US 9364473

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

Let me verify litigation and any 2026 appellate activity specifically tied to this patent.

Let me look for PTAB proceedings on the '473 patent and verify the actual claims.

US Patent 9,364,473 — Analyst Summary

Important preface on scope of verification: All facts below are grounded in the patent text you supplied, the Google Patents record, and litigation/PTAB documents retrieved today. I found no Federal Circuit 2026 docket activity naming US 9,364,473, and I found no IPR petition directed at the '473 patent itself. Where I cannot confirm something, I say so explicitly rather than guess.

---

1. Bibliographic data

Continuity (from the patent's own Cross-Reference section): the '473 patent is a continuation of Ser. No. 14/406,776 (filed Dec 10, 2014), which is the 35 U.S.C. §371(c) national phase of PCT/US2013/045495 (filed Jun 12, 2013), claiming benefit of the two provisionals above. Source: Google Patents US9364473B2.

FDA/Orange Book context: the '473 patent is listed in the Orange Book against NDA No. 207793 (Onivyde®, irinotecan liposome injection), use code U-1856, listed expiry Jun 12, 2033 (Google Patents; TheraRadar; DrugPatentWatch).

---

2. Abstract (verbatim)

"Provided are methods for treating pancreatic cancer in a patient by administering liposomal irinotecan (MM-398) alone or in combination with additional therapeutic agents. In one embodiment, the liposomal irinotecan (MM-398) is co-administered with 5-fluorouracil and leucovorin."

---

3. What the patent is about

The '473 patent claims dosing regimens for treating pancreatic cancer using liposomal irinotecan (MM-398 / "nal-IRI," now marketed as Onivyde®) — a unilamellar PEGylated liposome, ~80–140 nm, encapsulating irinotecan as a sucrose octasulfate salt — either as monotherapy or combined with 5-fluorouracil (5-FU) and leucovorin (LV). A central theme is UGT1A1*28 genotype-guided dose reduction: patients homozygous for UGT1A1*28 (the "7/7" genotype, reduced SN-38 glucuronidation) start at a lower dose because of elevated neutropenia risk.

The specification also describes a tumor-microenvironment mechanism (reduction of hypoxia/CAIX staining, increased microvessel density and small-molecule perfusion) supporting a "vascularity" aspect, and reports the Phase 3 NAPOLI-1 design (Arm A: MM-398 120 mg/m² q3w monotherapy; Arm B: 5-FU/LV control; Arm C: MM-398 80 mg/m² + 5-FU 2400 mg/m² + LV q2w).

---

4. Plain-language overview of the independent claims

Caveat — please note: the Google Patents text supplied did not include a verbatim claims section. The overviews below are drawn from the patent's Summary of the Invention (which is organized around the independent aspects) and the Detailed Description; the granted claim language should be confirmed against USPTO Patent Center full-text before being relied on for any legal purpose.

Claim family 1 — Liposomal irinotecan monotherapy regimen.
A method of treating pancreatic cancer in a patient by administering an effective amount of liposomal irinotecan, in 3-week cycles, where on Day 1 of each cycle the dose is 120 mg/m² — except that a patient homozygous for UGT1A1*28 receives 80 mg/m² on Day 1 of Cycle 1. In an embodiment, that patient's dose is escalated in 20 mg/m² increments after one cycle, up to a maximum of 120 mg/m².

Claim family 2 — Triple combination regimen (nal-IRI + 5-FU + leucovorin).
A method of treating pancreatic cancer by co-administering effective amounts of liposomal irinotecan, 5-FU and leucovorin in 2-week cycles, wherein per cycle:

• (a) liposomal irinotecan 80 mg/m² on Day 1 for non-UGT1A128-homozygotes; for UGT1A128 homozygotes, 60 mg/m² in Cycle 1 and 60–80 mg/m² (e.g., 60, 70 or 80 mg/m²) thereafter (embodiment: increased to 80 mg/m² after Cycle 1);
• (b) 5-FU at 2400 mg/m²; and
• (c) leucovorin at 200 mg/m² (l-form/levoleucovorin) or 400 mg/m² (l+d racemic).

Dependent embodiments add the sequencing/administration specifics: nal-IRI before leucovorin, leucovorin before 5-FU; nal-IRI IV over 90 minutes; 5-FU IV over 46 hours; leucovorin IV over 30 minutes; and premedication with dexamethasone and/or a 5-HT3 antagonist.

Claim family 3 — Kit.
A kit for treating pancreatic cancer comprising a dose of liposomal irinotecan (in another aspect, doses of all three of liposomal irinotecan, 5-FU and leucovorin) plus instructions for use according to the regimens described.

Claim family 4 — Formulation/use-type claim.
A formulation of liposomal irinotecan adapted for co-administration with 5-FU and leucovorin in at least one 2-week cycle, with the same dose elements (a)–(c) recited in claim family 2.

Claim family 5 — Tumor-vascularity aspect.
A method of improving chemotherapy outcomes by increasing tumor vascularity, comprising administering to a tumor-bearing patient an amount of irinotecan sucrose octasulfate salt liposome injection effective to increase tumor vascularity, and concomitantly administering an effective amount of a different (non-irinotecan) chemotherapy agent. A mirror "product-for-use" aspect is also disclosed: irinotecan sucrose octasulfate salt liposome injection for such concomitant administration.

Therapeutic outcome language: dependent/specification embodiments frame positive outcomes as pCR, CR, PR or SD, with the combination asserted to produce therapeutic synergy (Corbett log₁₀ cell-kill framework).

(Minor note for completeness: the Summary section contains a typographical error — "and affective amount of liposomal irinotecan" — which plainly should read "an effective amount." I mention it only so the text is not misquoted.)

---

5. Litigation — district court (confirmed)

• Case: Ipsen Biopharmaceuticals, Inc. et al. v. Conjupro Biotherapeutics, Inc. et al., D.N.J. No. 1:24-cv-04991 (RMB)(MJS), filed April 15, 2024; Judge Renee Marie Bumb, referred to Magistrate Matthew J. Skahill. Cause: 35 U.S.C. §271 (Hatch-Waxman / 505(b)(2) NDA No. 218923).
• US 9,364,473 is one of the asserted patents (patents-in-suit include 8,329,213; 9,339,497; 9,364,473; 9,452,162; 9,492,442; 9,717,724; 10,980,795; 11,369,597).
• Consolidations: 1:24-cv-08723 (patent 12,059,497) consolidated Oct 18, 2024; 1:25-cv-13647 (patent 12,364,691) consolidated Aug 21, 2025.
• Conjupro/CSPC filed counterclaims seeking declaratory judgment of non-infringement and invalidity of the '473 patent (among others). Reported status as of the paragraphfour/DrugPatentWatch entries: discovery.
• Plaintiff-side counsel: Sterlington LLP (Brian V. Slater); defense: Fox Rothschild (Gerard Norton).
• Google Patents also flags a Unified Patents litigation record for the New Jersey case: portal.unifiedpatents.com/litigation/New%20Jersey%20District%20Court/case/3%3A24-cv-04991.

6. PTAB / IPR — careful identifier note

• I found no IPR petition challenging US 9,364,473. The instituted PTAB proceeding in this family is IPR2025-00505, filed by Conjupro (RPIs: CSPC Pharmaceutical Group Ltd., CSPC Ouyi Pharmaceutical Co., Ltd., Conjupro Biotherapeutics, Inc.) against US 11,344,552 — a different patent. Apotex filed IPR 25-01531 against the same '552 patent; the Board joined it. Oral argument was scheduled May 19, 2026 (per paragraphfour).
• Identifier caution: in a PTAB document surfaced during searching, a "'936 patent" appears with an August 11, 2003 priority date and an Oliphant/Honig/Vinberg obviousness ground. That is not US 9,364,473 (whose priority date is June 13, 2012). Taking the identifiers literally, these are unrelated patents and must not be conflated.

7. CAFC 2026 dockets — explicit uncertainty

I could not identify any Federal Circuit proceeding in 2026 involving US 9,364,473. Targeted searches for a 2026 appellate docket tying "'936473" or "'473"/Ipsen/Onivyde together returned no results. Given that the D.N.J. case was reported in discovery as of the most recent update I retrieved, no final judgment appears to have ripened into a §1295 appeal on this patent. I do not have authoritative confirmation of any CAFC docket number, and I will not supply one speculatively. Note also that the Google Patents family page does reference global family litigation (patents.darts-ip.com/?family=48692681), which concerns the family, not an appellate docket.

---

Sources

• Patent full text (authoritative, supplied): https://patents.google.com/patent/US9364473/en
• FreePatentsOnline record (inventors, assignee, filing date, references): https://www.freepatentsonline.com/9364473.html
• D.N.J. litigation details/patents-in-suit: https://www.drugpatentwatch.com/p/litigation/casename/index.php?query=IPSEN+BIOPHARMACEUTICALS%2C+INC.+v.+CONJUPRO+BIOTHERAPEUTICS%2C+INC.%7C1%3A24-cv-04991
• P-IV/litigation tracker (case status, consolidations, IPR 25-00505 & 25-01531, OB expiries): https://paragraphfour.com/members/parryashford/activity/11735/
• Counterclaim/answer PDFs (D.N.J.): https://paragraphfour.com/wp-content/uploads/2024/04/njdc24cv8723A.pdf ; https://paragraphfour.com/wp-content/uploads/2024/04/njdc25cv13647A.pdf
• PTAB documents: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1557313](/patent/1557313)/ ; https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558590](/patent/1558590)/
• Orange Book patent stack: https://theraradar.com/drugs/onivyde/patents/

Bottom line: US 9,364,473 is Ipsen Biopharm's (originally Merrimack's) genotype-guided liposomal-irinotecan pancreatic-cancer dosing patent, expiring June 12, 2033, currently in Hatch-Waxman litigation in D.N.J. (1:24-cv-04991) against Conjupro/CSPC. No 2026 Federal Circuit docket for this patent is confirmed, and no IPR against this patent is confirmed — the live IPRs target US 11,344,552.

I'll verify the litigation record for US 9,364,473 against current sources before drafting.

Let me check for other defendants, other jurisdictions, and any appellate activity.

Known Litigation Involving U.S. Patent No. 9,364,473

Bottom line: Exactly one litigated patent family asserts US 9,364,473 — the Hatch-Waxman / § 271(e)(2) action brought by Ipsen against Conjupro Biotherapeutics and its CSPC affiliates in the District of New Jersey. That single lead case (1:24-cv-04991) has since absorbed two later-filed, consolidated complaints. There is no confirmed Federal Circuit appeal, no ITC action, and no second defendant group litigating this patent. I found no litigation naming the '473 outside the United States.

A note on the date field: the task header gives April 26, 2026, while the previously generated sections and system context are stamped 2026-09-28, and one retrieved source self-identifies as "Last Updated: August 3, 2026." Nothing in the case list turns on the difference; I flag it only so the timestamps aren't mistaken for a stale record.

---

1. Case list

Appellate: no Federal Circuit proceeding naming US 9,364,473 is confirmed. The district case remains in discovery, so no appealable final judgment has ripened. (Consistent with the earlier section's explicit-uncertainty finding — I am not supplying a CAFC docket number.)

---

2. Case No. 1 — the operative '473 suit

Ipsen Biopharmaceuticals, Inc. et al. v. Conjupro Biotherapeutics, Inc. et al., D.N.J. 1:24-cv-04991.

• Cause of action: 35 U.S.C. § 271 patent infringement — specifically § 271(e)(2)(A), infringement by submission of a 505(b)(2) NDA (No. 218923). Nature of suit on the civil cover sheet: "Patent – Abbreviated New Drug Application (ANDA)." Jury demand: none (bench case).
• Trigger: Conjupro's Paragraph IV notice letter under 21 U.S.C. § 355(b)(2)(A)(iv), received by Ipsen on or about March 5–7, 2024. Compliant filed within 45 days (April 15, 2024).
• The '473 patent's role: pleaded as Count V of the complaint — "Infringement by Conjupro/CSPC of U.S. Patent No. 9,364,473" — and attached as Exhibit C. Ipsen Biopharm Ltd. is pleaded as owner; the '473 is recited as issued June 14, 2016, and listed in the Orange Book against NDA No. 207793 (Onivyde®).
• Patents-in-suit in the lead complaint: U.S. 8,329,213; 9,339,497; 9,364,473; 9,452,162; 9,492,442; 9,717,724; 10,980,795; 11,369,597.
• Other patents in the notice letter but not initially asserted: 8,147,867; 10,456,360; 10,993,914.
• Relief sought as to the '473: a § 271(e)(4) order setting the Conjupro NDA's effective approval date no earlier than the '473's expiration (nominal 2033-06-12), injunctive relief, an exceptional-case finding under § 285, and attorneys' fees.
• Defendants' response: answered and asserted affirmative defenses of non-infringement, invalidity (¶¶101, 102, 103, 112), failure to state a claim, and prosecution history estoppel, and filed counterclaims seeking declaratory judgment that the '473 (among others) is not infringed and/or invalid and unenforceable. Those counterclaims are pleadings, not merits rulings.
• Service: waivers of service sent 2024-04-23; executed as to Conjupro (answer due 6/24/2024) and as to CSPC Pharmaceutical Group Limited and CSPC Ouyi (answer due 7/22/2024); responsive deadline extended by stipulation to 2024-07-08. A separate plea to the § 315(b) calculus was noted in the earlier section (waiver 2024-08-28 as a benchmark).
• Current status: discovery per the paragraphfour tracker's most recent update; no Markman order, no summary-judgment ruling, and no trial date surfaced in my searches.

Counsel of record (reported): plaintiff side — Brian V. Slater / Sterlington LLP (also Charles M. Lizza, Saul Ewing, on the D.N.J. filings); defense — Gerard P. Norton / Fox Rothschild LLP (with Paul W. Kalish, Jonathan J. Madara, Jonathan R. Lagarenne, and Joe G. Chen appearing).

---

3. Consolidated Cases Nos. 2 and 3 — why they matter to the '473

They do not add a new assertion of the '473 as a lead patent, but the '473 is swept into the DJ counterclaims and into the shared posture:

• No. 2 (1:24-cv-08723) — filed 2024-08-23, after issuance of U.S. 12,059,497 (~2024-08-13); consolidated 2024-10-18. Ipsen listed the '497 in the Orange Book against NDA 207793 on or about 2024-09-05.
• No. 3 (1:25-cv-13647) — filed 2025-07-22, after issuance of U.S. 12,364,691 on or about 2025-07-22; consolidated 2025-08-21. The '691 is described as non-Orange-Book (not listed at filing).
• In both, Conjupro/CSPC's counterclaims recite the same NDA No. 218923 P-IV certification covering the '473 and seek declarations with respect to the '473 among the "Counterclaim Patents-in-Suit."

---

4. Identifier discipline — near-miss "473" records I excluded

Per the strict literal-identifier rule, these surfaced in searching and are not US 9,364,473. Each is a different patent asserted in an unrelated case:

• U.S. 6,587,473 — asserted by IWS in a wireless-network (IEEE 802.11) infringement count; a 2013 demand-letter case.
• A Canadian "473 Patent" in a Federal Court (FCT-CF) pleading — a downhole drilling vibration-tool patent (invalidity based on Gilleylen and Eddison references). This is the same category of trap the earlier section flagged with the "'936 patent" with an August 11, 2003 priority date.
• U.S. 8,394,743 — a targeted-advertising patent in Consolidated Transaction Processing LLC v. Goodyear Tire & Rubber Co. (E.D. Tex. 4:25-cv-00019).
• U.S. 11,344,552 / 12,059,497 / 12,364,691 — same Ipsen family, different patents; the IPRs target them, not the '473.

None of these involve the Ipsen pancreatic-cancer patent.

---

5. Explicit gaps

1. No Federal Circuit docket. I could not confirm any 2026 CAFC proceeding in which US 9,364,473 is a patent at issue. With the district case in discovery, there is no final judgment to appeal.
2. No ITC action involving the '473 was found.
3. No non-U.S. litigation naming this patent was found. The Google Patents "family has litigation" flags point to (a) the D.N.J. case and (b) a Darts-ip global family aggregation link (patents.darts-ip.com/?family=48692681), which is a family-level database entry, not a confirmed foreign docket.
4. Docket prefix discrepancy, reported as-is: the case appears both as 1:24-cv-04991 (Justia's N.J. docket index, paragraphfour, DrugPatentWatch) and as 3:24-cv-04991 (Justia's parallel entry and the Google Patents sidebar). Both resolve to the same April 15, 2024 Ipsen v. Conjupro complaint. I have not "corrected" either prefix.
5. Current status is only as fresh as the aggregators. The most recent tracker update retrieved was dated August 3, 2026 ("Discovery"). For a current docket state, pull PACER directly.

---

Sources

• D.N.J. lead docket: https://dockets.justia.com/docket/new-jersey/njdce/1:2024cv04991/[546476](/patent/546476) and https://dockets.justia.com/docket/new-jersey/njdce/3:2024cv04991/[546433](/patent/546433)
• Case/patent/status detail: https://www.drugpatentwatch.com/p/litigation/casename/index.php?query=IPSEN+BIOPHARMACEUTICALS%2C+INC.+v.+CONJUPRO+BIOTHERAPEUTICS%2C+INC.%7C1%3A24-cv-04991
• Tracker (patents, expiry, counsel, consolidations, IPRs): https://paragraphfour.com/members/parryashford/activity/11735/
• Complaint Count V ('473): https://insight.rpxcorp.com/litigation_documents/15650730
• Answer / DJ counterclaims (No. 2 and No. 3): https://paragraphfour.com/wp-content/uploads/2024/04/njdc24cv8723A.pdf ; https://paragraphfour.com/wp-content/uploads/2024/04/njdc25cv13647A.pdf
• Unified Patents litigation record link (sidebar): https://portal.unifiedpatents.com/litigation/New%20Jersey%20District%20Court/case/3%3A24-cv-04991

Summary: US 9,364,473 is asserted in one pending Hatch-Waxman family — Ipsen Biopharmaceuticals, Inc. et al. v. Conjupro Biotherapeutics, Inc. et al., D.N.J. 1:24-cv-04991 (filed April 15, 2024, Judge Bumb), with two later complaints (1:24-cv-08723 and 1:25-cv-13647) consolidated into it. No outcome has been reached; the case is in discovery, the '473 remains valid and enforceable as issued, and the invalidity/non-infringement positions exist only as counterclaim pleadings. No PTAB proceeding and no Federal Circuit appeal on this patent is confirmed.

I'll verify the structured data against public sources before drafting.

Let me dig deeper on any IPR filings and the related litigation.

Let me check whether any IPR targets the '473 patent itself, and look at the litigation docket for validity challenges.

Proceedings overview

Zero. The structured USPTO Open Data Portal block for US 9,364,473 returns no AIA trial proceedings on file — no IPR, PGR, or CBM — and my independent web search surfaced none either. My verification did find AIA trials in the same patent family (the Onivyde/NALIRIFOX line), but not against the '473 itself. Bottom line for a defendant: you cannot point to a canceled claim of the '473 — the patent is untested at the PTAB and stands with all issued claims intact. That is a double-edged posture: no invalidating FWD exists to hand you leverage, but also no § 315(e)(2) estoppel has been created, so the full prior-art universe remains open to you in an IPR you file yourself.

One important correction to the framing: the '473 is not being asserted by a troll. It is an Orange Book method-of-use patent (NDA 207793, use code U-1856, nominal expiry 2033-06-12) owned by Ipsen Biopharm Ltd. and asserted against a generic filer — Conjupro Biotherapeutics / CSPC — in Hatch-Waxman litigation. Read the defensive-value analysis with that posture in mind.

---

Proceedings on US 9,364,473

None — no AIA trial proceeding has ever been instituted, denied, or filed against this patent

• Type: n/a
• Filed: n/a
• Status: No PTAB activity on file (USPTO ODP, as of most recent ingest; confirmed by independent web search on 2026-09-28). No IPR/PGR/CBM docket number exists to cite, and I will not invent one.
• Judge panel: n/a — no panel has ever been assigned.
• Petition grounds: n/a
• Institution decision: n/a
• Final Written Decision: n/a — no claim of the '473 has been canceled, confirmed, or construed by the Board.
• Settlement / termination: n/a
• Appeal: n/a — with no FWD, there is no PTAB appeal to the Federal Circuit on this patent. (I found no CAFC appeal involving the '473.)
• Defensive value: You get no free kill-shot, but you also inherit no estoppel. Every § 102/§ 103 ground a petitioner could have raised is still live for a fresh IPR — subject only to the § 315(b) one-year bar running from service of an infringement complaint on you or your privies.

---

Adjacent family proceedings — flagged, but NOT on the '473

These are the AIA trials I could verify in the Onivyde family. Neither challenges US 9,364,473. They are relevant only as intelligence on how the same patent owner and the same generic challenger are litigating.

IPR2025-00505 — CSPC Pharmaceutical Group Ltd., CSPC Ouyi Pharmaceutical Co., Ltd., and Conjupro Biotherapeutics, Inc. v. Ipsen Biopharm Ltd.

• Type: Inter Partes Review
• Filed: 2025-01-17 (one source, ipverse, lists the filing date as 2025-01-16 — I could not resolve the one-day discrepancy from public sources)
• Patent challenged: U.S. 11,344,552 (the "NALIRIFOX" first-line regimen patent) — not the '473
• Challenged claims: 1–15
• Status: "Final Written Decision"; institution decision dated 2025-08-13, termination date listed as 2026-08-12 (consistent with the statutory 1-year trial clock). I do not have the claim-level outcome of the FWD and will not guess at it.
• Judge panel: Not retrieved. Counsel of record per ipverse: Petitioner — Lukas Toft et al.; Respondent (Patent Owner) — James Glass et al.
• Petition grounds: Obviousness. Ground 1 — claims 1, 3–6, 8–14 obvious over Conroy (2011 NEJM FOLFIRINOX), the Conroy Protocol, the Conroy Appendix, and Mahaseth, in combination with Bayever (WO 2013/188586), Saif, Ko, and Cantore. Ground 2 — claims 2, 7, 15 obvious over the Ground 1 art plus Masi and Ginocchi. Core theory: substitute liposomal irinotecan for free irinotecan in FOLFIRINOX and drop oxaliplatin to 60 mg/m². (Source: Ex parte case record, IPR2025-00505.)
• Institution decision: Instituted. The Board also rejected Ipsen's real-party-in-interest challenge (Ipsen argued Cipla USA was an unnamed RPI); the panel declined to deny institution on that basis and deferred the issue to trial. (Source: institution decision text reproduced at ptacts.uspto.gov, Paper 6.)
• Settlement / termination: No settlement reported.
• Appeal: No Federal Circuit appeal identified on the public record as of 2026-09-28. If an FWD issued on/before 2026-08-12, a notice of appeal could have followed — unverified, treat as open.
• Defensive value for this patent (the '473): None directly. Different patent, different claims (the '552 claims the four-drug first-line regimen; the '473 claims the two-drug second-line 5-FU/LV regimen). No estoppel flows to the '473.

IPR2025-01531 — Apotex Inc. v. Ipsen Biopharm Ltd. (joined)

• Type: Inter Partes Review, filed with a motion for joinder to IPR2025-00505
• Filed: 2025-09-12
• Patent challenged: U.S. 11,344,552 — again, not the '473
• Status: Joined to IPR2025-00505. The Board granted the unopposed (after withdrawal of opposition) motion for joinder and directed that Apotex be added as a named petitioner under the existing schedule in IPR2025-00505.
• Petition grounds: Substantively identical to CSPC's petition; Apotex's expert (Dr. Dreicer) adopted CSPC's expert (Dr. Mark Ratain) opinions in toto, and Apotex committed to serve as a "silent understudy." Ipsen opposed on the ground that Apotex's last-business-day timing prejudiced Ipsen's ability to seek schedule modification.
• Sources: NYIPLA PTAB Committee materials quoting the joinder decision; Ipsen's discretionary-denial arguments.
• Defensive value for this patent (the '473): None directly. It signals, however, that generic challengers are pooling onto a single lead IPR rather than filing across the family — which is consistent with the absence of an IPR against the '473.

---

Strategic summary

Claim status of US 9,364,473. The entire issued claim set is UNTESTED at the PTAB. Nothing is canceled; nothing is confirmed; no claim has been construed by the Board. The nearest thing to an adjudicated validity event is the district court counterclaims of invalidity filed by Conjupro/CSPC in D.N.J. Civil Action No. 1:24-cv-04991 — and those are declaratory-judgment pleadings, not merits rulings. If someone hands you a demand letter citing the '473, the answer is not "that claim is dead"; it is "that claim has never been through an AIA trial, and I can be the first petitioner."

Estoppel landscape. There is no § 315(e)(2) estoppel attaching to the '473, because estoppel is proceeding-specific and no proceeding exists. Even the '552 IPR estoppel (which would bind CSPC, Conjupro, CSPC Ouyi, and, as a joined party, Apotex, and their privies) does not reach the '473. Practical consequence: any prior-art ground, including the Conroy FOLFIRINOX art and any printed publication that predates the '473's 2013-06-12 filing / 2012-06-13 priority date, is available in a fresh IPR. Note the family-specific trap: WO 2013/188586 ("Bayever"), the workhorse reference against the '552, is the '473's own family publication and therefore is not prior art to the '473 — do not build a petition on it.

Pattern signals. (1) Same-petitioner stacking: CSPC/Conjupro filed on the '552; Apotex then joined rather than run a parallel track. Three IPRs are not on file against this family — the challengers concentrated on the newest, narrowest, most vulnerable patent. (2) No defensive aggregator: the only "Unified Patents" reference in the record is a litigation data portal link in the Google Patents sidebar, not a Unified Patents IPR — Unified has not challenged this patent. (3) Patent owner posture: Ipsen is litigating aggressively (three consolidated N.D.J. complaints, 1:24-cv-04991, 1:24-cv-08723, 1:25-cv-13647), asserting a thicket of nine-plus patents including the '473, and has fought discretionary denial and RPI arguments at the Board on the '552. It has not yet had to defend the '473 at the PTAB.

Why the silence? An Orange Book method-of-use patent that has been listed since 2016-06-24 and asserted since 2024-04-15 with no IPR is worth pondering. The most likely explanation is strategic, not doctrinal: the '473 claims a dosage-regimen method (2-week cycles, 80 mg/m² liposomal irinotecan with 5-FU/LV, with UGT1A1*28 dose adjustments) that is harder to invalidate on the available art, whereas the '552's NALIRIFOX claims were already the subject of repeated § 103 rejections during prosecution — a roadmap for a petitioner. Absence of an IPR is not evidence of validity; here it looks like challenger triage.

---

Recommended next steps

If you are the defendant / prospective petitioner:

1. Treat the absence of PTAB art as an opportunity, not a shield. There is no FWD to rely on defensively, so your only path to PTAB leverage is to file. Confirm your § 315(b) clock: it runs one year from service of a complaint alleging infringement of the '473 on you or a privy/RPI. In the Conjupro action, service is documented as 2024-08-28 by waiver — a benchmark for how quickly that window closes.
2. Do not assume you can piggyback on IPR2025-00505. It challenges U.S. 11,344,552 only. § 315(e)(2) estoppel from that proceeding does not bar you from raising grounds against the '473, and equally gives you no cover.
3. Screen the '473 against the Conroy FOLFIRINOX art with care. The court record shows Ipsen already had to distinguish FOLFIRINOX-family art in the '552 prosecution, and the same Conroy reference predates the '473. But the '473's claims are directed to 5-FU/leucovorin plus liposomal irinotecan without oxaliplatin, at specific 2-week cycle doses with UGT1A1*28 stratification — the Bayever specification itself disclosed that Arm C regimen. The strongest § 102/§ 103 theory is likely to run through the '473's own pre-filing disclosures and the MM-398 clinical literature, not through the '552 dispute.
4. Confirm the '552 FWD outcome before budgeting. The '552 FWD was due on the statutory clock around 2026-08-12. Pull the actual FWD from PTAB E2E rather than relying on an aggregator's status field — if the '552 claims were canceled, that reinforces the challengers' triage logic and tells you which art the Board credited; if they were sustained, it tells you what the Board thinks of the FOLFIRINOX-substitution theory.
5. Watch for an appeal. If the '552 FWD went against Ipsen, expect a Federal Circuit notice of appeal; check CourtListener for the CAFC docket. A reversal or affirmance on the Bayever/Conroy combination will inform, though not control, your '473 strategy.

If you are the patent owner (Ipsen): the '473's clean PTAB record is an asset. The realistic risk is a follow-on petition by the same Conjupro/CSPC/Apotex group now that the '552 trial has run its course — note that a § 315(a)(1) declaratory-judgment counterclaim of invalidity, which Conjupro filed, can itself trigger the § 315(a)(1) bar (IPR filed before the DJ action) rather than § 315(b); if Conjupro/CSPC want an IPR against the '473, § 315(a)(1)/(b) timing should be assessed carefully.

Caveats I will not paper over: (a) the '552 FWD's claim-level disposition is unknown to me and I have not guessed it; (b) I could not retrieve the '473's PTAB docket because none exists, not because I failed to find it; (c) the IPR2025-00505 filing date carries a one-day public-source discrepancy (2025-01-16 vs. 2025-01-17); (d) no appellate disposition for any family proceeding is verified as of 2026-09-28.

I'll research the assignment chain for US 9,364,473. Let me run targeted searches.

Good — I have a lead on the Ipsen recording. Let me pin down reel/frame numbers for each link in the chain.

Let me pin down the reel/frame entries and the recording correspondents.

Assignment & Ownership Chain — US 9,364,473

Method note / scope of verification. My primary source was intended to be the USPTO Patent Assignment Search (https://assignmentcenter.uspto.gov/, legacy mirror https://assignment.uspto.gov/patent/index.html). That system is a JavaScript application that my retrieval tools could not query directly, and my operator runs hit the tool-step ceiling before I could reach an indexed Assignment Center record. Consequently I could not retrieve any reel/frame number or any recording correspondent for this patent directly from the USPTO record. Everything below is built from (a) the Google Patents legal-events/reassignment block in the authoritative patent text you supplied, and (b) corroborating public filings (Merrimack and Ipsen press releases and SEC filings). Where I am inferring rather than verifying, I say so. No assignment has been fabricated.

(Housekeeping, flagged once and not belabored: the previously generated sections were stamped 2026-09-28 while the task header says 2026-04-26. Nothing in the chain below turns on which date controls.)

---

Inventors

Seven named inventors, per the issued patent and the FPO record:

Employer determination. The applicant/assignee on the application is Merrimack Pharmaceuticals, Inc. (Cambridge, MA), and all seven inventors assigned their rights to Merrimack (two separate inventor-assignment recordings, below). Five of the seven list Massachusetts residences, consistent with Merrimack's Cambridge headquarters. I could not confirm from an authoritative, citable source in this pass which individual held which title at Merrimack, so I decline to assert specific roles; the securities record (Merrimack 8-K/10-K leadership disclosures) is the place to confirm.

Two patterns worth flagging:

1. Split inventor assignment — a real chain-of-title gap. The initial inventor assignment, recorded 2015-11-23, covers only six assignors (Niyikiza, Moyo, Bayever, Fitzgerald, Dhindsa, Laivins). The seventh inventor, Jaeyeon Kim, executed a separate assignment that was not recorded until 2016-05-03 — roughly eight months after the 2015-09-03 filing and only ~6 weeks before the patent issued on 2016-06-14. Until May 2016, Merrimack's title was incomplete on the face of the Assignment Center record. For standing purposes, verify whether the Kim assignment's execution date precedes the Nov 2015 recording.

2. The "all inventors depart within 12 months of filing" tell is NOT met on this record. I found no evidence of inventor departures tied to this application. The dislocations at Merrimack (the sale of the commercial business and transfer of >100 employees to Ipsen) occurred in April 2017 — about 19 months after this application's 2015-09-03 filing, and nearly five years after the 2012-06-13 priority date. Mark: not present / unclear. I am not treating the 2017 workforce transfer as a "post-filing inventor exodus" because the record does not tie named inventors to it.

---

Original assignee

Merrimack Pharmaceuticals, Inc., Cambridge, Massachusetts — a public (Nasdaq: MACK) biopharmaceutical company whose primary line of business was oncology drug development and commercialization (it also carried legacy diagnostics assets).

• Did it ship a product embodying the claims? Yes — this is the unusual and important part. The '473 claims dose regimens for MM-398 / nal-IRI, commercialized as ONIVYDE® (irinotecan liposome injection). Merrimack obtained FDA approval in October 2015, in combination with fluorouracil and leucovorin, for metastatic adenocarcinoma of the pancreas after gemcitabine-based therapy — i.e., the claimed subject matter. Merrimack built the manufacturing and sales infrastructure to sell it. This is therefore not a paper patent from a non-practicing originator.
• Current status: Wound down / in dissolution — not operating, and (on the record I retrieved) not a bankruptcy debtor. The sequence: April 3, 2017 — closed the asset sale to Ipsen for $575.0 million upfront plus up to $450 million in FDA-approval milestones (Merrimack 10-Q, Q3 2017: "On April 3, 2017, the Company closed the Asset Sale with Ipsen…"). May 30, 2019 — announced completion of a strategic review and a restructuring. March 27, 2024 — received the $225 million first-line pancreatic milestone from Ipsen, coupled with a Plan of Dissolution including a liquidating trust for any residual milestone receipts. Caveat: I could not confirm the precise date, if any, on which Merrimack ceased to exist as a legal entity; as of the most recent corporate communication I retrieved it was a milestone-contingent shell implementing dissolution.

---

Assignment timeline

Honest gap up front: I retrieved no reel/frame numbers and no correspondents of record for any of the seven links in this chain. The Google Patents legal-events block in the authoritative text lists party names, dates and conveyance types, but not reel/frame or the recording attorney. I flag the single reel/frame lead I did surface, clearly labelled as unverified for this patent.

---

• Executed 2015 (exact date not retrieved) / recorded 2015-11-23 — Reel not retrieved

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Eliel Bayever; Navreet Dhindsa; Jonathan Basil Fitzgerald; Peter Laivins; Victor Moyo; Clet Niyikiza
  • Assignee: Merrimack Pharmaceuticals, Inc.
  • Correspondent: not retrieved. No recurrence assessment possible.
  • Context: Ordinary employee-inventor assignment to the operating-company employer.
  • Note: Omits Jaeyeon Kim — see the next entry.

• Executed 2016 (exact date not retrieved) / recorded 2016-05-03 — Reel not retrieved

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Jaeyeon Kim
  • Assignee: Merrimack Pharmaceuticals, Inc.
  • Correspondent: not retrieved.
  • Context: Late-recorded curative assignment closing the chain-of-title gap left by the 2015-11-23 recording; recorded ~8 months post-filing.

• Executed 2015-12-22 (dating from the Collateral Agreement; recording date listed as 2015-12-28) / recorded 2015-12-28 — Reel not retrieved

  • Conveyance: Security Agreement / Security Interest (not an ownership transfer)
  • Assignor: Merrimack Pharmaceuticals, Inc.
  • Assignee: U.S. Bank National Association, as Collateral Agent
  • Correspondent: not retrieved.
  • Context: Securitization. Corroborated externally: the "Collateral Agreement dated as of December 22, 2015 among Merrimack Pharmaceuticals, Inc., as Issuer…" and a parallel Collateral Agreement naming U.S. Bank as Collateral Agent — i.e., the patent portfolio (including the '473) was pledged to secure Merrimack's notes.

• Recorded 2017-04-14 — Reel not retrieved

  • Conveyance: Release by Secured Party
  • Assignor: U.S. Bank National Association, as Collateral Agent
  • Assignee: Merrimack Pharmaceuticals, Inc.
  • Correspondent: not retrieved.
  • Context: Lien release clearing the 2015 collateral encumbrance to permit the clean transfer to Ipsen. Note the sequencing: deal closed 2017-04-03; the release recorded 2017-04-14, i.e., after closing but before the Ipsen assignment recording on 2017-05-01.

• Executed 2017-04-03 (asset-sale closing date) / recorded 2017-05-01 — Reel 042377/0696 (LEAD — see caveat below)

  • Conveyance: Assignment (Assignment of Assignors' Interest)
  • Assignor: Merrimack Pharmaceuticals, Inc.
  • Assignee: Ipsen Biopharm Ltd. (United Kingdom)
  • Correspondent: not retrieved.
  • Context: Asset acquisition, not a fire-sale to an aggregator — the $575M oncology asset purchase (ONIVYDE, manufacturing, and >100 employees) by a listed operating specialty-pharma group.
  • ⚠️ Reel/frame caveat: The reel 042377/0696 and the effective date 2017-04-03 surfaced on the Google Patents legal-events record for a sibling family patent (US 10,478,428 / US 2016/0074382), not from the '473's own record. Because the Merrimack→Ipsen transfer was a single bulk asset-sale assignment covering a whole patent schedule, the same reel/frame almost certainly covers the '473 — but I could not verify that, and I am not asserting it as the '473's reel/frame. Confirm in Assignment Center.

• Recorded 2017-05-17 — Reel 042522/0334 (same sibling-record caveat)

  • Conveyance: Change of Address of Assignee (not an ownership transfer)
  • Assignor: listed as Merrimack Pharmaceuticals, Inc. in the record
  • Assignee: Ipsen Biopharm Ltd.
  • Correspondent: not retrieved.
  • Context: Administrative only — housekeeping update of the assignee's address of record; no change in title.

One thing the Assignment Center will not show you

The 2017 deal was not a clean conveyance of all rights. Merrimack's 8-K (Jan 9, 2017) discloses an intellectual property license agreement at closing: Ipsen granted Merrimack an exclusive license back to a portion of the transferred patents relating to certain liposomal technology (outside Ipsen's commercial field) plus a non-exclusive license to the remainder, and Merrimack granted Ipsen a non-exclusive license to its remaining patents in Ipsen's field. The pre-existing Shire (ex-U.S.) and PharmaEngine (Taiwan) licences were also assigned to Ipsen as part of the deal. Licences are not recorded as assignments, so the Assignment Center chain will look like a simple, unencumbered Merrimack→Ipsen transfer when in fact the '473 family is royalty-encumbered and cross-licensed. Do not read the assignment record as a complete encumbrance picture.

---

Timeline diagram

timeline
    title Ownership of US 9364473
    2012 : Priority application filed
    2015 : Continuation filed by Merrimack
         : Six inventors assign to Merrimack
         : U S Bank takes security interest
         : Onivyde approved by FDA
    2016 : Kim assignment recorded
         : Patent issued June 14
    2017 : U S Bank releases security interest
         : Merrimack sells assets to Ipsen
         : Ipsen Biopharm records assignment
         : Ipsen records change of address
    2024 : Ipsen milestone paid to Merrimack
         : Merrimack Plan of Dissolution

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
There is no "IP / Licensing / Holdings / Ventures" entity anywhere in the chain. The only post-issuance ownership transfer is Merrimack (operating biopharma) → Ipsen Biopharm Ltd. (UK operating subsidiary of Ipsen, Euronext: IPN). The two U.S. Bank entries are a security interest (recorded 2015-12-28) and its release (recorded 2017-04-14) — encumbrances, not conveyances. No registered-agent address, no single-member LLC, no Delaware/Texas shell appears in the record.

2. Known asserter in the chain — NOT PRESENT.
Ipsen Biopharm Ltd. / Ipsen S.A. does not appear on the enumerated NPE lists (Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). Neither does Merrimack. Ipsen sues generic filers — as it is doing in D.N.J. — but that is ordinary Hatch-Waxman brand-side enforcement, not an NPE assertion model. The only "Unified Patents" reference in the entire public record for this patent is a litigation-data portal link in the Google Patents sidebar, which is a data aggregation citation, not evidence that Unified Patents has ever acquired or challenged this patent. (Nor has it challenged it: per the prior PTAB section, no IPR exists against the '473.)

3. Repeat correspondent across the chain — NOT ASSESSABLE (unclear, not "not present").
I could not retrieve a single correspondent of record for any link, so I cannot run the recurrence test that this signal turns on. I will not guess at a name. Two observations: (a) the signal requires recurrence across links, and a four-link operating-company chain would normally exhibit one IP counsel of record throughout — which is unremarkable, not suspicious; (b) to close this properly, pull each of the seven records in Assignment Center and compare the "correspondent" field. Verify via: https://assignmentcenter.uspto.gov/ (search patent number 9364473), then open each recorded assignment.

4. Cascading transfers — NOT PRESENT.
There is exactly one change of ownership in 11+ years: Merrimack → Ipsen, executed 2017-04-03, recorded 2017-05-01. No chained LLCs, no common-principal structure, no shared correspondent address. The "multiple consecutive assignments through chained shell entities in <24 months" fact pattern is absent.

5. Pre-litigation transfer — NOT PRESENT.
The first infringement suit naming the '473 is Ipsen v. Conjupro/CSPC, D.N.J. 1:24-cv-04991, filed 2024-04-15. The operative ownership transfer occurred ~7 years earlier (executed 2017-04-03 / recorded 2017-05-01). Nothing in the chain falls within six months of the complaint. The ownership position was settled long before the generic filed its Paragraph IV notice (~March 2024).

6. Bankruptcy fire-sale — NOT PRESENT (with an economic nuance).
No Chapter 7 or Chapter 11 filing by Merrimack appears in the record I retrieved. The patents did not pass through a bankruptcy estate. There is a distressed-divestiture flavor — Merrimack ran a strategic review and sold its crown-jewel asset — but the sale was a negotiated out-of-court asset sale (April 3, 2017) to a strategic acquirer, not a § 363 auction. Mark the bankruptcy tell not present; note the divestiture separately as context, not as a signal.

7. Privateering — NOT PRESENT.
Privateering is an operating company parking patents with an NPE to assert against competitors. The facts here are the inverse: an operating company sold a marketed product, its manufacturing facility and its workforce to another operating company, which then asserted the patents in its own name — the D.N.J. plaintiffs are Ipsen Biopharmaceuticals, Inc. and Ipsen Biopharm Ltd. There is no NPE front, no fee-sharing vehicle, no off-balance-sheet asserter.

8. Defensive aggregator (anti-NPE) — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents or OIN. It terminates at Ipsen Biopharm Ltd., an operating pharmaceutical company that sells ONIVYDE® — a product practising the claims. The patent is live and asserted, not neutralized.

---

Verdict

Operating-company assertion.

Justification. The ownership chain is a short, conventional, operating-company lineage — inventors → Merrimack Pharmaceuticals, Inc. → Ipsen Biopharm Ltd. — with exactly one change of ownership (executed 2017-04-03, recorded 2017-05-01, reel 042377/0696 per the sibling family record — verify) and two non-ownership U.S. Bank entries (security interest recorded 2015-12-28; release recorded 2017-04-14). The current assignee is the UK operating subsidiary of a listed specialty-pharma group whose marketed product, ONIVYDE® (irinotecan liposome injection), embodies the claimed methods, and the assertion in D.N.J. 1:24-cv-04991 (filed 2024-04-15) is against an actual generic/505(b)(2) challenger (Conjupro Biotherapeutics / CSPC), with two later complaints consolidated into it. None of the eight NPE signals registers as present; the one signal I cannot score (repeat correspondent) is unscored because I retrieved no correspondent data, not because anything suspicious surfaced.

Verification links:

• USPTO Assignment Center (search patent number 9364473): https://assignmentcenter.uspto.gov/
• Legacy Assignment Search mirror: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events for the '473: https://patents.google.com/patent/US9364473/en

Residual uncertainties I will not paper over: (a) no reel/frame numbers were independently verified for the '473 — the 042377/0696 and 042522/0334 figures come from a sibling family patent's record and are a lead only; (b) no recording correspondent was identified for any link, so signal 3 is genuinely unresolved; (c) the exact execution dates of the 2015-11-23 and 2016-05-03 inventor assignments were not retrieved; (d) Merrimack's precise entity-status date (dissolved vs. in dissolution) is unconfirmed.

I'll research the prior art references cited on the face of US 9,364,473. Let me run several targeted searches.

Let me pull the actual cited references from patent databases directly.

I have the core US patent references. Let me retrieve the complete front-page reference list, including foreign documents and non-patent literature.

Prior Art Cited Against / On the Face of US 9,364,473 ("the '473 patent")

Verification and scope note (read first)

I searched for the literal patent number 9364473 and confirmed it resolves to US 9,364,473 B2 — "Methods for treating pancreatic cancer using combination therapies comprising liposomal irinotecan," Merrimack Pharmaceuticals (now Ipsen Biopharm Ltd.) — the same patent analyzed in the previously generated sections. No similar-number results were substituted.

Two limitations on what follows, stated up front because they affect how much weight you can put on it:

1. The authoritative full text supplied for this analysis did not include the patent's front-page "References Cited" section. I retrieved the cited-reference list from the FreePatentsOnline record for US 9,364,473 (https://www.freepatentsonline.com/9364473.html). That list was truncated in the retrieved page, so the US-patent portion below is near-complete but the "Foreign Patent Documents" and "Other References" (non-patent literature) sections were not retrievable. I will not invent entries for those sections.
2. The granted claim language is not verbatim in the supplied text (flagged in the earlier summary). I therefore map each reference to the claim families identified previously rather than to specific claim numbers, and I flag where claim numbering is unverified.

Minor housekeeping: the task header states "Current Date: April 26, 2026," while the supplied patent record and system context are dated 2026-09-28. I proceeded on the supplied record; nothing in this answer turns on the difference.

---

A. The front-page "References Cited" — US patent documents

All entries below are from the FPO record for US 9,364,473 (front page). Dates are the listed publication dates (for patent applications) or issue dates (for granted patents); where the reference also has an earlier priority date I note it. The '473 patent's effective priority date is 2012-06-13 (prov. 61/659,211), PCT filed 2013-06-12.

The list continues past item 20 in the source but was cut off on retrieval, and I could not retrieve the foreign-document or NPL subsections. Treat the US list as "substantially complete as retrieved" and the foreign/NPL portions as unknown to me.

---

B. Which claims could each reference potentially anticipate under § 102

Key analytical point: Under § 102 a reference anticipates only if it discloses every limitation of the claim. The '473's independent claims are dosing-regimen method claims — q3w liposomal irinotecan 120 mg/m² (80 mg/m² for UGT1A128 homozygotes), or q2w liposomal irinotecan 80/60 mg/m² + 5-FU 2400 mg/m² + leucovorin 200/400 mg/m², with specified infusion times and sequencing. *The cited references are almost entirely liposome-formulation and background art; none of them, on its face, discloses those dosing regimens, sequencing, or the UGT1A128 stratification.* So the honest § 102 answer is: no cited reference appears to individually anticipate the '473's regimen claims; several are relevant only as § 103 secondary/background art. The references with the closest § 102 nexus are addressed individually below.

References with genuine (but limited) § 102 relevance

US 8,147,867 B2 — Hong et al., "Liposomes useful for drug delivery" (issued 2012-04-03; priority 2004-05-03).

• Description: Claims the PEGylated, sucrose-octasulfate-loaded liposomal camptothecin/irinotecan platform — i.e., the MM-398 composition — that the '473 expressly incorporates by reference.
• Potential § 102 target: The composition/formulation claim family (claimed as "a formulation of liposomal irinotecan for co-administration with 5-FU and leucovorin"). It would anticipate a bare composition claim to the liposomal irinotecan (sucrose octasulfate salt) formulation. However, because the '473 formulation claim additionally requires the co-administration use in a 2-week cycle with the recited 5-FU/leucovorin doses, Hong does not supply those missing limitations — so it does not anticipate the claim as a whole. Realistically it is a § 103 reference for the "irinotecan sucrose octasulfate salt liposome injection" limitation, not a § 102 killer.
• Caveat: I could not verify from the retrieved text whether the '473 claim family 4 is drafted as a true composition claim or a use-limited/formulation-for-use claim; if it is use-limited (as the Summary suggests), Hong's § 102 value drops to essentially zero and it is prior art of record only.

US 6,794,370 B2 — Achterrath, "Method for treating cancer using camptothecin derivatives and 5-fluorouracil" (issued 2004-09-21).

• Description: Method claims for treating cancer with a camptothecin derivative (e.g., irinotecan/CPT-11) combined with 5-fluorouracil — the genus that embraces FOLFIRI-type therapy.
• Potential § 102 target: The triple-combination claim family — but only at a genus level, and only incompletely. Achterrath discloses conventional (free) irinotecan + 5-FU (optionally with folinic acid/leucovorin), not liposomal irinotecan, not the q2w schedule, not the 80 mg/m² / 2400 mg/m² / 200–400 mg/m² doses, and not UGT1A1*28 stratification. It therefore cannot anticipate the '473 combination claims under § 102; its relevance is as § 103 background showing irinotecan + 5-FU co-administration was known.

US 8,067,432 (Anderson, 2011-11-29); US 7,060,828 (Madden, 2006-06-13); US 6,465,008 (Slater, 2002-10-15); US 7,829,113 (Okada, 2010-11-09); US 7,846,473 (Yoshino, 2010-12-07).

• Description: Each discloses a liposomal camptothecin/irinotecan or irinotecan formulation or liposome composition.
• Potential § 102 target: At most the compositional limitations (a liposomal irinotecan formulation). None discloses the pancreatic-cancer indication, the dosing regimen, or the 5-FU/leucovorin combination. No § 102 anticipation of any '473 claim; § 103 background on the state of the liposomal-irinotecan art only.

References that are NOT prior art to the '473 (flag this explicitly)

US 2015/0182521 A1 — Bayever et al. (published 2015-07-02).

• Why it is not prior art: This is the '473's own family publication (same inventors/applicant, US counterpart of WO 2013/188586). It published after the '473's 2012-06-13 priority date and is the applicant's own work. It is listed as a "US Patent Reference" for related-application/background purposes, not as § 102 art. Do not build a § 102 case on it.
• Consequence for the file: This is the same trap flagged in the earlier litigation summary — WO 2013/188586 ("Bayever") is the workhorse reference against sibling patents in the family but is not available against the '473 itself.

The remaining Hong-family documents (US 2015/0182460; US 8,992,970; US 2014/0154298; US 8,703,181; US 2013/0122081; US 8,329,213; US 2012/0171283; US 2007/0116753) and the delivery-vehicle documents (US 6,720,001; US 2003/0138481; US 2002/0192275; and the truncated US 2002/0146450).

• These disclose liposomal-delivery and emulsion technology. Even the earliest published ones (US 2007/0116753; US 6,720,001; US 2003/0138481; US 2002/0192275) predate the priority date and are technically available art, but they disclose only liposome/vehicle technology — no '473 claim limitation about pancreatic-cancer dosing. Under § 102 they anticipate nothing in the '473's claim set; under § 103 they are cumulative to Hong.

---

C. What the cited-reference list tells you (and what it doesn't)

1. The '473's cited art is overwhelmingly composition/formulation art, not method-of-use art. That is consistent with the patent being a dosing-regimen patent whose novelty resides in the regimen, not the drug product. The examiner's cited references were not used to reject the regimen claims on § 102 — they situate the MM-398 formulation and irinotecan/5-FU combinations in the art.
2. The most § 102-relevant citations are Hong (US 8,147,867 et al.) for the liposomal-irinotecan composition and Achterrath (US 6,794,370) for irinotecan + 5-FU, and even those fall short of any full claim because they omit the specific doses, schedules, sequencing, pancreatic-cancer indication, and (critically) the UGT1A1*28 genotype stratification.
3. The references that a real § 102/§ 103 challenge would most likely rest on are not on this list — they are the applicant's own pre-priority clinical disclosures (e.g., MM-398/PEP02 phase I–II reports; the PEP02 corporate webpage dated Aug. 4, 2011, which appears in the reference list of a sibling patent, US 11,369,597, not necessarily the '473's) and the NAPOLI-1 study design. I could not confirm which of these (if any) are cited on the '473's own face, because the "Other References" section was not retrievable. I will not assert that they are cited.

---

D. Precision caveats I will not paper over

• Source of the reference list: FreePatentsOnline (https://www.freepatentsonline.com/9364473.html). The supplied Google Patents full text did not include the "References Cited" block, so I could not cross-corroborate every date against a second database before running out of retrieval budget.
• Incomplete list: The FPO US-reference list was truncated at item 20, and the Foreign Patent Documents and Other References (NPL) subsections were not retrieved. My list is therefore not certified complete.
• Claim mapping is family-level, not claim-number-specific, because the granted claims were not in the supplied text. Any § 102 statement above should be re-checked against the issued claim language before being relied on for a legal purpose.
• § 102 framework: The '473 has a 2012-06-13 effective priority date. Whether pre-AIA § 102 or AIA § 102(a)(1)/(a)(2) governs a given claim depends on that claim's entitlement to the pre-March-16-2013 provisionals/PCT. I have not resolved that priority question and flag it as a threshold issue for any invalidity position.
• No § 102 anticipation is asserted by me. Based on the retrieved art, no cited reference anticipates any '473 claim in full; the closest are Hong (composition) and Achterrath (irinotecan + 5-FU combination), both incomplete against the regimen claims.

Bottom line: On the face of US 9,364,473 the cited prior art is liposome-formulation and background art, of which US 8,147,867 (Hong et al.) is the most relevant to the liposomal-irinotecan composition and US 6,794,370 (Achterrath) the most relevant to the irinotecan + 5-FU combination. None of the cited references discloses the '473's claimed dosing regimens (including the UGT1A1*28-based dose adjustment), so none anticipates the claims under § 102 as a matter of the retrieved record — and the family's own WO 2013/188586 / US 2015/0182521 (Bayever) is not prior art to the '473.

I'll gather the prior art references cited on this patent's record before building the § 103 analysis.

Let me pull the patent's cited non-patent literature and the granted claims, which are essential for a claim-by-claim § 103 analysis.

Obviousness Analysis — U.S. Patent No. 9,364,473 under 35 U.S.C. § 103

Analysis date: per task instruction, April 26, 2026. (Flag: the previously generated sections of this analysis are timestamped 2026‑09‑28. The discrepancy is in the task metadata, not in the substance; nothing below depends on which date controls, since no § 103-relevant event falls between them.)

Cross-reference note: The prior sections established the bibliographic record, the D.N.J. litigation, and the (empty) PTAB history of the '473. I do not repeat those. Two of their findings drive this analysis and I build on them directly: (i) no IPR has ever been filed against the '473, so there is no Board construction or FWD estoppel to work from; and (ii) the live IPR2025‑00505/‑01531 art is directed only at U.S. 11,344,552. That matters because the '552 petition's workhorse reference — WO 2013/188586 ("Bayever") — is not available against the '473 at all (explained in § 5 below). Any § 103 theory against the '473 must be rebuilt from pre‑2012 art.

---

1. Legal frame

POSA definition (proposed): a clinician-scientist with an M.D. or Ph.D. and ≥5 years in medical oncology/GI oncology, familiar with irinotecan's pharmacology and pharmacogenetics and with the FOLFIRI/FOLFIRINOX literature; optionally supported by a formulation scientist for the liposomal limitations.

---

2. The claim set actually at issue

This differs materially from the patent's own Summary section, and the difference is outcome-determinative. The granted claims are all directed to the UGT1A1*28‑homozygous, gemcitabine‑pretreated subpopulation — the Section‑Summary's broad "120 mg/m² q3w monotherapy" aspect is not in the granted claim set as it appears on the public record.

Claim 1 (independent; verbatim):

"A method of treating metastatic adenocarcinoma of the pancreas in a human patient in need thereof, the patient being homozygous for the UGT1A1*28 allele, the method comprising intravenously administering to the patient once in a two week treatment cycle 60 mg/m² of the antineoplastic agent MM‑398 liposomal irinotecan in combination with 200 mg/m² of the (l)-form of leucovorin or 400 mg/m² of the (l+d) racemic form of leucovorin, and 2400 mg/m² of the antineoplastic agent 5‑fluorouracil to treat the metastatic adenocarcinoma of the pancreas in the patient, wherein the patient has previously been treated with gemcitabine and no other antineoplastic agent is administered to the human patient in the two week treatment cycle…"

Claim 14 (independent): same population, "50 mg/m² or 60 mg/m² of MM‑398 … over 90 minutes, followed by 200 mg/m² of the (l) form of leucovorin over 30 minutes, followed by 2400 mg/m² of 5‑fluorouracil over 46 hours … day 1 of a two week treatment cycle."

Claim 17 (independent, closed): "consisting of once every two weeks: premedicating with an anti‑emetic followed by … a single infusion of 60 mg/m² of MM‑398 … 200 mg/m² of the (l)‑form of leucovorin, and … 2400 mg/m² of 5‑fluorouracil."

Claim 28 (independent): three sequenced two‑week cycles with de‑escalation to 50 mg/m² on Grade 3/4 AE or escalation to 60/80 mg/m² if none.

Claims 2–13 and 15–29 are sequencing/administration/leucovorin‑form/dose‑titration dependents.

Two structural observations that shape the analysis:

1. Every independent claim is a dose‑regimen + patient‑selection claim. There is no composition limitation; the liposome itself is not claimed. That collapses the case to (a) is the 5‑FU/LV/irinotecan regimen a known second‑line pancreatic regimen? (b) is substituting liposomal for free irinotecan motivated? (c) is genotype‑stratified dose reduction routine?
2. The UGT1A1*28‑homozygous limitation and the 60 (or 50) mg/m² figure are the two elements carrying all the patentable weight. Everything else is admittedly old — the '473's own Background concedes FOLFIRI and FOLFIRINOX "are also used to treat some pancreatic cancers," and concedes gemcitabine as standard of care.

---

3. The prior art of record (the "Prior Art" section of the patent)

These are the references listed on the face of the '473 (US patent references, foreign references, and other references), with their art status against a 2012‑06‑13 effective date:

A. Patents

B. Non‑patent literature (of record)

C. Art beyond the four corners of the patent that a POSA would have had

• Conroy, T., et al., "FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer," N Engl J Med 364(19):1817‑1825 (2011) — the 5‑FU 2400 mg/m² 46‑h + leucovorin 400 mg/m² q2w backbone in metastatic pancreatic cancer.
• Iyer / Ando / Innocenti & Ratain UGT1A1 line (1998‑2006) and the O'Dwyer, J Clin Oncol 24(28):4534‑8 (2006) review — the genotype‑toxicity relationship and the clinical case for genotype‑directed dosing.
• The "Means and method for improving cancer treatment based on UGT1A1" patent family (e.g., JP 2005‑505526 A / WO 03/015793) — expressly teaches modifying irinotecan treatment based on UGT1A1 genotype. This is third‑party and squarely discloses the genotype‑stratified dose‑adjustment element.
• (Lead to verify): the ClinicalTrials.gov record for NAPOLI‑1 (NCT01494506), first posted in late 2011, which on its face recites the trial's Arms A/B/C and the UGT1A1*28 homozygous dose reduction. If the posting date is confirmed pre‑2012‑06‑13, this is a potentially dispositive § 102(a)(1) / § 103 printed publication. I have not verified the posting date in this pass and will not assert it as fact.

---

4. Ground 1 — Claim 1 (and claims 2–13, 15–16, 19–27)

The combination

Yoo 2009 (or Zaniboni 2012 / Kozuch 2001) + WO 03/030864 and the Slater/Madden/Anderson/Okada liposome art + Sadzuka 1998 + Achterrath US 6,403,569 + the Camptosar 2010 label and the UGT1A1 literature.

Element‑by‑element

4.2 Why 60 mg/m² is not a patentable dose

The Camptosar label's own instruction is to reduce the irinotecan starting dose by one dose level in known UGT1A1 7/7 patients (e.g., 125 → 100 mg/m² weekly; 350 → 300 mg/m² q3w) — i.e., a ~20–30 % reduction. The '473 specification itself reports the MM‑398 combination MTD as 80 mg/m², and the NAPOLI‑1 Arm C protocol reduced the homozygous‑patient starting dose from 80 to 60 mg/m² — a single step. Choosing 60 from a known 80 mg/m² MTD, on a label instruction to step down one level, is the paradigmatic "result‑effective variable … within the routine skill of the artisan" case. Claim 14's alternative "50 mg/m²" is simply the next step down, and it appears in the same protocol as the de‑escalation dose.

4.3 The "consisting of" claims (17, 28)

Claim 17's closed transition plus claim 1's "no other antineoplastic agent" are negative limitations that exclude only what the prior art already omitted — Yoo's FOLFIRI.3 arm contains no fourth agent. Closed language does not confer patentability on a combination whose constituent elements are all disclosed; it only narrows the claim. Claim 28's toxicity‑driven titration (60 → 80 on no Grade 3/4 AE; → 50 on Grade 3/4) is the Camptosar label's dose‑modification algorithm applied to the substituted formulation — routine optimization under KSR.

---

5. Ground 2 — Claim 14 (sequential, timed infusion)

Claim 14 adds: 90‑min MM‑398 infusion → 30‑min leucovorin → 46‑h 5‑FU, q2w.

• 46‑hour 5‑FU / 30‑minute LV: directly from Yoo 2009, Conroy 2011 and the de Gramont schedule.
• 90‑minute MM‑398 infusion: the MM‑398 clinical protocol infusion time; the '473 specification describes it as the ordinary administration mode and it appears in the published PEP0201/PEP0203 dose‑escalation reports.
• Order (irinotecan → LV → 5‑FU): the standard FOLFIRI order and the order the art teaches for leucovorin potentiation. The Board in IPR2025‑00505 credited precisely this reasoning when it found a reasonable likelihood that an analogous sequence limitation was obvious, and expressly reasoned that a POSA would look to the FOLFOXIRI/FOLFIRI sequencing literature.

---

6. Exclusions and traps — do not build a petition on these

This is where a '473 challenge most often goes wrong, because practitioners carry over the '552 playbook.

1. WO 2013/188586 ("Bayever") is not prior art. It published 2013‑12‑19, after the 2012‑06‑13 priority date; it is the '473's own family publication; and the '473 specification's summary text is verbatim the Bayever disclosure. It is disqualified under AIA § 102(b)(2)(C) (common ownership) and/or § 102(b)(1)(A) (inventor's own disclosure), and under pre‑AIA § 103(c). Note the irony: the Board's '552 institution decision reasoned that "Bayever taught that liposomal irinotecan was superior to free irinotecan," which is exactly the substitution motivation one needs here — and it is unavailable here.
2. US 2015/0182521 (Bayever et al.) — the '473's own pre‑grant publication. Not prior art.
3. US 8,147,867 / 8,329,213 / 8,703,181 / 8,992,970 (Hong et al.) — the MM‑398 liposome patents. They appear to share the Merrimack assignee of the '473. If commonly owned at the relevant time, they are disqualified as § 103 art (pre‑AIA § 103(c); AIA § 102(b)(2)(C)). Verify the chain of title before relying on them. The good news: WO 03/030864, US 7,060,828, US 6,465,008, US 7,826,113, US 7,846,473 and US 8,067,432 are third‑party art and supply the same liposomal‑irinotecan teaching.
4. Ko, et al., Br J Cancer 109(4):920‑25 (2013) and Saif 2014 and Mahaseth 2013 post‑date the priority date and are not prior art to the '473 (unlike in the '552 case). Ko's 2011 ASCO poster/abstract (June 2011) is available and teaches MM‑398 monotherapy 120 mg/m² q3w in gemcitabine‑refractory pancreatic cancer — useful, but it teaches monotherapy and a q3w schedule, so it does not by itself give the q2w 60 mg/m² combination.
5. "Conroy Appendix"/"Conroy Protocol" (the NEJM supplementary protocol, 2011) — available in principle, but confirm the public‑accessibility date for § 102(a)(1) purposes.

---

7. Motivation to combine — the KSR rationales, enumerated

1. Substitution of one known element for another, with predictable results. Liposomal irinotecan is irinotecan in a known delivery vehicle (WO 03/030864; Slater; Madden). The '473 specification itself asserts the substitution changes only PK/tolerability, not the mechanism of action: "once irinotecan leaves the liposomes, it is catabolized by the same metabolic pathways as conventional (free) irinotecan." That is the patentee's own admission that the substitution yields a predictable result.
2. Known technique to improve a similar therapy. Sadzuka 1998 and the liposome patents teach that encapsulation reduces CPT‑11 toxicity — which is precisely the clinical problem that caps FOLFIRI dosing in gemcitabine‑refractory pancreatic patients (Yoo 2009 reports the dose‑limiting neutropenia/diarrhea of free irinotecan).
3. Design incentive / market demand. The '473's Background concedes there were "few approved treatment options" for gemcitabine‑refractory pancreatic cancer and no durable therapy. A POSA had an explicit incentive to improve second‑line FOLFIRI by reducing its toxicity.
4. "Obvious to try" with a finite number of identified, predictable solutions. Once MM‑398 was in Phase I/II in pancreatic cancer (Ko 2011), the candidate regimens were a small, known set: FOLFIRI, FOLFOX, and combinations thereof, at standard q2w FOLFIRI doses. That is the KSR "finite number of identified, predictable solutions" situation.
5. Genotype‑directed dosing was the known, mandated practice. The 2010 Camptosar label, the 2005 FDA label change, O'Dwyer 2006 and the UGT1A1 "means and method" patent all put genotype‑stratified irinotecan dose reduction squarely in the POSA's toolkit. Patient selection on a known biomarker with a known dose response is a result‑effective variable, not an inventive contribution.
6. Routine optimization of the remaining variables. Infusion times (90 min / 30 min / 46 h), the l‑form/racemic leucovorin equivalence (conceded in the specification), anti‑emetic premedication (Munstedt 1999; Camptosar label), and toxicity‑triggered dose level changes (Camptosar dose‑modification table) are all optimization of parameters the art identified as variable.

Reasonable expectation of success: high. The art teaches (a) the regimen works in this population with free irinotecan, (b) the liposomal version is better tolerated, (c) the 5‑FU/LV backbone and schedule work in metastatic pancreatic cancer, and (d) how to titrate for genotype and toxicity. Nothing in the combination required a new mechanism or an unpredictable leap.

---

8. Anticipated rebuttal — secondary considerations

---

9. Confidence assessment and where the patent survives

Where the patent has a genuine defence. (a) If the patentee establishes an effective filing date of 2013‑03‑14 or earlier for all claims, then everything the challenger can muster is the 2009‑2012 art — the art is thinner than it looks, because the richest MM‑398‑specific references (Ko 2013, Saif 2014, Mahaseth 2013, Wainberg 2016) are all post‑date. (b) If the challenger relies on the Hong/Merrimack liposome patents and those are commonly owned, the liposomal‑irinotecan element rests on weaker third‑party art (WO 03/030864 is a formulation patent, not a pancreatic‑cancer method patent). (c) The genotype‑stratified limitation combined with a specific reduced dose is the one limitation that a tribunal might find non‑obvious if the patentee can show that the 7/7 population in a 5‑FU/LV doublet responds differently than the general population — a factual question that depends on data I have not seen.

---

10. Explicit uncertainties I am not papering over

1. I did not verify the Ziniboni 2012 and NAPOLI‑1 registry publication dates against the priority cutoff; both are flagged as needing verification and NAPOLI‑1 could be dispositive.
2. I did not confirm the chain of title for US 8,147,867 et al. The common‑ownership exclusion in § 6 is hypothesis‑based on the shared Merrimack assignee; confirm before relying on it either way.
3. The exact grant‑date claims were taken from the Google Patents claim text and corroborating secondary sources (DrugPatentWatch), not from a USPTO Patent Center certified copy. The claim set is internally consistent with the specification, but for any legal purpose pull the grant from Patent Center.
4. The PEP0203 Phase 1 combination data (MTD 80 mg/m²) appears in the '473 specification as the applicant's own work; I have not confirmed a pre‑2012 public disclosure of it, so I have not treated it as prior art — only as an admission of the 80 mg/m² MTD.
5. No PTAB or court has construed a single term of the '473. Every construction above (especially "consisting of," "no other antineoplastic agent," and "in need thereof") is unlitigated and could move the analysis; "treating" in particular was contested at the Board in the '552 matter and could be argued either way here.

Bottom line. The '473's granted claims are narrow, genotype‑locked, second‑line dosage‑regimen claims. On the art of record, the strongest § 103 case is Yoo 2009 + WO 03/030864 and the Slater/Madden/Anderson liposome art + Sadzuka 1998 + Achterrath + the Camptosar 2010 label and the UGT1A1 literature, with 60 mg/m² established as routine one‑step dose‑level reduction from the disclosed 80 mg/m² combination MTD. The fatal flaw in transplanting the '552 petition is that its centrepiece — Bayever WO 2013/188586 — is the '473's own family publication and cannot be used at all. A challenger who rebuilds from pre‑2012 third‑party art has a strong, though not certain, case; a challenger who copies the '552 petition will fail on the reference dates alone.