Transcript PSA Fact or Fiction The debate as it stands

PSA: FACT OR FICTION
THE DEBATE AS IT STANDS
Dr Charles Chabert
PSA Screening
Charles Chabert
European randomised Screening for Prostate Cancer
Charles Chabert
ERSPC

Initiated in early 1990s
Aim was to evaluate the effect of PSA
screening on death rate from prostate cancer


Specifically whether PSA screening could
reduce the mortality of CAP by 25%
Charles Chabert
Methods



182000 men
Ages between 50-74 (core group 55-69yr)
Seven European countries

Randomly assigned into group offered PSA
screening on average every 4 year

Control group that received no screening
Charles Chabert
Study Design

Power of 86% to show a statistically significant
difference of 25% or more in prostate cancer
specific mortality with a p value of 0.05

Basis of F/U through to 2008

On basis of overall level of compliance of 82%
& 20% contamination in the control group a
25% reduction in CAP mortality in screening
arm equates to 14% reduction on intention to
screen
Randomisation
Charles Chabert
Screening tests and indications for biopsy

Most centres used PSA cut-off of >4.0ng/ml

Some centres also used DRE and F/T ratios

In Finland PSA cut-off of 10.0ng/ml between 1991-1994
was initially used

Initially sextant biopsies, in June 1996 these were
lateralised
Italy transperineal biopsies

Charles Chabert
Results

5990 CAPs detected in screening group and
4307 in control group

Cumulative incidence of 8.2% and 4.8%
respectively
Incidence of bone scan positivity was 0.23 vs
0.39 per 1000 in SCR vs CON
41% reduction in Sc group (p<0.0001)


Charles Chabert
Results TRUS Biopsy
Gleason 6
Gleason >6
Screening Group
72.2%
27.8%
Control group
54.8%
45.2%
Chabert
13%
Chabert
pT2 (57.6%)
Charles Chabert
87%
(GS=7 74% GS=8-10 13%)
pT3 (42.4%)
Prostate Cancer Mortality


31 Dec 2006
Median F/U 9.0 years
CAP Mortality
Screening
214 deaths
Control
326 deaths
Charles Chabert
ERSPC
Charles Chabert
Results: Intention to screen analysis





PSA screening : significant 0.71 prostatecancer deaths per 1000 after median F/U 9
years
Relative reduction of 20% of CAP related
death for men between ages of 55-69years
1410 need to be screened to prevent 1 death
48 men treated
This can be reduced by not treating indolent
cancers
Charles Chabert
Prostate, Lung, Colorectal and Ovarian screening trial ( PLCO)
Charles Chabert
Study Design





Exclusion criteria:
History of PLCO cancer, current cancer
treatment and from 1995 having had >1 PSA
test in preceding 3 years
Between ages 55-74 years
Enrolled at 10 centres
PSA> 4.0ng/ml indication for biopsy
Charles Chabert
Study Design

1:1 randomisation

76 793men Randomized

38 343 in Screening group

38 350 in control group
Charles Chabert
Study Design

91% and 98% power to show a 25% and 30%
reduction in CAP mortality

Assumption of 100% compliance with the
assignment of screening and control

No reference made to the power of the study
at time of this analysis
Charles Chabert
PLCO
Charles Chabert
PLCO Results

Median F/U 11.5 years

Compliance 85%

PSA screening in control group 40% in first
year

Increased to 52% in 6th year
Charles Chabert
Results
Screened
Control
Prostate Cancer
(7 years)
2820
2322
Prostate cancer
(10 years)
3452
2974
289
341
F/U 67%
Gleason score 8-10
50% had Gleason 5 or 6
Charles Chabert
PLCO Results
Charles Chabert
Results
Charles Chabert
Conclusion

PSA screening associated with 22% increase
in CAP diagnosis

Compliance with screening 85%( expected
90%)

Charles Chabert
No change on CAP mortality
Results
Charles Chabert
ERSPC & PLCO
Similar goals for both studies
 Pilot studies in both
 Screening: execution of biopsies under study
group not clinical judgement
 Treatment left to regional centres
 ERSPC 4 yearly PSA ( Sweden 2 yr)
 PLCO Pre-randomisation limited to 1 in prior 3
years
 Annual PSA & DRE then 2 yrs PSA
 Regional centres made call on TRUS

Charles Chabert
Take Home Points

ERSPC shows effect of screening on CAP
mortality at 9 years
This amounts to 20% on intention to treat
analysis and 31% for men who are screened



ERSPC NNT=48
PLCO shows no difference
Charles Chabert
Lancet Oncology (online early publication)

20 000 men Randomised (Swedish cohort from ERSPC)
Median upper limit screening 69 (67-71)
Primary end point prostate cancer specific mortality
First planned report
Median F/U 14 years
CAP incidence 12.7% vs 8.2%
RR in CAP death 44%
293 men need to be screened
12 diagnosed to prevent 1 CAP death
Charles Chabert
CAP Mortality
Charles Chabert
Summary


“GPs should be offering a PSA test to 40 year old men in
conjunction with a digital rectal examination (DRE) after
discussing with them the subsequent potential issues.”
“Those identified as being at higher risk should undergo
regular tests; those at low risk should consider less
frequent testing.”
Charles Chabert
Summary
“A PSA level higher than 0.6 in a 40 year old is
considered higher risk, as is a level of higher than 0.7 in
a 50 year old, and regular monitoring is recommended
for these groups.


“There is firm data that PSA testing reduces the risk of
being diagnosed with advanced disease, and that
treatment of prostate cancer at an early stage can lead
to a reduced risk of death.
Charles Chabert