103 · L2 Patenting in Biotechnology

Patent eligibility — the biotech trap

• Eligible: any new & useful process, machine, manufacture, composition of matter
• Exceptions: laws of nature, natural phenomena, abstract ideas
• Myriad (2013): isolated DNA not patentable (not "markedly different"); cDNA is (not naturally occurring)
• Ariosa v. Sequenom: detecting fetal cffDNA was "routine/conventional" → unpatentable natural phenomenon
• Fix: integrate into practical application / markedly different characteristics / unconventional methods

Obviousness & enablement

• Obviousness (KSR v. Teleflex 2007): teaching-suggestion-motivation + common sense. Overcome with unexpected effect, synergy, teaching away, commercial success
• Written description + enablement ("how to make & use" without undue experimentation). Biotech: use biological deposits + sequence listings

Patent lifecycle

Biopharma patent-eligibility problem zones

Biopharma sits on the fault line of the judicial exceptions. The slides flag specific recurring problem areas:

Abstract ideas — the third exception

Beyond laws of nature and natural phenomena, claims fail as abstract ideas when they cover:

• Mathematical algorithms / formulas / relationships
• Abstract steps merely performed by a computer
• Methods of organizing human activity
• Fundamental economic practices or concepts
• Human mental steps
• An idea with no particular concrete or tangible form

Myriad & Ariosa — the natural-product cases in detail

Myriad (2013): Myriad's invention related to isolated BRCA1 and BRCA2 genes. Although isolated genes were structurally different from natural genes (lacking covalent bonds linking them to the rest of the gene), they were not functionally different enough to avoid "improperly tying up the future use and study of these genes." Supreme Court: isolated DNA is not patent-eligible (not "markedly different"); cDNA is eligible because it is not naturally occurring. Myriad, 133 S. Ct. at 2111, 106 USPQ2d.

Ariosa Diagnostics v. Sequenom (U.S. Patent No. 6,258,540): a method for detecting paternally inherited fetal nucleic acid (cffDNA) in maternal plasma / serum. The Federal Circuit held the lab steps (amplification, detection, correlation) were "routine and conventional"; the claims started and ended with a naturally occurring phenomenon, so the only "invention" was discovery of a natural phenomenon → unpatentable.

Gilead's Remdesivir IP stakes (the case-study specifics)

The lecture dissects Gilead's actual patent family to show claim scope:

Novelty destroyers — the disclosure quiz

Which acts destroy novelty? The slide tests public disclosure vs. secret / protected act:

Parts of a patent application & who files

A patent application comprises: Title, Abstract, Description, Drawings, Claims. After filing, prosecution runs ~2–3 years (fast-track possible): examiner issues an office action, applicant responds, second OA / allowance, then grant.

Who can file? A registered patent agent / attorney, or (in the US) the inventors themselves.

Claim structure & the "chair" design-around

A claim has a structure: subject matter (preamble) → transition clause ("comprising") → elements (seat; four legs; back panel) → limitations (back attached to seat for resting the back). "Comprising" is open-ended — adding more elements does not escape infringement.

What you do with a granted patent

• Enforce — sue infringers
• Use as a leveraging tool in negotiation
• License to others
• Use for marketing purposes

Term extension & total market exclusivity (by region)

On top of the basic 20-year term you can add data / market exclusivity, orphan status, and pediatric exclusivity — these run from the basic patent term.

The patent attorney / agent

A patent attorney / agent is a specialized legal professional representing clients before a Patent Office. They draft, file and prosecute applications; handle design / trademark apps; do searches, due diligence, and advise IP strategy; and oppose granted patents.

Remdesivir's pre-COVID history (case-study background)

The card states Gilead patented Remdesivir's composition and coronavirus use, but omits the origin story the lecture gives. Per the source:

• Gilead originally developed Remdesivir to treat other diseases, including Paramyxoviridae and Ebola.
• Gilead filed numerous patent applications on Remdesivir — including the composition of matter and use in treating coronavirus infections (SARS, MERS) — from 2010 to the present.
• The Wuhan Institute of Virology filed its Chinese patent on Remdesivir for treating COVID-19 on 21 January 2020, building on Gilead's already-broad coronaviridae claims.

Obviousness — the statutory basis and Graham factors

The card cites KSR v. Teleflex (2007) for obviousness but omits the statutory anchor and the foundational case the lecture lists:

• Statute: 35 U.S.C. §103 — a claim may not be obtained if, notwithstanding novelty, the differences from the prior art would have been obvious before the effective filing date to a person of ordinary skill in the art (PHOSITA).
• Foundational case: Graham v. John Deere Co. — the Supreme Court case that frames the obviousness inquiry as three steps: (1) determine the scope and content of the prior art; (2) determine the differences between the prior art and the invention; (3) determine the level of ordinary skill in the art.
• The invention is obvious if any prior-art teaching would have motivated a PHOSITA to modify or combine the prior art to arrive at it.

Overcoming obviousness — the full toolkit

The card lists "unexpected effect, synergy, teaching away, commercial success." The lecture adds two more recognised routes:

Written description, enablement & Best Mode

The card mentions biological deposits and sequence listings, but omits two key requirements the lecture lists under getting protection:

• Written description (possession) — show you possess the invention (e.g. identify nucleic acids/proteins via Sequence Listings).
• Enablement — teach how to make and use without undue experimentation.
• Best Mode — disclose the inventor's preferred embodiment; the lecture notes that in China and Europe this is called SUFFICIENCY.

Prosecution timeline details — PCT publication & divisionals

The card's lifecycle table says "Provisional → 12 mo PCT/Paris → national phase 30 mo" but omits two milestone details from the filing-strategy slide:

• PCT publication at 18 months — the application is published ~18 months after the priority date (between the Paris Convention filing and national phase entry).
• Divisional applications (分案) — a parent application can spawn divisionals; the slide explicitly shows a "Divisional / 分案" branch off the family.

Career advice to biotechnology students (lecture close)

The card ends at the patent attorney role and never covers the speaker's closing career advice — which is explicit source material:

• Each path is different, and that's OK; take sufficient time to explore different options.
• You never know how each part of your journey helps later — even challenging experiences.
• Importance of "soft skills": connections / network, communication / people skills, never stop learning, and be humble — appreciate what others can teach you.
• Why the speaker moved from chemistry to law: patent law makes good use of a science foundation, keeps you at the edge of innovation, spans varied projects, and is big-picture / strategy work.

Markush claims — a large class held together by one structure

A Markush claim is the biotech and chemistry way of claiming a very large class of compounds together. The defining condition the source states is short but strict: the class must share a common structure. The shared structural core is what makes the class one claim rather than a wish-list.

The source's form for it:

Formula I supplies the common skeleton; the list of possible R groups generates the members. Every member of that giant set is covered by Claim 1 — that is the whole point of a Markush claim.

Genus vs species — the two-level claim structure

The source states the relationship in one line: Claim 1 is the genus — the class — and Claim 2 is the species — one member of it.

This is the same ladder as the paper-clip set, translated into molecules: a broad Claim 1, then successive levels of narrowing, each available as a fallback position.

For biologics, structure is expressed differently

Small molecules are defined by chemical structure; biologics must be defined by the tools that stand in for structure. The source gives three:

• Sequence listings — nucleic acids and proteins are identified by submitted sequence listings.
• CDR definition — "An isolated antibody comprising CDR1, CDR2, and CDR3, wherein …".
• Vector / host cell claims — the same invention claimed one level down the biological chain.

Sufficiency — the limit on how wide a genus can be

Breadth in a Markush or genus claim is not free. The source's rule for sufficiency / enablement is that the skilled person must be able to make the whole claimed range — not just one example. A genus supported by a single worked example is exposed: the claim reaches further than the disclosure teaches.

Biological deposits and sequence listings — the support tools

Two mechanisms exist precisely because some biological subject matter cannot be written down in words:

Both are enablement devices: they let a genus claim satisfy sufficiency where prose cannot reach, and they are the reason a biotech specification looks so different from a mechanical one.

Unpredictable arts — the double edge for genus claims

The source describes biology and chemistry as unpredictable arts, where a small change can have a large effect. That cuts both ways for a genus claim:

• Helps non-obviousness — an unexpected result from a small structural change is exactly what the inventive-step analysis rewards.
• Hurts sufficiency — if small changes have large, unpredictable effects, you cannot assume the whole claimed range behaves like the one example you tested. The broader the genus, the harder it is to enable.

Drafting checklist for a biotech genus claim

1. Is the class held together by a common structure? If not, it is not a Markush class.
2. Are the variable groups (R groups, sequences, modifications) defined clearly — not as "different" anything?
3. Can the skilled person make the whole claimed range, or only the example?
4. Is anything in the class a cell, strain or hybridoma that needs a biological deposit?
5. Are all nucleic acids and proteins identified by sequence listing?
6. Are the fallbacks drafted — sub-genus, species, salt, crystalline form, composition, use?