++ Pre-existing insulin resistance - Newcastle University · The twin cycle hypothesis. Substantial weight loss in people with type 2 ... 0 2 4 6 8 0 5 10 15 20 Fat % 0 2 4 6 8 0

The twin-cycle hypothesis: type 2 diabetes

Positive calorie balance

↑ liver fat

↑ basal insulin secretion

Resistance to insulin control of glucose

↑ plasma glucose

Liver cycle

Pre-existing insulin resistance++

Taylor, Diabetologia 2008; 51: 1781



↑ liver fat



↑ VLDL-TG

↑ plasma glucose

Liver cycle

Pancreas cycle ↑ islet fat


↓ acute insulin response to food




↑ liver fat



↑ VLDL-TG

↑ ↑ plasma glucose

Liver cycle

Pancreas cycle ↑ islet fat




Twin cycles grind on over at least a decade –

But the rapid rise in plasma glucose occurs over ~18 months

Decrease fat –normalise the insulin response to eating

Decrease fat –improve insulin action and normalise overnight blood sugar

The twin cycle hypothesis

Substantial weight loss in people with type 2 diabetes will:

and

Liver Pancreas

The Counterpoint Study

Weeks0 1 4 8

Very low calorie diet(~800 kcal/day)

Tests of: Beta cell functionLiver and muscle insulin sensitivityLiver and pancreas fat

(Counteracting Pancreatic inhibitiOn of INsulin secretion by Triglyceride)

New magnetic resonance method allows measurement of organ fat content

Bland-Altman reproducibility coefficients: Liver 0.5; Pancreas 0.9

Method based on 3-point Dixon developed by Dr Kieren Hollingsworth

Detailed anatomical MRI scan Fat map co-localised with scan

Body weight change during Counterpoint

0 2 4 6 8

80

90

100

110

120

Week

Wei

ght

(kg)

Total loss of 15.3 ± 1.2 kg

Lim et al, Diabetologia 2011

Matched non-DM controlsstudied at one time point

Counterpoint: Effect of VLCD on fasting glucose

0 2 4 6 8

4

6

8

10

Week

Fast

ing

plas

ma

gluc

ose

(mm

ol/l) 9.2 → 5.9 mmol/l (p=0.003)


All hypoglycaemic agents stopped

30% decrease after 1st week

Week 0 Week 1 Week 4 Week 80

5

10

15

20Li

ver f

at c

onte

nt(%

)


20

40

60

80

100

% In

sulin

sup

pres

sion

of H

GP

Dotted red lines show non-diabetic matched control data

As liver fat normalised -so did liver insulin sensitivity


Liver fat and liver insulin sensitivity - Counterpoint

Colour key:100%

0%

Change in liver fat during 8 weeks of low calorie diet and reversal of Type 2 diabetes to normal

Baseline scan –liver is green

= 36% liver fat

8 weeks scan –liver is black

= 2% liver fat

0

2

4

6

8

10

M v

alue

(mg.

kgff

m-1

.min

-1)

Muscle insulin sensitivity assessed by isoglycaemichyperinsulaemic clamp during Counterpoint


No change in muscle despite reversal of diabetes

0 1 4 8Week

Beta cell function: Control subjectsStepped Insulin Secretion Test with Arginine (SISTA)

-20 0 20 40 60 80

4

8

12

16 Arginine

Time (min)

Pla

sma

gluc

ose

(mm

ol/l)

-20 0 20 40 60 80

0.3

0.6

0.9

1.2

1.5

1.8

Time (min)

ISR

(nm

ol/m

in/m

2 )


IV glucose infused to achieve this plasma glucose profile:

Observed insulin secretion rate:

Beta cell function: Diabetes subjects

-20 0 20 40 60 80

0.3

0.6

0.9

1.2

1.5

1.8

Time (min)

ISR

(nm

ol/m

in/m

2 )

Baseline


No first phase insulin secretion


Week 1

-20 0 20 40 60 80

0.3

0.6

0.9

1.2

1.5

1.8

Time (min)

ISR

(nm

ol/m

in/m

2 )



Week 4

-20 0 20 40 60 80

0.3

0.6

0.9

1.2

1.5

1.8

Time (min)

ISR

(nm

ol/m

in/m

2 )



Week 8

-20 0 20 40 60 80

0.3

0.6

0.9

1.2

1.5

1.8

Time (min)

ISR

(nm

ol/m

in/m

2 )


Change in beta-cell functionand pancreas fat

Lim et al. Diabetologia 2011; 54: 2506–2514

0 2 4 6 8Week

0.0

0.2

0.4

0.6

0.8

1.0

Firs

t pha

se in

sulin

resp

onse

(nm

ol/m

in/m

2 )

0 2 4 6 8Week

3

6

9

12

Panc

reas

fat c

onte

nt (%

)

Type 2 diabetes is a simple condition of fat excess to which some people are more susceptible than others

Health-motivated people can reverse their diabetes

This knowledge must not be used as a stick with which to beat people who do not want to change their lives

Reversing the twin cycles of type 2 diabetes

Decrease fat –normalise the insulin response to eating

Decrease fat –improve insulin action and normalise overnight blood sugar

The twin cycle hypothesis

Substantial weight loss in people with type 2 diabetes will:

and

Liver Pancreas

0 2 4 6 80

5

10

15

20Fa

t %

0 2 4 6 80

5

10

15

Insu

lin s

ensi

tivity

Inde

x

0 2 4 6 80.0

0.2

0.4

0.6

0.8

a

Week

Firs

t pha

sein

sulin

sec

retio

n

0 2 4 6 83

6

9

12

Week

Fat %

The Counterpoint study – Type 2 diabetes, 800kcal diet

Liver

Pancreas

0 2 4 6 8

4

6

8

10

WeekFa

stin

g gl

ucos

e(m

mol

/l)

Reversibility of type 2 diabetes

The pathophysiological defects in the liver and pancreas are reversible –if diabetes duration <4yr


20

40

60

80

100

% In

sulin

sup

pres

sion

of H

GP

0.0

0.2

0.4

0.6

0.8

1.0

Firs

t pha

se in

sulin

resp

onse

(nm

ol/m

in/m

2 )

Can long duration type 2 diabetes also be reversed?

0 1 4 8

The Counterbalance Study –

Counteracting BetA cell failure by Long term

Action to Normalize Calorie intakE

Questions:Can people with longer duration type 2 diabetes reverse to normal?Is the reversal of type 2 diabetes durable if body weight remains stable?

Very low calorie diet

Protocol for CounterBalance study

T2DM < 4 yr

0 1 4 8 10 + 6 months

Weeks

T2DM > 8 yr

Stepped return to

normal eating

Weight maintenance

8 10

Counterbalance Subjects

Age (yr) 52.1 ± 2.6 61.6 ± 2.0

Weight (kg) 99.0 ± 3.7 96.9 ± 3.8

BMI (kg/m2) 34.6 (27.6-38.0) 33.0 (29.4-45.7)

FPG (mmol/l) 9.6 ± 0.7 13.4 ± 0.8

HbA1c (mmol/mol) 55 ± 2 70 ± 4

Fasting insulin 17.4 (3.9-48.9) 7.0 (4.1-31.9)

Diabetes duration 0-4y 8-23y(n=15) (n=14)

Steven et al Diabetic Med 2015

Results: weight loss during diet

Weight loss: 14.6 ± 0.8 % vs. 14.5 ± 0.7 %p=0.662

Fasting plasma glucose during diet


By diabetes duration

Spearman 0.501; p=0.006

<4 y 5.8 mM

8-12 y 6.2 mM

>12 y10.6 mM


Aetiology of type 2 diabetes: twin-cycle hypothesis


↑ liver fat


Resistance to insulin suppression of HGP

↑ VLDL triglyceride

↑ plasma glucose

Liver cycle

Pancreas cycle ↑ islet triglyceride



Taylor. Diabetologia 2008; 51: 1781

BMI distribution of participants –Start of UK Prospective Diabetes Study

Taylor & Holman, Clin Sci 128: 405-410 2015

BMI distribution of individuals with type 2 diabetes

20 25 30 4035Taylor & Holman, Clin Sci 128: 405-410 2015

Personal vs population

20 353025 40BMI

A

C

B

Indi

vidu

als

Indi

vidu

als

Indi

vidu

als

The top panel shows the BMI distribution of individuals at diagnosis of type 2 diabetes.

However, a generation ago, the alter egos of those people would have been ~15 mg lighter and would not have had diabetes (blue dots).

OR –

If those people with T2DM lose 15kg, they lose their type 2 diabetes



20 353025 40BMI

A

C

B

Indi

vidu

als

Indi

vidu

als

Indi

vidu

als

Take 3 individuals for example.Each lost 15kg and reversed their diabetes. But by BMI criteria they remain obese, overweight and normal respectively.

Each individual must have crossed a threshold of fat mass, personal to them.



20 353025 40BMI

A

C

B

Indi

vidu

als

Indi

vidu

als

Indi

vidu

als

explains the “Why me” factor –indivduals have personal capacity to store fat safely. If the PFT is exceeded, T2DM slowly develops.


0

20

40

60

80

<10 10-20 20-30 >30

Perc

enta

gere

vers

ing

diab

etes

Amount of weight lost and effect on reversal of diabetes – 92 people after bariatric surgery

Steven, Carey, Small & Taylor; Diabetic Med 2014

Weight loss (%)

The Pancreas Study –

Question:Is the decrease in triglyceride content of the pancreas specifically related to type 2 diabetes and it reversal?Or is it merely a generalised effect of loss of fat from the body?

Design:Compare pancreas triglyceride in matched groups of people with and without type 2 diabetes undergoing identical weight loss

Baseline subject characteristics

Type 2 DMn=18

NGTN=9

Age (yr) 49.1 ± 1.6 46.3 ± 2.1

Weight (kg) 121.0 ± 3.0 114 ± 5.0

BMI (kg/m2) 42.7 ± 0.7 41.3 ± 1.0

DM duration (yr) 6.9 ± 0.5

HbA1c (%) 7.6±0.4

Insulin treated 3

SU treated 9Steven et al, Diabetes Care 39(1):158-65; 2016

T2DM NGT30

40

50

60Fa

t mas

s (k

g)

Fat mass in type 2 diabetic and normal glucose tolerance groups – before and after weight loss

Decrease body weight 13.6±0.7% 12.8 ± 0.8%

T2DM NGT0

5

10

15

Plas

ma

gluc

ose

(mm

ol/l)

Fasting plasma glucose in T2DM and NGT –Before and after weight loss

p<0.001

Steven et al, Diabetes Care 39(1):158-65; 2016

Hepatic insulin resistance in T2DM and NGT –before and after weight loss

p<0.002p<0.001

T2DM NGT0

1

2

3

4

Hepa

tic in

sulin

resi

stan

ce(m

mol

.min

-1.k

g ffm

-1 p

mol

.l-1)


Liver fat in T2DM and NGT –before and after weight loss

T2DM NGT0

5

10

15

Live

r fat

(%)

p<0.05p<0.02


-20 0 20 40 60 80

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (min)

ISR

(nm

ol/m

in/m

2 )

-20 0 20 40 60 80

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (min)

ISR

(nm

ol/m

in/m

2 )

NGT

-20 0 20 40 60 80

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (min)

ISR

(nm

ol/m

in/m

2 )

-20 0 20 40 60 80

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Time (min)

ISR

(nm

ol/m

in/m

2 )

T2DM

Insulin secretion before and after weight loss

Baseline

8 weeks

T2DM NGT

0.0

0.1

0.2

0.3

0.4

0.5Fi

rst p

hase

insu

lin r

espo

nse

(nm

ol m

in-1

m-2

)

First phase insulin secretion in T2DM and NGT –before and after weight loss

p<0.005

p<0.02


T2DM NGT4

5

6

7

8Pa

ncre

atic

tria

cylg

lyce

rol (

%)

Pancreatic triglyceride in T2DM and NGT –before weight lossand after weight loss

p<0.01p<0.005

Panc

reat

ic tr

igly

cerid

e (%

)


T2DM NGT4

5

6

7

8Pa

ncre

atic

tria

cylg

lyce

rol (

%)

Pancreatic triglyceride in T2DM and NGT –before weight lossand after weight loss

p<0.005

Decrease of 1.2%Mean pancreas volume ~50ml

This represents ~0.6 grams of fatPa

ncre

atic

trig

lyce

ride

(%)


Conclusions

Weight loss over 8 weeks brings about loss of pancreatic triglyceride specifically in type 2 diabetes

It is likely that type 2 diabetes is caused by less than 1 gram of fat in the pancreas

30 Practices

DiRECT – a study in routine NHS General Practice

0 12 months 24 months

INTERVENTION2 weekly visits to 20 weeks then 4 weekly

CONTROLVisits at 0, 12 and 24 weeks

280 people

140 people

140 people

30 Practices

DiRECT – a study in routine NHS General Practice

0 12 months 24 months

INTERVENTION2 weekly visits to 20 weeks then 4 weekly

CONTROLVisits at 0, 12 and 24 weeks

280 people

140 people

140 people

What proportion of people with type 2 diabetes can be returned to ongoing normal health?

What psychological factors lead to success?

What metabolic changes ensure continuing normal metabolism?

Documents

++ Pre-existing insulin resistance - Newcastle University · The twin cycle hypothesis. Substantial weight loss in people with type 2 ... 0 2 4 6 8 0 5 10 15 20 Fat % 0 2 4 6 8 0