Drug Combination Inventive Over Prior Concomitant Use

Published March 20th, 2012

Allergan Inc. v Sandoz 2011 FC 1316


After conducting stability tests and a human trial, the Federal Court accepted that Allergan’s fixed dose (i.e., single formulation) drug combination for glaucoma was not obvious[1].  The combination had been previously used (as concomitant separate formulations) when monotherapy did not achieve sufficient reduction of ocular hypertension (paras. 19, 78).  Overall, it appears that several factors contributed to this finding:  (1) extensive human tests in the patent (para. 64); (2) specific concerns regarding combinations in the area of glaucoma (para. 115(i)); and (3) extensive work conducted by the patentee (para. 115(ii)).  The court distinguished an earlier decision regarding an obvious drug combination for ocular hypertension (Merck v Apotex 2010 FC 1042) (at paras. 119, 120).  The Allergen decision introduces possible ambiguity as to the requirements for a valid combination.  There may be no requirement for synergy with combinations (i.e., that a combination be better than the added efficacy of the known individual drugs).  It appears that one can rely on various unobvious results/properties of a combination, where these properties are included in the patent disclosure.

On its face, in some ways the court appears to have treated the Allergan patent as a selection (selection patents of course do not require the advantage or selection to be in a claim), but at the same time did not require Allergan to have demonstrated any improvement over the genus of concomitant combinations.  The Allergan patent contained only a comparison of the new combined formulation and the separate individual drugs.  There was no comparison to the known concomitant administration of the two drugs.


Brimonidine/Timolol Combination Patent Valid

As of 2002 (the priority date), brimonidine had been previously used as a drug for ocular hypertension for 6 years.  Timolol was one of the “mainstays in the treatment of glaucoma” (available since 1978) (paras. 16, 17).  The mechanism of action and formulations of the two drugs were known. 

Brimonidine was formulated in a 0.15% or 0.2% solution, with a preservative (initially 0.005% benzalkonium chloride (BAK), and then purite – in Alphagan P[2]), given twice daily (in Canada) (paras. 43, 80, 82, 87, 100); and three times daily (in the U.S.).  The FDA was concerned about afternoon decreased efficiency with twice daily (BID) dosing (para. 62).  Timolol was formulated in a 0.5% solution, with 0.01% BAK (paras. 63, 82).  BAK was a “commonly used preservative in ophthalmic solutions” (para. 41), including another timolol combination solution (dorzolamide/timolol) (para. 82).  However, BAK was known to have cytotoxic effects.  Although the court seemed to focus on the commercial 0.2% brimonidine/0.5% timolol formulation (with 0.005% BAK) (paras. 25, 87), only two of the claims in issue referred to the preservative (Claims 6 and 22):

Claim 22 – Topical use of a therapeutically effective amount of a composition according to claim 6 in an affected eye for treating glaucoma.

Claim 6 – A composition according to claim 3 further comprising from 0.001% by weight to less than 0.01% by weight of benzalkonium chloride.

Claim 3 – A composition according to claim 1, wherein the amount of brimonidine is 0.2 percent by weight and the amount of timolol is 0.5 percent by weight.

Claim 1 – An ophthalmic topical pharmaceutical composition for the treatment of glaucoma or ocular hypertension comprising an effective amount of brimonidine and an effective amount of timolol in a pharmaceutically acceptable carrier therefor.

It is important to note that Claims 1 and 3 contain no reference to the BAK formulation that the court focused on when examining the large amount of work of the inventors (paras. 96-103, 122).

The patent, and the parties, agreed that it was known to combine glaucoma drugs.  The patent described detailed Phase III test results that compared:  (1) 0.2% brimonidine/0.5% timolol twice a day; (2) 0.5% timolol twice a day; and (3) 0.2% brimonidine three times a day.  There was no comparison to the known brimonidine/timolol combination that was separately administered.  The court accepted Allergan’s view that although concomitant BID brimonidine/timolol (“B/T”) was known, “it had no obligation to expand its clinical trials to include a comparison with concomitant administration of those ingredients.” (paras. 67, 78)  There were no published reports dealing with B/T as prior art.  Sandoz tried to rely on later articles showing B/T concomitant administration had similar efficacy as the fixed dose B/T, however the court found they were irrelevant to the common general knowledge.  In any event, even if the effectiveness was the same, the court found that the fixed dose B/T was still an improvement because fewer drops were required in the fixed dose, and the afternoon trough with BID brimonidine in concomitant therapy was eliminated.  The court does not elaborate on this latter finding given its earlier conclusion regarding similar efficacy.

The court was clearly impressed by the tests in the patent.  Once the court made the initial determination that the inventive concept included the results of the human trial in the patent, it was clear that such concepts would not be obvious to try.  Notably however, after completing its obviousness analysis using the Apotex v Sanofi 2008 SCC 61 test, the court then states that even if the inventive concept excluded the patent results, the patent would still be valid and not obvious (para. 127).  In determining the inventive concept, the court relied on several selection patent cases – where it is accepted that the whole point of the invention is an asserted advantage over a larger genus group.  This approach may add uncertainty as to the definition of “inventive concept”.  The courts have previously employed a different analysis in similar circumstances where one is conducting work that, on its face, is a routine inquiry.  Such work can be rendered non-obvious by surprising and unexpected results where such results are supported by the patent in the specific ranges claimed.  Justice Hughes took this approach in Eli Lilly v Novopharm 2009 FC 235 at paras. 10, 96-98, where he found Eli Lilly’s raloxifene particle size patent was both obvious and invalid for no sound prediction.  Eli Lilly tried to argue it had found surprising results in a particular particle size range.  However, the patent did not show/soundly predict that the particular range gave the surprising result.

The inventive concept of Allergan’s patent included:  improved “remarkable” safety, no afternoon trough, and a reduction in daily preservative (paras. 58, 61-63, 84).  The court also considered the advantage of fewer drops (because in fact the test results in the patent showed there were no statistically significant differences between brimonidine 3/day and the B/T fixed dose BID regarding the afternoon trough) (para. 65).  The court focused on the specifics of actually creating the B/T fixed dose product (para. 72), and not the general concept of combining the two drugs, which Allergan’s patent admitted was known.  “I do not interpret Allergan to be taking the position that there is anything inventive in using both brimonidine and timolol to treat glaucoma or ocular hypertension.” (para. 75).

No Inducement to Infringe Non-Approved Use

The court also briefly dealt with a second Allergan patent directed at a new neuroprotective use of brimonidine (paras. 128, 131).  The three prong legal test for inducement to infringe had not been met (para 145).  Regarding the first prong:  the court accepted that “off-label” use of Sandoz’s brimonidine would occur for neuroprotection (paras. 149, 150), even where Allergan’s product had not yet been approved for such use (para. 166).  The court found that “ophthalmologists do in fact prescribe brimonidine for neuroprotection” based on a recent article describing a clinical trial, the patent, and the Allergan expert’s clinical experience (paras. 172, 191, 228).  Regarding the second prong:  Allergan argued unsuccessfully that “Neuroprotective Information” in Sandoz’s product monograph (a reference to rat neuroprotection and a cited article in the bibliography that discussed neuroprotection) would influence doctors to prescribe the Sandoz product for neuroprotection (paras. 153, 167).  This information was also in Allergan’s product monograph (para. 171).  The court commented on problems with both parties’ evidence on this point.  Allergan changed their position on this issue after filing their evidence (para. 161).  Sandoz’s U.S. ophthalmologist admitted he was not aware of the Neuroprotective Information, or practices of Canadian physicians, and that including the cited article in the list of Product Monograph references “is “possibly” to elucidate a mode of action of brimonidine in neuroprotection” (paras. 163, 165).  Regarding the third prong:  the court found Sandoz knew that its influence would result in infringement.  Sandoz had removed certain references from Allergan’s product monograph, but had maintained the Neuroprotective Information, with no “explanation of why” (para. 171).

Sandoz was unsuccessful in arguing inutility based on the uncertainty of the new use at the time (paras. 204, 208); Allergan had demonstrated a “mere scintilla” of utility.  The court also refused to consider Sandoz’s arguments that topically applied brimonidine was not soundly predicted to be neuroprotective in humans (this was not argued in their Notice of Allegation) (para. 214).  In any event, the court went on to find “no hesitation” in concluding that Allergan had a sound prediction based on the animal tests in the patent (para. 226).  Allergan’s expert said the person skilled in the art “would understand that the data disclosed [in the patent] made it highly likely that brimonidine would have some level of efficacy as a neuroprotective agent in humans” (para. 222).  Sandoz’s expert “conceded in cross-examination that the experiments disclosed in the ‘626 Patent demonstrated that brimonidine did in fact have a neuroprotective effect in rats” (para. 224).  The court considered previous decisions where animal tests provided a sound basis for human use.[3]

Points for Practice


(1) A patentee should try to include in its patent all results it has in its possession regarding the tests of an invention, particularly a new formulation or combination.  Typically, such specific results will be difficult to have predicted in advance.  This decision may make it very difficult to invalidate what is, at first glance, an obvious combination.  Although the Sandoz decision was not appealed, a second NOC case involving Apotex is due to be heard in May 2012[4] by the same judge, involving the same B/T combination patent (Apotex unsuccessfully requested that Mr. Justice Crampton recuse himself).

(2) Generics should reconsider their approach when arguing NOC cases.  The decision briefly discusses one of the problems with NOC cases – one does not necessarily obtain an accurate picture of the work conducted by the inventors.  The court had “some sympathy” for Sandoz because Allergan failed to provide notebooks regarding “the steps that were taken to develop the Composition” (paras. 112, 113).  However, Sandoz did not request this information from Allergan. 


(1) The court effectively rejected the evidence of one of Sandoz’s key experts (whose credibility had been challenged in other U.S. decisions).  Such a finding, particularly in an NOC case (where there is no viva voce evidence), can be determinative in a case based on the “grey” area of obviousness (paras. 32, 87). 

(2) When preparing product monographs, even vague references to “new uses” should be carefully vetted in view of the applicable patents, even where the brand name product is not approved for such uses.  It is not worth the risk to leave such unnecessary references in a product monograph (paras. 167, 171). 


[1] Allergan Inc. v Sandoz 2011 FC 1316 (no appeal was filed)

[2] Due to known cytotoxic side effects of BAK, Allergan developed and launched Alphagan P with Purite, which was approved by the U.S. FDA in 2001 (para. 98).

[3] The court does not address how the promise of the patent (discussed briefly under the non-infringement analysis at para. 144 – “The promise of the ‘626 Patent is therefore that the compounds specified in the claims, including brimonidine, will have this purported neuroprotective effect upon retinal or optic nerves in humans”) would be soundly predicted given that glaucoma is a chronic condition.  This has been a factor the court has relied on to find no sound prediction based on only animal tests (in Apotex v Pfizer 2011 FCA 236 (regarding latanoprost), and more recently in Eli Lilly v Novopharm 2011 FC 1288 (regarding olanzapine)).

[4] Court File No. T-1560-10