US 5591639

I'll search for authoritative information on this specific patent number and any 2026 CAFC docket activity.

Let me run a few more targeted searches to check for any 2026 litigation or CAFC activity and to confirm bibliographic details.

US Patent 5,591,639 — Summary

Bottom line on the "2026 CAFC docket" search: I found no Federal Circuit 2026 docket activity involving US 5,591,639, and none is plausible — the patent's anticipated expiration was 2014‑01‑07 and its legal status is "Expired – Lifetime." All enforcement activity I could locate for this patent is from ~2007. I flag this as a negative finding with the caveat that I am relying on web search rather than a logged‑in PACER/CAFC CM/ECF query.

Bibliographic data (from the patent text provided, which I treat as authoritative)

Chain of applications: this patent is a continuation of Ser. No. 07/896,797 (filed 1992‑06‑09, abandoned), itself a continuation of Ser. No. 07/339,615 (filed 1989‑04‑28, abandoned). A divisional, US 5,658,759, issued 1997‑08‑19 from app. 08/633,013. Family members include WO 1989/001036 A1, EP 0 323 997 B1, JP 2505268 B2, AT E88501 T1, DE 3880468 T2, GB 8717430 D0.

Abstract

Expression vectors containing the promoter, enhancer and substantially complete 5'-untranslated region including the first intron of the human cytomegalovirus major immediate early (hCMV‑MIE) gene; vectors in which the hCMV‑MIE DNA is linked directly to a heterologous coding sequence; host cells transfected therewith; a process for producing heterologous polypeptides; and use of the hCMV‑MIE DNA to express a heterologous gene.

Plain-language overview of the independent claims

Claim 1 — the specific plasmids. Claims three named plasmids as such: pCMGS, pEE6.hCMV and pHT.1. This is a product claim to tangible DNA constructs, not to a method. (Note the specification text references "pMT.1" when describing the TIMP‑containing intermediate, and later "pHT.1"; claim 1 recites pHT.1.)

Claim 2 — the generic expression vector. A recombinant expression vector comprising three hCMV‑MIE elements — (a) the promoter, (b) the enhancer, and (c) the complete 5' untranslated region including the first intron — operably linked to a heterologous coding sequence. The "complete 5' UTR including the first intron," linked directly to heterologous coding sequence, is the core of the invention (the specification stresses this yields high, gene‑independent translation).

Dependent claims (3–7), all referring back to claim 2:

• Claim 3 — adds a restriction site to facilitate insertion of the heterologous coding sequence.
• Claim 4 — the hCMV‑MIE promoter/enhancer/5'-UTR are linked directly to the heterologous coding sequence (no intervening sequence).
• Claim 5 — the vector further includes the hCMV‑MIE gene's translation initiation signal.
• Claim 6 — the initiation signal includes the sequence 5'-GTCACCGTCCTTGACACCATG-3'.
• Claim 7 — the initiation signal includes 5'-CCATGG-3' (an NcoI site); the specification explains this results from a single G→C change at position −1 that also better matches the Kozak consensus.

Scope note: Claim 1 is limited to three specific plasmids; claim 2 is the broad genus; claims 3–7 narrow claim 2 by adding features. Because claims 3–7 are all dependent on claim 2 (none on claim 1), the "linked directly" and "translation initiation signal" limitations only narrow the generic vector claim.

Known litigation / enforcement history

• D. Md. case 8:07‑cv‑03177 — flagged by Google Patents/Unified Patents as a US case involving this family (source: https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177).
• D. Del. case 1:07‑cv‑00467‑GMS, Lonza Group AG v. Northwest Biotherapeutics, Inc. — an AO‑120 report dated 7/30/2007 lists 5,591,639 among asserted patents (alongside 5,122,464; 5,658,759; 5,770,359; 5,827,739; 5,879,936; 5,891,693; 5,981,216).
• DrugPatentWatch lists 5,591,639 in connection with INTRON A (interferon alfa‑2b), BLA 103132, with a 2014‑09‑02 date (source: https://www.drugpatentwatch.com/p/biologics/patent/5591639).
• Patent listed as Expired – Lifetime; ownership traced Celltech → Alusuisse Holdings A.G. (1998) → Alusuisse Lonza Group AG / Lonza Group AG (2001) → Lonza Swiss Licences AG (2010/2014) → Lonza Licences AG + Healthcare Royalty Partners II LP (2013).

Explicit uncertainty / caveats

1. No 2026 CAFC appeal located. My searches of public sources returned no 2026 Federal Circuit docket for 5591639. Given the 2014 expiration and "Expired – Lifetime" status, a 2026 appeal of this patent is highly unlikely; I could not verify an authoritative CAFC docket either way.
2. Assignee naming is inconsistent across sources ("Aluguisse"/"Alusuisse," "Celltech Ltd."/"Celltech Therapeutics Limited"). I have reported the names literally rather than normalizing them.
3. Priority vs. filing date: the 1994‑09‑02 date is the filing date of this application; the operative priority is 1987‑07‑23. The Google Patents "Filing date 1994‑09‑02 / Priority date 1987‑07‑23" pairing should not be read as a 1994 priority.
4. Search coverage: I used open web search, not direct USPTO Patent Center/Public PAIR or PACER/CAFC CM/ECF queries. For a definitive prosecution‑history or docket record, those primary systems should be checked directly.

I'll research litigation involving US patent 5591639 using patent litigation sources.

Let me investigate the specific litigation referenced on the patent page and related litigation.

I found key cases. Let me dig deeper into these and look for others.

Litigation involving U.S. Patent No. 5,591,639 ("Recombinant DNA expression vectors")

I located litigation involving this patent. Note that US 5,591,639 (Celltech Ltd; priority 1987-07-23; granted 1997-01-07) is the U.S. sibling of WO 1989/01036 / EP 0323997 and belongs to the same Lonza/Celltech "GS system + hCMV-MIE promoter/intron" patent family as US 5,122,464, 5,658,759, 5,770,359, 5,827,739, 5,879,936, 5,891,693 and 5,981,216. The '639 patent expired on 2014-01-07 (anticipated expiration; status "Expired – Lifetime"), so no litigation post-2014 is possible.

I identified two related cases, both Lonza Group AG v. Northwest Biotherapeutics, Inc., involving the same patent family:

Case 1 — Delaware (first-filed)

• Plaintiff: Lonza Group AG
• Defendant: Northwest Biotherapeutics, Inc. ("NWBT")
• Jurisdiction: U.S. District Court for the District of Delaware
• Case No.: 1:07-cv-00467-GMS
• Filing date: July 27, 2007
• Patents asserted: U.S. 5,122,464; 5,591,639; 5,658,759; 5,770,359; 5,827,739; 5,879,936; 5,891,693; 5,981,216 (per the Form AO-120 Report on the Filing of an Action, docketed 7/30/2007, naming Celltech/Alusuisse holders)
• Outcome/status: Terminated. Lonza agreed to dismissal of its entire complaint, and NWBT's answer/counterclaims (Delaware Deceptive Trade Practices Act, unfair competition, tortious interference, abuse of process) were dismissed as well. Lonza then re-filed the same dispute in Maryland.
• Source: https://www.findforms.com/single_form.php/form/98659/ and NWBT press release, Dec. 20, 2007 (https://www.biospace.com/lonza-group-ltd-withdraws-majority-of-claims-in-patent-lawsuit-against-northwest-biotherapeutics)

Case 2 — Maryland (re-filed; the case flagged by Google Patents)

• Plaintiff: Lonza Group AG
• Defendant: Northwest Biotherapeutics, Inc.
• Jurisdiction: U.S. District Court for the District of Maryland
• Case No.: 8:07-cv-03177
• Filing date: 2007 (filed contemporaneously with the Delaware dismissal, i.e., late 2007; the docket is dated 2007)
• Patents asserted: Lonza's "gene expression patents" (the same family, which includes US 5,591,639)
• Outcome/current status: The last development I can confirm is the Amended Complaint filed December 13, 2007, which dropped all accused products except DCVax-Prostate (retaining related state-law claims). NWBT contested infringement, arguing it never used Lonza's expression system (an unrelated third-party manufacturer used a gene-expression system to make two batches of recombinant PSMA protein). I could not confirm from the sources retrieved what the final disposition of the Maryland case was (e.g., settlement, dismissal, or judgment). I am flagging this as unverified rather than asserting an outcome.
• Sources: Google Patents litigation link (https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177) and Dockets listing (https://dockets.justia.com/search?nos=830&...state=maryland, showing "Lonza Group AG v. Northwest Biotherapeutics, Inc., as 8:2007cv03177," cause of action 35 U.S.C. § 271).

Notes, caveats, and gaps

1. Google Patents records that the "family has litigation" (family ID 10621156) and links a single U.S. case: the Maryland District Court case 8:07-cv-03177 above. The Darts-IP "first worldwide family litigation" link points to the same family.
2. No CAFC appeal and no PTAB proceeding (IPR/PGR) involving US 5,591,639 were identified in my searches. Because the patent expired in 2014, an IPR would not have been available in any event.
3. I found no other defendants sued on US 5,591,639 specifically. Lonza/Celltech's related patent disputes with other parties (e.g., the Celltech–Genentech "Boss/Cabilly" antibody-expression dispute, and the MedImmune–Genentech Cabilly II case) concern different patents and should not be attributed to the '639 patent.
4. Verification recommended: The authoritative sources for a complete docket and final disposition are PACER (D. Md. 8:07-cv-03177; D. Del. 1:07-cv-00467-GMS) and the Unified Patents litigation portal. My search results establish the parties, jurisdictions, case numbers, and filing years but did not surface a final judgment or settlement document for the Maryland case.
5. Company-name note: the patent's ownership changed over time (Celltech Ltd → Alusuisse Holdings A.G. → Lonza Group AG → Lonza Swiss Licences AG / Lonza Licences AG / Healthcare Royalty Partners II, L.P.). The litigation was brought in the name of Lonza Group AG.

Bottom line: the only litigation I can identify naming US 5,591,639 is Lonza Group AG v. Northwest Biotherapeutics, Inc., filed first in D. Del. (1:07-cv-00467-GMS, July 27, 2007; dismissed/withdrawn) and re-filed in D. Md. (8:07-cv-03177, 2007; narrowed to DCvax-Prostate by the Dec. 13, 2007 amended complaint; final disposition not confirmed).

I'll verify the ODP result against public sources before reporting.

Let me check litigation and any PTAB/E2E traces directly.

Let me check the docketed district court case referenced on the patent's page for context.

Proceedings overview

Zero AIA trial proceedings on file. The USPTO Open Data Portal structured block for US5591639 reports no IPR, PGR, or CBM petitions — no institution decisions, no Final Written Decisions, no settlements, no appeals. My independent web searches for IPR/PGR/CBM numbers tied to "5,591,639" likewise returned nothing, so the ODP default holds: there is no PTAB activity on this patent. Because the patent issued 1997-01-07 and its term expired 2014-01-07, the defensive postures described in your template ("hardened by surviving IPRs" / "claims canceled") do not apply at all — the correct bottom line is that the patent is expired and off the board, and the only live question for a defendant is when the accused conduct occurred.

No proceedings to profile

I cannot populate the per-proceeding template because doing so would require me to invent a proceeding number, and that is off-limits under your instructions. Instead, here is what I verified and why the null result is credible rather than an indexing artifact.

What I checked

Why the null result is expected, not surprising

• IPR availability window was ~16 months. AIA IPR/PGR practice began 2012-09-16. The '639 patent expired 2014-01-07 (Google Patents: "2014-01-07 Anticipated expiration"; consistent with the 1997-01-07 grant plus the pre-URAA 20-year-from-filing term). Only a narrow window existed in which a commercially rational petitioner would spend ~$500k–$1M challenging claims with at most 16 months of enforceable life left.
• PGR was never available. PGR reaches only patents issued in the post-AIA regime; a 1997-01-07 grant is out of scope.
• CBM was never available. The '639 claims are directed to recombinant DNA expression vectors — a life-sciences technology, not a "financial product or service" under AIA § 18(d)(1). No CBM could have been brought.
• The one flagged piece of litigation is too old to boot-strap an IPR. Google Patents flags "[f]amily has litigation" at Maryland District Court, case 8:07-cv-03177 (ported from the Unified Patents litigation dataset, https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177). Any defendant served in that 2007 action was long past the § 315(b) one-year bar by the time IPR became available in 2012. I could not verify the parties or subject matter of 8:07-cv-03177 from public sources, and I am not asserting that US5591639 was the asserted patent in that case — the flag is a family-level litigation indicator only.
• I found no ex parte reexamination, inter partes reexamination, or EPO opposition record. That is an absence of evidence, not proof of absence — EPO opposition files are not reliably indexed in the sources I searched, so treat the European point as unverified.

Strategic summary

Claim status: all seven claims are UNTESTED and stand exactly as issued — and all seven expired on 2014-01-07. No claim of US5591639 has ever been canceled, narrowed, disclaimed, or confirmed in any AIA trial. The claim set is: claim 1 (plasmids pCMGS, pEE6.hCMV and pHT.1); claim 2 (recombinant expression vector comprising the promoter, enhancer and complete 5' untranslated region including the first intron of the hCMV-MIE gene operably linked to a heterologous coding sequence); claim 3 (further comprising a restriction site to facilitate insertion of the heterologous coding sequence); claim 4 (direct linkage to the coding sequence); claim 5 (further including the hCMV-MIE translational initiation signal); claim 6 (initiation signal including 5'-GTCACCGTCCTTGACACCATG-3'); and claim 7 (initiation signal including 5'-CCATGG-3'). Because the patent expired in 2014, "survived" is the wrong metaphor — nobody needed to kill these claims, and the term simply ran out. Damages liability is limited to infringing acts occurring on or before 2014-01-07 (plus any pre-expiration limitations analysis under § 286, which cuts off recovery six years back from filing, i.e. 2008-01-07 or earlier depending on the filing date of the complaint).

Estoppel landscape is empty and therefore irrelevant. There is no petitioner, so there is no § 315(e)(2) estoppel, no privity chain, and no IPR-barred ground. If you are being asserted against today, no prior-art ground is foreclosed by an AIA trial estoppel — but that is cold comfort, because the § 102/§ 103/§ 112 invalidity and § 101 arguments that matter in an expired-patent dispute are ones you would raise in district court, where the '639's prosecution history (1989 filing, 1994 continuation, 1997 grant) and the pre-AIA prior-art regime govern. Note the related continuation/division in the family, US5658759 (filed 1996-04-16, same 1987-07-23 priority, "Expired - Fee Related"), which the same owners controlled and which also shows no PTAB activity in the structured data — if a demand letter cites the family broadly, check both.

Pattern signals: none. No repeat petitioner, no serial filings, no patent-owner PTAB appellate activity, and no defensive aggregator in the chain. The interesting ownership pattern is corporate rather than litigational: Celltech Ltd → Alusuisse Holdings A.G. (1998-07-01) → Alusuisse Lonza Group AG (2001-08-13) → Lonza Group AG (2001-08-13) → Lonza Swiss Licences AG (2010-10-22) → Lonza Licences AG and Healthcare Royalty Partners II, L.P. (2013-12-27). A royalty-fund security interest (Healthcare Royalty Partners II) was recorded against the patent in 2013-12-27, roughly two weeks before expiration — a signal that the family was being monetized as a licensing asset up to the end of its term, consistent with the Lonza "GS System" licensing program (see the Lonza/CytoDyn license excerpt listing US5591639 as patent "5591639, Expiry Date 07.01.14").

Recommended next steps

1. If you are a defendant being asserted today: lead with expiration, not with validity. The patent expired 2014-01-07. Any claim chart citing post-expiration activity — a current GS-expression CHO cell line, a present-day biomanufacturing process — is asserting an expired patent against non-infringing conduct. Ask for the accused acts to be dated to a pre-2014-01-07 window, and for the § 286 damages-period computation. Do not concede the patent was ever valid or infringed for the residual pre-expiration window without running the numbers on recoverable royalties.

2. Say plainly that no PTAB proceeding exists, and use that affirmatively. A demand letter implying the patent survived a validity challenge, or threatening that "our IPR-proof patent" is at issue, is misleading. There has never been an AIA trial here; there is no FWD to link, because none was ever issued. If opposing counsel cites an IPR number for US5591639, ask for the E2E docket link — no such number exists in the ODP record.

3. If you are evaluating a pre-expiration damages exposure, don't budget for a PTAB defense. IPR on an expired claim is technically permissible as a validity determination, but it buys you nothing: there is no prospective relief to enjoin and no royalty to stop after 2014-01-07. Spend the budget on the district-court invalidity and damages-period arguments instead.

4. Verify the 8:07-cv-03177 Maryland matter before relying on it. The litigation flag on the patent page is family-level and dataset-derived. Confirm parties, outcome, and whether US5591639 itself was asserted by pulling the docket, rather than assuming — I could not substantiate it from public search results, and I decline to guess at the parties.

5. Do not treat this null as an endorsement of the claims. The claims were never tested. If you face a pre-expiration damages theory, the substantive work is a pre-AIA § 102/§ 103 attack on Boshart et al., Cell 41:521-530 (1985) and Spaete & Mocarski, J. Virol. 56(1):135-143 (1985) — both expressly discussed in the specification as the source of the PstI fragment — plus a § 112 written-description attack on the plasmid-deposit claim 1 (pCMGS, pEE6.hCMV, pHT.1), which recites specific biological materials without any sequence or deposit-number definition on the face of the claim.

Confidence statement: I am highly confident there are no AIA trial proceedings on US5591639, based on the canonical ODP block plus independent searches. I am not confident about the parties or substance of the 2007 Maryland case, and I have no verified record of any EPO opposition to the EP family member (EP0323997B1) or the JP family member (JP2505268B2). Those are open items I will not fill in by inference.

I'll search for the recorded assignment data before writing anything up. I must not invent reel/frame numbers.

Let me look for the specific reel/frame records and the correspondent of record.

Let me try to pull INPADOC legal-status records and the correspondent of record.

US 5,591,639 — Assignment Chain & NPE Screen

Data-provenance note (read first). My tooling for this run was open web search only; I could not return a live query against assignmentcenter.uspto.gov / assignment.uspto.gov, and I could not obtain reel/frame numbers or the correspondent of record for any link in this chain. The record-level data below (dates, conveyance types, assignors, assignees) comes from the USPTO-derived legal-events feed reproduced in the authoritative patent text supplied for this analysis (Google Patents, "Reassignment" entries). I am labelling every reel/frame and correspondent field as not retrieved rather than inventing one. Any statement not traceable to that feed or to a cited URL is flagged as context/unverified.

Inventors

Points to note:

• Single inventor. A one-inventor patent means the "all inventors departed within 12 months" fire-sale tell is not applicable — there is no co-inventor cohort to track.
• Name-form inconsistency across the family (flagging, not correcting). The US patent reads "Christopher R. Bebbington." Related Celltech filings render the name as "BEBBINGTON CHRISTOPHER ROBERT" (EP 0 491 875) and, in one PubChem/EP record, "BEBBINGTON ROBERT" alongside "WILSON HARRIS" on the GS patent DE 3772485 D1 / WO 87/04462. Per operating rules I report these literally; they appear to be OCR/ordering variants of one person, but I did not independently confirm identity.
• No departure record located. I found no evidence of Bebbington leaving Celltech within any specific window; he continued publishing Celltech GS work into the early 1990s. Treat "departure pattern" as not determinable from available sources.

Original assignee

Celltech Ltd. (Berkshire, United Kingdom) — later Celltech Therapeutics Limited (the name used as assignor in the 1998 recorded assignment), part of what became Celltech Group plc.

• Primary line of business: recombinant biopharmaceuticals and, critically for this patent, mammalian cell-culture technology — the glutamine synthetase (GS) gene-expression/amplification system and the pEE series of expression vectors. This patent is that platform: claims 1–2 cover the very plasmids (pCMGS, pEE6.hCMV, pHT.1) and the hCMV‑MIE promoter/enhancer/first-intron cassette used to drive them.
• Did they ship a product embodying the claims? Yes, in substance. Celltech commercialised/licensed the GS expression system as a research and manufacturing platform (the vectors and CHO/NS0 cell lines claimed), and the technology underpinned contract manufacture of recombinant proteins — the patent text itself reports TIMP, plasminogen activator and immunoglobulin light-chain expression from these constructs. DrugPatentWatch associates US 5,591,639 with INTRON A (interferon alfa‑2b) manufacture (https://www.drugpatentwatch.com/p/biologics/patent/5591639).
• Current status of the original assignee: acquired. Celltech Group plc was acquired by UCB S.A. in 2004. Important nuance: the biologics/cell-culture business relevant to this patent did not stay with Celltech to the UCB era — it was sold to the Alusuisse‑Lonza group in the mid‑1990s (see the 1998 recorded assignment to Alusuisse Holdings A.G.). That sale, not a bankruptcy, is what moved this patent out of Celltech. (The 1996 Celltech Biologics → Alusuisse‑Lonza transaction is context I did not re-verify in this run; flag as unverified.)

Assignment timeline

All entries below are the "Reassignment" records from the USPTO-derived legal-events feed. Each is listed with the single date the feed provides (recording/processing date as displayed) — I could not extract separate executed-vs-recorded pairs.

⚠️ Reel/Frame and Correspondent: NOT RETRIEVED. My search tooling could not open assignmentcenter.uspto.gov record detail, so no reel/frame number and no recording correspondent (attorney/firm) is reported anywhere below. These must be pulled via a live Assignment Center query. I did not fabricate any.

Closest adjacent data point, clearly labelled: the prosecution attorney of record on the granted patent (per FreePatentsOnline, https://FreePatentsOnline.com/5591639.html) was Suzanne E. Ziska, firm Spencer Frank and Schneider, Washington, DC. That is patent-prosecution counsel, not the assignment-recording correspondent, and the two roles should not be conflated.

1. 1998‑07‑01 — Reel not retrieved/not retrieved

   • Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
   • Assignor: CELLTECH THERAPEUTICS LIMITED (formerly known as "Celltech Limited")
   • Assignee: ALUSUISSE HOLDINGS A.G.
   • Correspondent: not retrieved
   • Context: Divestiture — Celltech's biologics/cell-technology business transferred to the Alusuisse‑Lonza group; this is the transaction that removed the patent from its inventor's employer.

2. 2001‑08‑13 — Reel not retrieved

   • Conveyance: Merger
   • Assignor: ALUSUISSE HOLDINGS, AG
   • Assignee: ALUSUISSE LONZA GROUP, AG
   • Correspondent: not retrieved
   • Context: Internal reorganisation — upstream holding-company merger; no change in ultimate control.

3. 2001‑08‑13 — Reel not retrieved

   • Conveyance: Assignment of assignors' interest
   • Assignor: ALUSUISSE LONZA GROUP, AG
   • Assignee: LONZA GROUP, AG
   • Correspondent: not retrieved
   • Context: Internal reorganisation — consolidation into the Lonza Group name. Same-day pairing with item 2 is a classic reorg signature.

4. 2006‑08‑01 — Reel not retrieved

   • Conveyance: Security Agreement
   • Assignor: PHARMATHENE, INC.
   • Assignee: MPM BIOVENTURES III-QP, L.P., as administrative agent
   • Correspondent: not retrieved
   • Context: Anomalous / unexplained. This is the only entry in the chain whose parties have no apparent relationship to Celltech or Lonza. PharmAthene and MPM BioVentures III are unrelated to the GS/expression-vector business. This looks like a legal-events mis-linking artifact in the source feed (or a stray USPTO record), not a step in this patent's chain. I am flagging it rather than treating it as a real transfer.

5. 2010‑10‑22 — Reel not retrieved

   • Conveyance: License
   • Assignor: LONZA GROUP AG
   • Assignee: LONZA SWISS LICENCES AG
   • Correspondent: not retrieved
   • Context: Internal licence to a dedicated IP-holding subsidiary (the "Intercompany Licence Agreement" structure described in Lonza Swiss Licences AG's Swiss commercial-register purpose, https://www.northdata.com/Lonza%20Swiss%20Licences%20AG,%20Basel/CHE-116.039.014).

6. 2010‑10‑22 — Reel not retrieved

   • Conveyance: Conditional Assignment
   • Assignor: LONZA GROUP AG
   • Assignee: LONZA SWISS LICENCES AG
   • Correspondent: not retrieved
   • Context: Financing-linked holding structure — the Swiss register purpose expressly ties this to a licence agreement plus pledge agreements with Drug Royalty II LP and Drug Royalty III, L.P. Same-day with item 5.

7. 2013‑12‑27 — Reel not retrieved

   • Conveyance: Assignment of assignors' interest
   • Assignor: LONZA GROUP AG
   • Assignee: LONZA LICENCES AG; HEALTHCARE ROYALTY PARTNERS II, L.P.
   • Correspondent: not retrieved
   • Context: Royalty monetisation — addition of a royalty-finance fund alongside the Lonza licensing vehicle.

8. 2013‑12‑27 — Reel not retrieved

   • Conveyance: License
   • Assignor: LONZA GROUP AG
   • Assignee: LONZA LICENCES AG
   • Correspondent: not retrieved
   • Context: Internal licence — same-day cluster with items 7 and 9.

9. 2013‑12‑27 — Reel not retrieved

   • Conveyance: Security Agreement
   • Assignor: LONZA LICENCES AG
   • Assignee: HEALTHCARE ROYALTY PARTNERS II, L.P.
   • Correspondent: not retrieved
   • Context: Collateral pledge securing a royalty-financing obligation — HC Royalty takes a security interest, not an asserting position. Matches the XOMA/HC Royalty "true sale + precautionary security interest" template seen in SEC exhibits.

10. 2014‑06‑26 — Reel not retrieved

    • Conveyance: Amended Conditional Assignment
    • Assignor: LONZA GROUP AG
    • Assignee: LONZA SWISS LICENCES AG
    • Correspondent: not retrieved
    • Context: Housekeeping — amendment to the 2010 conditional assignment, recorded five months after the patent's anticipated expiry.

Non-assignment anchor dates for context: application 08/300,063 filed 1994‑09‑02; patent granted 1997‑01‑07; anticipated expiration 2014‑01‑07; status "Expired – Lifetime."

Timeline diagram

timeline
    title Ownership of US 5591639
    1987 : Priority filed by Celltech Ltd
    1994 : US application 08 300 063 filed
    1997 : Patent granted 07 Jan
    1998 : Assigned to Alusuisse Holdings AG
    2001 : Merger into Alusuisse Lonza Group AG
         : Then assigned to Lonza Group AG
    2006 : Unrelated security agreement appears
    2010 : Licence to Lonza Swiss Licences AG
         : Conditional assignment same day
    2013 : Assigned to Lonza Licences AG
         : Healthcare Royalty security agreement
    2014 : Patent expired 07 Jan

NPE / troll-pattern signals

1. Shell-entity transfer — Not present

The patent does move into dedicated holding/licensing vehicles — Lonza Swiss Licences AG (2010‑10‑22 licence + conditional assignment; 2014‑06‑26 amended conditional assignment) and Lonza Licences AG (2013‑12‑27). But the concrete evidence cuts against the shell-entity call: Lonza Swiss Licences AG is a registered Swiss company (CHE‑116.039.014) with a detailed, published corporate purpose covering an intercompany licence agreement, conditional assignment, and a slate of named third-party licence and supply agreements (Regeneron, Glaxo Group, Genmab, Alexion, Biogen). That is a captive IP-holding/licensing subsidiary of an operating group, not a single-purpose anonymous Delaware/Texas LLC with a registered-agent address. No registered-agent-service address and no evidence of "no products in commerce" was found.

2. Known asserter in the chain — Not present

I compared every assignee against the standard list (Acacia, Marathon, Intellectual Ventures, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities, plus Unified/RPX high-frequency plaintiffs). None of the recorded assignees or security-interest holders match. Alusuisse/Lonza is an operating speciality-chemicals/CDMO group; Healthcare Royalty Partners II, L.P. is a royalty-financing fund holding a security interest, which is a creditor position, not an asserter position.

3. Repeat correspondent across the chain — Unclear / not assessable

This is the single most informative field per the brief, and I could not retrieve it. No recording correspondent or attorney name was obtainable for any of the ten chain entries, so I cannot test for a recurring recording attorney. Do not read this as "no recurrence" — it is "unverified." A live Assignment Center pull of the ten records is required.

4. Cascading transfers — Weak / explainable

There are same-day clusters: 2001‑08‑13 (two entries), 2010‑10‑22 (two entries), 2013‑12‑27 (three entries). Under the brief's test, however, the tell is chained transfers through unrelated LLCs sharing a correspondent address or principals within <24 months. Here the transfers run within one corporate group (Alusuisse → Alusuisse Lonza → Lonza Group → Lonza Swiss Licences → Lonza Licences), and the same-day clustering is the opposite of obfuscation — it is the signature of an internal restructure plus a financing that were papered together. The spans are also not <24 months across the whole chain (2001→2010→2013). Weak signal at most.

5. Pre-litigation transfer — Not present

The enforcement activity naming this patent is from 2007: D. Md. 8:07‑cv‑03177 (https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177) and D. Del. 1:07‑cv‑00467‑GMS, Lonza Group AG v. Northwest Biotherapeutics, Inc., whose 2007‑07‑30 AO‑120 report lists 5,591,639. The nearest preceding assignment is the 1998‑07‑01 transfer to Alusuisse and the 2001‑08‑13 reorg — roughly six years and six months before suit, far outside any 6-month window. All the 2010 and 2013 transfers post-date the litigation. No venue- or standing-driven last-minute assignment is present.

6. Bankruptcy fire-sale — Not present

No evidence of a Chapter 7/11 proceeding at Celltech, Alusuisse/Lonza, or any chain entity. The 1998 transfer out of Celltech reads as a business-unit divestiture, not an insolvency sale. The 2006‑08‑01 MPM BioVentures/PharmAthene security agreement is the only record resembling distress financing, and its parties are unrelated to this patent's chain — I treat it as a source-feed artifact, not as a fire-sale step.

7. Privateering — Not present

Privateering means an operating company hands the patent to an NPE that asserts on its behalf. Here the asserting party in the 2007 suits was Lonza Group AG itself — an operating company — suing in its own name against Northwest Biotherapeutics, Inc., apparently a licensee/commercial counterparty. There is no NPE interposed between Lonza and the courthouse.

8. Defensive aggregator — Not present

The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. The patent was simply allowed to run to its 2014‑01‑07 anticipated expiry, and its recorded status is "Expired – Lifetime."

Verdict

Operating-company assertion.

Justification: the chain never passes through an NPE — it runs Celltech Ltd. → Alusuisse Holdings A.G. (1998‑07‑01) → Alusuisse Lonza Group AG / Lonza Group AG (2001‑08‑13) → Lonza Swiss Licences AG (2010‑10‑22) → Lonza Licences AG (2013‑12‑27), ending with a royalty-financing security interest to Healthcare Royalty Partners II, L.P. rather than a transfer to an asserter; and the only enforcement naming this patent, Lonza Group AG v. Northwest Biotherapeutics (D. Del. 1:07‑cv‑00467‑GMS, AO‑120 report 2007‑07‑30) and the related D. Md. 8:07‑cv‑03177 action, was brought by the operating Lonza group in its own name against a commercial counterparty. Two caveats you should carry forward: (i) the reel/frame numbers and the recording correspondent are unretrieved, so the "repeat correspondent" test is unverified rather than negative; and (ii) the 2006‑08‑01 MPM BioVentures/PharmAthene "Security Agreement" entry is unexplained and appears to be a source-feed mis-link, not a real step in this chain.

Verify at USPTO Assignment Center (search by patent number 5591639, then open each record to read Reel/Frame and "Correspondent"): https://assignmentcenter.uspto.gov/ — alternate index: https://assignment.uspto.gov/patent/index.html . Cross-check the litigation entries at https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177 and the assignee-of-record view at https://patents.google.com/patent/US5591639/en.

I'll search for the specific patent number and its front-page cited references.

Let me look for the front-page references on the patent document itself.

Let me try to retrieve the actual front-page reference list and the Google Patents citations listing.

I hit the tool-call limit before I could pull the examiner's front‑page (56) reference list, so I need to be upfront about a material limitation before giving the analysis.

⚠️ Transparency note on the search

I could not retrieve the actual front‑page "References Cited" (item 56) list from the US 5,591,639 patent document. My open‑web searches repeatedly returned derivative documents (later patents that cite US 5,591,639 — e.g., EP 2 711 426 A1, EP 1 475 443 A2, CN 166 8749 A, CA 2 672 809 C, various Lonza/Genmab filings) rather than the patent's own front page.

Corroborating evidence that this list was unavailable: the authoritative full text supplied to me contains a "Cited By" section but no "Citations"/"References Cited" section — a strong indicator the (56) block was not captured.

Therefore: rather than fabricate U.S./foreign patent numbers I cannot verify, I am reporting only references that are expressly identified in the authoritative patent text (the specification's background and examples), plus the family/priority documents from the patent's own bibliographic record. I flag each item's source and confidence level. If a literal USPTO (56) list is required, it must be pulled from USPTO Patent Center / Patent Public Search (Patent Full‑Text) for 5,591,639 — I did not have logged‑in access.

One further point the user should note: the task asks for "each patent citation." In the authoritative text, only two patent documents are actually named as prior art/technology sources (EP 260148 and WO 87/04462); everything else in the document is non‑patent literature. I am not going to pad the list with unverifiable patent numbers.

Statutory framework

US 5,591,639 has a priority date of 1987‑07‑23 (GB 8717430), was filed as US 08/300,063 on 1994‑09‑02, and issued 1997‑01‑07. Because it was filed before 2013‑03‑16 and claims pre‑AIA priority, pre‑AIA 35 U.S.C. §§ 102/103 govern. The effective filing/priority date for §102(a)/(b) purposes is tied to the 1987‑07‑23 GB priority (via PCT/GB1988/000602) and, at the latest, the 1989‑04‑28 U.S. filing. So references published 1983–1986 are potential §102(a)/§102(b) art.

Part A — Patent documents identified in the authoritative text

Not prior art but on the same record (flagged to prevent confusion):

• WO 1989/001036 A1 / EP 0 323 997 B1 / JP 2505268 B2 / AT E88501 T1 / DE 3880468 T2 / GB 8717430 D0 — these are the patent family members of 5,591,639 (same 1987‑07‑23 priority). WO 89/01036 published 1989‑02‑09, i.e. after the priority date, so it is not §102(a) art against this patent — it is the patent's own international publication.
• US 5,658,759 — divisional of this patent (from app. 08/633,013, filed 1996‑04‑16). Same disclosure; a §102 candidate only in the "same‑family/double‑patenting" sense, not as extrinsic prior art.
• pSVLGS.1 (Fig. 1) and pEE6 are constructs described but not separately claimed.

Part B — Non‑patent literature expressly cited in the authoritative text

These are the substantive prior‑art references the applicant identified. I list them because "each patent citation" in the document is otherwise sparse; several are the classic hCMV‑MIE references. Dates below are as printed in the patent / standard bibliographic form; I have not re‑verified each publication date against a primary source.

Part C — Bottom‑line §102 assessment

No reference identified in the authoritative text anticipates any of claims 1–7. Specifically:

1. Claim 1 (pCMGS, pEE6.hCMV, pHT.1 as such) — none of the cited references discloses these named plasmids. Claim 1 is a product‑by‑name claim to constructs defined by their specific component ligations (Pst‑I hCMV fragment + synthetic NcoI adaptor + GS or TIMP cassettes). The GS‑system reference (A2/WO 87/04462) supplies only the pSV2.GS parent.

2. Claim 2 (the genus) — the invention's point of novelty is the complete hCMV‑MIE 5'‑UTR including the first intron operably linked to a heterologous coding sequence. The two closest references (B1 Boshart; B2 Spaete & Mocarski) each supply only the promoter/enhancer and, at most, part of the 5'UTR, and both rely on the heterologous gene's own 5'UTR for translation. A1 (EP 260148) is even more clearly distinguished: it puts a fragment of the hCMV 5'UTR into the heterologous gene's 5'UTR as a stabilizer. None anticipates; they are §103 material.

3. Claims 3–7 (dependent on claim 2) — same analysis; B7 (Kozak 1986) is the reference most relevant to the modified initiation signal/NcoI limitation of claims 6–7, but a consensus sequence is not the claimed vector.

Explicit caveats on this deliverable

1. The (56) front‑page reference list for US 5,591,639 was NOT retrieved. I could not access USPTO Patent Center / Patent Full‑Text or a Google Patents "Citations" tab. A definitive answer requires those primary sources. Any patent numbers a USPTO printout shows that are absent above are simply not verifiable by me here, and I declined to invent them.
2. Citation spelling is reported literally. The patent prints "Spaeta & Mocarski" (standard: Spaete), "Foecking and Horstetter" (standard: Hofstetter), and "Aced. Sci." for "Acad. Sci." I did not auto‑correct these.
3. Internal inconsistencies in the authoritative text (flagged, not corrected): the specification describes an intermediate as "pMT.1" but claim 1 recites "pHT.1"; the text uses "HSX" once where "MSX" (methionine sulphoximine) is evidently meant; and "pEE6.TIMP"/"pEE6TIMP" appear in both forms. These are relevant to the claim‑1 support question but are not prior‑art issues.
4. Dates for A1 (EP 260148) and the NPL items are as stated in the patent and standard bibliographic usage, not independently re‑verified against primary publications in this session.
5. Applicable law is pre‑AIA §102/§103 (pre‑2013 filing; 1987 priority). If any reference you eventually pull from the (56) list published between the 1987‑07‑23 priority date and the 1989‑04‑28 U.S. filing, its §102 status needs individual re‑examination.

Recommended next step: query USPTO Patent Public Search or Patent Center for application 08/300,063 and read the "References Cited" (56) field and the File Wrapper (which for this family also bears on the ~2007 litigation in D. Md. 8:07‑cv‑03177 and D. Del. 1:07‑cv‑00467‑GMS, where validity was put at issue). That is the only way to close the gap I have flagged above.

I'll ground the obviousness analysis with a few targeted searches before writing.

Excellent — the EP 260148 snippet about intron/splice-site necessity is highly relevant. Let me pull more detail on it and on the hCMV MIE sequence prior art.

Obviousness Analysis — US 5,591,639 under 35 U.S.C. § 103

Scope note. This builds on the Prior Art section already generated (references A1–A2 and B1–B12, with its stated limitation that the front-page (56) list was not retrieved). I re-ran searches and added verified substance and dates for the two most important references — EP 0 260 148 (previously "A1," unverified) and Foecking & Hofstetter (B6) — plus the actual abstract of Spaete & Mocarski (B2). Everything below is pre-AIA §103 (application filed 1994-09-02 from a chain reaching 1989-04-28, priority 1987-07-23). Because the patent expired 2014-01-07, this matters only to the pre-expiration damages window — but within that window it is the whole ballgame.

1. Bottom line

1. The single most dangerous reference is EP 0 260 148 (Genentech), and its content is materially worse for the patentee than the '639 specification admits. The specification characterizes EP 260148 as merely using "part of the 5'-untranslated region of the hCMV‑MIE gene as a stabilising sequence." The published document actually teaches that expression fails without a splice donor–intron–acceptor element, that the element must be 5′ of the heterologous cDNA, and that the hCMV‑MIE splice donor/intron is a preferred source of that element. That is a direct, explicit motivation to put an hCMV‑MIE intron in the 5′ leader immediately upstream of the heterologous coding sequence — i.e., claim 2 and claim 4.
2. Neverthless there is a serious date problem for the two most structurally similar references (EP 260148, WO 87/04462), both of which published after the 1987-07-23 priority date. Whether either is §102(b) art turns on a single, verifiable fact: the effective U.S. filing date (§ 120/§ 363 benefit) of application 07/339,615. I flag this as the gating issue rather than resolving it (§2 below).
3. The strongest §103 combination that does not depend on that gating issue is: Boshart (June 1985) + Spaete & Mocarski (Oct. 1985) + Stenberg (Jan. 1984) + Kozak (1986). All four predate 1987-07-23, all four are the applicant's own cited art, and together they supply (a) the hCMV‑MIE enhancer/promoter as an expression-vector component, (b) an actual hCMV‑MIE‑promoter/heterologous‑gene fusion, (c) the complete structural map of the MIE gene showing its 5′ leader is spliced and contains an intron, and (d) the sequence rules for translation initiation.
4. Claim 7 is the weakest claim. Its only added limitation — 5′-CCATGG-3′ — is the sequence the specification itself concedes "many eukaryotic genes contain … overlapping the translation start site." That is an admission against interest that largely decides the claim.
5. Claim 1 (three plasmids by name) is vulnerable to a "known elements, known methods, predictable result" attack, and its specification's own Table 1 (pCMGS 17 colonies vs. pSV2.GS 32) undercuts any unexpected-results story at the construct level.
6. The patentee's best defense is not a technical argument — it is the priority-date shield in §2, plus the fact that claims 2–7 were allowed by an examiner who had the hCMV references of record.

2. The gating issue: which references are actually prior art

Two facts I could not resolve from open sources, and which are dispositive:

(a) What is the §102(b) critical date? The chain is US 07/339,615 (filed 1989-04-28) → 07/896,797 (1992-06-09) → 08/300,063 (1994-09-02, the patent). But the Google Patents record also lists PCT/GB1988/000602 (international filing date 1988-07-22) in the priority chain. If 07/339,615 was a § 371 national-stage entry of that PCT and the patent derives §120/§363 benefit from the 1988-07-22 international filing date, then:

Note how tight this is: on the first reading the critical date is 1987-07-22 — one day before the GB priority date, and WO 87/04462 misses §102(b) by eight days. Verify the file wrapper for 07/339,615 (a §371 vs. a straight §119-filing) before relying on either reference.

(b) §102(e) fallbacks. Even if §102(b) fails, WO 87/04462 (Celltech, English-language PCT) may be §102(e) art as of its international filing date — a date I did not verify. But note §103(c): the '639 patent and WO 87/04462 were commonly owned by Celltech, which is an argument to disqualify it as §103 art to the extent it is only §102(e)/(f)/(g) art. (The amended, §102(e)-inclusive version of §103(c) post-dates the 1997 grant of this patent, so its scope at the relevant time needs checking.) Practically: do not build the §103 case on WO 87/04462 alone.

Similarly, EP 260148's own U.S. counterpart (Genentech, "Improved recombinant expression method, vector and transformed cells," EP priority 1986-09-12) should be identified and checked for a §102(e) date — a U.S. filing at or near 1986-09-12 would put it squarely in the art as of a date before the '639 priority.

(c) Regardless of the gating outcome, the patentee's own specification is evidence of the state of the art. The '639 text states that prior workers "have used sequences from the upstream region of the hCMV‑MIE gene in expression vectors," and that EP 260148's vectors "contain part of the 5'-untranslated region of the hCMV‑MIE gene as a stabilising sequence." That is an applicant statement about what the field was doing, usable in a §103 motivation analysis even for a reference whose statutory status is contested.

3. The reference set, with dates and status

4. Why EP 260148 is the reference that does the damage

The '639 specification's two-sentence characterization of EP 260148 is a concession plus a mischaracterization. The published EP 260148 A2 text (as reproduced in the Google Patents full text and the EP document PDF) teaches, in its own words:

• "a vector was constructed containing a cytomegalovirus promoter and enhancer, a cDNA encoding factor VIII, and a 3′ terminating sequence, absent any intron or constructed splice site. Neither transient nor stable expression of factor VIII was observed in any of the cell types tested";
• the same negative result with SV40 promoter/enhancer and no intron;
• "deletion of the Ig variable region intron and donor and acceptor sites, while maintaining the other control regions, resulted in elimination of transient expression … at least one splice donor-intron-acceptor sequence appears to be required for expression";
• "location of the stabilizing sequence is important … an intron 3′ to the cDNA … failed to express factor VIII";
• the DK family claims recite the splice donor "is from the immediate early gene of human cytomegalovirus," with the intron from hCMV plus the immunoglobulin variable region and an Ig acceptor; and
• "the CMV enhancer and promoter can be completely replaced by the analogous SV40 enhancer and promoter … factor VIII production is not dependent on the specific transcriptional start signal but rather is dependent on other parts of the control region such as the stabilizing sequence."

Three consequences:

1. Motivation to place a spliceable hCMV‑MIE element 5′ of a heterologous coding sequence is express, not inferred. The reference teaches both the what (an intron/splice donor–acceptor must be present) and the where (5′ of the cDNA, not 3′), and names hCMV‑MIE as a source.
2. The applicant's distinction collapses to a question of degree. The delta between EP 260148 and claim 2 is "chimeric, Ig-acceptor-terminated intron built from an hCMV splice donor" vs. "the hCMV‑MIE first intron in its native, complete 5′ leader." Given that the hCMV‑MIE first intron is delivered for free on the same PstI fragment that supplies the promoter and enhancer (Boshart/Spaete & Mocarski), substituting the native intron for the chimeric one is the paradigm KSR case — substitution of one known element for another to obtain its known advantages.
3. Note the date trap for the patentee. The specification cites EP 260148 in the "Background to the Invention" — an applicant who cites a document as showing what the prior art did has a hard time later arguing it is not prior art at all. Watch for the patentee running an inconsistent position: "EP 260148 is prior art for §112-best-mode purposes but not §103."

5. Claim-by-claim §103 analysis

Claim 2 — "promoter, enhancer and complete 5′ UTR including the first intron … operably linked to a heterologous coding sequence" (the genus)

Primary combination: R1 (Boshart) + R2 (Spaete & Mocarski) + R3 (Stenberg).

Motivation to combine (why a PHOSITA would have done this):

• The fragment is the path of least resistance. R2's PstI fragment already contains promoter, enhancer and the leader including intron A. Nothing in R1 or R2 tells the artisan to delete the intron; getting the strong enhancer in hand meant accepting the rest of the fragment as it came.
• The intron was known to be functionally useful, not inert. R3 showed the MIE mRNA is a spliced molecule with the ATG in the second exon, so an artisan building an MIE-driven cassette in 1987 was working with a leader whose natural structure includes the first intron. Combined with EP 260148's splice-requirement teaching (if available) and the general splicing/mRNA-stability literature of the era (e.g., the enhancer-in-intron work cited generally in the period), including the intron carried a known expected benefit and no known penalty.
• A commercial need existed. The '639 background itself states the goal: reliably high-level production from a transfected cell line, with transcription/translation efficiency as a named bottleneck — and states that heterologous-gene 5′ UTRs give "unreliable" translation "probably as a result of the 'hybrid nature' of the 5′-untranslated region."

KSR rationales engaged: (i) known elements arranged per known methods; (ii) simple substitution of one known element for another (complete/native MIE leader for the partial leader + heterologous leader); (iii) "obvious to try" — the identified options were finite and predictable (use the MIE leader whole, partially, or not at all), and the reference set supplied a reasonable expectation of success.

Secondary form of the same combination (if EP 260148 is available): R2 + R5 + R3, motivated by EP 260148's explicit statement that a splice donor–intron–acceptor upstream of the cDNA "appears to be required for expression" and that the CMV promoter can be swapped out without loss — which directs the artisan's attention to the 5′ stabilizing element rather than the promoter.

Anticipation-adjacent flag (verify): the actual title/abstract of R2 is a trans-activation study ("α and β promoters are trans activated…"), not a bioproduction paper — a genuine discrepancy from the '639 specification's characterization. The constructive question for a §102 attack (or a strong §103 sub-case) is: does the R2 lacZ construct contain the hCMV‑MIE 5′ leader including intron A fused in-frame to a heterologous coding sequence? If yes, claim 2 is arguably anticipated outright, and the "part of the 5′UTR" characterization in the '639 specification is wrong. I could not resolve this from the abstract; pull the full text and Figures 1–3 of R2.

Claim 4 — "…linked directly to the heterologous coding sequence" (the real battleground)

This is where the patentee will fight, because it is where the applicants asserted the unexpected result ("we have unexpectedly found that when the hCMV‑MIE derived DNA is linked directly to the coding sequence of the heterologous gene high levels of mRNA translation are achieved … consistently and … independent of the particular heterologous gene").

Combination: R2 + R5 + R6 (Kozak) + R7 (Pelletier & Sonenberg).

Motivation: R6 and R7 teach that features of the 5′ UTR — length, secondary structure, and the sequence context of the AUG — modulate translation initiation. That teaching cuts against carrying two 5′ leaders in series. An artisan trying to remove a confounder (a "hybrid" 5′ UTR formed by splicing two heterologous leaders together) is led to one defined leader and to place the heterologous ATG where the viral ATG naturally sits — which is exactly what R3's exon map tells him where to put it (ATG in the second exon, immediately after intron A). R5 reinforces this by teaching that the 5′ element, not the promoter or the transcriptional start, governs the productive outcome.

Prediction/predictability: the linkage question ("direct" vs. "with the gene's own 5′ UTR") is a binary design choice; the reference set identifies both options and supplies a reason to prefer direct linkage. This fits the In re Aller / In re Applied Materials line (optimization of a recognized variable by routine methods) as much as the "unexpected results" line.

The patentee's counter (see §8) is that the art affirmatively taught retaining the heterologous gene's own 5′ UTR — R2 and (per the '639 specification's own reading) R5 both did so. That is a "teaching away" theory. It is worth noting for the client that the Federal Circuit requires the art to criticize, discredit, or discourage the claimed solution (In re Fulton; Medichem v. Rolabo), and "the art used a different arrangement" falls short of that. Also, R5's own data locate the critical variable in the 5′ stabilizing element, not in preserving the heterologous leader.

Claims 5, 6, 7 — the translation-initiation-signal limitations

Claim 5 (vector "further includes the translation initiation signal of the hCMV MIE gene"): R3 is dispositive in substance. Stenberg et al. determined the MIE exon sequences and identified where the ORF begins (within exon 2). The MIE's own initiation context was therefore a published, sequenced, and locatable feature of the same gene whose promoter the artisan was already using. Combining R1/R2 (promoter) with R3 (the sequenced ATG context) requires no inventive act.

Claim 6 (initiation signal "includes the sequence 5′-GTCACCGTCCTTGACACCATG-3′"): two independent routes.

Route (a) — inherent disclosure. This 21-mer reads as the MIE leader sequence immediately upstream of the ATG, ending in …CACC|ATG. R3 reports the exon and splice-junction sequences and the location of the ORF within exon 2; the DNA sequence flanking the MIE ATG is therefore in a pre-1987 printed publication. If Stenberg's (or Boshart's deposited) published sequence matches, claim 6 is anticipated by, or at minimum obvious over, R3 in combination with R6.

Route (b) — obvious engineering. Note a subtlety in the claim's own wording: the specification states the synthetic oligomer "recreate[s] the complete 5′-untranslated sequence of the MIE gene with the single alteration of a G to a C at position −1 relative to the translation initiation codon." Counting the recited 21-mer (…G A C A C C A T G), the position immediately 5′ of ATG is C — i.e., claim 6 as written recites the engineered (Kozak-optimized) sequence, not the natural MIE sequence. That is decisive for obviousness: the claimed sequence is the product of applying R6's consensus rule (5′-ACCATGPu-3′) to a sequence already in hand, with a known and expected effect. It also means claim 6 does not read on the unmodified natural MIE leader.

Claim 7 (initiation signal "includes 5′-CCATGG-3′"): the weakest claim in the patent, and the specification pleads it away:

"Many eukaryotic genes contain an NcoI restriction site (5′-CCATGG-3′) overlapping the translation start site, since this sequence frequently forms part of a preferred translation initiation signal 5′ACCATGPu-3′."

So the applicant concedes that (i) CCATGG at the ATG is commonplace in eukaryotic genes, (ii) it is a preferred Kozak context, and (iii) it is a restriction site. The single G→C change needed to convert the natural MIE context into the claim-7 sequence is described in the specification as doing nothing more than making the sequence "resemble more closely the 'Kozak' concensus initiation signal" and "introduc[ing] an NcoI recognition site." A claim whose sole added limitation is the conventional restriction site the applicant chose for cloning convenience, over a Kozak consensus published in 1986 (R6), has a very short obviousness step. The obviousness case here is: R6 + the specification's own admission + official notice of the standard 5′NcoI/ATG cloning convention. Note also that many 1980s expression vectors were built around an NcoI (CCATGG) cloning site at the ATG; if any pre-1987‑07‑23 vector has that structure with an hCMV promoter, claim 7 is anticipated outright. I did not find one in this session — that search is worth running.

Claim 3 — "further comprising a restriction site to facilitate insertion of the heterologous coding sequence"

Essentially per se obvious, and again on the applicant's own record: the specification describes pEE6 as containing "a pSP64 … polylinker inserted in between the HindIII and EcoRI sites," with "the BamHI and SalI sites … removed from the polylinker by digestion." A polylinker/restriction site for inserting a coding sequence is the defining feature of a 1985-87 cloning vector. Combination: any of the R1–R2 promoter vectors + the standard polylinker practice the applicant itself describes (pSP64 polylinker / pCT54 backbone, R9). No motivation gap exists; the claim adds nothing to claim 2 but an admitted convention.

Claim 1 — plasmids pCMGS, pEE6.hCMV and pHT.1 (as such)

The specification supplies the complete parts list for each plasmid:

§103 case: every component is a pre-1987 published DNA element, every joining step is standard restriction/ligation/linker chemistry, and the result is a replication-competent plasmid that does exactly what its parts predict — it expresses GS and confers MSX resistance. The applicants verified this with a routine transfection (Table 1). Under KSR's "known elements + known methods + predictable result" branch, claim 1 is obvious. The only inventive contribution, if any, is the hCMV leader cassette — and a claim to it is claim 2, which is separately addressed. (Note the specification's own internal inconsistency flagged in the Prior Art section: the intermediate is called "pMT.1" in Example 2 but the granted claim recites "pHT.1." That cuts against claim 1 on §112 grounds and also creates a claim-construction question that the §103 analysis inherits.)

6. An additional, non-§103 validity theory worth preserving

Obviousness-type double patenting against the sibling US 5,658,759. The '759 patent (app. 08/633,013, filed 1996-04-16, same 1987-07-23 priority, "Expired – Fee Related") is a divisional of this application and shares the identical disclosure. Both were commonly owned (Celltech → Alusuisse → Lonza). If the '759 claims are not "patentably distinct" from the claims here, ODP is an independent invalidity/unenforceability defense that does not require any prior art at all. I have not examined '759's claims in this session — pull them and run the two-way distinctness test. (Relatedly, the Prior Art section already flagged the §112 written-description problem for claim 1's named-plasmid recitations; that is a claim-construction input to the §103 analysis as much as a standalone defense: if "pCMGS, pEE6.hCMV and pHT.1" is construed broadly by reference to the specification's component list, the §103 case above applies; if it cannot be construed, the claim is indefinite.)

7. Claim-by-claim summary table

8. The patentee's best rebuttal — what to expect

Run these as the counter-case so the analysis is not one-sided:

1. Priority-date shield. If the §102(b) critical date is 1987-07-22, EP 260148 (1988-03-16) and WO 87/04462 (1987-07-30) are out, and the §103 case must be built on Boshart + Spaete & Mocarski + Stenberg + Kozak alone. That combination is still respectable but is weaker on the motivation prong, because none of those four says "include the first intron" or "link directly." This is the patentee's single best argument and the reason §2 matters more than anything else in this memo.
2. Teaching away / consistency. The art the applicant itself cites (R2 for the hCMV promoter; R5 as the applicant reads it; the general practice described in the '639 background) used the heterologous gene's own 5′ UTR "such that translation initiation is from the natural sequence of the gene." The patentee will argue the field regarded that as necessary, so deleting it to fuse the viral leader directly to the heterologous ATG was a departure, not an optimization. Expect this to be backed by Kozak's PNAS 83:2850-2854 teaching that 5′-UTR sequences can inhibit translation — i.e., a long, structured viral leader placed at the ATG was arguably contraindicated.
3. Unexpected, gene-independent results. TIMP at 3×10⁸ molecules/cell/day from a single-copy line; amplification to 3×10⁹ at 20–30 copies/cell; PA-1 at 5.5×10⁷ and PA-2 at 1.1×10⁸; Ig light chain at ≥100 ng/ml transient. The breadth across four different proteins is the patentee's strongest secondary-considerations evidence.
4. Prosecution history. Claims 2–7 were allowed by an examiner with Boshart/Spaete & Mocarski-type art on the face of the specification. Under MPEP 2145-type reasoning that is weak (examiner allowance is not a validity presumption beyond the statutory one), but a patentee will use it.

Why each is beatable: (1) turns entirely on a verifiable docket fact — confirm it rather than assume it, and keep EP 260148 usable as an admission/background reference if not as statutory art; (2) "the art used a different design" is not teaching away absent criticism or discouragement, and R5's own data point to the 5′ stabilizing element as the critical variable; (3) the specification contains no comparative data against the closest art (nothing head-to-head against a partial-5′-UTR or heterologous-5′-UTR control), and its own Table 1 shows pCMGS conferring fewer MSX-resistant colonies than the prior-art pSV2.GS (17 vs. 32) — which undercuts a construct-level "unexpectedly superior" story; the headline 3×10⁹ figure required a second round of MSX selection, i.e., it is attributable in part to the known GS amplification system the applicants did not invent. Under the case law on unexpected results (commensurateness with claim scope; comparative evidence against the closest prior art), the evidentiary gap is real.

9. Verification checklist (facts I could not confirm — do these first)

1. File wrapper of 07/339,615: was it a § 371 national-stage entry of PCT/GB1988/000602 (critical date 1987-07-22) or a standalone U.S. filing of 1989-04-28 (critical date 1988-04-28)? This decides EP 260148's status.
2. WO 87/04462 international filing date (for a possible §102(e) date), and whether §103(c) common ownership disqualifies it as §103 art.
3. U.S. counterpart of EP 260148 A2 — filing date and any §102(e) date before 1987-07-23.
4. Full text and figures of Spaete & Mocarski 1985 — does the lacZ fusion include the hCMV‑MIE 5′ leader with intron A fused to the reporter? (Anticipation-grade question for claim 2.)
5. Stenberg 1984, Figure/sequence data — does the published MIE sequence include the ~20 nt immediately 5′ of the ATG (i.e., does it disclose claim 6's 21-mer)?
6. The (56) front-page list for 5,591,639, and the pre-1987-07-23 vector art with an NcoI site at the ATG (claim 7 anticipation check).
7. Claims of US 5,658,759, for the ODP analysis.
8. Literal citation note (not a correction): the specification prints "Kozak H. Cell 41 283-292 (1986)" — volume 41 is a 1985 volume and the standard initiation-context paper is Cell 44(2):283-292 (1986). Per operating rules I report the printed citation as-is; confirm which paper the examiner actually applied.

Confidence statement

• High confidence in the verified bibliographic facts now attached to EP 260148 (Genentech; priority 1986-09-12; published 1988-03-16), Foecking & Hofstetter (Oct. 1986), Boshart (June 1985), Spaete & Mocarski (Oct. 1985) and Stenberg (Jan. 1984).
• High confidence that the EP 260148 text teaches splice-dependence and 5′ placement of the stabilizing element, because I read those passages from the published document and its EPO abstract rather than from a summary.
• Low confidence on the §102(b) / §102(e) status of EP 260148 and WO 87/04462 against the 1987-07-23 priority. I set out the two competing computations rather than picking one, because the answer turns on a file-wrapper fact I could not retrieve.
• Explicit uncertainty: whether Spaete & Mocarski's construct includes the MIE first intron (this could convert the claim-2 analysis from obviousness to anticipation), and whether the published MIE sequence covers claim 6's 21-mer. Both are answerable from the primary documents and I did not have them in full text.
• No prior-art reference, combination, or docket fact above is invented. Where I lacked a source I said so; where a source exists, the URL/citation is given.