Why the Dapa Decision isn’t Interesting

Dapa is a small molecule drug used as a diabetes medication.

In July 2025, the Court of Appeal of England & Wales agreed with the first instance decision that the GB dapagliflozin (“dapa“) patent is neither sufficient nor inventive. Subsequently, the Supreme Court refused to hear a second appeal, so the patent has now been cleared from the path and a generic version has already been launched. 

The dapa GB patent decision has been in the headlines recently in view of differences in patentability standards of the England and Wales (E&W) courts versus those of the European Patent Office (EPO). 

Dapa is a small molecule drug used as a diabetes medication. The dapa patent in force in the UK was the GB designation of EP1506211 (EP’211), which granted in 2007. It was granted in the name of Bristol-Myers Squib and wasn’t opposed before the EPO. EP’211 bears a filing date of May 2003 and claims priority from a US continuation-in-part filed in May 2002 (The validity of the priority claim is not discussed in this article). The dapa patent was licensed to AstraZeneca, who received the European marketing authorisation for dapa in 2012. 

Flicking through the granted specification of EP’211, it is apparent that the dapa patent claims a specific compound, but contains zero data. Paragraph 14 states that the “compound of formula I possesses activity as inhibitors [sic] of the [target]”, but any patent practitioner knows that such statements can easily be written off as mere unsubstantiated allegation if there is no concrete supporting evidence. Paragraph 114 states “(…) inhibitor activity of compounds of the invention may be determined by use of an assay system as set out below”. Whilst said assay is described, there is no evidence that the assay was actually performed on the claimed compound. Indeed, when it comes to the application of the drug as a medicament, the patent is littered with statements such as the “compound of structure I will be employed (…)” — so, the use of the future tense may be telling here?

Furthermore, the crux of the inventive step argument seems to be only the addition of a -CH2- group to the methoxyl of a prior art disclosure. 

Why the Case Isn’t Interesting

The discussions on UK versus EPO plausibility aren’t germane — the dapa patent isn’t interesting on the topic of plausibility, the dapa patent is interesting because it made it to grant in the first place! 

Some further points that are also actually interesting:

• With the benefit of hindsight, was the license deal still financially worth it? And if yes, by how much?

• Why were no data included in the patent application as filed? What were the non-rosy circumstances that led to this omission? Or was the pressure to file early simply too great?

• Did the due diligence work highlight the discrepancy and who was clever enough to proceed nevertheless?

• The litigation surrounding dapa resulted in delayed generic market entry. Will there be any repercussions to this? And how will that affect the above answers? ;-)

Key Takeaway

While headline debates focus on differences between case law and EPO practice, the dapa decision mere underscores old lessons: always provide concrete experimental support in your patent application. Remember that patentability is subject to evolving standards. What might get through to grant now, might not survive revocation proceedings when the patent really counts.