Four Oppositions, Fifteen Years, One Grant: The Indian Patent Office's Ribociclib Decision

Novartis AG and Astex Therapeutics Ltd. have finally received a grant of a patent under number 594997 for their application relating to Ribociclib (Kisqali®), a selective CDK4/6 inhibitor indicated for the treatment of HR+/HER2- breast cancer, after contesting four pre-grant oppositions filed against the application in an exhaustive and detailed 198-page order. The application drew oppositions from Natco Pharma and three individual Opponents.

The application was filed in 2011 under number 1014/DELNP/2011, titled "Pyrrolopyrimidine Compounds as CDK Inhibitors". During the prosecution, the claims were substantially narrowed, ultimately covering the specific compound Ribociclib.

The order rendered by the Controller is significant and addresses the whole gamut of issues from the much controversial coverage versus disclosure debate, prior claiming under a Markush genus, person in the know standard, bioisosterism as an obviousness argument, and Section 3(d) efficacy in a potency/selectivity trade-off. Virtually every available statutory ground under the pre-grant opposition provision (Section 25(1)) was raised by at least one Opponent, including anticipation by prior publication, prior claiming, lack of inventive step, Section 3(d), insufficiency, and prior public knowledge/use.

Ground I - Prior claiming (Section 25(1)(c)):

The ground of prior claiming was the central and most heavily contested ground concerning Novartis's own earlier (genus) patent, IN 283133 ("IN'133"). IN'133 has an earlier priority date (May 26, 2006) than the impugned application (August 22, 2008) and was published after the impugned application's priority date squarely falling within the temporal window of Section 25(1)(c).

The Opponents relied not only on the Markush formula's breadth, but a substantial evidentiary record of the Applicants' conduct was also stressed upon. In this regard, much reliance was placed on Form 27 filed for IN'133 identifying "Kryxana 200 mg" (Ribociclib) as the commercially worked product, Novartis obtaining an interim injunction against Natco and filing of US Patent Term Extension (PTE) for IN'133's US counterpart (US 8,324,225). The Opponents also invoked the proposition that a patentee who enforces a genus patent against a specific compound cannot later deny, for validity purposes, that the compound falls within the genus claim invoking the doctrine against approbation and reprobation, referring to Delhi High Court's Division Bench decisions in AstraZeneca AB v. Intas Pharmaceuticals (2021) and the Boehringer Ingelheim v. Vee Excel Drugs Linagliptin litigation.

In response, the Applicant submitted a detailed claim-to-claim comparison besides relying on the F. Hoffmann-La Roche AG v. Natco Pharma judgement coverage/disclosure dichotomy. The Applicant also defended that Form 27 filings and PTE representations, filed years after the priority date may be relevant for coverage (relevant to infringement) but cannot retroactively convert a broad Markush formula into an individualised claim to Ribociclib for validity purposes.

After analysing the claim comparison, Controller concluded that Claim 1 of IN'133 is a Markush genus claim while Claim 1 of the impugned application is a species claim to a single compound, and the same are not literally identical.

On Form 27 and PTE filings, the Controller adopted the caution expressed in Kudos Pharmaceuticals v. Natco and Roche v. Natco against over-reliance on post-priority regulatory admissions to establish anticipation, treating such material as relevant to "commercial, regulatory or enforcement contexts" but not independently dispositive of patentability.

Ground II — Anticipation by prior publication (Section 25(1)(b))

This ground was dismissed essentially as a corollary of the prior-claiming analysis applying the "individualised form" standard from Dr. Reddy's Laboratories v. Eli Lilly and the "clear and unmistakeable directions" test from General Tyre v. Firestone.

Ground III — Obviousness (Section 25(1)(e))

On obviousness, Opponents contended that Example 338 of WO'222, differs from Ribociclib only by a phenyl-for-pyridyl substitution and is routine and motivated substitution in view of Toogood et al. and Grimm's Hydride Displacement Law. They further contended that Example 392 of WO'222, differ only by a C5-methyl group present in the prior art compound and absent in Ribociclib, also a routine methyl/hydrogen SAR variation within WO'222's own disclosed compound set.

Opponent (Natco) additionally invoked the "person in the know" doctrine by referring to AstraZeneca v. Intas and Novo Nordisk v. Dr. Reddy's (Semaglutide) to contend that where the inventors of the genus and species patents are substantially the same, obviousness should be assessed not from the perspective of an ordinary POSA but from that of an inventor intimately familiar with the genus patent's teachings , a materially lower bar for a finding of obviousness.

The Applicants contested that the invention was not properly characterised as a single atom substitution at all, but as the identification, out of a very large number of compounds disclosed in the earlier genus, of one specific compound possessing an unusual and unpredictable combination of high potency and very high selectivity for CDK4 over the related kinases CDK1 and CDK2. They also disputed the "person in the know" by pointing out that only two of Ribociclib's fifteen named inventors overlap with the earlier genus patent.

Upholding the ordinary skilled-person under Section 2(1)(ja) as applicable test, the Controller agreed with the Applicants' formulation of the inventive concept and declined to apply the heightened "person in the know" standard on the ground that this doctrine is drawn from cases where the genus and species patents shared substantially the same inventors.

A detailed analysis of the comparative biological data submitted by the Applicants was conducted by the Controller to analyse whether the selected compound showed an unexpected technical advantage over the compounds already disclosed in the prior art.

The data as submitted showed that while certain prior art compounds, including Example 338, were comparable or even somewhat more potent against CDK4 in isolation, they lacked Ribociclib’s selectivity. Ribociclib was over 11,000-fold selective for CDK4 over CDK1 and roughly 7,600-fold selective over CDK2, against a selectivity of only 300–400-fold for the closest prior art compound. Cell-based data reinforced the same picture, showing that Ribociclib arrested cells cleanly at the intended checkpoint across a wide range of concentrations, while the prior art compound began to show signs of off-target activity at much lower concentrations.

The Controller accepted that this distinction was scientifically significant. Improved selectivity, by reducing off-target kinase inhibition, was found to point toward an improved therapeutic index and a lower toxicity profile and the order treats selectivity, rather than raw potency, as the principal technical contribution of the invention. This was considered alongside the FDA breakthrough therapy designation and approval in 125+ countries as supporting a finding of a "genuine inventive selection" rather than a routine optimisation.

On bioisosterism specifically, the order adopts the EPO Boards of Appeal's longstanding caution (T 643/96 and progeny, cited via the Beecham, Hoechst Marion Roussel and Eisai decisions relied on by the Applicants) that bioisosteric replacement is "an empirical rule... which in each particular case needs to be experimentally verified," not a mechanical predictor of retained or improved activity.

Ground IV - Section 3(d)

Amongst other contentions, one point of contention was whether Section 3(d)'s "enhanced therapeutic efficacy" requirement could be satisfied by a compound that is, on raw potency, inferior to the closest prior art compound. The Opponents relied on Novartis's own "Discovery of Kisqali" publication which was filed by the Applicants as additional evidence exhibiting Example 338 is roughly ten-fold more potent against CDK4 (IC₅₀ 0.001 μM) than Ribociclib itself (IC₅₀ 0.01 μM) .

The Applicants submitted that the objection under Section 3(d) was misconceived at the outset, as Ribociclib was claimed as a distinct new chemical entity and not as a derivative of a known substance falling within the categories enumerated under Section 3(d). Without prejudice to this submission, the Applicants further contended that the comparative evidence on record established that the claimed compound possessed significant technical advantages over the prior art compounds.

The Applicants argued that therapeutic efficacy cannot be assessed solely on the basis of potency against a single kinase. It was submitted that the claimed compound demonstrated markedly improved selectivity for CDK4 over CDK1 and CDK2, and that such selectivity was a critical pharmacological characteristic having a direct bearing on the therapeutic profile of the compound. In support of this contention, the Applicants relied upon comparative biochemical data as well as fluorescence-activated cell sorting (FACS) studies to demonstrate that Ribociclib produced selective G1 cell-cycle arrest while minimising inhibition of other cyclin-dependent kinases.

The Controller's conclusion seems to stand on two limbs. First, treating Ribociclib as a new chemical entity and not a “new form of a known substance”. Second, placing reliance on the data exhibiting enhanced therapeutic efficacy, in the scenario this characterisation fails its stand.

After reviewing the comparative data relied upon by both parties, the Controller held that the Applicants had demonstrated substantially improved selectivity of Ribociclib for CDK4 over CDK1 and CDK2 when compared with the cited prior art compounds. The Controller further noted that the Applicants had placed material on record to show that inhibition of CDK1 and CDK2 is associated with undesirable off-target effects, whereas increased selectivity towards CDK4 contributes to an improved therapeutic profile.

Rejecting the Opponents' reliance on differences in CDK4 inhibitory activity, the Applicants' submission that selectivity is an equally relevant consideration in evaluating therapeutic efficacy was accepted by the Controller. The Controller considered the Applicants’ submission that the assessment of therapeutic efficacy cannot be confined to a comparison of potency values alone. The Controller also took into consideration the FACS data relied upon by the Applicants, which demonstrated that Ribociclib produced selective G1 arrest consistent with selective CDK4/6 inhibition, whereas the comparative compounds exhibited broader biological effects attributable to inhibition of additional cyclin-dependent kinases.

Accordingly, the objection under Section 3(d) was considered not sustainable by the Controller.

Ground V - Insufficiency (Section 25(1)(g))

A technical enablement objection was raised by one of the Opponents (Opponent 4) citing that the method as described in the specification concerns a substrate bearing a trifluoroethyl-piperazine substituent rather than Ribociclib's unsubstituted piperazine, with no explicit disclosure of how the trifluoroethyl group is removed (example 74).

The Controller acknowledged this concern but resolved it in the Applicants' favour on the basis that a POSA (a medicinal chemist) would be able to adapt a generic coupling procedure and combine it with the deprotection steps in "General Procedure A" without undue experimentation, reinforced by the fact that corresponding EP and US patents issued on substantially the same disclosure. Claim 2 (the composition claim) was deleted during prosecution in response to the excipient-support objection, mooting that portion of the ground.

Ground VI - Prior public knowledge/use (Section 25(1)(d))

In the absence of any documentary evidence , this ground was rejected.

Key Takeaways/ Points of interest:

· On the issue of Coverage Vs. Disclosure- In spite of patentee affirming (in litigation/ PTE filing/statement of working) that a compound falls within an earlier genus claim, the Patent Office may not treat that as establishing prior claiming or anticipation without an individualised, enabling disclosure in the genus document itself;

· Generic disclosures do not necessarily anticipate specific compounds. In the absence of clear and enabling disclosure of specific compound, a broad Markush formula or genus claim in the prior art may not be considered destroying the novelty of a species claim;

· Selection inventions remain patentable notwithstanding an earlier genus patent. Where a claimed compound is shown to possess an unexpected technical advantage over the compounds disclosed in the genus;

· Enhanced therapeutic efficacy may be established through improved selectivity and a better safety profile, even where the claimed compound is not the most potent compound on record. Section 3(d) is applied pragmatically, recognising efficacy through selectivity and safety rather than potency alone.

Ranjna Mehta-Dutt 

Partner | Attorney at Law |

ranjnamehtadutt@iprattorneys.com

Nanki Arneja

Partner | Attorney at Law

nanki@iprattorneys.com