1. SP on May 2, 2005 11:14 AM writes...

Did the patent issue, or is it just an application?
If it's only an application, you can do whatever the hell you want until it issues, and deal with it then.
If it's issued, I believe the case history that applies is when something is used as a research tool rather than sold directly for profit. That states that the only damages allowed, if any, are what would have reasonably been expected for A to charge B as a fee for using the compound as a research tool. A has no rights to any revenue stream from B's new product that was discovered with A's compound.
Now, if all that research was done before the composition of matter patent even issued, I don't think A is entitled to anything at all. The real question is who at B was dumb enough to tell someone at A they were using the compound to do internal research- would A have known otherwise? (Was the info about using the compound disclosed in an IND?)

2. MDL on May 2, 2005 1:19 PM writes...

Of course you're talking about Merck vs. Integra.
I agree with Derek that this somewhat unusual Supreme Court case has at least the potential to yield a fundamental decision regarding how pharmaceutical companies conduct early exploratory medicinal chemistry research. I for one, would appreciate some clarity on these issues.

I also disagree with the comment by SP, that simply obscuring this fact, e.g, the"question is who at B was dumb enough..." is an effective way to deal with type kind of infringment threat. At least for right now, this would kind of information is critical in the determination of liability for triple damages for willful infringement.

3. SRC on May 2, 2005 3:46 PM writes...

You're implictly referring to the "experimental use" exemption, which pertains to those whose experimentation involves ascertaining how the invention works and/or how to improve it, thereby advancing the public policy goal of promoting progress (sorry for the alliteration). In its absence, no one could attempt to improve an invention during the patent term without being liable for infringement. (Recall that it's perfectly possible to obtain a patent on an improvement that would infringe if the improvement were practiced.)

Merck v. Integra is quite different. First, it concerns the "safe harbor" provision of the Hatch-Waxman Act that allows otherwise infringing activities if they are reasonably related to preparing an FDA application. This provision was included to allow generic manufacturers to begin selling patented products as soon as the patent expired. (Otherwise, the patentee would in essence receive a non-statutory addition to his patent term.)

In my view, this is pretty clear cut. Experiments on a patented invention are OK; experiments involving use of a patented invention (in the fashion disclosed by the inventor) are not.

4. gcb on May 2, 2005 9:56 PM writes...

SRC,

While I think that you're on the right treack, the line between experiments "on" an invention and experiments "using" an invention isn't always that clear cut, especially if the invention is a method. I think a better way to express the dividing line is to look at the purpose underlying the experiment. If the goal is to study or improve upon the claimed invention, then there is no infringement.

For example, use of a claimed process as a control doesn't infringe. If, however, the invention is merely being used for its intended process--e.g. a Sharpless epoxidation, PCR--then there is infringement. I think that this is the dividing line that Justice Story was trying to express in his formulation of the experimental use exception. Once you get to R&D uses of inventions for their intended purpose, there are two questions--how to identify infringers and how to measure damages. That's what Judge Rader was saying in his Embrex concurrence and I think he was right.

Merck v. Integra raises a statuory construction issue. In the earlier parts of the case, Merck shot itself in the foot by arguing that the provision essentially immunized the entire pharmaceutical industry from liability because everything it did was "reasonably related" to data submitted to the FDA. That overreach probably cost them more than anything.

The danger here is that the Supreme Court will issue unclear rules that purposrt to attempt to clarify an inherently unclear concept--"reasonably related." Instead, the Court should accept that there is a line to be drawn and that deference should be accorded to the District Court's findings on the issue

5. SRC on May 2, 2005 11:39 PM writes...

gcb,

Thank you for paraphrasing my point so elegantly.