US 5879936

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

Let me run more targeted searches on the litigations and any recent activity.

US Patent 5,879,936 — Summary

Note on method/sources: I was asked to search the USPTO database and CAFC 2026 dockets. Public web searches surfaced the patent's full text (Google Patents) and a 2007 District Court filing, but I found no authoritative record of any 2026 CAFC docket activity involving this patent. The 2026 Federal Circuit materials returned by search reference unrelated patents (e.g., ironSource/Digital Turbine, AudienceScience/Google, FedEx/Qualcomm) and never cite 5,879,936. I therefore cannot confirm any pending 2026 appeal or PTAB matter for this number, and I state that as an explicit uncertainty rather than fabricating one. All bibliographic data below come from the authoritative patent text supplied.

Bibliographic data

Prosecution chain: This patent is a continuation of application Ser. No. 07/460,154 (filed 1990-01-25, abandoned) and claims priority to GB 8809129 (1988-04-18). It has a related continuation child, US 5,891,693 (application 08/376,380). A chain of reassignments runs from Celltech Therapeutics Limited → Alusuisse Holding A.G. → Alusuisse Lonza Group AG → Lonza Group AG → Lonza Swiss Licences AG / Lonza Licences AG.

Abstract

The patent concerns vectors for transforming a lymphoid cell line to glutamine independence. The vectors contain an active glutamine synthetase (GS) gene plus a heterologous gene of interest. Preferred embodiments use a strong promoter for the heterologous gene and a weak promoter for the GS gene — for example, the heterologous gene linked to the hCMV-MIE promoter and GS linked to the SV40 early region promoter.

Plain-language overview of the independent claims

Claim 1 — A method of making a glutamine-independent transformed myeloma (lymphoid) cell line, with four steps:

• (a) co-transforming a myeloma/lymphoid cell line with two vectors — one carrying a GS gene and another carrying a gene for a protein heterologous to the cell line;
• (b) growing the transformed line in medium containing glutamine;
• (c) continuing growth in a selection medium that contains glutamine plus asparagine (or an ammonia donor), with the glutamine then being progressively depleted from that medium; and
• (d) selecting for a glutamine-independent transformed cell line.

Claim 9 — The same co-transformation and glutamine-containing growth steps, but step (c) continues growth in a glutamine-free selection medium containing asparagine or an ammonia donor, followed by selection. In effect, Claim 9 is the "no-glutamine from the outset of selection" variant of Claim 1's "progressive depletion" approach.

Dependent claims 2–8 add: a selectable marker other than GS on the same vector (claim 2); co-transformation with a separate marker vector (claim 3); the marker being a gpt gene (claims 4 and 5); depletion of glutamine by dilution with asparagine-containing, glutamine-lacking medium (claim 6); the heterologous protein being an immunoglobulin heavy or light chain (claim 7); and the weak-GS-promoter / strong-heterologous-promoter arrangement directing synthesis preferentially to the heterologous protein (claim 8).

Litigation history (from search results, not the patent text)

• District of Maryland, case 8:07-cv-03177 — family litigation noted on the Google Patents record.
• Lonza Group AG v. Northwest Biotherapeutics, Inc., 1:07-cv-00467 (D. Del., filed July 27, 2007) — 5,879,936 was one of eight asserted patents (alongside 5,122,464; 5,591,639; 5,658,759; 5,770,359; 5,827,739; 5,891,693; and 5,981,216), asserted over Northwest's DCVax® products. The AO-120 filing lists the holder as "Aluguisse Holding A.G." (literal spelling in that document).

Caveats

• The Google Patents "current assignee" and "original assignee" fields carry an express disclaimer that assignee listings may be inaccurate; treat assignee/ownership statements as provisional.
• "ALUGUISSE HOLDING AG" is reproduced exactly as it appears in the record; I have not auto-corrected it despite parallel records using "Alusuisse."
• The patent expired (anticipated expiration 2016-03-09), so any 2026 appellate activity would be unusual; I found no evidence of any, and I make no representation that none exists.

I'll search for litigation involving US patent 5879936 across patent litigation sources.

Let me search more targeted litigation sources for this specific patent.

I have enough grounded information to answer. Note: my tool access indicated I've reached the search limit, so I'll report what was verified and flag what remains unconfirmed.

Litigation involving U.S. Patent No. 5,879,936

"Recombinant DNA Methods, Vectors and Host Cells" (Bebbington et al.), assigned to Lonza Group AG.

Confirmed case

Lonza Group AG v. Northwest Biotherapeutics, Inc.

Sources:

• Docket listing (Justia Dockets): https://dockets.justia.com/search?nos=830&cases=mostrecent&min-day=1&min-month=1&min-year=2004&max-day=21&max-month=11&max-year=2009&page=7&noscat=10&state=maryland
• Unified Patents / Google Patents litigation link for US 5,879,936 (D. Md., case 8:07-cv-03177): https://portal.unifiedpatents.com/litigation/Maryland%20District%20Court/case/8%3A07-cv-03177
• Complaint report (patentdocs.net, as reproduced at siliconinvestor.com, 12/11/2007): https://www.siliconinvestor.com/readmsg.aspx?msgid=24127441

Important caveats

1. Outcome not verified. I could not retrieve a PACER docket entry, CAFC opinion, or settlement/termination record establishing the disposition of 8:07-cv-03177. The US5879936A Google Patents record itself is flagged as "Family has litigation" with the Maryland case, but Google's litigation annotation typically does not report case outcomes. I therefore cannot state the judgment or current status with confidence and am not going to speculate. To confirm, pull the PACER docket for D. Md. 8:07-cv-03177 or search CAFC records (this case is old enough that it was likely terminated years ago, but I have no document confirming that).

2. "No other litigation found" caveat. My searches surfaced only this one case tied specifically to the literal number 5,879,936. I did not find any other district court case, ITC Section 337 investigation (EDIS), PTAB proceeding (IPR/PGR/CBM), or CAFC appeal naming US 5,879,936 as an asserted or challenged patent. That is a negative finding based on the searches run, not proof of absence.

3. A caution on similarly-numbered targets. Several other Lonza "GS-family" patents in the same family were asserted in the same 2007 complaint (see the patent list above), and U.S. 5,879,936 has a sibling — U.S. 5,891,693 ("Recombinant DNA Methods Vectors and Host Cells," also Bebbington) — which is a separate patent that is frequently confused with '936. Per your instruction to interpret IDs literally, I have kept '936 distinct and have not folded the sibling patent's activity into this answer. Also note that the unrelated hits with "5,879,936" (a Lithuanian NGO registration number, a Siebert Financial SEC filing accession, a PTAB exhibit number, an unrelated '993 patent) are not this patent and were excluded.

4. Other Lonza patent suits exist in the ecosystem (e.g., Vita-Herb Nutriceuticals, Inc. v. Lonza Group AG, D. Or. 6:23-cv-01829, filed Dec. 6, 2023), but that action asserts U.S. 8,895,060 and 8,846,082 — not 5,879,936 — so it is outside the scope of your request.

Recommendation

If you need a litigation-grade opinion on this patent's enforcement history, I'd suggest the following primary sources, which are more authoritative than the aggregator pages I reached:

• PACER – D. Md. docket 8:07-cv-03177 (docket sheet, all filings, and the terminating order/ judgment).
• Unified Patents Portal – the case link above (may show status/termination).
• CourtListener/RECAP – free docket if a party or user uploaded it.
• CAFC – for any appeal from the D. Md. judgment.

I can refine further if you can confirm whether you want me to also pull the sibling patent U.S. 5,891,693's litigation separately.

I'll start by confirming the canonical proceeding list and then run targeted searches to see whether any AIA trial activity exists outside the ODP feed.

Let me cross-check via a couple of independent angles before concluding.

One more check on the litigation/administrative-record backstory, which is the only place any challenge to this patent would appear.

Proceedings overview

Total AIA trial proceedings on file for US 5,879,936: ZERO. The structured "PTAB proceedings on file" block (canonical USPTO ODP ingest) returns no IPR, PGR, or CBM for this patent, and my independent searches surfaced no petition, institution decision, Final Written Decision (FWD), or Federal Circuit appeal involving it — so the breakdown is 0 active / 0 invalidated / 0 sustained / 0 settled / 0 institution-denied. The defensive posture this creates is unusual but straightforward: there is no PTAB record to fight over, and no § 315(e)(2) estoppel has attached to anyone — but that is cold comfort for the patent owner, because US 5,879,936 expired on 2016-03-09 ("Expired – Lifetime," anticipated expiration 2016-03-09, per the Google Patents bibliographic record at https://patents.google.com/patent/US5879936/en). A demand letter citing this patent in 2026 is asserting a patent whose enforceable term ended roughly a decade ago.

No proceedings to enumerate

Because the proceeding count is zero, there is no proceeding to rank "most-impactful first." I am explicitly not inventing an IPR/IPR number to fill the template — no petition number exists in the ODP structured data, and no proceeding number was returned by web search or PTAB-facing sources.

For completeness, here is the negative record on each element the template asks about:

• Type / Filed / Status / Judge panel / Petition grounds / Institution decision / FWD / Settlement / Appeal: Not applicable — no trial was ever instituted.
• Defensive value: The absence of any IPR is explained largely by timing, not by the patent's strength. AIA trials only became available 2012-09-16 (IPR effective date), and the only window in which an IPR could have been filed against a live patent ran from 2012-09-16 to the 2016-03-09 expiry — a ~3.5-year window. The patent's asserted life (the 2007 Lonza campaigns, below) was overwhelmingly in the pre-AIA era, when the available vehicles were district-court invalidity and reexamination, not IPR. So "no IPRs" here is not evidence that the claims were tested and survived; it is evidence that nobody had a commercial reason to test them before the term ran out.

Surrounding enforcement record (context — not PTAB)

The Google Patents family record flags litigation for this family, and I was able to corroborate the district-court story, which is worth knowing because it explains why no AIA petition followed:

I did not find a reported final judgment, claim construction, or invalidity holding on US 5,879,936 specifically. I am flagging that as unverified rather than guessing.

Related family members: US 5,891,693 (continuation, granted 1999-04-06, from Ser. No. 08/376,380) and EP 0 338 841 B1. The same GS-selection patent family is routinely cited in later antibody-expression patents (e.g., the § 215/§ 102 antecedent citations in EP 1 682 178 B1 and EP 1 957 538 B1).

Strategic summary

Claim status. All nine claims — independent claims 1 and 9, and dependent claims 2–8 (claims 4 and 5 reciting a gpt selectable marker, claim 6 recitation depletion-by-dilution, claim 7 reciting an immunoglobulin heavy or light chain, claim 8 reciting the weak-promoter/strong-promoter configuration) — stand UNTESTED before the PTAB and, more importantly, unamendable and unenforceable going forward. Nothing was canceled, because nothing was ever challenged in an AIA trial. The claims died of old age, not of prior art. I did not verify a post-issuance reexamination or reissue certificate for this patent, and I found none — treat any assertion that the claims were administratively narrowed as unverified.

Estoppel landscape. There is no estoppel. 35 U.S.C. § 315(e)(2) attaches only to a petitioner that obtained an institution decision, and no institution ever issued. Consequently every prior-art ground remains theoretically available — § 102, § 103, § 112, and § 101 — to any defendant. Practically, though, a defendant spending money on an IPR petition against an expired patent is lighting money on fire: the Board can institute on an expired patent (the claims can still be canceled, and a cancellation is a useful defensive "clean hands" outcome), but a petition on this docket is a pure defensive vanity purchase unless the patent owner is asserting it in a co-pending case.

Pattern signals. No defensive aggregator (Unified Patents, RPX, etc.) ever petitioned — consistent with a patent that went quiet after 2007. The patent owner never had a PTAB appeal to pursue, because it never had a PTAB loss, and the family's "first worldwide family litigation" flag traces back to the 2007 Lonza/NWBT campaign rather than any modern assertion campaign. Current assignee chain is Lonza Licences AG / Lonza Swiss Licences AG with Healthcare Royalty Partners II LP holding a security interest — i.e., a monetization vehicle, but not one that has shown any appetite for litigating this particular patent recently.

Timing point that dominates everything. The patent expired 2016-03-09. The last possible infringing act is on or before that date. Under 35 U.S.C. § 286, a suit filed now (2026-09-27) reaches back only six years, to roughly 2020-09-27 — a window entirely after expiry, meaning no compensable infringement period remains for any newly filed action. (This is my legal reasoning from the expiry date and § 286; I found no source addressing the point for this patent.) Caveat for a live pre-2016 dispute: the patent could still matter for historic conduct if a case were already pending and tolled, but there is no such publicly reported case.

Recommended next steps

1. If you are a defendant receiving a demand on US 5,879,936: the dispositive response is the expiry date, not a PTAB argument. Quote the bibliographic record — "Anticipated expiration 2016-03-09," status "Expired – Lifetime" (https://patents.google.com/patent/US5879936/en) — and note that under § 286 the recovery window for a complaint filed today does not reach a single day on which the patent was in force. There is no FWD to link because no FWD exists. Do not let opposing counsel imply the absence of IPRs means the claims were validated; as explained above, the AIA window overlapped the patent's life by only ~3.5 years.
2. If a co-pending case asserts this patent (rather than a bare demand): because no IPR has been filed by anyone, you retain the full invalidity toolkit (§ 102/§ 103 on any art, § 112, § 101), unencumbered by § 315(e)(2) estoppel — but weigh the cost of an IPR against the near-certainty that damages are zero or nominal.
3. If you are advising the patent owner: there is nothing to salvage here. The family has already been extensively cited as prior art in later Lonza/third-party expression patents, and its commercial relevance now runs through the successor patents rather than this one.
4. Verification I could not complete: I did not confirm the ultimate disposition of Lonza v. Northwest Biotherapeutics in Maryland (No. 8:07-cv-03177), and I found no ex parte reexamination or reissue certificate for US 5,879,936. If the Maryland docket matters to your analysis, pull it directly — PACER/Unified Patents at the link above. I also confirmed no CAFC appeal exists, which follows a fortiori from there having been no PTAB decision to appeal.

I'll research the assignment records for this patent, focusing on reel/frame numbers and correspondents of record.

Let me search for the specific reel/frame and correspondent details in the assignment records.

Assignment & Ownership Analysis — US 5,879,936

Sourcing limitation (read first). I could not reach a live, field-by-field USPTO Assignment Center result set for this patent number in this session; the queries returned the Google Patents legal-events mirror plus unrelated third-party documents. The Google Patents mirror exposes execution/recording dates, conveyance type, assignor and assignee, but it does not expose reel/frame numbers or the correspondent of record. Accordingly, every reel/frame field below is marked "not retrieved" and every correspondent field is marked "not retrieved." I am explicitly not inventing reel/frame or correspondent values. If you need those fields, they must be pulled directly from https://assignmentcenter.uspto.gov/ (search patent 5879936) or https://assignment.uspto.gov/patent/index.html — both surface the same underlying records.

A second, more important caution: the Google Patents "reassignment" module is an automated ingest of USPTO assignment data and is known to occasionally mis-join unrelated records into a patent's event stream. One entry in this patent's stream (the PharmAthene/MPM BioVentures security agreement, 2006-08-01) is almost certainly such a misfire — see the flagged entry below.

Inventors

Employer determination. The US application is a continuation of Ser. No. 07/460,154 (filed 1990-01-25), claiming priority to GB 8809129 (1988-04-18). The GB priority was a Celltech filing, and the 1997 recorded assignment runs from Celltech Therapeutics Limited (see timeline) — establishing the inventors were Celltech employees and that the invention was an employee-work-for-hire assignment to Celltech. Both names recur as inventors across the Celltech/Lonza glutamine-synthetase (GS) patent family (US 5,591,639; 5,658,759; 5,770,359; 5,827,739; 5,891,693; 5,981,216), which is consistent with a stable in-house research team rather than a departing-inventor pattern.

Departure pattern: No evidence of inventors leaving the assignee within 12 months of filing. The "all inventors departed → fire-sale" tell is not present on the available record. I could not independently verify each inventor's later employment history in this session, so I state that as an open item rather than a finding.

Original assignee

Named on the issued patent: ALUGUISSE HOLDING AG — spelled exactly that way in the Google Patents record and in the Espacenet applicant field. The parallel reassignment records and the AO-120 litigation filing spell the same entity "Alusuisse Holding A.G." and elsewhere "Alusuisse Holdings AG [CH]." Per operating rules I reproduce the record literally and flag that this is a spelling variant of the Swiss entity Alusuisse Holding AG, not a distinct company.

• Primary line of business: Alusuisse was a Swiss industrial group (aluminium/chemicals); its Lonza arm was the chemicals/life-science business. The GS technology itself originated at Celltech Limited, a UK biotechnology company (founded 1980, first UK biotech to list).
• Product embodiment: The claims are method/vector claims (transform a lymphoid cell line to glutamine independence). The commercial embodiment is the GS Gene Expression System — a technology/licensing product, not a pill. It was actively licensed: Cytodyn's SEC-filed Lonza agreement lists 5,879,936 among the licensed patents (https://www.cytodyn.com/investors/sec-filings/all-sec-filings/content/0001193125-15-[296302](/patent/296302)/0001193125-15-296302.pdf), and Leap Therapeutics' Lonza license Appendix 1 (Patent Rights) lists the family with "Registered Owner: Lonza Group AG" (https://leaptherapeutics.gcs-web.com/static-files/d8b4a2d8-8757-4bf8-87c2-06d0b9e207d2). So the patent sits inside a live, revenue-bearing outbound licensing program.
• Current status: Alusuisse Holding A.G. as such no longer exists independently — it merged into Alusuisse Lonza Group AG (2001), which became Lonza Group AG. Lonza is an operating company (a global CDMO) trading today. The residual "Celltech" brand was acquired by UCB S.A. in 2004 (trade sale, not bankruptcy). The original assignee is therefore acquired/renamed, not dissolved and not in bankruptcy.

Assignment timeline

Chronological, from the Google Patents legal-events mirror of the USPTO assignment record. All reel/frame and correspondent fields are "not retrieved" in this session (see limitation above).

• 1992-06-12 (filing) / recorded 1992-06-12 — Reel not retrieved

  • Conveyance: Application filed / priority to US 07/898,165
  • Assignor: (n/a — applicant of record)
  • Assignee: ALUGUISSE HOLDING AG
  • Correspondent: not retrieved
  • Context: The continuation application is filed directly in the name of the Alusuisse holding entity even though the invention originated at Celltech — i.e., the patent was already sitting inside the Alusuisse/Lonza corporate structure by filing.

• 1997-12-01 (executed) / recorded 1997-12-01 — Reel not retrieved

  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST (see document for details)
  • Assignor: CELLTECH THERAPEUTICS LIMITED (recorded parenthetically as "formerly known as Celltech Limited")
  • Assignee: ALUSUISSE HOLDING A.G.
  • Correspondent: not retrieved
  • Context: The true originating owner (Celltech) formally conveys to the Alusuisse/Lonza holding entity — a corporate-group transfer / internal reorganization, not a third-party sale.

• 2001-07-26 (executed) / recorded 2001-07-26 — Reel not retrieved

  • Conveyance: MERGER
  • Assignor: ALUSUISSE HOLDING, AG
  • Assignee: ALUSUISSE LONZA GROUP, AG
  • Correspondent: not retrieved
  • Context: Internal reorg — upstream merger of the holding company.

• 2001-07-26 (executed) / recorded 2001-07-26 — Reel not retrieved

  • Conveyance: ASSIGNMENT OF ASSIGNORS INTEREST
  • Assignor: ALUSUISSE LONZA GROUP, AG
  • Assignee: LONZA GROUP, AG
  • Correspondent: not retrieved
  • Context: Change-of-name / internal reorg completing the renaming to Lonza Group AG.

• 2006-08-01 (executed) / recorded 2006-08-01 — Reel not retrieved — ⚠️ ANOMALOUS — likely mis-ingested

  • Conveyance: SECURITY AGREEMENT
  • Assignor: PHARMATHENE, INC.
  • Assignee: MPM BIOVENTURES III-QP, L.P., AS ADMINISTRATIVE AGENT
  • Correspondent: not retrieved
  • Context: PharmAthene's disclosed IP is anthrax/BChE vaccine technology; MPM BioVentures III is a venture fund in PharmAthene's cap table (https://altimmune.gcs-web.com/static-files/c2c1dde5-d46e-42ff-baae-60878ba177d7). Nothing links PharmAthene or MPM to the GS patent family. This is very likely an automated mis-join by Google Patents' automated reassignment ingest and should not be treated as part of US 5,879,936's chain pending direct verification at Assignment Center. I flag it rather than delete it.

• 2010-10-22 (executed) / recorded 2010-10-22 — Reel not retrieved

  • Conveyance: LICENSE
  • Assignor: LONZA GROUP AG
  • Assignee: LONZA SWISS LICENCES AG
  • Correspondent: not retrieved
  • Context: Internal royalty/licensing transfer within the Lonza group (Lonza Swiss Licences AG is a defined-purpose Lonza licensing vehicle — see below).

• 2010-10-22 (executed) / recorded 2010-10-22 — Reel not retrieved

  • Conveyance: CONDITIONAL ASSIGNMENT
  • Assignor: LONZA GROUP AG
  • Assignee: LONZA SWISS LICENCES AG
  • Correspondent: not retrieved
  • Context: Internal — the "conditional" transfer that only ripens on the occurrence of defined events; mirrors the Intercompany Licence Agreement structure documented in the Swiss commercial register.

• 2013-12-27 (executed) / recorded 2013-12-27 — Reel not retrieved

  • Conveyance: LICENSE
  • Assignor: LONZA GROUP AG
  • Assignee: LONZA LICENCES AG
  • Correspondent: not retrieved
  • Context: Internal — re-papering the same licensing arrangement into a renamed Lonza licensing subsidiary.

• 2013-12-27 (executed) / recorded 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. (co-assignees of record)
  • Correspondent: not retrieved
  • Context: Royalty monetization — Healthcare Royalty Partners II is a royalty-finance fund, taking an interest alongside the Lonza entity.

• 2013-12-27 (executed) / recorded 2013-12-27 — Reel not retrieved

  • Conveyance: SECURITY AGREEMENT
  • Assignor: LONZA LICENCES AG
  • Assignee: HEALTHCARE ROYALTY PARTNERS II, L.P.
  • Correspondent: not retrieved
  • Context: Securitization — collateral assignment securing the royalty-finance arrangement.

• 2014-06-26 (executed) / recorded 2014-06-26 — Reel not retrieved

  • Conveyance: AMENDED LICENSE
  • Assignor: LONZA GROUP AG
  • Assignee: LONZA SWISS LICENCES AG
  • Correspondent: not retrieved
  • Context: Internal — amendment of the 2010 instrument.

• 2014-06-26 (executed) / recorded 2014-06-26 — Reel not retrieved

  • Conveyance: AMENDED CONDITIONAL ASSIGNMENT (two amended-copy entries)
  • Assignor: LONZA GROUP AG
  • Assignee: LONZA SWISS LICENCES AG
  • Correspondent: not retrieved
  • Context: Internal — amendment of the 2010 conditional assignment.

• 2016-03-09 — Legal status: Expired – Lifetime (anticipated expiration)

  • Context: Terminal event, not an assignment.

Timeline diagram

timeline
    title Ownership of US 5879936
    1988 : GB priority filed by Celltech
    1992 : US application filed
    1997 : Celltech assigns to Alusuisse Holding
    1999 : Patent issues
    2001 : Merger into Alusuisse Lonza Group
         : Renamed Lonza Group AG
    2007 : Lonza sues Northwest Biotherapeutics
    2010 : Licence to Lonza Swiss Licences
         : Conditional assignment recorded
    2013 : Lonza Licences AG records licence
         : Healthcare Royalty takes security
    2014 : Amended licence recorded
    2016 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — Not present.
The chain runs Celltech → Alusuisse Holding → Alusuisse Lonza Group → Lonza Group → Lonza Swiss Licences AG / Lonza Licences AG. The suffix "Licences" is an NPE-looking tell, but the evidence contradicts the shell-entity reading: Lonza Swiss Licences AG (Basel, CHE-116.039.014) is a registered Swiss company whose charter purpose is expressly to hold and exercise rights under the Intercompany Licence Agreement with its parent Lonza Group AG, i.e. a group-internal licensing/royalty vehicle, with defined pledge agreements to Drug Royalty II/III (https://www.northdata.com/Lonza%20Swiss%20Licences%20AG,%20Basel/CHE-116.039.014). That is an internal-treasury structure, not an anonymous third-party assertion shell.

2. Known asserter in the chain — Not present.
No Acacia, Marathon, Intellectual Ventures, Wi-LAN, Conversant, Vringo, Pendrell, Round Rock, MPHJ, Lumen View, or Spangenberg entity appears anywhere in the chain. The recurring principals are Celltech and Lonza — operating life-science companies.

3. Repeat correspondent across the chain — Unclear (data gap).
The correspondent-of-record fields are not retrievable in this session (see limitation), so this signal cannot be scored either way. This is the one hallmark I genuinely cannot test, and I will not guess at attorney/firm names.

4. Cascading transfers — Not present (in the NPE sense).
There is temporal clustering (two records on 2010-10-22, three on 2013-12-27, two on 2014-06-26), but the counterparties are not unrelated chained LLCs — they are all members of one corporate family plus a single royalty financier. Clustering here reflects internal re-papering plus one financing, not an LLC hop-scotch.

5. Pre-litigation transfer — Not present.
The asserted litigation (Lonza Group AG v. Northwest Biotherapeutics, 1:07-cv-00467 (D. Del.), filed 2007-07-27; and the family Maryland case 8:07-cv-03177) is not preceded within 6 months by any assignment of this patent. The nearest prior recorded event is 2001-07-26; the 2006-08-01 entry is a probable mis-join (see above) and post-dates the 2001 reorg by five years. The 2013/2014 records post-date the litigation.

6. Bankruptcy fire-sale — Not present.
No Chapter 7/11 by Celltech, Alusuisse, or Lonza on the record. Celltech's 2004 exit was a trade acquisition by UCB, not a liquidation.

7. Privateering — Not present.
The classic pattern (operating company → third-party NPE asserting on its behalf) is absent. Lonza asserted the patent directly, in its own name, against a licensee/user of the GS System — that is ordinary operating-company enforcement, not privateering. The Healthcare Royalty Partners II interest is royalty monetization / security, not an assertion vehicle.

8. Defensive aggregator — Not present.
The chain does not terminate at RPX, AST, LOT, Unified, or OIN. There is no neutralizing defensive acquisition.

Verdict

Operating-company assertion.

The chain is an unbroken corporate-family chain — Celltech → Alusuisse Holding A.G. (rec. 1997-12-01) → Alusuisse Lonza Group AG and Lonza Group AG (both rec. 2001-07-26) → Lonza Swiss Licences AG / Lonza Licences AG (rec. 2010-10-22, 2013-12-27, 2014-06-26) — with the invention's originator (Celltech) and its corporate successor (Lonza, a live CDMO) remaining the effective owner, and with Lonza asserting the patent in its own name in Lonza Group AG v. Northwest Biotherapeutics (1:07-cv-00467, D. Del., 2007). None of the hallmark NPE drivers — shell-entity transfer, known asserter, pre-litigation assignment — is present. The only third-party entry (Healthcare Royalty Partners II, L.P., security agreement rec. 2013-12-27) is a royalty-finance lender, not an assertion entity, and it post-dates the sole litigation. The single caveat is that reel/frame and correspondent fields could not be retrieved, so signal #3 is untested; that gap does not change the verdict, which is driven by the counterparty identities and dates.

Verify at: https://assignmentcenter.uspto.gov/ (patent number 5879936) — pull the reel/frame and correspondent fields to close the #3 gap and to confirm whether the 2006-08-01 PharmAthene/MPM entry is a genuine record or a mirror artifact.

Explicit uncertainties (not fabricated)

• Reel/frame numbers: not retrieved. Not invented.
• Correspondent of record: not retrieved. Not invented.
• The 2006-08-01 PharmAthene → MPM BioVentures III security agreement: flagged as probable mis-ingest; not deleted, but not relied upon.
• Inventor post-filing employment history: not independently verified.
• Dates and conveyance types above are taken from the Google Patents legal-events mirror of USPTO assignment data (a secondary source); the authoritative primary record remains USPTO Assignment Center.

I'll verify each cited reference against public records before analyzing its § 102 relevance.

Now let me verify the two European citations and the non-patent literature references.

Scope, method, and compliance notes

• I ran this analysis on the literal identifier US 5,879,936 only. I did not fold in its sibling US 5,891,693 or the EP counterpart EP 0 338 841, except where explicitly flagged.
• The prior-art set below is the citation set printed on the face of US 5,879,936 (the "Patent Citations"/"Citations" and "Non-Patent Citations" lists in the Google Patents record at https://patents.google.com/patent/US5879936/en). I retrieved and read the substantive text of US 4,656,134, US 5,122,464 and EP 0 244 677 (via the Google Patents and FreePatentsOnline/uspto.report/everypatent mirrors). I could not retrieve the substantive disclosure of EP 0 255 320, of the "Foecking et al." paper, or of the two textbook citations before my search budget was exhausted — I say so explicitly below rather than paraphrasing from memory as if verified.
• Applicable statute: priority 1988-04-18 (GB 8809129), US filing 1992-06-12 as a continuation of Ser. No. 07/460,154 (filed 1990-01-25). This is a pre-AIA case, so §§ 102(a), (b), (e) and (g) apply, not the AIA first-inventor-to-file provisions.
• Working critical dates. For § 102(a)/(g), the date is the 1988-04-18 priority date. For the § 102(b) one-year bar, benefit of the 1990-01-25 parent gives a working critical date of 1989-01-25 (a foreign § 119 priority date does not antedate a § 102(b) bar).
• One terminological caveat up front: the Google Patents lists give the references without examiner category codes (no X/Y/A or "Relevant to claim" columns). For US patents that data simply isn't on the printed face. So I can tell you what was cited, and I can map each reference to claims myself, but I cannot tell you from this record which reference the examiner actually applied as a § 102 rejection versus a § 103 rejection, and I won't pretend otherwise.

Consolidated citation table (9 listed entries; 7 distinct documents)

Important structural point: entries 2 and 3 are the same Celltech family — US 5,122,464 is the US national-phase/continuation of the PCT application published as WO 87/04462 (PCT/GB87/00039, § 371 date 1987-10-23). That family is also EP-A-0 256 055, which US 5,879,936's own specification repeatedly cites as the source of plasmids pSVLGS1 and pSV2GS. So the "five patent citations" reduce to four distinct prior-art families, and one of them is the applicant's own admitted starting point.

Reference-by-reference § 102 analysis

1. US 4,656,134 A (Ringold / Stanford) — issued 1987-04-07

Full citation: Ringold, G. M., "Gene amplification in eukaryotic cells," U.S. Patent 4,656,134, granted Apr. 7, 1987 (Board of Trustees of Leland Stanford Jr. University). https://patents.google.com/patent/US4656134 · https://www.freepatentsonline.com/4656134.html

Description (verified from text): claims a DNA unit replicable in an auxotrophic eukaryotic host containing two or more structural genes in tandem: first an "amplifiable gene" that complements the auxotroph and responds to selective pressure by amplification (the spec names DHFR, CAD, metallothionein genes, asparagine synthetase, and drug-resistance/surface-membrane genes), followed by the gene(s) of interest; selection is by inhibitor (e.g. methotrexate) or by "maintaining a low or zero concentration of an essential metabolite." The disclosed plasmid pMDSG carries the E. coli XGPRT (gpt) gene, and the disclosed genes of interest expressly include "immunoglobulins, the heavy and light chains or fragments thereof." Critically, the reference also teaches: "Desirably, the promoter for the amplifiable gene should be weaker than the promoter(s) for the succeeding gene(s)… at least about 1.5, more preferably at least two times as strong."

Anticipation assessment: It does not anticipate claims 1–9 as a whole — it discloses no glutamine synthetase gene, no lymphoid/myeloma host, and no glutamine-depleted or glutamine-free selection medium containing asparagine or an ammonia donor.

• Claim 8 ("GS gene comprises a relatively weak promoter and the heterologous gene a relatively strong promoter, whereby synthesis is directed preferentially to the heterologous protein") is the claim this reference most directly reaches: the weak-amplifiable-gene / strong-gene-of-interest promoter gradient is disclosed verbatim in substance. Claim 8 nevertheless depends from claim 1, so the GS-gene and lymphoid-host elements are still missing — this is a strong § 103 combination reference rather than a clean § 102 anticipation.
• Same reasoning for the gpt element of claims 2–5 (XGPRT/gpt disclosed) and the Ig heavy/light chain element of claim 7 (disclosed), both again missing the claim-1 base elements.

2. WO 87/04462 A1 (Celltech) — published 1987-07-30

3. US 5,122,464 A (Celltech/Glasgow, Wilson & Bebbington) — issued 1992-06-16, § 102(e) date 1987-10-23

Full citations: "Recombinant DNA sequences, vectors containing them and method for the use thereof," WO 87/04462 A1, published Jul. 30, 1987 (Celltech Limited); and "Method for dominant selection in eucaryotic cells," US 5,122,464 A, issued Jun. 16, 1992 (Celltech Limited; The University Court of the University of Glasgow). https://patents.google.com/patent/WO1987004462A1/en · https://www.everypatent.com/comp/pat5122464.html · https://uspto.report/patent/grant/[5122464](/patent/5122464)

Description (verified from the full claim set and description): the founding GS-selection disclosure. Claims recombinant DNA encoding complete glutamine synthetase, GS as a dominant selectable and co-amplifiable marker, co-amplification of a non-selected gene, and transformation of host cell lines to glutamine independence. Claim 1 covers an expression vector carrying both a GS gene and the desired protein gene; claim 9 covers the two-vector format — "(a) providing a first expression vector comprising a recombinant DNA sequence which encodes an active GS enzyme; (b) providing a second expression vector comprising… the desired protein…; (c) providing a eukaryotic host cell which is a GS prototroph; (d) transforming said host cell with both said first and said second expression vectors." Selection is by GS inhibitor (claims 2–3: phosphinothricin and methionine sulphoximine), with optional regulatable promoters (claims 5–7, heat-shock/metallothionein, "up-regulated during… selection… and down-regulated after selection"). Claim 8's desired protein is tPA; claim 18's host is CHO-K1.

Anticipation assessment: This is the closest prior art, and it is the reference the patent itself acknowledges. It discloses nearly every element of claim 1's step (a) — a GS-gene vector co-transformed with a heterologous-protein vector — and the whole purpose (conversion to glutamine independence) and downstream amplification.
It does not anticipate claims 1 or 9, because it is missing three claim elements:

1. the host is a myeloma (lymphoid) cell line — US 5,122,464 claims the opposite host type (a "GS prototroph," e.g. CHO-K1, expressly chosen so as to need an inhibitor); and
2. step (b) growth in glutamine-containing medium followed by progressive glutamine depletion, and
3. a selection medium comprising asparagine or an ammonia donor (its selection agent is MSX/phosphinothricin, not glutamine withdrawal).

So: no § 102 anticipation of claims 1–9. But because it supplies the GS gene, the two-vector co-transformation, and the glutamine-independence objective, it is the primary § 103 reference, and any claim that did not recite the lymphoid host and the depletion/medium limitations would be squarely anticipated by it. Note also that the four examples in the '936 specification that the applicant used "for comparative purposes" (pSVLGS1, pSV2GS) come from this same family.

"By another" wrinkle worth flagging: US 5,122,464 names Wilson and Bebbington; US 5,879,936 names Bebbington and Yarranton. The inventive entities differ, so US 5,122,464 is "by another" for § 102(e) purposes (In re Land; In re DeBaun), notwithstanding the shared inventor. Because this application was filed in 1992, pre-AIA § 103(c) common-ownership disqualification is not available as a shield.

4. EP 0 244 677 A2 (Daiichi) — published 1987-11-11 (US counterpart US 5,019,499)

Full citation: "Method of producing peptides, recombinant plasmid for use in the same and myeloma cells transformed with the same," EP 0 244 677 A2 (app. 87105559), priority 1986-04-14, published Nov. 11, 1987, Daiichi Pharmaceutical Co., Ltd. US counterpart: "Method of producing peptides by transforming myeloma cells with a recombinant plasmid," US 5,019,499. https://patents.google.com/patent/EP0244677A2/en · https://uspto.report/patent/grant/[5,019,499](/patent/5019499)

Description (verified): transforming myeloma cells with a recombinant plasmid in which the SV40 early promoter (and enhancer) flanks the heterologous gene on both the 5′ and 3′ sides, giving ~10-fold more product (worked example: human γ-IFN). The illustrative host lines are named expressly — P3-X63-Ag8-UI (nonsecretory, HGPRT⁻), X63-Ag8-6.5.3, and SP2/O-Ag14 — i.e. precisely the mouse plasmacytoma/hybridoma lineages used as hosts in US 5,879,936's examples (NSO, P3-X63Ag8.653, Sp2/0). The vector carries a selectable marker gene, specifically the Eco-gpt (xanthine-guanine phosphoribosyl transferase) gene under an SV40 promoter, or alternatively tk.

Anticipation assessment: No anticipation of any of claims 1–9 — there is no GS gene and no glutamine-free/glutamine-depleted selection anywhere in it. But it is the most pertinent § 102(a)/(b) art for the host-cell and marker elements:

• claim 1 / claim 9 — supplies "myeloma cell line transformed with a vector comprising a heterologous gene" (missing: GS gene; glutamine media steps);
• claims 2–5 — supplies the co-introduced non-GS gpt selectable marker in a myeloma transformation vector;
• claim 7 — its claim 4 lists the desired polypeptide as lymphokines, hormones, antigenic proteins, tissue plasminogen activator and somatomedins (Ig chains not recited there);
• and it is the natural § 103 partner with reference 2/3 (GS gene + two-vector co-transformation) to reach claims 1–7.

5. EP 0 255 320 A2 (Genzyme) — published 1988-02-03

Full citation: "Production of proteins in myeloma cells," EP 0 255 320 A2, priority 1986-07-28, published Feb. 3, 1988, Genzyme Corporation. https://patents.google.com/patent/EP0255320A2/en

Description: I could not retrieve this document's disclosure — the search budget was exhausted before I reached it, and I will not characterize its contents from memory. From the face of the '936 record, it is a § 102(a) only reference (published ~2½ months before the 1988-04-18 priority date, and inside the § 102(b) one-year window).

Anticipation assessment (provisional): On its title alone, it is directed to myeloma-cell protein production, so it plausibly bears on the myeloma host element of claims 1 and 9. Whether it anticipates anything turns entirely on whether it discloses (i) a GS gene, (ii) co-transformation with a heterologous-gene vector, and (iii) glutamine-withdrawal selection with asparagine or an ammonia donor — none of which can be inferred from the title. Treat this entry as an open item. I also note that a Genzyme myeloma-expression family appears in the '936 "Similar Documents" list as US 5,981,216 / AU 584417 B2 ("Transformed myeloma cell-line and a process for the expression of a gene coding for a eukaryotic polypeptide employing same"); whether EP 0 255 320 is the EP member of that family is a lead I could not confirm in this session.

6. Foecking et al., Gene 45:101 (1986)

Full citation (as printed): "Foecking et al Gene 45:101, 1986."

Description: on my general knowledge (medium confidence, not web-verified in this session) this is Foecking, M. K. & Hofstetter, H., "Powerful and versatile enhancer-promoter unit for mammalian expression vectors," Gene 45:101–105 (1986) — the standard description of the human cytomegalovirus major immediate-early promoter/enhancer, i.e. the hCMV-MIE element that US 5,879,936 uses as its "strong" promoter. The applicant's own reference no. 1 (Boshart et al., Cell 41:521–530, 1985) is the parallel hCMV-MIE citation.

Anticipation assessment: § 102(a)/(b) art as of 1986. It cannot anticipate any claim — it discloses no GS gene, no lymphoid host, and no selection method. It is pertinent only to the "relatively strong promoter" half of claim 8 (and to the hCMV-MIE promoter limitation that appears throughout the specification and FIGs 2–3).

7. Molecular Cloning: A Laboratory Manual, pp. 1645–1646 (1989)

Full citation (as printed, twice): "Molecular Cloning: A Laboratory Manual pp.1645 1646, 1989." and "Molecular Cloning: A Laboratory Manual pp.1645-1646, 1989."

Description / assessment: I could not verify the content of pp. 1645–1646, and I will not guess at it. What I can say, and what matters here, is the date problem: a 1989 publication (the Cold Spring Harbor Molecular Cloning manual, 2nd ed., was published in 1989, i.e. after the 1988-04-18 priority date and after the 1989-01-25 § 102(b) critical date). As printed, this entry cannot be § 102(a) or § 102(b) prior art against claims entitled to the 1988 priority date. If it was genuinely cited during prosecution, the likely explanations are (a) it was cited in a later-filed continuation (e.g. US 5,891,693) whose citation list has been merged into this one by the aggregator; (b) it was cited as background/method-enabling art under § 112 rather than as § 102 art; or (c) the priority claim to GB 8809129 was not perfected for some claim(s), making a later date controlling. This is a genuine anomaly and I flag it rather than paper over it. The fact that the same document is listed twice with different hyphenation suggests a duplicate/OCR artifact, not two documents.

8. Rodrued, Ch. 23, p. 369, in Review of Physiological Chemistry (1977)

Full citation (as printed): "Rodrued, Chapter 23 p. 369, in Review of Physiological Chemistry, 1977, Lange Medical Publication."

Description / assessment: a 1977 medical-biochemistry textbook chapter, almost certainly on amino-acid/nitrogen metabolism (glutamine, glutamate, ammonia handling) — i.e. the biochemical predicate for the patent's use of asparagine and of ammonia donors (ammonium chloride) in glutamine-free medium. I could not verify the chapter's actual contents, and I note the printed author name "Rodrued" is likely a drafting/OCR corruption of "Rodwell" (Victor W. Rodwell authored the metabolism chapters of Review of Physiological Chemistry/Harper's Review of Biochemistry) — I have not auto-corrected it, per instruction. It is clean § 102(a)/(b) art as to date, but it cannot anticipate any claim; on its face it discloses no vector, no cell line, and no selection protocol.

Claim-by-claim anticipation matrix

Data-integrity / literal-interpretation notes

Per the strict rule, I have not silently corrected any identifier. Recording the anomalies I hit:

• "ALUGUISSE HOLDING AG" (original assignee field) vs "ALUSUISSE HOLDINGS AG" (Espacenet applicant field) vs "ALUSUISSE HOLDING A.G." (reassignment records) — three literal spellings in the same family record. Not corrected.
• "Rodrued" — reproduced exactly; probable corruption of "Rodwell" (noted, not substituted).
• "Foecking et al Gene 45:101, 1986" — reproduced exactly; the "101" is a leading page, and the printed citation omits the co-author, the page range, and the article title.
• Molecular Cloning is listed twice (hyphenated and unhyphenated) with the same 1989 date and same pages — treated as one document/one artifact.
• Internal specification inconsistencies (they matter because they affect which document is which): the "Brief Description of the Drawings" assigns FIG. 3 to plasmid pST6, but Example 2 states "The Southern blot analysis is shown in FIG. 3" and the drawings list assigns the Southern blot to FIG. 4; and Example 2 refers to transfecting with "pGMGS" where every other passage says "pCMGS." The prior-generation summary in this analysis followed FIG. 4 for the Southern blot. Flagging rather than reconciling.

Explicit uncertainties (things I do not know with high confidence)

1. EP 0 255 320's actual disclosure is unverified. I could not retrieve it. My mapping of it is provisional and rests on its title and date alone.
2. Foecking et al.'s full citation is unverified this session (identified from general knowledge; medium confidence).
3. The content of Molecular Cloning pp. 1645–1646 is unverified, and the 1989 date cannot be reconciled with the 1988-04-18 priority date as § 102 art. I did not resolve which application in the family the citation actually belongs to.
4. No examiner category codes / rejection grounds were available for these citations, so which references were applied as § 102 versus § 103 during prosecution is unknown to me from this record.
5. I did not obtain the file history (USPTO PatentCenter / Global Dossier), so I cannot say whether any of these references actually featured in a § 102 rejection, or whether any was overcome by amendment or by a perfected § 119 priority claim.

Bottom line

The cited art divides cleanly into two groups that never overlap: (i) Celltech's GS dominant-selection family (WO 87/04462 / US 5,122,464 / EP-A-0 256 055), which supplies the GS gene, the two-vector co-transformation, and the glutamine-independence objective but uses a drug (MSX/phosphinothricin) and a GS-prototroph host; and (ii) myeloma-expression art (US 4,656,134 promoter-gradient; EP 0 244 677 myeloma host + gpt marker; Foecking hCMV-MIE; Rodrued biochemistry), which supplies host, marker and promoter elements but no GS selection. No cited reference anticipates claims 1–9 individually, and none anticipates them in any two-reference combination without a § 103 motivation to combine. The claim features that appear in no cited reference are (a) the myeloma/lymphoid host together with a GS gene for glutamine-independent selection, and (b) selection by progressive glutamine depletion or by a glutamine-free medium containing asparagine (or an ammonia donor) — which is precisely the ground the specification stakes out over EP-A-0 256 055 and which claims 1, 6 and 9 capture.

I'll research the prior art references cited on this patent's face to ground an obviousness analysis.

I found the key Celltech reference. Let me pin down the other cited references and the asparagine/glutamine-depletion aspect.

Let me verify the pre-1988 publication dates of the key Celltech references and check the European counterpart's opposition history, which bears directly on obviousness of these claims.

I've gathered the prior art of record and the key secondary references. I hit the search-step limit before I could verify two dates (EP 0 256 055's exact publication date and the outcome of the EP 0 338 841 opposition) — I flag those as explicit uncertainties below rather than guess.

I'll now build the §103 analysis on top of the summary/litigation sections already generated (which I don't repeat), using the Prior Art section of this record as the universe of references.

Obviousness Analysis — U.S. 5,879,936 under 35 U.S.C. § 103

0. Framework, dates, and which regime applies

• Application filing: US 07/898,165 filed 1992‑06‑12; continuation of US 07/460,154 filed 1990‑01‑25 (abandoned); earliest priority 1988‑04‑18 (GB 8809129). Pre‑AIA § 103 governs (application/priority predate 2013‑03‑16).
• Critical date: assuming the granted claims are supported by the 1988 priority document, art must predate 18 April 1988. Where a reference falls between 1988‑04‑18 and the actual U.S. filing it is not § 102(a)/(b) art but may still be evidence of the level of ordinary skill.
• Reference standard: Graham v. John Deere factors; KSR Int'l v. Teleflex rationales (predictable combination, substitution of known elements, use of known technique to improve a similar method, "obvious to try" over a finite set of predictable solutions).

1. The delta between the claims and the art (what is really at issue)

This is essential, because much of the specification's "inventive" discussion (single‑vector architecture, promoter occlusion, weak‑GS/strong‑product promoters) is not what independent claims 1 and 9 claim. Both independent claims recite co‑transformation with two vectors (a GS vector + a heterologous‑protein vector).

That two‑vector co‑transformation format is the central teaching of the Celltech GS art itself, so it is not a point of novelty. The genuine distance over the prior art is only:

1. Host: the method is performed in a myeloma (lymphoid) cell line; and
2. Selection protocol: an initial growth phase in glutamine‑containing medium, followed by a selection medium that (i) contains asparagine or an ammonia donor and either (claim 1) gradually depletes glutamine from a medium that also contains it, or (claim 9) is glutamine‑free.

Claim 8 adds the weak‑GS‑promoter / strong‑heterologous‑promoter polarity. Claims 2–7 add a second selectable marker (gpt), dilution‑based depletion, and Ig heavy/light chains.

2. The prior art of record, mapped to the claim elements

Note: US 5,891,693 (the '936 continuation sibling), EP 0 338 841 B1 (the European counterpart of this family) and US 5,827,739 appear on the "Similar Documents" list. They are family/sibling art, not independent prior art, and are excluded from the § 103 combinations below.

3. Combination 1 (the primary prima facie case) — claims 1 and 9

References: US 5,122,464 / WO 87/04462 (Celltech) in view of EP 0 244 677 (Daiichi) and/or EP 0 255 320 (Integrated Genetics).

Element mapping

• (a) co‑transformation with a GS vector + a heterologous‑protein vector → US 5,122,464, claim 20 (first vector GS, second vector desired protein, transform host with both). Expressly disclosed.
• (a) "myeloma (lymphoid) cell line" + "protein heterologous to said cell line" → EP 0 244 677 and EP 0 255 320 both describe myeloma cells engineered to express non‑myeloma (heterologous) proteins.
• (b) growth on glutamine‑containing medium → routine; the '936 specification itself states lymphoid/myeloma cells "cannot be grown in vitro on media lacking in glutamine," and standard DMEM contains glutamine.
• (c) selection medium with asparagine/an ammonia donor; glutamine‑free or progressively depleted → the only element not squarely in the GS art; supported by Rodwell (asparagine/ammonia as nitrogen donors feeding GS‑catalysed glutamine synthesis) and by routine cell‑culture media optimization. Progressive depletion is the ordinary technique of stepwise / adaptive selection.
• (d) selecting for glutamine‑independent transformants → US 5,122,464 ("transforming host cell lines to glutamine independence"; selection of transformants resistant to a GS inhibitor / able to grow without glutamine).

Motivation to combine (KSR rationales)

1. Use of a known technique to improve a similar method in the same way. GS was disclosed as a dominant marker usable across "a wide variety of cell lines" precisely to overcome DHFR's limitation (US 5,122,464, Background; US 4,656,134 supplies the DHFR baseline). A POSA seeking a selectable/amplifiable marker for a myeloma host would apply the known GS technique.
2. Substitution of a known element (host) to obtain a predictable result. EP 0 244 677 and EP 0 255 320 make myeloma cells the known production host for secreted proteins (they are "professional" secretory cells). Moving the Celltech GS system into that host is a substitution of one known host for another.
3. Art‑recognized problem + finite predictable solutions. The '936 specification and the Celltech art both recognize that lymphoid/myeloma cells are glutamine auxotrophs — i.e., the field already identified "convert lymphoid cells to glutamine independence with GS" as the goal (the '936 patent concedes the suggestion is made in EP‑A‑0 256 055).
4. "Obvious to try." Supplementing a glutamine‑free/‑reduced medium with asparagine or ammonium ion is a small, finite set of well‑understood nitrogen sources for the GS reaction; progressive depletion is standard selective‑pressure ramping (cf. MTX/MSX stepwise amplification, US 4,656,134).

Result: A strong prima facie case that claims 1 and 9 are obvious — with the caveat in §7 below.

4. Combination 2 — claim 8 (weak GS promoter / strong product promoter)

References: US 5,122,464 / WO 87/04462 + Foecking & Hofstetter 1986 + EP 0 255 320.

• Foecking supplies the fact that hCMV‑MIE ≫ SV40 in strength; EP 0 255 320 puts a CMV promoter in a myeloma expression vector; US 5,122,464 supplies GS as the marker.
• Rationale: if the goal is to direct synthesis preferentially to the product (claim 8's own stated purpose), a POSA would pair the strongest available promoter (hCMV‑MIE) with the product gene and a weaker promoter (SV40 early) with the marker — an ordinary resource‑allocation design choice.
• Counter‑evidence (cuts against obviousness): US 5,122,464 claims 11–13 teach the opposite polarity for the marker — a regulatable GS promoter that is up‑regulated during selection and down‑regulated afterwards. That is a teaching toward a strong‑during‑selection GS promoter, i.e., away from the '936's weak‑constitutive‑GS arrangement. This is a genuine, citable teaching‑away argument for claim 8.

5. Combinations for the dependent claims

• Claims 2 & 4 / 3 & 5 (a second selectable marker; gpt): Routine. The '936 specification itself states the GS vector "may" carry a gpt gene or be co‑transformed with a separate marker vector; gpt/XGPRT‑mycophenolic‑acid selection was a standard, well‑known dominant marker (the record's own operative example uses pEE6gpt). Combining a known marker with a known GS vector for a two‑step selection is a predictable combination.
• Claim 6 (deplete by dilution with asparagine‑containing, glutamine‑lacking medium): Routine medium exchange; supported by the same Rodwell‑based asparagine rationale.
• Claim 7 (heterologous protein is an Ig heavy/light chain): Strongly supported. The Celltech GS art was characterized in the field as transfecting "a gene that encodes the GS enzyme and the desired genes of the antibody's heavy and light chain," with selection in glutamine‑free medium. Myeloma hosts + Ig chains are the core of EP 0 244 677 / EP 0 255 320.

6. Secondary references that strengthen (or weaken) the case

• US 4,656,134 (Stanford) — general co‑amplification paradigm; supplies motivation to keep the marker and the product gene co‑amplifiable.
• US 5,747,308 (Celltech, "Recombinant DNA method") — surfaced in search and recites "culturing … in a medium which lacks glutamine or in which the amount of glutamine is progressively depleted, the GS gene being … so weakly transcribed that the cells in which the GS gene has been amplified are selected." ⚠️ I could not verify this patent's priority/filing date within the available steps. If it predates 1988‑04‑18 it would be highly material (it collapses claims 1, 6, 8 and 9 in one reference); if it postdates the priority (which I suspect, since it cites EP‑A‑0 256 055 as prior art "[1]"), it is not prior art and belongs only in the "state of the art" discussion. Do not rely on it without confirming the date.

7. Rebuttal / non‑obviousness considerations (why the claims might nonetheless stand)

A rigorous § 103 analysis must weigh these; several are unusually strong here because the patent's own background documents prior‑art failure:

1. Documented failure of the prior‑art suggestion (teaching away / "failure of others"). The '936 specification states that the EP‑A‑0 256 055 suggestion of transforming lymphoid cells to glutamine independence "has subsequently been found" not to work: hybridoma lines (SP2/0, YB2/0) generate spontaneous glutamine‑independent variants at frequencies of ~1/10⁵ and ~1/10², defeating GS selection; and for myelomas "transfection with a GS gene and growth … in a glutamine‑free medium does not result in significant survival rates." Under In re Dow / DePuy‑type reasoning, prior‑art recognition of failure weighs against obviousness.
2. The selection protocol is not merely "routine." The Examples (Table 2) show that an abrupt switch to glutamine‑free medium (protocols B, D, E) yielded only 3–8 colonies/10⁶ cells, whereas progressive depletion (protocol A) yielded 130, and mock‑transfected controls yielded 0. A ~20–40× dependence on the specific ramp argues that the protocol was a critical, non‑predictable parameter — a classic "unexpected results" position.
3. The asparagine finding is counter‑intuitive. The Examples report that glutamate — the actual substrate of GS — would not sustain growth even at 2 mM, whereas 500 µM asparagine did. That is a surprising result that a predictable‑combination argument does not anticipate.
4. The weak‑GS‑promoter teaching runs opposite to the art. US 5,122,464 teaches up‑regulating GS during selection (claims 11–13); the '936 claims a weak GS promoter. That is art pointing away from at least claim 8.
5. Secondary considerations (post‑filing, but probative). The GS‑NS0 system became an industry‑standard platform (see the inventors' own Bio/Technology 10:169 (1992), the reference at the top of the GS literature), with long‑felt need for high‑yield myeloma antibody production and subsequent commercial significance (the Lonza licensing/royalty arrangements reflected in the assignment chain already summarized). These can rebut a prima facie case if causally linked to the claimed subject matter.

Where these rebuttals are weakest: them = the medium‑supplementation/ramp elements themselves. A challenger will characterize Table 2's differences as routine optimization and the asparagine supplementation as an obvious‑to‑try nitrogen source; the unexpected‑results argument must therefore be tied to the claimed combination (two‑vector GS co‑transformation + asparagine/ammonia‑donor selection medium + depletion), not to isolated data points.

8. Bottom line

• Independent claims 1 and 9: A legally cognizable prima facie § 103 case exists via US 5,122,464 / WO 87/04462 (GS co‑transformation & glutamine‑independence selection) + EP 0 244 677 and/or EP 0 255 320 (myeloma host + heterologous protein) + Rodwell (asparagine/ammonia‑donor biochemistry). The two‑vector co‑transformation limitation is squarely met by the Celltech art, so the obviousness fight collapses onto the host choice and the selection‑medium/protocol elements. The host choice is weak ground for the patentee; the protocol/medium elements are where non‑obviousness is most defensible, given the documented failure of the prior suggestion and the surprising asparagine‑vs‑glutamate result.
• Claim 8: Obvious over Foecking + EP 0 255 320 on a predictable‑design theory, but US 5,122,464's regulatable‑strong‑promoter teaching (claims 11–13) provides a credible teaching‑away defense.
• Claims 2–7: Individually weak; each is a conventional, art‑recognized addition (gpt marker; media dilution; Ig chains).
• Net assessment: The claims are more vulnerable than the specification's "surprisingly found" rhetoric suggests, but the strongest invalidity theory must (i) rely on the pre‑1988 WO/EP publications rather than the 1992‑issued US 5,122,464, and (ii) overcome the documented prior‑art failure and the asparagine/glutamate unexpected‑result evidence.

9. Explicit uncertainties / caveats

• I could not verify the exact publication date of EP‑A‑0 256 055 (the Celltech reference the '936 specification itself cites) within my available steps; I assume it is the EP counterpart of the WO 87/04462 / US 5,122,464 family (priority 1986‑01‑23) and published January 1988. Confirm before relying on it as § 102 art.
• I could not verify the priority/filing date of US 5,747,308, which if pre‑1988 would be materially more damaging (see §6).
• EP 0 338 841 B1 is the European counterpart of this family; I did not confirm whether it survived opposition or in what amended form. Its prosecution/opposition record would be highly probative of how the same subject matter was assessed on the merits, and is the single most valuable follow‑up source.
• The "Molecular Cloning … 1989" citation postdates the 1988 priority and is not § 102 art against these claims as written; treat it as state‑of‑the‑art only.
• This is a technical/legal‑analytic assessment, not a validity opinion; a real § 103 conclusion depends on claim construction (especially "progressively depleting" and "myeloma (lymphoid)"), the ordinary‑skill level, and the priority‑support analysis for each claim.
• Assignee spelling is reproduced literally from the record ("ALUGUISSE HOLDING AG"), consistent with the summary section; I have not auto‑corrected it.