US 5891693

I'll verify current information on this patent, including any 2026 CAFC activity, before summarizing.

US 5,891,693 — verification note

I ran targeted searches for the exact identifier 5891693 (and for any 2026 CAFC activity). No 2026 Federal Circuit docket or appeal involving US 5,891,693 surfaced. The patent's legal status is Expired – Lifetime with an anticipated expiration of 2016‑04‑06, which is consistent with the absence of current appellate activity. I cannot claim authoritative USPTO/CAFC docket completeness from these searches; treat the litigation items below as reported records, not verified 2026 dockets.

Bibliographic data (from the patent record)

Data caveats (interpreted literally, not auto-corrected): The Google Patents family listing shows US 07/898,165 as a continuation with a priority date of 1988‑04‑18 and a corresponding document US5879936A titled "Recombinant DNA methods, vectors and host cells," while the 5891693 record itself claims priority to 1990‑01‑25. These two statements are internally inconsistent as displayed. I am flagging the discrepancy rather than resolving it.

Reported litigation: A Darts‑IP family litigations link (family 27039591) and a Unified Patents link to a Maryland District Court case, 8:07‑cv‑03177. I did not independently confirm the parties or outcome.

Abstract (verbatim)

The present invention relates to vectors useful for transforming a lymphoid cell line to glutamine independence. The vectors comprise an active glutamine synthetase (GS) gene as well as a heterologous gene of interest to be expressed. The preferred embodiments encompass vectors wherein the heterologous gene is expressed from a relatively strong promoter and the GS gene is expressed from a relatively weak promoter. In one example, the heterologous gene is operatively linked to the hCMV-MIE promoter and the GS gene is operatively linked to the SV40 early region promoter.

Independent claims in plain language

There are three independent claims: 1, 17, and 18.

Claim 1 — Method of conferring glutamine independence to a myeloma cell line.
Transform a myeloma cell line with a vector that carries (a) a glutamine synthetase (GS) gene and (b) one or more genes encoding protein(s) foreign to the myeloma cell ("heterologous"). The vector's genes must be arranged so that the GS gene can be expressed and glutamine‑independent myeloma colonies can be produced. The gist: the gene layout must not shut down GS expression.

Claim 17 — Method of selecting transfected myeloma cells.
A specific three‑step selection protocol:

1. Plate the transfected cells in one volume of non‑selective medium containing glutamine;
2. 24 hours later, add two volumes of glutamine‑free medium (i.e., dilute rather than abruptly remove glutamine);
3. Incubate 7 days and recover the myeloma colonies.

This claim is directed to the "Protocol A" gradual‑depletion procedure the specification identifies as giving the highest survival of transfectants (Table 2: 130 colonies/10⁶ cells vs. 0 for mock).

Claim 18 — Method of conferring glutamine independence to a lymphoid cell line.
The same subject matter as claim 1 but reciting a "lymphoid cell line" generally rather than a myeloma cell line specifically — i.e., the broader genus, with independent claims 1 (myeloma) and 18 (lymphoid) standing separately.

Core technical point common to all three: the specification attributes failures with earlier constructs (e.g., pSVLGS1 and pAb2GS) to "promoter occlusion"/transcriptional interference — a strong promoter driving an upstream heterologous gene reads through into a downstream GS gene with a weak promoter, starving the cell of GS and collapsing transfection frequency. The claimed solutions are: order/orient the genes so transcription does not run through GS (see dependent claims 2–8, 19–25), use a weak GS promoter with a strong heterologous promoter (claims 3, 5, 20, 22), or use the gradual‑depletion selection of claim 17.

Dependent claims — structure and notable features

• Claims 2–16 depend (directly or indirectly) on claim 1; claims 19–25 mirror them on claim 18.
• Promoter pairs: claims 3 and 5 (and 20, 22) specify the SV40 early region promoter (weak, for GS) paired with the hCMV‑MIE promoter (strong, for the heterologous gene).
• Orientation/position rules: claim 2/19 — GS promoter placed upstream of the heterologous gene; claim 4/21 — GS promoter directs expression in the opposite direction; claims 6, 7, 23, 24 — heavy‑chain gene downstream of GS with the light‑chain promoter oriented oppositely; claims 8 and 25 — all three genes (GS, light chain, heavy chain) transcribed in the same direction with GS upstream.
• GS promoter source: claim 9 (SV40 early region), claim 10 (hCMV‑MIE), claim 11 (derived from pSV2GS), claim 12 (derived from pCMGS).
• Claim 13–14: all genes expressed from the same type of promoter, specifically hCMV‑MIE.
• Claim 16: vector is pSV2GScLccHc or pST6.

Claim‑drafting anomalies worth noting (interpreted literally):

• Claims 15 and 16 are written as "The vector of claim 14" and "The vector of claim 8" even though claims 8 and 14 are method claims in the chain from claim 1 — a claim‑type/antecedent mismatch on the face of the text.
• Claim 8 (dependent on claim 1, which is expressly limited to a "myeloma cell line") recites "heterologous to said lymphoid cell line" — an antecedent‑basis mismatch relative to its parent claim.

These are textual features of the granted claims as printed, not corrections I am making.

Uncertainties

1. No verified 2026 CAFC docket for this patent was found; the 2016 expiration makes such activity unlikely, but I did not access a live CAFC docket system to confirm a negative.
2. Assignee chain includes both Lonza Licences AG and Lonza Swiss Licences AG plus Healthcare Royalty Partners II LP per recorded security/licence instruments; the Google Patents listing itself warns assignee data "may be inaccurate."
3. The 1988‑04‑18 vs. 1990‑01‑25 priority inconsistency noted above is unresolved from the sources available.
4. I could not independently confirm the Maryland District Court 8:07‑cv‑03177 parties, claims asserted, or outcome.

I'll search for litigation involving US patent 5891693.

The initial searches were noisy. Let me search more specifically.

Found the key case. Let me dig into outcomes and check for other suits.

Litigation Involving U.S. Patent No. 5,891,693

I ran targeted searches (Google Patents litigation record, Unified Patents litigation portal, Justia dockets, patentdocs/biospace reporting, SEC filings) for the specific number 5,891,693 ("Recombinant DNA methods vectors and host cells"). Here is what I can confirm, and where I cannot confirm.

1. Confirmed case

Lonza Group AG v. Northwest Biotherapeutics, Inc.

Patents asserted in this complaint (8 in total): U.S. 5,122,464; 5,591,639; 5,658,759; 5,770,359; 5,827,739; 5,879,936; 5,891,693; and 5,981,216. The assertion was based on Northwest's manufacture, use, and sale of its DCVax® products, including DCVax®-Prostate.

Important prosecution/litigation history around this case:

• Lonza first sued Northwest on July 27, 2007 in the U.S. District Court for the District of Delaware, alleging that all of NWBT's DCVax® products infringed "certain gene expression technology patents." I could not confirm from the sources retrieved that 5,891,693 specifically was asserted in that first Delaware complaint. Lonza did not serve NWBT within the 120-day period; NWBT filed an answer and counterclaims (Delaware Deceptive Trade Practices Act, unfair competition, tortious interference, abuse of process). The parties then agreed to dismiss the entire Delaware action (complaint and counterclaims), and Lonza simultaneously refiled in Maryland (the 8:07-cv-03177 case above, which does list 5,891,693).
• On December 13, 2007, Lonza filed an Amended Complaint in Maryland withdrawing all infringement allegations as to NWBT products other than DCVax®-Prostate. NWBT publicly disputed the remaining claim, stating it never used Lonza's expression system and that only an unrelated third-party contract manufacturer had used a gene-expression system (to make two batches of recombinant PSMA protein) years earlier. (Source: BioSpace / PRNewswire item, Dec. 20, 2007.)

⚠️ Caveat: The sources I retrieved establish the case, parties, jurisdiction, case number, filing date, and the fact of amendment/settlement activity, but I could not retrieve a definitive final disposition order (e.g., dismissal with/without prejudice, consent judgment, or settlement terms) for 8:07-cv-03177. I will not fabricate a specific outcome.

2. Possible related case — cannot confirm 5,891,693 was asserted

Lonza Sales AG v. Progenics Pharmaceuticals, Inc., No. 1:15-cv-00562-UNA (D. Del.), referenced in CytoDyn Inc.'s 2015 licence agreement with Lonza. In the agreement, CytoDyn's payment obligations were expressly tied to "Lonza's claims asserted in the case of Lonza Sales AG vs. Progenics Pharmaceuticals, Inc." The same agreement's patent schedule lists U.S. 5,891,693 (expiry 06.04.16) as part of Lonza's licensed "Patent Rights."

• This strongly suggests the Progenics suit involved the same GS/glutamine-synthetase patent family, but the documents I retrieved do not expressly state which patents were asserted in that case. I therefore list it as possibly involving 5,891,693 but not confirmed.

3. What I did not find

• No CAFC appeal specifically identified for a case in which 5,891,693 was the asserted patent.
• No IPR, PGR, reissue, or ex parte reexamination of 5,891,693 identified.
• No other affirmative infringement suits naming 5,891,693 identified.

Summary

The only confirmed litigation involving U.S. Patent No. 5,891,693 is:

Lonza Group AG v. Northwest Biotherapeutics, Inc., No. 8:07-cv-03177 (D. Md.), filed Nov. 27, 2007 — patent infringement under 35 U.S.C. § 271, asserting 5,891,693 among eight Lonza gene-expression patents, targeting NWBT's DCVax®-Prostate. The case was narrowed to only DCVax®-Prostate by amended complaint (Dec. 13, 2007) and is no longer active; the asserted patents expired in April 2016. A definitive disposition document was not retrieved.

Note also the underlying IP was later administered by Lonza Swiss Licences AG / Lonza Licences AG, with Healthcare Royalty Partners II, L.P. holding a security interest (per the Google Patents reassignment record for this patent) — meaning any past recovery ran through that ownership chain, though no additional assertion of 5,891,693 itself was found.

If you need certainty on the final disposition of 8:07-cv-03177 or the patent-by-patent assertion list in the Delaware and Progenics cases, the next step would be a PACER docket pull and a review of the complaints themselves, which are not free-web-searchable in full text.

I'll verify the structured-data finding with targeted searches before concluding.

Let me run a few more targeted checks before concluding.

PTAB / AIA Trial Proceedings — US 5,891,693

Proceedings overview

Total AIA trial proceedings on file: zero (0). The structured USPTO Open Data Portal ("PTAB proceedings on file") block returns no inter partes review, post-grant review, or covered business method proceeding for US 5,891,693 — no institution-denied cases, no settled cases, no final written decisions, and therefore no claims canceled and none sustained. The bottom-line defensive posture is therefore not "the patent survived two IPRs and is hardened," and also not "claims 1–5 were canceled." It is: the patent was never tested at the Board, and it expired on 2016‑04‑06 (anticipated expiration per the record), so there is no live IPR target and no § 315(e)(2) estoppel in place. Any defensive strategy built on "we'll IPR it" is moot; an invalidity position would have to be run in district court, and only if there is a pre‑2016 damages theory on the table.

Verification performed

I could not query the live PTAB E2E system or the ODP API directly in this session, so I corroborated the structured data with targeted web searches for any proceeding naming this patent (searches for the patent number paired with IPR/PGR/CBM identifiers, Lonza/Celltech GS-selection challenges, and Board dockets). No proceeding number, petition, institution decision, FWD, or Board appeal involving 5,891,693 surfaced. Consistent with that, the only litigation-of-record for this patent is the district court action discussed in the earlier section (Lonza Group AG v. Northwest Biotherapeutics, Inc., No. 8:07‑cv‑03177 (D. Md.)), which never migrated to the Board.

No proceedings to list. Per the instruction not to invent proceeding numbers, I am not fabricating placeholder IPR entries. The canonical PTAB list is empty.

⚠️ Contradiction / refinement vs. the previously generated section

The earlier litigation section said it "could not confirm from the sources retrieved that 5,891,693 specifically was asserted" in the first Lonza v. Northwest Biotherapeutics action in Delaware. A retrieved district-court AO‑120 form (Patent/Trademark Report to Commissioner, D. Del.) for Case 1:07‑cv‑00467‑GMS, plaintiff Lonza Group AG v. Northwest Biotherapeutics, Inc., filed 2007‑07‑27, lists eight patents, and 5,891,693 is one of them (with "Aluguisse Holdings A.G." as holder of record — as spelled in the form). This resolves the prior open question: 5,891,693 was listed in the Delaware filing as well. This strengthens, rather than contradicts, the earlier conclusion that the patent was asserted in the 2007 Lonza/NWBT campaign; it does not create any PTAB footprint.

Strategic summary

Claim status: everything is UNTESTED at the Board. No claim of 5,891,693 — independent claims 1, 17, 18 or dependents 2–16 and 19–25 — has ever been canceled, confirmed, or construed in an AIA trial. There is no FWD to cite. Anyone who tells you a claim of this patent "was invalidated in IPR" is wrong; anyone who tells you it was "upheld in IPR" is equally wrong. The absence of Board activity is explained by timing: the patent issued 1999‑04‑06 and expired 2016‑04‑06, and the AIA trial regime began 2012‑09‑16. The patent was thus eligible for roughly three and a half years of IPR availability, and no competitor chose to file — a plausible reflection of the fact that the dispute that did occur was resolved in district court (Maryland), where the case was narrowed by amended complaint and went inactive long before expiration.

Estoppel landscape: clean. Because no IPR/PGR was ever instituted, § 315(e)(2) estoppel does not attach to anyone on this patent. There is no petitioner, no privy, no real party in interest bound to any ground the Board addressed or "reasonably could have raised." For a defendant being darted today, that means there is no PTAB-imposed narrowing of the invalidity case and no estoppel exposure — but it also means there is no free roadmap (no FWD-credited art, no Board claim constructions) to borrow. Note separately that § 315(b)'s one-year bar would apply to anyone served with a complaint alleging infringement more than a year ago, but that is academic here: the patent term ended 2016‑04‑06, the earlier section records the record status as "Expired – Lifetime," and there is no ongoing enforceable right to infringe for post‑expiration conduct.

Pattern signals. No serial petitioner, no defensive aggregator (Unified Patents et al.) in the PTAB chain for this patent, and no patent-owner PTAB appeal strategy. The "Family has litigation" / Darts‑IP flag on the Google Patents record and the Unified Patents case link both point to the Maryland district court action, not to any Board proceeding. The ownership/royalty administration trail (Lonza Licences AG, Lonza Swiss Licences AG, Healthcare Royalty Partners II LP security interest) is a monetization signal from the litigation/licensing side, not a PTAB one.

Recommended next steps

• If a demand letter or complaint asserts 5,891,693 against you: the threshold response is temporal, not merits-based. The patent's anticipated expiration is 2016‑04‑06; absent a laches/damages theory reaching pre‑2016 conduct, there is no ongoing royalty base. Confirm the accused-conduct dates before spending anything on validity.
• Do not plan an IPR. There is no live proceeding, and the 2016 expiration plus (for most defendants) the § 315(b) one-year bar makes a petition pointless. Nothing in the structured PTAB record gives you an institution or FWD to lean on.
• If pre‑2016 damages exposure is real, the invalidity case has to be built in district court from scratch — the Board never construed these claims, so there is no FWD claim construction to import. The relevant prior art is the EP‑A‑0 256 055 (Celltech) family and the eight references on the face of the patent (US 4,399,216; US 4,656,134; US 4,797,359; WO 86/06409; WO 87/04462; US 5,122,464; EP 0 244 672; EP 0 255 320), as catalogued in the earlier the patent-record sections.
• Verify the negative independently before relying on it in a brief. Pull the patent's proceeding list directly from PTAB E2E (https://ptab.uspto.gov/) and the USPTO Open Data Portal, and check CourtListener (https://www.courtlistener.com/) for any Federal Circuit appeal of a Board decision naming 5,891,693. My searches found none, but I could not query those systems live, so treat "zero proceedings" as the strong probability it is rather than a certified docket negative.

I'll attempt to pull the actual USPTO Assignment Center records (reel/frame, correspondent) before writing anything.

Search results are noisy. Let me try more targeted queries on the specific recorded instruments.

Let me make a few more attempts to retrieve the actual reel/frame and correspondent fields.

I was unable to retrieve the primary-source Assignment Center records themselves — my only retrieval channel for this task was general web search, which returned no reel/frame numbers and no correspondent fields for US 5,891,693. Everything below is reconstructed from the USPTO assignment records as mirrored in the Google Patents legal-events list for this patent (those events are transcribed from USPTO assignment data, including conveyance type, assignor and assignee). Reel/frame numbers and correspondent-of-record identities are a gap in this analysis, not a "none recorded" finding. I have not guessed or back-filled them. I flag below exactly what is missing and what would close it.

Inventors

Pattern assessment — no inventor-departure fire-sale signal. All three are Celltech personnel and the patent stayed with Celltech for ~7.5 years post-filing. The 1997 transfer was not preceded by inventor attrition triggering a portfolio sale; it was a divestiture of the whole Celltech Biologics business unit to Alusuisse-Lonza, with the patent riding along as part of the transferred technology. Note the caveat: I could not independently verify each inventor's employment dates from primary HR/corporate records; the employer attribution rests on the assignment record plus the patent's own reference list.

Original assignee

Two entities are relevant, and the record is internally awkward:

• Applicant of record at filing (1990): Celltech Limited / Celltech Therapeutics Limited (UK). Celltech is the originator of the glutamine synthetase (GS) selectable-marker expression system at issue — the patent's own background cites Celltech's earlier EP-A-0 256 055. Celltech was a UK biotech operating company; it commercialised the GS system by licensing, not by selling a gene-therapy product. Celltech Biologics was its contract-manufacturing arm.
• Assignee of record at issue (1999): Alusuisse Holdings AG (Swiss) — the Google Patents "Original Assignee" field names this entity, which is why it appears as the original assignee even though Celltech filed the case.

Business context / status:

• Celltech sold Celltech Biologics to Alusuisse-Lonza in the 1996–97 window (contemporaneous trade press describes Alusuisse-Lonza's purchase of Celltech's interest in Celltech Biologics, with further earn-out payments tied to the former Biologics division's sales). Celltech Therapeutics Limited was later subsumed into Celltech Group plc, which was acquired by UCB S.A. in 2004. So Celltech as a distinct entity no longer exists; it is not a dissolved/bankrupt seller but an acquired one.
• Alusuisse Holdings AG → Alusuisse Lonza Group AG → Lonza Group AG. Lonza is today a major operating CDMO that still runs a real GS Gene Expression System licensing business (multi-party licence agreements appear in SEC filings of Acumen, Tracon, Context Therapeutics, Avidity, NGM Bio, Vaxcyte and others). This matters directly to the verdict below: the owner at the assertion date was a genuine operating company with a live licensing programme, not a shell.

Assignment timeline

Reel/frame: NOT RETRIEVED for any entry. Correspondent of record: NOT RETRIEVED for any entry. The dates, conveyance types, assignors and assignees below are from the USPTO-derived legal-events list for US 5,891,693.

• 1997-08-19 (recorded) — Reel not retrievable

  • Conveyance: Assignment of assignors' interest
  • Assignor: Celltech Therapeutics Limited
  • Assignee: Alusuisse Holdings A.G.
  • Correspondent: not retrievable
  • Context: Divestiture — transfer of the Celltech Biologics business (and this patent with it) to Alusuisse-Lonza.

• 2001-07-26 (recorded) — Reel not retrievable

  • Conveyance: Assignment of assignors' interest
  • Assignor: Alusuisse Lonza Group, AG
  • Assignee: Lonza Group, AG
  • Correspondent: not retrievable
  • Context: Internal reorganisation / group renaming — Alusuisse Lonza Group → Lonza Group after the packaging business went to Alcan.

• 2001-11-01 (recorded) — Reel not retrievable

  • Conveyance: Merger
  • Assignor: Alusuisse Holdings, AG
  • Assignee: Alusuisse Lonza Group, AG
  • Correspondent: not retrievable
  • Context: Internal reorganisation — upstream entity absorbed into the group company.
  • ⚠️ Chronology anomaly: as displayed, the 2001-07-26 transfer to "Lonza Group, AG" is dated earlier than the 2001-11-01 merger creating "Alusuisse Lonza Group, AG," which the 2001-07-26 record treats as the assignor. Execution vs. recording dates may diverge, or the listing order is unreliable. Flagged, not corrected.

• 2006-08-01 (recorded) — Reel not retrievable

  • Conveyance: Security Agreement
  • Assignor: PharmAthene, Inc.
  • Assignee: MPM BioVentures III-QP, L.P., as Administrative Agent
  • Correspondent: not retrievable
  • Context: Securitisation — but see the anomaly note below; this record does not sit naturally in the Lonza title chain.
  • ⚠️ Unresolved anomaly: PharmAthene, Inc. is a US biodefense company backed at the relevant time by MPM Capital; it is not an assignee of 5,891,693 anywhere in the chain. The most plausible explanations are (a) a blanketed "all intellectual property" security grant that swept in licensed rights (PharmAthene was a Lonza Biologics customer), or (b) a record-mapping artifact, since one assignment record can cover many properties. I could not verify which. This is the single entry in the chain I would not rely on.

• 2010-10-22 (recorded) — Reel not retrievable

  • Conveyance: License
  • Assignor: Lonza Group AG
  • Assignee: Lonza Swiss Licences AG
  • Correspondent: not retrievable
  • Context: Internal reorganisation / carve-out — Lonza moves its licensing rights into a dedicated Swiss vehicle.

• 2010-10-22 (recorded) — Reel not retrievable

  • Conveyance: Conditional Assignment
  • Assignor: Lonza Group AG
  • Assignee: Lonza Swiss Licences AG
  • Correspondent: not retrievable
  • Context: Internal reorganisation — conditional (trigger-based) transfer of the licensed IP, paired with the licence above.
  • Corroboration: the Swiss commercial register entry for Lonza Swiss Licences AG (CHE-116.039.014) states its purpose is to hold rights under an Intercompany Licence Agreement and Conditional Assignment Agreement with Lonza Group AG, including as successor to Lonza Sales AG under the Regeneron, Glaxo Group, Genmab, Alexion and Biogen licence agreements. That is the same instrument pair recorded here.

• 2013-12-27 (recorded) — Reel not retrievable

  • Conveyance: Assignment of assignors' interest
  • Assignor: Lonza Group AG
  • Assignee: Lonza Licences AG; Healthcare Royalty Partners II, L.P.
  • Correspondent: not retrievable
  • Context: Royalty monetisation — assignment of the licensing/royalty rights into a newly formed entity alongside the royalty fund.

• 2013-12-27 (recorded) — Reel not retrievable

  • Conveyance: License
  • Assignor: Lonza Group AG
  • Assignee: Lonza Licences AG
  • Correspondent: not retrievable
  • Context: Royalty monetisation (intercompany licence feeding the assignment above).

• 2013-12-27 (recorded) — Reel not retrievable

  • Conveyance: Security Agreement
  • Assignor: Lonza Licences AG
  • Assignee: Healthcare Royalty Partners II, L.P.
  • Correspondent: not retrievable
  • Context: Securitisation — the fund takes security over the assigned royalty-bearing IP.
  • Corroboration: Lonza Licences AG (CHE-323.471.004, statute dated 09.12.2013) has a registered purpose that is purely holding/licensing and expressly prohibits granting security except under the "Intellectual Property Security Agreement and Pledge" with HEALTHCARE ROYALTY PARTNERS II, L.P. The registration date (Dec 2013) matches the 2013-12-27 recording.

• 2014-06-26 (recorded) — Reel not retrievable (three records)

  • Conveyance: Amended License; Amended Conditional Assignment; Amended Conditional Assignment
  • Assignor: Lonza Group AG
  • Assignee: Lonza Swiss Licences AG
  • Correspondent: not retrievable
  • Context: Internal reorganisation — amendment only; no change in ultimate control.

• 2016-04-06 — Anticipated expiration (Google Patents legal status). No assignments recorded after this date.

Correspondent gap: because not one correspondent entry was retrievable, the task's central discriminator — the repeat-player attorney across the chain — cannot be run on this patent. I am explicitly declining to name a correspondent rather than infer one from "who usually files for Lonza."

Timeline diagram

timeline
    title Ownership of US 5891693
    1990 : Priority filing date
         : Celltech files the application
    1995 : Continuation application filed
    1997 : Celltech Therapeutics assigns to Alusuisse Holdings
    2001 : Alusuisse Holdings merges into Alusuisse Lonza Group
         : Alusuisse Lonza Group assigns to Lonza Group AG
    2006 : PharmAthene security agreement to MPM BioVentures III-QP
    2010 : Lonza Group AG licenses to Lonza Swiss Licences AG
         : Conditional assignment to Lonza Swiss Licences AG
    2013 : Lonza Group AG assigns to Lonza Licences AG
         : Healthcare Royalty Partners II takes security interest
    2014 : Amended licence and conditional assignment recorded
    2016 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — Present in form, benign on the evidence.
Lonza Swiss Licences AG and Lonza Licences AG are genuinely single-purpose, product-free Swiss vehicles — their registered purposes are holding/licensing only and exclude other business (NorthData/Lixt/graph.swiss register entries, CHE-116.039.014 and CHE-323.471.004). That satisfies the "no products, licensing-only" tell. But the tell is normally paired with an arm's-length acquisition by an unrelated acquirer; here both vehicles are captive Lonza subsidiaries created for intra-group carve-out and royalty monetisation (recorded 2010-10-22 and 2013-12-27). The counterparty that is not a Lonza affiliate — Healthcare Royalty Partners II, L.P. — takes a security interest plus an assignment of the royalty stream, i.e. a royalty-fund monetisation, not a patent-acquisition-for-assertion play. Marked present but not probative of NPE conduct.

2. Known asserter in the chain — Not present.
No assignee in the chain matches Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity. The counterparties are Celltech (UK biotech), Alusuisse/Lonza (Swiss industrial → CDMO), and Healthcare Royalty Partners (royalty fund). Cited against the 1997-08-19, 2001-07-26, 2001-11-01, 2010-10-22, 2013-12-27 and 2014-06-26 entries.

3. Repeat correspondent across the chain — Unclear / not retrievable.
No correspondent-of-record data was retrievable for any of the twelve recorded events. This is the one signal I most wanted and could not obtain. A PACER-independent route to close it is a manual reel/frame lookup on Assignment Center (link below) for each of the dated events above.

4. Cascading transfers — Partially present.
Two same-day clusters: 2010-10-22 (licence + conditional assignment to Lonza Swiss Licences AG) and 2013-12-27 (assignment to Lonza Licences AG + HCR II, licence, security agreement to HCR II). These are rapid multi-instrument cascades, and the 2013-12-27 cluster does run Lonza Group AG → Lonza Licences AG → HCR II in a single day. But the cascade spans three years, not "<24 months through chained LLCs," and all parties are disclosed affiliates or a named fund. Marked partially present, intra-group.

5. Pre-litigation transfer — Not present.
The first suits naming the patent family were filed 2007-07-27 (D. Del.) and 2007-11-27 (D. Md., No. 8:07-cv-03177). The nearest preceding recorded assignment is 2001-11-01 — over five years earlier. The only near-window event is the 2006-08-01 PharmAthene/MPM security agreement (~16 months before the Delaware filing), which is outside the 6-month window, is a security interest rather than a title change, and is the anomalous entry flagged above. No pre-litigation transfer.

6. Bankruptcy fire-sale — Not present.
Neither Celltech (acquired by UCB in 2004) nor Alusuisse/Lonza (name change and reorganisation) entered bankruptcy proceedings in connection with these transfers. The 1997 Celltech Biologics transfer was a strategic divestiture, and trade press confirms earn-out payments tied to the division's sales — inconsistent with a distressed fire-sale. (PharmAthene later underwent distress and merged into Altimmune, but its connection to this patent is unverified, so it is not counted.)

7. Privateering — Not present; the inverse is present.
The 2007 assertion in Lonza Group AG v. Northwest Biotherapeutics, Inc. was brought by Lonza Group AG itself, the operating parent that owns and licenses the GS system. Assertion by the operating owner against a counterparty is the opposite of transferring to an NPE to assert on one's behalf. No evidence of a Lonza→NPE privateering transfer appears anywhere in the chain.

8. Defensive aggregator — Not present.
The chain terminates at Lonza Licences AG with Healthcare Royalty Partners II, L.P. holding a security interest. No RPX, AST, LOT Network, Unified Patents or OIN entity appears. The patent was not neutralised by defensive aggregation; it expired on 2016-04-06 while still in the Lonza royalty structure.

Verdict

Operating-company assertion.

The chain runs from a genuine technology originator (Celltech) through a real industrial acquirer (Alusuisse-Lonza → Lonza Group AG) that operates a live GS Gene Expression System licensing business with numerous named licensees, and the patent was asserted directly by Lonza Group AG as plaintiff (Maryland No. 8:07-cv-03177, filed 2007-11-27, asserting 5,891,693 among eight Lonza gene-expression patents). The 2010-10-22 and 2013-12-27 entries are a captive intra-group carve-out plus royalty-stream monetisation — Lonza Licences AG is a Lonza subsidiary whose registered purpose is limited to holding the licensed IP and which may pledge assets only under the Healthcare Royalty Partners II security instrument — followed by the 2014-06-26 amendments, which change no ultimate control. There is no arm's-length transfer to an unrelated asserter, no pre-litigation transfer within six months, and no defensive-aggregator terminus.

Residual uncertainty, stated plainly: (i) if the 2006-08-01 PharmAthene / MPM BioVentures III-QP security record is a genuine grant over this patent rather than a mirroring artifact, the chain has an unexplained third party in it; and (ii) absent correspondent data for any of the twelve events, signal 3 could not be tested at all. Neither gap is sufficient to move the verdict away from operating-company assertion, because the assertion itself was made by the operating parent and the downstream entities are disclosed Lonza affiliates and a royalty fund.

Verify / close the gaps here: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (legacy interface: https://assignment.uspto.gov/patent/index.html). Search by patent number 5891693 and capture the reel/frame and correspondent for each dated event above; also pull D. Md. 8:07-cv-03177 via PACER for the disposition.

Cross-reference note versus the earlier sections: this reconstruction is consistent with the previously generated litigation and summary sections on parties, dates and the 2016 expiration. One item to reconcile: the earlier summary lists the 2006-08-01 PharmAthene/MPM record without comment, whereas I treat it as the chain's one genuinely anomalous entry. The unresolved 1988-04-18 vs 1990-01-25 priority inconsistency flagged earlier remains unresolved and is unaffected by the assignment analysis.

I'll search for the specific patent and its cited prior art references.

Let me gather details on the remaining cited references.

Prior Art Analysis — US 5,891,693 ("Recombinant DNA methods vectors and host cells")

A. USPTO lookup note (read first)

I ran targeted searches against the USPTO material available to me and against the authoritative full text already supplied. A point of candor: my searches did not return a standalone USPTO PatentCenter/Pair page dedicated to 5891693; the record is corroborated by (i) the Google Patents authoritative text supplied for this task, and (ii) USPTO-derived third-party records for the closely related Celltech filings (e.g., the uspto.report grant record for US 5,122,464 / application 07/595,733). I therefore treat the provided full text as ground truth and do not auto-correct any identifier. Per your strict rule, I interpret all numbers literally.

The eight references below are the ones listed on the face of US 5,891,693 under "Patent Citations (8)" — all marked as cited by the examiner (*). Two of them (WO 87/04462 and US 5,122,464) are members of the same Celltech glutamine-synthetase (GS) family that the 5891693 specification itself discusses by number as EP‑A‑0 256 055. That intra-family citation is important context: the examiner was aware of applicant's own earlier GS work.

B. Critical art-date framing for § 102

For every one of these references to anticipate (not merely be relevant), its effective prior-art date must precede the 5891693 invention date. The 5891693 record gives a priority date of 1990‑01‑25 (with the family-level inconsistency — US 07/898,165 shown as priority 1988‑04‑18 — flagged in the earlier section; that earlier date, if correct, only narrows the usable art window). Two consequences:

• US 5,122,464 issued 1992‑06‑16, which is after the 1990 priority date. It can only be § 102(e) art (a US patent granted on an application whose effective filing date predates the invention). Its chain runs to PCT/GB87/00039, filed 1987‑01‑23, which comfortably predates 1990 — so it is available, but only as § 102(e) art, and only for what it enables.
• All other seven references published between 1983 and 1989, i.e., all pre‑date the 1990 priority date and are available as § 102(a)/(b) printed publications.

Bottom line up front: none of the eight references, taken alone, discloses every element of any independent claim (1, 17, 18). The only references with genuine anticipation potential are the two Celltech GS references, and their reach is strongest against claims 1 and 18 (and their mirror dependents) — not against claim 17's specific dilution protocol.

C. Reference-by-reference analysis

D. Claim-by-claim anticipation assessment

Independent claim 1 — transformation of a myeloma cell line with a vector carrying (a) a GS gene and (b) a heterologous gene, arranged so GS is expressed and glutamine‑independent colonies arise.

• WO 87/04462 (#7): discloses a vector encoding GS plus a desired protein, used as a dominant selectable marker, and expressly extends use to myeloma/hybridoma cells and to conferring glutamine independence. This comes closest to a § 102(a)/(b) attack on claim 1. Weakness: the reference frames GS under a regulatable promoter and emphasizes co‑amplification/dominant‑marker selection; it is not a clean, express disclosure of the exact "GS‑expressed‑so‑colonies‑arise" arrangement limitation, nor of the specific strong/weak promoter geometry. Expect this to be a § 102 or § 103 battleground, not a clean anticipation.
• US 5,122,464 (#8): same family; as § 102(e) art it supplies GS vector + desired‑protein + glutamine/GS‑inhibitor selection in lymphoid cells. Same limitation gap as #7.
• All other references (#1–#6): no GS gene, no glutamine‑independence element → cannot anticipate claim 1.

Independent claim 17 — the three‑step selection protocol (plate in glutamine‑containing medium; at 24 h add two volumes of glutamine‑free medium; recover colonies at 7 days = "Protocol A").

• No cited reference discloses this protocol. None of #1–#8 describes the specific 1‑volume/2‑volume dilution timing. Claim 17 is the least exposed of the three independent claims to this art. Its novelty rests on the specification's own finding (Table 2) that gradual depletion (Protocol A, 130 colonies/10⁶) vastly outperforms abrupt glutamine withdrawal (Protocols B/D/E, 3–8 colonies/10⁶).
• The Celltech references describe MSX‑based amplification and glutamine‑free growth generally, but not this graduated‑dilution plating scheme.

Independent claim 18 — same as claim 1 but for a "lymphoid cell line" generally.

• Same analysis as claim 1, and #7/#8 are even more directly on point because they speak to lymphoid cells (myeloma/hybridoma) rather than the narrower "myeloma" recitation. If WO 87/04462 is held to anticipate claim 1, it would a fortiori reach claim 18 (the broader genus). Contradiction to flag: claim 8 (dependent on claim 1, which is myeloma‑limited) recites "said lymphoid cell line" — a mismatch already noted in the earlier claims section; this mismatch has no bearing on the prior‑art analysis but is worth remembering if claim scope is construed.

Dependent claims 2–5 and 19–22 (strong heterologous promoter + weak GS promoter; SV40‑early/HCMV‑MIE pair):

• EP 0 255 320 (#6) teaches a CMV promoter driving a non‑Ig gene in myeloma cells — relevant to the "strong promoter" element (claims 3/5/20/22), but silent on a weak GS promoter paired therewith.
• WO 87/04462 (#7) contemplates regulatable GS promoters, not the SV40‑early "weak‑promoter" solution. No anticipation; § 103 combination possible.

Dependent claims 6–8 and 23–25 (heavy/light chain gene orientation relative to GS):

• #6 (EP 0 255 320) concerns non‑Ig genes, so it is off‑point for the Ig heavy/light orientation limitations. #5 (EP 0 244 677) and #6 involve myeloma expression of heterologous genes but neither teaches the specific GS/light‑chain/heavy‑chain orientation geometry. No anticipation.

Dependent claims 9–12 (GS from SV40‑early or hCMV‑MIE; GS/promoter derived from pSV2GS or pCMGS):

• US 5,122,464 (#8) and WO 87/04462 (#7) disclose GS expression constructs (including SV40‑driven GS in the family's pSV2.GS‑type work). These are the most relevant to claims 9 and 11; the hCMV‑MIE/GS combination (claims 10, 12) is the applicant's own later development and is not squarely in the cited references. Possible § 102(e) for claim 9/11 via #8; not for 10/12.

Dependent claims 13–14 (all genes from the same promoter, specifically hCMV‑MIE):

• #6 (EP 0 255 320) teaches CMV promoter use in myeloma but does not teach using the same promoter type for both GS and the heterologous gene. No anticipation.

Dependent claims 15–16 (vector‑type / specific plasmids pSV2GScLccHc, pST6):

• These are specific structural claims; no cited reference discloses these plasmids. No anticipation. (Note the claim‑type anomalies flagged earlier — claims 15 and 16 are drafted as "The vector of…" inside a method chain; that defect is independent of prior art.)

E. Summary ranking of the cited art

1. WO 87/04462 A1 (#7) and US 5,122,464 A (#8) — Celltech GS family. The only references that reach the GS‑plus‑glutamine‑independence core of claims 1 and 18. #7 is § 102(a)/(b) art (pub. 1987); #8 is § 102(e) art (effective filing 1987‑01‑23). These are the references a challenger would lead with, and the ones the applicant's own specification distinguishes as EP‑A‑0 256 055.
2. EP 0 255 320 A2 (#6, Genzyme) and EP 0 244 677 A2 (#5, Daiichi). Relevant to the myeloma host and CMV‑promoter heterologous‑gene elements (claims 1/18, 3/5/10/12/14/20/22), but contain no GS/glutamine‑independence teaching — § 103 material, not § 102.
3. US 4,399,216 (#1), US 4,656,134 (#2), WO 86/06409 (#4). Foundational dominant‑marker/co‑amplification background (Axel‑Wigler cotransformation; Ringold amplification; Genetics Institute amplification). Support a general "amplifiable selectable marker" motivation but do not anticipate any claim.
4. US 4,797,359 (#3). Yeast heat‑shock expression — largely peripheral; relevant only as generic regulatable‑promoter background.

Non‑patent citations on the face of the patent (Sanders & Wilson 1984 EMBO J. 3:65‑71 on hamster GS cloning; Young & Ringold 1983 JBC; de Saint Vincent 1981 Cell; Kaufman 1986 PNAS; Foecking & Hofstetter 1986 Gene 45:101‑105 on promoter units) supply background on GS cloning and promoter/enhancer units but likewise do not disclose the claimed lymphoid‑cell GS selection arrangement.

F. Explicit uncertainties / caveats

• I could not retrieve a native USPTO PatentCenter page for 5891693; the verified record rests on the authoritative full text provided plus corroborating USPTO‑derived records for sibling cases. I have not fabricated an independent USPTO citation.
• The reference EP 0 255 320 appears with kind codes A2 (1988‑02‑03) and A3 (1989‑09‑06); both pre‑date the 1990 priority date. I list the A2 publication as the primary citable document.
• EP 0 244 677 returned no full‑text fetch (search step limit); my description derives from its title/abstract‑level records and the JP counterpart (JPS63126483A). The specific selectable markers (gpt, tk) I attribute to it come from the closely related Daiichi disclosure (US 5,019,499 record retrieved), so treat the marker detail as corroborated but not from the EP document itself.
• Whether WO 87/04462 / US 5,122,464 rise to anticipation or only obviousness of claims 1/18 turns on whether their GS‑vector disclosure is read to expressly disclose the claimed promoter‑geometry/expression arrangement. I assess this as genuine § 102 potential but not a clean anticipation; a full anticipation finding would require reading the reference's regulatable‑promoter GS teaching into the claim's "arranged such that GS can be expressed" limitation, which I flag as contestable.

I'll verify the content and teaching of the key cited references before building the §103 combinations.

Key prior-art content confirmed. Let me now verify the myeloid/lymphoid host-cell references and the state of the art on transcriptional interference.

Two final verification checks: the identity of the Foecking enhancer-promoter unit cited on the face of the patent, and the dates of the closely-related Celltech "progressive depletion" disclosure.

Obviousness Analysis — U.S. 5,891,693 under 35 U.S.C. § 103

This builds on the earlier sections' bibliographic, claim, litigation and PTAB treatment. I do not repeat them. Where a new finding sharpens or corrects an earlier statement, I flag it.

1. Analytical frame and the one date that matters

This is a pre-AIA § 103 case: the application (08/376,380) was filed 1995-01-23 as a continuation, ultimately tracing to 07/460,154 filed 1990-01-25 (the priority date shown on the record). Per In re Hilmer, a foreign priority claim cannot be used to antedate a § 102(b) reference, so the operative § 102(b) critical date is 1989-01-25 on the face of the record (or as early as 1987-04-18 if the 1988-04-18 priority date appearing in the family listing for 07/898,165 is the true earliest U.S. filing — the inconsistency the earlier section flagged as unresolved). Every reference relied on below predates even the later (1989-01-25) date, so the combination analysis is date-robust.

The Graham v. John Deere factors control: scope/content of the prior art, differences from the claims, PHOSITA level (here: a molecular biologist with several years of mammalian expression-vector experience — the specification's own level of routine practice, e.g. electroporation, linker conversion of restriction sites, Southern blotting, ELISA), and objective evidence. KSR supplies the legal gloss: a finite number of identified, predictable solutions, with a documented reason to pursue them, yields obviousness; so does "a known technique to improve similar devices in the same way" and "obvious to try."

2. What each reference actually teaches (verified)

3. Combination 1 — Claims 1 and 18 (the independent method claims)

Proposed combination: WO 87/04462 / EP 0 256 055 / US 5,122,464 (GS gene as dominant selectable marker, vectors, transformation to glutamine independence, co-amplification of Ig/tPA/hGH/TIMP) + EP 0 244 672 (myeloma host, product gene and marker on one plasmid) + US 4,399,216 (transformation methodology).

Claim 1 requires only: (a) a myeloma host; (b) a vector with a GS gene and a heterologous gene; (c) the genes arranged so GS can be expressed and glutamine-independent colonies produced.

• (a) is met by EP 0 244 672, which names P3-X63-Ag8 and SP2/0-Ag14 — the very lineage (P3 plasmacytoma) from which the patent's NSO and P3-X63Ag8.653 hosts derive.
• (b) is met by WO 87/04462, which claims a vector "capable, in a transformant host cell, of expressing both a recombinant DNA sequence which encodes an active GS enzyme and the recombinant DNA sequence which encodes the complete amino acid sequence of the desired protein other than GS," and names immunoglobulin polypeptides among the products.
• (c) is met by the combination, because WO 87/04462's stated object is precisely to confer glutamine independence by transformation with a GS vector.

Motivation (documented in the references themselves, not inferred):

1. EP 0 256 055's own stated problem is that DHFR/MTX requires a DHFR⁻ host ("at present only one cell line known which lacks the gene encoding DHFR") and a toxic drug ("MTX is a potential carcinogen"). Its stated object is "a dominant selectable marker which was applicable to a wide variety of cell lines." A PHOSITA reading that is directed to try the GS system in further cell types — including myelomas.
2. EP 0 244 672 and EP 0 255 320 both teach that myeloma cells are the advantageous host for heterologous protein production (high protein-synthetic capacity, growth to high density, ability to assemble/secret Ig). The '693 spec's own background concedes that lymphoid hosts "are at present being appraised."
3. The references are all in the same field of endeavour (mammalian expression-vector design) and are combinatorially compatible: EP 0 256 055 already names Ig as the co-amplified product; EP 0 244 672 already puts the marker and the product gene on one plasmid in a myeloma; EP 0 255 320 already uses a CMV promoter in a myeloma.
4. WO 87/04462 further reports that GS vectors work as dominant markers even in cells containing an active endogenous GS gene, removing the principal technical objection to use in a GS⁺ myeloma.

Result: claims 1 and 18 are, on their face, an aggregation of elements each disclosed for the same purpose in the same cell type, with the references supplying the reason to combine. This is the combination an Examiner or defendant would lead with, and it requires no art outside the patent's own cited set except the admitted EP-A-0 256 055.

4. Combination 2 — Claims 2, 4, 7, 8, 19, 21, 24, 25 (the "arrangement" claims)

These claims require ordering/orienting the GS gene and the heterologous gene(s) so transcription of the heterologous gene does not run through the GS gene (GS promoter upstream; or GS promoter in the opposite orientation; or the light-chain promoter opposed to the GS/heavy-chain promoters).

Proposed combination: Combination 1 + Adhya & Gottesman (1982) + WO 88/01645 (and/or Gibson's Lorist2 terminator-insulation), plus Sambrook (1989) for the mechanical construction.

WO 88/01645 states the operative rule expressly:

"Promoter occlusion occurs when a gene without a transcription termination signal lies upstream of another gene and in the same transcriptional orientation. Transcription from the upstream promoter may interfere with normal transcription of the downstream gene."

and then prescribes the remedy — an inserted transcription-termination sequence, or reversal of orientation. Adhya & Gottesman had characterized the phenomenon in 1982; Bateman & Paule (1988) described it for rRNA and noted that termination upstream of a promoter prevents occlusion in tandem arrays; Gibson's Lorist2 (1987) used terminators specifically to insulate vector genes "from interference by promoters within the insert."

Motivation: Where two transcription units are placed on one plasmid to be co-amplified, a PHOSITA designing the construct has an obvious reason to avoid the one arrangement (strong promoter upstream, same orientation, no terminator) that the art identifies as causing read-through — that is the "known technique to improve similar devices in the same way" branch of KSR. The claimed fixes (put the weakly expressed gene first; point the promoters away from each other; interpose a terminator — the last being the Express alternative at column 5 of the specification) are all in the published menu. Claim 7's and claim 8's recitations of "so that transcription of the heterologous gene does not run through the GS gene" are statements of the intended result of applying that known technique.

The key caveat, and the strongest non-obviousness rebuttal: none of this art applies promoter occlusion to a GS selection cassette, and none quantifies the consequence the specification reports — a collapse from hundreds of colonies per 10⁶ cells to zero (Table 4: pCMGS 250 → pSV2GS 18 → pcLc2GS 0) and the recovery of only 4 colonies from pAb2GS (Example 1), all of which produced "very low level" antibody. If the applicants offer this as unexpected results under In re Soni, the arrangement claims are the ones where the objective evidence is strongest. The counter-counter is that the patent's own data show both the same-direction (claims 8/25) and the opposed-direction (claims 6/7/23/24) arrangements work (Table 4; Table 8 pST-6), which under KSR tends to prove the result is insensitive to which non-interfering arrangement is chosen — i.e., a predictable alternative rather than a discovery.

5. Combination 3 — Claims 3, 5, 20, 22 (SV40-early GS promoter + hCMV-MIE product promoter)

Proposed combination: Combination 2 + the applicant's own admitted pSV2GS (GS cDNA under the SV40 early promoter, EP-A-0 256 055) + EP 0 255 320 (CMV promoter/enhancer in myeloma) + Boshart (1985) and/or Foecking (1986) (strong mammalian promoter/enhancer units, both cited on the face of the patent).

Motivation: the GS-side element is literally the applicants' admitted starting material; the product-side element is a named option in myeloma art (EP 0 255 320) and a well-characterized strong unit in the general expression-vector art (Boshart). Selecting a weak promoter for the marker and a strong one for the product is a standard, goal-driven design choice, and the specification itself supplies the rationale for it (lower MSX needed for amplification; synthesis directed to the product), which is usable as evidence that the choice was result-effective and predictable, not inventive.

Teaching-away counterpoint worth preserving: WO 87/04462 "suggests that the GS gene should include a regulatable promoter which is switched up during selection and amplification and subsequently down-regulated" (as characterized in the sibling Celltech case EP 0 491 875). That is an express direction toward a strong/up-regulated GS promoter — i.e., away from the claimed weak-GS-promoter configuration — and is the best teaching-away argument available on claims 2–5, 9 and 11.

6. Combination 4 — Claim 17 (the selection protocol)

Claim 17 is a three-step process: plate in one volume of glutamine-containing non-selective medium; after 24 h add two volumes of glutamine-free medium; recover colonies after 7 days.

Proposed combination: Combination 1 + WO 86/06409 (stepwise/graduated amplification and selection) + Kaufman (1986) (multi-step ADA amplification) + Sambrook (1989) (routine media/culture manipulation), under KSR's "obvious to try" and the long line of optimization cases (In re Aller; In re Applied Materials — optimizing a recognized result-effective variable is within the skill of the art).

Motivation: the only variable is the rate of glutamine removal. WO 87/04462 already teaches selection "for resistance to progressively increased levels" of inhibitor, so graduated, rather than abrupt, selection was an established principle. The specification's own Table 2 shows a smooth continuum — Protocol A 130, C 72, D/E 8, B 3 colonies per 10⁶ — which reads as an optimization curve, not a binary discovery.

Non-obviousness counterpoint: the specification reports that complete removal (aspiration) "severely reduces the number of surviving colonies" while progressive depletion succeeds, and that the essential medium additive is asparagine (500 µM), not the GS substrate glutamate (Table 5). That latter finding is genuinely counter-intuitive. But claim 17 does not recite asparagine, and it does not recite the GS-vector arrangement; it claims only the dilute-then-wait protocol. A defendant would argue that the claim captures the routine part and leaves the surprising part unclaimed. I assess claim 17 as the most obvious of the three independent claims, saved principally by its narrow numeric specificity (1:2 volumes; 7 days).

7. Dependent-claim vulnerabilities

• Claims 9, 10, 11, 12 (GS from SV40-early or hCMV-MIE; "derived from pSV2GS" / "derived from pCMGS"): claims 9/11 are close to admitted prior art plus a heterologous gene. The specification says pSV2GS and pCMGS (pCMGS differs only by swapping the SV40 early promoter for the hCMV-MIE Pst-Im fragment) are EP-A-0 256 055 constructs. A claim to "the GS gene and promoter derived from pSV2GS" in a myeloma host has almost no clear space.
• Claims 13, 14 (all genes from the same promoter type, hCMV-MIE): predictable uniformity choice; the materials (pEE6hCMV, pCMGS) are in the record.
• Claims 6, 7, 23, 24 (GS weak promoter, heavy chain downstream with strong promoter, light chain promoter opposed): the specificity is a construction detail; the record already builds Ig H+L vectors (Whittle et al., Protein Eng. 1:499, 1987, the B72.3 source, cited in the specification).
• Claims 15, 16: independent of obviousness, these carry the claim-type/antecedent defects flagged in the earlier claim section ("The vector of claim 14"; "The vector of claim 8"), which are § 112 issues but will also complicate any § 103 defence of claim 16's pST6/pSV2GScLccHc recitation.

8. Bottom line

Caveats I will not paper over:

1. I have not reviewed the file history of 08/376,380. The examiner had US 5,122,464 and the other seven references of record before him; the § 103 question is whether allowing was correct, and the answer may turn on argument/evidence in the prosecution record I have not seen.
2. Own-work art issue. The primary reference is the applicants' own earlier work (Bebbington & Wilson; Celltech). EP 0 256 055 published 1987-07-30 is more than one year before the earliest filing date on the record, so it is § 102(b) art regardless of inventorship overlap — but whether pre-AIA § 103(c) common ownership would disqualify any of it for obviousness purposes (and which version of § 103(c) applies to a 1995-filed application) is a legal question I flag rather than resolve.
3. Unresolved date discrepancy. The earlier section's flag stands: the family listing shows 07/898,165 with a 1988-04-18 priority while this patent's record shows 1990-01-25. I have shown the combinations work under either date, but the discrepancy should be resolved before any of this is filed.
4. New art alert. Adhya & Gottesman (1982) and WO 88/01645 are not on the face of the patent. If this analysis is used offensively, they must be presented as new art of record; if a validity opinion is being prepared, they are the references that most need a second look, because they convert the patent's central insight ("if a gene containing a strong promoter is located upstream of a GS gene having a weaker promoter … occlusion of the downstream promoter") from a discovery into an application of a 1982 printed rule.