US 9486426

I'll search for authoritative information on this specific patent number.

Let me run additional targeted searches to confirm the litigation status and rule out any CAFC 2026 activity specific to this patent number.

Let me verify the claim set and check for any 2026 Federal Circuit activity specifically involving the '426 patent.

US Patent 9,486,426 — Summary

Note on sourcing: I retrieved the patent's own full text (Google Patents, fetched 2026‑10‑01) plus two independent third‑party databases (DrugPatentWatch, RPX/Insight) that reproduce the granted claim set. Where sources conflict or were truncated, I flag it below. I ran targeted searches on the exact number 9,486,426 / "9486426" and filtered out nearby numbers (e.g., 9,486,426 vs. 9,050,302 vs. 10,213,400, which are different, related patents and are not reported here).

Bibliographic data

Abstract

"One embodiment of the present invention is to improve the safety and efficacy of the administration of GHB or a salt thereof to a patient. It has been discovered that the concomitant administration of an MCT inhibitor, such as diclofenac, valproate, or ibuprofen, will affect GHB administration. For example, it has been discovered that diclofenac lowers the effect of GHB in the body, thereby potentially causing an unsafe condition. Furthermore, it has been discovered that valproate increases the effect of GHB on the body, thereby potentially causing an unsafe condition."

Plain-language overview of the independent claims

The patent is a dosing / drug‑interaction method‑of‑use patent: it covers adjusting a patient's sodium oxybate (Xyrem®) dose when an MCT‑inhibitor drug — chiefly divalproex sodium (valproate) — is given at the same time, because valproate raises GHB exposure.

Claim 1 — Treating cataplexy‑in‑narcolepsy or excessive daytime sleepiness‑in‑narcolepsy in a patient already on GHB/salt, by: (a) giving a dose of divalproex sodium together with a GHB dose, and (b) reducing the daily GHB dose by about 5%–50%, where the patient's non‑concomitant daily GHB dose would have been 4.5 g to 9 g.

Claim 9 — Same patient/indication population, but framed without the "administering divalproex" step: reducing the daily GHB dose by about 5%–50% during concomitant divalproex sodium administration, compared with the 4.5 g–9 g daily GHB dose the patient uses absent divalproex.

Claim 15 — (text partly truncated in the sources I retrieved; independent) A variant directed to a patient currently taking divalproex sodium, beginning with "administering to the patient a starting daily dosage amount of gamma‑hydroxybutyrate…". I could not retrieve the full claim language — treat this description as incomplete.

Claims 22 and 28 — Listed as independent by the defendants' contentions, but I was not able to retrieve their full text in the sources available. I am not asserting what they recite.

Selected dependents (fully retrieved): claim 2/11 (monitor patient response and adjust GHB dose); claims 3–6 and 12 (baseline non‑concomitant dose is 4.5 g, 6 g, 7.5 g, or 9 g); claims 7–8 (reduction of about 15%–50% and about 20%–50%); claim 13 (about 15%–50% reduction); claim 14 (at least 20% reduction).

Litigation posture (as of the search results)

• This patent has been asserted in numerous Hatch‑Waxman / ANDA actions in the District of New Jersey against generic Xywav®/Xyrem® filers — e.g., Jazz Pharmaceuticals Ireland Ltd. v. Lupin (2:21‑cv‑14271; 2:23‑cv‑00329) and Jazz Pharmaceuticals Ireland Ltd. v. Teva (2:23‑cv‑01617, terminated 2023‑12‑15), among others. Jazz's complaint treats the '426 patent as one of a family of "co‑administration patents" sharing priority to the March 1, 2013 provisionals (along with U.S. 9,050,302; 8,772,306; 10,213,400; 10,864,181; 11,253,494).
• Also cited in Jazz Pharmaceuticals, Inc. v. Avadel CNS Pharmaceuticals, LLC (D. Del. 1:21‑cv‑01138) and in antitrust/REM S‑related suits such as City of Providence v. Jazz Pharmaceuticals PLC (N.D. Cal. 3:20‑cv‑04064).
• No Federal Circuit 2026 docket specifically naming the '426 patent was found. The 2026‑dated Federal Circuit activity I located — Jazz Pharmaceuticals, Inc. v. Avadel CNS Pharmaceuticals, LLC, No. 24‑2274 — concerns other patents (e.g., the '782 patent and related sustained‑release/REM‑S patents), not the '426 patent. The Google Patents record lists several 2025–2026 New Jersey district‑court entries (e.g., 2:25‑cv‑14606, 2:26‑cv‑01739, 2:26‑cv‑01740) but provisioned no docket detail in my results.

Uncertainty / caveats

1. Claim text for independent claims 15, 22, and 28 is incomplete or unavailable in the sources retrieved; the overview above is authoritative only for claims 1–14.
2. The Google Patents "Definitions" block is an OCR/summarized rendering, not the literal specification, and contains typographical anomalies (e.g., "reduced at least 10% to 30%, or at least +15%"). I therefore relied on the third‑party claim reproductions for verbatim claim wording.
3. One court filing snippet references "Claim 18 of the '426 patent is invalid under pre‑AIA 35 U.S.C." Because the same document also addresses ten other patents, I cannot confirm whether that heading applies to the '426 patent; the pre‑AIA question is at least arguable given the March 1, 2013 priority date. I flag this rather than resolve it.
4. I found no evidence of a granted IPR or Federal Circuit decision holding the '426 patent invalid; its legal status is listed as Active.

Sources: Google Patents (patents.google.com/patent/US9486426/en); drugpatentwatch.com/p/patent-claims/9486426 and /p/patent-exob/9486426; insight.rpxcorp.com/patent/US9486426B2; Docket Alarm Orange Book records for NDA 212690 (XYWAV) and NDA 021196 (XYREM); CourtListener (D.N.J. 2:21‑cv‑14271 invalidity contentions); Justia (Fed. Cir. 24‑2274).

I'll research litigation involving US 9,486,426 specifically. Let me run several targeted searches.

Let me dig into the specific case numbers and parties.

Summary of known litigation involving US 9,486,426 ("the '426 patent")

Method note / caveats up front: The '426 patent (Jazz Pharmaceuticals Ireland Ltd.; inventor Mark Eller; issued Nov. 8, 2016; priority Mar. 1, 2013; application 14/707,914) is listed in the Orange Book against both Xyrem® (sodium oxybate) and Xywav® (calcium/magnesium/potassium/sodium oxybates, NDA 212690), so it has been swept into Jazz's large multi-patent oxybate campaigns. Two important cautions:

1. Google Patents' litigation box on this page is labeled "Family has litigation" — i.e., some listed case numbers may involve sibling patents in the same family (e.g., 8,772,306; 9,050,302; 10,213,400), not necessarily the '426 patent itself. I have tried to distinguish verified from unverified.
2. I hit a search-step limit before resolving every case number, so items below marked UNVERIFIED should be checked against PACER/CourtListener before being relied on.

A. Cases I verified as asserting the '426 patent

1. Jazz Pharmaceuticals Ireland Ltd. v. Lupin Ltd., Lupin Inc. & Lupin Pharmaceuticals, Inc.

• Jurisdiction: U.S. District Court for the District of New Jersey
• Case No.: 2:21-cv-14271
• Filed: July 28, 2021
• Basis: Lupin's ANDA No. 215911 for generic Xywav®; complaint expressly lists 9,486,426 (the '426 patent) among the patents-in-suit (ten patents: '922, '306, '173, '302, '107, '426, '168, '400, '258, '181).
• Status: Answer/counterclaims filed Oct. 4, 2021 (Lupin counterclaimed invalidity/non-infringement). Consolidated with later Lupin suits (2:22-cv-02773, filed May 11, 2022; 2:23-cv-00329, filed Jan. 19–20, 2023; 2:24-cv-08786, filed Aug. 27, 2024) and, on Dec. 15, 2023, consolidated with the Teva case for all purposes. Active through at least early 2026 (Jan. 26, 2026 opinion addressing expert/invalidity disputes). No final merits judgment on the '426 patent found.

2. Jazz Pharmaceuticals Ireland Ltd. v. Teva Pharmaceuticals, Inc.

• Jurisdiction: D.N.J.
• Case No.: 2:23-cv-01617
• Filed: March 21, 2023 (complaint docketed)
• Basis: Teva ANDA for generic Xywav®; complaint lists 9,486,426 (the '426 patent) among the patents-in-suit (along with '107, '168, '400, '173, '302, '258, '181, '494, '373, '102).
• Status: Teva filed invalidity contentions and counterclaims. Consolidated with the Lupin actions on Dec. 15, 2023. Ongoing; no final judgment on '426 found.

3. Jazz Pharmaceuticals Ireland Ltd. v. Alkem Laboratories Ltd.

• Jurisdiction: D.N.J.
• Case No.: 2:23-cv-03182
• Filed: June 9, 2023
• Basis: Xyrem®-related ANDA; the complaint asserts the '426 patent (Exhibit C) plus '306, '302, '400, '181, '494.
• Outcome: Voluntarily dismissed — notice of voluntary dismissal Oct. 4, 2023; order under Fed. R. Civ. P. 41(a)(1)(A)(i) signed Oct. 5, 2023 (Judge Stanley R. Chesler). Case terminated.

4. Jazz Pharmaceuticals Ireland Ltd. v. Tris Pharma, Inc. (two related suits)

• Jurisdiction: D.N.J. (Judge Stanley R. Chesler / Magistrate Judge Jessica S. Allen)
• Case Nos.: 2:26-cv-01739 and 2:26-cv-01740
• Filed: February 20, 2026 (docketed; case assignment Feb. 21, 2026)
• Basis: Tris Pharma's January 2026 Section 505(b)(2) NDA for generic Xyrem/Xywav; both dockets list 9,486,426 among asserted patents (2:26-cv-01739 also lists '400, '181, '494, '446, '306, '302).
• Status: In 2:26-cv-01740, the court denied Tris Pharma's motion to dismiss (Opinion & Order, Judge Chesler, docketed July 8, 2026). Both matters appear active.

5. Jazz Pharmaceuticals, Inc. v. [Amneal Pharmaceuticals LLC](/litigations/by-plaintiff/Amneal%20Pharmaceuticals%20LLC)

• Jurisdiction: D.N.J.
• Case No.: 2:17-cv-01440
• Filed: 2017 (Amneal's answer/counterclaims dated July 10, 2017)
• Basis: sodium oxybate ANDA No. 203631; Amneal's counterclaims expressly assert "Each claim of the '426 Patent is invalid…" — confirming '426 was in suit.
• Status: Outcome not confirmed in the sources retrieved; flag as unresolved (this family of Jazz/Amneal Xyrem cases was ultimately settled, but I could not verify the disposition of the '426 claims specifically).

B. Case numbers listed on the Google Patents page that I could NOT verify as involving the '426 patent (UNVERIFIED)

These appear in the "Family has litigation" list but I did not confirm that the '426 patent itself was asserted, nor the parties/outcomes:

• 2:18-cv-08267 — D.N.J. (2018)
• 2:18-cv-00029 — D.N.J. (2018)
• 2:17-cv-05487 — D.N.J. (2017)
• 2:17-cv-04439 — E.D.N.Y. (2017)
• 2:25-cv-14606 — D.N.J. (2025)

Given Jazz's 2017–2018 wave of Xyrem® ANDA suits, some of these are plausibly gegen generic filers, but I cannot confirm the party names, whether '426 was asserted, or the outcomes.

C. Non-infringement matters citing the '426 patent

New York State Teamsters Council Health and Hospital Fund v. Jazz Pharmaceuticals, Inc. (and related Xyrem antitrust/class actions)

• Jurisdiction: U.S. District Court for the Northern District of California (Judge Richard G. Seeborg)
• Case No.: 3:20-cv-04056
• Filed: June 18, 2020
• Nature: Antitrust (15:15), not patent infringement. The complaint lists 9,486,426 among the patents at issue in Jazz's Orange Book/patent-settlement conduct. This is part of the broader Xyrem/Xywav pay-for-delay antitrust litigation (N.D. Cal.), which also produced the Aetna and MSP actions discussed in Jazz's SEC filings.
• Status: Antitrust class litigation; separate from any infringement claim on '426.

D. Definitively NOT about the '426 patent (to avoid confusion)

• Jazz Pharmaceuticals, Inc. v. Roxane Laboratories, Inc., 2:10-cv-06108 (D.N.J.) and the 2011–2013 consolidated Xyrem formulation cases (e.g., 2:11-cv-00660, 2:11-cv-02523, 2:12-cv-06761, 2:13-cv-00391) predate the '426 patent and assert the '431, '889, '219, '506, '059, '650, '275, '730, '106, '107 patents — not '426.
• Jazz v. Avadel (Lumryz) suits in D. Del. and the D.C. Circuit FDA appeal concern sustained-release formulation patents (e.g., the '782 patent) and orphan-drug exclusivity — the '426 patent was not among the asserted patents in the Avadel trial described in Jazz's filings.

E. It is worth noting

• I found no IPR/PGR or other PTAB validity challenge specifically against the '426 patent.
• I found no CAFC appeal specifically adjudicating the '426 patent (the Federal Circuit activity in this space concerns the Avadel Orange Book delisting order, a different patent).

If you need certainty on the UNVERIFIED docket numbers (Section B) or the Amneal disposition, the reliable next step is a PACER/CourtListener query for each of 2:17-cv-01440, 2:17-cv-04439, 2:17-cv-05487, 2:18-cv-00029, 2:18-cv-08267, and 2:25-cv-14606, filtered to confirm whether "9,486,426" appears in the respective complaints.

I'll verify the structured data against live sources before reporting.

The searches surfaced references to IPRs against the "306 family" (which includes the '426 patent). Let me pin down specific proceedings on 9,486,426.

Now I need to determine whether any IPR specifically targeted the '426 patent (which issued 2016-11-08, after the main wave of petitions).

Proceedings overview

Total AIA trial proceedings on US 9,486,426: 0. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming this patent, and my web verification found none either — no petition, no institution decision, no Final Written Decision, no appeal, in the eleven years since the patent issued on 2016-11-08. The breakdown is therefore: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. Bottom line for a defendant: every claim of the '426 patent is untested and intact. There are no canceled claims to point to, no FWD to leverage, and — critically — no § 315(e)(2) statutory estoppel attaches to this patent from anyone. The patent is not hardened by PTAB wins, but it is also not narrowed, and the sibling-patent IPRs that Jazz defeated (below) tell you exactly which obviousness theories the Board has already found unpersuasive.

One caveat you should not miss: the patent is the third member of the "306 family" (8,772,306, 9,050,302, 9,486,426), and two of its three siblings were hit with IPRs. Follow-on antitrust/consumer complaints repeatedly say "when the '306 patents were challenged under inter partes review…" — that phrasing refers to the family, not to the '426 patent. No petition ever named 9,486,426.

Related proceedings (sibling patents — not on 9,486,426)

I include these because a defendant will inevitably run into them, but they are not AIA trials on this patent and they confer no estoppel against the '426.

IPR2016-00024 — Ranbaxy Inc. v. Jazz Pharmaceuticals, Inc. (U.S. 8,772,306)

• Type: Inter Partes Review
• Filed: 2015-10-07
• Status: Terminated — settled (2016-05-23), no FWD
• Judge panel: Not verified in the sources retrieved.
• Petition grounds: § 103 obviousness against claims 1–34 of the '306 patent (declarant Dr. David Rotella). The theory was that reducing the GHB dose when valproate is co-administered was routine optimization of a known drug–drug interaction.
• Institution decision: Partially instituted 2016-04-12 — trial instituted on claims 19–34 (16 claims), and denied as to claims 1–18. The Board credited Jazz's teaching-away evidence (the prior-art label discouraged co-administration at any dose) and found no reasonable expectation of success, holding that "further experimentation" would have been required. On claims 19–34, the Board agreed that the "warning" and "recommending" steps were likely printed matter entitled to no patentable weight.
• Final Written Decision: None — the case never reached one.
• Settlement / termination: Terminated 2016-05-23, 41 days after institution, in connection with the global Ranbaxy settlement announced 2016-05-09 (license to launch generic Xyrem on or after 2025-12-31; terms confidential). Ranbaxy's counsel filed a refund request the next day confirming "No substantive post-institution proceedings have taken place."
• Appeal: None.
• Defensive value: The Board's claims 1–18 denial is the single most useful roadmap Jazz has handed a future defendant and the most useful warning: the Board bought the teaching-away/no-expectation-of-success story on the record presented, but expressly invited fuller printed-matter briefing. The reasoning is not binding on a fresh panel and not estoppel.

IPR2016-00002 — Par Pharmaceutical, Inc. v. Jazz Pharmaceuticals, Inc. (U.S. 8,772,306)

• Type: Inter Partes Review
• Filed: 2015-10-06
• Status: Institution denied in its entirety (2016-04-12)
• Petition grounds: § 103, claims 1–34; overlapping art with Ranbaxy.
• Institution decision: Denied in full. Same teaching-away and no-reasonable-expectation-of-success reasoning.
• Appeal / settlement: None.
• Defensive value: Confirms the Board's unwillingness to institute on the dose-reduction claims on the 2015–2016 art record.

IPR2016-00546 — [Amneal Pharmaceuticals LLC](/litigations/by-plaintiff/Amneal%20Pharmaceuticals%20LLC) v. Jazz Pharmaceuticals, Inc. (U.S. 8,772,306)

• Type: Inter Partes Review
• Filed: petition accorded filing date 2016-02-08
• Status: Institution denied in its entirety (2016-07-28)
• Petition grounds: § 103, claims 1–34, with a substantively different attack — a detailed GHB metabolic-clearance-pathway / valproate-inhibition-of-GHB-dehydrogenase mechanism argument (declarant John R. Horn, Pharm.D.).
• Institution decision: Denied. Amneal's better mechanistic record did not cure the teaching-away and reasonable-expectation-of-success defects the Board had identified in the Ranbaxy/Par petitions.
• Defensive value: This matters for the '426 patent. The obviousness theory with the strongest mechanistic hook — MCT inhibition plus GHB dehydrogenase inhibition — was tried and failed at the institution stage.

IPR2016-00738 — Ranbaxy Inc. v. Jazz Pharmaceuticals, Inc. (reported as U.S. 9,050,302)

• Type: Inter Partes Review
• Filed: 2016-03-10
• Status: Institution denied; terminated in connection with the Ranbaxy settlement (Jazz's Q2 2016 10-Q: "both of the IPR petitions filed by Ranbaxy were terminated").
• Petition grounds: § 103 against the second divalproex-coadministration patent in the family.
• Note: I could not independently retrieve the institution-decision text or confirm the patent number mapping with complete certainty; treat the '302 mapping as reported, not verified.

Strategic summary

Claim status on the '426 patent. No claim has been canceled, disclaimed, or amended through any AIA trial. The entire claim set — including the independent claims directed to orally administering an adjusted GHB dosage during concomitant valproate (reduced dose) or diclofenac (increased dose), and the corresponding warning/labeling claims — is UNTOUCHED. I could not verify the '426 claim text from the sources retrieved, so I will not quote claim numbers; the patent is in force with an anticipated expiration of 2033-03-15 (2033-09-15 with pediatric exclusivity), and it is Orange Book–listed (use code U-1532) for both Xyrem and Xywav.

Estoppel landscape. This is the headline for a defendant. Because no IPR reached a Final Written Decision on the '426 patent, § 315(e)(2) estoppel does not apply to it at all. Even the sibling-patent proceedings produce no estoppel here — estoppel is patent-specific, and in any event IPR2016-00024 was terminated pre-FWD (which generally does not trigger § 315(e)) and the other three never instituted. Every § 102/§ 103 ground, every reference, and every combination remains available against the '426, subject only to: (i) the one-year bar of § 315(b) running from service of a complaint alleging infringement of this patent on you or a privy; and (ii) the '426's 2013-03-01 priority date (with a 2013-03-12 second provisional), which sharply limits citable art — you need references public before March 2013, and the real fight will be over whether the two 2013 provisionals support the specific dose-adjustment limitations under § 112.

Pattern signals. No petitioner ever filed multiple IPRs against the '426. There is no defensive aggregator in the chain — Unified Patents appears on this patent's page only as a litigation-database source for the many New Jersey and E.D.N.Y. Hatch-Waxman dockets, not as a challenger. Jazz has litigated the '426 aggressively and continuously (Alkem 2:23-cv-03182, Lupin 2:21-cv-14271, Teva 2:23-cv-01617, Granules 2:25-cv-14606, Tris 2:26-cv-01740, plus earlier Amneal/Par/Ranbaxy actions), and yet none of those defendants — including sophisticated generics with full paragraph IV invalidity contentions in hand — ever pulled the IPR trigger on the '426. The likeliest explanations are the family-wide teaching-away precedent from the '306 IPRs, the tight 2013 art window, and defendants choosing district-court invalidity (Lupin's second amended invalidity contentions run hundreds of pages against the '426) over a PTAB bet. Separately, follow-on complaints allege Jazz withheld its 2012 Xyrem label revision from the PTO and misrepresented the label to the PTAB in the '306 IPRs, and expressly frame this as inequitable conduct — that is a district-court unenforceability theory, not a PTAB issue, but it is a live defense worth developing.

Recommended next steps

• If you have been served with a complaint asserting 9,486,426: calendar the § 315(b) one-year bar immediately. You are the first mover on this patent at the PTAB, so there is no estoppel against you and no adverse FWD to distinguish — but you also have no free-win art package from a predecessor, so budget for a real prior-art search keyed to pre-2013-03-01 publications (the 2002/2005/2012 Xyrem labels, GHB/valproate pharmacokinetic literature, and MCT-inhibitor literature).
• Do not re-run the losing theories. The Board has now three times (IPR2016-00002, -00024, -00546) rejected "routine optimization of a known drug–drug interaction," even when fortified with GHB-clearance-pathway mechanism evidence. Your petition needs either (a) art that expressly discloses a quantified dose reduction of GHB with valproate, or (b) a § 112 written-description/priority attack on whether the 2013-03-01 and 2013-03-12 provisionals support the claimed dose-adjustment ranges — the latter is the angle the family IPRs never got to test.
• Exploit the priorities/printed-matter split. The Board's preliminary view in the '306 IPR that "warning of a potential drug/drug interaction" and "recommending reducing the dose" recites printed matter is a live, unresolved issue. If the '426's independent claims contain those steps, that ground deserves full post-institution treatment that IPR2016-00024 never received.
• Consider the two-track approach. Because there is no FWD on the '426, an IPR does not give you § 315(e)(2) estoppel protection against a parallel district-court theory — you would litigate both. Given the extensive district-court invalidity contentions already developed by Lupin and others, an IPR may be the supplement, not the substitute.
• Verify the negative. Before filing, pull the '426 patent's PTAB record directly at PTAB E2E and USPTO PatentCenter rather than relying on this summary; the ODP ingest block in the prompt is the canonical source and it shows zero proceedings, but confirm nothing was filed recently. Absence of PTAB activity on a patent asserted this aggressively for this long is itself the signal — but it is a signal that a challenge is available, not that one is easy.

I'll research the assignment record for US 9486426. Let me start with parallel searches.

I found a key reference in an IPR document. Let me dig deeper for the specific reel/frame numbers for the '426 patent and the correspondent.

Ownership & Assignment Analysis — US 9,486,426 B2

Important methodological caveat up front: the USPTO Assignment Center (https://assignmentcenter.uspto.gov/ and the parallel https://assignment.uspto.gov/patent/index.html) is an interactive search application and could not be queried directly from this session. The record below is reconstructed from (a) the Google Patents legal-events tab supplied in the authoritative patent text, and (b) a PTAB filing (IPR2016-00546, Paper 8, Notice of Related Matters) that states the reel/frame numbers for the sibling Eller patent US 8,772,306. I flag explicitly where a reel/frame pertains to the sibling patent rather than to the '426 itself. No reel/frame is invented.

Inventors

• Only one inventor — no multi-inventor team. This is a "label-expansion / drug–drug interaction" method patent built off Jazz's existing Xyrem programme, not a bench-science team.
• No unusual pattern of simultaneous inventor departure is observable from the record (single inventor; he assigned into Jazz rather than out of it). I cannot confirm his employment status beyond the assignment record — the assignment to Jazz is strong circumstantial evidence of an employee/contractor relationship at filing, but I am not asserting it as fact.

Original assignee

• Entity on the issued patent: Jazz Pharmaceuticals Ireland Ltd (Dublin, Ireland — a subsidiary of Jazz Pharmaceuticals plc, Nasdaq: JAZZ). The Google Patents face shows Original Assignee = Jazz Pharmaceuticals Ireland Ltd and Current Assignee = Jazz Pharmaceuticals Ireland Ltd.
• Line of business: specialty biopharmaceutical company; the '426 is an Orange Book-listed method-of-use patent (U-1532) for XYREM / XYWAV (sodium oxybate) covering "treating EDS and/or cataplexy in narcolepsy patients with sodium oxybate when divalproex sodium is concomitantly administered."
• Product embodying the claims: Yes, directly. Xyrem (sodium oxybate, NDA 21-196, first approved 2002) and Xywav (mixed oxybate salts, NDA 212690, approved 2020) are the marketed products; the '426 is listed against both. The claimed subject matter (dose reduction of GHB when valproate is co-administered) is reflected in the product labelling.
• Current status: Operating. Jazz Pharmaceuticals plc is a publicly traded, revenue-generating commercial-stage company. No bankruptcy, dissolution, or fire-sale. Its 10-Ks describe ongoing Xyrem/Xywav litigation and settlements.

Assignment timeline

The Assignment Center exposes two kinds of entries here: true assignments of title (the Eller→Jazz→Jazz Ireland chain) and security interests recorded against the patent as collateral. I separate them.

1. 2013-03-15 (executed, inferred from parent filing) / recorded 2015-05-14 — Reel/Frame: not confirmed for the '426 (see note)

• Conveyance: Assignment (inventor to company)
• Assignor: Mark Eller
• Assignee: Jazz Pharmaceuticals, Inc.
• Correspondent: Not determinable from the sources retrieved — a direct Assignment Center lookup is required. *(For the sibling '306 the recorded assignment from Eller to Jazz was Reel/Frame 30836/0953, recorded 2013-07-09, per IPR2016-00546 Ex. — but that is the '306's record, not established as the '426's.)*
• Context: inventor-to-employer assignment of title.

2. 2013 (executed, inferred) / recorded 2015-05-14 — Reel/Frame: not confirmed for the '426

• Conveyance: Assignment (intra-group transfer)
• Assignor: Jazz Pharmaceuticals, Inc.
• Assignee: Jazz Pharmaceuticals Ireland Limited
• Correspondent: Same caveat as above. *(Sibling '306: Jazz → Jazz Ireland at Reel/Frame 33480/0220, recorded 2014-07-29.)*
• Context: internal corporate reorganisation — the IP was parked in the Irish NDA-holding subsidiary, with the U.S. parent retaining an exclusive licence.

3. 2016-07-05 — Reel/Frame not retrieved

• Conveyance: Security Agreement (recorded as "SUPPLEMENT NO. 1 TO SECURITY AGREEMENT")
• Assignor: Jazz Pharmaceuticals Ireland Limited and Jazz Pharmaceuticals, Inc.
• Assignee: Bank of America, N.A., as Collateral Agent
• Correspondent: Not retrieved.
• Context: securitisation — lien over the IP portfolio supporting Jazz's credit facility. Not an ownership transfer.

4. 2021-05-05 — Reel/Frame appears in a sibling-patent legal-events feed as 056151/0010 (unverified for the '426)

• Conveyance: Security Agreement
• Assignor: Cavion, Inc.; Celator Pharmaceuticals, Inc.; Jazz Pharmaceuticals Ireland Limited; Jazz Pharmaceuticals, Inc.
• Assignee: U.S. Bank National Association
• Correspondent: Not retrieved.
• Context: securitisation / refinancing — new collateral agent replacing Bank of America. Not an ownership transfer.

5. 2021-05-05 — Reel/Frame not retrieved

• Conveyance: Release by Secured Party
• Assignor: Bank of America, N.A.
• Assignee: Jazz Pharmaceuticals, Inc.; Jazz Pharmaceuticals International Limited; Jazz Pharmaceuticals International III Limited; Celator Pharmaceuticals, Inc.; Cavion, Inc.; Jazz Pharmaceuticals Ireland Limited
• Correspondent: Not retrieved.
• Context: release of the 2016 Bank of America lien on refinancing. Not an ownership transfer.

Bottom line: there is one operating chain of title — Eller → Jazz Pharmaceuticals, Inc. → Jazz Pharmaceuticals Ireland Limited — plus three financing/lien records. No shell entity, no licensee-of-last-resort, no asserter ever appears.

Timeline diagram

timeline
    title Ownership of US 9486426
    2013 : Parent application filed
         : Eller assigns rights to Jazz
    2015 : Continuation application filed
         : Assignments recorded at USPTO
    2016 : Patent issued to Jazz Ireland
         : Bank of America security interest
    2021 : U S Bank security interest
         : Bank of America lien released
    2023 : First suit naming the 426 patent

NPE / troll-pattern signals

Additional non-signal worth noting: the presence of Bank of America and U.S. Bank security interests (2016-07-05 and 2021-05-05) is a routine portfolio-collateralisation pattern for a levered specialty pharma. It is common in normal operating-company financing and should not be read as an NPE tell.

Verdict

Operating-company assertion.

The entire chain of title runs from the sole inventor Mark Eller to Jazz Pharmaceuticals, Inc. and then to Jazz Pharmaceuticals Ireland Limited (recorded 2015-05-14 per Google Patents legal events) — an intra-group reorganisation, not a shell transfer. Jazz ships the product embodying the claims (Xyrem and Xywav, both Orange Book-listed against U-1532) and is suing actual competing ANDA/505(b)(2) filers in its own name — e.g. Jazz Pharmaceuticals Ireland Ltd v. Teva Pharmaceuticals, Inc., D.N.J. 2:23-cv-01617 (filed 2023-03-21), which expressly names the '426 patent. The only other recorded entries (2016-07-05 Bank of America security agreement; 2021-05-05 U.S. Bank security agreement and Bank of America release) are financing liens, not conveyances of title, so no cascading/asset-stripping pattern exists.

Verification links:

• USPTO Assignment Center (search patent 9486426): https://assignmentcenter.uspto.gov/
• Legacy Assignment Search: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events (source for the 2015/2016/2021 entries): https://patents.google.com/patent/US9486426/en

Open items requiring a direct Assignment Center session: the exact reel/frame for each of the five entries, the execution dates (as distinct from recording dates), and the name/firm of the correspondent of record on each. Reel/frames 30836/0953 and 33480/0220 are confirmed only for the sibling US 8,772,306 (per IPR2016-00546) and should not be attributed to the '426 without verification.

I'll research US patent 9486426 and its cited prior art references.

Let me dig deeper into the specific patent's citation list.

I have good context. Let me get the actual "Patent Citations" list from the Google Patents and USPTO records.

Prior-Art Analysis — US Patent No. 9,486,426 B2

1. Patent Under Analysis (confirmed, interpreted literally)

The specification states the '426 patent is a continuation of 13/837,714 and claims benefit of 61/771,557 and 61/777,873. The controlling prior-art date is therefore the earliest priority, 1 March 2013; the sibling '302 claims are worth noting because they define the same inventive concept (treating narcolepsy cataplexy/EDS, administering divalproex sodium, and reducing the GHB daily dose by ≥20% from a 4.5–9 g baseline).

2. Important Scope Limitations (stated explicitly, not glossed over)

Two honest caveats before the tables:

1. The authoritative Google Patents text supplied for this task does not contain the front-page "Patent Citations" table. That extraction reproduced the abstract, definitions, description, and the in‑specification "incorporated by reference" lists — but not the examiner's citation tables (Patent Citations / Non‑Patent Citations / Cited By). My follow-up searches for that exact table were cut off by a tool-step limit. I therefore cannot claim to have enumerated the complete examiner citation list for US 9,486,426. What I can do reliably is analyze every reference that the patent itself identifies (and that the related litigation/IPR record identifies).
2. The supplied text does not include the issued claim set of US 9,486,426. Accordingly, the § 102 mapping below is keyed to (a) the claim-like embodiments recited in the specification and (b) the claims of the directly related parent US 9,050,302, with the basis for each mapping labeled. Where I cannot tie a reference to a specific numbered claim of the '426 patent, I say so rather than inventing a claim number.

3. Part A — Patent References Cited / Incorporated by Reference in US 9,486,426

Every one of the following appears in the "Detailed Description" of the '426 specification as expressly incorporated by reference. Dates are given where I have high confidence; where I do not, I flag it.

A1. GHB formulation / salt patents (teach concentrations and pH)

A2. Valproate chemistry patents

A3. Pharmacy-management / REMS patents — the strongest § 102 candidates among the cited patents

The '426 specification expressly states that the pharmacy‑management/REMS system "can be a REMS system as shown in U.S. Pat. Nos. 7,895,059; 7,797,171; and 7,668,730 and also include monitoring for concomitant use of diclofenac, valproate, or ibuprofen."

A4. Other incorporated references

• US 6,780,889 (Cook et al.), issued 2004‑08‑24 — confirmed date from the IPR record. GHB/GABA‑related art; also asserted as prior art in the parallel IPR against US 8,772,306. Relevant to the pharmacology background, not to dose adjustment.
• US 7,262,219; US 7,851,506 — Jazz GHB‑formulation family (dates not verified here).
• Ser. No. 13/071,369; Ser. No. 12/264,709; PCT/US2010/033572; PCT/US2009/061312; US 2009/0137565; US 2012/0076865; Prov. 61/317,212 — GHB formulation/analog family; background.

Bottom line for Part A: none of the patent references cited in US 9,486,426 discloses the core inventive concept — that an MCT inhibitor (valproate/diclofenac/ibuprofen) alters GHB exposure and that the GHB dose must be decreased (valproate) or increased (diclofenac). They anticipate at most: (i) GHB salt/composition claims, (ii) GHB concentration/pH limitations, or (iii) the pharmacy‑distribution‑with‑alert embodiment.

4. Part B — Non-Patent Prior Art (where the real anticipation fight is)

These are the references actually relied on in the co‑pending IPR against the sibling Jazz patent US 8,772,306 (same inventor, same family, same 1 March 2013 priority) — i.e., the art a challenger would use against US 9,486,426. Per the Ranbaxy and Amneal IPR petitions (PTAB), these were framed as pre‑AIA 35 U.S.C. § 102(b) references published more than one year before the 1 March 2013 priority date.

5. Part C — § 102 Anticipation Mapping

Because the '426 claim set was not in the supplied text, I map to the disclosed subject matter below and flag confidence.

Claim group 1 — Valproate-directed dose reduction ("orally administering an adjusted dosage amount of GHB … when the patient is receiving concomitant valproate"; reduction of ~1–50%, e.g. ≥10–30%, ≥15%).

• Best § 102 candidates: Hechler (1997), Shinka (2003), Vayer (1988), Snead (1980).
• Assessment: These teach that valproate interacts with the GHB/GABA system, but none of them discloses administering a reduced GHB dose in a patient on valproate. Anticipation requires all elements in a single reference, including the dosage-adjustment step. On the record I have, no single reference anticipates this group under § 102; this is a § 103 (obviousness) battleground, which is precisely how Jazz's counterpart patents were attacked in the IPRs.
• Only if the '426 claims are directed to the valproate dose‑reduction method (as US 9,050,302 is), the anticipation case would rest on a hypothetical single reference combining a Xyrem‑label‑type disclosure with an explicit valproate–GHB interaction — no such single reference appears in the material I retrieved.

Claim group 2 — Diclofenac-directed dose increase.

• No cited patent or NPL reference discloses increasing the GHB dose to compensate for diclofenac. This group is not anticipated by anything in the retrieved record.

Claim group 3 — Pharmacokinetic findings (ibuprofen doubles renal excretion; diclofenac lowers PD effect; divalproex raises AUC ~26%, renal clearance +30%).

• These are data disclosures; the closest art is Bhattacharya (2006) (MCT transport inhibition → GHB DDIs), which supports obviousness rather than strict anticipation.

Claim group 4 — Pharmacy distribution / REMS method with MCT‑inhibitor warning.

• Best § 102 candidates: US 7,895,059; US 7,797,171; US 7,668,730; US 7,765,106; US 7,765,107 and the generic pharmacy‑management patents (5,758,095; 5,833,599; 5,845,255; 6,014,631; 6,067,524; 6,112,182; 6,317,719; 6,356,873; 7,072,840). If any of these discloses a distribution workflow that conditions dispensing on a drug‑interaction alert, that reference could anticipate a claim to distributing GHB with MCT‑inhibitor risk information. This is the most realistic § 102 exposure among the cited patents — and the '426 specification itself cites these as the system to be modified.

Claim group 5 — Formulation limitations (GHB 350–750 mg/mL, pH 6–10; single/mixed salts).

• Potentially anticipated by US 6,472,431 / US 8,263,650 / US 8,324,275 / Ser. No. 13/739,886 / US 4,393,236 to the extent those teachings overlap the recited concentration/pH/salt ranges. But such claims would be tied to a distinctly narrower inventive scope than the drug‑interaction methods.

Claim group 6 — Avoiding concomitant diclofenac/valproate; using diclofenac to reduce GHB toxicity; using valproate to potentiate GHB.

• The "avoidance" concept is a negative‑limitation/instructional claim; the cited patents do not disclose it. The "diclofenac to treat GHB toxicity" and "valproate to potentiate GHB" embodiments are the mirror image of the interaction teaching and are not anticipated by any single reference retrieved.

6. Part D — Real-World Prior-Art Challenges (context)

The related Jazz family patent US 8,772,306 (filed 2013‑04‑29, same inventor/family) was challenged by Ranbaxy and Amneal in IPRs at the PTAB (Petitions 1463743 and 1463786), with primary references Maitre (1997), Okun (2001), the Xyrem® PI (2007), the Xyrem® Titration Schedule (2008), and, in the Amneal petition, Hechler (1997), Shinka (2003), Cagnin (2011), the Depakote label (2011), the FDA Drug‑Interaction Guidance (2012), Kaufman (1991), Weiss (2012), Vayer (1988), Snead (1980), Waszkielewicz & Bojarski (2004), and the Orange Book listings. These are the references that matter most for validity of US 9,486,426, because US 9,486,426 shares the same priority and disclosure. Note also that litigation over the family is documented (multiple New Jersey D.N.J. actions against Roxane, Amneal, Mallinckrodt, Par, and others, including 2:18‑cv‑00029 and 2:18‑cv‑08267 indexed on the '426 Google Patents page).

7. Bottom Line

1. Patent most directly affected: US 9,486,426 is a continuation of 13/837,714 → US 9,050,302, and its closest prior‑art exposure mirrors the '302/'306 family.
2. Cited patents: none of them discloses the core invention (MCT‑inhibitor‑driven GHB dose adjustment). Their § 102 value is confined to (i) GHB salt/composition and (ii) pharmacy‑distribution/REMS claims (US 7,895,059 / 7,797,171 / 7,668,730 and the generic pharmacy‑management patents being the best candidates).
3. Cited non‑patent art is the real threat. The most relevant anticipation/obviousness references are Hechler (1997) (valproate blocks GHB effects), Shinka (2003) (valproate alters GHB metabolites), and especially Bhattacharya (2004, 2006) (GHB carrier/MCT transport and predicted GHB drug–drug interactions), together with Maitre (1997), Okun (2001), and the Xyrem® PI (2007) for the narcolepsy/dosing background.
4. No single retrieved reference anticipates the valproate dose‑reduction or diclofenac dose‑increase methods under § 102; those claims rise or fall on § 103 obviousness, which is how the family was actually challenged in the IPRs.
5. Explicit limitation: the front‑page examiner "Patent Citations" table for US 9,486,426 and the issued claim set of the '426 patent were not present in the authoritative text I was given, and my attempts to retrieve them were halted by a search tool‑step limit. I have therefore analyzed every reference the patent and the litigation/IPR record identify rather than presenting a guessed citation list. If you supply the patent's "References Cited" front‑page table (or the claims), I can produce a per‑claim § 102 chart with specific claim numbers.

Sources: Google Patents US9486426 (patent text, family, priority, litigation); PTAB/Unified Patents records for the sibling US 8,772,306 IPRs (Ranbaxy Pet. 1463743; Amneal Pet. 1463786); DrugPatentWatch entries for US 9,050,302; Espacenet US11986446 family data; GreyB/Pharsight Jazz patent‑portfolio and Xyrem/Xywav expiry pages.

I'll research the prior art cited on this patent's page and any related challenges before building the §103 analysis.

Let me dig into the specific claims and the prior art used against this patent family.

Let me confirm the claim set and any IPR/litigation outcomes specific to the '426 patent.

Obviousness Analysis — U.S. Patent No. 9,486,426 B2

Analyst's note on sources and method. The Google Patents page for US9486426 as fetched lists only prior‑art keywords ("ghb, patient, salt, daily dosage, dosage amount") and does not surface a discrete, itemized "Prior Art" reference list. I therefore built the reference set from (a) the prior‑art references expressly discussed and incorporated in the '426 specification itself, and (b) the references actually asserted against the three sibling patents that share the same provisional priority (US 8,772,306; US 9,050,302) in the IPR petitions and district‑court invalidity contentions. Where I could not verify something (e.g., whether an IPR was ever filed against the '426 patent specifically), I say so explicitly. All patent numbers and identifiers are read literally.

1. The Patent and Its Effective Date

Governing law. Because the '426 patent is a continuation of an application (13/837,714) filed before 16 March 2013, and it claims benefit to provisionals filed 1 and 12 March 2013, the pre‑AIA version of 35 U.S.C. § 102/§ 103 applies (see the Federal Circuit's statement in the family litigation that "the applications for each of the patents in suit were filed before March 16, 2013," so pre‑AIA § 102 governs — CourtListener, N.D. Cal. 3:16‑cv‑…‑361158). The legal framework is Graham v. John Deere Co., 383 U.S. 1 (1966) and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

2. The Claims at Issue

Claims 1–9 are the issued claims (per RPX Insight and DrugPatentWatch). Notably, no claim recites any monocarboxylate‑transporter (MCT) limitation, notwithstanding the title and abstract — the claims are pure methods‑of‑treatment dose‑adjustment claims. Claim 1, decomposed:

(a) "A method for the treatment of cataplexy in narcolepsy or excessive daytime sleepiness in narcolepsy"
(b) "in a patient who is currently taking GHB or a salt thereof"
(c) "comprising: administering to the patient a dose of divalproex sodium concomitant to a dose of GHB or salt thereof; and"
(d) "reducing the daily dosage amount of GHB… between about 5% and about 50%"
(e) "wherein the daily dosage amount of GHB… in the absence of concomitant administration of divalproex sodium is between 4.5 g to 9 g."

Dependents:

• 2 – monitoring patient response and adjusting the GHB dose to maintain effect
• 3, 4, 5, 6 – baseline (no‑valproate) dose is 4.5 g / 6 g / 7.5 g / 9 g, respectively
• 7 – reduction of about 15%–50%
• 8 – reduction of about 20%–50%
• 9 – independent claim, same as claim 1 but omitting the step of administering divalproex; recites "reducing the daily dosage amount… during concomitant administration of divalproex sodium, compared to the daily dosage amount of between 4.5 g and 9 g… currently used in the absence of concomitant administration."

Claim 9 is materially broader than claim 1 (no affirmative co‑administration step) and is the closest analogue to sibling US 8,772,306 claim 1, which the PTAB and district courts have already addressed. It is the most exposed claim.

3. Prior Art Available Before 1 March 2013

Family‑specific note: In consolidated N.D.N.J. invalidity contentions, Lupin expressly contended that "claims 1–12 of the '426 patent" are invalid for obviousness, incorporating by reference its '306‑patent analysis (Xyrem Label, Cagnin, Hechler, Kaufman 1987/1991, Maitre, Okun, Feldman, Frucht, Harvey, the Xyrem EPAR, the FDA Drug Safety Communication, and the Xyrem 2005/2012 PIs). See E.D.N.J./D.N.J. contentions, 2:21‑cv‑478299, D.I. 244‑2.

4. The Obviousness Grounds

Ground 1 — Xyrem® PI (PA‑1) + Depakote® Label (PA‑2) + Cagnin (PA‑3) + Waszkielewicz (PA‑4) [± FDA Guidance (PA‑10)]

This is the ground Par Pharmaceutical advanced against the sibling '306 patent (IPR2016‑00002), and it maps onto '426 claim 1 nearly element‑for‑element:

Motivation to combine. All four references are in the same field (CNS pharmacology / the GHB and valproate arts), were familiar to a POSA, and the Xyrem label itself supplies the express motivation: it tells the prescriber that when a CNS depressant must be used with Xyrem, dose reduction should be considered. The Depakote label confirms valproate is a CNS depressant and standard texts (PA‑12, PA‑7) confirm it inhibits GHB dehydrogenase, so the "dose reduction" instruction is not merely generically applicable — it is directly on point. PA‑3 supplies a real‑world adverse‑event signal that the combination is dangerous, i.e., a safety motivation of the strongest kind.

Reasonable expectation of success. GHB is a titrate‑to‑effect drug with a documented, stepwise 1.5 g titration ladder (4.5 → 6 → 7.5 → 9 g). A POSA adjusting the dose downward one or more label steps would land on objectively predictable figures:

• 9 g → 7.5 g = 16.7%
• 7.5 g → 6 g = 20%
• 6 g → 4.5 g = 25%
• 4.5 g → 3 g = 33.3%

Each of those falls inside the claimed "about 5% to about 50%" (claim 1/claim 9) and inside "about 15%–50%" (claim 7) and "about 20%–50%" (claim 8). This is the paradigm "result‑effective variable" case: the dose of a drug already known to be titrated against clinical effect was a recognized parameter to optimize (In re Boesch, 617 F.2d 272 (CCPA 1980); In re Applied Materials, 692 F.3d 1289 (Fed. Cir. 2012)). Using known options within known ranges is prima facie obvious (In re Peterson, 315 F.3d 1325 (Fed. Cir. 2003)).

Ground 2 — Xyrem® PI + Hechler (PA‑5) + Kaufman (PA‑6) + Depakote® Label (PA‑2) ± Shinka (PA‑8)

Amneal's ground. PA‑5 and PA‑6 establish mechanistically that valproate inhibits GHB dehydrogenase and raises brain GHB levels (1.4‑fold in rats), which is precisely the mechanism the '426 specification alleges ("GHB dehydrogenase inhibition caused systemic exposure (plasma AUC) to be increased 26%"). A POSA who knows the enzyme is inhibited and has a label instructing dose reduction where a CNS depressant is co‑used has an express rationale for reducing the GHB dose, and a rational expectation that reduced input compensates for reduced clearance. PA‑8 supplies a human clinical corroboration in the SSADH‑deficiency population that PA‑1 already flags.

Ground 3 — Maitre (PA‑7) + Xyrem® PI + Okun (PA‑9) [± Kaufman, ± Sandson]

Ranbaxy's ground against sibling '302. PA‑7's statement that valproate "lead[s] to the accumulation of GHB in brain," combined with the Xyrem PI's titration instructions and PA‑9's recitation of GHB's narrow therapeutic index, yields the claimed method for all of claims 1–9, with the specific gram baselines (claims 3–6) literally disclosed in the Xyrem PI titration table.

Ground 4 — Any of the above in view of the Xyrem® EPAR (PA‑12)

PA‑12 is unusually direct: it states that GHB is metabolised by GHB dehydrogenase and that "there is a potential interaction with drugs that inhibit this enzyme," expressly "[includ[ing]… valproate…" This closes any residual "no human data" gap because it puts the valproate–GHB dehydrogenase interaction squarely in a regulator‑reviewed product document for Xyrem itself.

Ground 5 — Claim‑specific mappings

5. Claim 1 Is Broader Than Claim 9 — and Both Are Broader Than the Spec's "MCT Inhibitor" Theme

Two structural vulnerabilities worth flagging:

1. No claim recites an MCT limitation, and no claim recites diclofenac or ibuprofen. The title, abstract, and much of the specification are directed to MCT inhibition and to increasing GHB dose with diclofenac, but the issued claims cover only the valproate/divalproex dose‑reduction embodiment. The MCT prior art (PA‑11) is therefore not needed to invalidate the claims, but it is highly relevant to written‑description/support and to § 101 — which is exactly the attack Lupin raised ("Claims 1–12 of the '426 patent are invalid under 35 U.S.C. § 101 for claiming unpatentable subject matter… The interaction between GHB and valproate is a natural phenomenon, and the reduction in the GHB dosage is not [an inventive concept]").

2. Claim 9 omits the step of administering divalproex sodium, reciting only that the reduction occurs "during concomitant administration." That removes the affirmative co‑administration act that Jazz can point to as the inventive departure from the prior art's "don't co‑administer" teaching. Claim 9 is, on its face, the closest analogue to '306 claim 1.

6. The Counter‑Arguments (and Why a Challenger Could Still Lose)

This is where the analysis must be candid: the PTAB has already rejected substantially these grounds once. In decisions dated 12 April 2016, the Board denied institution in both IPR2016‑00002 (Par) and IPR2016‑00024 (Ranbaxy) as to the sibling '306 patent, and in IPR2016‑00546 the Board again refused to institute over Amneal's Xyrem Label + Hechler + Shinka + Cagnin + Depakote grounds. The Board reasoned:

• Teaching away. The Xyrem label carries a black‑box warning against use with other CNS depressants and contraindicates GHB in SSADH deficiency (a condition valproate mimics by inhibiting SSADH). The Board held that "[a] reference teaches away from a claimed invention if it 'criticizes, discredits, or otherwise discourages' modifying the reference to arrive at the claimed invention," citing In re Fulton, 391 F.3d 1195, 1201 (Fed. Cir. 2004), and KSR, 550 U.S. at 416. Where the prior art tells the physician to stop co‑administering, the argument runs, the discovery of a workable co‑administration regimen is non‑obvious.
• Unpredictability. Jazz successfully argued that GHB is eliminated through alternate pathways not inhibited by valproate, and that some of those pathways could decrease GHB levels — making the net effect of valproate on human GHB exposure (and therefore the correct direction and magnitude of any dose adjustment) unpredictable. See Amneal/Patent Owner Preliminary Response, IPR2016‑00546.
• No direction as to how much. The Board found no sufficient basis to conclude that elevated brain GHB from valproate "could have been predictably compensated for by a corresponding decrease of at least 5% in the amount of GHB orally administered to patients."

Rebuttals available to a challenger:

• Those decisions were institution denials, not final written decisions; the "reasonable likelihood" standard under 35 U.S.C. § 314(a) is not the preponderance standard of § 318(a). A district‑court or later proceeding evaluating the same art under a preponderance standard is not bound.
• The Xyrem 2012 label revision is a double‑edged sword: it moved from "should not be used in combination" to "If use of these CNS depressants in combination with Xyrem is required, dose reduction or discontinuation of one or more CNS depressants (including Xyrem) should be considered." That is no longer a teaching away from co‑administration; it is an express recipe for the claimed dose reduction. Compare the two labels side‑by‑side as set out in CourtListener, N.D. Cal. 3:16‑cv‑361158.
• The '426 claims differ from the '306 claims in ways that cut both ways. Because claim 1 affirmatively requires administering divalproex, a POSA treating a patient who needs both agents (e.g., narcolepsy plus epilepsy or bipolar disorder, each an approved indication for divalproex) has a direct motivation the Board did not squarely address when the prior art is read as forbidding co‑administration. But this also means Jazz can argue the claim solves the very co‑administration problem the art avoided.
• Claim 9, which omits the co‑administration step, is materially weaker than claim 1 on the teaching‑away point.

Secondary considerations (Jazz's likely rebuttal):

• Unexpected results / unpredictability, as accepted at the institution stage.
• Long‑felt need / skepticism toward combining GHB and valproate.
• Commercial success of Xyrem — but note the challenger's rejoinder that any success is attributable to the active ingredient and the 2001–2002 restricted‑distribution architecture in the prior art, not to the claimed dose‑adjustment (see the argument in IPR2015‑00545 petition, paragraphfour.com: "where the commercial success can be attributed to characteristics of the invention that were already in the prior art, non‑obviousness is not shown," citing Dippin' Dots, Inc. v. Mosey, 476 F.3d 1337, 1345 (Fed. Cir. 2007)).
• Nexus is the battleground: Jazz must tie any objective indicia to the specific 5%–50% reduction / 4.5–9 g baseline claim elements, not to GHB therapy generally.

7. Bottom Line

1. On the intrinsic record and family prior art, claims 1–9 of US 9,486,426 are prima facie obvious under pre‑AIA § 103(a). The strongest single combination is Xyrem® PI (PA‑1) + Depakote® Label (PA‑2) + Cagnin (PA‑3) + Waszkielewicz (PA‑4), optionally with the Feb. 2012 draft FDA Guidance (PA‑10); the strongest mechanism‑based alternative is Xyrem® PI + Hechler (PA‑5) + Kaufman (PA‑6) + Depakote® Label, with Maitre (PA‑7) + Xyrem® PI + Okun (PA‑9) as a third independent route.
2. The critical fact that drives the analysis is that the claimed numbers are the label's own numbers. Claims 3–6 (4.5, 6, 7.5, 9 g) are literally the Xyrem® titration steps, and the claimed 5%–50% / 15%–50% / 20%–50% reduction windows are exactly what one obtains by stepping down the label's 1.5 g increments (16.7%, 20%, 25%, 33.3%). Obviousness of a numerical range disclosed by the prior art is well settled (In re Peterson).
3. The defense that has worked so far is "teaching away" plus "unpredictability of the net PK effect." That defense carried the day at the § 314(a) institution stage for the sibling patents and is Jazz's principal shield here. It is strongest against claim 1 (which requires administering divalproex) and weakest against claim 9 (which requires only dose reduction during concomitant use).
4. I could not confirm from the available material whether an IPR petition was ever filed against US 9,486,426 itself, nor the current validity status of its claims. The challenges I verified (IPR2016‑00002, IPR2016‑00024, IPR2016‑00546, and the Lupin/Sandoz‑style district‑court contentions) are directed to US 8,772,306, US 9,050,302, and the broader Xyrem® Orange Book family. Treat the '426‑specific conclusions above as an analytical projection from those verified family proceedings rather than a report of a decided outcome.

This is a technical patent‑analyst work product, not legal advice or an attorney opinion on validity, infringement, or enforceability.