When do we need a freedom to operate opinion for our cell therapy product?
Only once the product is finalised and heading into the clinic. A full opinion on a construct, a formulation and a process that are all still moving will be out of date before it is delivered, so the right moment is when the lead, the vector, the payload and the manufacturing route are fixed. Before that point, freedom to operate is a scientific discipline rather than a legal exercise. Follow the science, solve a real scientific and commercial problem in the field, and generate your own innovation rather than repurposing or reusing someone else’s approach. That is what gives investors confidence the technology is genuinely yours, and it is also what leaves you with a robust patent estate of your own rather than a list of licences. At the early stage, an IP landscape search of competitor companies and technologies is of far more value than an expensive belts and braces FTO search for a product or technology that is not yet finalised.
Read more: Overcoming freedom to operate and patentability challenges | Bridging the innovation and commercialisation gap
Should we patent our manufacturing process?
Often the manufacturing process should not be patented, or at least not patented on its own. Process claims are designed around, become obsolete as the process changes, can be cited against your own later filings, and are hard to enforce because infringement is undetectable from the finished product. Where the process really is the advantage, a managed trade secret is usually the better instrument. That means identifying what is genuinely secret, restricting access to it, documenting it and covering it in employment terms and supplier contracts, because a trade secret exists only for as long as it is actively kept. The trade-off is that secrecy does not stop independent development, so a feature a competitor could arrive at alone, and could be caught using, is often still worth patenting. A hybrid IP strategy is therefore often needed, patenting aspects of the process that are discoverable or that can easily be reverse engineered, whilst protecting genuine trade secrets.
Read more: T 0868/23 | Beyond the process
Our CAR or targeting molecule is novel. Do we have freedom to operate?
Not necessarily, because novelty and freedom to operate are two separate questions. Novelty asks whether anyone has published your construct before. Freedom to operate asks whether making, using or selling it would fall within someone else’s claims, and a genuinely novel CAR can still sit squarely inside a broader granted right. Patents on targeting molecules cover the molecule per se, so an antibody, ADC or bispecific patent on your target can block you even though that patentee never contemplated a cell therapy. The same applies across the rest of the product. The vector, the payload, the editing platform, the manufacturing process and the cell source each need to be cleared, and clearance has to cover every territory you intend to manufacture or sell in. Changing modality does not solve it either, because an allogeneic or in vivo route brings its own crowded landscape with it.
Read more: The first CAR-T for a solid tumour | IP strategy for the universal cell therapy revolution
When should we file our first patent application?
When the data supports the claim you want, and not before. Filing too early produces an application that cannot be enabled across its scope. Filing too late risks a competitor publication, or your own conference abstract, becoming prior art against you. The workable answer is to align filings with the clinical programme, so the priority application is drafted once the lead is chosen and the supporting data exists, and to time later filings against regulatory interactions when release criteria and markers are settled.
Read more: Too broad, too early (T 0660/22) | Juno v Kite
Do we have to license third-party vector or delivery technology?
Often yes, but less often than companies assume, and the cost of assuming is high. Each component of a vector or a lipid nanoparticle may carry its own in-licence, and those royalties stack, so a product that is clear on paper can still be uneconomic once every licence is layered up. The better route is not to engineer around other people’s claims but to follow your own science. A programme that solves a real technical problem with its own innovation, rather than reusing an established approach, usually ends up outside the crowded space anyway, and it generates something patentable in its own right that is attractive to investors.
Read more: Navigating the complex LNP IP landscape | IP strategy for the universal cell therapy revolution
What IP do investors look at in diligence?
Whether the market is protected, for how long, and why a competitor cannot simply design around. Beyond that they test whether the company owns what it says it owns, so the chain of title from founders, universities and contractors matters, as does whether in-licences cover the field and territory actually needed. They will also look at whether claims cover the clinical product rather than an earlier construct, whether freedom to operate has been assessed at the right stage, and whether manufacturing advantages are documented and genuinely kept secret.
Read more: Bridging the innovation and commercialisation gap | Patents versus regulatory exclusivity
