do not

Genetic risk alone is neither very sensitive nor specific

3/20/2014

10

Antibody testing

Antibody: Protein made by the immune system to identify and destroy foreign* objects (like infections)

*in autoimmune disease the immune system makes a mistake and destroys “self” not foreign objects

5 year risk Longer term risk

No antibodies Less than 1% Likely  less than 3%

One antibody ~3% Likely less than 5%

Two antibodies 35% Likely more than 90%

Two antibodies and abnormal glucose

85% Likely almost everyone

Risk of diabetes among those with genetic risk (family members)

1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0Sur

viva

l Dis

trib

utio

n F

unct

ion

1 2 3 4 5 6 7

Years Followed

8

More antibodies = greater risk

n = 26799

1 ab

2 ab

3 ab4 ab

ICA, IAA, GAD, IA-2ICA, IAA, GAD, IA-2

ALL babies with multiple antibodies will get T1D

Ziegler, et al, JAMA 2013

3/20/2014

11

PHILOSOPHY

PHILOSOPHY: When should we stop the

immune attack?

3/20/2014

12

100 %

Bet

a ce

ll fu

nct

ion

Time

Clinical onset of disease

Pre-islet autoimmunity Islet

autoimmunityAbnormal Glucose Tolerance

Stages of Diabetes

Honeymoon

Longstanding DM with or without insulin secretion

Parikka et al; Diabetologia (2012) 1936

# ch

ildr

en

Age (yrs)0 2 4 6 8 10 12 14

When do autoantibodies occur? (How soon does “diabetes” start?)

64% of children who got T1D before puberty had antibodies by age 2

95% of children who got T1D before puberty had antibodies by age 5

Is having two or more antibodies a “disease’”?

5 year risk Longer term risk

No antibodies Less than 1% Likely  less than 3%

One antibody ~3% Likely less than 5%

Two antibodies 35% Likely more than 90%

Two antibodies and abnormal glucose

85% Likely almosteveryone

Islet Autoimmunity

3/20/2014

13

Is having two or more antibodies a “disease’”?

Islet AutoimmunityHaving islet autoimmunity has no symptoms, but it puts you at risk for getting diabetes

Hypertension (high blood pressure)Having mild high blood pressure has no symptoms, but it puts you at risk for getting heart disease and stroke

Islet Autoimmunity

PHILOSOPHY: Thought experiment

Disease Hypertension Islet autoimmunity

Consequence within 4-5 years

~5/100 get coronary heart disease or stroke

35/100 get T1D

Relative risk reduction (effect size) of treatment

Treating hypertension reduces heart disease by 16% and stroke by 40%

Prevention studies designed with effect size of 40%

Absolute benefit of treatment

Treating 100 patients with high blood pressureprevents 2 people from getting heart disease or stroke

Treating 100 people would keep 14 from getting T1D

Severity of event Heart disease or stroke –severe disability or death

T1D is a manageabledisease

Risk of therapy Blood pressure treatment is costly, but adverse effects tolerated and alternatives available

Cost, risk of adverse events, psychological effects, treating children

PHILOSOPHY: Thought experiment

3/20/2014

14

What therapy is being tested  to treat islet autoimmunity?

Oral Insulin Daily capsule“oral tolerance”

Abatacept

• Age 2 or older• Two antibodies• One antibody is insulin 

autoantibody

Monthly IV infusion for 1 year

• Age 6 or older• Two antibodies, but not 

insulin autoantibody

Teplizumab 14 days IV infusion just once

• Age 8 or older• Two antibodies and 

abnormal glucose

4 year delay to diabetes onset in a subgroup of people treated with oral insulin

(Post‐hoc analysis)

Years

Pro

port

ion

Fre

e of

Dia

bete

s

0 1 2 3 4 5 6

0.0

0.2

0.4

0.6

0.8

1.0

Oral InsulinPlacebo

Log-rank P=0.01

Proportion without T1D

Oral Insulin; N=63 

Placebo; N=69

Does immunotherapy scare you?

3/20/2014

15

Immunotherapy is used to treat autoimmune disease.  There are more than 80 million Americans with 

autoimmune disease

Alopecia areataAnkylosing spondylitisAddisons diseaseHemolytic anemiaAutoimmune HepatitisThrombocytopenic purpuraBehcets diseasePemphigusCrohns diseaseDermatomyositisLupusGraves diseaseHashimotos ThyroiditisMultiple sclerosis

Myasthenia gravisPernicious anemiaPolyarteritisPolychondritisPolymyositisPsoriasisRheumatoid arthritisSclerodermaSjogren’s syndromsStiff man syndromeGiant cell ArteritisUlcerative colitisVasculitisUveitisVitiligo

many of these are treated with immunotherapy

Rituximab (anti CD20)Rituxan

T CellB Cell

Adult Rheumatoid Arthritis

0.8

Rituximab

Overallp < 0.001

0.4

0.5

0.6

0.7

Time in months0 3 6 12

Placebo

Rituximab

**

*

*p < 0.020

C‐pep

tide 

pmol/ml 

Treatment period

Pescovitz, NEJM 2006

3/20/2014

16

Abatacept (CTLA4 Ig)Orencia

T CellB Cell

Adult Rheumatoid Arthritis

Juvenile Idiopathic Arthritis (JIA), age 6 or older

Abatacept (CTLA4‐Ig)(co‐stimulation blockade)

Orban, Lancet 2012

Treatment period

Teplizumab (anti CD 3)

T CellB Cell

3/20/2014

17

Teplizumab (hOKT31 (ala‐ala): AbATE

Change in C‐peptide over time (primary endpoint)*

P=0.002

*Solid lines connect mean values; stars denote medians.  Bars represent 25th and 75th percentile.

Herold, Lancet Endocrinology 2013

0 6 12 18 24Months

Treatment periods

How does treating islet autoimmunity help people with type 1 diabetes?

Many people with type 1 diabetes make insulin long after diagnosis

47%

26%

7% 7% 8%

85%

68%

38%

23%

12%

0

10

20

30

40

50

60

70

80

90

100

3‐5 years(n=113  n=58)

6‐9 years(n=104  n=57)

10‐19 years(n=100  n=100)

20‐40 years(n=107  n=102)

>40 years(n=104  n=50)%

 of peo

ple still making some of their own insulin

T1D Duration

Diagnosed ≤18 years of age

Diagnosed >18 years of age

3/20/2014

18

It is common for adults with type 1 diabetes to still make insulin!!!!!

Many people with type 1 diabetes make insulin long after diagnosis

Maybe the same therapies will be useful in people long after diagnosis as well

• Math:  – Family members have a 15 X increased risk of getting T1D

• History:– T1D is an immune mediated disease

• English:– Risk for diabetes can be accurately predicted 

• Philosophy:– Having two or more antibodies IS a disease

• Civics

3/20/2014

19

Civics

Civics is the study of the great theoretical and practical aspects of citizenship, its rights and duties; the duties of citizens to each other as members of a political body and to the government

New treatments will require more from the very people already burdened with type 1 

diabetes

People with type 1 diabetes and their families are those that will need to participate in research

Raise your hand if you encourage families to  participate in diabetes research

3/20/2014

20

Are you in compliance with ADA 2014 standards of care?

Standards of Medical Care in Diabetes—2014

Screening for Type 1 Diabetes

• Inform type 1 diabetic patients of the opportunity to have their relatives screened for type 1 diabetes risk in the setting of a clinical research study. 

How to participate in research: families

• Be a research ambassador –

– make sure that everyone you know with type 1 diabetes knows that it “runs in families”

– Make sure that family members know they can be tested for risk (testing is “free”)

– Let them know that they may be eligible for research trials to see if we can keep islet autoimmunity from becoming type 1 diabetes

3/20/2014

21

How to participate in research: those with diabetes

• Join the Benaroya Research Institute Diabetes (BRIDGE) Study: 

– Jointly at BRI and Seattle Children’s 

• Join the T1D Exchange Biobank

– Check with the BRI team

How to participate in research: those recently diagnosed with diabetes

• HOT NEWS!If someone is 18 to 35, and diagnosed with diabetes recently – there is a clinical trial enrolling now…with several more to come which will also include younger subjects

HAVE YOUR PATIENTS: Sign up for our news

Join a clinical study

Join BRIDGE

Visit our website atBenaroyaResearch.org

3/20/2014

22

Deborah Hefty, RN, CDE