CASELAW - reviews of EPO Boards of Appeal decisions

T 437/21 – What can be considered a problem invention?

chat_bubble 8 comments access_time 7 minutes

The patent relates to a “Pharmaceutical composition comprising ivabradine hydrochloride polymorph IV”

Brief outline of the case

Claim 1 as granted was lacking IS over D2=WO 2013/064307. This document was the first document cited in the ESR and was qualified as X,D,Y. The same applied to AR1.

The OD thus revoked the patent

The proprietor appealed the revocation of its patent.

The board confirmed the revocation for lack of IS over D2.

The case is interesting as the proprietor tried to defend its patent as being the result of a problem invention.

The technical problem underlying the invention

The problem underlying the invention is to provide pharmaceutical compositions comprising ivabradine HCl polymorph IV which are stable and suitable for use on a commercial scale, see paragraphs [0003]-[0004] of the patent.

This problem is solved, following claim 1, by stabilizing the tablet by a moisture barrier as claimed, created by means of packaging the tablet in blister pack material.

There is no debate that D2 is a suitable starting point for the assessment of inventive step.

The assessment of IS depended on whether or not, the claimed invention qualify as a “problem invention”

The proprietor’s position

According to the proprietor, the objective technical problem is the provision of a stabilized pharmaceutical composition of ivabradine HCl form IV. The proprietor relied on paragraphs [0025]-[0028] of the patent in support.

Paragraphs [0025]-[0028] of the patent compare the stability, at 40°C and 75% RH for periods of 2.5 to 6 months, of ivabradine HCl form IV tablets (uncoated and coated) in different blister pack materials. It is found that the use of a cold form foil (CFF, or Alu-Alu foil), which has a WVTR value within the claimed range (see paragraph [0015]), prevents the conversion of form IV into form ß.

In contrast, the use of blister pack materials with lesser moisture barrier properties, i.e. higher WVTR values (PVC, Duplex) leads to partial conversion into form ß.

According to the proprietor, D2 describes form IV as stable.

The problem of instability of ivabradine HCl form IV as part of a pharmaceutical composition would not have been observed as part of the skilled person’s routine and was hitherto not recognized in the art.

The board’s position

The Board was of the opinion that a technical effect of stabilizing form IV against conversion into form ß in tablets under such humid conditions is credibly shown.

The technical problem may be thus formulated as the proprietor suggests, namely as the provision of a stabilized pharmaceutical composition of ivabradine HCl form IV, at least to the extent that the stability refers to this stability against this polymorphic conversion.

The Board did not share the proprietor’s view. 

D2 firstly discusses general aspects of polymorphic behaviour of drugs (see page 4, second paragraph). The document differentiates between the most stable polymorphic form and metastable forms, and mentions that the transformation of metastable polymorphs to the stable polymorph may be facilitated by moisture (see page 4).

D2 then emphasizes the need to produce ivabradine HCl Form IV in “a form having constant physical properties”, and indicates that form IV is “a thermodynamically stable polymorphic form under certain experimental conditions of isolation or purification” (see page 5, emphasis added by the Board).

The skilled person would not understand this qualified statement as indicating that form IV is the most stable polymorphic form of ivabradine HCl. The further statement of D2 that ivabradine HCl form IV is “stable and easy to handle” (see page 5, lines 8-14) also does not teach that form IV is perfectly stable.

D2 does not teach either that form IV is perfectly stable when formulated in a tablet. In particular, in example 26 (see page 17), the description of the process for preparing a tablet comprising ivabradine HCl form IV is followed by the conclusion “The absence of any transformation of the crystalline Form IV of ivabradine hydrochloride was confirmed”. This statement does not refer to storage stability or stability upon exposure to moisture, and would rather be read as referring to stability during the preparation steps.

Thus, there is no teaching in D2 that ivabradine HCl form IV is perfectly stable, whether as such or formulated in a tablet.

In the case at hand, D2 neither says whether the stability over time of ivabradine HCl form IV in a tablet formulation may or may not be improved.

However, the skilled person would in any case have had to address the stability issues of formulated form IV as part of routine works towards development of the tablet for commercial use.

As acknowledged by the proprietor, and as explained in D2 (see the paragraph bridging pages 4 and 5), for the purpose of registering a pharmaceutical product, the stability of the active ingredient within the pharmaceutical composition needs to be examined in any case.

It is also part of the common general knowledge that not only the drug substance (here: ivabradine HCl form IV) but also the finished dosage forms (here: immediate release tablets containing it) must be tested for stability and constant physical properties.

Thus the skilled person would in any case have to verify the stability of the immediate-release tablet comprising ivabradine HCl polymorph IV, irrespective of any statements regarding its stability or lack thereof in the prior art D2.

The proprietor did thus not demonstrate that observing the instability of ivabradine HCl form IV in a pharmaceutical composition would have required any undue effort. On the contrary, the post-published evidence D24 (Stability experiment carried out on tablet formulations of ivabradine hydrochloride (Form IV) according to Example 26 of D2) shows that, when subjecting the tablets of example 26 of D2, containing ivabradine HCl form IV, to the standard stress conditions 40°C / 75% RH for a month, a complete loss of form IV is observable.

Contrary to the proprietor’s position, the use of such accelerated stability testing conditions is not uncommon and is recommended in D32, see Annex 5  (“WHO Expert Committee on Specifications for Pharmaceutical Preparations” Thirty-fourth report of 1996, p 1-194.

Special considerations on problem inventions in the decision

An inventive step may, in certain circumstances, be acknowledged on the basis of the discovery of an unrecognised problem, even if the claimed solution is in itself obvious (see also the Case Law of the Boards of Appeal, 10**(th) edition, 2022, I.D.9.12).

However, the posing of a problem which the skilled person would have encountered in the course of routine work cannot represent a contribution to the inventive merit of a claimed subject-matter.

It is therefore not sufficient that the prior art is silent about a problem for the claimed invention to qualify as a problem invention. As explained in T 252/10, cf. point 8 of the reasons, the absence of a hint in the prior art that there might still be a desire for further improvement does not mean that an unrecognized problem has been discovered.


When discussing IS and the solution of the problem, the board used “credible” instead of “plausible”. One wonders if it is not a tribute to G 2/21.

The notion of problem inventions is resurfacing regularly. Rarely with success for the applicant/proprietor. See for instance T 703/18, T 791/92, T 630/92, and T 119/82 among many other decision with the same conclusion.

It is indeed rare that a board concludes that there is a problem invention, i.e. that the invention lay in the formulation of the problem, whereby the solution to this problem appears trivial.

There are very few decisions actually positively dealing with this topic.

Courtesy of T. Debled, Professional Representative, I can show two of those.

Anti-flatulence tablet

In T 2/83, the problem with which the claimed invention was concerned was to provide an improved anti-flatulency effect side by side with an antacid effect in the stomachs of patients. It was standard to avoid a barrier layer isolating the antacid from the simethicone, but here the effect was surprising.

Layered varieties of tablets, which were also obtainable on the market, showed a substantially reduced anti-foaming activity in comparison with a tablet after prolonged storage containing a barrier according to the application under appeal. The modification provides a striking difference in performance, although the presentation of simethicone in a solid adsorbed form on at least 20 to 40 times the quantity of lactose together with starch and other carriers should have been sufficient to prevent migration and inactivation.

In T 2/83, the catchword reads:

The discovery of an unrecognised problem may in certain circumstances give rise to patentable subject-matter in spite of the fact that the claimed solution is retrospectively trivial and in itself obvious (“problem-inventions”).

In a case where the applicant had supplemented a known layered tablet by the provision of a barrier between the layers, the Board held that the proper question to be asked was not whether the skilled man could have provided the barrier but whether he would have done so in expectation of some improvement or advantage.

The tablets claimed in the present application are therefore novel, and show, in consequence of the inserted barrier, an improved performance.

The improved cat flap

If a cat, when trying to go through a cat flap which is top hung, the cat changes its mind and decides to stay on the same side, the paw gets stuck and the cat hurts itself. The harder the cat pulls back its paw, the tighter the flap jams. The solution to this problem is to provide an enlarged clearance for the cat’s paw between the tunnel portion and the bottom edge of the flap when the flap is displaced in at least one direction.

Paw trapping incidents may well have occurred repeatedly but hardly with the same cat since it either would have learnt how to proceed through the door or would have avoided the door in the future.

The succession of incidents would thus have concerned different doors and different cats at different places. In some cases the cat would have freed itself but of course could not report to the owner why the accident had happened, indeed the cat door might not even have been suspected to be the cause.

It cannot be expected that the cat owner is present each time the cat enters or leaves the building because the purpose of a cat flap is precisely to make such presence unnecessary.

The board thus has substantial doubts as to whether the recognition of the paw injury problem was obvious and accordingly gives the benefit of the doubt to the applicant on this point, so that the recognition of this particular problem contributes to the inventive step of the subject-matter of Claim 1.


Problem inventions do exist, but much less frequently as argued by representatives.

Share this post


8 replies on “T 437/21 – What can be considered a problem invention?”

francis hagelsays:

Dear Mr Thomas,

I agree that this case is not a persuasive problem invention. But problem inventions are not so infrequent. There are famous cases such as the Windsurfer and the baby’s drinker cup fitted with a valve to avoid spilling.

I reviewed a case recently relating to a container for specific items also designed for entertainment. The potential for entertainment was a new problem. This was clear from the description and to some extent from the original claims, but both the applicant and the ED missed the point and this made the exam unnecessarily protracted. The responsability was primarily the applicant’s because of a fraught claim drafting.

It seems difficult for professionals to realise and highlight the novelty of the problem, when that occurs. I suspect the EQE training is so efficient and the PSA doctrine so deeply ingrained in European patent attorneys that they simply cannot conceive of a case in which the PSA should not apply, or if they do, they would perhaps fear that an argument to show that the PSA does not apply would be perceived as offensive by EDs.

Do not get me wrong, I think the PSA is a good tool, more predictable esp. than the US approach. But it has its limits, as in the case of problem inventions, and I see no reason to turn a blind eye. And its use often entails difficulties. For example, the principe that there must be one technical problem is often over-simplistic. In real life, an inventor typically faces a host of constraints determined by the field of use, which must simultaneously be met by the invention. In that sense, the « problem » is a complex of objective(s) and constraints. Or the need to define a « problem » may lead to a definition which comes close to being meaningless, as may be the case when the problem is to « find an alternative ».

Another common difficulty arises from claim drafting which ignores the specific field of use of prime interest to the applicant. This is a sort of reflex for many practitioners (probably because field of use claim limitations are adverse on what they consider their “added value”). If the applicant has to narrow down the claims to that specific field of use, the « art » of the skilled person changes and also, as a result, the closest prior art.

For this reason, I think it would be advisable to emphasise as a first step of the PSA the definition of the relevant field, instead of starting from the definition of the closest prior art. This would help practitioners pay attention to the field of use of interest to the applicant.

Max Dreisays:

Mr Hagel, I am not convinced that the patentability of that rare bird, a genuine “problem invention”, is any less susceptible of analysis under EPO-PSA than any routine invention (where there was no inventive insight in perceiving the OTP). I wonder what Daniel Thomas would say about that proposition.

For me, the debate about what is obvious is always in danger of spiralling down into arm-waving and subjective opinions. What the Americans call “TSM” (teaching-suggestion-motivation) helps to exclude ex post facto (subjective) analysis but, unfortunately, TSM doesn’t work because it is not constrained by the strict methodology of the EPO’s version of TSM, namely PSA. TSM seems to have fallen into disrepute in the USA, which doesn’t help the Americans to believe in EPO-PSA.

What is so problematic about applying PSA to a problem invention. I (the inventor) state at the outset that it was inventive to perceive the OTP. Can you convincingly refute that assertion? Yes or No. In the vanishingly rare case that the answer is No, then my invention wasn’t obvious. End of story

If yes, however, then the debate defaults to the further defined steps of EPO-PSA.

Of course, the debate as to whether it was a patentable contribution to the art to identify the OTP might involve a lot of arm-waving on both sides. But as it is only going to happen on rare occasions, where’s the harm in that?

Avatar photoDaniel X. Thomassays:

Dear Mr Hagel,

I am following Max Drei’s point of view rather than yours.

Like Max Drei, I am also not convinced that the patentability of that rare bird, a genuine “problem invention”, is any less susceptible of analysis under EPO-PSA than any routine invention (where there was no inventive insight in perceiving the OTP).

There is a claim and a piece of prior art considered the CPA and then we look at the difference, the effect of the difference and subsequently we define the OTP. It will be only in very rare occurrences that the problem will, as such justify, inventive step.

When the problem is to find an alternative it simply means that the invention claim does not produce any effect different than that already obtained with the prior art. Where is the problem?

EQE training is simply reflecting the position taken by the boards when they devised the PSA in order to asses IS. Nowadays, you will hardly find a decision of a BA not using the PSA. It was devised in chemistry and has slowly spread to the remaining technical areas. The last ones adopting the PSA were mechanical boards.

Having met chores of European patent attorneys a few years after passing the EQE, I would not think that the PSA doctrine so deeply ingrained in them. What I have seen and heard, looked like a PSA, but was anything than a proper application of PSA.

I would agree with you that the PSA is no more than tool, and as long as I have been an examiner. I never used it, as it simply did not exist at that time. It is by training candidates to the EQE that I became most knowledgeable of the PSA, but not in order to be successful at the EQE. It was rather to apply the case law of the boards.

In a separate comment on T 2324/18, I gave you my thoughts on the choice of the PSA as it can be derived from recent case law of the boards.

In spite of what you think, problem inventions are relatively rare. Besides the examples you have quoted and those I found, I do not know any other.

Max Dreisays:

Perhaps the “blue squash ball” case is an example of a problem invention, even though there is a debate whether its objective problem is technical.

Claim: Squash racquets ball, characterized in that it is blue.

Problem: how to enable players more quickly to assess the flight of the ball (against the white walls of a squash court).

Did anybody even dream that the black ball, conventional back then, suffered from such a problem? Could anybody seriously have asserted that the problem was “known”?

But the inventor’s rigorous research had revealed that, against a plain white background, the human brain can more quickly pick up the flight path of a small flying object when it is blue rather than black.

These days, does anybody still use a black squash ball? Does any player use anything but blue?

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

I would not consider the blue squash ball as a problem invention.

It is the result of the discovery that “the human brain can more quickly pick up the flight path of a small flying object when it is blue rather than black”.

This discovery has a practical and technical application to a squash ball. In this context the colour has a technical function and patentability should not be at stake.

Quite a few improvements are the result of discoveries which have been translated in technical items. Just think of the blue or the white LEDs.

Max Dreisays:

Well yes, Daniel, I also think that the objective problem is technical, but putting to one side the entire debate “technical Y/N” I do not yet understand your reasoning, why the blue squash ball is not a member of the set we call “problem invention”.

My understanding of the meaning of the expression “problem invention” is one where the skilled person was not conscious of the OTP, where the OTP was not “known” to the imaginary person skilled in the art, where the OTP was not a consideration within the thinking proceessses of the skilled person.

Back to the blue squash ball. Suppose that in military research circles it was already known that pilots can see against a white background a small flying object faster if it is blue than if it is black. So, as soon as the squash ball maker reads the military research report they are motivated to colour their balls blue. The solution is obvious, as soon as you formulate the OTP.

The point is though, that none of those squash ball makers ever supposed that there was any “problem” for a squash player to pick out the flight path of their conventional black balls. They never for one second supposed that the pick out time of their black ball against a white background was anything less than already optimal and therefore incapable of any improvement.

Is this not a fine example of a “problem invention”? If not, why not?

I realise that in running a hypothetical past you I am on dangerous ground. It is just that you and F Hagel seemed to be desirous of having more “problem invention” cases to discuss, so I though I might be able to help.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Thanks for your comments. For a start, I was not desirous of having more “problem invention” cases to discuss. I do however appreciate that you wanted to help

That a problem is always underlying an invention should not be in debate. I cannot imagine an inventor not wanting to improve a product or process which has shown some drawbacks. Even for pioneering inventions, like for instance photocopying as we presently know it (with the selenium drum), the inventor wanted to get read of the existing means for reproducing documents.

The thinking process of an inventor thus always implies a problem. The OTP is derived from the effect of the difference of the claimed invention when taking into consideration the CPA.

In some cases the OTP is the same than the problem the inventor faced. In the absence of a different effect, the OTP is less ambitious and boils down to find an alternative. I would thus claim that in most of the cases the OTP is not known to the person skilled in the art. This does not make all inventions to become problem inventions.

A problem invention is an invention in which the solution can be, as such, obvious but for which the skilled person never thought there could be a problem. The best example to me is the cat flap.

In the case of the blue squash ball, with the prior art you mention, there cannot be any inventive step. If in military research circles, it was already known that pilots can see against a white background a small flying object faster if it is blue than if it is black, then the application to a squash ball is straightforward and cannot imply inventive step.

Even without any prior art, I cannot see the blue squash ball being a problem invention. The original problem is how to improve the visibility of the squash ball flying against a white background. It appears indeed counter intuitive that a black colour does not lead to the best visibility.

On the other hand, it appears irrelevant to me that squash ball makers ever supposed or not that there was any “problem” for a squash player to pick out the flight path of their conventional black balls.

It is the result of the discovery that “the human brain can more quickly pick up the flight path of a small flying object against a white underground when it is blue rather than black”. The considerations relating to the visibility of the blue squash ball can be considered as the result of an academic study. A manufacturer of squash balls would never a priori spend one € on this kind of study.

I agree with you that coming with hypothetical examples is always delicate, but to me I cannot see that manufacturers of squash balls ever found there was a problem. Applying the PSA in this situation can lead to recognizing IS, but I could also envisage a board coming to a different conclusion. In any case the colour of the ball has a technical effect.

Max Dreisays:

Daniel, I thank you for your time and effort, to get across to me your understanding of what “problem invention” means, namely: an invention in which the solution can be, as such, obvious but for which the skilled person never thought there could be a problem. That is my understanding too. In each case (such as a cat flap or squash ball) the parties could argue the issue, adducing whatever evidence they choose, and the tribunal would then freely evaluate that evidence and decide whether or not the claimed subject matter can be designated a “problem invention”. And I think we also agree, that all of this can be accommodated without problem (ha ha) under the umbrella of the EPO problem-solution approach. Hooray!

Leave a Reply

Your email address will not be published. Required fields are marked *