Dean Stell

1
Are any of these biomarkers really worth anything?
2
Types of inventions

• DNA
• SNPs, repeats, deletions, haplotypes, etc.
• mRNA
• Array chip type biomarkers
• Protein
• Antibody array mass spec.

3
Should you be patenting these things?
4
THE BAD NEWS
5
Market isnt developed yet.

• Diagnostic companies unsure of what to do with
  the information vomit coming from academia
• Searching biomarker on PubMed on May 4, 2004
  resulted in 275,647 hits!!!!!
• Genomics tests are akin to pharmaceuticals in
  that they must be marketed to physicians
• Different sales force for diagnostic company than
  selling reagents to perform electrolytes
• Difficulty for physicians to keep up with cutting
  edge genomics (e.g. they have a lot to do besides
  read Nature Genetics)

6
Market Problem - continued

• Eventual regulatory issues associated with the
  tests
• Required for reimbursement in many cases
• DNA-based tests have limited market potential
• One time only
• We are not on the verge of sequencing the
  patients genome
• Sowhere is the line between relevance
  irrelevance
• Weakness of method-of-use patents

7
The Good News
8
The Good news

• The market IS emerging.
• More genomic tests out there every day
• Diagnostic companies want unique exclusive
  tests which command higher margins
• My fundamental belief that all this genomic data
  will not sit on the shelf forever
• Some of it will be used

9
So, how do you decide what to patent????
10
Questions to ask

• Is the biomarker novel?
• Good Novel biomarker ? much better patent
  protection will be available
• Bad Known biomarker ? you may have a lot of
  trouble policing the patent
• How would the test be used?
• Good Routine screening (e.g. companion to PSA
  test?)
• Bad Only if other risk factors exist (e.g.
  parent diagnosed with cancer)

11
Questions - continued

• Prognostic or diagnostic?
• Good Protein based assay? larger market,
  diagnostic, tracking, many uses per patient
• Bad DNA? prognostic, smaller market, one-time
  only
• Is the disease prevalent?
• Good Prostate cancer, diabetes, etc.
• Bad Obscure genetic disorder

12
Questions - continued

• How relevant is your biomarker?
• Good 100 predictive/diagnostic
• Bad It indicates a 7 increased risk of
  developing cancer in the future
• So what? How would knowing this change your
  behavior or treatment?
• What population was studied?
• Good 800 Caucasian men
• Bad Inbred mountain family

13
Sofire at will IF

• You find a NOVEL biomarker that would be used for
  SCREENING and TRACKING based on GENE EXPRESSION
  of PREVALENT DISEASE that is 100 ACCURATE in a
  LARGE POPULATION GROUP

14
Andrun away IF

• You have a NON-NOVEL DNA BASED biomarker that
  would be USED RARELY to provide EQUIVOCAL
  PROGNOSTIC susceptibility to an OBSCURE DISEASE
  in a RARE POPULATION