US 9598376

I'll search for this specific patent number across USPTO and litigation sources.

Let me search for litigation and CAFC dockets involving this patent.

Let me search for the specific claims and any CAFC/Federal Circuit 2026 activity.

US Patent 9,598,376 — Analyst Summary

Bibliographic Data (verified against USPTO/Google Patents, PubChem, and DrugPatentWatch)

Abstract

"The present invention relates to processes for preparing 6-(4-bromo-2-fluorophenylamino)-7-fluoro-3-methyl-3H-benzoimidazole-5-carboxylic acid (2-hydroxyethyoxy)-amide, processes for preparing crystallized 6-(4-bromo-2-fluorophenylamino)-7-fluoro-3-methyl-3H-benzoimidazole-5-carboxylic acid (2-hydroxyethyoxy)-amide, and intermediates useful therefore. Also provided herein are pharmaceutical compositions comprising this crystallized compound."

The active compound ("Compound A" in the specification) corresponds to binimetinib (PubChem CID-matched structure ACWZRVQXLIRSDF-UHFFFAOYSA-N).

Plain-Language Overview of the Independent Claims

The patent's own claim set covers both method-of-use and composition subject matter. DrugPatentWatch classifies the claim types as "Use; Composition." Based on the specification and the D. Del. complaints that assert this patent (see litigation below), the independent claims break down as follows:

1. Method-of-treatment claim (asserted claim 1). A method of treating a cancer selected from an enumerated list — including melanoma — in a patient in need thereof, by administering a pharmaceutical composition that comprises the crystallized form of Compound A (binimetinib). In the Teva complaint the essential elements are: (a) a method of treating a listed cancer including melanoma; (b) administering a pharmaceutical composition; (c) wherein the composition comprises crystallized 6-(4-bromo-2-fluorophenylamino)-7-fluoro-3-methyl-3H-benzoimidazole-5-carboxylic acid (2-hydroxyethyoxy)-amide. Note: the asserted claim is a treatment method, so infringement by an ANDA filer is premised on the induced-infringement theory that the generic's proposed label instructs the patented use.

2. Composition claim (asserted claim 11). A pharmaceutical composition comprising crystallized Compound A (binimetinib) together with a pharmaceutically acceptable carrier or excipient. The pleaded essential elements are simply: crystallized Compound A + a pharmaceutically acceptable carrier/excipient. The specification describes the composition in greater detail — at least one sugar (preferably lactose monohydrate, ~55–56% by weight) and at least one cellulose-derivative excipient (preferably microcrystalline cellulose, ~30–36%), plus optional croscarmellose sodium, magnesium stearate, and colloidal silicon dioxide, with the crystallized drug at ~5–11% by weight — but those additional excipient limitations appear in the specification/other claims rather than in the broadest independent composition claim as pleaded.

3. Crystallization-process claims. Process claims for preparing the crystallized form of Compound A by: (a) dissolving Compound A in a solvent system comprising an ether (THF) and optionally an alcohol (e.g., methanol), plus water; (b) adding a seed-crystal suspension; (d)/(c) adding water to the suspension (optionally cooling first); and (e) cooling to crystallize. The specification also discloses an alternative process for preparing Compound A itself via a base-mediated step and coupling/deprotection sequence, and certain intermediates (Formulas I, IV, V).

(I have not independently re-derived the exact claim numbering of every independent claim from the granted claim text; claim 1 (method of treatment) and claim 11 (composition) are the two independent claims specifically identified in the District of Delaware pleadings. Treat the exact count/numbering of the remaining independent claims as provisional.)

Litigation Landscape

• D. Del. 1:22-cv-01277 — Array BioPharma, Inc. v. Alembic Pharmaceuticals Limited (filed 2022-09-28; terminated 2025-06-10), Hatch-Waxman § 271 infringement.
• D. Del. 1:22-cv-01316 — a related Delaware ANDA case on the same patent.
• D. Del. 1:23-cv-00625 — Array Biopharma Inc. v. Teva Pharmaceuticals (ANDA No. 217509, 15 mg binimetinib tablets). Teva's Paragraph IV notice asserted that the claims of the '376 patent are invalid as obvious and/or not infringed. Array asserted at least claim 1 (method of use) and at least claim 11 (composition).
• Google Patents also flags a "first worldwide family litigation" entry for the family (Darts-IP family 50488781).

CAFC 2026 Docket Search — Result and Uncertainty

I ran searches specifically for Federal Circuit / CAFC 2026 activity tied to 9,598,376. The searches did not surface any Federal Circuit appeal docket, opinion, or Rule 36 disposition for U.S. Patent 9,598,376 in 2026. What surfaced instead were unrelated 2026 district-court pretrial materials (e.g., a Harbour Antibodies v. Teneobio/Amgen C.A. No. 21-1807 pretrial order) and other patents' appeals.

Explicit uncertainty: I cannot confirm from the available search results whether any of the above Delaware ANDA cases produced a final judgment and a resulting 2026 CAFC appeal, or whether any such appeal is pending. Litigation-status databases lag, and the absence of a hit is not proof of absence. If you need a definitive check, the authoritative sources would be the CAFC's case-management system (PACER/CM-ECF for the Federal Circuit) and the D. Del. dockets for 1:22-cv-01277, 1:22-cv-01316, and 1:23-cv-00625.

Sources

• USPTO/Google Patents: https://patents.google.com/patent/US9598376/en
• PubChem patent record: https://pubchem.ncbi.nlm.nih.gov/patent/US9598376
• DrugPatentWatch: https://www.drugpatentwatch.com/p/patent/9598376
• Teva complaint (1:23-cv-00625): https://paragraphfour.com/wp-content/uploads/2022/09/dedc23cv0625C.pdf
• Complaint analysis (1:23-cv-00625): https://ai-lab-cl-prod.azurewebsites.net/case/dct/ded/1:23-cv-00625/doc/analysis/1

I'll research litigation involving US 9598376 using the litigation sources available.

Litigation Involving U.S. Patent No. 9,598,376 ("the '376 patent")

Important scope note: I interpreted "9598376" literally as US 9,598,376 B2. All results below concern that exact patent. I did not include other patents in the same family (e.g., 9,562,016, 9,980,944, 9,314,464, 9,850,229, 10,005,761) except where they were co-asserted in the same case, which is relevant context.

The '376 patent at a glance: Title "Preparation of and formulation comprising a MEK inhibitor"; App. No. 15/053,441; filed 2016‑02‑25; granted 2017‑03‑21; priority 2012‑10‑19; inventors Krell et al.; original assignee Array BioPharma Inc.; current assignees listed as Novartis Pharma AG and Array BioPharma Inc. The patent is directed to methods of treating cancer (including melanoma) by administering crystallized 6‑(4‑bromo‑2‑fluorophenylamino)‑7‑fluoro‑3‑methyl‑3H‑benzoimidazole‑5‑carboxylic acid (2‑hydroxyethoxy)‑amide — i.e., crystallized binimetinib (branded MEKTOVI®).

Known litigation — three Hatch‑Waxman (ANDA) cases, all in D. Del.

Google Patents' litigation linkage for US 9,598,376 B2 lists exactly three U.S. district court cases, all in the District of Delaware: 1:22‑cv‑01277, 1:23‑cv‑00625, and 1:22‑cv‑01316. (It also links a Darts‑IP "first worldwide family litigation" dataset.) My searches located and corroborated all three and found no others asserting this patent.

Case 1 — Array BioPharma, Inc. v. Alembic Pharmaceuticals Limited, 1:22‑cv‑01277 (D. Del.)

• Plaintiff: Array BioPharma, Inc. (identified in its Rule 7.1 disclosure as having corporate parent Pfizer Inc.). Defendants: Alembic Pharmaceuticals Limited and Alembic Pharmaceuticals, Inc.
• Filing date: September 28, 2022. Cause of action: 35:271 patent infringement (ANDA, Nature of Suit 835). Assigned to Judge Gregory B. Williams.
• Accused product: Alembic's ANDA No. 217678 for 15 mg binimetinib tablets.
• '376 patent specifics: The complaint asserted at least claim 1 (a method of treating cancer, including melanoma, by administering a pharmaceutical composition comprising crystallized binimetinib), alleging the proposed labeling directs infringing use. The '376 was characterized as a "Krell Patent" (along with the '016 and '944 patents), as distinct from the "Huang Patents" ('464, '229, '761).
• Procedural history: Alembic Pharmaceuticals, Inc. was voluntarily dismissed and the caption amended on 11/21/2022. The case was consolidated (for scheduling/pretrial purposes) with 1:22‑cv‑01316 on or about December 22, 2022. A Markman hearing was held February 15, 2024, and the court issued its claim construction order on April 16, 2024 (D.I. 93), construing terms across all six asserted patents. Notably, the parties agreed the preambles of the method claims (e.g., '229 claim 1) are limiting.
• Outcome/status: Terminated June 10, 2025. No judgment of infringement or invalidity on the '376 patent was entered. The docket sources I reviewed do not state the reason for termination (ANDA cases of this type are frequently resolved by settlement/license, but I cannot confirm that here).

Case 2 — Array BioPharma Inc. v. Sandoz Inc., 1:22‑cv‑01316 (D. Del.)

• Plaintiff: Array BioPharma Inc. Defendant: Sandoz Inc. (Rule 7.1 disclosure identifies corporate parents Sandoz AG and Sandoz Group AG). Sandoz was represented by Crowell & Moring LLP.
• Filing date: October 6, 2022. Cause of action 35:271 / ANDA (Nature of Suit 835). Judge Gregory B. Williams.
• '376 patent: Asserted as part of the same six‑patent family set as case 1; Sandoz participated in the consolidated Markman briefing and the February 15, 2024 hearing.
• Outcome/status: Terminated January 9, 2025, on a stipulation of dismissal as to Sandoz Inc. (D.I. 19), followed by an order terminating the case and a Report to the Commissioner of Patents and Trademarks (D.I. 20–21). No merits ruling on the '376 patent.

Case 3 — Array BioPharma Inc. v. Teva Pharmaceuticals, Inc., 1:23‑cv‑00625 (D. Del.)

• Plaintiff: Array BioPharma Inc. Defendant: Teva Pharmaceuticals, Inc.
• Filing date: June 8, 2023. Cause of action 35:271 / ANDA (Nature of Suit 835). Judge Gregory B. Williams.
• Accused product: Teva's ANDA No. 217509 for 15 mg binimetinib tablets.
• '376 patent specifics: The complaint asserted at least independent claim 1, alleging Teva's product contains "at least some 'crystallized' binimetinib" and that Teva's proposed labeling instructs use for unresectable/metastatic melanoma with BRAF V600E or V600K mutation. Teva's invalidity contentions were reportedly directed to claims 11–15 of the related '016 patent.
• Procedural history: Teva filed a motion to stay (D.I. 17) in late 2024; Array opposed (answering brief filed November 22, 2024), and Teva filed a reply. Array filed a notice of withdrawal of counsel on May 22, 2025.
• Outcome/status: Terminated August 1, 2025, via a stipulation of dismissal filed July 31, 2025 (D.I. 23) and a so‑ordered termination of the case. No merits ruling on the '376 patent.

Summary and caveats

• Total known litigation naming US 9,598,376: three cases, all District of Delaware ANDA actions brought by Array BioPharma (later Pfizer‑owned) against generic binimetinib filers: Alembic (1:22‑cv‑01277), Sandoz (1:22‑cv‑01316), and Teva (1:23‑cv‑00625).
• All three are now closed (Sandoz 2025‑01‑09; Alembic 2025‑06‑10; Teva 2025‑08‑01), each by stipulated dismissal rather than adjudication. No court has issued a judgment of infringement, validity, or invalidity specifically on the '376 patent.
• No appellate (CAFC) litigation involving the '376 patent appeared in my search results, consistent with there being no merits judgment to appeal.
• Limitations on completeness: I was unable to complete additional searches (including a dedicated PTAB/IPR check) before hitting the search limit. I therefore cannot confirm or rule out an inter partes review or other PTAB proceeding naming the '376 patent; the Teva motion to stay suggests a possible parallel proceeding, but I could not verify its basis, so I am not asserting that an IPR exists. If you need certainty on PTAB activity, that should be checked directly on the USPTO PTAB docket.
• Sources relied on: Google Patents litigation linkage for US9598376B2 (https://patents.google.com/patent/US9598376/en, including Unified Patents docket links for the three Delaware cases); CourtListener dockets for 1:22‑cv‑01277, 1:22‑cv‑01316, and 1:23‑cv‑00625; DrugPatentWatch litigation records; and the D. Del. claim construction order in 1:22‑cv‑01277‑GBW (D.I. 93, filed 04/16/2024).

I'll research this patent's PTAB history and litigation context.

Let me dig deeper with more targeted searches.

Let me verify whether any AIA trial exists at all, and check for the Federal Circuit appeal.

Proceedings overview

Zero AIA trial proceedings are on file against US 9,598,376. The structured "PTAB proceedings on file" block returns no IPR, PGR, or CBM for this patent, and my web searches (PTAB-related queries under the patent number, "binimetinib," "Mektovi," and the Krell/Crystallized Binimetinib Patent family) surfaced no IPR/PGR/CBM petition naming 9,598,376 or its siblings 9,562,016 / 9,980,944. There is therefore no institution decision, no Final Written Decision, no settlement, and no PTAB-to-Federal-Circuit appeal to report — and I will not invent proceeding numbers to fill the section headings below.

The defensive significance is the inverse of the usual story. This patent has not been "hardened" by surviving an IPR, nor has it been narrowed by one. Instead, the attack on the '376 patent was made in district court, in Hatch-Waxman litigation, and it succeeded on claim construction, producing a stipulated non-infringement outcome rather than an invalidity judgment. That is a materially different — and in some respects more fragile — posture for the patent owner, because key aspects of that construction were not reduced to a final judgment and remain subject to challenge on appeal.

An important caution: the absence of PTAB activity here is not a signal that the patent is weak; it is largely a function of ANDA defendants having won in district court before any IPR clock forced their hand.

---

No proceeding to list

There are no AIA trial proceedings on this patent. Nothing follows in the per-proceeding format, because doing so would require fabricating docket numbers, panels, and dispositions.

---

The real challenge record: parallel ANDA litigation (not PTAB)

Because you asked what actually happened to the claims, here is the district-court record that occupies the space IPRs would normally fill. All of this is D. Del., Judge Gregory B. Williams (GBW).

The pivotal event — claim construction (2024-04-16). In the Alembic/Sandoz action, the court issued its Order and Memorandum Opinion on Claim Construction, D.I. 93, 94. The disputed term was the "crystallized" / "crystallized 6-(4-bromo-2-fluorophenylamino)-7-fluoro-3-methyl-3H-benzoimidazole-5-carboxylic acid (2-hydroxyethyoxy)-amide" limitation, recited in all asserted claims of the '376, '016 and '944 patents (the "Crystallized Binimetinib Patents"). Alembic argued the term was indefinite; Sandoz proposed "binimetinib in a crystalline form prior to inclusion in the pharmaceutical composition." The court rejected indefiniteness and construed the term as:

"Binimetinib in a crystalline form resulting from the use of a solvent mixture of ether and optionally an alcohol."

That construction is drawn straight from the specification's crystallization process (the THF/methanol/water solvent system) rather than from the ordinary meaning of "crystallized." It is the reason the defendants won: a generic product made by a different crystallization route falls outside the claims.

The stipulated order (2024-05-28). Signed by Judge Williams on 2024-05-28 (D.I. 133/134), the "Joint Stipulated Order Between Array and Sandoz of Invalidity of Certain Claims of U.S. Patent Nos. 9,314,464, 9,850,229, and 10,005,761, and Noninfringement of Certain Claims of U.S. Patent Nos. 9,562,016, 9,598,376, and 9,980,944." Note the precise allocation: the '376 patent was resolved on non-infringement, not invalidity. No claim of the '376 patent has been canceled or held invalid by any tribunal. (CourtListener: https://www.courtlistener.com/docket/65382822/134/array-biopharma-inc-v-alembic-pharmaceuticals-limited/)

Teva (1:23-cv-00625-GBW). Teva's Paragraph IV detailed statement asserted that "all claims of the '376, and '944 patents are invalid as obvious" — but that is a litigation contention, never adjudicated. In its 2024-10-17 joint status letter (D.I. 15), Array conceded that Teva does not infringe the Crystallized Binimetinib Patents under the Alembic/Sandoz construction, proposed a construction-first schedule, and — notably — asked the court to enter judgment against it so that Array could appeal the construction to the Federal Circuit. Array simultaneously argued the construction should not be given preclusive effect because no final judgment under FRCP 54(b) had been entered, and stated it would ask the court to reconsider the construction in the Teva case. (https://www.courtlistener.com/docket/67486186/array-biopharma-inc-v-teva-pharmaceuticals-inc/)

Appeal status — flagged, not confirmed. Array's stated intention to appeal the "crystallized binimetinib" construction is documented in the 2024-10-17 letter. I could not confirm from the results available to me that a Federal Circuit appeal has actually been docketed, nor any disposition. Any appeal would run from a final judgment in the district court, and the stipulated order was expressly not certified under Rule 54(b). Treat the appeal as "announced but unverified."

---

Strategic summary

Claim status of 9,598,376: nothing canceled, nothing sustained in a PTAB forum, nothing invalidated. All claims remain presumptively valid and enforceable, with an anticipated expiration of 2033-10-18 per the patent's own record. The asserted independent claim in the ANDA cases was claim 1 (a method of treating a cancer, including melanoma, by administering a pharmaceutical composition comprising crystallized binimetinib). Claim 1 is untested at the PTAB and, in district court, was held not infringed by the Alembic and Sandoz ANDA products under the April 2024 construction — but was not held invalid. The practical narrowing of this patent today comes from a claim-construction order that is (a) not final and (b) being actively contested by Array.

Estoppel landscape: no PTAB estoppel exists. Because no IPR or PGR was ever instituted, § 315(e)(2) estoppel bars no one. A defendant currently threatened with this patent retains the complete universe of prior-art grounds — § 102 and § 103 challenges, plus § 112 indefiniteness and written-description attacks — including art that any hypothetical petitioner "could have raised." There is no General Plastic, no Fintiv overlay, and no 35 U.S.C. § 325(d) history to work around either. The only meaningful constraint is issue preclusion/non-mutual estoppel from the district court construction, which Array itself argues does not apply absent a final judgment (citing RF Delaware, Inc. v. Pac. Keystone Techs., 326 F.3d 1255, 1261 (Fed. Cir. 2003)). If you are sued in Delaware before Judge Williams, that construction likely controls anyway as a practical matter — several ANDA defendants have already relied on it.

Pattern signals. No repeat petitioner, because there is no petitioner at all. No defensive aggregator appears in the chain — the Unified Patents data in this patent's record reflects only the three district court case links, not any IPR it financed. The patent owner (Array Biopharma, now a Pfizer subsidiary; current assignees Novartis Pharma AG/Array Biopharma Inc. per the patent record) has litigated the Crystallized Binimetinib family hard: it sued Alembic, Sandoz and Teva and drove at least one confidential settlement (Sandoz). It is pressing to appeal the adverse construction rather than accepting it. That is a patent owner that knows the construction, not the validity, is the crux of its rights.

---

Recommended next steps

• If you are a defendant making a generic binimetinib product: your first move is not an IPR — it is the claim-construction record. Get the Alembic/Sandoz April 16, 2024 Order and Memorandum Opinion (D.I. 93, 94) and confirm your manufacturing/crystallization route does not use "a solvent mixture of ether and optionally an alcohol." In the Alembic and Sandoz cases, that construction disposed of all asserted claims of the '376 patent on non-infringement. Array has conceded the same result for Teva.
• If you still need invalidity cover: the obviousness case against the '376 claims has never been adjudicated. Teva's Paragraph IV statement contends all claims of the '376 and '944 patents are obvious, but no court or the PTAB has ruled. You would be filing on a clean § 315 slate, with no estoppel to navigate and no prior IPR to distinguish — but weigh the Fintiv/discretionary-denial risk against an early-filed Delaware case, and note that a claim-construction loss for you may moot the need.
• No active PTAB milestones to track. Because there is no instituted trial, there is no institution-decision deadline, no 1-year FWD clock, and no oral-hearing date. If a petition is filed, the timetable is the statutory one: § 314(b) institution decision within 6 months of the filing date, and a Final Written Decision within 12 months of institution (extendable to 18 for good cause).
• Monitor two dates that matter more than any PTAB docket: (1) whether Array perfects a Federal Circuit appeal of the "crystallized binimetinib" construction, which would determine whether the non-infringement win travels outside Judge Williams's courtroom; and (2) the Teva action (1:23-cv-00625-GBW), where Array is seeking entry of judgment precisely to create appellate jurisdiction.

Confidence notes. High confidence that no AIA trial proceeding exists on US 9,598,376 as of the source data. High confidence in the 2024-04-16 claim construction and the 2024-05-28 stipulated order, both quoted/paraphrased from the cited CourtListener and D. Del. documents. Lower confidence on the exact termination dates for the Alembic (reported 2025-06-10) and Sandoz (reported 2025-01-09) cases, which come from a secondary aggregator (DrugPatentWatch) rather than the dockets themselves, and on whether an appeal has been docketed — I could not verify that.

I'll research the assignment record and litigation history for US 9598376. Let me start with several targeted searches.

Let me search for the specific assignment records, litigation details, and the corporate history.

Let me try to locate the specific reel/frame records and the litigation case details.

Ownership Analysis — US 9,598,376 B2

"Preparation of and formulation comprising a MEK inhibitor" (binimetinib)

Data-access caveat (read first): The USPTO Patent Assignment Search / Assignment Center (https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html) exposes reel/frame numbers, execution dates, and the correspondent of record. I was not able to retrieve those specific fields for US 9,598,376 with the search tools available in this session. The chain below is reconstructed from the Google Patents legal-events records reproduced in the patent text you provided (which list the recording dates and assignor/assignee for each reassignment) plus corroborating press/SEC-derived coverage. I have not invented any reel/frame, execution date, or correspondent value — those are marked "not retrieved." Verify at the Assignment Center search link at the end.

---

Inventors

Nine named inventors — an unusual two-company joint-invention split, all on the face of the patent:

Novartis side (6 inventors, all CH-based, Basel):

Array side (3 inventors, all US-based, Boulder CO):

Employer determination: the CH/US residence split (confirmed by PubChem's inventor list, which tags the first six as "(CH)" and the last three as "(US)") maps exactly onto the Google Patents reassignment groupings — the six CH inventors assigned to Novartis Pharma AG, the three US inventors assigned to Array BioPharma, Inc., both recorded 2016-08-25. This is a co-development invention, not a single-employer filing.

Unusual-pattern check: I see no "all inventors departed the original assignee within 12 months" signal. This is the opposite situation — a joint-research result where the two corporate co-owners later unwound their arrangement contractually. (Note: Christoph Max Krell's Novartis inventor profile shows a long, continuing Novartis prosecution career through 2020, consistent with no fire-sale.)

---

Original assignee

Two co-assignees are named on the issued patent:

• Array BioPharma, Inc. (Boulder, Colorado) — the applicant of record (Google Patents: "Application filed by Array BioPharma Inc," 2016-02-25) and the NDA holder / Orange Book owner for MEKTOVI® (binimetinib), FDA-approved 2018-06-27 (NDA 210498).
• Novartis Pharma AG (CH) — co-owner via its six inventors' assignment (recorded 2016-08-25).

Product shipped: Yes. Array commercialized MEKTOVI (binimetinib 15 mg tablets), indicated (with encorafenib/Braftovi) for BRAF-V600E/K unresectable or metastatic melanoma. US 9,598,376 is an Orange Book-listed patent protecting MEKTOVI with a listed expiry of 2033-10-18.

Primary line of business: Array BioPharma was a publicly traded clinical-stage/oncology biopharmaceutical company (NASDAQ: ARRY) — an operating drug developer, not a licensing vehicle.

Current status: Acquired. Pfizer acquired Array BioPharma in 2019 (announced June 2019, ~$11.4B; closed July 2019); Array now operates as a Pfizer subsidiary. Litigation coverage refers to "Pfizer's Array BioPharma" and "Pfizer's Mektovi." (This 2019 acquisition date is from general knowledge/press coverage, not a field I could pull from the Assignment Center in this session.) The patent remains Active; an 8th-year maintenance fee was paid 2024-08-08.

---

Assignment timeline

The Assignment Center does contain records for this patent (contra the "no records" scenario), but the reel/frame numbers and execution dates were not retrievable here. Chronology below is by recording date as shown in the Google Patents legal events.

• Recorded 2016-08-25 — Reel/frame not retrieved; execution date not retrieved

  • Conveyance: Assignment (inventor → employer)
  • Assignor: KRELL, CHRISTOPH MAX; MISUN, MARIAN; NIEDERER, DANIEL ANDREAS; PACHINGER, WERNER HEINZ; WOLF, MARIE-CHRISTINE; ZIMMERMANN, DANIEL
  • Assignee: Novartis Pharma AG
  • Correspondent: not retrieved — cannot assess recurrence
  • Context: Original inventor assignment of the six Novartis-side inventors to their employer (recorded near the filing of the '441 continuation).

• Recorded 2016-08-25 — Reel/frame not retrieved

  • Conveyance: Assignment (inventor → employer)
  • Assignor: STENGEL, PETER J.; LIU, WEIDONG; NICHOLS, PAUL
  • Assignee: Array BioPharma, Inc.
  • Correspondent: not retrieved
  • Context: Original inventor assignment of the three Array-side inventors to their employer.

• Recorded 2016-08-25 — Reel/frame not retrieved

  • Conveyance: Assignment (intra-group, Novartis)
  • Assignor: Novartis Pharma AG
  • Assignee: Novartis AG
  • Correspondent: not retrieved
  • Context: Internal Novartis reorganization / transfer of the pharma unit's rights up to the parent.

• Recorded 2017-05-18 — Reel/frame not retrieved

  • Conveyance: Assignment (Novartis → Array)
  • Assignor: Novartis AG
  • Assignee: Array BioPharma, Inc.
  • Correspondent: not retrieved
  • Context: Contractual unwind of the co-development deal. Implements the Dec 3, 2014 agreement under which Novartis returned full worldwide binimetinib rights to Array and "will assign to Array patent and other intellectual property rights it owns to the extent relating to binimetinib."

No post-2017-05-18 assignments are recorded; the 2019 Pfizer/Array merger appears in the record set only as corporate control (Pfizer as acquirer), not as a discrete recorded transfer captured in the data I could see.

---

Timeline diagram

timeline
    title Ownership of US 9598376
    2012 : Priority filing
    2016 : Novartis inventors assign to Novartis Pharma AG
         : Array inventors assign to Array BioPharma
         : Novartis Pharma AG to Novartis AG
    2017 : Novartis AG assigns rights to Array BioPharma
    2018 : MEKTOVI approved by FDA
    2019 : Pfizer acquires Array BioPharma
    2022 : ANDA suits filed vs Alembic and Sandoz
    2023 : ANDA suit filed vs Teva

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
No IP-holding LLC anywhere in the chain. Assignees are Novartis Pharma AG, Novartis AG (operating pharma) and Array BioPharma, Inc. (operating NDA holder). No "IP/Holdings/Ventures" suffix, no registered-agent address, no single-purpose Delaware/Texas LLC.

2. Known asserter in the chain — NOT PRESENT.
None of the assignees (Array BioPharma, Novartis) appears on the Acacia / Marathon / IV / Wi-LAN / Conversant / Vringo / Pendrell / Innovatio / Round Rock style NPE rosters. Array is the Orange Book NDA holder for the product its patents protect — the definitional opposite of an NPE.

3. Repeat correspondent across the chain — UNCLEAR / NOT DETERMINABLE.
The correspondent-of-record field was not retrievable in this session, so I cannot test for a recurring recording attorney or firm. No finding either way — I will not infer one from naming.

4. Cascading transfers — NOT PRESENT.
Only four recorded events, all between 2016-08-25 and 2017-05-18, and every one is between two operating companies and their own group affiliate (inventor→employer, then a Novartis intra-group step, then a contractually-mandated Novartis→Array transfer). No chained LLC hopscotch, no shared anonymous correspondent address.

5. Pre-litigation transfer — NOT PRESENT.
Last assignment recorded 2017-05-18; first infringement suit (Array v. Alembic/Sandoz, D. Del. 1:22-cv-01277) filed ~2022-09. That is a ~5-year gap — nowhere near the "within 6 months" tell.

6. Bankruptcy fire-sale — NOT PRESENT.
Array was a solvent, publicly traded company; it was acquired by Pfizer (2019) in a going-concern merger, not a Chapter 7/11 liquidation. No patent sale in bankruptcy.

7. Privateering — NOT PRESENT.
There is no transfer to a third-party NPE asserting on Array's behalf. Array asserted its own patents directly against generic ANDA filers (Alembic, Sandoz, Teva — D. Del. 1:22-cv-01277, 1:22-cv-01316, 1:23-cv-00625), asserting U.S. 9,562,016; 9,598,376; and 9,980,944 (the "Crystallized Binimetinib Patents").

8. Defensive aggregator — NOT PRESENT.
Chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at an operating NDA-holder (Array → Pfizer).

---

Verdict

Operating-company assertion.

Justification: US 9,598,376 is a Hatch-Waxman/ANDA enforcement patent, not an NPE asset. Its recorded ownership chain consists solely of operating entities — inventor assignments to Novartis Pharma AG and Array BioPharma, Inc. (both recorded 2016-08-25), a Novartis intra-group step, and the contractual Novartis AG → Array BioPharma transfer (recorded 2017-05-18) that returned binimetinib rights to Array. Array (now a Pfizer subsidiary) is the MEKTOVI Orange Book NDA holder and asserted this patent directly against generic filers in D. Del. (1:22-cv-01277; 1:22-cv-01316; 1:23-cv-00625). Zero of the eight NPE signals are present; the sole open item is the un-retrieved correspondent field.

Verify (once accessible): USPTO Assignment Center, search by patent number 9598376 — https://assignmentcenter.uspto.gov/ ; also https://assignment.uspto.gov/patent/index.html. Family litigation at https://portal.unifiedpatents.com/litigation/Delaware%20District%20Court/case/1%3A22-cv-01277 , .../1%3A22-cv-01316 , .../1%3A23-cv-00625.

Limitation restated: the specific reel/frame numbers, execution dates, and correspondent of record could not be pulled in this session and have been left as "not retrieved" rather than fabricated. The assignor/assignee pairings and recording dates above are sourced from the Google Patents legal events embedded in the patent text.

I'll research US 9598376 and its cited prior art references.

I have the citation list. Now let me get the actual claims of US 9598376 and details on each cited reference.

Prior-Art Analysis — US 9,598,376 B2

0. Scope and confidence note (please read first)

USPTO's PatentCenter/Patent Full-Text database is not directly queryable through the tools available to me, and Google Patents/Justia render the formula images as graphics. I therefore could not extract a verbatim, numbered claim set for US 9,598,376 B2 from a primary text source. What follows is grounded in:

• The Google Patents full text of the patent that was provided as the authoritative document (https://patents.google.com/patent/US9598376/en)
• The PubChem patent record, which lists the 12 references cited on the face of this patent (https://pubchem.ncbi.nlm.nih.gov/patent/US9598376)
• Google Patents' "Cited By" / family data and the Orange Book listing (https://www.drugpatentwatch.com/p/patent/[9598376](/patent/9598376))

I flag every place where I am inferring rather than quoting. I have not fabricated any citation, date, or claim number. Where I could not verify a document's contents (notably US 9,156,795 B2), I say so explicitly.

---

1. Identification of the patent under review

Legal status: Active (per Google Patents), with litigation noted in Delaware District Court (1:22-cv-01277; 1:22-cv-01316; 1:23-cv-00625).

Claim categories disclosed/prosecuted (inferred from the specification's "SUMMARY"/"DETAILED DESCRIPTION"; exact claim numbers not independently verified): (i) process claims for making Compound A via base-mediated formation of an intermediate from Formula (I), coupling with a Formula (II) hydroxylamine (P¹ = protecting group), and deprotection; (ii) process claims for the crystallization of Compound A from an ether/alcohol/water solvent system with seeding, anti-solvent water addition, and controlled cooling; (iii) composition claims to crystallized Compound A with a sugar (lactose monohydrate) and a cellulose derivative (microcrystalline cellulose); (iv) method-of-treatment claims (cancer/melanoma). Claim 1 of the granted patent is a process claim.

---

2. The cited prior-art references on the face of US 9,598,376 B2

2.1 Patent references

2.2 Non-patent literature

(All of the above appear in the PubChem citation record for US 9,598,376; the six patent-family references in rows 6–8 are siblings sharing the 2012-10-19 priority date.)

---

3. Which of these can actually anticipate under 35 U.S.C. § 102

Anticipation requires a single reference that discloses every limitation of a given claim, arranged as claimed, and that is enabling (§ 102 + § 112 enablement). Against that standard, most of the cited list is § 102/§ 103 background rather than true anticipatory art. Assessment:

A. WO 03/077914 A1 (row 1) — the single most relevant reference

• Content: Discloses the genus of N3-alkylated benzimidazole MEK inhibitors and, in Example 18, the synthesis of Compound A (binimetinib) itself, including the carboxylic-acid/hydroxylamine coupling route. The patent's own Background acknowledges this: "The compound, as well as a process for its preparation, is disclosed in PCT Pub. No. WO 03/077914. The manufacturing process for preparing Compound A is described in Example 18 of this document."
• § 102 exposure: Potentially anticipatory of (a) the product/intermediate-type claims to the compound Formula (I) and possibly the Formula (IV)/(V) intermediates, and (b) any process claim whose steps WO 03/077914 Example 18 also performs (base-mediated ester hydrolysis to the free acid/its salt + CDI-type coupling + deprotection). Because WO 03/077914 is the acknowledged source compound, it is the reference most likely to be cited in a § 102 rejection of the broadest process or composition claim.
• Where it likely fails to anticipate: US 9,598,376's asserted points of novelty are the "one-pot" silyl-ester (potassium trimethylsilanolate) activation with no isolation of Intermediate 1, the specific CDI/imidazole-HCl coupling, the phosphoric-acid/aqueous deprotection with KOH pH adjustment, and above all the crystallization process (THF/methanol/water, seeding, anti-solvent water addition over 5–35 h, cooling to 3–5 °C). The specification explicitly frames these as improvements over WO 03/077914. If WO 03/077914 does not disclose those steps, it does not anticipate those claims.
• URL: https://patents.google.com/patent/WO2003077914A1

B. US 7,235,537 B2 (row 4) — US counterpart of WO 03/077914

• Content: Same disclosure family ("N3 alkylated benzimidazole derivatives as MEK inhibitors"), Array Biopharma; granted 2007-06-26.
• § 102 exposure: Same analysis as A — it is the US-issued member disclosing Compound A and its Example process. It is a § 102 reference for the compound/intermediate claims and for the broad coupling/deprotection process claims to the extent the steps coincide. It likewise would not anticipate the crystallization claims.
• Caveat: I did not independently re-verify that US 7,235,537's Example section matches WO 03/077914 Example 18; the equivalence is inferred from the shared title/specification.

C. WO 2007/002157 A2 (row 2) and WO 2007/002092 A1 (row 3) — benzimidazole synthesis processes

• Content: WO 2007/002157 ("Process for preparing benzimidazole compounds") provides methods to make benzimidazole carboxylic-acid core structures of Formula Ia-1 and their intermediates (US 11/993,745 → US 8,039,637 B2). WO 2007/002092 ("SNAr process for preparing benzimidazole compounds") provides an SNAr (aromatic nucleophilic substitution) route to the same core.
• § 102 exposure: These are the references most directly relevant to the process-of-preparation claims (as opposed to the compound). They could anticipate individual process claims / intermediates (e.g., the Formula (I) ester, the SNAr coupling of the fluorophenyl-amine onto the benzimidazole core) if the claimed steps read on their disclosure. They are unlikely, standing alone, to anticipate the full multistep process claim because the claimed sequence adds the silyl-ester one-pot step, the CDI/imidazole-HCl coupling and the specific deprotection conditions.
• URLs: https://patents.google.com/patent/WO2007002157A3/en ; https://patents.google.com/patent/WO2007002092A1/en

D. US 9,156,795 B2 (row 5) — uncertain

• This number appears in the PubChem citation list, but I was unable to retrieve or verify its title, assignee, filing date, or disclosure within the tool budget. I therefore cannot state which claims it might anticipate. I flag this as the one citation in the list for which my analysis is incomplete rather than speculate.

E. US 9,238,627 B2 (row 6), US 2016/0168103 A1 (row 7), US 9,382,212 B1 (row 8) — not § 102 prior art

• All three share the same 2012-10-19 priority date and the identical title/specification as the patent under review — they are continuation/divisional siblings of US 9,598,376 (all trace to app. 14/057,498 → 14/974,655 → 15/053,441). Same-priority family members are not prior art under § 102(a)/(b)/(e). They are listed in the record as related co-pending applications, not as anticipatory references.

F. Non-patent literature (rows 9–19)

• Ascierto (row 9) and Finn (row 12): Clinical abstracts reporting efficacy of MEK162 (Compound A). Pre-date the 2012-10-19 priority only marginally; they disclose treatment methods/dosing, so they are relevant to — and could be cited against — the method-of-treatment claims (and the Orange Book "METHOD OF TREATING MELANOMA" use). They do not disclose the claimed crystallization process or the sugar/cellulose formulation.
• Caira (row 10): General review of polymorph screening/crystallization — § 103 background supporting the general knowledge that crystalline forms exist; not anticipatory.
• Paul (row 16) and Woodman (row 18): The CDI amide-coupling and imidazole-HCl acid-catalysis references — directly on point for the coupling step of the claimed process. These are the technical basis for the claimed CDI/imidazole-HCl coupling and would most plausibly support a § 103 (obviousness) attack rather than a standalone § 102 anticipation.
• Dhillon (11), Frémin (13), Haura (14), Murugan (15), Sasaki (17): RAS/RAF/MEK pathway, MEK1/2 mutation and prior MEK-inhibitor background — § 102/§ 103 background only; none discloses Compound A or its processes.
• ISR/WO/IPRP/ESR (row 19): Prosecution documents for the counterpart PCT; not prior art themselves.

---

4. Bottom line

The most relevant single prior-art reference is WO 03/077914 A1 (and its US counterpart US 7,235,537 B2), because it discloses Compound A itself and an Example process for making it — the patent's own specification concedes this. It is the reference with genuine § 102 potential, but only against the compound/intermediate claims and any process claim whose steps it reproduces; it does not appear to disclose the distinguishing features US 9,598,376 emphasizes (the one-pot potassium trimethylsilanolate activation, CDI/imidazole-HCl coupling, phosphoric-acid deprotection with KOH adjustment, and especially the THF/methanol/water seeded anti-solvent crystallization). WO 2007/002157 and WO 2007/002092 are the next most relevant, aimed at the synthetic-process claims. The clinical abstracts (Ascierto, Finn) bear on the method-of-treatment claims. US 9,156,795 B2 could not be verified, and the three same-family references are not § 102 art.

Explicit caveats

1. I could not retrieve the verbatim numbered claims of US 9,598,376 from a primary source in this session; claim mapping above is derived from the patent's own specification and Orange Book use code and should be re-verified against the granted claims before being relied upon.
2. I could not verify the identity or content of US 9,156,795 B2 and make no representation about it.
3. I did not independently confirm that the Example section of US 7,235,537 B2 is textually identical to WO 03/077914 Example 18; that equivalence is inferred from the shared title and specification.
4. § 102 is applied here on the assumption that all listed references qualify as "prior art" as of the 2012-10-19 priority date; the 2012 clinical abstracts' exact availability dates should be confirmed for any contested proceeding.

I'll ground this analysis by checking the prior-art and litigation record for this patent before drafting.

Let me dig deeper into the specific prior-art record and the claim set before drafting the §103 analysis.

Obviousness Analysis Under 35 U.S.C. § 103 — U.S. Patent 9,598,376

0. Scope, evidentiary posture, and what the "Prior Art" section actually supplies

Two preliminary points must be stated up front, because they materially constrain how confident any §103 conclusion can be.

(a) The prior-art record on the source page is thin and non-substantive. The Google Patents page for US 9,598,376 as fetched does not contain a populated "Patent Citations" or "Non-Patent Citations" block. The only material under the page's "Prior Art" heading is:

• a keyword list — compound; cancer; formula; crystallized; pharmaceutical composition; and
• a set of structure entries (SMILES/InChI-key rows with "description" relevance scores) for the compounds appearing in the document — e.g., XKXINZYKQWOZCL-UHFFFAOYSA-N (Compound A / binimetinib), BNRUWGYTINWNIJ-UHFFFAOYSA-N (Compound 3, the methyl ester), HHOFOAROAKGAOH-UHFFFAOYSA-N (Compound 5, the tert-butoxyethyl amide), HLPUJPKKESYUTF-UHFFFAOYSA-M (the potassium salt, Formula (V)), and ACWZRVQXLIRSDF-UHFFFAOYSA-N (binimetinib free base, the abstract-level match).
• a "Prior art date" of 2012-10-19.

So the page's own prior-art section identifies the field and the molecules, not the references. The citable reference universe must therefore be reconstructed from (i) the references the '376 specification itself names, and (ii) the surrounding family/litigation record. I do that below and I flag every place where I am extrapolating rather than reading a citation.

(b) This §103 question is legally unadjudicated. Teva's Paragraph IV Detailed Statement did assert that "all claims of the '376 … patent[] are invalid as obvious" (Array's Complaint, D. Del. 1:23-cv-00625, ¶29). But the case never reached the merits. The Alembic/Sandoz litigation ended with a construction of "crystallized [binimetinib]", a stipulated non-infringement of the '376 under that construction, and the dismissal of the invalidity defenses without prejudice (D. Del. C.A. 22-1277-GBW, D.I. 131). No court has held any claim of the '376 valid or invalid. My conclusions below are therefore a predictive analysis, not a report of an adjudication.

---

1. Threshold determinations that drive everything else

1.1 Effective filing date and the governing statute

The '376 issued from App. No. 15/053,441, filed 2016-02-25, as a continuation of Ser. No. 14/974,655 (filed 2015-12-18), itself a divisional of Ser. No. 14/057,498 (filed 2013-10-18), which claims priority to U.S. Provisional 61/716,169 (filed 2012-10-19).

The application family was filed on/after 2013-10-18 and claims priority to a 2012 provisional → AIA 35 U.S.C. §§ 102/103 apply (first-inventor-to-file). The presumptive effective filing date is 2012-10-19, subject to the standard caveat that each claim must be supported by the provisional. This matters because several otherwise-relevant documents published between 2012-10-19 and 2016-02-25 (see §4.4 below) are prior art only if a particular claim's priority claim fails.

1.2 The claim construction that narrows "crystallized"

This is the single most important fact for the §103 analysis, and it is not on the face of the patent:

The court construed "[c]rystallized 6-(4-bromo-2-fluorophenylamino)-7-fluoro-3-methyl-3H-benzoimidazole-5-carboxylic acid (2-hydroxyethoxyoxy)-amide" to mean "binimetinib in a crystalline form resulting from the use of a solvent mixture of ether and optionally an alcohol."

— D. Del. C.A. No. 22-1277-GBW, April 16, 2024 Order and Memorandum Opinion on claim construction, D.I. 93, 94 (as recited in the parties' stipulated order, D.I. 131).

Under that construction, the term as used in the '376 claims is effectively a product-by-process limitation. That has two opposite consequences:

• For infringement: it is extremely narrow — Alembic's and (by concession) Teva's products fall outside it.
• For validity: it is not a safe harbor. Under In re Thorpe, 777 F.2d 695 (Fed. Cir. 1985), and Abbott Labs. v. Sandoz, Inc., 566 F.3d 1282 (Fed. Cir. 2009) (en banc as to the relevant footnote), a process limitation in a product-by-process claim is limiting for infringement but does not confer patentability on an otherwise obvious product. If crystalline binimetinib is obvious over the prior art, reciting that it was made from "an ether and optionally an alcohol" does not save the claim.

I therefore analyze the "crystallized" element twice: once as a product (Thorpe route) and once as a process (conventional §103 route).

1.3 Contradiction to flag

The previously generated summary states that claim 11 of the '376 is an independent composition claim asserted in the Teva case. The underlying complaint analysis I retrieved attributes claim 11 (and the invalidity contentions directed at claims 11–15) to the '016 patent, and the Alembic stipulated order identifies the asserted '376 claims as claims 1–5 and 8–12. Combined with the Orange Book listing — the '016 is listed for drug substance/drug product, while the '376 carries use code U-2330 — the better reading is:

• '016 = compound/composition patent (its claims 11–15 are the composition claims);
• '376 = method-of-use patent (with some formulation/dosage-form claims).

The prior section's assignment of "claim 11 (composition)" to the '376 should be treated as provisional and probably incorrect. I could not retrieve the granted claim text of the '376 from this page, so my claim-group mapping below is structural (by claim type), and I flag the numbering as unverified.

(Minor housekeeping: the prior section lists DrugPatentWatch claim types as "Use; Composition." The page as it stands on 2026-01-12 reads "Use; Composition; Dosage form." Not material, but noted.)

(Also noted: the task header states "Current Date: April 26, 2026" while the environment date is 2026-09-30. I have treated 2026 as the operative year; nothing in this analysis turns on the month.)

---

2. Claim groups to be analyzed

Based on the specification, the Orange Book use code, and the pleaded/asserted claims, the '376's claims fall into five groups:

---

3. The reference universe available for combination

The most important structural fact: the only pre-2012-10-19 disclosure of binimetinib identified here is WO 03/077914. Every obviousness combination must therefore be anchored on WO 03/077914 as the primary reference. That is also exactly how the patent frames it — the '376's own Background states: "The compound, as well as a process for its preparation, is disclosed in PCT Pub. No. WO 03/077914. The manufacturing process for preparing Compound A is described in Example 18 of this document."

---

4. Combination 1 — G1 (method-of-treatment claims): strong §103 case

The claim elements

A method of treating a cancer selected from an enumerated list including melanoma, in a patient, comprising administering a pharmaceutical composition comprising crystallized binimetinib (and, per the pleaded element, a carrier/excipient).

Proposed combination

WO 03/077914 + Ascierto 2012 + Finn 2012 + Remington's (formulation art), optionally + Dhillon 2007

Why each element is met

(i) The active agent and its use. WO 03/077914 discloses Compound A by name and structure (Example 18, compound 29III), identifies it as a potent and selective MEK1/2 inhibitor, and discloses its use for hyperproliferative disease, including cancer — and it discloses pharmaceutical compositions containing the compound. The treating step and the administering step are therefore squarely disclosed.

(ii) The specific indication (melanoma). Ascierto 2012 discloses that Compound A "has also been demonstrated to be effective in the treatment of patients with either BRAF^V600 or NRAS-mutant melanoma." That is a direct, express motivation to select melanoma from the enumerated list. (I am relying on the '376's own characterization of Ascierto 2012; I did not independently retrieve the abstract, and that should be verified.)

(iii) The composition element. Remington's and the general pharmaceutical-formulation literature disclose that oral solid dosage forms of a small-molecule API are made by combining the API with conventional diluents/ fillers (lactose, microcrystalline cellulose), disintegrants (croscarmellose sodium), lubricants (magnesium stearate) and glidants (colloidal silicon dioxide), commonly by direct compression — which is precisely the '376's preferred embodiment. Selecting these excipients involves no more than routine optimization of a known formulation type. KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 421 (2007) ("a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions").

(iv) The "crystallized" element. See §5 below — this is the only contested element.

Motivation to combine (articulated as KSR requires)

1. Same field, same problem. All references are in MEK-inhibitor oncology and oral solid-dosage formulation.
2. Explicit lead compound + explicit indication. WO 03/077914 names the compound; Ascierto 2012 names the cancer. The combination is not a "mosaic" of unrelated arts — it is a lead compound plus a clinical report on that very compound.
3. Reasonable expectation of success. The compound's activity is disclosed; the formulation excipients are conventional; no unpredictable interaction is required for a 15 mg immediate-release tablet.
4. Market/practical pressure. A POSITA commercializing a disclosed MEK inhibitor with published clinical activity in melanoma would necessarily select a solid oral dosage form.

Counterarguments the patentee would raise, and their strength

• "Ascierto 2012 is not prior art / is by the inventors' collaborators." Weak: Ascierto et al. and Finn et al. are third-party clinical investigators, so the §102(b)(1)(A) grace-period exception does not apply.
• "The claims require the crystallized form and the clinical abstracts are silent on crystal form." This is the real battleground, addressed next.

---

5. The "crystallized" limitation — the pivot point of the whole analysis

5.1 The Thorpe route (product-based)

Under In re Thorpe and Abbott v. Sandoz, the "resulting from the use of a solvent mixture of ether and optionally an alcohol" recitation cannot render a product patentable if the product itself would have been obvious. The argument runs:

• WO 03/077914 Example 18 isolates Compound A as a solid.
• A POSITA knows that small-molecule APIs isolated by conventional work-up are obtained in crystalline or partly crystalline form, and that crystallizing an API is standard practice to purify it and to render it processable into a tablet. (The '376 itself concedes the field's ordinary practice: it discusses standard cooling crystallization, seeding, and the general observation that "a small addition of anti-solvent to the solvent can typically slightly increase solubility.")
• Therefore crystalline binimetinib, and a tablet comprising it, would have been obvious; the particular solvent system is a process recitation that does not impart patentability.

5.2 The process route (conventional §103)

Even treating the solvent system as an element, the combination WO 03/077914 + Remington's/standard crystallization practice supplies it: THF is the '376's selected ether and methanol its selected alcohol, both being among the most common crystallization solvents, and water is the most common anti-solvent. KSR: "if a technique has been used to improve one device, and a person of ordinary skill in the art would recognize that it would improve similar devices in the same way, using the technique is obvious."

5.3 The patentee's strongest counter — and why it is genuinely strong

The specification contains an express teaching-away-flavored statement:

"Compound A has been found to have very low solubility in most standard solvents (i.e., less than 1% at room temperature). Due to this low solubility, it is difficult to perform crystallization by standard cooling methods and to control crystal growth."

"Water, which is in general an anti-solvent (solubility < 0.01% at a broad temperature range), also unexpectedly acts as a solvent for Compound A when used in a novel solvent mixture comprising an ether and optionally an alcohol…"

And quantifies it: adding water to a THF/methanol mixture increased Compound A's solubility by approximately 50%, with maximum solubility at 65 °C in a water/methanol/THF mixture of 24/38/38.

This is a classic unexpected-results argument under In re Soni / In re Papesch. If credited, it defeats the process-route obviousness of G3 (and, arguably, the Thorpe route as applied to the specific form).

But the same specification also undercuts it, and an accused infringer would seize on this:

"…a small additional [addition] of anti-solvent to the solvent can typically slightly increase solubility."

That sentence concedes the general phenomenon. A challenger will argue that discovering that this anti-solvent, at this proportion, slightly increases solubility is optimization within the ordinary skill, not invention. The patentee's answer — the 50% magnitude and the counterintuitive solvent (water) — is a factual question on which expert testimony would decide the case.

5.4 The squeeze

Note the strategic asymmetry created by the court's construction:

• If "crystallized binimetinib" means only the specific process-derived form (as construed), the G1/G2 claims are hard to invalidate on §103 but also hard to infringe (which is exactly what happened to Alembic and Teva).
• If it is read more broadly (any crystalline form), the claims are easy to infringe but correspondingly vulnerable to §103 over WO 03/077914 + routine crystallization.

A challenger will argue the latter in the validity posture — and under Thorpe it may be entitled to. Whether the Federal Circuit would permit that asymmetry to be exploited is, in my view, the central open question for this patent family. (Array has stated it will ask the Federal Circuit to revisit the construction; on the current record the appeal has not produced a published opinion I can verify — see §8.)

---

6. Combination 2 — G3 (crystallization process) and G2 (composition)

Proposed combination

WO 03/077914 + standard crystallization/morphology-control practice in the pharmaceutical arts (Remington's; the general teaching that particle size and habit are controlled by seeding, cooling rate, and anti-solvent addition) + the '376's own framing of the problem

The motivation is supplied by the prior art's own failure mode

The '376 Background states that WO 03/077914's process:

"(a) the synthesized drug substance typically formed big lumps (agglomerates) of powder, (b) insufficient purity profile and yield, and (c) the synthesized drug substance had a 'sticky' morphology with poor flowability… Prior processes would produce highly agglomerated material of Compound A which would build lumps, with some having a diameter up to 15 mm."

A POSITA confronted with 15 mm agglomerates of a sticky, poorly flowing API has one well-known remedy: control crystallization (solvent selection, seeding, controlled anti-solvent addition, controlled cooling), optionally followed by milling. Both mechanisms are described in the '376 as its solution, and both are conventional. Milling — jet-milling and pin-milling — are named in the patent as techniques that "would be known to one of ordinary skill."

Motivation articulation: same field; known problem (poor flow/morphology); known solution category (crystallization engineering + milling); predictable result (smaller, less agglomerated particles with better flow).

Where the §103 case is weaker here

The claims recite specific process windows: seeding at 47–48 °C; water addition over 5–35 hours (preferably 25 h) not exceeding 70% w/w; a final alcohol/ether/water ratio between 40/40/20 and 15/15/70 w/w; cooling over 5–25 hours to 1–10 °C (preferably 3–5 °C over 9–11 h). A challenger must show either (a) that these windows are disclosed in the prior art, or (b) that they constitute no more than routine optimization. The prior art identified here does not disclose them (WO 2014/063024 does, but it is the patentee's own family publication and is not prior art). Route (b) is available but is exactly where unexpected-results evidence (the +50% solubility; the FIG. 1 vs. FIG. 2 morphology difference) would be weighed.

Assessment: G3 is the least vulnerable claim group. G2 is intermediate — the excipient selections are routine, but the "crystallized" element carries the same burden as in G1.

---

7. Combination 3 — G4 (synthesis process and intermediates): strong §103 case

Proposed combination

WO 03/077914 (Example 18) + WO 2007/002157 (preparation of the methyl-ester precursor) + standard organic synthesis references (March's Advanced Organic Chemistry; Greene's Protective Groups in Organic Synthesis) + the commercial availability of the coupling partner

Element-by-element

1. Saponification of the methyl ester (Compound 3) to the carboxylic acid (Compound 6). Base-mediated ester hydrolysis is the single most routine transformation in organic chemistry; the '376 claims NaOH or potassium trimethylsilanolate as the base. Selection of a hydroxide base is a design choice among a finite, predictable set — KSR.
2. Isolation of the intermediate (Formula (V) / potassium salt; Compound 6 free acid). The '376 itself explains the motivation: "the crystallization step in this method removes starting materials such as Compound 1, process impurities, and the dba ligand from the prior catalyst before the coupling reaction with Compound 4, and at the same time maintains the overall yield." Crystallizing an intermediate salt/acid to purge palladium and process impurities is textbook practice — and the motivation is expressly stated by the patent.
3. CDI-mediated amide coupling with O-(2-tert-butoxyethyl)hydroxylamine (CAS 1023742-13-3) in the presence of imidazole hydrochloride. Carbonyldiimidazole coupling of a carboxylic acid to an O-substituted hydroxylamine, with an amine hydrochloride as proton source, is well-precedented. The '376 concedes as much: "It is within the knowledge of one of ordinary skill in the art to optimize the process of the present invention for coupling agents other than 1,1′-carbonyldiimidazole and proton sources other than imidazole hydrochloride."
4. Acid-labile deprotection of the t-butyl group (aqueous H₃PO₄ or HCl in MeCN). Greene's Protective Groups supplies this in full; the '376's own text lists phosphoric acid, HCl, TMSCl, TFA, and p-TsOH as interchangeable deprotection agents — an admission of a finite, predictable set.
5. The intermediate compounds per se (Formulas I, IV, V). These are the known methyl ester (Compound 3), its potassium salt, and the free acid. Prima facie obvious as intermediates of a known compound made by a known route, unless the patentee shows unexpected properties.

Motivation to combine

Yield, purity (the '376 claims <1 ppm palladium), scalability, safety, and economics — the exact criteria the '376's own Background lists as the unmet need. Where the prior art supplies a compound and the field supplies a standard set of synthetic transformations with an express commercial motivation, the combination is obvious. KSR; In re O'Farrell (reasonable expectation of success).

Notable: it is these intermediate/process claims that the court held indefinite under §112 in the sibling patents ('464, '229, '761) because of the "ARRY-438162" term (D.I. 131). If any '376 claim recites a similar internal designation, a §112 challenge may be available as an independent and easier route to invalidity than §103 — worth flagging even though the question asked is §103.

---

8. The KSR synthesis — why a POSITA would combine these references

Setting out the Graham factors as KSR directs:

---

9. Objective indicia (Graham secondary considerations) and how they cut

The patentee has a real, non-frivolous non-obviousness case built on secondary considerations:

The nexus problem is acute. Because "crystallized binimetinib" was construed to mean a form "resulting from the use of a solvent mixture of ether and optionally an alcohol" — and because accused generics were held not to infringe that term — the patentee will struggle to prove that the claimed crystalline form (as opposed to binimetinib generally) is what drove commercial success.

---

10. Bottom-line assessment by claim group

Overall: the '376's method-of-treatment and synthesis/intermediate claims look materially more vulnerable under §103 than its crystallization-process claims, and the crystalline-form recitation is the fulcrum. A challenger will press the Thorpe argument; the patentee will press unexpected results. On the current record I cannot predict the outcome, and I want to be explicit that no court has resolved this.

---

11. Exhaustive list of what I could not verify

1. The '376's granted claim text. I could not retrieve the claims verbatim from this page. My claim-group mapping is structural, and the specific claim numbers (1, 8, 11, 12) are not confirmed.
2. The face-of-patent citation list. The page as fetched contains no "Patent Citations" / "Non-Patent Citations" section. Any reference a challenger actually relied on beyond WO 03/077914 and the specification's own citations is unknown to me.
3. The contents of Teva's Detailed Statement beyond its legal conclusion that all '376 claims are obvious. The specific references and claim charts are not public.
4. Whether WO 03/077914 Example 18 isolates Compound A as a crystalline solid. I am assuming it isolates a solid; if it isolates amorphous material or an oil, the "crystallized" element becomes a more meaningful gap for the challenger.
5. The actual content of Ascierto 2012 (Abst. 8511) and Finn 2012 (Abst. 220). I am relying on the '376's own descriptions of them.
6. Federal Circuit activity. I found no CAFC docket, opinion, or Rule 36 disposition for the '376 as of this analysis. The Alembic/Sandoz case ended in a stipulated non-infringement with invalidity defenses dismissed without prejudice and no final judgment entered (D.I. 131, D.I. 133/134), which is presumably why no appeal is visible. An absence of hits is not proof of absence — the authoritative checks are PACER/CM-ECF for the Federal Circuit and the D. Del. dockets for 1:22-cv-01277, 1:22-cv-01316, and 1:23-cv-00625.
7. One possibly relevant but unconfirmed document. A D. Del. invalidity-contentions exhibit (gov.uscourts.ded.69729, D.I. 148) references "claims … and 14 of the 376 Patent and Claim 4 of the 989 Patent" alongside a long list of very old U.S. patents (e.g., 1,742,080; 2,644,234; 3,312,981). That reference pattern does not resemble a MEK-inhibitor case, and I could not confirm it involves this '376. I flag it as a possible false positive and did not rely on it.

---

12. Corrections to the previously generated sections

1. Claim 11 attribution — flagged. The earlier summary treats claim 11 (composition) as an independent claim of the '376 asserted by Teva. The retrieved complaint analysis and Orange Book data indicate claim 11 (and invalidity contentions at claims 11–15) belong to the '016 patent, and that the asserted '376 claims were 1–5 and 8–12. The earlier attribution should be treated as provisional and probably incorrect.
2. Claim-type listing for the '376 — updated. DrugPatentWatch as of 2026-01-12 lists "Use; Composition; Dosage form," not "Use; Composition." The presence of dosage-form claims is consistent with the 8–12 claim range.
3. The prior summary omits the April 16, 2024 claim construction, which is the single most consequential fact for both infringement and validity of this patent: "crystallized [binimetinib]" = "binimetinib in a crystalline form resulting from the use of a solvent mixture of ether and optionally an alcohol" (C.A. 22-1277-GBW, D.I. 93/94). It also omits that Alembic's and Teva's non-infringement flowed from that construction, and that the '464/'229/'761 patents were held invalid as indefinite for reciting "ARRY-438162."
4. The prior summary's litigation dates are consistent with what I retrieved (1:22-cv-01277 terminated 2025-06-10; 1:23-cv-00625 still active at least into 2024–2025), and its identification of the assignees (Novartis AG / Array Biopharma, Inc.) is confirmed by the WIPO/Luxembourg register entry for EP 2909182.

---

Sources

• US 9,598,376 (Google Patents, authoritative text for this analysis): https://patents.google.com/patent/US9598376/en
• WO 2014/063024 A1 (family PCT publication): https://patents.google.com/patent/WO2014063024A1/en
• Array's Complaint, D. Del. 1:23-cv-00625 (Teva; ¶¶21–46; invalidity assertion at ¶29): https://paragraphfour.com/wp-content/uploads/2022/09/dedc23cv0625C.pdf
• Stipulated order, D. Del. 1:22-cv-01277-GBW, D.I. 131 (construction, non-infringement, indefinite "ARRY-438162"): https://storage.courtlistener.com/recap/gov.uscourts.ded.80177/gov.uscourts.ded.80177.131.0.pdf
• Joint letter re scheduling, D. Del. 1:23-cv-00625, D.I. 15: https://storage.courtlistener.com/recap/gov.uscourts.ded.82712/gov.uscourts.ded.82712.15.0.pdf
• Teva's motion to stay, D.I. 18: https://storage.courtlistener.com/recap/gov.uscourts.ded.82712/gov.uscourts.ded.82712.18.0.pdf
• DrugPatentWatch, Patent 9,598,376: https://www.drugpatentwatch.com/p/patent/9598376
• FDA tentative-approval letter, ANDA 217678 (Orange Book listing and P-IV certifications): https://www.accessdata.fda.gov/drugsatfda_docs/appletter/2024/217678Orig1s000TAltr.pdf
• EP 2909182 register entry (priority, filing, publication dates): https://patent.public.lu/fo-eregister-view/search/details/601165470_EPV/0/0/1/10/0/0/0/null/null?locale=en
• CN 105820124 A (binimetinib synthesis; background identifying WO 2007/002157 and WO 2013/142182): https://patents.google.com/patent/CN105820124A/en