Prolaris to help make treatment decisions in localised prostate cancer

Prolaris -

a gene expression assay for

assessing long-term risk of

prostate cancer progression

Marc Laniado MD FEBU FRCS(Urol)

Hertzelia/Israel

26 May 2017

Difficult decisions in many scenarios

because prognostication poor

Risk on prostate

biopsies uncertain

30% BCR after

radical

prostatectomy

30% fail

“active surveillance”

Can a genomic test help us

make better decision

Very Low

Low

Intermediate

Individual Risk

Add independent predictive information

beyond standard clinical & pathological measures

Biomarkers: measurable entity whose

presence signifies a disease or condition

•Sources

•Nucleic acid (RNA/DNA)

•Protein

•Metabolite

Associated with meaningful endpoints

Provide additional information above nomograms

Allow more accurate risk assessment

Influences physician treatment decisions

Individualised treatment decisions

PROLARIS:

cell cycle progression (CCP)

Score

- Score = ratio of 31

proliferation genes

to 15 normal genes

- RNA expression

assay RT PCR

Genes correlated with CCP

score include DNA repair genes

DNA Repair Genes

Some CCP genes are target of

chemotherapy & radiotherapy

Anthracycline

Gemcitabine

premetrexed

Taxanes

Fluorouracil

Radiotherapy

Oncotype: 1

proliferation gene only

TPX2

CCP score in published

studies use DIFFERENT

scale

THE CURRENT PROLARIS SCORE

IS

THE OLD CCP SCORE +4

CCP score has a normal

distribution in men prostate cancer

Porpiglia 2015 EAUMedian -0.4

Num

er

of m

en

Level 1B evidence for biomarkers

score exists for Prolaris

LOE LOE category Study Design Validation studies required

I A Prospective Preferred, but not required

BProspective using

archived samples≥ 1 with consistent results

II BProspective using

archived samplesNone or inconsistent results

CProspective/Obser

vational≥ 2 with consistent results

III CProspective/Obser

vational

None or 1 with consistent results

or inconsistent results

IV-V DRetrospective/Obs

ervationaln/a

Simon 2009 JNCI

CCP score evaluated in biopsy/TURP/ RP

tissue & after conservative/radical treatment

Study Sample typeNo of patients

(events)Endpoint Ref

TURP WW TURP chips 337 (76) Death (PCa) Cuzick 2011

TRUS biopsy WW Biopsy 349 (90) Death (PCa) Cuzick 2012

TRUS biopsy WW Biopsy 585 (100) Death (PCa) Cuzick 2015

RP Biopsy 582 (166, 12)BCR,

metastasisBishoff 2014

EBRT Biopsy 141 (19) BCR Freedland 2013

RP (low risk) Biopsy 188 (56) BCR Tosian 2017

RP Surgical specimen 353 (132) BCR Cuzick 2011

RP Surgical specimen 413 (82) BCR Cooperberg 2013

EBRT after RP Surgical specimen 47 (36)BCR/metasta

sisKoch 2016

RP or RT men (AA 36%) Biopsy 767 (281) Metastasis Bardot 2017

CCP score evaluated in cohorts

similar to today’s patients

StudySample

type

No of

patients

(events)

EndpointCharacteristc: age, PSA,

ISUP grade group

CCP

scoreRef

TURP WWTURP

chips337 (76)

Death

(PCa)age 70, PSA 8, GG 1 0.67 Cuzick 2011

TRUS biopsy

WWBiopsy 349 (90)

Death

(PCa)age 70, PSA 19, GG 2 1.03 Cuzick 2012

TRUS biopsy

WWBiopsy 585 (100)

Death

(PCa)Age 71, PSA 15, GG 2 0.40 Cuzick 2015

RP (low risk) Biopsy 188 (56) BCR Age 61, PSA 6, GG 1 -0.15 Tosian 2017

RP Biopsy582 (166,

12)

BCR,

metastasisAge 62, PSA 6, GG 1

-0.4 to

0.3Bishoff 2014

RPSurgical

specimen353 (132) BCR age 68, PSA 7, GG 1 0.16 Cuzick 2011

RPSurgical

specimen413 (82) BCR Age 63, PSA 6, GG 1 -0.37 Cooperberg 2013

EBRT Biopsy 141 (19) BCR Age 66, PSA 8, GG 2 0.12 Freedland 2013

EBRT after RPSurgical

specimen 47 (36)

BCR/metast

asisAge 60, PSA 9, GG 2 0.4 to 0.6 Koch 2016

RP or RT men

(AA 36%)Biopsy 767 (281) Metastasis Age 65, PSA 6, GG 2 - Bardot 2017

CCP score varies within same

ISUP prognostic grade groupCCP score -1 CCP score 1 CCP score > 1

Increasing ratio of proliferation gene activity

Gleason score 6 & 7 tumours

have have low & high CCP score

Porpiglia 2015 EAU

CCP score distribution similar amongst

Caucasian & African American

Freedland 2013

Bardot 2017 AUA

CCP score poorly correlated to

PSA, Gleason Score or stageStudy Sample type

No of patients

(events)

PSA correlation Gleason score

correlationRef

TURP WW TURP chips 337 (76) 0.27 0.57 Cuzick 2011

TRUS biopsy WW Biopsy 349 (90) 0.14 0.37 Cuzick 2012

TRUS biopsy WW Biopsy 585 (100) 0.3 0.4 Cuzick 2015

EBRT Biopsy 141 (19) <0.33 <0.33 Freedland 2013

RP Biopsy 582 (166, 12) 0.09 to 0.13 0.18 to 0.30 Bishoff 2014

RPSurgical

specimen353 (132) 0.21 0.22 Cuzick 2011

RPSurgical

specimen413 (82) 0.11 0.18 Cooperberg 2013

EBRT after RPSurgical

specimen 47 (36) “weak” “weak” Koch 2016

This is a benefit!

Gleason score, PSA and stage poorly

predict who benefits from treatment

There is a need to better stratify the risk for patients

Wilt 2017 AUA PIVOT study

Is the CCP score a useful

biomarker?

Associated with meaningful endpoints

death

Metastases





CCP score independently associated

with metastasis/death in multiple studies

Study

No of

patients

(events)

EndpointCCP

scoreP PSA

Gleason

scoreRef

TURP WW 337 (76) Death (PCa)Univ. 2.9 <10-21 <10-13 <10-18

Cuzick 2011Multiv. 2.6 <10-10 <10-7 <10-4

TRUS biopsy

WW349 (90) Death (PCa)

Univ. 2.0 <10-9 <10-4 <10-7

Cuzick 2012Multiv. 1.7 <10-4 0.017 0.002

RP (bx) 582 (12) MetastasisUniv. 5.4 <10-7 <10-8 <10-3

Bishoff 2014Multiv. 1.5 <10-4 <10-5 0.02

TRUS biopsy

WW585 (100) Death (PCa)

Univ. 2.1 <10-14 <10-8 <10-11

Cuzick 2015Multiv. 1.8 <10-6 - -

Salvage EBRT

after RP (RP)47 (36) Met/progress

Univ. 3.7 0.006 0.83 0.007Koch 2016

Multiv. 10.4 0.003 0.55 0.002

RP or RT treated

men (AA 36%)767 (281) Metastasis

Univ. 2.8 <10-11 - -Bardot 2017

Multiv. 2.0 <10-5 - -

CCP score predicts death in

conservatively managed men

CCP score predicts death in

conservatively managed men after TURP

Cuzick 2011 Lancet Oncology

CCP Score < 0

CCP Score 0 to 1

CCP Score 1 to 2

CCP Score >2

Typical pt: age 70, PSA 8, Gleason score 3+3=6

CCP Score stratifies mortality

risk in transrectal biopsy cohort

• 442 men in 6 UK cancer

registries (90 to 96)

• Transrectal prostate

biopsies 1990 to 1996

• Average 3 cores/pt

• Median age 70, PSA 19,

GS 3+4=7,

CCP score 1.03

• 20% died overall

CCP >3

CCP 2-3

CCP <2

Cuzick 2012 B J Cancer

HR 2.56 (CI 1.9 to 3.5)

CCP score predicts death inTRUS bx cohort

after conservative mx in more recent cohort

Cuzick 2015 BJC

n = 761 men

2002

Average 6 cores

age 71,

F/U 10 y

Diagnoses 2002

GS 3+4=7

PSA 15

Analysis of most significant lesion may be

important: CCP score drops slightly as

move away from tumour in RP specimens

n=35 overall, but bigger drop

0.8 unit in CCP for Gleason

score 8-10 tumours (n=5)

Berman 2013 EAU

PSA failure & metastases

proportional to CCP score in

needle bx & RP specimens

CCP score independently associated

with PSA recurrence after treatment

Study

No of

patients

(events)

EndpointCCP

scoreP PSA

Gleason

scoreRef

RP (RP

specimen)353 (132) BCR

Univ. 2.0 <10-8 <10—17 <10-9

Cuzick 2011

Multiv. 1.7 <10-5 <10-8 0.028

RP (RP

specimen)413 (82) BCR

Univ. 2.1 <10-5 0.003 <10-5

Cooperberg

2013Multiv. 2.0 <10-4 0.12 0.17

EBRT

(TRUS bx)141 (19) BCR

Univ. 2.6 0.002 <10-3 0.051

Freedland 2013

Multiv. 2.1 0.035 0.054 0.20

RP (TRUS

bx)582 (166, 12) BCR

Univ. 1.6 <10-6 <10-8 <10-4

Bishoff 2014

Multiv. 1.5 <10-4 <10-5 0.02

RP (GS6 low

risk) -TRUS

Bx

188 (56) BCR

Univ. 1.8 0.002 0.001 -

Tosian 2017

Multiv. 1.8 0.003 - -

Risk of progression after RP proportional to

TRUS biopsy CCP score - margin -ve only

Bishoff 2014 J Urol

Oncotype: only 1 proliferation

gene test - not significant

CCP score > 2 on biopsies predicted

25% with metastases after RP at 5

years

HR 5 (univariate)

HR 4 (multivariate)

n=582 Bishoff 2014 J Urol

CCP score > 2 on biopsies predicted 25%

with metastases after RP/RT at 5 years

independent of race, treatment type

Bardot 2017 AUAYears after diagnosis

Me

tasta

tic d

isease

n=767CCP 2 to 5

CCP 1 to1.99

CCP < 1

New Prolaris reporting forms show metastasis

risk at 10 years with definitive treatment

BCR after radiotherapy

predicted by CCP score

Freedland 2013 Int J Rad Oncol

CCP Score < 0

CCP Score < -1

CCP Score < 1

CCP Score >1

CCP Score predicts BCR,

metastasis & death

✔️Associated with meaningful endpoints ✔️

death ✔️

Metastases ✔️





Does biomarker provide additional

information above nomograms?


death ✔️

Metastases ✔️

Provide additional information above

nomograms e.g. CAPRA score




Clinical Risk Tools: (CAPRA)

CAncer of the Prostate Risk Assessment

Cooperberg JNCI 2009

Predicts:

• PSA failure

• Metastases

• Death

CAPRA score combined with CCP score

“clinical cell-cycle risk [CCR]score” -

predict 10 year chance of death

CCR = (0.39 x CAPRA + 0.57 x CCP)

CCR (CCP score combined with CAPRA)

improves risk classification on CAPRA alone

Cuzick 2105

Provides additional information

compared to nomograms


death ✔️

Metastases ✔️

Provide additional information above

nomograms ✔️





death ✔️

Metastases ✔️

Provide additional information above nomograms ✔️


Active surveillance Y/N?



Does the CCP score allow more

accurate risk assessment?

Conservative management fails because

risk poorly predicted at diagnosisKlotz 2015 JCO

Hamdy 2016 NEJM PROTECTWilt 2017 AUA

Moore 2017

Combined score (CCR) improves risks prediction

in candidates for conservative management

Cuzick 2015

CCR upgraded 14%

to >4% DSM

CCR downgraded 5%

to PCM < 4%

Most

useful

Capra

2 to 3?

Which CAPRA threshold?

Depends on life expectancy & man’s values

Life Expectancy

Typical patient for AS?

Cooperberg JNCI 2009

CAPRA 2

If CCR <0.8 no deaths at 10 y - good

candidate for conservative management

Stone 2014 SUOGS ≤ 3+4, < 25% core +ve, PSA < 10, Clin stage ≤ T2a

(AUA/ASTRO/SUO favourable intermediate risk or below)Cuzick 2015 AUA

CCR ≧ 0.8

CCR < 0.8

n = 505

New Prolaris reporting form shows the man

has very low chance of dying at 10 years

Young healthy men need to know

AS will “miss a window of

opportunity”

Currently, no reliable tools or markers that can predict

outcome for 20 years or more

Life expectancy

Age

Missing the window of opportunity would mean a

rising PSA after treatment

CCP > 0 predicts rising PSA after

RP in grade group 1 tumours Only low risk NCCNLow & Intermediate risk

Implication: CCP < 0 not to miss window of opportunity in young men

Which test to use or not?

Depends on life expectancy & man’s values

Life Expectancy

Age

groupComorbidity

Life

ExpectancyVery low Low

Favourable

intermediate

Unfavourable

IntermediateHigh risk

80 Healthy 12 WW WW WW CCR > 0.8 Treat

Normal 8 WW WW WW CCP > 2 Treat

Unhealthy 4 WW WW WW CCP > 2 WW

70 Healthy 20 CCP > 0 CCP > 0 CCP > 0 Treat Treat

Normal 14 CCP > 0 CCP > 0 CCP > 0 Treat Treat

Unhealthy 7 WW WW WW CCP > 2 Treat

60 Health 31 CCP > 0 CCP > 0 CCP > 0 Treat Treat


Unhealthy 11 WW WW WW Treat Treat

50 Healthy 44 CCP > 0 CCP > 0 CCP > 0 Treat Treat


Unhealthy 15 CCR > 0.8 CCR > 0.8 CCP > 0 Treat Treat

This is my speculation!

Treatment algorithm in 50 - 60 y.o. man with

localised, single lesion on mpMRI fusion biopsy

GG 2, 3(CR- & IDC-)

ProlarisCCP < 0

GG 2, 3

FT

GG 4/5

RP/RT

CCP 0 to 1

CCP > 1

AS:

PSA

mpMRI

Rebiopsy & redo CCP score < 3-5 y

GG 1

<3 mm ≧3 mm

CR+ or IDC+CR+

CCP threshold are MY arbitrary values


death ✔️

Metastases ✔️



Active surveillance/Watchful waiting ✔️

Influences physician treatment decision and

individualised treatment decisions

The CCR score allow more

accurate risk assessment


death ✔️

Metastases ✔️



Active surveillance/Watchful waiting ✔️

Influences physician treatment decision &

Individualise treatment decisions

Does the CCR score

influence decision making?

CCR score reclassifies low and intermediate

risk into lower or higher risk categories

Original Category

Reclassification with CCR score

Stone 2017 ASCON=16,442

If low risk initially, and CCR reclassifies

risk to intermediate, suggest treatment

AUA,

ASTRO,

SUO

2017

guideline

If intermediate risk, & CCR reclassifies

to low risk, consider active surveillance

AUA,

ASTRO,

SUO

2017

guideline

The reclassification makes clinicians

more or less likely to advise on treatment

Absolute change in visual analogue score after CCR scoreGonzalgo 2014 SUO

The reclassification by CCR score changed

treatment chosen by patients & urologist in 44%

Gonzalgo 2014 SUO. Crawford 2014 CMRO

44% changed choice 32% chose less treatment

CCP/CCR score: meets the

criteria for a useful Biomarker

•Predicts men at risk of metastases/death who are

conservatively managed or treated✔️

• CCR score (CCP & CAPRA) better than

conventional variables alone✔️

•Predicts men suitable for conservative management

or in need of adjuvant treatment after surgery✔️

• Influences clinicians and patients choices so more

choose conservative management ✔️

• Individualised treatment decisions ✔️

Case Examples

CASE

STUDY

Actual case

65

CASE

STUDY

Actual case

66

CASE

STUDY

1

67

CASE

STUDY

2

68

CASE

STUDY

2

Actual case reproduced with permission from Dr. Thomas Lanchoney

69

CASE

STUDY

2

70

CASE

STUDY

3

Actual case

71

CASE

STUDY

3

Actual case

72

CASE

STUDY

3

Actual case

73

Would what

would you do

with this man?

Age 55 years

PSA 6.5

14% free PSA,

clinical stage Stage T2

No Family history

No comorbidity

55 year old, PSA 6.5,

prostate vol 45 cc

mpMRI

T2

T2

DCE

ADC

55 year old, PSA 6.5,

Gleason Score 3+3=6, prostate vol 45 cc

mpMRI

T2

T2

DCE

ADC

CCP score poorly correlated with

mpMRI ADC value

CCP score:

< 0(blue)

> 0 (red)

CCP score & ADC value correlation coefficient 0.36, P=0.0003

ADC

Max T

um

our

length

Renard-Penna 2015 J Urol

Years after diagnosis

Dead from

prostate

cancer

Dead

from

other

causes

4% chance of dying after 10 y for low risk

disease in PSA era in 65 year old

Contains Gleason pattern 3 only

1 2 3 4 5

Prolaris 10 year prostate cancer specific

mortality rate was reduced to 1%

Continued with active surveillance

2 3 4 5 6 7 8 Prolaris score

72 years, PSA 8, pT2, Gleason pattern 3+4=7,

apical positive margin - adjuvant radiotherapy?

CAncer of the Prostate Risk

Assessment Post-Surgical CAPRA-

S

• CAPRA-S 0-2

low risk

• CAPRA-S 3-5

Intermediate risk

• CAPRA-S >5

High risk

CCP score & CAPRA-s

reclassifies low risk to greater risk

Cooperberg

2013

J Clin Oncol

Low risk patients

(CAPRA-s 0 to 2)

CCP score & CAPRA-s reclassifies

higher risk to lower risk

Cooperberg

2013

J Clin Oncol

Higher risk patients

(CAPRA-s ≥ 3)

Consider no adjuvant

treatment if CCP < -1

Prolaris post robotic prostatectomy -

15% chance of PSA failure at 10 years

Chose no adjuvant radiotherapy

&

Very Happy!

91

Prolaris Advantages

• independent predictor of death or metastases

• CCP score best used in combination with other

predictors e.g. CAPRA, CAPRA-s

• High biomarker scores at diagnosis indicate higher

chance of metastatic disease & need for adjuvant

Rx after primary treatment

Tumour adjacent ‘normal’ prostate

mutations may not affect survival

Weischenfeldt J, Cancer Cell 2013;23:159–70

Schlomm Eur Urol 2015

TMPRSS2:ERG fusion:

• commonest genomic alteration in PCa

• Seen in normal ‘prostate’ tissue’

CCP score & mpMRI

Prolaris to help make treatment decisions in localised prostate cancer

Health & Medicine