DG-041 Single Dose Results

1
Genetic Risk Tests for early detection of
Prostate Cancer Jeff Gulcher MD PhD Chief
Scientific Officer and co-Founder Decode Genetics
2
Prostate cancer detection

• 27,000 men died in 2007, African-Americans at
  higher risk for cancer and for death due to
  cancer
• (15 years ago 40,000 deaths per year before PSA
  used )
• Best treatment is early detection
• We have tools for curative treatment if detected
  early
• More emphasis needs to be made on early detection
  vs late treatment
• The only major risk factors are ethnicity and
  family history of early prostate cancer (younger
  than 65)
• Need new ways to focus PSA and diagnostic
  procedures to those who are highest risk

3
(No Transcript)
4
(No Transcript)
5
Prostate cancer genetic risk markers discovery
and wide replication

• Discovery in white populations 2500 patients vs
  20,000 controls using 300,000 markers
• All 8 markers replicated in 4000 to 12,000 white
  patients and 30,000 white controls
• 4 of these markers have been replicated in
  smaller numbers African-American patients (small
  collections of 150 to 500)
• The remaining markers may not adequately tested
  in small numbers of patients may need to
  recruit more participants in these important
  studies
• In some cases the markers which are more
  important to African-Americans are different but
  near the ones first found in whites
• In other cases, the strongest markers for whites
  are not important for African-Americans and vice
  versa (example- the 9p21 heart attack markers do
  not impact AA risk)

6
Prostate cancer genetic risk test (deCODE
ProstateCancer)

• Measures 8 SNP markers over 6 regions originally
  discovered by Decode and confirmed in numerous
  populations covers 70 of the genetic risk
• The genetic risk profile ranges from 0.4 to over
  6 fold
• 10 of the white population is at 2 fold risk
  (33 lifetime risk), 1 has 3 fold risk (49
  lifetime risk)
• This is a risk test not determinative
• Defines patients more likely to have more
  aggressive cancer
• The risk profile is independent of risk due to
  family history

7
Case Study in the use of deCODE Prostate Cancer

• 48 year old white male in good apparent health,
• father diagnosed with localized prostate cancer
  at age 68
• ACS guidelines recommend screening with PSA
  beginning at age 50 unless family history of
  early prostate cancer lt 65
• deCODE Prostate Cancer results
• Relative risk 1.88 fold compared to general
  population risk for white males.
• Calculated lifetime risk 1.88 X 16 30
• Modestly higher risk for aggressive vs.
  non-aggressive disease

8
(No Transcript)
9
Case Study in the use of deCODE Prostate Cancer

• High risk prompted early PSA test by primary care
• PSA midrange normal at 2.0ng/ml
• High risk prompted urologist to perform
  TRUS-guided biopsy
• Positive in 3 out of 12 core biopsies 15
  volume
• Gleason score of 6 (3/3) intermediate grade
• Negative workup for cancer spread
• Radical prostatectomy with nerve sparing for
  likely cure
• Final pathology on resected prostate showed
  Gleason 7 (high-grade) in both lobes
• Followup PSA now 0.0 (presumed cure)

10
Risk Family History
5
15 High Risk account for 30 of prostate cancers
High Risk
Family History Alone
deCODE ProstateCancer

95
Family History
Average Risk
11
Prostate cancer early detection needs for
African-Americans

• Need larger collections of patients and controls
• To ensure rapid translation of discoveries made
  in large studies in white to African-Americans,
  who are at even higher overall risk
• Need to use in African-Americans the
  industrial-scale genetic approaches that have
  worked so well in whites
• Solutions
• Collaborate with patient organizations like PHEN
  and NMA to recruit more participants
• Continue and expand funding for collections of
  patients and controls by medical centers
• Combine the numerous projects at NIH and funded
  by NIH
• Do not cut the SCOR grant program which funds
  collections
• Begin to fund genome-wide association discovery
  projects for African-Americans (does not exist
  because large numbers of patients have not been
  collected Catch-22)