Can “public prior use” be relied upon to invalidate European patents?

The “state of the art” for assessing novelty and inventive step at the European Patent Office includes “…everything made available to the public by means of a written or oral description, by use, or in any other way…”. Most of the time – in fact, nearly all of the time – written, printed publications form the basis of novelty and inventive step objections in examination / attacks in opposition proceedings. But not every product that is sold to the public (i.e. made available by “public prior use”) is described in detail through printed publications. So, while something that has already been sold cannot, in principle, be validly claimed in a new patent application, how do you prove the public prior use?

Consider an example.

Let’s say you’re a toothbrush manufacturer. You watch your competitors’ activities closely. New toothbrushes come on the market, are sold for a while, and then disappear from the shops again. You diligently buy packets of those new toothbrushes, keep the receipts, study them in your laboratory and then put them in your carefully catalogued archive. In parallel, you carry out searches for any patents or patent applications covering those new toothbrushes (and repeat those searches periodically for approximately 18 months to capture any newly published patent applications). You find no patents / applications covering those toothbrushes, either from their supplier or a third party. You therefore advise your internal business client that, if they wish to make and sell those toothbrushes in Europe, they can do so, i.e. that they have freedom to operate because there are no patents / applications standing in their way. Significantly, you also believe that you can maintain that freedom to operate if any new patent applications are filed covering those toothbrushes because you can rely on the prior use that you have carefully recorded to invalidate them. Right? Not necessarily.
 
The EPO requires that a party alleging lack of novelty by prior use shows:

1. When the use occurred

2. What was used

3. The circumstances of the use

We reviewed the EPO’s case law on each of these requirements in our recent article here.
 
Regarding i), you kept the receipt.
 
Regarding ii), you characterised the toothbrushes and carefully stored them in your archive.
 
Regarding iii), you have a signed declaration from the person who purchased the packets of toothbrushes stating that they purchased them over the counter.
 
You might therefore think that because you have evidence of the required “when, what, and which circumstances”, you can invalidate a patent covering those toothbrushes at the EPO based on a prior use attack that is certain to succeed. However, this is where things get interesting. The EPO has different standards of proof – “balance of probabilities” and “up-to-the-hilt” (or “beyond reasonable doubt”) – for assessing whether the evidence is sufficient to prove the prior use.

Sure, if your evidence is assessed on the balance of probabilities, you may well have done enough. However, the established case law at the EPO requires that in the case of lack of novelty due to public prior use, where all the facts lie within the control of the opponent, the standard of proof is not balance of probabilities; it is up-to-the-hilt. 

You might think that because the sales that form the prior use were not by you, the opponent, then the strict criteria should not apply because the patent proprietor could equally have walked into a shop and purchased the toothbrushes when they were on sale. But that is not how the EPO looks at it. The rationale is that the proprietor does not have access to any counterevidence to disprove the when, what, and which circumstances; they are limited to pointing out holes in your evidence.
 
A rare case in which the EPO did apply the balance of probabilities standard is T 363/96. This case concerned diapers (“nappies” here in the UK). The diaper wars between Procter & Gamble (P&G) and Kimberly-Clark (KC) in the ‘80s and ‘90s spawned many patent litigations, which in turn provided an unusual business opportunity for the firm Marketing Technology Services, Inc. (MTS). This firm bought absorbent articles, e.g. diapers, from around the world, kept the receipts, and took them back to their archive in Kalamazoo, Michigan. They then offered these absorbent articles and evidence of purchase (supported by declarations from the owner regarding the purchase) for sale to anyone who requested them, such as opponents in European opposition proceedings. The net effect was that these diapers were not ephemeral in the way that fast-moving consumer goods usually are; they were publicly available at the time they were purchased over the counter by MTS, and remained publicly available from MTS’s archive from then onwards. 

So, in T 363/96, the opponent (P&G) argued that KC’s patent lacked novelty over a diaper they had bought from MTS, who themselves had bought it over the counter in Japan. The Board decided that because the prior art diaper was not a product sold by P&G, but was a product bought by MTS on the open market and then sold to P&G, KC were in an identical position to P&G: they too could have bought the diaper from MTS (and, as it happens, P&G provided them with a sample diaper). Consequently, the opponent, P&G, did not have all the power to obtain evidence relating to the prior use; the proprietor, KC, had the same power to obtain evidence, meaning that the balance of probabilities test applied instead of up-to-the-hilt.
 

So, what can you do to make sure you can rely on the prior use that you have recorded? 

A first option is to ensure that the evidence that you collect would pass the up-to-the-hilt test, e.g. carefully record every step of the journey of the prior use product from its purchase to your archive and any testing done on it, noting any unique identifiers that confirm that all the evidence you gather relates to that product. It’s onerous, but achievable (e.g. see T 1682/09 regarding the prior use of a weighing and packing system).
 
A second option is to consider how you could move to the balance of probabilities test. For instance, can you align your position towards the facts in T 363/96? Again, it might be onerous, but if the ultimate effect is that you are able to use a product from your archive to invalidate a patent that threatened your freedom to operate, it may well be worth the effort.