US 5643575

I'll search for this specific patent number to verify details and check for any litigation.

US 5,643,575 — Verification and Summary

Scope of what I could verify. I searched Google Patents (which mirrors USPTO bibliographic data), the full patent text you supplied, and secondary/news sources for litigation. I did not obtain a live hit from the USPTO PatentCenter or from a CAFC 2026 docket listing — see the "litigation" caveat below.

---

Bibliographic data

Abstract (verbatim): "Branched, substantially non-antigenic polymers are disclosed. Conjugates prepared with the polymers and biologically active molecules such as proteins and peptides demonstrate extended circulating life in vivo. Substantially fewer sites on the biologically active material are used as attachment sites. Methods of forming the polymer, conjugating the polymers with biologically active moieties and methods of using the conjugates are also disclosed."

Related family / continuations (same priority date, useful context): 08/696,198 → US 5,681,567; 08/821,055 → US 5,919,455; 09/222,463 → US 6,113,906; 09/545,066 → US 6,566,506. Corresponding foreign members include EP 0 788 515 B1, EP 1 055 685 B1, CA 2,174,325 C, JP 3626494 B2, WO 95/11924 A1.

---

Plain-language overview of the independent claims

Claim 1 — the branched polymer itself (composition of matter).
Covers a branched, substantially non-antigenic polymer of the formula (R)ₙL–A, where:

• (R) = a water-soluble, substantially non-antigenic polymer bearing a C₁–C₄ terminal group (e.g., methoxy-capped PEG); n = 2 or 3 (two or three polymer "arms" off a common root);
• (A) = a functional group able to bond with a biologically active nucleophile selected from proteins, peptides, polypeptides, enzymes, and chemotherapeutic molecules — or a moiety that can be functionalized to do so;
• (L) = a specific aliphatic linking moiety, selected from the Markush structures depicted in the claim, where (α) is 1 to 5.
  The point of novelty is the non-planar, "umbrella-like" branched architecture (expressly contrasted with planar triazine double-substitution in the background). Dependent claims 2–16 narrow this (straight-chain vs. branched R; poly(alkylene oxide) R; α-substituted PAO; PEG homopolymers/alkyl-capped PEO/block copolymers; MW 200–20,000, preferably 2,000–10,000, most preferably ~5,000; n = 2; A selected from hydroxyl/amino/carboxyl/carbonyl/thiol/methylene hydrogen; A = succinimidyl or p-nitrophenyl carbonate; A plus spacer; R further functionalized for nucleophiles), and claims 36–40 add the methoxy terminal group, further A groups, and specific structural/spacer definitions.

Claim 17 — method of making a conjugate.
Contacting a biologically active nucleophile (proteins, peptides, polypeptides, enzymes, chemotherapeutic molecules) with the activated branched polymer (R)ₙL–A as defined above, so the nucleophile covalently attaches to the polymer's activated (A) group.

Claim 18 — the conjugate product.
A polymer conjugate of formula [(R)ₙL–A¹]₂–(nucleophile), i.e., the reaction product of the branched polymer and the biological material; (A¹) is the resulting covalent linkage and (z) is the number of branched polymers attached to the nucleophile. Dependent claims 19–30 narrow polymer type/MW, branched R, non-planar L, and specific nucleophile classes — notably antibodies/monoclonal antibodies/fragments/single-chain binding antigens (claim 28) and a list including anti-neoplastics, anti-infectives, analgesics, cardiovascular agents, etc. (claims 29–30, with taxol/taxanes/anthracyclines/methotrexate recited in claim 30). Claim 31 is a method-of-treatment claim depending from claim 20 (which depends on 18), covering administration of a therapeutically effective amount of the conjugate to a mammal.

Claim 32 — method of making the succinimidyl carbonate active ester.
A two-step synthesis: (i) contact the branched polymer (R)ₙL–A with p-nitrophenyl chloroformate to form the p-nitrophenyl carbonate active ester; then (ii) react that intermediate with N-hydroxysuccinimide. Claim 33 adds that (A) may include a spacer moiety.

Claim 34 — alternative synthesis route.
Making the succinimidyl carbonate active ester by contacting the branched polymer (R)ₙL–A directly with bis-succinimidyl carbonate. Claim 35 adds the optional spacer limitation.

---

Litigation — what the record shows, and what I could not confirm

• Enzon, Inc. v. Shearwater Polymers, Inc. — Google Patents' "Family has litigation" field links this family to a U.S. case in the Alabama Northern District Court, case 5:00-cv-01310 (sourced to Unified Patents litigation data). News coverage confirms Enzon was asserting the '575 patent against Shearwater, and that Shearwater challenged validity on several grounds, including Yamasaki-authored articles (Pharmaceutical Online, Sept. 5, 2000).
• A BioCentury item reports a separate Enzon infringement suit against Shearwater filed in the U.S. District Court for the District of New Jersey also alleging infringement of the '575 patent.
• The related patent in the same family, US 6,113,906, was asserted by Enzon against Hoffmann-La Roche / Roche Laboratories (PEGASYS) in New Jersey in September 2000.

Important caveats:

1. I found no CAFC 2026 docket activity for the '575 patent in my searches, and none would be expected: the patent's legal status is Expired – Lifetime, with the anticipated expiration dated 2014-07-01 (the ~20-year term from the 1993-10-27 filing date, which the record shows as a July 2014 date, consistent with a patent term adjustment/extension). Any 2026 activity would most likely concern the later family members (e.g., the Roche/PEGASYS-related '906 patent) rather than the '575 itself.
2. I could not directly load a USPTO PatentCenter or CAFC docket page; the litigation entries above come from Google Patents/Unified Patents, Darts-ip family data, and contemporaneous trade press. Treat the case number 5:00-cv-01310 as reported by a secondary aggregator rather than a first-hand court record.
3. Reported rather than independently verified: the "current assignee" listing. The record shows Enzon Pharmaceuticals Inc as current assignee plus a 2003 Pfizer/Eyetech sublicense confirmation entry; I have no document confirming an outright transfer of the '575 to Pfizer.

If you need authoritative claim-by-claim status or a certified litigation history, the definitive sources would be USPTO PatentCenter/Assignment records for application 08/143,403 and the PACER docket for N.D. Ala. 5:00-cv-01310 — neither of which I could query directly here.

I'll research litigation involving US 5,643,575. Let me run several targeted searches.

Good leads. Let me dig deeper into each case and verify case details.

US 5,643,575 — Litigation Report

Patent identification (as recorded, not corrected): US 5,643,575 A ("Non-antigenic branched polymer conjugates"), Application No. 08/143,403, filed 1993‑10‑27, granted 1997‑07‑01, inventors Anthony Martinez and Richard B. Greenwald, original assignee Enzon, Inc. (current assignee of record listed as Enzon Pharmaceuticals Inc.). Status: "Expired – Lifetime," anticipated expiration 2014‑07‑01. The '575 patent is in a family that includes WO 1995011924 A1, US 5,919,455, US 6,113,906, and US 6,566,506.

Below are the litigation matters I could verify. Note that only two US actions appear to have involved branched‑PEG patents of this family, and only one of them is expressly documented as asserting the '575 patent by number.

---

1. Enzon, Inc. v. Shearwater Polymers, Inc. (the '575 case)

Status / outcome details. In this action Enzon asserted the '575 patent against Shearwater. Shearwater defended on invalidity, relying in part on articles authored by Yamasaki and colleagues. Enzon publicly stated: "Enzon's new '906 patent issued from the same family as Enzon's earlier U.S. Patent 5,643,575, which Enzon is currently asserting against Shearwater Polymers, Inc. (Shearwater), in federal court in Alabama. In the Alabama case, Shearwater asserted that the '575 patent is invalid on a number of grounds, including certain articles authored by Yamasaki and colleagues." (Enzon SEC/press filing, 2000‑09‑05.)

The case ended by settlement, not judgment: in January 2002 Enzon announced a broad strategic alliance with Nektar (Shearwater's parent), under which "Enzon made a $40.0 million investment in our convertible preferred stock, and we agreed to the settlement of the patent infringement litigation filed by Enzon against our subsidiary, Shearwater" (Nektar Therapeutics SEC filing).

Discrepancies you should flag to the client:

• BioCentury reported that Enzon "filed a patent infringement suit in the U.S. District Court for the district of New Jersey against Shearwater, alleging infringement of ENZN's U.S. Patent No. 5,643,575" (https://www.biocentury.com/article/[112212](/patent/112212)). This conflicts with the Alabama venue on the face of the docket.
• Enzon's own 2001‑08‑13 press release refers to "the patent infringement lawsuit it filed in December 1998, against Shearwater Polymers Inc. for infringement of Enzon's branched PEG patents."
• The most likely reconciliation is that the dispute was commenced in December 1998 (possibly in New Jersey) and/or the docketed N.D. Ala. action 5:00‑cv‑01310 is the operative vehicle for that dispute by 2000. I could not confirm a transfer order or the 1998 complaint's case number/jurisdiction from the sources available; treat the venue history as unresolved.

---

2. Enzon, Inc. v. Hoffmann‑La Roche Inc. and Roche Laboratories Inc.

Important caveat on this one: the press coverage identifies the asserted patent as U.S. 6,113,906, which issued 2000‑09‑05 from the same family as the '575 patent. Enzon described the action generally as asserting that PEGASYS "infringes Enzon's branched PEG patents," and the later sublicense covered "Enzon's branched PEG patents (among others)." I did not find a source confirming that the '575 patent was specifically pleaded in the Roche complaint. Do not represent to the client that '575 was asserted against Roche without confirming the complaint.

The Roche action was resolved by the Schering‑Plough/Roche in‑licensing arrangement announced 2001‑08‑13: Schering‑Plough reported a licensing agreement with Roche settling all patent disputes over the two companies' peginterferon products; Schering‑Plough cross‑licensed and sublicensed Enzon's branched PEG patents to Roche solely for pegylated alpha interferon; consequently "Enzon has agreed to dismiss the patent infringement lawsuit it filed in September 2000 against Roche asserting that PEGASYS infringes Enzon's branched PEG patents." All US and European litigation was to be dismissed, and Enzon and Schering‑Plough withdrew their EPO oppositions to Roche's European peginterferon patent (https://investor.enzon.com/news-releases/news-release-details/enzon-announces-schering-plough-licensing-agreement-roche).

---

Matters NOT confirmed as involving US 5,643,575

• Enzon's actions against Gen‑Probe/Chugai/biOMérieux/Becton Dickinson (D. Del./S.D.N.Y., 1999), Amersham/Perkin‑Elmer/Sigma (S.D.N.Y., 2002), Affymetrix (2003), Digene (D. Del., 2002), Roche Diagnostic GmbH (S.D.N.Y., 2004), Applera/Tropix (D. Conn., 2004) — these concern Enzo Biochem/Enzo Life Sciences patents (e.g., U.S. 4,900,659; 6,221,581), a different corporate family, and are not '575 matters.
• Board of Trustees of the University of Alabama v. Nektar Therapeutics (filed 2005‑07‑12) — appears in Nektar's SEC filings in proximity to Shearwater‑related disclosures, but I found no basis to attribute it to the '575 patent. Do not list it without verifying the asserted patents.
• No CAFC appeal of the '575 patent itself was found. Given that both actions ended in settlement/license rather than judgment, nothing in the record suggests a Federal Circuit merits decision on the '575 patent.
• No current litigation. The patent expired 2014‑07‑01, and its Google Patents record shows only the N.D. Ala. case (5:00‑cv‑01310) plus a Darts‑IP "first worldwide family litigation" entry for family ID 22503911.

---

Bottom line

Only one action is documented as asserting U.S. 5,643,575 by number: Enzon, Inc. v. Shearwater Polymers, Inc., No. 5:00‑cv‑01310 (N.D. Ala., filed 2000‑05‑15), in which invalidity was raised (Yamasaki references) and which was settled in January 2002 as part of the Enzon–Nektar alliance. The related Enzon v. Hoffmann‑La‑Roche (D.N.J., filed September 2000) action asserted the same‑family '906 patent and was dismissed in August 2001 under the Schering‑Plough/Roche license. Case numbers for the Roche action and for the alleged December 1998 New Jersey filing were not located.

Research limitations: these conclusions rest on the patent's Google Patents record, RPX Empower's docket entry for 5:00‑cv‑01310, Enzon's 2000/2001 press releases and SEC filings, and Nektar's SEC filings. I was unable to complete additional docket/PACER and CAFC follow‑up searches in this session (search budget exhausted), so the venue discrepancy in the Shearwater matter and any undocketed later filings should be confirmed against PACER before being relied upon.

I'll start by running targeted searches to check whether any AIA trial proceeding exists for US 5643575 despite the empty ODP structured record.

Let me verify there's no PTAB proceeding hiding under a differently formatted citation, and check the litigation/appeal chain for any PTAB-adjacent activity.

Let me confirm the absence is real across the patent family and check whether any sibling-claim proceedings or Unified Patents activity exists.

PTAB proceedings overview

Total AIA trial proceedings on US 5,643,575: 0. The structured USPTO Open Data Portal ("PTAB proceedings on file") block is empty, and independent web searching surfaced no IPR, PGR, or CBM naming this patent — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials. All 40 claims issued on 1997-07-01 remain exactly as granted, and the bottom-line defensive posture is therefore not "the patent is hardened by surviving IPRs" — it is "the patent has never been tested at the PTAB at all." For a defendant, that cuts two ways: you inherit no estoppel baggage, but you also inherit no free kill — you would be the first petitioner, and the § 102/§ 103 art that would have to carry the case has never been PTAB-screened.

---

No proceedings to enumerate

There is no proceeding number to report. I am not going to construct one. The canonical structured record supplied in this prompt states that ODP returns no AIA trial proceedings as of the most recent ingest, and my independent searches were consistent with that:

• No IPR/PGR/CBM petition, institution decision, or Final Written Decision naming U.S. Patent No. 5,643,575 appeared in search results.
• The patent's Google Patents record carries a "Family has litigation" flag and a district-court entry (Alabama Northern District Court, case 5:00-cv-01310), but district-court litigation is not an AIA trial. That flag is commonly misread as PTAB activity; it is not. The same record shows only routine prosecution/reassignment events (Enzon, Inc. assignee; 2003-09-09 confirmation of exclusive patent sublicense to Pfizer Inc. via EyeTech Pharmaceuticals) and an anticipated expiration date of 2014-07-01 with status "Expired - Lifetime."
• The "Cited By" list on the record includes later IPRs and PGRs, but those are citations to this patent as prior art or background — not proceedings against it.

Search hits that are NOT this patent (flagged to prevent false positives)

These surfaced for "'564" or "5,643,575"-style queries and must not be attributed to US 5,643,575:

Separately, NOF Corporation v. Nektar Therapeutics, IPR2019-01395 (challenging Nektar's US 7,872,072) is a real PEG-chemistry IPR, but the patent owner is Nektar, not Enzon, and the patent is not 5,643,575. Enzon's WO 1998/041562 appears in that record as an exhibit — again, cited art, not a challenged patent.

---

Strategic summary

Claim status: all 40 claims UNTESTED — none canceled, none sustained through PTAB. Because no AIA trial has ever been instituted, there is no claim-by-claim PTAB record for claims 1–40. Claims 1 and 18 (the independent polymer and conjugate claims) and claim 32/34 (the succinimidyl-carbonate preparation methods) are unadjudicated. If someone hands you a demand letter citing 5,643,575, there is no FWD to quote back at them, and equally no canceled claim to point to as "dead." The honest posture is untested, not vulnerable and not hardened.

Estoppel landscape is clean — and that is the whole ballgame. There is no § 315(e)(2) estoppel against anyone, because estoppel only attaches to a petitioner that obtained an institution decision on the merits. Every prior-art ground — § 102 and § 103 — remains available both in an IPR and under § 282 in district court. Note the historical signal: in the Alabama action, Shearwater asserted invalidity of the '575 patent in district court on a number of grounds, including articles authored by Yamasaki and colleagues (see Enzon's SEC filings at https://enzon.gcs-web.com/node/9711/html). That is a § 282 defense, not an IPR, and it generated no estoppel and no PTAB claim-level ruling. It does, however, tell you where at least one earlier defendant thought the vulnerability lay.

Pattern signals. No repeat petitioner exists (there are no petitioners). No defensive aggregator — Unified Patents or otherwise — appears anywhere in the chain for this patent. The patent owner (Enzon, now Enzon Pharmaceuticals Inc.; sublicense confirmation to Pfizer via EyeTech in 2003) has never litigated an appeal from a PTAB decision on this patent, because there has never been one. The patent's enforcement history is entirely pre-AIA: Enzon, Inc. v. Shearwater Polymers, Inc., filed December 1998 in the District of New Jersey and later litigated in the Northern District of Alabama (Case No. 5:00-cv-01310), and Enzon v. Hoffmann-La Roche, Inc./Roche Laboratories, Inc., filed 2000-09-05 in the District of New Jersey, asserting the related '906 patent against PEGASYS.

The expiration point materially changes the calculus. The record shows an anticipated expiration of 2014-07-01 and a status of "Expired - Lifetime." An IPR can still technically be filed against an expired patent, but the practical value is normally low (no prospective injunctive exposure, no claim-amendment path, and damages confined to the pre-expiration window subject to the § 286 six-year recovery bar). If you are being asserted today, the more urgent question may be whether the assertion is actually under 5,643,575 or under a family member — US 5,919,455, US 6,113,906, or US 6,566,506 — because those carry different expiration dates and different claim sets.

---

Recommended next steps

• Treat "no PTAB activity" as the answer, and say so to your client plainly. There is no FWD to link to, no disposition to quote, and no institution decision to distinguish. I will not manufacture one. The canonical source for confirmation is the USPTO's PTAB E2E / PTAB Decisions portal (https://ptab.uspto.gov / https://developer.uspto.gov/ptab-api) and the patent's Google Patents page (https://patents.google.com/patent/US5643575/en), which shows no AIA trial entries.
• Verify which patent you are being accused of infringing before any invalidity spend. Confirm against the actual demand letter that the asserted patent is 5,643,575 and not a sibling. The family includes US 5,919,455 (continuation-in-part of the '575 application), US 6,113,906, and US 6,566,506; also US 5,681,567 (method of preparing polyalkylene oxide carboxylic acids). Family members have different priority/expiration profiles and their own (also apparently empty) PTAB histories.
• If the assertion is under the expired '575, attack the demand, not the patent. Since expiration is 2014-07-01, interrogate: what pre-expiration acts are alleged, and are they within the § 286 six-year recovery window measured from the complaint? An expired patent asserting only backward-looking royalties is a very different case from a live injunction threat.
• If you do decide to file an IPR, you are the first movers — build the record accordingly. There is no prior petition to borrow from, no PO preliminary response template on this patent to anticipate, and no Board guidance on the claim terms. Expect to have to develop the Yamasaki-reference ground flagged in the Alabama litigation yourself, and remember that for an expired patent the Board will not entertain claim amendments, so the fight is purely about whether the as-issued claims are unpatentable.
• No trial-stage milestones to track. Because no proceeding is pending, there is no institution-decision deadline, no oral hearing date, and no statutory one-year FWD deadline running. If you file, the one-year clock under 35 U.S.C. § 316(a)(11) begins only upon institution.
• Caveat on completeness. The ODP structured block is authoritative for indexed AIA trials, and web search turned up nothing further; if a very recently filed petition is not yet indexed by ODP and not yet publicly posted to PTAB E2E, I would not have seen it. A direct PTAB E2E search on the patent number is the belt-and-suspenders check before you rely on "zero proceedings" in a written opinion or a clearance memo.

I'll research the assignment chain and corporate history for US 5,643,575. Let me run several targeted searches.

Good — strong corporate-history hits. Now let me nail down the assignment/ownership records and Enzon's post-2010 corporate fate.

Assignment & Ownership Reconstruction — US 5,643,575

Source-access disclaimer (read first). I could not load the USPTO Assignment Center (assignmentcenter.uspto.gov / assignment.uspto.gov) or its API in this session, and my search budget was exhausted before I could reach it. I therefore have no reel/frame numbers and no correspondent-of-record names for this patent, and I will not invent them. Everything below is reconstructed from (a) the Google Patents legal-events record supplied in the source text (which is itself USPTO-derived), and (b) SEC filings and press releases of the assignee. The two placeholders that would normally carry reel/frame and correspondent are marked [not captured]. Verify directly at https://assignmentcenter.uspto.gov (search "5643575") before relying on this in a written opinion.

---

Inventors

Basis for the employer call: the 1993-10-27 USPTO assignment event names both men as assignors conveying to ENZON, INC. — the classic employee-inventor assignment. Both are also named inventors on a long run of contemporaneous Enzon-assigned filings (e.g., US 5,343,001 and US 5,405,877, priority 1993-01-18; US 5,605,976 / US 5,681,567, "Method of preparing polyalkylene oxide carboxylic acids," assignee "Enzon, Inc. (Piscataway, NJ)"; US 5,122,614, cited in the '575 specification itself). Both inventors' stated residences are New Jersey locations consistent with Enzon's NJ headquarters. Taxol/camptothecin conjugate claims in the sibling '567 patent are also consistent with Enzon's internal PEG-conjugate program.

Unusual-pattern check — negative finding. There is no evidence of the inventors departing Enzon within 12 months of filing, and no evidence of a pre-fire-sale inventor exodus. The opposite: Greenwald continued to be named as an Enzon inventor on filings at least a decade later — e.g., EP 1 534 334 / EP 2 338 513 A3 ("Releasable polymeric conjugates based on aliphatic biodegradable linkers," applicant Enzon, Inc., Bridgewater, NJ, filed 2003-08-13, priority US 218167 of 2002-08-13, inventors including Greenwald, Richard B.). I could not determine either inventor's exact end-of-employment date; treat that as unverified.

---

Original assignee

Enzon, Inc. (Piscataway, NJ; later Enzon Pharmaceuticals, Inc., Bridgewater, NJ after a corporate reorganization/name alignment; Nasdaq: ENZN).

• Primary line of business: a biopharmaceutical company built on PEGylation (polyethylene glycol conjugation) chemistry. The '575 comes directly out of Enzon's PEG-technology platform, which the company described in SEC filings as its "industry-leading PEGylation technology platform" and later as its "Customized PEGylation Linker Technology."
• Did it ship products embodying the claims? Enzon itself shipped PEG-conjugated drugs — Adagen (pegademase bovine) and Oncaspar (pegaspargase) — and the '575 specification's examples conjugate the claimed branched polymer to EPO and TNF. More directly on point, third-party products manufactured under Enzon branched-PEG licenses are the reason the royalty stream existed: Enzon's 2013/2016 filings list royalty revenue on seven marketed products using its PEGylation platform — PegIntron, Sylatrel/Sylatron, Macugen, CIMZIA, OMONTYS, Oncaspar and Adagen — and Enzon's own 10-K states its cross-license with Nektar produced sublicenses to "Valeant/Pfizer's Macugen, UCB's CIMZIA, Affymax and Takeda's OMONTYS, Hoffmann-La Roche's PEGASYS and an undisclosed Pfizer product." The 2003 Pfizer/Eyetech recording on the '575 is the documentation of one of those product sublicenses (Macugen / pegaptanib, a branched-PEG aptamer).
• Current status: dissolved by plan of liquidation (not bankruptcy). Trajectory, all from Enzon's own filings:
  • 2010-01-29 — consummated sale of its entire specialty pharmaceutical business (Oncaspar, Adagen, DepoCyt, Abelcet + the Indianapolis plant) to the sigma-tau Group for ~$300M + milestones; Enzon retained royalties, PEG-SN38, LNA and PEG technology.
  • 2012-12 → 2013-04-23 — Lazard-run review of a sale of the company or its assets; no deal; special dividend and cash-return strategy adopted; clinical development substantially suspended.
  • 2013 — R&D wound down permanently.
  • 2016-02-04 — Board adopted a Plan of Liquidation and Dissolution under DGCL §§ 280/281(a), subject to stockholder approval. Final dissolution date not confirmed in my sources.

Correction to a common misreading: Google Patents' "current assignee = Enzon Pharmaceuticals Inc" is accurate as the recorded owner of the '575, but "Enzon" today is a wound-up shell, not an operating company. Any "operating company" characterization below describes Enzon during 1993–2013, not now. Also note the 2003-09-09 Pfizer entry is a sublicense confirmation, not a title transfer — it does not make Pfizer the owner of the '575.

---

Assignment timeline

Google Patents' legal-events record shows exactly two post-filing ownership-related recordings for the '575, plus the filing/priority events. I found no other recorded assignment. Reel/frame and correspondent could not be retrieved (see disclaimer).

• 1993-10-27 (execution date and recording event dated the same in the record) — Reel [not captured]/[not captured]

  • Conveyance: Assignment of interest ("ASSIGNMENT OF INTERESTS (SEE DOCUMENT FOR DETAILS)")
  • Assignor: GREENWALD, RICHARD B.; MARTINEZ, ANTHONY (recorded in that order)
  • Assignee: ENZON, INC.
  • Correspondent: [not captured]
  • Context: Founder/employee assignment — the inventors conveyed the application to their employer on the filing date; this is the root of title and is why the patent issued as Enzon's property.

• 2003-09-09 — Reel [not captured]/[not captured]

  • Conveyance: Confirmation of Exclusive Patent Sublicense (a license recordation, not a transfer of title)
  • Assignor: EYETECH PHARMACEUTICALS, INC. (recorded as the conveying party)
  • Assignee: PFIZER INC.
  • Correspondent: [not captured]
  • Context: Commercial sublicense perfection — documentation of Pfizer's exclusive sublicense under Enzon's branched-PEG rights in connection with the Eyetech/Pfizer ophthalmic product (Macugen). Ownership of the '575 remained with Enzon.

• 1997-03-20 / 1998-12-29 / 2000-04-07 — these dates appear in the family record as priority claims and continuation filings (US 5,919,455; US 6,113,906; US 6,566,506; plus US 08/821,055 and 09/222,463), not as assignments. Flags: the record shows US 5,919,455 as a continuation-in-part of the '575 application and US 6,113,906/US 6,566,506 as later continuations/continuations-in-part. All were originally Enzon-assigned. For completeness, note the record also shows US 5,681,567 (from application 08/696,198, filed 1996-08-13) as a continuation-in-part of the '575 application.

• 2013-04 (executed; date not confirmed as recorded) — Reel [not captured — no recordation confirmed]

  • Conveyance: Asset purchase agreement (sale of "all right, title and interest to our Customized PEGylation platform and related assets," including "(i) intellectual property and know-how associated with the PEGylation platform, (ii) patents and know-how related to PEG SN-38, (iii) patents and know-how associated with certain of our internal clinical programs and (iv) certain related supplies and equipment") — assignee/transferee: Belrose (see Enzon Form 10-K filed 2016-03-21).
  • Context: Distressed/strategic IP divestiture, not a bankruptcy § 363 sale. I could not confirm that the '575 itself was among the assets conveyed, nor that any assignment of the '575 to Belrose was recorded. The '575 was ~15 months from its 2014-07-01 expiration at the time, so its inclusion would be plausible but is unverified. Treat this line as a lead for document review, not as a finding.

• Related (context only, not a '575 assignment): 2014-03 Enzon novated its Hisun PEG-SN38 collaboration to Belrose; 2016-02-04 Enzon adopted its Plan of Liquidation and Dissolution.

---

Timeline diagram

timeline
    title Ownership of US 5643575
    1993 : Application filed 27 Oct
         : Inventors assign rights to Enzon Inc
    1997 : Patent issues 1 Jul
    2000 : Enzon sues Shearwater in Alabama
    2002 : Shearwater suit settled in Nektar deal
    2003 : Pfizer records exclusive sublicense
    2010 : Enzon sells specialty pharma business
    2013 : PEGylation platform sold to Belrose
    2014 : Patent expires 1 Jul
    2016 : Enzon adopts liquidation plan

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present (with one unclear step).
No transfer to a licensing-only LLC is recorded for the '575. The only title event is the 1993-10-27 employee assignment to Enzon, Inc. The 2003-09-09 Pfizer record is a sublicense confirmation, not a deed. The one unresolved item is the April 2013 Belrose asset purchase of the "Customized PEGylation platform and related assets" (Enzon 10-K, 2016-03-21). I have no evidence that Belrose is a shell — no registered-agent address, no single-purpose-LLC data, and no confirmation the '575 was in the conveyed schedule — so I mark this Unclear, not "present."

2. Known asserter in the chain — Not present.
No entity in the chain matches the listed NPE rosters (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). Every recorded party is either an operating pharma (Enzon, Pfizer) or a product-development partner (Eyetech). Enzon's enforcement posture was that of a technology-owning drug company, not a pure licensing plaintiff.

3. Repeat correspondent across the chain — Unclear / not retrievable.
This is the one signal I was specifically asked to capture and I could not do it. The Assignment Center exposes the correspondent of record; I could not reach it in this session, and no correspondent name appears in the Google Patents legal-events data or in the filings I retrieved. I am not going to name a firm. Omission here is a source-access failure, not a negative finding — if you need it, the Assignment Center entry for the 1993-10-27 and 2003-09-09 recordings will show it.

4. Cascading transfers — Not present.
No chain of LLCs, no <24-month sequence, no shared-correspondent cluster. The entire chain is two recordings 10 years apart, plus one unverified 2013 asset sale.

5. Pre-litigation transfer — Not present.
First enforcement is Enzon, Inc. v. Shearwater Polymers, Inc. — the family record ties the patent to N.D. Ala. Case No. 5:00-cv-01310 (filed 2000-05-15), and contemporaneous press reports an earlier N.D.N.J. filing in December 1998. The nearest recorded assignment is 1993-10-27 — roughly seven years before the earliest possible suit date. Nothing was recorded within six months of the complaint; Enzon asserted a patent it had owned since filing.

6. Bankruptcy fire-sale — Not present (analogue noted).
No Chapter 7 or Chapter 11 for Enzon appears anywhere in the record. Enzon's IP liquidation took the form of (a) the January 2010 sale of the specialty pharmaceutical business to sigma-tau and (b) the April 2013 sale of the PEGylation platform to Belrose, followed by a February 2016 solvent Delaware plan of liquidation and dissolution. The 2013 transaction is economically analogous to a wind-down asset sale, but it was a board-approved APA, not a bankruptcy sale, and it is not a § 363-type event.

7. Privateering — Not present.
Enzon did not hand the '575 to an NPE to sue on its behalf; it sued in its own name. Two adjacent facts are worth distinguishing carefully: (i) the Nektar relationship — Enzon's cross-license/option agreement with Nektar gave Nektar the lead role in granting sublicenses under certain Enzon PEGylation patents (Nektar's right to grant new sublicenses was narrowed effective January 2007), and the 2002 Enzon–Nektar settlement/alliance had Enzon invest $40M in Nektar convertible preferred while Shearwater's patent litigation was settled. That is a commercial settlement plus a licensing channel, not privateering. (ii) The Eyetech/Pfizer sublicense is a product license for a real drug, not a transfer-to-asserter. Late-stage friction with Nektar over "immunity fees" (Enzon 10-K: "Nektar has contested our right to receive an immunity fee and we are currently in litigation in connection with this issue") is a licensing/royalty dispute between operating companies.

8. Defensive aggregator (anti-NPE) — Not present.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents or OIN. There is no evidence the '575 was ever purchased into a defensive pool.

---

Verdict

Operating-company assertion.

The only recorded transfer of title to US 5,643,575 in the entire chain is the 1993-10-27 assignment from inventors Greenwald and Martinez to Enzon, Inc., and the patent was enforced by that same operating company — Enzon, Inc. v. Shearwater Polymers, Inc. in N.D. Ala. Case No. 5:00-cv-01310 (filed 2000-05-15), with the related family asserted against Hoffmann-La Roche over PEGASYS. Enzon was a real drug manufacturer (Adagen, Oncaspar) with a royalty stream from seven PEGylated products, including its own product embodying the branched-PEG sublicense recorded on 2003-09-09 to Pfizer/Eyetech; no NPE, no shell LLC, no cascading transfers, and no defensive-aggregator terminus appears anywhere in the record.

Two caveats that qualify the verdict. First, the characterization describes the enforcement period, not today: Enzon wound down R&D in 2013, sold its PEGylation platform in April 2013, and adopted a Plan of Liquidation and Dissolution on 2016-02-04 — so the current recorded assignee is a dissolved entity that ships nothing. Second, the April 2013 Belrose asset sale may (or may not) have carried the '575, and I could not confirm a recordation. If post-2013 ownership matters to your analysis, that single question is the one to resolve at the Assignment Center.

Verification links: Assignment Center search — https://assignmentcenter.uspto.gov/ (patent no. 5643575) · legacy index — https://assignment.uspto.gov/patent/index.html · Google Patents legal events — https://patents.google.com/patent/US5643575/en · Enzon 10-K (filed 2016-03-21, Belrose APA and liquidation plan) — https://getfilings.com/sec-filings/[160321](/patent/160321)/ENZON-PHARMACEUTICALS-INC_10-K/ · Enzon 2013 royalty/dividend release — https://www.biospace.com/enzon-pharmaceuticals-inc-concludes-review-of-possible-asset-sales-sale-of-the-company-announces-1-60-special-dividend-and-periodic-royalty-divid.

I'll research the prior art cited against (and by) US 5,643,575. Let me run targeted searches.

Let me get the actual front-page "References Cited" list from US 5,643,575 itself.

Let me pin down the front-page "References Cited" list for the '575 itself.

Prior-Art Analysis — US 5,643,575 ("Non-antigenic branched polymer conjugates")

Sourcing caveat (read this first)

The authoritative Google Patents text supplied for US 5,643,575 includes the "Cited By (558)" block but omits the front-page "(56) References Cited" block for the '575 itself. That matters: the patent's own examiner-cited list is what you asked me to enumerate, and I could not retrieve it first-hand in this session. What follows is therefore a reconstruction, built from three corroborating sources, with the provenance of each entry labeled:

• (EP) — the "References cited" block of EP 0 788 515 B1 (application 94932032.9, PCT/US94/12237), which is the European member of the same application (priority US 08/143,403, 1993‑10‑27). This is the single best proxy for what the examiner considered.
• (SPEC) — references discussed in the body of the '575/EP 0 788 515 specification, including documents "cited in the application."
• (SIB) — the front-page "References Cited" list of US 5,919,455, the continuation-in-part of the '575 application (the '455 list expressly cites "5,643,575 7/1997 Martinez et al."). Same field, same examiner art, but not necessarily identical to the '575 list, and it contains references that post-date the '575 filing date and therefore cannot be '575 prior art. I flag each one.

Cross-check against my earlier sections: nothing here contradicts the previously generated summary, litigation, or PTAB sections. One gap to note explicitly: my earlier "Patent summary" listed only the in-body cited patents (4,179,337; 5,122,614; 4,946,778; 5,093,531). It did not purport to give the front-page citation list, so there is no contradiction — but the front-page list is not in the supplied record and I am reconstructing it rather than reading it.

Confidence statement: I am confident about the identity, dates and gist of the references marked (EP)/(SPEC) because those descriptions come from the '575 family's own specification text. For several (SIB)-only entries I could not verify the subject matter in this session, and I say so rather than inventing a description.

---

1. The legal frame (this is a pre-AIA patent)

US 5,643,575 has a filing/priority date of 1993‑10‑27 and was not amended into AIA practice. So the § 102 categories are:

Two consequences that drive everything below:

1. Family members are not prior art against the '575. US 5,681,567, US 5,919,455, US 6,113,906, US 6,566,506 and US 5,605,976 are later continuations/CIPs/divisions. They were filed after the '575 and share its disclosure; they cannot be § 102 art against it.
2. Anything publishing after 1993‑10‑27 is § 102(e)-only. That knocks out several references the examiner "considered" but which are chronologically incapable of anticipating.

---

2. The cited patent references, with § 102 mapping

2a. The structurally on-point references (the ones that actually matter)

2b. The enabling/background references (no anticipation, but they set up § 103)

2c. SIB-only entries whose dates make them incapable of being '575 prior art

These appear in the '455 references-cited list but issued after 1993‑10‑27, so they cannot be § 102(a)/(b) art against the '575 (at most § 102(e) if earlier-filed). I list them so they are not mistaken for '575 citations: US 5,298,643 (Greenwald, 1994‑03‑29), US 5,321,905 (Greenwald, 1994‑06‑14), US 5,324,844 (Zalipsky, 1994‑06‑28), US 5,349,001 (Greenwald et al., 1994‑09‑20), US 5,281,698 (Nitecki, 1994‑01‑25), US 5,514,379 (Weissleder, 1996), US 5,543,158 (Gref, 1996), US 5,545,698 (Barany, 1996). The last three are plainly too late to touch the '575.

Earlier (SIB)-listed patents that could be § 102(b) art against the '575 (pre-1992‑10‑27) but whose contents I did not verify: US 4,101,380 (Rubinstein, 1978, immobilized enzymes), US 4,275,000 (Ross, 1981), US 4,680,338 (Sundoro, 1987), US 4,846,548 (Klainer, 1989, = WO 88/08992 optical-fiber sensor), US 4,904,582 (Tullis, 1990), US 5,091,176 (Braatz, 1992), US 5,214,131 (Sano, 1993), US 5,217,564 (Zalipsky, 1993), US 5,229,490 (Tam, 1993).

Foreign documents (SIB): EP 0 236 987, EP 0 510 356, EP 0 539 167, WO 91/02763, WO 93/24476 (see #2). I did not verify the subject matter of the first four.

---

3. Cited non-patent literature (all from the EP search report / specification)

---

4. Art asserted in litigation but "cited" only by adverse counsel

The invalidity contentions in Enzon, Inc. v. Shearwater Polymers, Inc., No. 5:00‑cv‑01310 (N.D. Ala.) rested in part on "certain articles authored by Yamasaki and colleagues" (Enzon SEC/press filing, 2000‑09‑05, https://enzon.gcs-web.com/node/9711/html). Those are not examiner citations on the '575 face, and I do not have the specific Yamasaki citations in this session — I will not fabricate them. They are, however, the closest thing on the record to a § 102/§ 103 attack that someone actually built. Separately, the "Yamasaki" name also appears in later art listings paired with Kodera's branched-PEG work (e.g., EP 1 565 217 A2 lists "Kodera Y., Bioconjugate Chem. 5:283–288 (1994); and Yamasaki et al."). Treat the identification of the specific Yamasaki references as an open verification item.

---

5. Bottom line on § 102

1. No reference in the cited set is a clean § 102 anticipation of independent claim 1, 17, 18, 32 or 34 as issued. The reason is consistent across every reference: the '575 claims require, in combination, (a) 2 or 3 water-soluble non-antigenic polymer arms, (b) each bearing a C₁–C₄ terminal group, joined through (c) a specific aliphatic Markush (L) to (d) an activating group (A). Each cited reference supplies at most two of the four.
2. Ranked § 102 candidates, strongest first:
   • WO 93/24476 — branched/star PEG, taxol conjugate. Only if it qualifies as § 102(e) art (US-designating PCT filed before 1993‑10‑27). If it does, it is a real § 102 challenge to claims 18, 29 and 30, but still faces claim 1's (L).
   • US 5,183,660 — the closest structural reference (two PEG arms, activated, peptide-directed, expressly the anti-triazine alternative). But the aromatic benzene linker places it outside claim 1's aliphatic (L); this is a § 103 case, not § 102.
   • WO 88/08992 / US 4,846,548 — 2- and 3-arm branched PEO–antibody conjugates, but no aliphatic linking moiety and MW ≤ 3,000 (the specification itself distinguishes it). § 103.
   • Nathan et al. (1990) — PEG–lysine copolymers; potentially the most dangerous single reference for the lysine-ester species (claim 8; EP divisional claim 1), but unverified as cited art.
3. Everything else is § 103 (or mere background) material. US 4,179,337, US 5,122,614, WO 90/13540, US 4,766,106, US 5,093,531 and the six NPL papers individually teach at most an element — linear mPEG conjugation, the succinimidyl-carbonate group, branched polyol-PEGs, or phenylglyoxal chemistry. They only bite in combination.
4. Watch the dates. Several references in the family's citation lists (US 5,281,698; 5,298,643; 5,321,905; 5,324,844; 5,349,001; 5,514,379; 5,543,158; 5,545,698) issue after 1993‑10‑27 and therefore cannot be § 102(a)/(b) art against the '575. If any of them is offered as '575 prior art, the date alone defeats it (subject to a § 102(e) earlier-filing-date check).

---

6. What I would verify before relying on any of this

• Pull the certified front page of US 5,643,575 from USPTO PatentCenter (application 08/143,403) or the patentimages PDF (https://patentimages.storage.googleapis.com/…/US5643575.pdf) to obtain the actual "(56) References Cited" list. My table is a reconstruction; the certified list is the ground truth and will be shorter for some categories and may add items (e.g., the examiner may have cited art that appears only on the '575 and not on the EP counterpart or the '455).
• Verify § 102(e) status of WO 93/24476 — its international filing date and whether it designated the US. This single fact decides whether the most on-point candidate is even available as prior art.
• Verify US 5,183,660's full disclosure — specifically whether it contains any aliphatic branched core, not merely the aromatic di-substituted benzene. If it does, it moves from § 103 to a § 102 candidate for claim 1.
• Retrieve Nathan, Zalipsky & Kohn (1990) and Ouchi (1992) — confirm whether they were cited on the '575 and, more importantly, whether Nathan discloses a bis-PEG-lysine architecture with an activatable handle.
• Obtain the Yamasaki references from the N.D. Ala. invalidity contentions (PACER, 5:00‑cv‑01310) if a § 102 record on this patent is ever needed.
• Verify the two unexplained EP citations, US 4,889,916 and US 5,091,542, whose subject matter I could not confirm and have deliberately left undescribed.

Honesty note: the specific descriptions of items 1–5 in § 2a, and of every non-patent reference in § 3, are quoted or paraphrased from the '575 family's own specification and the EP search report and are reliable. The descriptions I marked "not verified" are exactly that — I did not retrieve those documents, and I have not guessed at their contents.

I'll ground the § 103 analysis in the actual prior-art references associated with this patent. Let me search for the examiner-cited art and the key references named in the patent's own background section.

Obviousness Analysis — US 5,643,575 under 35 U.S.C. § 103

0. Evidentiary basis and a provenance caveat you need to know about

The page as supplied does not contain a "References Cited" table. The full text you provided for US 5,643,575 runs from the bibliographic header, through the classification block and the "Definitions" field, to the description, 40 claims, family data, and a "Cited By (558)" list — but it contains no examiner-cited prior-art list. What the page does give on the prior-art side is:

I want to be explicit about one trap: the "Cited By" list cannot supply § 102/§ 103 art here. Every entry I can date in it post-dates 1993-10-27 (e.g., US 5,874,500 — 1995-12-18; US 5,932,462 / Shearwater branched PEG — priority 1995-01-10; WO 98/05363). The only same-day entry is US 5,919,455, which is in the same family. If someone hands you "the Cited By list" as an invalidity packet, it is worthless as art.

To do the § 103 job, I therefore reconstructed the actual examiner-cited art from the same-family sibling US 5,919,455 (a CIP of the '575 application, sharing the specification), whose "References Cited" section is available here: https://patentimages.storage.googleapis.com/a1/99/41/55e7a9c126b841/US5919455.pdf and https://uspto.report/patent/grant/[5,919,455](/patent/5919455). I am treating that list as the operative prior-art set and flag it as reconstructed, not read off the '575's own face. I also used the '575's own background admissions, which are prior art by admission regardless of the reference list.

---

1. Legal framework applied

The '575 is a pre-AIA patent (filed 1993-10-27, no earlier priority). The governing test is Graham v. John Deere: scope/content of the prior art; level of ordinary skill; differences between the claim and the art; and objective indicia. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), the combination need only be "obvious to try," and the enumerated rationales apply — including (A) combining familiar elements by known methods, (B) simple substitution of one known element for another to obtain predictable results, (C) use of a known technique to improve a similar device in the same way, (D) applying a known technique to a known device ready for improvement, and (F) design incentives/market forces. Where a combination is sought from a finite number of identified, predictable solutions, that is sufficient. A prima facie case requires articulated motivation plus a reasonable expectation of success (e.g., Wyers v. Master Lock, 616 F.3d 1231, 1238 (Fed. Cir. 2010)). Claimed numeric ranges overlapping disclosed ranges are prima facie obvious (In re Aller; In re Woodruff).

POSITA definition (proposed): a synthetic polymer/organic chemist, Ph.D. or M.S. with 3–5 years' experience, or equivalent, in PEGylation/protein–polymer conjugation as of October 1993 — i.e., someone who routinely selected activated mPEG reagents and protein attachment chemistry. This is a crowded, incremental, highly predictable art as of 1993, which materially favors obviousness.

One important structural point about the claim: claim 1's (A) is "a functional group capable of bonding with a biologically active nucleophile … or a moiety capable of being functionalized to react with said nucleophile," and dependent claim 12 expressly recites (A) = hydroxyl, amino, carboxyl, carbonyl, thiol, methylene hydrogen. So claim 1 reaches the unactivated branched polymer (e.g., the U-PEG-OH diol of Example 1), not just activated esters. That substantially widens the § 103 attack surface.

---

2. The prior-art references (reconstructed set), and what each supplies

---

3. Claim 1 — the § 103 case

Claim-1 element chart

Ground 1 (primary): Davis '337 + Zalipsky '614 + Nishimura/Kimura

The combination. Davis establishes mPEG conjugation of proteins as safe, non-antigenic and life-prolonging. Zalipsky '614 supplies the activated, amine-reactive mPEG (SC-PEG) that reacts under mild aqueous conditions to give a stable urethane bond. Nishimura/Kimura establish that two PEG strands can be hung on a single protein attachment site through a branch point, and that the resulting "activated PEG₂" modifies proteins more effectively than single-chain PEG.

Motivation — and the patent supplies it itself. The '575's background concedes the entire inventive premise: after describing the triazine double-PEG art, it states that triazine "is a toxic substance which is difficult to reduce to acceptable levels after conjugation," is "a planar group and can only be double-polymer substituted," and that "the planar structure rigidly locks the two polymer chains in place," concluding "non-triazine-based activated polymers would offer substantial benefits to the art. The present invention addresses this need." That sentence is a roadmap: keep the two-armed architecture, replace the triazine core with something non-toxic, aliphatic and non-planar. This is KSR rationale (B) — substitution of one known element (aliphatic amine linker) for another (triazine ring) with predictable results — and rationale (D), applying a known improvement technique to a known device ready for improvement. It is also a long-felt need admission that cuts against the patentee, because a disclosed need known to the art is a motivation, not a defense.

Reasonable expectation of success. Both halves were routine. Davis had done PEG–protein conjugation for 14 years; Zalipsky '614 gave a shelf-stable amine-reactive mPEG; and aliphatic di-/triamines (diethylenetriamine, 1,3-diamino-2-propanol) and amino acids (lysine) were standard branch points. The '575's specification itself admits the synthesis route is routine: branched polymers "can also be formed by reacting aliphatic linking compounds substituted with nucleophilic functional groups such as di- or tri-amino, mercapto alcohols or alkyl triols with an activated or functionalized polymer chain such as SC-PEG… Such methods are preferred because functionalized polymer chains and suitable aliphatic linking groups are either commercially available or readily synthesized."

Ground 2 (independent, and arguably stronger): Davis '337 + Zalipsky '614 + Nathan (PEG–lysine copolymers, 1990)

Nathan discloses polyethylene glycol–lysine copolymers — i.e., lysine as a branch point carrying multiple PEG arms. Combine that with Zalipsky '614's activated mPEG→urethane chemistry and Davis's teaching of the therapeutic purpose, and you arrive at (R)₂L–A where L is the lysine-derived Markush member and A = the residual lysine carboxyl/ester or an activated carbonate derived from it. This is a two-reference-plus-Davis combination where the branch point is not merely suggested but disclosed. Note that a later third-party document independently characterizes the '575's disclosure as "a compound wherein two polyethylene glycol are branched using lysine as a branching point (WO 96/21469, U.S. Pat. No. 5,643,575)" — https://patentimages.storage.googleapis.com/3d/3f/3d/ba18004ab63bbe/US20030219404A1.pdf (US 2003/0219404 A1, ¶5). That is a party admission about what the '575's core species is, useful for framing.

n = 3. Extending the two-arm lysine core (two α-amines) to a three-arm core via diethylenetriamine (the specification's own Formula IV) or a triol is the routine, predicted next step — KSR rationale (C)/(E), and squarely within "inherently obvious to try."

Ground 3: Yamasaki 1988 + Zalipsky '614 + Davis '337

Yamasaki et al. is the art the prior defendant actually relied on (see § 7). Before I can assign it a specific element-by-element role I need its full text; on its face ("Novel polyethylene glycol derivatives for modification of proteins") it is directed at exactly the (R)/linker/(A) platform. Action item: obtain the full paper and read the actual derivatives disclosed before building a ground on it.

Ground 4 (secondary art lead worth developing): the 1989–1991 Japanese branched-PEG publications

A third-party (Kyowa Hakko) application describes the pre-1993 branched-PEG landscape as including JP Published Unexamined Application Nos. 72469/91 and 95200/91 (double-chain PEG built on cyanuric chloride, mPEG ~5,000), JP 153088/89 (comb-shaped PEG–maleic anhydride copolymer), and JP 88822/91 (PEG chains branched through a benzene ring via cinnamic acid). See US 2003/0219404 A1, ¶5, linked above. Those 1989–1991 JP publications are § 102(b) art if published before 1993-10-27. They are cumulative on the branch-point teaching but useful because they show the art systematically worked through planar cores (triazine, benzene) and would have pointed a POSITA toward the non-planar alternative the '575 claims as novel. I have these only as described by a 2003 third party; the JP documents themselves must be pulled.

Similarly, a contemporaneous review states that a two-chain "activated PEG₂" was commercially available from Seikagaku Kogyo Co. Ltd. (Tokyo), and that Ono et al. had prepared homogeneous activated PEG₂ — see Prog. Polym. Sci. 23(7) (1998) at 46. A commercial two-arm PEG reagent before the critical date is powerful market-forces motivation (KSR (F)) and, depending on its structure, close § 102 art. Verify the Ono publication date and the product's structure.

---

4. Dependent claims

---

5. Claims 17 and 18 (conjugate + method)

Claim 17 is a method of contacting a biologically active nucleophile with the claim-1 polymer; claim 18 is the resulting [(R)ₙL–A¹]₂–(nucleophile) product. Once claim 1 falls, these fall with it — Davis '337 supplies the conjugation step and its conditions; Zalipsky '614 supplies pH 7–9.5 aqueous conjugation with a molar excess of activated polymer; the '575's own conjugation protocol (0.1 M phosphate pH 7.0, 1 h, RT, 2–4× excess) is textbook and differs in no patentable respect.

• Claim 28 (antibodies/mAbs/fragments/single-chain binding antigens): the '575 itself cites U.S. 4,946,778 for single-chain binding antigens, and antibody–PEG conjugates were known. Weakest link is the breadth of the recitation, but the claim is not entitled to weight where it merely lists known nucleophiles.
• Claims 29–30 (anti-neoplastics; taxol/taxanes/anthracyclines/methotrexate): the art of PEG–drug conjugates and of linking drugs through polymer carboxylates is well developed; low-molecular-weight-drug PEGylation was conventional. Flag: I did not find, within the reconstructed reference set, a specific pre-1993 taxane–PEG reference; this sub-genus needs targeted art (Enzon's own Greenwald-activated PAO patents and the taxane prodrug literature). Do not represent claims 29–30 as "surely obvious" without that search.
• Claim 31 (method of treatment, depending via 20 on 18): a method claim reciting administration of an effective amount of a known conjugate to a mammal is obvious when the conjugate is obvious; any patentability must come from the conjugate, and no unexpected therapeutic result is credited by the spec beyond the in vitro ED₅₀ data discussed in § 6.

---

6. Claims 32 and 34 — the strongest § 103 ground

Claim 32 (contact branched polymer with p-nitrophenyl chloroformate, then react with N-hydroxysuccinimide) and claim 34 (contact branched polymer with bis-succinimidyl carbonate) are almost pure process claims over known activation chemistry, applied to a known core.

• Zalipsky '614 / '460 and the '614 family (DE 690 34200 T2, https://www.freepatentsonline.com/DE69034200T2.html) disclose exactly this activation sequence on a PEG hydroxyl: conversion of PEG to the chloroformate followed by reaction with N-hydroxysuccinimide, and — critically — BSC-PEG, the bis-succinimidyl carbonate of PEG, which is the reagent recited in claim 34. The German family member recites mPEG 5,000, preferred MW 2,000–20,000, and states the utility of the succinimidyl carbonate for "aminogroups of proteins."
• The p-nitrophenyl chloroformate route to PEG–p-nitrophenyl carbonate is likewise standard (and is the '575's own Example 9, m-PNP-PEG, used as a reagent in Example 8).
• The only difference from the art is which substrate the known activation chemistry is performed on: a branched (R)ₙL–OH instead of a linear mPEG–OH. Because the substrate is obvious (Grounds 1–2), the process claim falls with it. This is the cleanest KSR (A)/(B) case in the patent — "combining familiar elements (known activation chemistry + known branched core) according to known methods."

---

7. The real-world signal: what the prior defendant argued

In the Alabama action, Shearwater "asserted that the '575 patent is invalid on a number of grounds, including certain articles authored by Yamasaki and colleagues" (Enzon filing, 2000-09-05, https://enzon.gcs-web.com/node/9711/html). That tells you where the closest art was believed to be, and it corroborates my characterisation of Ground 3. It is not a PTAB record and creates no estoppel — but as an invalidity theory it is a starting point, not a conclusion: the case settled in January 2002 without any merits ruling, so no court ever accepted or rejected it.

---

8. Counterarguments, and where the § 103 case is actually weakest

Strength of the case, honestly stated: claim 1 is not a broad genus claim. Its (L) is a closed Markush of specific aliphatic linkers with α = 1–5 only — narrower than the sibling US 5,919,455, whose independent claim adds the (m), (p) and X variables. So the attack must land on the specific linker structures, which are the 1,3-diamino-2-propanol, ethanolamine/2-(2-aminoethoxy)ethanol and lysine-ester cores of Examples 1–8 and the diethylenetriamine core of Formula IV. That is a narrower target, but the specific cores are precisely the commercially available, "readily synthesized" aliphatic amines the spec admits to using, which makes the target easier, not harder, to reach.

Likely patentee responses and how they fare:

1. "The art taught away — triazine was the way to branch." Fail. The '575's own background says the opposite: it identifies triazine's toxicity and rigidity as problems, and the Seikagaku/Ono and JP 72469/91 and 95200/91 work shows the field actively seeking alternatives. A reference that criticizes a prior approach motivates departure from it; that is the opposite of teaching away.
2. "Unexpected results" — the TNF data (branched ED₅₀ 0.29 ng/mL vs. linear SC-PEG 8–19 ng/mL). The strongest available rebuttal, but vulnerable: (a) the comparison is branched ≈10 kDa vs. linear 5 kDa — not "polymers of equivalent molecular weight," so it is not commensurate with the claim's scope; (b) it is a single in vitro cytotoxicity assay; (c) the specification itself attributes the effect to more polymer mass per attachment site, which is the very rationale the triazine art already supplied (i.e., a difference in degree, not kind, and arguably predictable). Expect to need a rebuttal expert on commensurability.
3. "The aliphatic, non-planar linker was a design choice with no predictability." Fail on this record: Nathan discloses a PEG–lysine copolymer branch point, and the art had worked through triazine and benzene cores precisely because the branch-point structure was the variable being optimized.
4. "Commercial success (PEGASYS/PEG-Intron)." Requires a nexus to these claims; the marketed conjugates use branched PEGs whose linker structures must be mapped species-by-species to the '575 Markush before any nexus argument works. Do not concede it, but do not rely on defeating it.

Caveats I must flag rather than paper over:

• The '575's own "References Cited" table was not in the supplied page; the art set above is reconstructed from US 5,919,455's table. A prosecution-history pull from PatentCenter for application 08/143,403 should be used to confirm exactly which references the examiner actually cited and considered.
• I have not read Yamasaki 1988, Nathan 1990, Zalipsky 1983, Ouchi 1992, or the EP/WO references in full. Their contents above are characterised from titles, the family's citation record, and third-party descriptions. Every ground must be re-verified against the references' own text before it is asserted in a petition or an invalidity contention.
• Publication dates for WO 93/24476 matter: a December 1993 publication date would post-date the '575's 1993-10-27 filing and put it outside § 102 entirely absent an earlier priority. Verify each foreign reference's actual publication date.
• Filing/priority dates for the US patents in § 2 are needed to establish § 102(e)/§ 102(b) status individually; issue dates alone are not sufficient.

---

9. Bottom line

The § 103 case against US 5,643,575 is strong but not self-executing, and it is strongest on the process claims and weakest on the taxane/antineoplastic sub-genus.

• Best ground: claims 32 and 34 — Zalipsky's own SC-PEG/BSC-PEG activation chemistry applied to an obvious branched core. Nearly pure KSR (A)/(B).
• Core ground for claim 1: Davis '337 + Zalipsky '614 + Nishimura (1981)/Kimura (1988), with motivation supplied by the patent's own admission of triazine's toxicity and planarity.
• Alternative/independent core ground: Davis '337 + Zalipsky '614 + Nathan (PEG–lysine copolymers, 1990), which discloses the branch point rather than merely suggesting it.
• Independent verification needed on: Yamasaki 1988 (the reference actually asserted in the Shearwater defense), Nathan 1990, the 1989–1991 JP branched-PEG publications, and the Seikagaku "activated PEG₂" reagent — and a PatentCenter pull of the '575's real References Cited list, since the page you supplied omitted it.
• Do not use the page's "Cited By (558)" list as prior art. It is forward-citation data and post-dates the critical date.