US 9050302

I'll search for authoritative information on this patent, including USPTO records and any 2026 CAFC activity.

Let me check for 2026 CAFC activity and confirm the granted claim set.

Let me verify the granted claim set and look specifically for 2026 Federal Circuit docket entries naming this patent.

US 9,050,302 — Verified Summary

Important scope note up front: I was able to confirm bibliographic data, prosecution/PTAB history, and much of the claim structure from the sources below. I could not confirm any 2026 Court of Appeals for the Federal Circuit docket naming 9,050,302. Details on that point are in the last section — treat that request as unverified rather than answered.

1. Bibliographic data (high confidence)

Source: USPTO IPR exhibit of the printed patent (Ranbaxy Ex. 1001, IPR2016-00738), https://gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2016-00738/Inter_Partes_Review_of_U.S._Pat._9050302/03-10-2016-Petitioner/Exhibit-1001-Ex_1001___US__Patent_No_9,050,302/ ; https://www.drugpatentwatch.com/p/patent/9050302

2. Abstract (verbatim, as printed on the patent)

"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."

3. Plain-language overview of the disclosure

The specification reports human drug-interaction studies on Xyrem® (sodium oxybate, i.e., GHB salt) co-administered with three monocarboxylate transporter (MCT) inhibitors:

• Valproate / divalproex sodium → increases GHB effect (GHB-dehydrogenase and MCT inhibition; ~26% increase in plasma AUC; renal clearance up ~30%). Reported to worsen cognitive/attention impairment.
• Diclofenac → reduces GHB effect (PD effects significantly reduced) — which the patent frames as potentially unsafe because the patient may be under-dosed.
• Ibuprofen → pharmacokinetic changes consistent with MCT inhibition; renal excretion of GHB doubled; plasma levels ~5% lower.

From this the specification claims methods of adjusting the GHB dose (down for valproate, up for diclofenac), avoiding/discontinuing the interacting drug, warning/monitoring via pharmacy management (REMS-style) systems, and package inserts/kits. It also proposes the reverse use — diclofenac to reduce GHB toxicity, valproate to potentiate GHB benefit — and addresses GHB dose ranges (e.g., 1–4.5 g/day or 6–10 g/day), concentrations (350–750 mg/ml; 450–550 mg/ml), pH (6–10; 6.5–8), and single vs. mixed salts (Na.GHB, K.GHB, Mg.(GHB)₂, Ca.(GHB)₂).

Caveat on identifiers/typos: the OCR'd text contains obvious string artifacts — e.g., "GLB" for GHB, "oxysorbate", "ibruprofen", "divalproate". Per your instruction I have not silently corrected them in quotations, but I flag them as transcription noise, not distinct compounds.

4. Independent claims (as granted)

The granted claims are much narrower than the published application US 2014/0249115 A1 (which had claims up to 96 covering valproate, diclofenac, ibuprofen, monitoring, and distribution). The Ranbaxy IPR petition against the '302 patent states the granted patent has independent claims 1, 8, 13, and 20, each of which requires concomitant GHB + valproate administration and a ≥20% reduction in the GHB daily dose relative to the 4.5 g–9 g/day range used absent valproate.

• Claim 1 (verbatim per DrugPatentWatch): "A method for the treatment of cataplexy in narcolepsy or excessive daytime sleepiness in narcolepsy in a patient who is currently taking gamma-hydroxybutyrate (GHB) or a salt thereof comprising: administering to the patient a dose of divalproex sodium concomitant to a dose of GHB or salt thereof; and reducing the daily dosage amount of GHB or salt thereof administered to the patient by at least 20% wherein the daily dosage amount of GHB or salt thereof in the absence of concomitant administration of divalproex sodium is between 4.5 g to 9 g."
• Claim 8 — same ≥20% reduction framework, phrased around "concomitant administration of [GHB and] divalproex sodium."
• Claim 13 and Claim 20 — framed around a patient "who is currently taking divalproex sodium." Claim 20, per a litigation chart: a starting daily GHB dose "20% lower than the starting daily dosage amount of between 4.5 g to 9 g GHB that would otherwise have been recommended to the patient if the patient was not currently taking divalproex sodium."
• Dependent claims add monitoring/titration to effect (claim 2), and specific baseline doses of 4.5 g, 6 g, 7.5 g and 9 g, plus reduced amounts.

Uncertainty flag: DrugPatentWatch lists a "claim 31" reading "A method for the treatment of cataplexy in narcolepsy… in a patient who is currently taking divalproex sodium comprising… administering… a dose of GHB… lower than 2.25 g; and… a second dose… lower than 2.25 g." That text conflicts with the IPR petition's identification of claims 1, 8, 13 and 20 as the only independents, and the same wording appears in litigation charts for the sibling '426 patent (also 31 claims, same 2013 provisional family). I therefore cannot state with confidence which of claim 31's verbatim text belongs to the '302 patent versus the '426 patent. Sources: https://www.drugpatentwatch.com/p/patent-claims/9050302 ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1463743](/patent/1463743)/ ; https://storage.courtlistener.com/recap/gov.uscourts.njd.[478299](/patent/478299)/gov.uscourts.njd.478299.244.2.pdf

5. Litigation and PTAB record naming 9,050,302

• IPR2016-00738, Petitioner Ranbaxy Inc. v. Jazz Pharmaceuticals Ireland Limited, filed Mar. 10, 2016; institution denied and proceeding terminated May 23, 2016 (joint motion to terminate; settlement with confidential business information request). Respondent application 13/837,714, Tech Center 1600. https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2016-00738
• IPR2016-00024 (against U.S. 8,772,306) cites U.S. 9,050,302 as Exhibit 1025.
• District court / MDL matters in which the '302 patent appears: D.N.J. ANDA cases (e.g., 2:15-cv-00469 Roxane; 2:15-cv-08229 Sun/Ohm/Ranbaxy — both answer the '302 infringement counts with §101/§102/§103/§112 invalidity defenses); N.D. Ill. antitrust matters 1:20-cv-03543 and 1:20-cv-03673; D. Minn. 0:21-cv-00737; N.D. Ohio MDL 1:17-md-02804. https://www.drugpatentwatch.com/p/alphasignals/litigation/patent/9050302 ; https://paragraphfour.com/uploads/cases16/njdc16cv0469A.pdf ; https://paragraphfour.com/uploads/cases15/njdc15cv8229A.pdf
• Related family: the '306 family — 8,772,306, 9,050,302, and 9,486,426 — all from provisional 61/771,557.

6. CAFC 2026 dockets for 9,050,302 — not found / cannot confirm

I ran targeted searches for Federal Circuit activity in 2026 involving the '302 patent and found no docket, order, or opinion naming U.S. 9,050,302 in a 2026 CAFC proceeding. Specifically:

• The 2025 Jazz v. Avadel appeals I retrieved — CAFC Nos. 24-2274, 24-2277, 24-2278, decided May 6, 2025 ("reversed-in-part, vacated-in-part, and remanded") — concern a different patent set (US 10,966,931; 11,147,782; 8,731,963; 10,758,488; 10,959,956; 11,077,079; 10,813,885), not the '302 patent. https://www.docketalarm.com/cases/US_Court_of_Appeals_Federal_Circuit/24-2274/ ; https://cases.justia.com/federal/appellate-courts/cafc/24-2274/24-2274-2025-05-06.pdf
• The '302 patent's own PTAB proceeding (IPR2016-00738) terminated by settlement in 2016, with no appeal.

Bottom line: based on the searches available to me, there does not appear to be a 2026 Federal Circuit docket for US 9,050,302. I cannot rule out a very recent or sealed 2026 filing that is not yet reflected in the indexed databases; I am flagging this as an information gap rather than asserting a negative as fact. If you need certainty, the authoritative check is the CAFC docket/RECAP search under Jazz Pharmaceuticals Ireland Limited plus the patent number, which I was unable to complete within the allotted searches.

I'll research the litigation history for US patent 9050302. Let me search multiple sources.

Let me search the specific district court case numbers and the PTAB proceeding.

Litigation involving U.S. Patent No. 9,050,302

I researched the patent's own Google Patents litigation record (the authoritative source supplied), plus PTAB/PACER-derived databases (Unified Patents portal, Docket Alarm/IPVerse, Justia, Stanford NPE database, Paragraph IV). Below is what is verifiable for U.S. Patent No. 9,050,302 specifically. Note the patent (Jazz Pharmaceuticals Ireland Ltd.; inventor Mark Eller; filed Mar. 15, 2013; granted June 9, 2015) is listed in the Orange Book for Xyrem® under use code U-1532, and is described as a "DDI" (drug-drug-interaction) patent in the '306 family.

1. District court cases listed in the patent's own litigation record

The Google Patents record for US 9,050,302 lists these matters (I did not independently verify the parties in every 2021–2026 action before my step budget ran out — flagged below):

The record also notes: "First worldwide family litigation filed" (family 51031733), i.e., the '302 patent is part of Jazz's global Xyrem enforcement family.

2. PTAB proceeding

The Google Patents record confirms this as "PTAB case IPR2016-00738 filed (Settlement)."

3. Important caveats

• The '302 patent itself vs. its family. Many Jazz Xyrem suits asserted the '306 family collectively — U.S. 8,772,306 ('306), 9,050,302 ('302), and 9,486,426 ('426). The '302 patent was asserted together with family members (e.g., the 2015 Amneal pleadings expressly count infringement "of the '302 Patent"). Where a case targeted only a sibling patent, it is not a '302 case, and I have not listed it as one.
• Several widely-reported Jazz v. Avadel / MSN cases in D. Del. and elsewhere involve different Jazz patents (e.g., '488, '782, '963, '986) and are outside the scope of this request; they are not '302 litigation.
• I could not, within my research budget, positively confirm plaintiff, defendant, and disposition for each of the 2015–2026 D.N.J. docket numbers in the table. Those rows are drawn from the patent's official record and should be verified on PACER / the Unified Patents litigation portal (https://portal.unifiedpatents.com/litigation/caselist?patents=9050302) before being relied upon.

Summary: Yes — litigation involving U.S. 9,050,302 is known. It is centered on Hatch-Waxman ANDA challenges to Jazz's Xyrem® sodium oxybate product, led by Jazz Pharmaceuticals, Inc. / Jazz Pharmaceuticals Ireland Limited as plaintiff/patent owner, filed predominantly in the District of New Jersey, plus one E.D.N.Y. action (v. Ascent) and one PTAB IPR (Ranbaxy, IPR2016-00738, settled/terminated in May 2016). If you need the parties and outcomes for the specific 2015–2026 D.N.J. dockets, I recommend pulling each docket directly from PACER, as I was not able to fully verify each one here.

I'll start by verifying the structured PTAB block against the litigation record, which showed a potential contradiction (IPR2016-00738 appears in the litigation record but not the ODP block).

I've found the key proceeding. Let me dig into the institution/termination details and check for any other '302 proceedings.

I have the core record. Let me confirm the patent's total claim count and rule out any other '302 proceedings.

AIA Trial Proceedings — U.S. Patent No. 9,050,302 ("Method of administration of gamma hydroxybutyrate with monocarboxylate transporters")

⚠️ Contradiction flagged up front. The structured "PTAB proceedings on file" block in my instructions states the USPTO ODP API returns no AIA trial proceedings for this patent. That is incorrect / incomplete. A full IPR was filed against this exact patent: IPR2016-00738, Ranbaxy Inc. v. Jazz Pharmaceuticals, Inc. / Jazz Pharmaceuticals Ireland Ltd., filed 2016-03-10, and terminated 2016-05-23. The petition, its exhibit list, the patent owner's mandatory notices, and the Board's termination Judgment are all public on PTAB E2E. Because the ODP block is not consistent with the primary record, I have treated the web-sourced PTAB docket as authoritative and built the analysis on it, per the instruction to prefer search results over stale/default data and to flag any proceeding the ODP hasn't indexed. This is also consistent with (not contradictory to) the earlier litigation section of this patent, which already identified IPR2016-00738 as "filed (Settlement)."

Proceedings overview

Total AIA trial proceedings on US 9,050,302: exactly one — IPR2016-00738 (Ranbaxy), status: settled/terminated in the preliminary-proceeding stage before any institution decision. Breakdown: 0 active, 0 claims invalidated, 0 claims sustained on the merits, 1 settled (pre-institution), 0 institution denials on the merits. Bottom line for a defendant: the '302 patent has never been substantively tested at the PTAB — no Final Written Decision, no canceled claims, no Federal Circuit appeal. It is untested, not "hardened." An IPR/§ 103 obviousness defense built on the GHB + valproate prior art remains fully available, and the single prior challenge left every claim of the patent intact.

IPR2016-00738 — Ranbaxy Inc. v. Jazz Pharmaceuticals, Inc. and Jazz Pharmaceuticals Ireland Ltd.

• Type: Inter Partes Review (post-AIA, 35 U.S.C. §§ 311–319)
• Filed: 2016-03-10 (Petition accorded a filing date 2016-03-18)
• Status: "Judgment – Termination of the Proceeding" (verbatim docket status; some commercial databases compress this to "Institution Denied" — see caveat below). Plain English: the parties settled and the Board killed the case before deciding whether to institute trial.
• Judge panel: Administrative Patent Judges Erica A. Franklin (writing), Jacqueline Wright Bonilla, and Zhenyu Yang. Tech Center 1600.
• Petition grounds: Petitioner requested cancellation of claims 1–31 under 35 U.S.C. § 103 (obviousness), supported by the Declaration of Dr. David Rotella (Ex. 1002). The theory: the challenged claims recite methods of treating cataplexy/EDS in narcolepsy by administering a reduced dose of GHB to patients also receiving valproate (divalproex sodium); the asserted "unexpected discovery" that valproate increases GHB's effect was allegedly not new, because the art already taught that valproate inhibits GHB dehydrogenase (GHB-DH) — the primary GHB-metabolizing enzyme — thereby raising blood/brain GHB levels and making dose reduction an obvious safety step. Representative references: Maitre (1997), Okun (2001), Waszkielewicz (2004), Broughton (1979/1980), Cash, Mamelak, Scharf, Bernasconi (1992), Hechler (1997, valproate/GABA-B), Kaufman, Knerr, Löscher, Vayer, Weiss, Cagnin, Morris, Bhattacharya, plus the 2007 Xyrem PI and titration schedule.
• Institution decision: None issued. The Board's Judgment states verbatim: "This case is in the preliminary proceeding stage; a decision whether to institute trial has not been made." So there is no § 314(a) institution analysis, no claim-by-claim institution ruling, and no panel reasoning on the merits to cite. Anyone telling you the Board "denied institution on the merits" is misreading a database field.
• Final Written Decision: None. No FWD issued; no claim was canceled and no claim was sustained on the merits, because the case terminated before institution. Claim-level outcomes for claims 1–31 are all untested.
• Settlement / termination: On 2016-05-12 the parties filed a joint motion to terminate under 35 U.S.C. § 317(a), plus a true copy of their written settlement agreement (Ex. 2002) and a joint request under § 317(b) / 37 C.F.R. § 42.74(c) that the agreement be treated as business confidential and kept separate from the patent's file. The Judgment (2016-05-23) granted both requests and ordered the proceeding terminated. The parties represented that the settlement also resolved the related D.N.J. litigation (Jazz v. Amneal, 2:13-cv-00391 / 13-391-ES-JAD) via a stipulation and order of dismissal. Terms are confidential.
  • Commercial context (from Jazz's own SEC filings): Ranbaxy's settlement was entered ~2016-05-09; Jazz granted Ranbaxy a license to market generic Xyrem on or after 2025-12-31, or earlier on certain triggering events. Petitioner counsel: Knobbe, Martens, Olson & Bear (Reisman, Pitzel Cruz, Taylor). Patent-owner counsel: Quinn Emanuel (Cerrito, Shih, Calvosa).
  • Postscript: Board Notice of Refund 2016-05-26; Ranbaxy refund request 2016-05-24.
• Appeal: None. With no FWD and no institution, there was nothing appealable; no CAFC docket exists for this proceeding. (The later, unrelated Jazz v. Avadel, Fed. Cir. 24-2274 / 24-2278, decided 2025-05-06, involves different Jazz patents and is not an appeal of a '302 proceeding.)
• Defensive value: A defendant today gets essentially a clean slate on the '302 patent — IPR2016-00738 creates no estoppel (no FWD, so § 315(e)(2) never attached to Ranbaxy/Sun or anyone else) and no adverse precedent. The corollary risk for defendant is the mirror image: the patent is unnarrowed and un-air-tested, so a plaintiff can assert all 31 claims with no IPR-eroded scope — but the same fact means a well-built § 103 petition (GHB + valproate, GHB-DH inhibition, routine titration) has never been rebutted on the merits by this Patent Owner.

Caveat on the "Institution Denied" label. IPVerse, Docket Alarm, and some aggregators list the status as "Institution Denied" and give an "Institution Decision Date" of 2016-05-23. The Board's own Judgment shows that date is the termination date and that no institution decision was ever made. Treat "Institution Denied" here as a database artifact, not a merits holding. (PTAB E2E; IPVerse case page; Judgment – Termination PDF.)

Strategic summary

1. Claim status: 100% UNTESTED. No claim of US 9,050,302 has been canceled, confirmed, or construed by the PTAB. The petition attacked claims 1–31, but termination occurred before institution, so the entire claim set — independent claims 1, 8, 13, 20, and 27 and all dependents — remains as issued. There is no "surviving-claims" list to give because no claim was ever adjudicated. (I did not independently confirm in this run whether the issued patent contains exactly 31 claims; "claims 1–31" is the scope the petitioner itself identified. Verify the face of the patent before relying on a total claim count.)

2. Estoppel landscape — nothing binds anyone. IPR estoppel under § 315(e)(2) (and § 315(e)(1)) is triggered only by a Final Written Decision. IPR2016-00738 produced no FWD — it terminated in the preliminary stage — so no IPR estoppel attaches to Ranbaxy, its privies (e.g., Sun Pharmaceutical), or any other party. Also note this is a pre-2016-00738-asset: it predates SAS, so even if it had instituted, claim-by-claim estoppel contours would differ. Practically, for a defendant now being asserted against: every prior-art ground is still on the table — § 102/§ 103 art against claims 1–31, including the Rotella/Maitre/Hechler-type GHB + valproate references, § 112 written-description/enablement attacks on the "reduce the dose" genus, and validity challenges under §§ 101/102/103 in district court. Nothing in IPR2016-00738 was decided, so nothing is foreclosed.

3. Pattern signals. Ranbaxy was a repeat PTAB challenger against the same Jazz "306 family" — it also filed IPR2016-00024 on sibling U.S. 8,772,306, and Jazz's filings confirm both Ranbaxy IPRs were terminated together as part of one settlement. The parallel family petitions — IPR2016-00002 (Par Pharmaceutical) and IPR2016-00546 (Amneal) also targeted the '306 patent, not the '302 patent — and, per Jazz's 10-Q, the Board denied Par's petition in its entirety (April 2016) and denied Amneal's in its entirety (July 2016), while instituting Ranbaxy's '306 petition on 16 of 34 claims (also later terminated on settlement). No defensive aggregator (e.g., Unified Patents) appears in the chain — this was generic-ANDA-filer-driven litigation. Patent Owner appeal behavior is aggressive in general (Jazz has litigated Xyrem patents to the Federal Circuit repeatedly), but on the '302 patent specifically there is no appeal history because none was possible.

4. One wildcard worth noting (allegation, not adjudicated fact). A later complaint in the Xyrem litigation (e.g., the Avadel/C.A. matter) alleges that during the '306-family IPRs Jazz withheld its 2012 Xyrem label — which, the complaint says, undercut Jazz's "teaching away" argument on CNS-depressant co-administration — and that the PTAB's institution denials on the '306 patent relied on that teaching-away theory, raising inequitable-conduct / unenforceability exposure. Because the '302 and '306 patents share the same spec and the same disclosure conduct, a defendant could explore an unenforceability theory against the '302 patent. Treat this strictly as a pleading allegation — it has not been adjudicated, and it is pleaded against siblings/other parties, not against the '302 patent in an IPR.

Recommended next steps

• You are a defendant being asserted the '302 patent: the PTAB gave you a blank canvas, not a shield. There is no FWD to cite and no claim to point to as canceled. Do not tell a court or adversary that "claims 1–31 were canceled" or that "the PTAB denied institution" — both are false. If you want a claim canceled, you must file your own IPR; nothing from IPR2016-00738 helps or hurts.
• Copy the Ranbaxy playbook, but improve on it. Ranbaxy's petition theory (valproate inhibits GHB-DH → raised GHB levels → dose reduction obvious) is a tested and coherent § 103 frame, but it was never adjudicated, so it carries no precedential weight for or against you. A fresh petition can reuse the Rotella-style art while adding art Ranbaxy did not cite. A petition today must also clear § 315(b) (one year from service of an infringement complaint) and the § 325(d) / Fintiv-style discretionary considerations.
• Consider the enforceability theory in parallel. If, on investigation, the 2012-label-withholding allegation has teeth as applied to the '302 patent's prosecution and the '306 IPRs, an inequitable-conduct defense may be more valuable than a merits IPR — but it is fact- and intent-intensive and currently unproven.
• Timing / milestones. There are no live PTAB proceedings, so there is no institution-deadline, oral-hearing, or 1-year FWD clock running on the '302 patent. Any new IPR you file would run the standard statutory timeline — institution decision within ~6 months of a complete petition, FWD within 12 months of institution (35 U.S.C. § 316(a)(11)).
• Bottom line to management: the absence of a merits challenge is the signal here. A patent that has been asserted across a decade of Xyrem ANDA litigation but has never once reached a PTAB Final Written Decision leaves both sides exposed — the owner has no PTAB-blessed validity, and a challenger has no pre-cleared roadmap. The '302 patent is untested ground.

Primary sources: PTAB E2E petition file — https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1463743](/patent/1463743)/; Board Judgment (Termination) — https://www.docketalarm.com/cases/PTAB/IPR2016-00738/Inter_Partes_Review_of_U.S._Pat._9050302/docs/05-23-2016-Board/Termination___Settlement-11-Judgment___Termination_of_the_Proceeding.pdf; case index — https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2016-00738.

Nothing above is legal advice; verify the docket and the face of the patent on PTAB E2E and USPTO PatentCenter before relying on any row.

I'll attempt to pull the recorded assignment details (reel/frame and correspondent) before reconstructing the chain, since the Google Patents legal-events feed gives me the events but not those fields.

I found the key reel/frame data. Let me try to retrieve the recorded correspondent and confirm the security-agreement filings.

I retrieved the reel/frame data from a primary-source party admission (Jazz's IPR2016-00738 Notice of Related Matters), but I hit my step limit before I could pull the recorded correspondence field. Here is the ownership reconstruction, with gaps flagged rather than filled.

Inventors

Unusual patterns: None of the fire-sale precursor type. There is one inventor, and he did not depart the assignee — Eller remained the named inventor across the whole Jazz oxybate family (Xyrem®/Xywav® method patents) for years afterward. Single-inventor, single-assignee internal filings of this kind are the opposite of the "inventors leave within 12 months → portfolio sale" pattern: the inventor's rights were captured by the employer on day one and never re-entered the market.

Original assignee

• Named on the issued patent: Jazz Pharmaceuticals Ireland Limited, Dublin, Ireland (per the printed face of the patent and the Orange Book listing).
• Original assignee at filing / first-recorded owner: Jazz Pharmaceuticals, Inc., 3180 Porter Drive, Palo Alto, CA (a Delaware corporation).
• Primary line of business: Specialty pharmaceutical company; commercialization of the narcolepsy drugs Xyrem® (sodium oxybate) and Xywav® (mixed oxybate salts).
• Did it ship a product embodying the claims? Yes. The '302 patent is listed in the Orange Book against Xyrem® and Xywav® (use code U-1532), and the specification's working examples are the Xyrem®/divalproex and Xyrem®/diclofenac human interaction studies.
• Status: Operating. Jazz Pharmaceuticals plc (Nasdaq: JAZZ) remains an active public company; the Irish subsidiary holds the family and is the plaintiff in the ongoing ANDA litigation.

Assignment timeline

The Assignment Center/PEDS record does contain entries for this patent. The two ownership transfers are recited verbatim (with reel/frame) in Jazz's own PTAB filing; the security-interest entries appear in the Google Patents legal-events feed but I could not retrieve their reel/frame or the recorded correspondent within my research budget — flagged below.

1. Executed c. 2013 (recorded 2013-05-24) — Reel 30652 / Frame 0164

   • Conveyance: Assignment of Assignors' Interest
   • Assignor: Mark Eller (sole inventor)
   • Assignee: Jazz Pharmaceuticals, Inc., Palo Alto, CA
   • Correspondent: Not retrieved (see signal 3 — I could not pull the correspondence field for any link in this chain).
   • Context: Initial capture of inventor rights by the employer — standard operating-company acquisition, not a transfer to an asserter.
   • Discrepancy flag: Google Patents' legal-events feed dates this assignment 2013-06-20, whereas the PTAB Notice of Related Matters states it was recorded 2013-05-24 at Reel 30652/0164. I could not reconcile the ~4-week gap (possibly an execution-vs-recordation difference or a second confirmatory filing). Reel/frame is from the party admission and is the more reliable figure.

2. Executed c. 2014 (recorded 2014-07-29) — Reel 33480 / Frame 0135

   • Conveyance: Assignment of Assignors' Interest
   • Assignor: Jazz Pharmaceuticals, Inc.
   • Assignee: Jazz Pharmaceuticals Ireland Limited, Dublin, Ireland
   • Correspondent: Not retrieved.
   • Context: Internal corporate reorganization / IP-holding migration within the Jazz group. Jazz Pharmaceuticals, Inc. retained only an exclusive license going forward (Jazz's IPR filing: "Jazz is currently the exclusive licensee"). This is an intra-group title move, not an arm's-length sale.
   • Discrepancy flag: Google Patents dates this event 2014-08-06; the PTAB filing says recorded 2014-07-29. Same unresolved date gap as above.

3. 2015-06-19 — Reel/Frame not retrieved

   • Conveyance: Security Agreement (grant of collateral, not a title transfer)
   • Assignors/Chargors: Jazz Pharmaceuticals International III Limited; Jazz Pharmaceuticals International Limited; Jazz Pharmaceuticals Ireland Limited; Jazz Pharmaceuticals, Inc.
   • Assignee/Secured Party: Bank of America, N.A., as Collateral Agent
   • Correspondent: Not retrieved.
   • Context: Securitization — the '302 patent (and sibling family members) pledged as part of the Jazz group's credit facility collateral package. No change in beneficial ownership.

4. 2021-05-05 — Reel/Frame not retrieved

   • Conveyance: Security Agreement
   • Assignors/Chargors: Cavion, Inc.; Celator Pharmaceuticals, Inc.; Jazz Pharmaceuticals Ireland Limited; Jazz Pharmaceuticals, Inc.
   • Assignee/Secured Party: U.S. Bank National Association (later U.S. Bank Trust Company, N.A., as collateral trustee)
   • Correspondent: Not retrieved.
   • Context: Securitization — re-pledge of the portfolio under the May 5, 2021 Jazz credit agreement (collateral trustee U.S. Bank), a refinancing that also secured the GW Pharmaceuticals acquisition financing. Title remains with Jazz Ireland.
   • SEC corroboration: Jazz's 10-K/8-K exhibits describe the "Credit Agreement, dated as of May 5, 2021 … Bank of America, N.A., as administrative agent, and U.S. Bank National Association, as collateral trustee," and list the chargors including Jazz Pharmaceuticals Ireland Limited. (Jazz investor-relations static filings, e.g. https://investor.jazzpharma.com/static-files/a8427a25-7532-490f-b59a-1d57c21099cd.)

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

   • Conveyance: Release by Secured Party
   • Assignor (releasing party): Bank of America, N.A.
   • Assignee (released parties): Jazz Pharmaceuticals Ireland Limited; Jazz Pharmaceuticals International Limited; Jazz Pharmaceuticals International III Limited; Celator Pharmaceuticals, Inc.; Cavion, Inc.; Jazz Pharmaceuticals, Inc.
   • Correspondent: Not retrieved.
   • Context: Securitization housekeeping — release of the 2015 Bank of America collateral liens upon the 2021 refinancing. Not a transfer of ownership.

Net ownership position: Jazz Pharmaceuticals Ireland Limited holds legal title today. Every recorded event after the 2013 inventor capture is either (a) an intra-Jazz title migration (2014) or (b) a lien/release (2015, 2021). No assignment has ever moved this patent outside the Jazz corporate family, and no ownership assignment is recorded after 2014.

Timeline diagram

timeline
    title Ownership of US 9050302
    2013 : Filed by Jazz Pharmaceuticals Inc
         : Eller assigns rights to Jazz Inc
    2014 : Jazz Inc moves title to Jazz Ireland
    2015 : Patent issues to Jazz Ireland
         : Bank of America security interest
    2021 : Bank of America lien released
         : US Bank security interest granted

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The chain's terminal owner is Jazz Pharmaceuticals Ireland Limited, an operating pharmaceutical subsidiary of a Nasdaq-listed company (Jazz Pharmaceuticals plc). No "IP / Licensing / Holdings / Ventures" LLC appears, no registered-agent service address appears, and there is no single-purpose Delaware/Texas shell. The 2014 transfer (Reel 33480/0135) is a name-adjacent group entity, not an anonymous assignee.

2. Known asserter in the chain — Not present. No current or prior assignee matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Round Rock, etc. Jazz is the plaintiff suing generic ANDA filers (e.g., Jazz Pharmaceuticals Ireland Ltd. v. Teva, D.N.J. 2:23-cv-01617; v. Lupin, D.N.J. 2:21-cv-14271), i.e., an operating company asserting against actual competitors — the inverse of the NPE pattern.

3. Repeat correspondent across the chain — Unclear / not established. I was unable to retrieve the recorded correspondent field for any of the five entries above, so I cannot make a recurrence finding either way. Adjacent (non-assignment) counsel context, offered only as context and not as an assignment-correspondent finding: prosecution of the family was handled by Jones Day; the PTAB respondent counsel for Jazz was Francis Cerrito; Jazz's litigation counsel in the Xyrem ANDA suits has included Charles M. Lizza / Richard G. Greco (Saul Ewing); and a Jazz terminal disclaimer in a sibling application was signed by David D'Zurilla. None of these is a corroborated assignment-record correspondent, so I am not treating any of them as a signal.

4. Cascading transfers — Not present. There are exactly two ownership links, both in 2013–2014 and both intra-Jazz. No chained LLC sequence, no shared correspondent address, no common-principal shells.

5. Pre-litigation transfer — Not present. The title move to Jazz Ireland was recorded 2014-07-29; the first suits naming the '302 patent were filed September 2015 (2:15-cv-06562 Amneal; 2:15-cv-06548 Lupin), roughly 13 months later and outside the 6-month window. More importantly, the transfer ran to the group's principal operating IP holder, not to an assertion vehicle, so the timing carries no arranged-assertion meaning.

6. Bankruptcy fire-sale — Not present. No Chapter 7/11, no distress sale. The only secured-lien event (Bank of America, 2015) was a routine credit-facility pledge that was released in 2021 as part of a refinancing, not foreclosed.

7. Privateering — Not present. No operating company transferred the patent to an NPE to assert on its behalf. Jazz itself is the asserting plaintiff across the family.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

Verdict

Operating-company assertion.

The ownership chain is a two-step, wholly intra-corporate sequence — inventor Mark Eller → Jazz Pharmaceuticals, Inc. (Reel 30652/0164, recorded 2013-05-24) and Jazz Pharmaceuticals, Inc. → Jazz Pharmaceuticals Ireland Limited (Reel 33480/0135, recorded 2014-07-29) — after which the only recorded events are a Bank of America security interest (2015-06-19), a U.S. Bank security interest (2021-05-05), and a Bank of America release (2021-05-05), none of which transfers title. The current owner is an operating pharmaceutical company whose Xyrem®/Xywav® products practice the claims and which asserts the patent against generic ANDA filers; zero NPE signals are present.

Verify at: USPTO Assignment Center search for patent 9,050,302 — https://assignmentcenter.uspto.gov/ (mirror: https://assignment.uspto.gov/patent/index.html). The reel/frame citations above are also recorded in Jazz's own PTAB filing, IPR2016-00738, Notice of Related Matters (Mar. 31, 2016), https://www.docketalarm.com/cases/PTAB/IPR2016-00738/ — the authoritative confirmation that only these two assignments exist and that title sits with Jazz Ireland.

Two open items I could not close: (a) the recorded correspondent for each entry (so signal 3 is genuinely unresolved, not negative), and (b) the reel/frame numbers for the three security/release filings, which are not in the Google legal-events feed and which I ran out of budget to pull from Assignment Center directly.

I'll research this systematically — first confirming the '302 record, then pulling its actual cited prior-art references.

Let me try to retrieve the actual front-page "References Cited" list and the cited-by data for the '302 patent.

I have the IPR record and part of the front page. Now let me pull the actual "U.S. Patent Documents" cited on the face of the '302 patent.

I hit the search-step limit before I could pull the complete OCR of the printed front page. Below I've built the analysis from (a) the authoritative full text you supplied, (b) the verified front‑page fragment from the '302 exhibit in IPR2016‑00738, and (c) the Ranbaxy IPR exhibit list. I flag every gap rather than filling it speculatively.

0. Record check on the number itself

Search results confirm — and do not contradict — that US 9,050,302 B2, "Method of administration of gamma hydroxybutyrate with monocarboxylate transporters," Mark Eller, Jazz Pharmaceuticals Ireland Ltd., application 13/837,714, filed Mar 15, 2013, granted Jun 9, 2015, is Orange‑Book listed against XYREM/XYWAV under use code U‑1532, with a pediatric‑exclusivity expiry of ~Sep 15, 2033. No conflicting "similar‑number" patent surfaced, so I have treated the '302 as the sole subject throughout. (This is consistent with the previously generated summary; no contradiction to flag there.)

One contradiction to flag up front: the earlier "Patent summary" states the granted independent claims are 1, 8, 13, and 20. The Ranbaxy IPR2016‑00738 petition, however, identifies the "alleged advance" with reference to "Claims 1, 8, 13, 20, and 27." I could not resolve whether claim 27 is a fifth independent or a dependent; treat the independent‑claim set as {1, 8, 13, 20} with claim 27 in doubt.

AIA point relevant to §102: the application was filed March 15, 2013 — one day before the AIA first‑inventor‑to‑file provisions took effect (Mar 16, 2013) — and claims provisionals from March 1/March 12, 2013. Pre‑AIA §102 therefore governs (§102(a)/(b)/(e)/(g)), and the §102(b) one‑year bar date is on or about March 1–15, 2012.

1. Method / scope limitation

The task asks me to "look at each patent citation for 9050302." Two important framing points before the tables:

1. I could not retrieve the complete "(56) References Cited — U.S. Patent Documents" block from the printed '302 front page within my search budget. The IPR exhibit OCR I obtained (Ex. 1001) captured only portions of the "Other Publications" subsection. The U.S.‑patent‑document list below is reconstructed from the specification's express incorporation‑by‑reference (which in Jazz's practice mirrors the face‑of‑patent citations) and is marked accordingly. Rows I could not positively confirm as face‑of‑patent citations are flagged [unverified as face citation].
2. §102 vs. §103: virtually every reference the '302 patent cites is background/context art (GHB formulations, GHB distribution/REMS systems, GHB pharmacology). None is a single reference disclosing the combination + dose‑adjustment that the granted claims require. That is consistent with why the only PTAB challenge (Ranbaxy) pleaded §103 obviousness, not §102 anticipation. I say this plainly rather than manufacturing anticipation mappings.

2. U.S. patent documents cited / incorporated by reference

Bottom line for the patent documents: none of items 1–27, alone, discloses all of (i) cataplexy/EDS‑in‑narcolepsy treatment + (ii) a patient already on GHB + (iii) concomitant divalproex sodium + (iv) a ≥20% reduction of the GHB daily dose + (v) from a 4.5–9 g/day baseline. No §102 anticipation of granted claims 1, 8, 13, or 20 (or 27) is supported by any cited patent document. The distribution patents (items 6–11, 14–22) are the only ones that touch the spec's warning/monitoring embodiments, and those embodiments are not in the granted claim set.

3. Non‑patent literature on the face of the patent (the genuine §102 candidates)

These are from the verified "Other Publications" fragment of the '302 front page (Ranbaxy Ex. 1001) and thus are confirmed face‑of‑patent citations. All predate Mar 2012 unless noted, so they are §102(b) art unless flagged otherwise.

(Also appearing on the front page per the OCR: PCT International Search Report/Written Opinion mailed Jun 24, 2014 in PCT/US2014/019217; USPTO Notice of Allowance mailed Nov 3, 2014 in 13/873,000; §1.84/§1.97 and paragraph‑IV notification letters (Par, Ranbaxy, Watson). These are prosecution/notification documents, not prior art, and cannot anticipate.)

4. Additional art from the Ranbaxy IPR (the operative challenge set)

Ranbaxy's exhibits (IPR2016‑00738) are the most probative prior‑art collection actually deployed against the '302 claims. Notably they were combined under §103, not §102:

Maitre 1997 (Ex. 1003); Okun 2001 (1004); Xyrem PI (1005); Xyrem‑TS 2008 (1006); Waszkielewicz 2004 (1007); Broughton 1979/1980 (1008‑1009); Cash 1994 (1010); Mamelak 1986 (1011); Scharf 1985 (1012); Scharf 1998 (1013); Bernasconi 1992 (1014); Hechler 1997 (1015); Kaufman (1016‑1017); Knerr (1018); Loscher (1019‑1020); Vayer (1021‑1022, and 1038 = Vayer 1988); Weiss (1023); Cagnin (1024); Morris I (AAPS, 1025) and Morris II (2011, 1026); Bhattacharya (1027); Depakene PI (1029); FDA Draft Guidance — Drug Interaction Studies (1030); Lamictal PI (1031); Havelaar (1039); Chateauvieux (1040).

The Ranbaxy declaration's theory was expressly obviousness: valproate's elevation of GHB was known (Hechler, Snead), GHB dosage was known to be titratable (Xyrem PI/TS), so reducing the GHB dose for a valproate‑co‑treated patient would have been obvious. That framing is itself evidence that no single reference was believed to anticipate the claims.

5. §102 conclusion (claim‑by‑claim)

Net: the '302 patent's own citations are background art, not anticipatory art. The closest any cited reference comes to §102 is the GHB‑dosing art for the baseline element (Xyrem PI/PDR/Titration Schedule) and the valproate‑GHB‑metabolism art for the interaction element (Hechler 1997; Snead 1980) — and neither alone bridges the gap. This is corroborated by the fact that the sole PTAB challenge was pleaded as §103 obviousness and was resolved by settlement (joint motion May 12, 2016; termination entered May 23, 2016) with the settlement agreement kept confidential under 35 U.S.C. §317(b).

6. Explicit information gaps (not asserted as negatives)

• Complete face‑of‑patent "(56) U.S. Patent Documents" list: I could not OCR the full block. Items 1–13 and 23–24 are strongly supported by the specification's incorporation‑by‑reference; items 14–22 (pharmacy‑management patents) are cited in the specification and are likely also face citations, but I could not confirm each one appears on the printed front page. Verify against the printed patent (Ex. 1001) or the USPTO PatentCenter "References Cited" tab.
• Inventor names / exact issue dates for U.S. 6,112,182, 6,317,719, and 6,356,873 were not verified within my budget.
• Claim 27's status (independent vs. dependent) is unresolved on the sources retrieved.
• These are gaps in my retrieval, not findings that the references don't exist.

Sources relied on: the authoritative full text you supplied (https://patents.google.com/patent/US9050302/en); Ranbaxy Ex. 1001 front page (US 9,050,302 printed patent) via the PTAB exhibit mirror (https://gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2016‑00738/…/Ex_1001___US__Patent_No_9,050,302/); IPR2016‑00738 petition and exhibit list (https://ptacts.uspto.gov/ptacts/public‑informations/petitions/[1463743](/patent/1463743)/download‑documents…; https://ipverse.greyb.com/ptab‑web/cases/case‑details/IPR2016‑00738); IPR2016‑00024 Ex. 1025 (U.S. 9,050,302) at https://gaeflexstaging‑dot‑docketupdate.appspot.com/cases/PTAB/IPR2016‑00024/…/Exhibit‑1025‑US_Patent_No_9050302/.

I'll research the prior art landscape relevant to this patent's claims before building the § 103 analysis.

Let me pull the specific IPR petition against the '302 patent itself and prior art on GHB transport/pharmacokinetics.

Obviousness Analysis — U.S. Patent No. 9,050,302 under 35 U.S.C. § 103

Scope note / cross-reference: This analysis builds directly on the two prior sections. It does not repeat the bibliographic or litigation summaries. Where the earlier "Patent summary" flagged an unresolved question — whether the "first and second dose each less than 2.25 g" language belongs to claim 31 of the '302 patent or to the sibling '426 patent — the Ranbaxy IPR2016-00738 petition located below resolves it in favor of the '302 patent's dependent claims (claims 16, 17, 23, 24, 29, 30 per the petition's numbering), while the DrugPatentWatch attribution to "claim 31 of 9050302" may be a record mix-up with the '426 patent. Flagged, not silently corrected.

1. Governing framework and temporal law

The '302 patent's earliest effective filing date is March 1, 2013 (provisional 61/771,557); the non-provisional was filed March 15, 2013 — one day before the AIA first-inventor-to-file provisions took effect (March 16, 2013). Pre-AIA § 103(a) therefore governs. Prior art available includes § 102(a)/(b)/(e) art predating the March 1, 2013 critical date, with the one-year § 102(b) bar reaching anything public before March 1, 2012.

Graham v. John Deere, 383 U.S. 1 (1966), controls: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) secondary considerations. Under KSR Int'l v. Teleflex, 550 U.S. 398 (2007), a claim is obvious where a POSA would have (a) combined prior-art elements per known methods or (b) pursued a known, finite set of identified, predictable solutions with a reasonable expectation of success.

Level of ordinary skill (POSA): a person holding a Pharm.D., Ph.D. (pharmacology, pharmaceutics, pharmacokinetics), or M.D. (sleep medicine/neurology) with several years' experience in CNS-drug pharmacology and clinical pharmacokinetics — a skill level the Ranbaxy petitioners and Jazz appear to have accepted (IPR2016-00738 Declaration of John R. Horn, Pharm.D., FCCP).

2. The claims at issue — what actually has to be proven

Per the Ranbaxy IPR2016-00738 petition (filed March 10, 2016, against the '302 patent), the granted '302 patent has independent claims 1, 8, 13, and 20, each requiring:

1. Treatment of cataplexy in narcolepsy or excessive daytime sleepiness (EDS) in narcolepsy;
2. In a patient currently taking GHB or a salt thereof;
3. Concomitant administration of divalproex sodium (valproate);
4. Reducing the daily GHB dose by ≥ 20%, relative to a baseline of 4.5 g to 9 g/day used absent valproate.

Dependent claims add monitoring/titration, specific baselines (4.5 g, 6 g, 7.5 g, 9 g), specific reduced amounts, and first/second doses each < 2.25 g.

Critical contrast with the sibling '306 patent: the '302 claims recite a ≥ 20% reduction, not the '306's "at least 5%." That difference drives everything below. It makes the claims narrower (harder to invalidate on the elements directly read from prior art) but also places the 20% figure squarely inside the numerical-range/routine-optimization doctrine of In re Aller, 220 F.2d 454 (CCPA 1955), and In re Boesch, 617 F.2d 272 (CCPA 1980).

Source: Ranbaxy IPR2016-00738 Petition, https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1463743](/patent/1463743)/ ; https://www.docketalarm.com/cases/[PTAB](/ptab)/IPR2016-00738/

3. Prior-art references in the record (all pre-date March 1, 2013)

The Ranbaxy petition's exhibit list confirming these references appears at https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1463786](/patent/1463786)/ ; the '302-specific petition is directed to "Ground 1: Waszkielewicz + Xyrem® PI" and "Ground 2: Maitre + Xyrem® PI" for claims 1–31.

4. Claim-element mapping — independent claims 1, 8, 13, 20

The arithmetic matters. A POSA following the Xyrem® PI's own titration ladder reaches the ≥ 20% threshold with a single downward step whenever the baseline is ≤ 7.5 g (7.5→6 g = 20.0%; 6→4.5 g = 25.0%), and with two steps from the 9 g maximum (9→6 g = 33.3%). The claim therefore covers precisely the dose points the label already discloses as available dose levels.

5. The obviousness grounds

Ground 1 — Maitre (or Hechler/Waszkielewicz) in view of the Xyrem® PI

Maitre teaches that valproate increases brain GHB by inhibiting GHB dehydrogenase. Hechler independently confirms valproate inhibits GHB dehydrogenase. The Xyrem® PI teaches (i) GHB's identity/dosing, (ii) that excess GHB causes oversedation, coma and death, and (iii) that the dose should be titrated in fixed 1.5 g decrements and halved when elimination is compromised. Combining: a POSA seeking to safely co-administer GHB and valproate would (a) recognize the elevated-GHB risk and (b) reduce the GHB dose using the label's own decrement. Any downward step from ≤ 7.5 g yields ≥ 20%.

Ground 2 — Cagnin (or Weiss) in view of the Xyrem® PI

Cagnin discloses an actual clinical adverse interaction (seizures) in a patient on GHB + valproate, and explains the mechanism (SSADH inhibition → up to 30-fold ↑ GHB). Weiss discloses the same mechanism for a mechanistically identical enzyme-inhibiting AED, with a measured 2.8-fold plasma GHB increase, and explicitly calls for dose-management in such combinations. These are the strongest "known problem → known solution" references: a POSA confronted with a documented, mechanistic GHB–valproate interaction would reduce the GHB dose.

Ground 3 — FDA DDI Guidance (2012) + Xyrem® PI + Depakote label

The 2012 FDA Guidance establishes that when a concomitant drug changes exposure to a substrate, the response is to characterize the change and adjust dosing/labeling accordingly. Applied to GHB (substrate) + valproate (inhibitor), this supplies an explicit, regulatory-grade motivation to adjust the GHB dose. The Depakote label supplies valproate's PK/DDI profile.

Ground 4 — The Xyrem® PI's own internal evidence, for the 20% figure specifically

The '302 specification itself reports that valproate raised GHB systemic exposure (plasma AUC) by 26%. A POSA reading the specification's own data would see that a ~20–26% dose reduction is the natural arithmetic compensation for the reported 26% exposure increase. This converts the "at least 20%" limitation from a non-obvious design choice into a result-effective, predictable numeric adjustment — the classic In re Aller/In re Boesch situation. (Corroborated by JAZZ's own later Xyrem label, which states that GHB 6 g/day + valproate 1250 mg/day "resulted in an increase in systemic exposure to sodium oxybate of approximately 25%" and that "the dose should be adjusted accordingly" — https://fass.se/health/product/20040607043715/fass-text.)

Ground 5 (specification-enabled but outside the granted independent claims) — the MCT-specific subject matter

The '302 specification, and its sibling patents, additionally disclose diclofenac/ibuprofen interactions, dose increases, monitoring, REMS distribution, and concentration/pH limits. To the extent any claim or amendment in this family reaches MCT inhibition generally, the Morris/Wang/Morse line of references (2005–2012) is directly on point: GHB is an MCT1/2/4 substrate (Wang & Morris 2007), renal reabsorption is MCT-mediated (Wang 2006), and MCT inhibition increases GHB renal clearance — expressly framed as a therapeutic "detoxification strategy" (Morris 2005). Diclofenac, ibuprofen and valproate are all monocarboxylate-containing acidic drugs recognized in the art as MCT-interacting; the '302 specification itself states "Valproate, diclofenac, and ibuprofen are monocarboxylate transporter inhibitors." That admission, read against Halestrap & Meredith (2004), supplies both the motivation and the expectation for the MCT-based embodiments.

6. Motivation to combine (why a POSA would have done this)

1. Same field, same problem. Both the Xyrem® PI and Maitre/Cagnin/Weiss concern GHB's safety when co-administered with other CNS drugs — no field-crossing leap (KSR prong (a)).
2. Known problem, known solution. Documented valproate-driven GHB elevation → the label's own instruction to reduce GHB in "compromised elimination." The solution is explicitly telegraphed.
3. Predictable variation. Reducing a dose in the label's own 1.5 g decrements is a mechanical, predictable variation; the 4.5/6.0/7.5/9.0 g grid is finite and identified (KSR prong (b)).
4. Regulatory pull. The 2012 FDA DDI Guidance created a concrete, contemporaneous incentive to manage exposure-altering interactions by dose adjustment.
5. Economic/practical pull. Valproate is widely prescribed (epilepsy, bipolar disorder); a POSA treating a narcolepsy patient who is also on valproate would have a strong reason to keep the patient on GHB by dose-adjusting rather than discontinuing either drug.

7. Reasonable expectation of success — and the strongest counterarguments

For obviousness. GHB is an endogenous compound whose dose–effect relationship was well characterized and titratable. The exposure increase (~25–26%) is of the same order as the claimed 20% reduction, so a POSA could reasonably expect the reduced dose to restore near-baseline exposure. The Weiss reference expressly identifies GHB dehydrogenase as GHB's main elimination route inhibited by valproate, giving a mechanistic basis for the expectation.

Against obviousness — and this is a genuine, on-the-record problem for the invalidity case. In the sibling '306 IPRs, the PTAB denied institution on claims 1–18 (Ranbaxy IPR2016-00024; Par IPR2016-00002, decisions April 12, 2016), crediting Jazz's evidence that:

• The prior art taught away from GHB + valproate co-administration (contraindication/caution language in the Xyrem label and adverse event reports such as Cagnin);
• A POSA would not have had a reasonable expectation that reduced GHB doses would still treat the claimed sleep disorders without side effects, because GHB is eliminated through alternate pathways not inhibited by valproate (a point Amneal emphasized with its "Inhibitory Effect of Valproate on Metabolic Pathways for GHB" figure);
• Ranbaxy's "routine optimization" theory was insufficient given evidence of unpredictability in the alternate clearance pathways.

See https://natlawreview.com/node/61672/ ; https://natlawreview.com/node/61870/ ; Jazz 10-K disclosure at http://investor.jazzpharma.com/static-files/3ca0692c-83fe-4378-b77d-d616278faa60.

How much does that carry to the '302? Two important caveats:

• The PTAB never adjudicated the '302 patent. Ranbaxy's IPR2016-00738 (the '302 petition) was terminated by settlement on May 23, 2016, before institution. The '302's ≥20% claims were never given a merits ruling. The '306 denial is therefore persuasive but not a holding as to these claims.
• The '306 denial rested on the broader "≥ 5%" claims. The '302's ≥ 20% figure is measurably closer to the specification's own 26% exposure change. If anything, the narrower numeric limit makes the result more predictable — a point that cuts toward obviousness under In re Aller, even as the narrower scope cuts toward validity under a literal-infringement analysis. The two effects must not be conflated.

The single strongest non-obviousness argument for the '302 is the teaching-away / unpredictability combination the PTAB accepted — but that record is (a) about a different claim scope and (b) about a different patent, and it was never applied by the PTAB to the ≥20% claims.

8. Secondary considerations (Graham prong 4)

• Unexpected results (Jazz's assertion): Jazz argued during prosecution that the direction and magnitude of the GHB–valproate interaction was unexpected. But the asserted "unexpected" result is contradicted by Maitre (1997), Hechler (1997), Snead (1980), and Cagnin (2011), which all point to valproate increasing GHB. An unexpected-result argument is weak where the prior art discloses the same direction of effect.
• Teaching away: supported by the Xyrem label's caution/contraindication language and Cagnin's adverse case report — the strongest secondary consideration Jazz can deploy, and the one the PTAB credited for the '306.
• Commercial success / long-felt need: Xyrem®/XYWAV® are commercially major, but nexus to the specific ≥20% dose-reduction step will be contested (the commercial success is largely attributable to sodium oxybate itself, long predating the '302 claims).
• Licensing/industry praise: no evidence located tying either to the claimed numeric limitation.

9. Bottom line

On the record available, independent claims 1, 8, 13, and 20 of US 9,050,302 are vulnerable to a § 103 challenge. The strongest combination is Maitre (or Hechler/Cagnin/Weiss) + the Xyrem® PI, optionally with the 2012 FDA DDI Guidance:

• The Xyrem® PI supplies every structural element except the interaction-triggered reduction — the disease (EDS/cataplexy in narcolepsy), the drug (GHB/sodium oxybate), the baseline (4.5–9 g/day), and the mechanism for adjustment (1.5 g decrements; halve if elimination compromised);
• Maitre/Hechler/Cagnin/Weiss supply the reason (valproate inhibits GHB dehydrogenase/SSADH → elevated GHB → toxicity risk);
• The arithmetic of the label's own titration ladder lands directly on the ≥20% threshold (7.5→6 g = 20%; 6→4.5 g = 25%);
• The specification's own 26% AUC figure makes a ~20% reduction a predictable, result-effective adjustment under In re Aller.

However, this is not a one-sided conclusion. The PTAB's refusal to institute on the broader '306 claims 1–18, based on a teaching-away record and evidence of unpredictable alternate GHB clearance pathways, is a real and citable obstacle — and the '302 patent's own IPR (IPR2016-00738) settled before institution, so no tribunal has ever passed on the validity of the ≥20% claims. A defendant seeking to invalidate claims 1/8/13/20 would need to (i) overcome the teaching-away evidence and (ii) rebut the "reasonable expectation of success" problem with concrete PK evidence (e.g., the Morris/Wang/Morse MCT-clearance data plus the 25–26% exposure data) showing that a 20% reduction was a predictable dose response rather than guesswork.

Confidence: Moderate on the motivation-to-combine and element-mapping; lower on the ultimate legal conclusion, precisely because the only PTAB merits rulings in this family went against obviousness on the closely-related '306 claims, and the '302 claims were never adjudicated.

Key sources

• Ranbaxy IPR2016-00738 petition & exhibits ('302): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1463743/ ; https://www.docketalarm.com/cases/PTAB/IPR2016-00738/
• Ranbaxy/Amneal IPR petitions & PTAB decisions ('306): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1463786/ ; https://www.docketalarm.com/cases/PTAB/IPR2016-00024/
• PTAB decisions commentary: https://natlawreview.com/node/61672/ ; https://natlawreview.com/node/61870/
• Hechler 1997 (PubMed): https://pubmed.ncbi.nlm.nih.gov/[9152382](/patent/9152382)/
• GHB/MCT science (Felmlee, Morse, Morris): https://pubmed.ncbi.nlm.nih.gov/29280004/ ; https://europepmc.org/articles/pmc3250051
• Current Xyrem label (25% exposure increase + dose adjustment): https://fass.se/health/product/20040607043715/fass-text
• Litigation invalidity contentions referencing Cagnin/Weiss/Shep: https://storage.courtlistener.com/recap/gov.uscourts.njd.[478299](/patent/478299)/gov.uscourts.njd.478299.244.2.pdf
• DrugPatentWatch '302 record: https://www.drugpatentwatch.com/p/patent/9050302