What Makes an Invention Patentable? The Four Core Requirements

Before a patent application can be granted, the invention it describes must satisfy four fundamental legal requirements under United States patent law. These requirements are codified in Title 35 of the United States Code — the federal statute that governs patents — and each one represents a distinct hurdle that an application must clear before a patent can be granted. Understanding these requirements is important not just for the filing process itself, but for how a startup thinks about its innovations from the very beginning. An invention that fails any one of these tests cannot be patented, no matter how scientifically impressive or commercially valuable it may be. The four requirements are: patentable subject matter (Section 101), novelty (Section 102), non-obviousness (Section 103), and adequate written description and enablement (Section 112).

Section 101: Patentable Subject Matter

The first question patent law asks is a foundational one: is this the kind of thing that can be patented at all? Section 101 defines the categories of invention eligible for patent protection, which includes processes, machines, manufactures, and compositions of matter. In practical terms, this covers a wide range of life sciences innovations — a new drug compound, a method of treatment, a diagnostic process, a medical device, or a genetically engineered organism, among many others. The scope of what qualifies is intentionally broad, and for most tangible inventions in life sciences, Section 101 is not the primary obstacle.

However, Section 101 also establishes important exclusions that have become particularly significant in life sciences and biotechnology. The challenges that Section 101 poses for diagnostic patents stem largely from two landmark Supreme Court decisions: Mayo Collaborative Services v. Prometheus Laboratories, decided in 2012, and Association for Molecular Pathology v. Myriad Genetics, decided in 2013. In Mayo, the Court held that a method of optimizing drug dosage based on the relationship between metabolite levels in a patient’s blood and therapeutic outcome was not patentable, because that relationship was a law of nature and the additional steps in the claimed method were too conventional to transform it into something patentable. In Myriad, the Court held that naturally occurring DNA sequences — including the BRCA1 and BRCA2 gene sequences at the center of that case — could not be patented because they are products of nature, even when isolated from the human genome. Together, these decisions sent significant shockwaves through the diagnostics and personalized medicine industries, calling into question the patentability of a wide range of innovations that depend on detecting or measuring naturally occurring biological markers to guide clinical decisions.

The practical consequence for life sciences startups developing diagnostic products is that claiming a method simply because it identifies a correlation between a biomarker and a disease state — no matter how scientifically valuable that correlation may be — is unlikely to survive Section 101 scrutiny without more. Patent applicants in this space must work carefully with counsel to draft claims that go beyond the natural phenomenon itself and incorporate meaningful additional elements that reflect genuine inventive contribution: novel sample preparation techniques, unconventional detection methods, innovative data analysis approaches, or specific technological implementations that are not routine in the field. This requires close collaboration between the scientific team and patent counsel from an early stage, so that the inventive elements most likely to satisfy Section 101 are identified, documented, and protected before filing. While the landscape remains challenging and continues to evolve — with ongoing legislative debate about whether Congress should act to clarify the boundaries of patentable subject matter in diagnostics — startups that understand these constraints and plan around them are far better positioned to build a defensible and commercially valuable IP portfolio.

Section 102: Novelty

Even if an invention falls within a patentable category, it must also be genuinely new. Section 102 sets out the novelty requirement, which provides that a patent cannot be granted for an invention that was already known to the public before the applicant filed. More specifically, an invention lacks novelty — and therefore cannot be patented — if every element of the claimed invention was previously disclosed in a single prior publication, patent, product, or other public source. That prior disclosure is referred to as “prior art,” and the patent examiner’s job includes searching for prior art that might undermine the applicant’s novelty claim. If even one prior art reference describes everything the applicant is claiming, the application will be rejected under Section 102.

For life sciences startups, the novelty requirement carries a particularly important practical implication: public disclosure before filing can be fatal to patent rights. Presenting research findings at a conference, publishing a paper, or even pitching to investors in a setting without a confidentiality agreement can constitute a public disclosure that triggers the novelty clock. In the United States, the law provides a limited one-year grace period following such a disclosure during which the inventor can still file, but most other countries offer no such grace period at all — meaning that an unprotected public disclosure can permanently destroy international patent rights. This is why life sciences startups are strongly advised to file at a minimum a provisional patent application before making any public disclosure about a new invention.

Section 103: Non-Obviousness

Novelty alone is not sufficient. Even if no single prior art reference discloses every element of an invention, a patent will still be refused if the differences between the invention and the existing body of knowledge would have been obvious to a person of ordinary skill in the relevant field at the time the invention was made. This is the non-obviousness requirement of Section 103, and it exists to ensure that patents are reserved for genuine inventive contributions rather than incremental or predictable advances that any competent scientist or engineer would have arrived at in the normal course of their work. It is, in many respects, the most judgment-intensive of the four patentability requirements.

In life sciences, non-obviousness questions frequently arise when a startup is seeking to patent improvements to known compounds, modifications of established drug classes, or new uses for previously characterized biological targets. For example, a medicinal chemist modifying a known molecule must demonstrate that the specific modification they made — and the result it produced — was not something a skilled colleague could have predicted with reasonable confidence from the existing literature. Courts and patent examiners consider factors such as whether the invention produced unexpected results, whether others in the field tried and failed to solve the same problem, and whether there was skepticism in the field that the approach would work at all. Strong evidence of unexpected results or long-felt unmet need can be powerful arguments for non-obviousness, and startups should document these factors carefully throughout the research and development process.

Section 112: Written Description and Enablement

The final requirement addresses not whether an invention is worth protecting, but whether the patent application itself does its job properly. Section 112 imposes two closely related obligations on the applicant. The first is the written description requirement, which demands that the application clearly convey that the inventor actually possessed the invention at the time of filing — in other words, that the claims being made are grounded in real, demonstrated science rather than speculative aspiration. The second is the enablement requirement, which demands that the application describe the invention in sufficient detail that a person of ordinary skill in the field could make and use it without undue experimentation. Together, these requirements ensure that the public receives the genuine disclosure of knowledge that the patent system is designed to incentivize.

For life sciences startups, Section 112 is frequently where well-intentioned patent applications run into serious trouble. The temptation to file early, before experiments are fully complete, can result in applications that claim broad protection for technologies the applicant has not yet fully demonstrated. Courts have increasingly scrutinized overly broad claims in biotechnology and pharmaceutical patents, finding that applications which claim entire classes of antibodies, gene therapies, or drug compounds without providing sufficient experimental support across the full scope of the claim fail the enablement requirement. This means that while filing early to secure a priority date is important, the application must still be drafted with enough scientific substance to hold up under scrutiny. Working closely with a patent attorney who understands both the legal requirements and the underlying science is essential to striking the right balance between breadth of protection and defensibility of the claims.

This article is part of our Life Sciences Startup IP Resource Center.