clinical applications of risk prediction models

Clinical Applications of Risk Prediction Models

Laura Esserman, M.D., M.B.A.Laura Esserman, M.D., M.B.A.Professor of Surgery and RadiologyProfessor of Surgery and Radiology

Director, UCSF Carol Franc Buck Breast Care CenterDirector, UCSF Carol Franc Buck Breast Care Center

Agenda

•Current Clinical Climate for Prevention

•Potential for Risk Tools to Refine Risk, motivate interventions

•Framework for Decision Aids: the need for tools that provide information in a decision ready context

•How risk models can be integrated into clinical consultations

•Insights from using decision aids, models

Current Clinical Decision Making

Calculate 5-year

Gail Risk

Gail Risk Below 1.67%

Screening

80-95% choose screening

Offered Tamoxifen

(50% risk reduction)

Gail Risk Above 1.67%

5-20% take Tamoxifen

Rush-Port, Vogel et al.

The Gail Model Does Not Identify a Truly High Risk Group of Women

4%14%

33%

65%

90%

44%

0%

25%

50%

75%

100%

45-49 50-54 55-59 60-65 65-70 All Ages

Rockhill et al.

Percent of Nurses Health Study Above the High-Risk Cutoff Point(5 yr Gail Score of 1.67%)

The Gail Model Does Not I dentif y a Truly High Risk Group of Women

4%14%

33%

65%

90%

44%

0%

25%

50%

75%

100%

45-49 50-54 55-59 60-65 65-70 All Ages

Rockhill et al.

Percent of Nurses Health Study Above the High-Risk Cutoff Point(5 yr Gail Score of 1.67%)

What should compel Providers to be concerned with preventionAge and Competing Causes of Death

0

10

20

30

40

50

60

Age

Per

cen

t of

All

Dea

ths

Cardiovascular

Breast Cancer

Lung Cancer

Phillips, et al, NEJM, Vol. 340, No. 2, 1999

High Risk Patients Don’t Choose Tamoxifen

2/43 high risk patients chose to take Tamoxifen for breast cancer prevention

Educational sessions had no influence

Fear of side effects

*Rush Port E, et al Ann Surg Oncol, Vol.8, No. 7, 2001

Decision Making in the Clinical SettingBreast Cancer Prevention Decisions are complex

Uncertainty about Outcomes

Dis

agre

emen

t on

Pre

fere

nces

High

High

Low

Chaos Chaos

Low

Zone of ComplexityZone of Complexity

Order Order

Uncertainty about Outcomes

Dis

agre

emen

t on

Pre

fere

nces

High

High

Low

Chaos Chaos

Low

Zone of ComplexityZone of Complexity

Order Order

What compels women at high risk to consider an

intervention?

1. Evidence that their risk is significant compared to others

2. Evidence that there is an intervention that will help THEM specifically

3. Evidence that the intervention will not have significant side effects

4. Evidence that the intervention is working

Improving the signal-to-noise ratio

Decision AnalysisDecision aid strives to provide the basic elements of a decision: frame,

alternatives, information, preferences and logic

Adult LearningDecision aids should let women choose what they want to learn

– What are people ready to receive? – Layers of complexity (start simple, detail is optional)

Cognitive Science (Tufte)Decision aid should use graphical formats that require the least amount of

cognitive processing – Train people on small number of formats, stick to them

Risk CommunicationRelative risk presentations are confusing, misleading, and bias patients toward

intervention

Clinically Accessible Biomarkers

Biomarkers Risk Discrimination

Detection Tool

Cost Targeted intervention

Atypia ++ rFNADuctal LavageOpen Bx

+++++++

+Tamoxifen, ?AIs

Breast Density

++/+++ MammoMRI

+++++

?Soy, Tam?

Serum Estradiol

+ Blood Test ++ Tamoxifen, Raloxifen

Serum Testosterone

+ Blood Test ++ Tamoxifen, Raloxifen

LCIS ++ BxMRI

+++++

+Tam

DCIS +++ MammoMRI, Bx

+++++

? Tam ?AI ?Statins ?IGFR1 ?

BRCA 1,2 mutations

++++ Blood Test +++ Propylactic surgery,Tam (BRCA2), oophorectomy

Anna Bella Smith Main5yr Gail Score: 2.1%Lifetime Gail Score: 17.3%

Prevention Decision ModelCarol Franc Buck Breast Care Center | UCSF Medical Center

©20

02

Elissa Ozanne, Laura Esserman


How does my risk compare to other women?

Tests to learn more about breast cancer risks

and benefits of therapiesWhat can I do to lower my risk?

What is my risk of breast cancer?

Learning About Your Risk Getting Perspective

Prevention Options Risks and Benefits


Prevention Decision Model :

Learning About Your Risk: What is my risk of breast cancer?

Next

Anna Bella Smith 4Main5yr Gail Score: 2.1%Lifetime Gail Score: 17.3%

Prevention Decision Model : Learning About Your Risk

Sou

rce:

Fis

her

B,

et

al,

JNC

I, v

ol 9

0,

No.

18

, 1

99

8

My Breast Cancer Risk Over Time

Next

9.7%By age 79

6.5%By age 69

2.7%By age 59

NABy age 49

NABy age 39

Claus Risk

17.3%Lifetime

2.1%5 year

Gail Risk

PostMenopausal Status

noneNumber of years of HRT usage

51Age

Anna Bella SmithGAI L MODEL:

The Gail Risk Assessment Model is a

statistical model for estimating the risk

of developing breast cancer in women

undergoing annual screening.This tool

was developed to assist in providing

women with a realistic and

individualized risk estimate of short

and long term breast cancer risk.

CLAUS MODEL:

The Claus model estimates the

probability that a woman will develop

breast cancer based on her family

history of cancer. This includes the

number of first and second-degree

relatives with breast cancer and the

age of cancer onset.



Getting Perspective:How does my risk compare to other women?

Next

A n n a B e l la S m it h M a i n5 y r G a i l S c o r e : 2 .1 %L if e t im e G a i l S c o r e : 1 7 .3 %

P r e v e n t io n D e c i s i o n M o d e l : G e t t in g P e r s p e c t iv e

So

urc

e:

B.R

ock

hill

, N

HS

da

ta

E X A M P L E :A 6 5 y e a r o l d w o m e n w i t h a fi v e y e a r G a i l S c o r e o f 3 % w o u l d f a l ls o m e w h e r e i n t h e t o p 2 5 % o f t h i s d i s t r i b u t i o n .

0 %

Ris

k o

f D

iag

no

sis

2 %

4 %

6 %

8 %

1 0 %

1 2 %

1 4 %

< 3 0 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5 7 0 7 5 8 0

T o p 2 5 %

M id d le 5 0 %

B o t t o m 2 5 %

1 0 0 %

A g e

W h a t D o e s M y G a i l S c o r e M e a n ? W h a t i s M y R i s k C o m p a r e d to O t h e r s ?

N e x t

0

2 0

4 0

6 0

8 0

1 0 0

1 2 0

1 4 0

1 0 0 0

Ra

te/1

00

0 W

om

en

1 .6 7 %F D A t h r e s h o ld f o r

t a m o x if e n u s e .


Average Risk of Breast Cancer Diagnosis for Women (Age 50~60)

40%

30%

20%

10%

0

Avg.

Ris

k o

f B

C D

iagn

osi

s

50%

60%

70%

100%

80%

90%

Prevention Decision Model : Getting Perspective

Within 10 yrs

2.8%

Within 20 yrs

6.2%

Within 30 yrs

9.6%

Lifetime Risk

12.0%

30~40 40~50 50~60 60~70|20~30 | NextPrev

Sou

rce:

Su

rvei

llance

, E

pid

em

iolo

gy,

an

d E

nd R

esu

lts

(SEE

R)

Can

cer

Sta

tist

ics

Rev

iew

19

73

-1

99

8.

400

300

200

100

0

500

600

900

1000

Rate

/10

00

wom

en

700

800


Average Chances of NOT Being Diagnosed with Breast Cancer (Age 50~60)

40%

30%

20%

10%

0

Ch

an

ces

of

not

bein

g d

iagn

ose

d

50%

60%

70%

100%

80%

90%


Within 10 yrs

97.2%

Within 20 yrs

93.8%

Within 30 yrs

90.4%

Lifetime Risk

88.0%

30~40 40~50 50~60 60~70|20~30 | NextPrev

Sou

rce:

Su

rvei

llance

, E

pid

em

iolo

gy,

an

d E

nd R

esu

lts

(SEE

R)

Can

cer

Sta

tist

ics

Rev

iew

19

73

-1

99

8.

400

300

200

100

0

500

600

900

1000

Rate

/10

00

wom

en

700

800

2.8%6.2%

9.6%12.0%



I n the next ten years, an average 50 year old woman has…

Next

Sou

rce:

Jou

rnal

of

the N

ati

onal C

ance

r In

stit

ute

, V

ol.

94

, N

o.

11

, Ju

ne 5

, 2

00

2.

30~40 40~50 50~60 60~70|20~30 |Prev

0.4 ~ 1.4%Heart Attack

38%Visit the doctor about the flu

8%Injured in an automobile accident

1 ~ 2%Increase in breast cancer for each year of HRT use

Other Risks this year alone:

0.2%Accidents

0.1 ~ 0.2%Pneumonia (smoker)

0.2 ~ 0.5%Lung Cancer (non-smoker)

2.1 ~ 6.5%Lung Cancer (smoker)

0.5 ~ 0.7%Breast Cancer

Risk of Death from:

2.8%Breast Cancer

Risk of Diagnosis from:

4.2

0.75-1.0%

Lifestyle Changes Chemoprevention Surgery


Prevention Options: What can I do to lower my risk?

Next

Lifestyle Changes

Sourc

e:

Ross

D, 2

3rd

an

nual San A

nto

nio

Bre

ast

Cance

r S

ym

posi

um

, 2

000

: S

um

mary

by P

ritc

hard

, K

I V

ogel V

G, C

ance

r Jo

urn

al fo

r C

linic

ians,

V

ol.

50

, N

o. 3

, 2

00

0

These moderate modifications are recommended for all women as potential

risk reduction strategies, in addition to vigilant surveillance.

-Weight control

-No cigarette smoking

-Decreased alcohol consumption

-Exercise

Click to learn about Hormone Replacement Therapy and

Breast Cancer Risk.

Prevention Decision Model : Preventative Measures

Lifestyle Changes Chemoprevention Surgery Next

here

InvasiveBreast Cancer

0

Rate

/10

00

Placebo

Tamoxifen

Chemoprevention

Sourc

e:

Gail,

et

al, JN

CI, v

ol 91

, N

o. 3

, 1

99

9

50

100

150

200

Non-InvasiveBreast Cancer

Vascular EventsFractures EndometrialCancer

0%

5%

10%

15%

20%

1000 100%

Benefits Risks

Lifestyle Changes Surgery Next

Prevention Decision Model : Prevention Options

Chemoprevention

50-6035~49 |

3.35% 1.89% 1.34% 0.68%0.36% 0.14% 0.07% 0.11% 0.36% 0.45%

Benefits and Risks of Tamoxifen Usage (Ages 35~49): 5 Year Estimates

60+|


0

Chemoprevention

100%


0%

5%

10%

15%

20%

Benefits Risks



Chemoprevention

Placebo

Tamoxifen

3.1%1.6% 1.9% 0.6%

0.4% 1.2% 1.0% 1. 1%

Benefits and Risks of Tamoxifen Usage (Age 50-60): 5 Year Estimates

Sourc

e:

Gail,

et

al, JN

CI, v

ol 91

, N

o. 3

, 1

99

9

|35~49 50-60 60+|

Rate

/10

00

50

100

150

200

1000


1.34% 0.68%


0

Chemoprevention

100%


0%

5%

10%

15%

20%

Benefits Risks



Chemoprevention

Placebo

Tamoxifen

3.67%1.67%

5.6%

1.9%0.7%

2.1% 3.1%3.8%

Benefits and Risks of Tamoxifen Usage (Age 60+): 5 Year Estimates

Sourc

e:

Gail,

et

al, JN

CI, v

ol 91

, N

o. 3

, 1

99

9

|35~49 50-60 60+|

Rate

/10

00

50

100

150

200

1000


1.34% 0.68%

GeneticTesting

Ductal Lavage andFine Needle Aspiration

SerumEstradiol


Risks and Benefits:

Next

Tests to learn more about breast cancer risks and benefits of therapies

Ductal Lavage and Fine Needle Aspiration

Sourc

e:

Fish

er

B, et

al, JN

CI, V

ol 90

, N

o. 1

8, 19

98

Expected Breast Cancer Risk Over Five Years

Placebo

Tamoxifen

Prevention Decision Model : Risks and Benefits

Next

All Women

3.4%1.7%

Atypical Hyperplasia

5.1%

0.7%0%

5%

10%

15%

20%

100%

0

50

100

150

200

1000

Rate

/10

00

Wom

en

50% relative risk reduction with tamoxifen

86% relative risk reduction with tamoxifen

SerumEstradiol


Sh

ort

term

(~

5 y

r) R

isk

of

Bre

ast

Can

cer

Women on tamoxifen had about 50% of the number

of breast cancers seen in the placebo group –

50% relative risk reduction.

The absolute benefit is smaller - only 3.4% high-risk

women are expected to develop breast cancer as

compared to 1.7% in women using tamoxifen – 1.7%

absolute risk reduction over 5 years.

Women with atypical hyperplasia on

tamoxifen had about 14% of the number

of breast cancers seen in the placebo

group – 86% relative risk reduction.

The absolute risk decreased from an

expected 5.1% to 0.7% - a 4.4% absolute

risk reduction over 5 years.

Atypical Hyperplasia Predicts Benefit from Tamoxifen

GeneticTesting


Sourc

e:

Saute

r, 1

99

7;

Fabia

n C

J, e

t al, JN

CI V

ol.

92

, N

o. 1

5, 20

00


SerumEstradiol

NextDuctal Lavage andFine Needle Aspiration

Lowest risk groupFor women with 5 yr Gail risk less than 2%, risk decreases to below 1% over 3 years for both women with AH and no AH.

Middle risk groupFor women with 5 yr Gail risk greater than 2% but with no AH, risk is about 4% in 3 years.

Highest risk groupFor women with 5yr Gail risk is greater than 2% with the presence of AH, risk is about 15% in 3 years.

Sh

ort

term

(~

3yr)

Ris

kof

Bre

ast

Can

cer

5 yr Gail Score > 2%5 yr Gail Score < 2%

0%

5%

10%

15%

20%

100%

No Atypia

Atypia

LowestRisk

Group0%

MiddleRisk

Group4%

HighestRisk

Group15%

0

50

100

150

200

1000

Rate

/10

00

Wom

en

Learning from Atypical Hyperplasia (AH)

GeneticTesting

Fabian JNCI 2001


Sourc

e:

Saute

r, 1

99

7;

Fabia

n C

J, e

t al, JN

CI V

ol.

92

, N

o. 1

5, 20

00


SerumEstradiol


Sh

ort

term

(~

3yr)

Ris

kof

Bre

ast

Can

cer

Middle Risk Group5 yr Gail Score > 2%

No finding of AH

Lowest Risk Group5 yr Gail Score < 2%

Independent of AH findings

0%

5%

10%

15%

20%

100%

Each Less than 1% 4%

15%

0

50

100

150

200

1000

Rate

/10

00

Wom

en

Atypical Hyperplasia and the Benefit from Tamoxifen

Highest Risk Group5 yr Gail Score > 2%

Finding of AH

No Treatment

50% Risk Relative Reduction with Tamoxifen Use

86% Risk Relative Reduction with Tamoxifen Use

2.1%2%

GeneticTesting

Serum Estradiol

Sourc

e:

Cu

mm

ing

s S

. et

al, JA

MA

, 2

87:

22, 20

02

0

Short Term Breast Cancer Risk

>0 to <5 5 to 10 >10

LowestRiskGroup0.6%

HighestRiskGroup3%

Serum Estradiol Level (pmol/L)



Next

1.2% 1.8%

Sh

ort

term

(~

4yr)

Ris

kof

Bre

ast

Can

cer

0%

5%

10%

15%

20%

100%

0

50

100

150

200

1000

Rate

/10

00

Wom

en

SerumEstradiol

Women with the highest estradiol level had about a three fold risk of breast cancer

as compared to the women with the lowest estradiol level.

Higher hormone levels in the blood are associated with a higher risk of breast cancer.

Learning From Serum Estradiol Level: Postmenopausal Women

GeneticTesting

Serum Estradiol

Sourc

e:

Cu

mm

ing

s S

. et

al, JA

MA

, 2

87:

22, 20

02

76% relativerisk reductionFrom Raloxifen

Placebo

Raloxifene



Next

0

Short Term Breast Cancer Risk

>0 to <5 5 to 10 >10

Serum Estradiol Level (pmol/L)

Sh

ort

term

(~

4yr)

Ris

kof

Bre

ast

Can

cer

0%

5%

10%

15%

20%

100%

0

50

100

150

200

1000

Rate

/10

00

Wom

en

0.6% 0.6%1.2%

0.4%1.8%

0.8%3%

0.7%

SerumEstradiol

Women with the highest estradiol levels on raloxifene had about 24% the number of

breast cancers seen in the placebo group. The absolute risk decreased from 3% to 0.7%.

As hormone levels in the blood is higher, the benefits of raloxifene increase. Side effects

of raloxifene are similar to those of tamoxifen but do not include endometrial events.

Learning From Serum Estradiol Level: Postmenopausal Women

GeneticTesting


SerumEstradiol


Genetic Testing

Genetic Testing and the Benefit of Prevention Options

Sourc

e:

AS

CO

Pro

ceed

ing

s 2

00

2

GeneticTesting

Life

tim

e R

isk

of

Bre

ast

Can

cer

Higher Risk EstimateFor Genetic Carriers

Lower Risk EstimateFor Genetic Carriers

0%

20%

40%

60%

80%

100%

50%

0

200

400

600

800

1000

Rate

/10

00

Wom

en

No Treatment

50-70% Relative Risk Reduction from Oophorectomy

90-95% Risk Relative Reduction from Mastectomy

20%

3.75%

85%

34%

6.4%

Insights

There is a critical need for dynamic models that enable us to assess the impact of interventions-

– that is what patients want

Biomarkers that predict effectiveness of interventions will increase willingness/motivation to accept interventions

There is a hierarchy of risk models – e.g. BRCA trumps Gail– Determines impact of and discussion about options,interventions

Risk that motivates patients to choose an intervention:– 10-15% risk at 5 years– Risk of recurrence after surgery for non-comedo DCIS

10-12% at 5 years, 20% risk at 10 years– Maybe DCIS is the best opportunity for prevention?

Cost Benefit ModelElissa Ozanne PhD; Laura Esserman MD MBA

GoalsUnderstand value of biomarkers for breast cancer riskEvaluate cost effectiveness using atypia as an example

MethodsMarkov model, evidence from clinical studies

Strategies Examined:1. 1. Screening: Routine screening (mammography) all women 2. 2. Tamoxifen: Tamoxifen therapy for all women3. 3. Lavage: Attempt lavage, tam use if DL possible and atypia found4. 4. FNA: 4 quadrant FNA all women, tam use only for atypia

Ozanne, Esserman 2004, Cancer Epidemiology and Biomarkers, accepted

•biomarker relative risk prediction increases cost effectiveness• FNA and DL are more CE if atypia is a good predictor

•more effective intervention increases CE • If biomarker predicts more effect of drug, CE increases

• inexpensive tests offer highly cost effective strategies • If it is expensive/painful to get biomarker, treating everyone

is more CE

•inexpensive interventions offer highly cost effective strategies • Expensive effective interventions not very cost effective

Mammography 50-70

Sensitivity

What is the yearly hazard rate for progression to cancer for . . .

Annual Hazard

DCIS 1-3%

Atypia Gail Risk > 2 Gail Risk < 2

4%1%

LCIS family history none

1-2%0.5-1%

BRCA1/2 1-5%

5 yr Gail Risk >5

1-2%

60 yr old Gail <2

0.3-0.5%

CBC for pt with Ca

0.5%

How do the treatments vary? . . .Treatment

DCIS BCSBCS + XRTBCS + XRT+TamMastectomy

Atypia Gail Risk > 2 Gail Risk < 2

ScreenTamBilat Mastectomy

LCIS family history none

ScreenTamBilat Mastectomy

BRCA1/2 ScreenOophorectomyTamBilat Mastectomy

High RiskGail>1.7; Inv Ca

Screen Consider Tam

What makes DCIS treatment hard to

change?

• Perspective not optimal

• Poor understanding of Risk, timing of progression

Prevention ParadigmHigh Risk Conditions

Normal cells

DCIS

Atypia

LCIS

Neoadjuvant Therapy?

Breast Cancer

Improvements

The Prevention Tool we developed is a physician decision aid

evidence is organized using common outcome: Risk at 5,10 years

Patient Physician Aids should include more layering of information

Decisions can be layered by side effects: serious vs. QOL

Trial of tool vs. notdesire for risk stratificationchoice of interventions

Side Effects

Serious

Yes Weigh risk vs benefit

No Review side effects

Trial of medication

Sx No Sx

ContinueWeigh Sx vs. benefits

A Good Decision Aid

Enables insight

Facilitates dialogue among providers, patients, families

Reduces confusion

Motivates change in approach based on personal preferences

Requires models that provide risk in perspective, and enable tailoring of risk based on interventions

clinical applications of risk prediction models

Documents

risk reductiongail risk

risk tools

high risk patients dont

year gail riskgail risk

high risk group of women4

increased risk of cavalidated

gail model

agesrockhill et