From October 2016 to March 2017 the team is joined by Guest Kats Rosie Burbidge and Eibhlin Vardy, and by InternKats Verónica Rodríguez Arguijo, Tian Lu and Hayleigh Bosher.

Thursday, 22 May 2014

When structure is wrong, patents may fall?

This moggy noticed this morning this article in Chemical & Engineering News (the magazine of the American Chemical Society) concerning an error that has been found in the structure of a compound known as TIC10 or ONC201.  The compound, which is the subject of various patents including US 8673923 and the pending EP 2701708 (in the name of Penn State Research Foundation, but apparently licensed to a compnay called Oncoceutics), is currently in clinical trials as an anticancer therapy.

Incorrect structure (top)
Correct structure (bottom)
But researchers at Scripps have found that the structure is wrong, and have applied for a patent on the correct structure.  This raises a whole host of questions - Is the first patent invalid and/or will it not cover the actual compound of interest?  Can the second patent be valid?

This Kat is put in mind of cases where the originally determined sequence of proteins has been found to be incorrect.  In the cases he is thinking of, the correct sequence has already been identified in the priority year, and so the downside has been loss of priority, not loss of the whole patent (or effective scope of patent protection). This happened for example in decision T 0080/05 where the European patent for the BRCA1 breast cancer gene EP 0 699 754 was maintained in amended form after the priority claim was held invalid because of a sequence error in the original US application.  The consequence of incorrect sequences is not always the same.  If by luck the difference is a "silent mutation", where different nucleotide sequences would in fact encode for the same amino acid sequence because of the degeneracy of the code, then there may be no problem. If there is an issue with the amino acid sequence itself, however, then the applicant is likely to encounter difficulties under EPO practice.

Some practitioners have felt that this is too strict, and that the law should accept the reality of sequencing errors, and should not punish the applicant who was able to correct the error during the priority year.  Such an approach however has not yet found favour at the EPO. Each situation will turn on its own facts, and depending on the technical consequence of the error, the legal consequence will likely be different.  Thus it is not possible to say in general terms what the consequence is.

Errors in small molecule structure are rarer, and this moggy cannot bring another one to mind.  Again, the precise consequence will depend on the facts of exactly how the first patent is framed; whether the claims are directed towards a structure or a compound defined in another way; and what other information there may be in the specification to enable the skilled person to determine the correct structure.  This Kat cannot say what applies in the present case, not least because he has not seen the second patent application.  It is logically possible however for neither patent to be valid.  If the first patent has claims directed towards the wrong structure, it may well not be infringed by the correct compound, and it is conceivable that this may not be correctable.  There may be issues of enablement as well if the actually claimed compound is inactive.  On the other hand, the patent may disclose enough for a subsequent application to the correct structure to lack novelty (based on the inevitable result of applying the prior art) or at least be obvious.  It will be jolly interesting to see what actually happens.

The article notes surprise from chemists that this compound progressed so far into clinical trials before this error was spotted.  It may be a while before we get another case like the present one, so Merpel advises to make the most of it.

Thanks to Dr Ross Forgan for alerting this Kat to the article.


Suleman Ali said...

On sequence cases one can sometimes use a definition based on the incorrect sequence to attempt to cover the correct sequence. For example a claim to homologues which have at least 70% identity to the incorrect sequence would normally cover the correct sequence and retain priority (as long as the incorrect sequence was retained on Convention filing, which doesn't always happen!). As the article points out though, once you've made an error it's easy to fall in the gap where there's no way of salvaging things from the old case or the new case.

Anonymous said...

This seems to have some similarities to the well-known Synthon BV v SmithKline Beecham case in the UK. IPKat readers will surely need no reminders, but this was a case in which Synthon's patent for paroxetine methanesulfonate identified the patented compound by reference to incorrect spectroscopic data:

[2005] UKHL 69, judgment at

Subscribe to the IPKat's posts by email here

Just pop your email address into the box and click 'Subscribe':