The role of patents in biomarker development

1
The role of patents in biomarker development

• Fiona Murray
• Class of 1922 Career Development Professor of
  Management
• MIT Sloan School
• February 2006

2
Balancing the Role of Patents
PROS
CONS

• Cost of monopoly disincentives for investment
  by others, high prices exclusion

• Incentives for further development of ideas due
  to monopoly exclusionary rights

• Opportunities to trade in the market for ideas
  exchange patents, license etc.

• Costs of having to trade fragmented pieces of
  intellectual property

3
How can you patent a biomarker?
4
Does it matter? Parties that can be impacted by
patents

• Physicians ( therefore patients)
• Impediments to research?
• Research exemptions?
• Other commercial firms ( therefore patients)
• Impediments to development?
• Patent pools or pre-competitive research?

5
Current patent landscape in biomarkers DNA
diagnostics

• Recently completed analysis of the patent
  landscape of the entire human genome
• Patents can include diagnostics, biomarkers,
  therapeutic uses, use in target identification
• Match this broad DNA patent landscape to patents
  on current gene tests

Jensen Murray, Science Vol. 310, 14 October
2005, pp. 239-240.
6
To map patents to human genes we assembled a
database of nucleotide sequences from issued US
patents

• 686,864 sequences from Genbank
• 109,766 parsed from patent full-texts
• Only from patents with a valid sequence listing

7
From this database, we extracted only sequences
that are explicitly mentioned in the claims

• Natural language modeling using simple type-I
  grammars
• Applicable only to claims using SEQ ID
  nomenclature

8
Our analysis show that 4,382 of the 23,688 genes
in the human genome are claimed in granted U.S.
patents
9
Private US-based firms tend to own the most human
gene IP
10
What about patents on existing gene-based
biomarkers? Gene-based diagnostic tests as a
proxy

• Data from genetests.org (funded by NIH)
• Of tests for 1200 diseases 200 use direct
  DNA-based analysis (sequencing of either select
  exons, targeted mutations, or the entire gene)
• Examined mapping from disease gtgenes genes gt
  patents

11
Preliminary Conclusions
TESTS Direct DNA-based tests on 200 diseases
PATENTS Of these 267 unique genes 97 (36) are
patented by 176 patents which means that 45
tests have to navigate patents (gt3 per gene e.g.
PS-1)
GENES Tests include analysis of 267 unique
genes (total of 325 different gene-disease
combinations)
One patent many genes One gene many patents
12
Conclusions

• Biomarker landscape already cluttered with
  patents for DNA-based biomarkers may be similar
  for other categories.
• Recommendations
• Need to examine map this landscape more
  systematically across all types of biomarkers
• Put in place solutions to deal with potential
  problems in advance role for industry, USPTO,
  NIH academia