US 9908907

I'll search for current information on this patent number, including any 2026 CAFC docket activity.

Let me run a couple more targeted searches to check for litigation, IPR, or Federal Circuit activity tied specifically to this patent number.

US Patent 9,908,907 — Verified Summary

Search scope note: I searched generally-indexed USPTO/PTAB/Federal Circuit material and patent databases. I did not find any 2026 Court of Appeals for the Federal Circuit docket or opinion naming this patent, and I did not find an IPR/PGR petition naming it. A general web search is not a substitute for a live CAFC/PTAB docket query, so treat that negative finding as not fully authoritative. I also did not auto-correct the number; all results below are for 9,908,907 literally.

Bibliographic data (high confidence — corroborated by the full patent text and secondary databases)

Assignment/security history: security interest to Hamilton SA LLC recorded 2022-12-30; release by secured party recorded 2023-09-20 (assignors: Helsinn Birex Pharmaceuticals Ltd., Helsinn Healthcare SA, Helsinn Therapeutics (U.S.), Inc.).

Orange Book linkage: The patent is listed for AKYNZEO (fosnetupitant chloride hydrochloride / palonosetron hydrochloride), NDA 210493, with patent-use codes DS/DP and expiration 2032-05-23 (drugpatentwatch.com; pharmakb.com). No generic version is approved as of the sources reviewed.

Abstract (verbatim substance)

"Disclosed are compounds, compositions and methods for the prevention and/or treatment of diseases which are pathophysiologically mediated by the neurokinin (NK1) receptor. The compounds have the general formula (I) …"

Plain-language overview of the claims

⚠️ Sourcing caveat: the Google Patents full text supplied here is truncated before the claims. The claim set below comes from a secondary claims database (drugpatentwatch.com/p/patent-claims/9908907), which lists six claims total. I could not independently verify the exact claim wording against the granted patent document, so treat the wording as moderate confidence and the claim structure as reliable.

• Claim 1 (independent — compound): A compound of formula (VI), or a pharmaceutically acceptable salt, restricted so that (a) R2, R5 and R6 are each hydrogen; (b) m, p and s are each zero; and (c) R200 and R300 are each hydrogen or methyl. In plain terms: a narrow genus of the fosnetupitant-type (phosphonooxy)methyl piperazinium salt/zwitterion scaffold.
• Claim 2 (independent — salt form): A chloride hydrochloride salt of a compound of formula (VI). This is the specific salt form of the active moiety (fosnetupitant chloride hydrochloride).
• Claim 3 (independent — composition): A parenteral pharmaceutical composition comprising a compound of formula (I) or a pharmaceutically acceptable salt, plus one or more liquid pharmaceutical excipients.
• Claim 4 (dependent on 3): The parenteral composition of claim 3 further comprising palonosetron or a pharmaceutically acceptable salt thereof.
• Claim 5 (independent — composition): A parenteral pharmaceutical composition comprising a chloride hydrochloride salt of a compound of formula (I).
• Claim 6 (dependent on 5): The parenteral composition of claim 5 further comprising palonosetron or a pharmaceutically acceptable salt thereof.

So the patent contains three independent claims (1, 2, 3, 5 plus the salt-form claim 2), covering: (i) the compound genus, (ii) the chloride hydrochloride salt, and (iii) parenteral compositions, including palonosetron-containing combination compositions.

Technical/legal context worth noting

• The specification describes the compounds as NK1-receptor antagonists/4-phenyl-pyridine derivatives, and states a particular preference for the chloride hydrochloride salt of GA1 (fosnetupitant), described as "tremendously resistant to decoupling of the oxo-phosphonomethyl" and reversion to the parent drug.
• The specification describes a one-step, acid-free synthesis using chloromethyl dialkyl phosphate esters, and asserts stabilization via two equivalents of HCl, expressly contrasting with prior art dibasic (phosphooxy)methyl quaternary ammonium salts.
• Prior art discussed includes U.S. 6,297,375 and U.S. 6,747,026 (Hoffmann-La Roche) and U.S. 5,985,856 (University of Kansas — consistent with co-inventor Valentino J. Stella).
• The patent family includes later continuations cited on the Google Patents page: US 10,208,073 B2, US 10,717,721 B2, US 11,312,698 B2, US 12,071,421 B2. Note the recurring oddity that the Google Patents record labels these "Priority to" events; they are better characterized as continuations/divisionals in the same family.

Litigation found (district court only)

• Helsinn Healthcare S.A. v. Gland Pharma Limited, D.N.J. No. 3:22-cv-04635 (ZNQ)(LHG), filed 2022-07-18. The '907 patent was one of ten Helsinn patents asserted against Gland's ANDA No. 217374 for generic fosnetupitant chloride hydrochloride/palonosetron hydrochloride IV. Consent Judgment and Dismissal Order entered 2023-01-18 (Doc. 50), dismissing all claims with prejudice; Gland admitted the ANDA submission was a technical act of infringement under 35 U.S.C. § 271(e)(2)(A). (courtlistener.com/docket/63601016; paragraphfour.com)
• The Google Patents record also lists New Jersey District Court litigation links for case 2:22-cv-04635 in addition to 3:22-cv-04635; I did not verify whether the "2:" entry is a distinct case or a data artifact, so I flag it rather than resolve it.
• No CAFC appeal and no PTAB proceeding naming the '907 patent was found. The Azurity IPRs I located (e.g., IPR2025-00946, filed 2025-05-01) target other Helsinn patents — US 10,828,297, 9,186,357, 8,623,826 and 9,943,515 — not 9,908,907 (docketalarm.com; PTAB filings). Again, absence of a hit in web search is not conclusive proof that no such proceeding exists.

Statement of uncertainty

• Claim language: sourced from a secondary database, not the granted claims text (the provided full text truncates before the claims). Verify against the USPTO PatentCenter/original PDF before relying on exact wording.
• CAFC 2026 docket: I have no authoritative information that US 9,908,907 is before the Federal Circuit in 2026; my searches returned unrelated Federal Circuit matters.
• Expiration: 2032-05-23 is the database/Orange Book figure; term is subject to any PTA/PTE and to terminal disclaimers, which I did not verify.

I'll research litigation involving US 9908907 using the sources you specified.

Let me dig deeper into the specific case and check for other cases involving this patent.

Let me check the Unified Patents portal directly and look for any additional defendants involving this patent.

Litigation involving US Patent 9,908,907

One litigation is confirmed for US 9,908,907 specifically (an ANDA/Paragraph IV case in D.N.J.). Details below, followed by negative findings and caveats. I searched Unified Patents' litigation portal, CourtListener/RECAP, DrugPatentWatch, and IPWatchdog roundups.

Case 1 — Helsinn Healthcare S.A. v. Gland Pharma Limited

Patents-in-suit (10 patents, '907 among them): U.S. 8,426,450; 8,895,586; 9,186,357; 9,403,772; 9,908,907; 10,208,073; 10,624,911; 10,717,721; 10,828,297; and 11,312,698.

Procedural background

• Triggered by Gland's Paragraph IV notice letters dated June 2, 2022 (as to the '450, '586, '357, '772, '907, '073, '911, '721 and '297 patents) and July 11, 2022 (as to the '698 patent), stating that ANDA No. 217374 contained Paragraph IV certifications. The ANDA product is generic EQ 11.75 mg/mL fosnetupitant chloride hydrochloride and 0.0125 mg/mL palonosetron hydrochloride, 235 mg/0.25 mg per 20 mL single-dose vials for IV administration.
• Gland filed an Amended Answer, Affirmative Defenses and Counterclaims on Oct. 11, 2022 (D.I. 25), seeking declaratory judgments of non-infringement and invalidity as to each patent-in-suit, including the '907 patent. Helsinn answered the counterclaims, denying Gland's entitlement to relief (D.I. 31).

Outcome (Consent Judgment, D.I. 50)

• Gland admitted that submission of ANDA No. 217374 "was a technical act of infringement of the Helsinn Patents under 35 U.S.C. § 271(e)(2)(A)," and did not rebut the statutory presumption that the patents are valid and enforceable (expressly without prejudice to invalidity defenses).
• Gland, its officers, agents, employees, attorneys and those in active concert are enjoined from manufacturing, using, offering for sale, selling, or importing the ANDA No. 217374 products in the U.S. "until [patent expiration] or at such earlier date as agreed to by the Parties." (The text is OCR-garbled in the copy I retrieved; the operative date appears where the order text reads "... that are the subject of ANDA No. 217374 uni or at such earlier date...")
• Both parties waived all rights of appeal; the court retained jurisdiction to enforce the order.
• All claims, counterclaims and affirmative defenses were dismissed. Note a conflict in the record: the docket description reads "dismissed with prejudice," while the body of the order (¶6) says dismissed "without prejudice."
• A subsequent order (D.I. 51, July 6, 2023) granted a motion to seal, keeping D.I. 43 and the Consent Judgment under seal.

Note on the two case numbers. Unified Patents' portal lists both 2:22-cv-04635 (status "Open") and 3:22-cv-04635 (status "Closed") with the identical parties, filing date (July 17/18, 2022) and cause of action. This is the same single case appearing under two docket-number prefixes in the databases (D.N.J. recasting/reassignment); the operative judgment was entered in 3:22-cv-04635. Google Patents' "Family has litigation" link likewise lists only these two New Jersey entries for this patent.

Negative findings

• No Court of Appeals for the Federal Circuit appeal involving 9,908,907 was found. (The prominent Helsinn v. Teva appeal, 855 F.3d 1356 (Fed. Cir. 2017), concerned U.S. 7,947,724 / 7,947,725 / 7,960,424 / 8,598,219 — not the '907 patent — and should not be conflated with this case.)
• No PTAB proceeding (IPR/PGR) naming 9,908,907 as the challenged patent was found in my searches.
• No other U.S. district court case asserting 9,908,907 was found. The many other Helsinn cases surfaced in the search (e.g., Helsinn v. Hospira, Helsinn v. Baxter, Helsinn v. Dr. Reddy's, Helsinn v. Teva, and the many ALOXI®/palonosetron ANDA suits) assert different patents.

Caveats

1. Google Patents' entry for this patent also carries a "First worldwide family litigation filed" marker linking to Darts‑ip family 48094815. I could not retrieve verified details of foreign proceedings within that family, so I make no representation about them beyond noting the flag exists.
2. My searches were web-based (Unified Patents portal, CourtListener/RECAP, DrugPatentWatch, IPWatchdog, Docket Alarm, UniCourt). This is not a substitute for an exhaustive PACER or Docket Navigator/Lex Machina run, and recently filed or sealed actions may not appear. If you need negative-clearance confidence, a PACER docket search on "9,908,907" plus a full Docket Navigator patent‑level search is warranted.
3. Per your instruction, I did not auto-correct any identifiers; where source documents conflict (e.g., the 2:22 vs. 3:22 case prefix, and "with" vs. "without" prejudice), I have reported the conflict rather than resolving it.

I'll start by verifying the structured block against web searches, since the ODP feed can lag behind recently-filed proceedings.

Let me search more specifically for IPRs on this patent and its siblings, plus the Helsinn/Azurity and Helsinn/Gland litigation chains.

Let me specifically check whether any IPR targets the '907 patent itself or its continuation siblings, and check the patent's PTAB/litigation page.

PTAB / AIA Trial Landscape — US 9,908,907

Analyst note: I sourced the proceeding list from the structured "PTAB proceedings on file" block (USPTO Open Data Portal), then ran independent web searches to catch anything ODP hadn't indexed. Both sources agree. Report as of 2026-09-19.

Proceedings overview

Zero AIA trial proceedings are on file against US 9,908,907 — no IPR, no PGR, no CBM, at the PTAB. The breakdown is therefore: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denied.

The defensive posture this gives a defendant is not "hardened," but it is also not "dead claims." It is best described as untested and fresh: the '907 claims have never been through an AIA trial, so there is no FWD to point to, no canceled claim to hide behind, and no § 315(e) estoppel to leverage. Every invalidity ground remains live and unburdened by Patent Office adjudication. That is a mixed signal — the patent family is under active, high-intensity attack (see the sibling-patent section below), but the '907 compound claims specifically have so far been left alone, and the patent owner is paying to keep them alive (8th-year maintenance fee paid 2025-08-21).

No proceedings on this patent — the absence of a per-proceeding record

There is no IPR / PGR / CBM number to report for 9,908,907. Per the operating rule, I am not inventing one. The default stated in the task prompt — "no PTAB activity on file" — is confirmed by independent search.

What is on file for the '907 patent (non-AIA context, so you can situate it):

• District court litigation flag (from the patent record): a US case filed in the New Jersey District Court — Case Nos. 2:22-cv-04635 / 3:22-cv-04635, Helsinn Healthcare S.A. v. Gland Pharma (Helsinn's § 271(e)(2) ANDA suit over Gland's fosnetupitant/palonosetron ANDA No. 217374). The '907 patent is one of the asserted patents there. Gland's Paragraph IV notice letters (2022-06-02 and 2022-07-11) certified invalidity/non-infringement of the '907 patent. This is an Article III validity fight, not a PTAB one — so no PTAB estoppel flows from it.
• Maintenance fee, 8th year, large entity, paid 2025-08-21 — the patent is being actively maintained.
• Anticipated expiration 2032-05-23 (subject to any PTA/PTE changes); Orange Book listing as both DS (drug substance) and DP (drug product) for Akynzeo (NDA 210493).

Context: live IPRs on sibling patents (NOT proceedings on '907)

These matter strategically but must not be misread as challenges to 9,908,907. In May 2025, Azurity Pharmaceuticals, Inc. filed five IPR petitions against related Helsinn patents in the same Akynzeo fosnetupitant/palonosetron family:

Key facts surfaced for these (for cross-referencing only):

• Filed: 2025-05-01 by Azurity (counsel Wilson Sonsini Goodrich & Rosati — Richard Torczon et al.). Patent owner counsel: Paul Hastings (Eric Dittmann et al.).
• Institution: granted 2025-11-19 by panels including APJs Michael J. Fitzpatrick, Sheridan K. Snedden, and Christopher G. Paulraj. The then-Acting Director denied Helsinn's discretionary-denial requests and referred the petitions to the Board on 2025-09-19 (the Advanced Bionics / material-examination-error theory).
• Grounds: pre-AIA § 103 obviousness over Herrstedt, Bös, and secondary references (Herrington, Hargreaves, Bonadeo, ALOXI). Azurity's expert is Dr. Stephen J. Peroutka; Helsinn's is Dr. Rudolph M. Navari.
• Notably absent: US 9,908,907 (and its continuations 10,208,073 / 10,717,721 / 11,312,698) are NOT among the challenged patents. Azurity challenged the method-of-use patents, not the '907 compound claims.
• A PGR (PGR2025-00035, U.S. 12,115,254) and other Azurity petitions exist against Heron, Exelixis, etc. — unrelated to this patent.

I flag these solely because the same petitioner has built a coordinated, multi-patent attack on the Akynzeo family. They create no estoppel against '907 and no issue-preclusion effect on '907's claims.

Strategic summary

Claim status of 9,908,907. All six claims — 1 (compound of formula (VI)), 2 (chloride hydrochloride salt), 3 (parenteral composition), 4 (composition + palonosetron), 5 (parenteral composition of the Cl·HCl salt), 6 (composition of claim 5 + palonosetron) — are UNTESTED. None canceled, none sustained, none construed by the PTAB. For a defendant, this is the central fact: there is no FWD to quote, so you cannot say "claim 1 is dead" and you cannot say "the PTAB blessed these claims." The compound claims (1–2) are the commercially significant ones (DS/DP listing) and have never faced an AIA validity challenge.

Estoppel landscape. Because no IPR/PGR was filed against '907, no § 315(e)(2) estoppel attaches to the '907 patent. A current defendant is free to raise any § 102/§ 103 ground in district court, unconstrained by PTAB ground-pool estoppel. Conversely, the Azurity IPRs on the sibling patents do trigger § 315(e)(2) estoppel for Azurity (and privies) as to those instituted patents' claims — but that estoppel runs to the '826/'357/'515/'297 claims, not to '907. If the same petitioner later pursued '907, it would face a fresh § 315(b)/§ 325(d) and discretionary-denial fight, not an estoppel bar from the earlier siblings.

Pattern signals. (1) Same petitioner, coordinated campaign: Azurity filed five petitions in one day (2025-05-01) across four Helsinn patents — a classic knock-out-the-family strategy — but conspicuously did not touch the '907 compound claims. (2) Aggressive patent-owner defense: Helsinn sought discretionary denial in every one of those IPRs and lost at the Director level on 2025-09-19; expect the same posture if '907 is petitioned. (3) No defensive aggregator: I found no indication of a Unified Patents or similar defensive aggregator in the chain; the challenges are being driven by a commercial rival (Azurity), not a nonprofit. (4) Parallel district-court invalidity case (Gland, D.N.J.) is the current proving ground for '907's validity.

Recommended next steps

• Do not represent to a court or a counterparty that any '907 claim has been canceled or sustained. No AIA trial has touched this patent. The absence is the story: a well-asserted, revenue-critical compound patent that has attracted a five-petition family attack yet remains AIA-untested on its core compound claims.
• If you are a defendant facing a '907 demand/assertion: you are the first mover. You carry no § 315(e) estoppel and face none. The obvious IPR handles are the '907 compound claims (1–2) and the parenteral-composition claims (3–6). Prior-art families worth scoping: the Bös 4-phenyl-pyridine lineage (US 6,297,375; US 6,479,483), the University of Kansas N-phosphoryloxymethyl prodrug work (US 5,985,856), Krise I (J. Med. Chem. 42:3094–3100, 1999), DeGoey (J. Med. Chem. 52:2964–2970, 2009), and Oslob (Bioorg. Med. Chem. Lett. 19:1409–1412, 2009) — the same art family Gland pleaded in its '907 counterclaims. Note the '907 prosecution relied on a § 103 rejection history and on the stability-of-the-chloride-hydrochloride-salt story (spec, Tables 1–2, FIG. 1), so secondary-considerations / unexpected-stability evidence is the likely patent-owner rebuttal; scope your obviousness case to neutralize it up front.
• Watch timing against the siblings. Institution of the sibling IPRs occurred 2025-11-19, so their statutory final-written-decision due dates fall around November 2026 (PTAB's one-year clock from institution, § 316(a)(11)). A loss on the closely-related method-of-use claims could embolden a follow-on petition against '907; a win for Helsinn could deter one. Either way, the sibling FWDs are the leading indicator for the '907 patent's next 12 months.
• If any '907 petition surfaces that ODP hasn't ingested, treat the search-confirmed record here as the baseline and re-check the USPTO PTAB E2E docket and Docket Alarm case pages (e.g., the IPR2025-0094x series) for a newly docketed '907 proceeding.

Sources / links

• USPTO PTAB Structured "PTAB proceedings on file" block for US 9,908,907 — empty (canonical: no AIA trial).
• Related-patent institution decision (context): Azurity Pharmaceuticals, Inc. v. Helsinn Healthcare S.A., IPR2025-00948 (Pat. 9,943,515) — institution granted 2025-11-19; and IPR2025-00947 (Pat. 9,186,357) — institution granted 2025-11-19. Docket pages: docketalarm.com/cases/PTAB/IPR2025-00948/... and .../IPR2025-00947/....
• District-court record: Helsinn Healthcare S.A. v. Gland Pharma, D.N.J. Case Nos. 2:22-cv-04635 / 3:22-cv-04635 — Gland counterclaims at storage.courtlistener.com/recap/gov.uscourts.njd.499051/gov.uscourts.njd.499051.25.0.pdf (asserts '907 and the related patents; Paragraph IV notices 2022-06-02 / 2022-07-11).
• Patent record / claims and maintenance: patents.google.com/patent/US9908907B2/en; patents.justia.com/patent/9908907 (six claims, 8th-year maintenance fee paid 2025-08-21).

If you need the claim-by-claim text of the '907 claims or the sibling FWDs once issued, say so and I'll pull the exact language rather than paraphrase.

I'll research the assignment chain for US 9,908,907, prioritizing the USPTO Assignment Center records and cross-referencing the legal events already established.

Let me pull the specific assignment records (reel/frame) and check the security-interest/lender chain in more detail.

Let me try to obtain the specific reel/frame numbers and the recorded correspondent for these assignments.

I've reached the tool-call limit for this run, so I'll report exactly what the record establishes and flag the specific gap (reel/frame and correspondent) that I could not pull from the Assignment Center directly.

Inventors

Eight named inventors (source: issued face of patent; Justia/PubChem inventor lists, which agree):

Pattern notes:

• Seven of eight inventors reside in the Swiss-Italian border cluster around Helsinn's R&D footprint (Giubiasco/Mendrisio CH; Como/Varese IT) — a normal corporate medicinal-chemistry team, not an outsourced or nominee group.
• Valentino J. Stella is the outlier and the historically significant name. He is a University of Kansas pharmaceutical-chemistry professor and is the first-named inventor on U.S. 5,985,856 (Kansas, N‑phosphoryloxymethyl prodrugs of secondary/tertiary amines) — the reference the '907 itself cites and the platform the '907 compound embodies. This is the same co-inventor/prior-art overlap flagged in the Prior Art section, and it explains why a US academic appears among an otherwise all-Helsinn team.
• No "all-inventors-departed" pattern is evidenced. I found no record of the Helsinn inventors leaving within 12 months of filing. This does not fit the pre-fire-sale precursor pattern. (Caveat: I could not run employment/HR records; this is a negative finding from the assignment and litigation record only.)

Original assignee

Helsinn Healthcare SA (Lugano/Pazzallo, Switzerland) — named as assignee on the issued patent and as the applicant of record.

• Primary line of business: Operating, privately held specialty pharmaceutical group (Helsinn Group), focused on cancer-supportive care / antiemetics. It is a product company, not a licensing vehicle. Publicly reported as a foreign corporation in USPTO PatentsView, ~51 granted US patents.
• Did it ship a product embodying the claims? Yes. The '907 is listed in the Orange Book for AKYNZEO (NDA 210493) as DS (drug substance) and DP (drug product), expiration 2032-05-23, for fosnetupitant chloride hydrochloride / palonosetron hydrochloride IV — i.e., the claimed compound is the commercial active. This is the single most important fact for the verdict (see below).
• Current status: Operating. It is actively maintaining the patent (8th-year maintenance fee paid 2025-08-21) and actively asserting it. It granted a group-level security interest in Dec 2022 (see timeline) and obtained a release in Sep 2023 — indicative of routine secured corporate financing, not distress. No bankruptcy, no dissolution, no acquisition of the patent-owning entity was found.
• Litigation posture: Plaintiff in Helsinn Healthcare S.A. v. Gland Pharma Ltd., D.N.J. 3:22‑cv‑04635 (and the parallel-listed 2:22‑cv‑04635), asserting the '907 and nine sibling patents against a generic ANDA filer under § 271(e)(2)(A) — a classic operating-company-vs-generic case, not an NPE assertion.

Assignment timeline

Sourcing caveat (read first): I could retrieve the Google Patents legal-events mirror of the Assignment Center records (conveyance type, assignor, assignee, dates) for this patent, and those events are confirmed below. I was cut off before I could retrieve the specific reel/frame numbers and the recorded correspondent of record for the '907 patent itself. I am therefore not inventing reel/frame numbers or naming a correspondent. The reel/frame format for Helsinn filings is known from a related Helsinn document (e.g., Reel 030478/Frame 0960), but none of those numbers is confirmed as belonging to the '907. Treat reel/frame and correspondent as to be verified at the Assignment Center.

Three recorded events, chronological:

• Recorded 2017‑11‑06 (execution date not retrieved) — Reel not retrieved/Frame not retrieved

  • Conveyance: Assignment (assignment of assignors' interest)
  • Assignor: Stella, Valentino J; Cannella, Roberta; Fadini, Luca; Giuliano, Claudio; Lovati, Emanuela; Manini, Peter; Pietra, Claudio; Venturini, Alessio (all eight inventors)
  • Assignee: HELSINN HEALTHCARE SA
  • Correspondent: not retrieved — cannot flag recurrence.
  • Context: Original inventor-to-company assignment recorded during prosecution (patent granted 2018-03-06). Consistent with the application being filed by/on behalf of Helsinn. Not an acquisition; the standard corporate vesting step.

• Recorded 2022‑12‑30 (execution date not retrieved) — Reel not retrieved/Frame not retrieved

  • Conveyance: Security Interest (Security Agreement)
  • Assignor: HELSINN BIREX PHARMACEUTICALS LIMITED; HELSINN HEALTHCARE SA; HELSINN THERAPEUTICS (U.S.), INC. (three Helsinn group entities)
  • Assignee: HAMILTON SA LLC
  • Correspondent: not retrieved — cannot flag recurrence.
  • Context: Securitization / collateral financing. A group-level lien over the Helsinn patent estate granted to a secured party. A security interest is a lien, not a transfer of title — Helsinn continued to own and use the patent throughout. It is not a transfer-to-asserter.

• Recorded 2023‑09‑20 (execution date not retrieved) — Reel not retrieved/Frame not retrieved

  • Conveyance: Release by Secured Party
  • Assignor: HAMILTON SA LLC
  • Assignee: HELSINN HEALTHCARE SA; HELSINN BIREX PHARMACEUTICALS LIMITED; HELSINN THERAPEUTICS (U.S.), INC.
  • Correspondent: not retrieved — cannot flag recurrence.
  • Context: Lien release/termination. The security interest was discharged ~9 months after recording — consistent with repayment/refinancing of the underlying facility, not default or foreclosure. No bankruptcy or fire-sale overtones.

Bottom line on the chain: the patent never left the Helsinn group. There is exactly one ownership transfer (inventors → Helsinn, 2017) plus one security interest and its release (2022–2023). The "current assignee" per the record is Helsinn Healthcare SA.

Note on the family: the Google Patents entry also carries "Priority to" events for US 15/874,325 (2018‑01‑18), 16/228,835 (2018‑12‑21), 16/896,135 (2020‑06‑08) and 17/699,522 (2022‑03‑21). These are continuation/divisional filings within the same Helsinn family, not assignments — do not miscount them as ownership changes. (This mirrors the "Priority to" vs. continuation caution flagged in the Patent summary section.)

Timeline diagram

timeline
    title Ownership of US 9908907
    2011 : Earliest priority date
    2016 : Filed by Helsinn Healthcare SA
    2017 : Inventors assign to Helsinn Healthcare SA
    2018 : Patent granted
    2022 : Gland Pharma infringement suit filed
         : Security interest to Hamilton SA LLC
    2023 : Security interest released by Hamilton

NPE / troll-pattern signals

Verdict

Operating-company assertion.

Justification: the record shows a single inventor-to-company assignment to Helsinn Healthcare SA (recorded 2017‑11‑06), followed only by a security interest to Hamilton SA LLC (recorded 2022‑12‑30) that was released (recorded 2023‑09‑20) — a corporate-financing lien, not a transfer to an asserter — leaving Helsinn as owner. Helsinn is an operating pharmaceutical company that ships a product embodying the claims (AKYNZEO, NDA 210493, with the '907 listed as DS/DP in the Orange Book) and asserts the patent directly against generic ANDA filers (Helsinn v. Gland Pharma, D.N.J. 3:22‑cv‑04635). No NPE signals are present; the only unresolved item is signal 3 (correspondent recurrence), which is a data gap rather than evidence.

Verification links: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ ; legacy Assignment Search — https://assignment.uspto.gov/patent/index.html (search patent number 9908907).

Explicit limitations of this run

1. Reel/frame numbers were not retrieved for any of the three recordings; the timeline above uses conveyance type, parties, and recording dates confirmed via the Google Patents legal-events mirror. Reel/frame must be pulled from the Assignment Center to complete the record.
2. Correspondent of record was not retrieved, so NPE signal 3 is reported as unclear, not as a finding.
3. Execution dates for each assignment were not separately retrieved (recording dates are shown).
4. The security-interest assignors are three Helsinn entities (Helsinn Birex Pharmaceuticals Limited, Helsinn Healthcare SA, Helsinn Therapeutics (U.S.), Inc.) appearing on a single group financing; I did not verify the underlying credit agreement or the identity/role of Hamilton SA LLC beyond its status as secured party on the recorded instrument. I have not characterized it as any specific lender or fund because I could not verify that.
5. No auto-correction of identifiers was performed; the "9908907" literal also collides with an unrelated international application number (US 1999/008907), as flagged in the Prior Art section.

I'll research the specific prior-art citations for US 9,908,907. Let me run targeted searches.

I found the exact "Referenced Cited" list. Now let me research each cited reference in detail.

I have the core citation list. Let me now research the remaining specific references.

Prior-Art Analysis — US 9,908,907 B2 ("Substituted piperaziniums for the treatment of emesis")

0. Scope, method, and a sourcing caveat

Number confirmed literally. Searches returned only US 9,908,907 B2 (application 15/194,984; Helsinn Healthcare SA; granted 2018‑03‑06). I did not fold in any similar number.

⚠️ Literal-search hazard worth recording: a search for the bare string "9908907" also returns PCT application US1999/008907 ("US 9908907 W", filed 1999‑04‑23), which is the Medicus biopsy‑tract hemostasis PCT (see the family of EP 1,075,219 / WO 1999/056632). That is an international application number, not a US patent number, and is unrelated to the '907 patent. Anyone searching on the numeric string alone will get false hits. I report it rather than silently filtering it, per the no-auto-correction rule.

Method / limitations. I retrieved the patent's face-of-document citation list and confirmed it against the PubChem patent record for US‑9908907‑B2. I was cut off by a tool-call limit before I could independently verify two of the ten cited U.S. patents (US 6,719,996 and US 7,211,579) and several of the foreign documents. Those are flagged below. Where I could not verify, I say so rather than characterising the reference.

Resolving the previous section's caveat. My earlier summary flagged that the claim text came from a secondary database and could not be checked against the granted document. That concern is now resolved: the Espacenet claims page for US9908907 (B2) reproduces claims 1–6 verbatim and matches the Justia reproduction, so the six-claim structure (independent claims 1, 2, 3, 5; dependent claims 4, 6) is now high confidence. Note, however, that the granted claim 2 is broader than my earlier summary implied — it recites the same (a)/(b)/(c) limitations as claim 1, not merely "a chloride hydrochloride salt."

1. The citation list actually on the face of US 9,908,907

The patent cites the following (source: Justia patent copy of "Referenced Cited", cross-checked against the PubChem US‑9908907‑B2 citation list, which agrees item-for-item on the U.S. and foreign documents).

1a. U.S. patent documents

1b. Foreign patent documents

1c. Non-patent literature cited

2. Which claims could each reference anticipate under 35 U.S.C. § 102?

2.1 The threshold point

The '907 patent's effective filing/priority date is 2011‑11‑29. Every cited reference except WO 2013/082102 was published well before that date and is therefore available as prior art under AIA § 102(a)(1). Availability, however, is not anticipation. Anticipation requires that a single reference disclose every element of the claim as arranged in the claim — including, for claims 1–6 here, the specific quaternary N‑(phosphonooxy)methyl piperazin‑1‑ium 4‑phenyl‑pyridine scaffold (formula (VI) as narrowed), and for claims 4/6 the co-formulation with palonosetron.

On the record I have, no cited reference discloses fosnetupitant (GA1), its chloride hydrochloride salt, or a parenteral fosnetupitant/palonosetron composition. The examiner allowed the claims (primary examiner Douglas M. Willis), which is consistent with that reading. The citations are therefore best characterised as § 103 obviousness references rather than § 102 anticipatory references — with two caveats identified below (US 5,985,856 and the composition claims).

2.2 Reference-by-reference

3. Points that deserve explicit flagging

3.1 The strongest § 103 combination on the face of the patent is Stella '856 + Bös '375/'483. Stella supplies the N‑phosphoryloxymethyl-quaternary-ammonium prodrug platform, the zwitterion/internal-salt concept, the sodium/ammonium cation variants, and the IV/lyophilised composition teaching. Bös supplies the 4‑phenyl‑pyridine NK1 antagonist and its exact structure (including the 3,5‑bis(trifluoromethyl)phenyl and o‑tolyl groups and the N‑methylpiperazine). The distinguishing features the '907 relies on are the one-step acid-free synthesis, the two-equivalents-of-HCl stabilisation (contrasted against prior-art dibasic salts, which the specification says "reform the underlying drug during storage"), and the chloride hydrochloride salt stability — plus the palonosetron co-formulation of claims 4/6. That is where the non-obviousness argument actually lives, not in § 102.

3.2 The Stella '856 is a co-inventor conflict-of-interest footnote. Valentino J. Stella is a named inventor on the '907 patent and the first-named inventor on the University of Kansas '856 patent. The '907 specification concedes the '856 patent explicitly in the background. This is the reason the reference appears both in the specification text and on the face of the patent.

3.3 "Chloride hydrochloride salt" is a claim-term worth scrutinising. Claim 2 (and claims 5/6) recite a "chloride hydrochloride salt." The '907 specification calls the "chloride hydrochloride HCl salt of GA1" a particular preferred compound and asserts it is "tremendously resistant to decoupling of the oxo‑phosphonomethyl." Whether the '856's generic recital of "an anion A" and "a cation X" (which includes chloride and hydrogen) discloses a chloride hydrochloride salt for § 102 is doubtful — the '856 nowhere names such a salt, and salt-form selections are normally § 103 territory. But a litigant could argue the genus reads on it.

3.4 Claims 3 and 5 are the widest claims and the ones most exposed to § 102. They recite "a compound of formula (I)" — the broad genus — plus "one or more liquid pharmaceutical excipients," in a parenteral composition. Unlike claims 1/2, they are not limited to the formula (VI) narrowing. Their vulnerability therefore turns entirely on (i) how broadly formula (I) was construed, and (ii) whether any pre‑2011 reference discloses a liquid parenteral composition of a formula (I) compound. On the current record, the only cited reference that even plausibly does is US 6,719,996, which I could not verify.

3.5 WO 2013/082102 needs a proper "is it my own art?" analysis. It is dated after the priority date, so it is not § 102(a)(1) art. If it is the applicant's own PCT in the fosnetupitant family, then even as § 102(a)(2) art the § 102(b)(2)(C) common-ownership exception (or the pre-AIA § 103(c) analogue for the priority year) would likely remove it. Its presence on the citation list is most likely an IDS formality, not an examiner rejection.

3.6 The "prior art date 2011‑11‑29" shown on Google Patents is a priority-date assumption, not a legal conclusion. The '907 was filed 2016‑06‑28 as a continuation; the 2011‑11‑29 date depends on a valid benefit chain through the parent PCT/US2012/066778 (whose ISR is cited on the patent) back to the earliest applications. If any link in that chain fails (e.g., for the palonosetron-combination claims 4/6, which may have been added later), the effective date for those claims moves forward and the prior-art universe expands. I have not verified the benefit chain document-by-document.

3.7 Litigation context (from the prior section, unchanged). The '907 was asserted in Helsinn Healthcare S.A. v. Gland Pharma Ltd., D.N.J. No. 3:22‑cv‑04635, and resolved by a consent judgment of dismissal with prejudice entered 2023‑01‑18, with Gland admitting a § 271(e)(2)(A) technical act of infringement against ANDA 217374. That outcome is a validity non-adjudication: none of the § 102 arguments above was tested in that case.

4. Uncertainties I am not able to resolve from this run

• Full text of US 6,719,996 (Kuentz) and US 7,211,579 (Funk) — not retrieved. The '996 patent is the one reference that could matter to claims 3/5; the '579 patent cannot be assessed at all.
• WO 2006/099,968, WO 2009/138393, WO 2011/061622 — dates given (Sep 2006 / Nov 2009 / May 2011) are consistent with the citation-list ordering and with general publication practice, but I did not open these documents.
• JP 2001‑527083 and JP 2008‑534454 — my characterisation as national-phase counterparts of WO 99/33846 and WO 2006/099,968 respectively is inference from date and pairing, not verification.
• The literal scope of formula (I) as issued — I have the Markush text but not the drawings; whether formula (I) reads on the parent netupitant governs the whole claim 3/5 analysis.
• Benefit chain / effective date per claim — not verified. All dates above are stated as availability dates on the assumption that 2011‑11‑29 controls; if it does not for some claims, additional references published 2011‑2016 become available and would need to be searched fresh.
• I did not find any 2026 Federal Circuit docket or PTAB proceeding naming US 9,908,907, consistent with the earlier section; the Azurity IPRs (filed 2025‑05‑01; instituted 2025‑11‑19) target US 10,828,297, 9,186,357, 8,623,826 and 9,943,515 — not the '907 patent. Absence of a hit in general web search is not conclusive.

I'll build the § 103 analysis on the prior-art record already established, but first let me try to close the two evidentiary gaps flagged in that section (US 6,719,996 and US 7,211,579), since both bear directly on the composition claims.

§ 103 Obviousness Analysis — US 9,908,907 B2

Building on the Prior-Art section above (cross-referenced, not repeated). Two evidentiary gaps flagged there are now closed by the searches run for this section:

• US 6,719,996 (Kuentz) — verified as "Galenic composition for low bioavailability medicaments," Hoffmann‑La Roche, priority 2000‑12‑14, granted 2004‑04‑13. It claims an oral composition (0.01–15% w/w active, 30–80% w/w edible lipid matrix, 1–20% emulsifier) of an active with bioavailability ≤20%; claim 13 lists "4‑phenylpyridin derivatives," and claim 15 specifically names netupitant — 2-(3,5-bis-trifluoromethyl-phenyl)-N-methyl-N-[6-(4-methyl-piperazin-1-yl)-4-o-tolyl-pyridin-3-yl]-isobutyramide (PubChem US‑6719996‑B2; FreePatentsOnline).
• US 7,211,579 (Funk) — verified as "NK‑1 receptor antagonists," Hoffmann‑La Roche, appl. 11/376,042, granted 2007‑05‑01 (uspto.report). Same Roche NK1 programme; content beyond the title/assignee not retrieved.

Net effect: the parent drug (netupitant) is expressly named in a granted, pre‑priority pharmaceutical-composition patent that frames it as a low-bioavailability molecule — the single most useful motivation-to-combine fact in the whole record.

⚠️ Contradiction flagged (per instructions). The "Patent summary" section characterised claim 2 as "a chloride hydrochloride salt of a compound of formula (VI)." The "Prior Art" section reports the granted claim 2 as reciting the same (a)/(b)/(c) limitations as claim 1 plus the salt limitation. Both can be true simultaneously (a salt claim with the same narrowing), but the wording differs between sections and I am not resolving it here. Confirm against the granted claims before relying on it.

A. Legal frame and the two threshold issues that drive everything

Standard. Obviousness is the Graham v. John Deere inquiry (scope/content of the prior art; differences from the claims; level of ordinary skill; secondary considerations), applied through the KSR Int'l v. Teleflex lens: a combination is obvious where the improvement is a predictable use of prior-art elements according to known methods, the design is driven by market demand, and the solution is one of a finite number of identified, predictable solutions ("obvious to try"). Conversely, a combination is not obvious where the prior art teaches away or where there was no reasonable expectation of success (In re O'Farrell).

Threshold issue 1 — effective date. The patent's face carries "prior art date 2011‑11‑29." The application was filed 2016‑06‑28 as a continuation (parent PCT/US2012/066778, ISR cited on the patent). If every claim benefits from 2011‑11‑29, the pre‑AIA §§ 102/103 regime (with pre-AIA § 103(c) common-ownership disqualification limited to § 102(e)/(f)/(g) art) most likely governs. If any claim ever had a post‑2013‑03‑16 effective filing date, the AIA regime applies and § 102(b)(2)(C) applies instead. I did not verify the benefit chain claim-by-claim, and for claims 4/6 (palonosetron combinations) specifically the effective date is the most likely to slip, which would expand the art. Flagged, not resolved.

Threshold issue 2 — pre-AIA § 103(c) / § 102(b)(2)(C) and Helsinn's own publications. Several cited foreign documents appear to be the applicant's own family (e.g., WO 2013/082102, discussed in the prior section; and WO 2011/061622, which the epublished Helsinn combination documents EP2722044/EP2722045/EP2744497 cite in their reference lists). A printed publication by the same inventive entity/COT within the year is not automatically art — but pre‑AIA § 103(c) does not disqualify § 102(a)/(b) printed publications even if commonly owned. This is a genuine asymmetry; whether Helsinn's own pre-priority publications can be used depends on inventorship identity and on which regime applies. Flagged.

B. The structural delta that the § 103 case turns on

The whole dispute is a single, well-characterised prodrug "handle" bolted onto a known molecule:

Verified chemical identity of the active (PMDA/MedChemExpress/J‑GLOBAL): 4-[5-{2-[3,5-bis(trifluoromethyl)phenyl]-N,2-dimethylpropanamido}-4-(2-methylphenyl)pyridin-2-yl]-1-methyl-1-[(phosphonooxy)methyl]piperazin-1-ium chloride monohydrochloride, C₃₁H₃₆ClF₆N₄O₅P·HCl, MW 761.52, CAS 1643757‑72‑5.

Takeaway: the claimed compound is the parent NK1 antagonist plus a known solubility-enhancing prodrug moiety, quaternised onto the exact nitrogen that Stella '856 teaches to quaternise.

C. Combination I (primary) — Bös '375 / Bös '483 + Stella '856, optionally + Kuentz '996 → claims 1 and 2

Element supply.

Motivation to combine (the KSR factors, consolidated):

1. Same field, same problem, same molecule. Stella '856 is a prodrug platform patent expressly aimed at "water soluble prodrugs of secondary and tertiary amine containing drugs." Netupitant is a tertiary-amine drug. Kuentz '996 supplies the express statement that netupitant has bioavailability of 20% or less — i.e., the prior art itself identified the exact deficiency the '907 claims are designed to fix. This is the strongest possible motivation fact: it is not inferred, it is stated in a granted Roche patent naming the compound.
2. Market/practical demand (KSR prong 1). CINV therapy is given to patients who are vomiting or about to be; an IV option is a recognised clinical need. The '907 specification itself recites that netupitant "is currently under clinical development in combination with palonosetron … for the prevention of CINV" — an applicant-admitted clinical programme that supplies both the demand and the commercial framing.
3. The prior art had already chosen prodrug-on-the-amine as the fix for this scaffold. US 6,593,472 / 6,747,026 / 6,806,370 (Roche N‑oxides) are amine-oxide prodrugs of the very same 4‑phenyl‑pyridine class, and the '907 specification admits they were "intended to overcome limitations … such as solubility or pharmacokinetic limitations." So the art taught: for this scaffold, the way to fix PK/solubility is to derivatise the basic nitrogen. Stella '856 supplies the alternative, better-characterised derivatisation (phosphoryloxymethyl). Combining "derivatise the amine" (Roche) with "use the phosphonooxymethyl handle" (Stella) is a predictable substitution of one known prodrug strategy for another on the same scaffold — the classic KSR "known options within the finite set" situation.
4. The mechanism was known and expected to work. Krise/Stella (J. Med. Chem. 42:3094–3100, 1999; and the '856 examples on quinuclidine, cinnarizine, loxapine, amiodarone) taught that the phosphate ester is cleaved by ubiquitous endogenous phosphatases to release the parent amine. A POSITA would therefore expect both the solubility gain and the in-vivo reversion to netupitant.

Reasonable expectation of success. High. The chemistry is a one-step N-alkylation of a tertiary amine with a chloromethyl dialkyl phosphate — squarely within the '856 disclosure and within routine skill. The '907 specification effectively concedes the reaction is doable (it claims the one-step acid-free improvement over the multi-step prior art), which is a concession that the prior art route worked, just less economically.

What the combination does not supply (where the patent owner's real estate is): the specific chloride hydrochloride salt, its unexpected storage stability vs the prior-art-preferred dibasic salts, and the low dimer/decoupling profile. That is claim 2's battleground, not claim 1's.

Assessment: Claim 1 is at high risk of § 103 invalidity. Its genus (R₂‌₀₀/R₃‌₀₀ = H or methyl) is the acid/zwitterion and the dimethyl ester — a tiny, predictable set squarely inside Stella's disclosed genus combined with Bös's disclosed scaffold. Claim 2 (the Cl·HCl salt) is the defensible claim, and only if the stability data establish a true unexpected-result nexus (see §I).

D. Combination II (alternative/adjunct) — Roche N‑oxide prodrug line as the "reason to modify the amine"

Even standing alone, US 6,593,472 + US 6,747,026 + US 6,806,370 supply the conceptual teaching that the 4‑phenyl‑pyridine NK1 class is amenable to amine-derivatised prodrugs for solubility/PK reasons, and the '907 specification's own background repeat that teaching. A petitioner will pair that with Stella '856 to argue an even tighter motivation chain: Roche told the artisan to make an amine-derivatised prodrug of this scaffold; Stella told the artisan which derivatisation to use and how. The '907's attempted distinction — that the '026 patent reports "no physicochemical or biological data" — is a data-void argument, not a teaching-away argument, and carries little weight: an absence of data in a reference does not remove its teaching.

E. Combination III — claims 3 and 5 (parenteral, liquid-excipient compositions)

Motivation: formulating a known, water-soluble prodrug salt as an injectable is the reason the prodrug was made. KSR makes the leap from "compound made more water-soluble for IV use" to "aqueous IV composition" nearly automatic.

Note the Kuentz '996 near-miss. My earlier concern was whether '996 might anticipate claims 3/5. It does not: it is an oral galenic (chewable/lipid) composition, not parenteral, and its active is the parent tertiary amine, not the quaternary phosphonooxymethyl salt. So '996 is § 103 motivation art only, not § 102 art, for claims 3/5.

Assessment: claims 3 and 5 at high risk, and they are the broadest claims (they recite formula (I), not the narrowed formula (VI)), so they are the most exposed.

F. Combination IV — claims 4 and 6 (compositions further comprising palonosetron)

Motivation is essentially admitted in the specification. The '907 background states netupitant "is currently under clinical development in combination with palonosetron (a 5‑HT₃ receptor antagonist) for the prevention of CINV" — an applicant-admitted prior-art combination. Add:

• Helsinn-family publication WO 2011/061622 (cited on the face of '907; published May 2011, i.e. before 2011‑11‑29). Its title/subject was not verified in this run, but it is referenced in the reference lists of the Helsinn netupitant + palonosetron ± dexamethasone family (EP2722044A1 / EP2722045A2 / EP2744497B1 surfaced in search), which claim combinations of netupitant + palonosetron + dexamethasone for CINV. That is exactly the combination claim 4/6 recites, on the parent drug.
• The 5‑HT₃ + NK1 dual-antagonist antiemetic rationale (Kramer 1998; Gesztesi 2000; and the clinical literature in the EP documents).

Motivation to combine: the NK1/5‑HT₃ co-administration was known, clinically supported, and commercially obvious. Swapping the NK1 component from netupitant to its phosphonooxymethyl prodrug (Combination I) does not change the rationale for co-formulating with palonosetron. Under KSR, where the combination of actives was known and the substitution is of a known bioequivalent prodrug for its parent, the composition claim follows.

Caveat that may save claims 4/6. Co-formulation is not purely mechanical: an IV fosnetupitant/palonosetron co-formulation raises real compatibility/stability questions (the EP3626231 family shows the later development work on pH, EDTA, mannitol, and the netupitant:fosnetupitant ratio). If Helsinn can show that the claimed salts/compositions solved a specific, unexpected co-formulation instability, that is its best secondary-considerations hook for claims 4/6. That evidence is not in the '907 record I have (FIG. 1/Tables 1–2 concern salt degradation generally, not the palonosetron combination).

Assessment: claims 4 and 6 at moderate-to-high risk, contingent on whether the record shows a genuine co-formulation surprise; absent that, the claim is a known combination with a known prodrug swap.

G. The patent owner's case — where the nexus is strong and where it is not

Strong (product claims):

• Unexpected stability of the chloride hydrochloride salt. The specification states that the prior art "preferred the use of dibasic salts of (phosphooxy)methyl substituents," and that "the present invention had found that such salts are unstable and reform the underlying drug during storage." FIG. 1 and Tables 1–2 then show mass-percent degradation over time for multiple salts, with the disodium salt as the benchmark and only a "few salts" outperforming it. The preferred Cl·HCl salt is said to be "tremendously resistant to decoupling of the oxo‑phosphonomethyl." If the data genuinely show the prior-art-preferred salt class is unstable and the claimed salt class is stable, that is a teaching-away + unexpected-results case for claim 2 (and possibly 5/6). This is the patent's centre of gravity.
• Low dimer / low parent-reversion. The crystalline-forms family (US 2017/0096442 A1) defines fosnetupitant by reference to "less than 1.0 wt% dimer" and "less than 1.0 wt% parent." A low-impurity product not taught by the prior art can support a nexus — but note that figure comes from a different family member, not from the '907 specification, and secondary considerations must be tied to the claimed invention.

Weak (and this is where a petitioner should attack):

• The "one-step, acid-free synthesis" argument is largely irrelevant to claims 1–6. The '907's asserted advance over the prior art is a process advance ("the prior art had required multiple synthetic steps … proton scavengers … strong acid to deprotect"). Claims 1 and 2 are product claims and claims 3–6 are composition claims. A process advantage does not rebut product-claim obviousness unless it produces a structurally/physically different product or there is a showing of nexus to the claimed subject matter. Expect Helsinn to lose this argument on the product claims.
• "The '856 doesn't disclose how the moiety would affect structure, stability, cost, selectivity" is an absence-of-teaching argument. Under KSR, where the prior art discloses a finite set of predictable options, the burden is on the patentee to show the result was unexpected, not merely undisclosed.
• Claim 1's breadth cuts against the patentee. A genus spanning the free acid/zwitterion and the dimethyl ester is hard to defend as a narrow, unexpected selection.

H. Claim-by-claim risk matrix

Structural reading: the compound genus and the parenteral-composition claims (1, 3) are the most vulnerable; the salt-specific claims (2, 5) and the combination claims (4, 6) are where the § 103 fight will actually be won or lost, and the outcome is data-dependent.

I. What a complete § 103 record still needs (and what I could not verify)

1. The granted claims, verbatim — the Markush scope of formulae (I) and (VI) determines whether claims 3/5 are as broad as they appear. Sourced second-hand in both prior sections; the claim-2 wording discrepancy noted above remains unresolved.
2. Full text of WO 2006/099,968, WO 2009/138393, WO 2011/061622 and EP 1,103,545 — these are the likely "same-family vs. third-party art" hinge, and § G above depends on what they actually disclose.
3. The '907 prosecution file history. The prior section notes a § 103 rejection history; the examiner's grounds and Helsinn's rebuttal are the single best roadmap to both the strongest combination and the weak point in the patentee's story.
4. A POSITA declaration defining the ordinary artisan (I would propose: a PhD/PharmD-level medicinal chemist or pharmaceutical scientist with ~3–5 years in CNS/antiemetic drug development, familiar with prodrug strategies and parenteral formulation) and opining on motivation and expectation of success.
5. Secondary-considerations evidence tied to the claims — full degradation datasets behind FIG. 1/Tables 1–2, the identity of the "few salts [that] manifested more desirable results," and any contemporaneous evidence of unexpected results. Without these, the teaching-away argument is narrative, not proof.
6. The pre-AIA/AIA determination and the § 103(c)/§ 102(b)(2)(C) common-ownership analysis for the Helsinn-family publications.
7. Additional art a petitioner would develop beyond the face citations (identified in the litigation/prior-art sections): Krise/Stella 1999 (J. Med. Chem. 42:3094–3100), DeGoey 2009 (J. Med. Chem. 52:2964–2970), Oslob 2009 (Bioorg. Med. Chem. Lett. 19:1409–1412) — i.e., the general phosphoryloxymethyl/phosphate-prodrug literature — plus the general parenteral-formulation and salt-selection art.

J. Statement of uncertainty

This is a technical analyst's § 103 assessment, not a legal opinion; obviousness is ultimately a legal conclusion reserved to the factfinder. I have built only on the prior-art record and citations already established, plus two newly verified references (US 6,719,996 and US 7,211,579), and I have not auto-corrected any identifier. The following are not established by anything in this run and are flagged rather than assumed: the verbatim granted claim scope (and the claim-2 wording contradiction between the two prior sections); the content of WO 2006/099,968, WO 2009/138393, WO 2011/061622, EP 1,103,545 and the JP documents; the per-claim effective filing date; whether the pre-AIA or AIA regime applies; and whether any actual secondary-considerations evidence beyond the specification's own FIG. 1 narrative exists in the file. No PTAB proceeding, Federal Circuit appeal, or other holding on the '907 claims' validity was found (consistent with the earlier sections), so none of the § 103 grounds discussed here has ever been adjudicated.