US 8673939

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

Let me look for CAFC docket activity and the actual claim text.

Let me search specifically for 2026 CAFC/appellate docket activity involving this patent.

US Patent 8,673,939 — Analyst Summary

Scope note / confidence flags. The authoritative patent text supplied to me is truncated before the claims section, and my searches returned litigation-record quotations rather than the verbatim granted claims. Where I reconstruct claim language below, I say so explicitly. I found no evidence of any 2026 Federal Circuit docket activity for this patent; see the litigation section for what I could and could not confirm.

1. Bibliographic data

Abstract (as published): "Methods of treating, preventing and/or managing cancer as well as and diseases and disorders associated with, or characterized by, undesired angiogenesis are disclosed. Specific methods encompass the administration of an immunomodulatory compound alone or in combination with a second active ingredient. The invention further relates to methods of reducing or avoiding adverse side effects associated with chemotherapy, radiation therapy, hormonal therapy, biological therapy or immunotherapy which comprise the administration of an immunomodulatory compound. Pharmaceutical compositions, single unit dosage forms, and kits suitable for use in methods of the invention are also disclosed."

Note the mismatch worth flagging: the title is specific (treating multiple myeloma with pomalidomide), but the abstract is generic (cancer and anti-angiogenic disorders generally). The generic abstract is a residue of the shared specification; it does not reflect the operative scope of the '939 claims.

2. Regulatory / Orange Book context

• '939 was listed in the Orange Book for POMALYST (NDA 204026), with expiry May 15, 2023 (U-1360) and Nov 15, 2023 with PED (U-2254).
• It issued from a continuation in a family that traces back to the May 2002 provisional and includes, inter alia, 7,968,569, 8,198,262 (the '262), 8,648,095, 8,673,939 (the '939), and 8,735,428 (the '428). The '939, '428, and '262 are the three "method of treatment" (MOT) patents that share a common specification.

3. Plain-language overview of the independent claims

Caveat: The claims did not appear in the text supplied to me, and I did not retrieve a verbatim claim set from a primary source in my searches. The following is reconstructed from the parties' and the court's descriptions in the New Jersey ANDA litigation and related antitrust complaints. Treat wording as approximate, not as a certified claim reproduction.

Independent claim 1 (representative MOT claim). A method of treating multiple myeloma comprising administering to a patient having multiple myeloma a therapeutically effective amount of pomalidomide (4-amino-2-(2,6-dioxopiperidine-3-yl)isoindoline-1,3-dione, depicted by structure in the claim) or a pharmaceutically acceptable salt, solvate, stereoisomer, hydrate, clathrate, or prodrug thereof. The preamble recites the patient as one who has received prior therapy (e.g., thalidomide, lenalidomide, or a proteasome inhibitor) and is relapsed and/or refractory — i.e., the claimed method is directed to pomalidomide after failure of prior anti-myeloma therapy.

Independent claim directed to a cyclical dosing regimen. A method of treating multiple myeloma comprising administering pomalidomide, or a solvate thereof, wherein the compound is administered in one or more cycles, each of which comprises administering the compound for a period of time followed by a period of rest. Celgene's own antitrust/validity pleadings describe the '428 and '939 as having "the same" independent claims, differing only in that the '428 hard-codes the schedule ("21 consecutive days followed by seven consecutive days of rest in a 28-day cycle") while the '939 states the cyclical concept generically. (One litigation exhibit labels this claim as claim 26 of the '939, but I could not independently confirm the exact claim number.)

Dosage-range claims (dependent). Dependent claims recite "about 1 mg to about 5 mg per day" of pomalidomide or a salt/solvate/stereoisomer thereof. Claims 8–11, 12, 17, and 28–29 of the '939 appear in the parties' Markman disclosures as the dose-related claims. This 1–5 mg/day range is the scope that mattered commercially — it corresponds to the POMALYST label dose (4 mg once daily, days 1–21 of a 28-day cycle).

Mechanistic summary: the '939 monopolizes, in plain terms, the use of pomalidomide to treat relapsed/refractory multiple myeloma, including on a cyclical (dosed-then-rest) schedule at roughly 1–5 mg/day. It is a method-of-use patent, not a compound or formulation patent.

4. Litigation posture (including what I could and could not verify for 2026)

Confirmed from the record I retrieved:

• Asserted by Celgene in the first wave of ANDA suits beginning May 4/11, 2017, alongside the '262, '428 (all MOT patents) and the '427 (formulation). Representative consolidated New Jersey case: Celgene Corp. v. Hetero Labs Ltd., No. 2:17-cv-03387 (D.N.J.).
• Key Markman ruling (June 16, 2020; D.I. 735): the court held the preamble "A method of treating multiple myeloma" is not limiting and refused to import an efficacy requirement into the claim. This was a decisive loss for Celgene, which had argued that patentability "hinged upon" the efficacy-limited construction (the patents had been allowed after Celgene asserted unexpected results for pomalidomide in lenalidomide-resistant myeloma via the Thakurta Declaration).
• Also construed: "about 1 mg to about 5 mg per day of a compound having the formula …" and "lubricant" (formulation patents). See the ruling at: https://cases.justia.com/federal/district-courts/new-jersey/njdce/2:2017cv03387/[348812/735](https://assignmentcenter.uspto.gov/search/patent/reelFrameDetail?reelFrame=348812-0735)/0.pdf
• Invalidity theories asserted by generics included §102/§103 over Lentzsch, Hideshima, D'Amato, the '471 and '517 patents; and §112 for lack of enablement and written description (no clinical data in the specification). See Hetero's counterclaims: https://storage.courtlistener.com/recap/gov.uscourts.njd.[348812](/patent/348812)/gov.uscourts.njd.348812.26.0.pdf
• Federal Circuit: the Google Patents family record lists one CAFC matter, No. 21-1154, and district court cases in D.N.J. (2:17-cv-03159; 2:17-cv-03387; 2:18-cv-10775; 2:19-cv-05802; 2:21-cv-02111; 2:22-cv-01993) and N.C. M.D. (1:18-cv-00540). This is a 2021 appeal, not a 2026 one.
• Follow-on antitrust/Walker Process litigation references the '939 ("'3939") heavily — e.g., The Cigna Group v. Celgene Corp., S.D.N.Y. No. 1:25-cv-05237-ER, and an earlier S.D.N.Y. matter, No. 1:23-cv-07871-ER. Plaintiffs allege the '939 was obtained through fraud on the PTO (the Thakurta Declaration) and that the '939 suits were sham litigation. Notably, the Southern District of New York dismissed the Walker Process and sham-litigation theories as to the '939 in a March 2025 opinion (sustaining other claims), per the Justia opinion at law.justia.com (case 1:23-cv-07871).

What I could NOT confirm — explicit uncertainty:

• I found no 2026 CAFC docket entry, opinion, or briefing naming patent 8,673,939. The patent expired on/around May 15, 2023 (Nov. 15, 2023 with PED), and the last ANDA settlements contemplated generic entry in Q1 2026 — so an appellate patent-validity appeal in 2026 is unlikely, though a 2026 appeal of the antitrust rulings (e.g., in the S.D.N.Y. Walker Process cases) is conceivable and I cannot rule it out. Do not treat the absence of a 2026 CAFC hit in my searches as affirmative proof that none exists.
• I did not retrieve the verbatim granted claim text from USPTO PatentCenter or Patent Public Search. The claim summaries in §3 are reconstructions from litigation documents.

5. Bottom line

US 8,673,939 is a Celgene method-of-use patent on pomalidomide for treating multiple myeloma, invented by Jerome B. Zeldis, filed March 1, 2013, issued March 18, 2014, claiming priority to May 17, 2002, and now expired. Its independent claims cover (a) administering pomalidomide to a relapsed/refractory multiple myeloma patient and (b) cyclical dosing (treatment period followed by rest), with dependent claims reciting about 1–5 mg/day. It was one of the three pomalidomide "method of treatment" patents litigated against generic ANDA filers, and it was the subject of the June 2020 Markman ruling that the "method of treating multiple myeloma" preamble is non-limiting. If you need the exact, verbatim claim set or a confirmed 2026 CAFC docket check, that requires a direct pull from USPTO PatentCenter/Patent Public Search and the Federal Circuit's PACER/CM-ECF docket — which my search tools did not surface, and I will not fabricate.

I'll research litigation involving US patent 8673939. Let me search multiple sources.

Let me search for the specific ANDA litigation cases involving this patent.

Let me search for more detail on the ANDA infringement cases and their statuses.

Litigation Involving U.S. Patent No. 8,673,939

Patent identification (literal)

• U.S. Patent No. 8,673,939 B2 — "Methods for treating multiple myeloma with 4-(amino)-2-(2,6-dioxo(3-piperidyl))-isoindoline-1,3-dione."
• Inventor: Jerome B. Zeldis. Assignee: Celgene Corporation (now a Bristol Myers Squibb subsidiary).
• Application No. 13/782,728; filed March 1, 2013; granted March 18, 2014; priority date May 17, 2002.
• The compound claimed (pomalidomide) is the active ingredient in POMALYST® (NDA 204026).
• Term: expired May 15, 2023 (Google Patents shows "anticipated expiration 2023-05-15" and status "Expired – Fee Related"; the Orange Book listing shows a pediatric extension to Nov. 15, 2023).

The '3939 patent is one of Celgene's three pomalidomide "method of treatment" patents (along with U.S. 8,198,262 and U.S. 8,735,428), and was Orange‑Book‑listed against Pomalyst and asserted in the Hatch‑Waxman litigation described below.

A. Hatch-Waxman / ANDA patent-infringement litigation (the '3939 asserted as a patent-in-suit)

1. Celgene Corp. v. Hetero Labs Ltd., et al. (first wave, 2017)

• Court: U.S. District Court for the District of New Jersey
• Case No.: 2:17-cv-03387 (ES)(MAH)
• Filed: 2017 (Docket No. 2:17-cv-03387)
• Patents asserted: U.S. 8,198,262; 8,673,939; 8,735,428; 8,828,427 (the three method-of-treatment patents plus the then-existing formulation patent). Defendants' invalidity contentions expressly addressed the '3939 patent. (See Celgene v. Hetero, No. 17-cv-3387 (D.N.J.), Defendants' Invalidity Contentions re U.S. Patent Nos. 8,198,262; 8,673,939; 8,735,428; 8,828,427.)
• Status: This became the lead consolidated pomalidomide case and was the subject of heavy claim-construction litigation. In June 2020 the court issued a claim construction rejecting Celgene's proposed "efficacy" limitation on the preamble "a method of treating multiple myeloma" — a ruling that, per the plaintiffs' later antitrust complaints, undermined the method-of-treatment patents. I do not have a confirmed final judgment in this docket; the record indicates the case was ultimately resolved in connection with the 2020–2021 settlement wave. (Treat the outcome as uncertain absent the actual docket.)

2. Celgene Corp. v. Synthon Pharmaceuticals, Inc., et al. (Synthon / Alvogen)

• Court: U.S. District Court for the District of New Jersey
• Case No.: 2:18-cv-10775 (ES)(MAH)
• Filed: June 19, 2018 (motion practice and docket entries begin June–July 2018; scheduling conference Sept. 2018)
• Defendants: Synthon Pharmaceuticals, Inc., Synthon B.V., Synthon s.r.o., and Alvogen Pine Brook, LLC (ANDA No. 210232)
• Patents asserted: Count I – '262; Count II – the '939 patent (the '3939); also '428 and '427 (First Amended Complaint filed Nov. 20, 2018, adding '467).
• Parallel case: Celgene Corp. v. Synthon Pharmaceuticals, Inc., Nos. 1:18-cv-00540 (M.D.N.C.), filed June 21, 2018 (same defendants, same ANDA).
• Outcome / status: Resolved by settlement/consent judgment. In mid-2019 the co-defendants in the lead New Jersey case (Apotex, Aurobindo, Mylan, Teva, Hetero) moved to compel production of the "Celgene–Synthon/Alvogen settlement documents" underlying the consent judgment, which confirms a settlement occurred. (See the '3939's litigation entry for M.D.N.C. 1:18-cv-00540 on DrugPatentWatch.)

3. Celgene Corp. v. [Dr. Reddy's Laboratories, Ltd.](/litigations/by-plaintiff/Dr.%20Reddy's%20Laboratories%2C%20Ltd.) and Dr. Reddy's Laboratories, Inc.

• Court: U.S. District Court for the District of New Jersey
• Case Nos.: 2:19-cv-15343 (consolidated) and 2:21-cv-02111 (ES)(MAH)
• Filed: July 12, 2019 ('3939 asserted; per the Cigna and CenterWell complaints)
• Patents asserted: '262, '3939, '428, '427, and '467 (ANDA No. 213234)
• Outcome: Consent Judgment entered (signature block dated January 28, 2022). The court enjoined DRL from infringing the patents-in-suit (which include U.S. 8,673,939) until their expiration, and dismissed all claims, counterclaims, and defenses with prejudice. (Consent Judgment, D.N.J. 2:19-cv-15343 / 2:21-cv-02111, at 1.)

4. Ancillary discovery proceeding

• In re Motion to Compel Production of Documents Pursuant to Subpoena Duces Tecum
• Court: U.S. District Court for the District of Massachusetts
• Case No.: 1:20-mc-91045
• Filed: Jan. 21, 2020 — Terminated: Feb. 6, 2020
• This is a third-party subpoena-enforcement matter docketed against the '3939's litigation entry (DrugPatentWatch), ancillary to the pomalidomide patent litigation.

B. Antitrust / purchaser-overcharge litigation in which the '3939 is pleaded as part of the alleged monopolization scheme

These are not patent-infringement suits; the '3939 patent is cited in the complaints as one of the patents allegedly procured by fraud and asserted through sham litigation to delay generic pomalidomide.

In each, the '3939 is expressly identified (e.g., Cigna Compl. ¶¶ referencing the "'3939" and the method-of-treatment patent tree; CenterWell Compl., patents listed as 6,281,230; 6,555,554; 7,968,569; 8,198,262; 8,648,095; 8,673,939; 8,735,428; 8,828,427; 9,993,467; 10,555,939).

C. Litigation links appearing in the '3939 patent family record (Google Patents "Family has litigation")

The Google Patents page for the '3939 lists the following family-level litigation links. Caveat: these are aggregated at the patent‑family level (family 32314544), which spans several Celgene pomalidomide/lenalidomide patents, so a given link does not by itself prove the '3939 was asserted in that case:

• D.N.J. 2:17-cv-03387 and 2:17-cv-03159 (2017 first-wave Pomalyst suits)
• D.N.J. 2:18-cv-10775 (Celgene v. Synthon/Alvogen)
• D.N.J. 2:19-cv-05802 — Celgene v. Mylan (venue-based dismissal, 2020 WL 12570814 (D.N.J. Sept. 25, 2020))
• M.D.N.C. 1:18-cv-00540 (Celgene v. Synthon)
• D.N.J. 2:21-cv-02111 (Dr. Reddy's)
• D.N.J. 2:22-cv-01993
• Fed. Cir. 21-1154 (appeal)

Important caveats

1. Verification limits. My searches were cut off before I could fully confirm final dispositions in every docket (notably the 2017 first-wave cases and the CAFC appeal 21-1154). Where I could not confirm an outcome, I have said so rather than guess.
2. The '3939 vs. other pomalidomide patents. Celgene's pomalidomide portfolio is large and the '3939 ('939) is frequently confused with the formulation patent U.S. 10,555,939 ("'5939"). Complaints expressly distinguish them ("We refer to this patent as the '3939 as there is another Celgene (formulation patent 10,555,939)… we refer to the formulation patent … as the '5939."). Ensure any docket you pull actually names 8,673,939.
3. Venue practice. Several first-wave cases were affected by venue rulings (e.g., Mylan's improper-venue dismissal), which is why some dockets moved between D.N.J. and other districts (e.g., M.D.N.C.).

Bottom line: Yes — litigation is known. The '3939 was asserted in the 2017–2019 Pomalyst ANDA suits (Hetero, Synthon/Alvogen, Dr. Reddy's, and related first-wave defendants), resolved through settlements/consent judgments, and is now pleaded in a series of 2023–2025 purchaser antitrust suits against Celgene and Bristol Myers Squibb in S.D.N.Y. and D.N.J.

If you want, I can pull the specific PACER/CourtListener dockets (2:17-cv-03387, 2:18-cv-10775, 2:19-cv-15343) to pin down exact filing dates and final judgment entries.

Proceedings overview

Zero AIA trial proceedings are on file for US 8,673,939 as of 2026-10-01 — no IPR, PGR, or CBM petition was ever instituted or denied against this patent (ODP structured data returns an empty set, and independent searches of Google Patents' litigation tab, DrugPatentWatch's PTAB table for patent 8,673,939, and the pomalidomide drug-level PTAB table surfaced no petition naming the '939 patent). Defensive posture: the patent was never PTAB-tested, so no claim has been canceled, narrowed, or sustained — but it is also expired (anticipated expiration 2023-05-15, "Expired – Fee Related"), and its invalidity story lives entirely in district court Hatch-Waxman and antitrust pleadings rather than in any Final Written Decision.

No proceeding to report

There is no {PROCEEDING_NUMBER} to populate. Every field the task asks for — judge panel, petition grounds, institution decision, FWD, settlement, PTAB appeal — is empty because no petition exists. I will not manufacture a docket number.

Caveat on completeness (stated rather than papered over): the prompt's canonical block says "no AIA trial activity on file," and that is consistent with the three independent sources I checked. I did not find a PTAB E2E page for this patent, but I also did not run an exhaustive E2E query (a live ptacts.uspto.gov search under the patent number is the definitive check — see recommended steps). If a petition were filed and denied institution before ODP ingest began, it would be an unpublished-in-part event that shows up only on E2E, so treat "zero" as high-confidence but verify.

Adjacent PTAB activity on the same family/product (context, not proceedings against this patent)

These are not proceedings on US 8,673,939. They are the Celgene PTAB matters that do exist, and they explain why the '939 patent was never attacked at the Board.

The Bass/CFAD campaign hit Celgene's REMS and compound patents; the Board's denial in IPR2015-01169 was Celgene's first PTAB win against CFAD and is documented at IPR2015-01169, Paper 22. Celgene appealed IPR2015-01092 and -01102 to the Federal Circuit in Celgene Corp. v. Peter, Nos. 18-1167/18-1171; the court affirmed the obviousness holdings and rejected Celgene's Fifth Amendment retroactivity/takings challenge (CAFC opinion, 2019-07-30; cert. petition denied). None of that touches the '939 patent — it belongs to the REMS/composition side of the portfolio.

Strategic summary

Claim-by-claim status: everything is UNTESTED. No claim of US 8,673,939 has been canceled, confirmed, or amended through an AIA trial. All issued claims — beginning with claim 1 (a method of treating relapsed, refractory, or relapsed-and-refractory multiple myeloma by administering ~1–5 mg/day of pomalidomide to a patient who has received prior therapy, in cycles of treatment followed by rest) and its dependents (2–24 at least, covering previous therapy with thalidomide/proteasome inhibitors/stem-cell transplant, 1/2/3/4 mg/day, oral or capsule administration, 21-days-on/7-days-off 28-day cycles, 4 mg on days 1–21, and dexamethasone combinations) — stand exactly as issued. This is the opposite of the premise in the task template: there is no canceled claim to point a defendant toward, and any assertion of claims 1–24 cannot be defeated by collateral estoppel or a § 315(e) estoppel record.

Estoppel landscape: none exists, and that cuts both ways. Because no IPR reached a Final Written Decision, no petitioner is subject to § 315(e)(2) estoppel on this patent, and no privies are either. A defendant being asserted today is therefore free to run any prior-art ground — § 102 or § 103, any reference, any combination — in district court or in a new IPR, subject only to the ordinary IPR time bar in § 315(b) (one year from service of an infringement complaint) and § 325(e) bars. Conversely, the patent owner has never had to defend its claims before the Board, so its claim-construction/validity positions are untested; the invalidity case against the '939 patent was developed in the Pomalyst ANDA litigation invalidity contentions (e.g., Celgene v. Hetero, No. 2:17-cv-03387 (D.N.J.)) and in the follow-on antitrust/inequitable-conduct complaints, not in an FWD.

Pattern signals. Three observations matter. (1) The same petitioner did not file multiple IPRs against this patent — CFAD/Bass went after Celgene's REMS patents ('501, '720) and the compound patent ('517), never the pomalidomide method-of-use patents. (2) The generic ANDA filers chose district court over the PTAB. Teva, Aurobindo, Apotex, Hetero, Mylan, Breckenridge, Par, Synthon and Dr. Reddy's all took Paragraph IV certifications on the '262/'939/'428/'427 patents (Teva's 2017-03-30 notice letter expressly certified against the '939 patent) and litigated; those cases were resolved by settlement with negotiated generic entry, so the Board was never asked to evaluate the '939 claims. (3) No defensive aggregator appears in the chain for this patent. The "Unified Patents Litigation Data" string on the Google Patents page is a data-source attribution, not evidence that Unified filed anything; the litigation entries are Hatch-Waxman suits and later antitrust suits, not aggregator challenges. The one Court of Appeals for the Federal Circuit entry in the family (appeal No. 21-1154) is a district-court appeal in the Celgene Revlimid/Pomalyst litigation family, not a PTAB appeal — I could not confirm its disposition from the sources reviewed and will not guess at it.

Recommended next steps

• If you are a defendant today, there is no FWD to cite. You cannot argue claims are canceled or that the patent owner is collaterally estopped. Your invalidity case must be built from scratch on the district-court record: verified prior art, the Pomalyst ANDA invalidity contentions (D.N.J. 2:17-cv-03387, ECF 922 Ex. D), and the prosecution history of the '939 patent (Application 13/782,728) — including the references and arguments identified in the pending antitrust/inequitable-conduct complaints (e.g., S.D.N.Y. 1:23-cv-07871). Those complaints allege fraud on the PTO and obviousness; they are allegations only, unadjudicated, and should not be represented as findings.
• IPR is technically available for this patent but practically worthless. IPR may be petitioned at any time during a patent's life, and the Board can institute on an expired patent, but the '939 patent expired 2023-05-15 (Orange Book PED to 2023-11-15) and its legal status is "Expired – Fee Related." An FWD canceling already-expired claims secures no market benefit and the patent owner cannot amend; expect a strong discretionary-denial argument and weigh the § 315(b) one-year clock from service of any complaint before spending on a petition.
• Verify the null result directly before relying on it: run the patent number through PTAB E2E / PTAB Center (https://ptacts.uspto.gov/) and confirm no petition was filed and terminally denied pre-2012-index. The three independent checks performed here — the ODP structured block, Google Patents US8673939B2 (no PTAB entry under "litigation"), and DrugPatentWatch's patent-level PTAB table — agree.
• The absence is itself the signal. The '939 patent was asserted against nine ANDA filers and listed in the Orange Book, yet attracted no IPR in thirteen years of eligibility. Combined with the fact that it expired in 2023 and that its sibling formulation patent (US 10,555,939) now carries the Pomalyst exclusivity to 2030, the '939 patent is a legacy asset: useful as historical context in a damages/market-definition argument, but not a live hazard you would expect to be enforced today.

I'll start by researching the ownership history, assignment records, and litigation for this patent.

Let me dig into the litigation records and the Celgene–BMS merger assignment history.

Let me verify the recorded assignment data and litigation case identities more precisely.

US 8673939 — Ownership & Assignment Reconstruction

Patent: US 8,673,939 B2 — "Methods for treating multiple myeloma with 4-(amino)-2-(2,6-dioxo(3-piperidyl))-isoindoline-1,3-dione" (pomalidomide; POMALYST®)
App. No.: 13/782,728 · filed 2013‑03‑01 · issued 2014‑03‑18 · priority 2002‑05‑17 (prov. 60/380,842)
Verified against: Google Patents legal events, Celgene litigation exhibits on CourtListener, Orange Book / FDA records, BMS–Celgene merger SEC filings.

Sourcing caveat (important): I was unable to load the USPTO Assignment Center query page directly in this session. The assignment reconstruction below is built from (a) the recorded-assignment event surfaced in Google Patents' legal-events panel (which mirrors ODP assignment data) and (b) an assignment cover sheet reproduced verbatim as a trial exhibit in Celgene v. Hetero, D.N.J. 2:17‑cv‑03387 (ECF 250‑30, p. 53). Where a reel/frame or correspondent is not visible in those sources, I say so rather than fill it in.

Inventors

Unusual-pattern check: No inventor-exodus pattern. There is a single inventor, a long-tenured Celgene officer (not a rank-and-file engineer), and the patent was prosecuted and asserted by Celgene itself. The classic "all inventors departed within 12 months → portfolio fire-sale" tell is absent. Note the related family-level quirk (not an assignment issue): the same family's prosecution history is the subject of fraud/unenforceability allegations in the pomalidomide antitrust MDL, and the prosecuting attorney of record for the sibling ’262 patent was Anthony Insogna (Pennie & Edmonds → Jones Day). That is prosecution counsel, not the assignment correspondent — see signal 3.

Original assignee

Celgene Corporation, Summit, NJ (assignment documents of this era give 7 Powder Horn Drive, Warren, NJ 07059).

• Product embodying the claims: Yes — POMALYST® (pomalidomide), FDA NDA 204,026 approved 2013‑02‑08, listed in the Orange Book with the ’939 patent under use code U‑1360. Celgene's own press release library is cited as prior art on the face of the patent.
• Primary line of business: Branded biopharmaceuticals (hematology/oncology; REVLIMID®, POMALYST®/IMNOVID®, THALOMID®, OTEZLA®, ABRAXANE®).
• Current status: Operating; acquired. Bristol‑Myers Squibb completed its $74B acquisition of Celgene on 2019‑11‑20, after which "Celgene became a wholly owned subsidiary of Bristol‑Myers Squibb Company" (BMS press release, 2019‑11‑20; BMS 10‑K for FY2019, purchase accounting dated at the acquisition date). Because Celgene Corporation survived as a subsidiary (stock merger, not an asset sale), no assignment to BMS was required and none appears in the record — the assignee of record remains Celgene Corporation. This is why the chain has exactly one link.

Assignment timeline

The Assignment Center / ODP record for this patent contains only the original inventor→employer assignment. There are no post-issuance assignments, no security interests, no name changes, and no releases recorded against US 8,673,939.

1. 2013‑03‑01 (executed) / recorded 2013‑03‑01 — Reel/frame not surfaced in sources reviewed

• Conveyance: Assignment of assignors' interest ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
• Assignor: ZELDIS, Jerome B.
• Assignee: CELGENE CORPORATION
• Correspondent: Not visible in the sources reviewed. (Flagged only as an open item — no recurrence can be established, so I do not treat this as a correspondent signal.)
• Context: Inventor-to-employer assignment, executed contemporaneously with the filing of continuation 13/782,728. Routine, not a transfer to a third party.

Family-level record (probable, not confirmed for ’939 specifically)

• Reel 013982 / Frame 0697 — assignment to CELGENE CORPORATION, 7 Powder Horn Drive, Warren, New Jersey 07059. This cover sheet is reproduced as an exhibit in Celgene v. Hetero, D.N.J. 2:17‑cv‑03387 (ECF 250‑30 at 53). Reel 013982 corresponds to recordings from the 2002–2003 window, consistent with the ’939 family's 2002 priority/2003 parent filing (Ser. No. 10/438,213). Caveat: the exhibit set covers the sibling patents ’262, ’428 and ’427 as well, and I could not confirm that this specific cover sheet is the ’939 record rather than a sibling's. Treat as family-level, unverified.

Litigation overlay (post-issuance, but not assignments): per the "Family has litigation" data and the pomalidomide antitrust pleadings, Celgene asserted the ’3939 in the 2017 ANDA wave — Celgene v. Hetero, D.N.J. 2:17‑cv‑03387, where defendants' invalidity contentions expressly address "U.S. Patent Nos. 8,198,262; 8,673,939; 8,735,428 and 8,828,427" (quoted in the Cigna complaint, ¶306 n.146). Other dockets in the same campaign: 2:17‑cv‑03159, 2:18‑cv‑10775, 1:18‑cv‑00540 (M.D.N.C.), 2:19‑cv‑05802 (Celgene v. Mylan), 2:21‑cv‑02111, 2:22‑cv‑01993, and Fed. Cir. 21‑1154. These are assertions by the owner, not transfers.

Timeline diagram

timeline
    title Ownership of US 8673939
    2002 : Priority application filed by Celgene
    2013 : Continuation filed March 1
         : Inventor Zeldis assigns rights to Celgene
    2014 : Patent issued March 18
    2017 : Celgene asserts patent against generic filers
    2019 : Celgene becomes a BMS subsidiary
    2023 : Patent expires May 15

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The only recorded conveyance is Zeldis → Celgene Corporation, an operating branded pharma. No "IP / Holdings / Ventures / Licensing" LLC ever appears as assignee, and no registered-agent address of record exists in this chain because there is no LLC.

2. Known asserter in the chain — NOT PRESENT. Neither the current nor any prior assignee (Celgene Corporation; parent Bristol‑Myers Squibb) appears on the Acacia / Marathon / IV / IPNav / Wi‑LAN / Conversant / Vringo / Pendrell / Innovatio / MPHJ / Round Rock / Spangenberg lists, nor in the Unified Patents or RPX high-frequency-plaintiff directories. Both are Fortune‑500 pharmaceutical manufacturers.

3. Repeat correspondent across the chain — UNCLEAR. With a single recorded assignment I could not retrieve the recording correspondent from the sources available, so there is no basis to find recurrence. Separately: the prosecuting attorney of record for this patent family, Anthony Insogna (Pennie & Edmonds, then Jones Day), recurs across the sibling patents (’262 and others) per the antitrust complaints — but prosecution counsel is not the assignment-recording correspondent, and I decline to bootstrap that into a signal. Marked unclear on the record available; not a finding.

4. Cascading transfers — NOT PRESENT. One assignment, executed pre-issuance, inventor → employer. No chained LLCs, no shared correspondent addresses, no sub‑24‑month cascade.

5. Pre-litigation transfer — NOT PRESENT. The sole assignment is dated 2013‑03‑01, roughly 4.5 years before the first ’3939 assertion in Celgene v. Hetero (2:17‑cv‑03387, 2017). It is an inventor-to-employer instrument, not a transfer arranged to set venue or manufacture standing.

6. Bankruptcy fire-sale — NOT PRESENT. Celgene never filed Chapter 7/11. Its exit was a $74B stock-for-stock acquisition by Bristol‑Myers Squibb (completed 2019‑11‑20); the FTC-conditioned remedy was the divestiture of OTEZLA to Amgen, not a patent fire-sale.

7. Privateering — NOT PRESENT. Celgene/BMS retained the ’3939 and asserted it in its own name against generic ANDA filers (Hetero, Mylan and the other 2017 first-to-file cohort). There is no operating-company→NPE transfer, and no third-party asserter appears anywhere in the chain.

8. Defensive aggregator — NOT PRESENT. The chain terminates at Celgene Corporation (now a BMS subsidiary). No RPX / AST / LOT / Unified / OIN acquisition. Inverse signal does not apply — the patent was actively asserted, not neutralized.

Verdict

Operating-company assertion.

Justification: the entire recorded chain is a single 2013‑03‑01 inventor-to-employer assignment from Jerome B. Zeldis to Celgene Corporation, with no subsequent assignments — Celgene (now a BMS subsidiary following the 2019‑11‑20 merger) remains owner of record. The patent is Orange-Book-listed (U‑1360) against POMALYST®, a product Celgene actually commercialized, and it was asserted by the owner in its own name against competing generic ANDA filers in Celgene v. Hetero, 2:17‑cv‑03387 (D.N.J.) and the parallel 2017–2022 dockets. Not one of the eight NPE signals is present on the record; the "unclear" correspondent signal (3) is an evidentiary gap, not evidence.

Verify: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 8673939); Google Patents legal events page — https://patents.google.com/patent/[US8673939B2](/patent/US8673939B2)/en.

I'll research US patent 8673939 and its prior-art citations. Let me run several searches.

Let me dig into the specific citation lists on this patent.

Let me retrieve the actual citation lists and claim text.

Prior-Art Analysis — US 8,673,939 B2

1. Identification of the patent (literal identifiers)

Sources: https://patents.google.com/patent/[US8673939B2](/patent/US8673939B2)/en ; https://www.freepatentsonline.com/8673939.html ; family chain shown on https://patents.google.com/patent/[US9283215B2](/patent/US9283215B2)/en

Scope of the claims. The claims (not reproduced in the specification text supplied to me, which is description‑only) are, per the title and abstract, method‑of‑treatment claims directed to treating multiple myeloma by administering 4‑(amino)‑2‑(2,6‑dioxo(3‑piperidyl))‑isoindoline‑1,3‑dione ("Actimid™"/CC‑4047; today's INN pomalidomide), with dependent claims reflecting the dosing/route/combination language in the specification (e.g., ~0.1–1 mg/day, ~0.1–5 mg every other day, oral administration, optional dexamethasone, cycling regimens). I could not retrieve the verbatim claim text within my search budget — the claim‑mapping below should be re‑checked against the printed claims before being relied on.

2. Methodological caveat (important)

You asked me to work from "each patent citation for 8673939." I attempted to pull the complete front‑page "References Cited" pane (USPTO PatentCenter / Google Patents / FPO). My searches returned only a partial list (FPO truncates after ten entries) and the specification's own "incorporated by reference" passages. I therefore split the analysis into:

• Part A — references expressly identified in the 8,673,939 specification itself (directly documented in the patent text; high confidence), and
• Part B — citations visible in the face‑of‑patent/related‑family citation data (partial; moderate confidence), and
• Part C — the references that would actually matter under §102 (non‑patent literature on IMiDs in multiple myeloma), flagged by confidence.

Nothing below is invented; items I could not verify are marked [UNVERIFIED].

3. Legal framework — which §102 applies, and the critical dates

• The application 13/782,728 was filed 2013‑03‑01, i.e. before the AIA first‑inventor‑to‑file changeover (16 March 2013), and all claims carry an effective filing date of 2002. Pre‑AIA 35 U.S.C. §102(a)/(b)/(e) governs.
• Effective filing date: 2002‑05‑17 (provisional 60/380,842), with 2002‑11‑06 (60/424,600) for subject matter added there.
• Therefore:
  • §102(a)/(e) art must predate 17 May 2002 (or be a U.S. application/patent with an earlier effective date).
  • §102(b) art must predate the one‑year critical date — at the latest 15 May 2002 (parent filing) and, if the provisional supports the claims, 17 May 2001.
• Anticipation vs. obviousness: A §102 rejection requires one reference disclosing every limitation, arranged as claimed. Nearly all compound‑disclosure citations here disclose the compound/genus, not treatment of multiple myeloma, and therefore are §103 art, not §102 art, as to method claims. No cited reference anticipates the method claims (with the possible NPL exception flagged in Part C).

4. Part A — References expressly cited/incorporated in the 8,673,939 specification

5. Part B — Citations visible in the face‑of‑patent / family citation data (partial list)

The FPO record for 8,673,939 lists these as "US Patent References" (list truncated at ten; [UNVERIFIED as to completeness]):

Key point: every reference actually displayed in this list post‑dates the 2002 priority date and shares the same inventor/family. Under pre‑AIA §102, they are not available as prior art at all, and under pre‑AIA §103(c) commonly owned art is disqualified for obviousness. Any §102 challenge built on this cluster would fail.

6. Part C — The references that would actually support a §102 attack

Because the claims are method‑of‑treatment claims with a 2002‑05‑17 effective date, only a single reference disclosing treatment of multiple myeloma with 4‑amino‑2‑(2,6‑dioxopiperidin‑3‑yl)isoindoline‑1,3‑dione (CC‑4047) can anticipate. The realistic candidates are non‑patent literature (I could not fully verify which of these appear on the 8,673,939 face; confidence flagged):

1. Lentzsch S, Rogers MS, LeBlanc R, et al., "S‑3‑Amino‑phthalimido‑glutarimide inhibits angiogenesis and growth of B‑cell neoplasias in mice," Cancer Research 62(8):2300‑2305 (2002‑04‑15). — If it discloses growth inhibition of myeloma/B‑cell neoplasia by the claimed compound, it is a §102(b) candidate only if published before the critical date (15 May 2002/17 May 2001) — publication on 15 April 2002 falls inside the one‑year grace if the parent‑filing date is the measure, so it is more likely §102(a) (swear‑behind) or §103 art. Confidence that it is cited somewhere in this family: moderate. Confidence that it anticipates: low (murine model; may not disclose human MM treatment "as arranged in the claim"). [Not verified in this session.]
2. Corral LG, et al., J. Immunol. 163:380‑386 (1999) — differential cytokine modulation/T‑cell activation by the two IMiD classes (CC‑5013/CC‑4047). §102(b) as to compound activity; no myeloma treatment → §103, not §102. [UNVERIFIED whether cited on this patent.]
3. Anderson et al., "Novel biologically based therapies for myeloma," VIII International Myeloma Workshop, Banff, Alberta, Canada, May 4‑8, 2001 — an abstract that (if it names Actimid™/CC‑4047 in myeloma) would be a §102(b) printed publication and the strongest single‑reference anticipation candidate. I saw this reference on a related Celgene patent's list, not confirmed for 8,673,939. [UNVERIFIED — verify against the patent's own face‑of‑patent list.]
4. Mitsiades et al., "Apoptotic signaling induced by immunomodulatory thalidomide analogs in human multiple myeloma cells," Blood 99(12):4525‑4530 (2002‑06‑15) — human MM cells + IMiD analogues. Publication post‑dates 2002‑05‑17, so it is not §102(a)/(b) art against this priority date (it could only be relevant if the Nov‑2002 provisional subject matter is claimed). [UNVERIFIED whether cited.]
5. D'Amato RJ et al., "Thalidomide is an inhibitor of angiogenesis," PNAS 91:4082‑4085 (1994) — anti‑angiogenic rationale; §102(b) but no Actimid/MM disclosure → §103 only. [UNVERIFIED whether cited.]
6. Dimopoulos et al. 2004 (MDT), Attal et al. 2012, Alexanian et al. 2002/2004, Barlogie et al. 2001‑2003, Baz et al. — these appear on the reference list of the related reissue RE48,890 (lenalidomide post‑transplant). All 2003‑2012 → not §102 art as to the 2002 priority date; §103 relevance only, and then only if the point of novelty is the dosing/combination regimen. (Source: https://patents.justia.com/patent/[RE48890](/patent/RE48890))

7. Bottom line

• No patent citation on US 8,673,939 anticipates its multiple‑myeloma method claims under §102. The cited U.S. patents (A1–A11) disclose the compound genus (notably U.S. 5,635,517, which discloses the 4‑amino‑benzyl‑substituted 2‑(2,6‑dioxopiperidin‑3‑yl)isoindoline‑1,3‑diones encompassing the claimed compound), the second active agents (GM‑CSF/G‑CSF), or dosage‑form technology — all as §103 bases.
• The face‑of‑patent "US Patent References" cluster (Part B) is the applicant's own family with the same 2002 priority date; those are not §102 prior art (and are §103(c)‑disqualified).
• The only plausible §102 candidates are non‑patent literature on IMiDs in multiple myeloma published before the critical date, and their quotation on this patent's face is [UNVERIFIED] — the decisive §102 showing would come from Lentzsch (Cancer Res. 2002) and/or the Banff 2001 International Myeloma Workshop abstract, both of which I flagged as needing verification against the printed reference list.

8. Recommended next steps to complete the record (not possible within my tool budget here)

1. Pull the verbatim "References Cited" pane from USPTO PatentCenter for application 13/782,728 (and the underlying IDS in 10/438,213), which will give the authoritative patent and non‑patent citation lists — including any foreign references (e.g., WO publications) I could not see.
2. Pull the printed claims of US 8,673,939 B2 to confirm the exact method‑step limitations before finalizing any §102 mapping.
3. Mine the litigation records already indexed for this patent — the New Jersey District Court actions (2:19‑cv‑05802; 2:21‑cv‑02111; 2:22‑cv‑01993; 2:17‑cv‑03159; 2:17‑cv‑03387; 2:18‑cv‑10775), the North Carolina Middle District action 1:18‑cv‑00540, and the Federal Circuit appeal 21‑1154 (links listed on the Google Patents page) — the defendants' invalidity contentions and IPR petitions in those cases contain the fully developed §102/§103 prior‑art charts.

Confidence summary: identification/priority/family data — high (multiple corroborating sources cited above); specification‑cited references A1–A16 — high (drawn from the patent text supplied); face‑of‑patent list — partial only (FPO truncation); Part C NPL anticipation candidates — low to moderate and explicitly unverified.

I'll ground this analysis in the actual prior-art references of record. Let me search for them.

Let me pull the claim text and the examiner's original obviousness rejections.

Obviousness Analysis — US 8,673,939 B2 under 35 U.S.C. § 103

Note on the "Prior Art" section of the source page. The Google Patents page for US 8,673,939 contains only a thin "Prior art keywords" field — administered, compound, patients, day, cancer — plus an auto-generated "Chemical compound" tag list (including UVSMNLNDYGZFPF-UHFFFAOYSA-N = pomalidomide and GOTYRUGSSMKFNF-UHFFFAOYSA-N = lenalidomide) and a "Prior art date: 2002-05-17." Those keywords are auto-extracted from the specification and are not references. The operative prior art for this patent is the set of record in the '939 prosecution and in the parallel ANDA litigation, which Hetero's own counterclaim enumerates: "the Lentzsch reference, the Hideshima reference, the D'Amato reference, the '471 patent, and the '517 patent." (Hetero counterclaim ¶68) I analyze §103 against that record plus the references the Examiner actually applied.

Cross-reference to the earlier sections (and two corrections/refinements). The prior summary hedged that the claim-1 preamble and the "claim 26" label were reconstructions. Both are now corroborated:

• Claim 1 does recite a prior-therapy limitation — the Examiner's July 9, 2013 rejection expressly addressed claims "wherein the previous therapy is, inter alia, thalidomide, lenalidomide or a proteasome inhibitor."
• Claim 26 is confirmed as the cyclical claim: the litigation claim table reproduces it as "[pomalidomide] or a solvate thereof, wherein the compound is administered in one or more cycles, each of which comprises administering the compound for a period of time followed by a period of rest." (Pls.' Br., claim chart)
• One new internal inconsistency worth flagging: the '939 specification's own dosing statement for the claimed compound is "from about 0.1 to about 1 mg per day, or alternatively from about 0.1 to about 5 mg every other day," escalating to 50 mg/d — while the asserted dose claims recite about 1–5 mg per day. The claimed range is thus a subset of the patent's own disclosed escalation schedule and of the range disclosed in the art (Schey, infra).

1. The claim scope to be analyzed (element decomposition)

Two structural facts drive everything below: (i) the compound is old — pomalidomide is a disclosed species of the '517 genus; and (ii) the only disputed patentability hook was efficacy, which Celgene itself conceded ("[p]atentability of the claimed methods depends upon efficacy") and which the court then read out of the claim.

2. Level of ordinary skill

A POSA here is a medicinal chemist or oncologist with an M.D. or Ph.D. and several years in hematologic oncology / IMiD development — the profile of the Hideshima, Lentzsch, Anderson, D'Amato, Schey and Davies author groups. The field was small, the compound class was a handful of thalidomide analogues, and the workers were largely the same people publishing on both sides of the priority date.

3. The prior art of record

4. Combinations that render the claims obvious

Combination A — '517 (+ '471/'230/'554) in view of Hideshima (2000): the core "administer pomalidomide to an MM patient" claim

• '517/'554 supply the compound and the method-of-treatment framing.
• Hideshima supplies the disease (multiple myeloma), the patient population (MM refractory to conventional therapy, including melphalan/doxorubicin/dexamethasone), the mechanism (direct apoptosis/G1 arrest in MM cells), and the class-wide teaching that IMiDs — the very class to which pomalidomide belongs — overcome drug resistance.
• Motivation: same field, same compounds, same mechanism (TNF-α modulation + direct anti-MM activity), and a finite, identified set of thalidomide analogues. KSR makes this the paradigm case: "a finite number of identified, predictable solutions" → obvious to try.
• Reasonable expectation of success: Hideshima reports the result in patient MM cells ex vivo — not mere conjecture.

Combination B — '517 + D'Amato (2001): pomalidomide, specifically, for multiple myeloma

D'Amato (2001) is close to anticipating: it names the compound and states it "directly inhibit[ed] myeloma proliferation" and was more efficacious than thalidomide in vitro and in vivo. Combined with '517's structural disclosure of pomalidomide, every element of the claim-1 body is present. The only gap is the prior-therapy limitation, which Hideshima (2000) supplies (resistance to melphalan/doxorubicin/dexamethasone) — yielding D'Amato (2001) + Hideshima (2000) + '517.

Combination C — Lentzsch (2002) + Schey (2002): relapsed/refractory MM in a human

• Lentzsch supplies drug-resistant myeloma, complete responses, and no toxicity — i.e., the relapsed/refractory population.
• Schey supplies the human clinical data in exactly that population, plus the dose ceiling (up to 5 mg/day) that frames the claimed 1–5 mg/day range.
• Motivation: once a potent, well-tolerated analogue shows in vivo activity in drug-resistant MM, moving it into relapsed/refractory patients is the ordinary therapeutic logic of oncology — and Schey shows the field had already made that move.

Combination D — Cohen (1982) + Kyle (2001) + Schey (2002): the cyclical claim (claim 26) and the dose range

• Cohen (1982) teaches the 28-day cycle, 21 days on / 7 days off.
• Kyle (2001) teaches cyclical thalidomide + dexamethasone in MM.
• Schey (2002) teaches pomalidomide in the claimed population at up to 5 mg/day.
• Motivation: cyclic dosing is bedrock oncology practice (the NCI's own explanation of a "cycle" as treatment followed by rest); claim 26's language ("a period of time followed by a period of rest") is broader than the hard-coded 21/7 schedule and therefore even easier to reach. For the 1–5 mg/day claims, the range sits inside the art's disclosed MTD; narrowing a disclosed range on the basis of routine dose titration is obvious absent a showing of a critical, unexpected threshold (In re Peterson; In re Boesch).

Combination E — the Examiner's own five-reference rejection: Kyle (2001) + Davies (2001) + Corral (1999) + Muller (1999) + '554

This is not a reconstruction: the PTO did reject the claims as "prima facie obvious to one of ordinary skill in the art," and gave the "obvious to try" rationale in terms:

"one of ordinary skill in the art would have understood that [pomalidomide] would provide benefits in treating [MM] whether the patient had previous therapy with thalidomide, lenalidomide, proteasome inhibitor, etc. The skilled artisan would have at least found it obvious to try in these patients as well as others with MM."

The claims issued only after the Thakurta Declaration was filed. That procedural fact matters twice over: it establishes that the prima facie case was made on the merits, and it isolates "unexpected results" as the sole rebuttal.

5. Motivation to combine — synthesis

Each articulated KSR rationale applies:

1. Same field of endeavor / same problem — treating multiple myeloma with IMiDs.
2. Same compound class, same mechanism — thalidomide analogues inhibiting TNF-α and directly killing MM cells.
3. Finite, small, predictable set of analogues — with potency differences already quantified (Stirling declaration; D'Amato 2001; Davies 2001).
4. Obvious to try — the Examiner said so, in writing.
5. Design/optimization incentives — dose-ranging and cycle design were routine (Kyle, Cohen, Coleman).

6. Reasonable expectation of success

Not marginal: Hideshima (2000) showed activity in patient-derived, multidrug-resistant MM cells; Lentzsch (2002) showed complete and sustained regressions of drug-resistant myeloma tumors in vivo; Schey (2002) showed clinical activity in relapsed/refractory MM patients. A POSA would expect success, and to a substantial degree the art had already demonstrated it.

7. Objective indicia and the rebuttal problem

Celgene's only non-obviousness theory was "unexpected results" — that pomalidomide works in lenalidomide-refractory MM, because resistance between the two compounds is not reciprocal (Thakurta Declaration, Oct. 7, 2013: "surprisingly found that the resistance … is not reciprocal"). Four problems:

1. No nexus after the Markman ruling. The court held the "method of treating multiple myeloma" preamble not limiting and refused "reading into the claim an efficacy limitation." Celgene had conceded patentability "hinge[d] upon" the preamble. The asserted unexpected result is an efficacy result; once efficacy is out of the claim, the supposed nexus dissolves.
2. The result was disclosed. Hideshima (2000) taught IMiDs overcome drug resistance of MM cells; Lentzsch (2001/2002) taught pomalidomide regresses drug-resistant myeloma. The antitrust complaints put it bluntly: "there was nothing surprising about the fact that the more potent thalidomide analog … would be used where the multiple myeloma patient had become relapsed or refractory to less potent analogs."
3. The direction was predictable. The Stirling Declaration and D'Amato (2001) had already published pomalidomide's far greater potency; a more potent analogue working where a less potent one failed is the expected ordering, not a surprise.
4. Prosecution-procedural weakness. The unexpected results entered via an examiner interview and a §1.132 declaration; the interview summary records the Examiners "agreed that Applicant has shown unexpected results" only at the interview stage. That is classic Wands-style attorney argument rather than a robust nexus.

8. Caveats I will not paper over

• No court has adjudicated the '939's validity on the merits in the record I retrieved. The D.N.J. case was Markman'd (June 16, 2020) and settled; the '939 expired May 15, 2023 (Nov. 15, 2023 with PED). Assertions that the patents "would be found invalid" are party allegations, not holdings.
• The S.D.N.Y. March 2025 opinion dismissed the Walker Process and sham-litigation theories as to the '939 on pleading grounds — "Plaintiffs have failed to state a Walker Process claim as to the '467 and '5939 patents" (Justia). That is not a validity or enforceability determination, and it cuts in Celgene's favor procedurally.
• I still have not verified the verbatim granted claim set from USPTO PatentCenter, and I could not confirm the D'Amato "'539" document's number. The single CAFC matter in the family record is No. 21-1154 (2021) — there is no 2026 Federal Circuit appellate activity I could find, consistent with an expired patent.

Bottom line. On this record, the '939 claims are most vulnerable on a Hideshima (2000) + '517/'471 combination for the core treating claim, D'Amato (2001) + Hideshima as an anticipation-adjacent attack, and Cohen (1982) + Kyle (2001) + Schey (2002) against the cyclical and 1–5 mg/day claims. The decisive vulnerability is structural rather than evidentiary: the June 2020 construction removed the efficacy limitation on which Celgene's entire non-obviousness case was built, leaving claims that read on administering a compound the art had already disclosed, to a patient population the art had already treated, at a dose the art had already disclosed.