Research data and what it does for a patent

Where data does real work

Inventive step. A measured comparison against the closest prior art is far more persuasive than an assertion that the invention is better.

Section 3(d). A new form of a known substance needs evidence of a significant difference in properties with regard to efficacy - for a medicine, therapeutic efficacy, judged strictly. Improved bioavailability alone has been held insufficient.

Section 3(e). A composition needs evidence of synergy, compared against each component alone rather than against an untreated control.

Sufficiency. Where a claim covers a range or a class, results at more than one point are what make the claim credible.

The design decision

These comparisons are usually not the ones a research group would run to answer its own question. So the useful step is to ask, while the experiment is still being designed, what the patent objection will look like, and add the arm that answers it.

Data generated after the priority date can often still be filed, but it is more vulnerable, and it cannot rescue a specification that never asserted the effect it demonstrates. Assert the effect in the document as filed.

What to actually do

• Design the comparison the objection will demand, not just the one your paper needs.
• Compare against each component alone, not only against a control.
• Assert the effect in the specification as filed.
• Keep the protocol and the raw data.