Cipla Took the Delhi High Court for a Ride

Finally after quite a hectic work schedule, we got some time to bring analysis on recent judgment by the Delhi HC in a high-profile patent litigation case between Roche and Cipla. On 24^th April, the divisional bench at the Delhi HC dismissed an appeal made by Roche against a last year single judge decision (dated 19^th March 2008) that rejected their appeal for grant of temporary injunction to restrain Cipla from manufacturing, offering for sale, selling and exporting the drug Erlotinib. Almost every leading Indian newspaper covered judgment story, some even provided sort of expert analysis on the judgment. In fact, daily English newspaper Mint quoted “judgment certainly go down as a landmark ruling.” We made this judgment publicly available and downloadable on our blog for our readers on April 27. Hope many of our readers have already gone through the judgment and may also reviewed the decision for its merit both on legal and technical grounds. However, before we proceed with our (as usual) critical analysis on this judgment we would like to know from our readers – how many of our readers think the decision is a landmark judgment? And how many consider the judgment not only lacked to take into account basic technical knowledge but also lacked basic ABC of pharmaceutical patent, reflecting sheer inability of judiciary to handle even not so complex patent issues? Will this judgment possibly go down as a low of Indian patent law? We would surely love to know our readers’ comments on this. This judgment is not a mere judgment but an eye opener to many believing in intellectual property to rethink before investing in research and innovation in India. We do not know what sort of precedent will be set by this judgment but feel that it will considerably question and distort image of Indian patent law. Let us move to our analysis on the judgment. Please note statements included in bracket in bold are our remarks.

After the single judge rejected Roche’s appeal for temporary injunction against Cipla, Roche appealed the single judge decision. However, before deciding the case, the divisional bench reviewed the proceedings and conclusion of the single judge. The divisional bench particularly noted two points in Roche’s plaint made before single judge – (1) Patent No. 196774 (the ‘774 patent) is granted for Erlotinib hydrochloride marketed as Tarceva, and (2) no details of the specification of the ‘774 patent or the x-ray diffraction of Tarceva or Cipla’s generic product (Erlocip) was indicated [wonder why XRD data is asked by the single judge/divisional bench, particularly when Roche filed for infringement of Erlotinib drug compound patent not a particular polymorph patent?]. The divisional bench also noted arguments made by Cipla before single judge against the injunction application which include the following notable points.

(1) questioned the date of grant of patent, and also argued patent could not be presumed to be valid unless it was more than six years old,

(2) argued invalidity of patent under section 3(d) and obvious to a person skilled in the art, particularly argued the alleged patented product (Erlotinib) is nothing but a derivative from Gefitinib,

(3) argued patent is not worked fully and commercially in India, further adding no sales figures for the product for India provided in the plaint.

(4) argued public interest issue that the drug should be made available at cheap and affordable prices.

(5) placed US Pat. No. 6,900,221 (the ‘221 patent) on records filed by OSI for polymorphic Form B of Erlotinib hydrochloride. [this was interestingly deceptive move by Cipla]

The divisional bench also noted counter-claim arguments made by Cipla before single judge, most importantly the argument stating that the ‘221 patent clearly stated that the compound Erlotinib hydrochloride was a mixture of two polymorphs A&B and that one needed to separate and purify that polymorphic Form B so as to get to the claimed compound (Erlotinib) for acceptable efficacy. Further adding that the ‘221 patent clearly defeated the inventive step of alleged invention (the ‘774 patent). [Does this mean that later dated patent is a prior art for earlier dated patent? Quite a new concept framed by Cipla and that too for “compound per se patent”] Cipla further argued [now this is the trick that made real impact on deceiving and confusing both single judge as well as the divisional bench] the ‘774 patent had been obtained by the plaintiffs by suppression of material information stating that the patentee knowingly suppressed the fact that the claimed product is in the form of polymorph.

Cipla further filed an application to dismiss injunction suit before the single judge [this move put the final nail in the coffin]. Cipla argued that plaintiffs filed two separate applications for grant of patent in respect of the same chemical compound for polymorphic Form B. Further arguing the ‘774 patent is mixture of polymorphs A and B which was known to plaintiffs but never stated in the application made before the Patent Controller. Cipla also provided x-ray diffraction data of Tarceva before the single judge to show Tarceva tablets are polymorphic Form B, not mixture of A and B polymorph and based on x-ray data argued that Tarceva sold by Roche in India is covered by pending patent applications and not by granted ‘774 patent. [what a statement made against compound per se patent? Does this mean if an active ingredient of a drug is sold in a polymorphic form then active ingredient is not covered by compound per se patent?] Cipla further argued that case made by plaintiff is completely false and incorrect and also suppressed the fact that it has made two further patent applications for the same compound in polymorphic Form B.

During proceedings, the divisional bench particularly referred to polymorphic Form B story raised by Cipla and was of opinion that the ‘774 patent is for erlotinib hydrochloride polymorph A&B mixture and Tarceva is polymorphic Form B for which the plaintiffs did not yet hold a patent and therefore no prima facie case was made out by the plaintiffs as they were seeking an injunction against Cipla in respect of a drug for which they did not yet hold a patent. [Now this sounds something out of the world analysis that compound per se patent is circumvented by polymorph patent] The divisional bench also was of opinion that the story framed by Cipla had been suppressed by the plaintiffs both before the Controller of Patents as well as in the suit and on this sole ground injunction ought to have been refused. [Now what to expect further when the divisional bench already got preconceived by Cipla’s arguments, better Roche to forget thinking about fair judgment]

The divisional bench further was of opinion that the plaintiff were trying to mislead both the Court as well as Controller of Patents to the effect that polymorph B was subsumed in Polymorphs A and B after the divisional bench found patentee (plaintiff) argument made before the Controller of Patents that the closest prior art i.e. US ‘498 did not teach a compound of polymorph B free of polymorph A contradictory to later statement dated 18^th August 2008 by plaintiff that the earlier compound (disclosed in the ‘498 patent) included all known and unknown polymorphs. [A statement technically and patently correct and widely known to any patent practitioner in pharmaceutical area] But the divisional court opined that if Tarceva correspond to polymorphs A and B then there was no need for the plaintiffs to have applied for a separate patent in respect of polymorph B. [Obviously the divisional bench seems not able to understand basic ABC of pharma patenting]. The divisional bench also pointed that polymorphic Form B of Erlotinib hydrochloride was not known to the plaintiffs at the time they applied for a patent for Erlotinib hydrochloride as a combination of polymorphs A and B and therefore polymorphic Form B could not be said to be subsumed in the compound of combination of polymorphs A and B.

The divisional bench further discusses more issues pertaining to polymorphic Form B, polymorphs A and B mixture which was unfortunately crude, irrational and technically absurd and bizarre to read and understand. We do not even know how to pen down into words, observations made by the divisional bench is not only laughable but also annoying to read. Interestingly the divisional bench was of opinion that an applicant for a patent of a pharmaceutical product be bound to disclose the details of all other applications made by the applicant for grant of patent of derivatives or forms of such product. Also, there are some statements made by the division bench which we have no clue such as this one – “the Controller of Patents has confused the tests of inventiveness with obviousness (page 42).” Interestingly, the divisional bench also was of opinion that the Controller finding about obviousness of Erlotinib hydrochloride during pre-grant opposition is not obviousness but was about anticipation. [Does the divisional bench really understand what anticipation means in patent law?] We would request our readers to please go through the text of judgment, line by line to feel depth of absurdity.

While considering arguments framed by Cipla and observations made by the divisional bench, the Court concluded that Cipla has been able to demonstrate prima facie case that Roche do not hold a patent yet for the drug Tarceva, which is the polymorphic Form B of the substance for which they hold a patent. [Landmark ruling that active ingredient sold in polymorphic form is not covered by drug compound patent!!!] The court also concluded that Roche failure to bring the fact about filing of patent application for polymorphic Form B to the notice of the Controller of Patents was not consistent with the requirement of a full disclosure. Finally the divisional bench fined Roche to pay Rs. 5 lakhs to Cipla. [God knows why?] Is Roche penalized because the story framed by Cipla was suppressed by them?