For Biotech & Pharma Founders
How to Patent a Biotech Invention
Biotech is the field where patents are most central — a drug or diagnostic can take a decade and a fortune, and the patent is what makes that investment possible. It is also the field where patent eligibility is most treacherous. Here is the strategy, and the §101 minefield to navigate.
Educational guide, not legal advice. Biotech IP is highly specialized — work with patent counsel experienced in life sciences.
Why patents are existential in biotech
No field depends on patents more than biotech and pharma. Bringing a drug or diagnostic to market can take a decade and enormous capital, mostly spent before any revenue. Without a patent to exclude copyists during that window, no rational investor would fund it — generics could launch the day after approval. The patent is what makes the whole investment possible.
Biotech is also the field where patent eligibility is most treacherous. Two Supreme Court cases — Mayo and Myriad — drew hard lines around what counts as an unpatentable product or law of nature, and getting on the wrong side of them can sink a patent. Strategy here is as much about navigating Section 101 as about the science.
Composition of matter — the gold standard
The strongest patent in biotech is a composition-of-matter claim: a patent on the molecule, compound, or composition itself. It is the broadest protection because it covers the substance regardless of how it is made or used — anyone who makes, uses, or sells that composition infringes.
For a genuinely new chemical entity, a composition-of-matter patent is the crown jewel. It is why the first patent filed on a new drug candidate — claiming the compound — is so valuable, and why so much turns on whether the molecule is truly novel and non-obvious over the prior art.
The Section 101 minefield: Myriad and Mayo
You cannot patent a product of nature, a law of nature, or a natural phenomenon. Two cases define the boundary.
Myriad (2013): naturally-occurring DNA is not patent-eligible simply because it has been isolated from the body — it is a product of nature. But synthetic cDNA, which does not occur in nature, is patentable. The lesson: a natural molecule in its natural form is off-limits; a non-natural or markedly different form can be claimed.
Mayo (2012): a diagnostic claim that observes a natural correlation (such as a metabolite level indicating drug dosage) and adds only routine, conventional steps is not patent-eligible — it claims a law of nature without an inventive concept. The lesson: diagnostics need a specific, non-routine technical contribution beyond detecting the natural relationship.
Navigating these is the central §101 challenge in biotech. The science may be brilliant, but if the claim reads as "a natural thing" or "a natural correlation plus conventional steps," it can fail eligibility.
Method-of-use and method-of-treatment patents
When you cannot get (or have lost) a composition-of-matter patent — for example, when the compound is known — you can often still patent a specific new use. A method-of-treatment patent claims using a compound to treat a particular condition, at a particular dose or regimen.
These are narrower than composition patents (they cover the use, not the substance) but genuinely valuable. Repurposing a known drug for a new indication, or a novel dosing regimen, can be protected this way — and method-of-treatment claims have generally fared better under Section 101 than diagnostic claims.
Publish or patent: the academic trap
Most biotech inventions are born in academic and research labs, where the culture is to publish. This collides head-on with patent law: your own publication, poster, preprint, or conference talk is a public disclosure that becomes prior art.
In the US you have a 12-month grace period after your own disclosure. But most of the rest of the world has no grace period — a single public disclosure before filing permanently destroys your patent rights there. Countless valuable inventions have been lost to a paper or a poster that went out before a patent application was filed.
The rule is absolute: file at least a thorough provisional BEFORE you publish, present, or post. Coordinate with your institution's technology-transfer office early — racing the publication clock is the single most important discipline in academic biotech IP.
The long game: patent term and extensions
A patent lasts 20 years from filing — but in biotech, years of that term are consumed by the regulatory review required before a product can be sold. To compensate, Patent Term Extension can restore a portion of the term lost to regulatory review (up to a statutory cap) for products subject to that review.
Patents also work alongside regulatory exclusivities, which run independently of the patent term. The interplay between patent protection and these exclusivities is central to how long a biotech product is actually protected from competition — and a major reason biotech IP strategy looks further ahead than almost any other field.
The university and licensing model
Because so much biotech originates in federally-funded academic research, the Bayh-Dole Act framework matters enormously: it lets universities patent inventions made with federal funding and license them to companies. The entire biotechnology industry was built on this model — the Cohen-Boyer recombinant-DNA patent, licensed broadly by Stanford, is the canonical example.
If you are building a biotech company on university-originated technology, the license terms and a clean chain of title are foundational. Investors will scrutinize whether the company actually controls the core IP — so get the licensing and assignment right from the start.