'Clerical errors' and Calvinball: a disappointment for Lundbeck

In South Africa the Court of the Commissioner of Patents has handed down its 20 June judgment in Lundbeck v Cipla Medpro which deals with escitalopram (a substance that has recently been the subject of judgments in the UK, Australia, Germany and the USA). In short, patent owner Lundbeck makes an antidepressant, CIPRALEX, the active ingredient of which is escitalopram -- an enantiomer of citalopram. Lundbeck sought two things: the amendment of what it maintained was errors in two of its patent's claims and an interim interdict against Cipla Medpro, to stop it selling a generic escitalopram product in South Africa.

Although argument was heard on the validity of claim 1, this judgment deals only with whether certain admitted errors in claims 6 and 7 could be corrected as “clerical errors”. These errors had been found in all the corresponding patent specifications in the early 1990s. Lundbeck had corrected/amended the errors in other jurisdictions at that time, but had not done so in South Africa. The issues addressed in the judgment are thus somewhat different to those addressed in the judgments in relation to the corresponding patents in other countries.

Two of the three errors related to the designation of a (+)-enantiomer of an intermediate (in each of claims 6 and 7), while a proper reading of the specification revealed that the process described in the body of the specification used the (­)-enantiomer of that intermediate. Lundbeck argued that the errors were “clerical errors”, and that the court should therefore simply correct these errors under section 50 of the Patents Act and grant the interdict. Lundbeck also argued that, in the event the court found the errors not to be “clerical”, the court should grant the “corrections” by way of an amendment under section 51 (the difference between correction of clerical errors is important, as an amendment under section 51 cannot be granted if the claims of the patent after amendment would cover matter which was not previously covered by the claims of the patent before amendment).

The difficulty faced by Lundbeck was that the draftsman of the specification was dead, and there was no direct evidence available as to how the errors arose. Lundbeck filed affidavits by experts who stated that the skilled addressee would have recognised the errors, and would have realised that the references to one enantiomer in each of the claims were meant to be references to the opposite enantiomer [Perhaps by hidden operation of the Calvinball “opposite pole” rule?] Cipla Medpro’s experts agreed that the claims contained errors; and Cipla Medpro argued that the patent was invalid for, inter alia, lack of fair basis as a result of these very errors. None of Lundbeck’s experts provided any explanation as to how the errors could have arisen. However, in 1993, the inventors had made a declaration to the USPTO in support of an application for re-issue of the corresponding US patent, in which they stated that corresponding errors in the US patent had apparently arisen as a result of “unfamiliarity with, or poor choice of, nomenclature” by the now-deceased in-house patent agent.

In the circumstances, Southwood J found that

* Lundbeck’s failure to correct the errors, knowing that they affected the validity of the patent and all the while maintaining its patent monopoly, constituted good reason for refusing the corrections;
* There was no evidence that the errors were in fact “clerical” in nature, and they could thus not be corrected under section 50;
* The corrections sought by Lundbeck had the effect of broadening the scope of the claims, and were thus not permissible as amendments under section 51.

Having refused to make the corrections/amendments sought by Lundbeck, the court refused to grant an interim interdict against Cipla Medpro.

This information was kindly supplied by Tim Ball (Brian Bacon & Associates).