Novartis Pharmaceuticals Canada Inc v Teva Canada Ltd (NOC) 2015 FC 770 O'Reilly J
2,255,951 / deferasirox / EXJADE
As discussed in Monday’s post, the ‘951 patent relates to iron chelators, which are used to treat
disorders caused by excess iron in the body [7]. As usual, the utility attack turned on the promise
of the patent. O'Reilly J’s Deferasirox decision illustrates once again that under the promise
doctrine the validity of the patent will turn on a meticulous reading of the disclosure. It is also
interesting as applying the recently developed distinction between an “overarching” promise, which will affect all
claims, and a promise that is directed at one claim only. One point I would quibble with is that
there seems to have been an implicit assumption that the stated utility is the same as the inventive
concept. Consequently, while O’Reilly J explicitly stated that (for the compound claims) he applied the low standard for
utility that applies in the absence of an explicit promise, the standard
he actually applied was higher than a mere scintilla.
The key sentence in the disclosure was that the compounds of
the patent “have valuable pharmaceutical properties when used in the treatment of disorders”
related to iron retention [15]. Teva argued that this was an explicit promise that the patented
compounds had been tested in humans and had been found to be therapeutically valuable [16]. While O’Reilly J remarked that this statement was “infelicitous,” and Teva’s
interpretation was a possible construction of the sentence, after reading it in context he
concluded that the better reading was that this statement made no express promise. I will not
review O’Reilly J’s reasoning in detail, except to say that while his careful analysis of the
disclosure does strike me as reasonable, that the validity of the patent turns on such a meticulous
analysis of the disclosure, which is intended to disclose the invention, not define it, is one of the
major problems with the promise doctrine.
While he found there was no overarching promise, following the FCA decision in Celecoxib
2014 FCA 250 (blogged here), O’Reilly J distinguished between an overarching promise which
applies to all claims, and a promise that applies to only some claims. In this case, the so-called
Formula II compounds, including deferasirox, were novel compounds and consequently were
claimed both as compounds and for the use in treatment of iron excess diseases [14]. The key
statement discussed above would have been an overarching promise which would have applied
to all claims, including the compound claims. Because O’Reilly J did not construe that statement
to be a promise, there was no promise which applied to the compounds claims [35], which
consequently survived the utility attack [39]. However, even Novartis’ experts acknowledged
that the use claims contained an explicit promise of a specific result, and O’Reilly J held that that
claim-specific promise had not been demonstrated or soundly predicted [41]. (Because the compound claims
were held to be valid, Novartis’ application for an order of prohibition was granted.)
Because no explicit promise applied to the compound claims, O’Reilly J held that “the usual, relatively low, utility requirement applies to those claims” [35]. However, the utility actually
applied was that compounds were useful “for their marked iron-binding characteristics as shown both
in vitro and, in animal studies, in vivo” [33], [39]. While this stops short of therapeutic utility in
humans, it is higher than the alternative suggested by Teva, which was simply that the
compounds had the ability to bind iron [18]. Teva argued for that alternative construction of the
promise in order to set up a utility / obviousness squeeze, on the basis that similar compounds
with an affinity for iron were known in the prior art, and it was obvious that the claimed
compounds would have similar characteristics [48]. O’Reilly J avoided this squeeze with a construction which came down the middle: “the inventive concept of
the ‘951 patent [is] a class of compounds with a capacity to bind to iron, which are soluble in
vivo and capable of inducing excretion of the iron complex” [49]. This, he held on the facts, was
not obvious.
My only quibble with this analysis is that O’Reilly J apparently treated the stated utility and the
inventive concept as being the same. So, he “would construe the compound claims of the patent
as follows: Claims 5 to 37 relate to the novel formula II compounds, including DFS [deferasirox]
(claim 32), which are useful for their marked iron-binding characteristics as shown both in vitro
and, in animal studies, in vivo [and] are sufficiently soluble to induce excretion of the iron
complex [33]. He then applied this same construction in assessing utility [39] and obviousness
[49]. In my view, there is no reason in principle why the utility and the inventive concept should
be the same, and indeed, generally, they are not. For example, if a compound is useful in
treating cancer, the inventive step may lie in discovering a method of synthesizing it, as for example in the separation of enantiomers. Nor do the inventive concept and
stated utility need to be part of the claim as construed.
It is now normal to talk about “construing”
the promise, but this is not the same as construing the claim. It is the invention as claimed which
must have the promised utility, but the utility is not necessarily part of the claim itself. In this
case, it seems to me that the compound claims should be construed simply as being the specified
compounds, namely deferasirox in the case of Claim 32. The inventive concept was a compound
which solved the problem at hand, so that would encompass the advantages over the prior art,
which is to say in vivo iron binding with sufficient solubility, which is the concept O’Reilly J
applied in the obviousness analysis. But a scintilla of utility could be less than that. So, iron
binding alone might be sufficient, even without sufficient solubility to allow oral administration;
after all, compounds in the prior art had been clinically used for treating humans even though
they could not be orally admininistered, so surely that must be sufficient utility to support a
patent. In other words, it strikes me that while O’Reillly J stated that he was applying the low
standard of utility that applies in the absence of an explicit promise, he actually applied a higher
degree of utility than the mere scintilla necessary to support a patent. I suspect he did this in order
to avoid the utility / obviousness squeeze. But there is no need to do so; the squeeze is avoided
simply because there is no reason why the stated utility and the inventive concept should be the
same. With all that said, I may be parsing O’Reilly J’s decision too finely, as the unnecessarily elevated “scintilla”
of utility did not make any difference in the final analysis.
No comments:
Post a Comment