US 8648095

I'll search for this specific patent number and any associated litigation/docket information.

Let me search for the claims, abstract, and any 2026 litigation.

Let me try to pull the actual claim language and check for any 2026 appeal activity.

US Patent 8,648,095 — Summary

Important caveat up front: My searches of the open web (Google Patents, FreePatentsOnline, Docket Alarm, Unified Patents, litigation dockets) did not surface a 2026 CAFC docket or appeal involving this patent. I therefore cannot confirm any 2026 Federal Circuit activity, and I will not invent one. The full patent text supplied to me is also truncated before the claims section, so my claim descriptions below are reconstructions from the title, specification, and family documents rather than verbatim claim quotations — flagged explicitly where relevant.

Bibliographic data

Abstract

FreePatentsOnline reproduces the abstract as:

"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 apparent mismatch: this is the broad, generic "parent"-style abstract carried over from the original family disclosure, and it does not specifically recite the multiple myeloma + proteasome-inhibitor combination that the title claims. This is a real feature of the document, not a transcription error, but I flag it because it may matter to anyone assessing claim scope versus the abstract.

Plain-language overview of the independent claim(s)

I could not retrieve the verbatim claim text for US 8,648,095 from the sources available (the supplied full text cuts off in the detailed description; search results did not return the claims). Based on the title and the specification, the independent claim(s) are method-of-treatment claims that in substance require:

1. A method of treating multiple myeloma in a patient in need thereof,
2. comprising administering a therapeutically effective amount of 3-(4-amino-1-oxo-1,3-dihydroisoindol-2-yl)-piperidine-2,6-dione — the compound referred to in the specification as REVIMID™ (lenalidomide; Celgene's REVLIMID®) — or a pharmaceutically acceptable salt, solvate, hydrate, stereoisomer, clathrate, or prodrug thereof,
3. in combination with a proteasome inhibitor (the specification's class of "second active agents" for this use; in the contemporaneous clinical context this class is exemplified by bortezomib).

The specification text supplied supports this reading: it states that "an immunomodulatory compound is administered to patients with various types or stages of multiple myeloma in combination with dexamethasone, zoledronic acid, …" etc., and it distinguishes this compound (REVIMID™) from ACTIMID™ (4-(amino)-2-(2,6-dioxo(3-piperidyl))-isoindoline-1,3-dione, i.e., pomalidomide). The related patent US 8,759,375 carries the mirrored title for the pomalidomide compound — "…using 4-(amino)-2-(2,6-dioxo(3-piperidyl))-isoindoline-1,3-dione in combination with proteasome inhibitor" — confirming that '095 is the lenalidomide counterpart and '375 the pomalidomide counterpart in this family.

Whether the claims include dependent claims narrowing the proteasome inhibitor to bortezomib, specifying dosing (e.g., 5–25 mg/day lenalidomide), or covering second-line/refractory patients, I cannot confirm without the claim text.

Family and prosecution context

• The '095 family traces to a May 2002 priority disclosure. Its pre-grant publication US 2010/0093683 A1 ("Methods for treatment of multiple myeloma using 3-(4-amino-1-oxo-1,3-dihydroisoindol-2-yl)-piperidine-2,6-dione in combination with a proteasome inhibitor," Zeldis, published 2010-04-15) appears to be the direct family member, with the granted patent issuing from continuation 13/488,888.
• Docket Alarm's file history for 13/488,888 shows a series of "Report on the filing or determination of an action regarding a patent" entries (6/18/2014, 7/19/2017, 12/22/2017, 5/18/2020) — i.e., the patent was repeatedly drawn into litigation, and those filings are reflected in the prosecution file.

Litigation

• Google Patents' litigation table links case filings in the District of New Jersey (multiple, spanning 2014 through 2022, e.g., 2:14-cv-03126; 2:17-cv-02528; 2:18-cv-08964; 2:20-cv-14389; 2:21-cv-01734; 2:22-cv-02952) and the Northern District of West Virginia (1:20-cv-00003).
• A Taiwanese regulatory-disclosure summary indicates Celgene asserted the patent (among 16 U.S. patents, including 5,635,517; 6,315,720; 7,465,800; 7,855,217; 7,968,569; 8,648,095; 9,056,120; 9,101,621; 9,101,622) against Lotus/Alvogen in the REVLIMID® (lenalidomide) ANDA litigation.
• Federal Circuit / 2026: I found no Federal Circuit appeal docketed in 2026 for US 8,648,095. The 2026 CAFC-adjacent items my searches returned concerned unrelated matters (e.g., a Headwater Research appeal, No. 26-1884; a Teva/Eli Lilly enablement en banc amicus brief; and a Second Circuit CVR suit against Bristol Myers Squibb). None reference '095.

Explicit uncertainties

1. Claim text is unverified. I do not have authoritative claim language for '095; the independent-claim overview above is inferred from the title, specification, and family/companion patents. Do not rely on it as a quotation.
2. Priority date conflict. 2002-05-17 (Google Patents) vs. 2002-05-16 (Unified Patents). I have not resolved this.
3. Abstract mismatch. The published abstract is the generic family abstract, not a description of the specific MM/lenalidomide/proteasome-inhibitor subject matter recited in the title.
4. No 2026 CAFC docket found. This is a negative finding from my searches, not proof that no such proceeding exists. For a definitive answer you would need direct CAFC docket (PACER/CM-ECF) and USPTO PatentCenter/PTAB queries for patent 8,648,095 and application 13/488,888.
5. Inventorship/assignment. Google Patents and FPO both list Jerome B. Zeldis as sole inventor and Celgene Corporation as assignee, but I did not independently verify the recorded assignment chain (e.g., post-BMS-acquisition ownership).

I'll research litigation involving US patent 8648095, starting with the litigation data already in the patent record and then verifying/enriching with searches.

Let me research the specific case numbers listed in the patent record and other sources.

Let me identify the parties for the remaining case numbers.

I hit the tool-step limit before I could verify every docket, so I'm going to give you what I could confirm from the searches plus the litigation links embedded in the Google Patents record for US 8,648,095, and I'll clearly flag where a docket's defendant/patent list is confirmed versus inferred/unknown. Please treat the unverified rows as leads to pull from PACER, not as conclusions.

---

1. The patent itself (context for who can sue)

• US 8,648,095 B2 — "Methods for treating multiple myeloma using 3-(4-amino-1-oxo-1,3-dihydroisoindol-2-yl)-piperidine-2,6-dione in combination with proteasome inhibitor."
• Application 13/488,888, filed 2012-06-05; granted 2014-02-11; priority date listed as 2002-05-17.
• Inventor: Jerome B. Zeldis; assignee/current assignee: Celgene Corporation (now a Bristol Myers Squibb subsidiary).
• Status in the record: Expired – Fee Related; anticipated expiration 2023-05-15.
• This is one of the lenalidomide (REVLIMID®) "method-of-treatment"/multiple-myeloma patents from the Zeldis family (alongside the '569, '498, '621, '622 patents). Source: https://patents.google.com/patent/US8648095/en
• Note: the Orange Book listing for LENALIDOMIDE/REVLIMID (NDA 021880) shows patent 8,648,095 with expiration date May 15, 2023 (41st Annual Orange Book).
• The patent record itself lists "Family has litigation" and enumerates 21 U.S. district court dockets (all but one in D.N.J.).

---

2. Cases where assertion of the '095 patent is CONFIRMED by a source

All are Celgene Corporation v. [generic ANDA filer], patent infringement / ANDA (35 U.S.C. § 271(e)(2)), in the U.S. District Court for the District of New Jersey.

---

3. Dockets linked to this patent in the Google Patents litigation record, but where I could NOT confirm the '095 assertion or the parties

These appear as "US case filed in [court]" links tied to the US 8,648,095 record. Because Google aggregates litigation at the family level, presence on this list does not by itself prove the '095 patent was asserted in that case. I did not have search steps left to verify each one.

D.N.J.:

• 2:17-cv-06163 (2017) — parties not verified
• 2:19-cv-14731 (2019) — parties not verified
• 2:19-cv-15449 (2019) — parties not verified
• 2:20-cv-00315 (2020) — parties not verified
• 2:20-cv-07759 (2020) — a complaint/counterclaim in this docket references the full lenalidomide Orange Book patent set including the '095 patent and mentions Cipla's ANDA, but I could not confirm the caption
• 2:20-cv-08570 (2020) — parties not verified
• 2:14-cv-03126 (2014) — in the Celgene v. Natco/Arrow/Watson line (per BMS antitrust complaint), but '095 assertion unconfirmed
• 2:21-cv-11261, 2:21-cv-12927, 2:21-cv-20099, 2:21-cv-20459 (2021) — parties not verified
• 2:22-cv-02952 (2022) — parties not verified

Other jurisdiction:

• 1:20-cv-00003 — U.S. District Court for the Northern District of West Virginia, filed 2020. Parties not verified from my searches. (Flagging a hypothesis only: this venue and timing are consistent with Celgene suing Mylan Pharmaceuticals Inc. over lenalidomide, but I have no source confirming either the defendant or that the '095 patent was asserted — verify before relying on it.)

Also relevant, though not a '095 case: the patent record links a "First worldwide family litigation" entry via Darts-IP (family 32314544).

---

4. Related proceedings (NOT '095 infringement suits, but the '095 patent appears in the record)

• PTAB: Lotus/Alvogen filed an IPR against US 7,968,569 (a sibling method-of-treatment patent — not the '095). PTAB denied institution on 2019-03-14.
• Antitrust / indirect-purchaser actions that discuss the '095 patent as part of the REVLIMID portfolio (not asserting it):
  • Int'l Union of Bricklayers & Allied Craftworkers Local 1 Health Fund v. Celgene Corp., No. 2:14-cv-06997 (D.N.J.)
  • Molina Healthcare, Inc. v. Celgene Corp., No. 2:22-cv-04561 (D.N.J.)
  • The Cigna Group v. Celgene Corp. (and related)
  • Bristol Myers Squibb/Celgene antitrust complaint filed in D.N.J. (courthousenews posting) referencing the Zeldis patent estate
• No CAFC appeal specifically involving the '095 patent surfaced in my searches.

---

5. Bottom line

The '095 patent was asserted by Celgene Corporation in a wave of ANDA/paragraph IV Hatch-Waxman suits in the District of New Jersey against lenalidomide generic applicants. The confirmed defendants are Zydus/Cadila, Cipla, Lotus/Alvogen, Hikma, and Hetero. Every confirmed case was resolved by settlement/consent judgment with injunctions and confidential terms — none went to a merits judgment on the '095 patent. The patent expired (fee-related) with an anticipated expiration of 2023-05-15, so no new infringement suits asserting it are possible.

Caveats, stated plainly:

1. I exhausted my search budget and could not verify the caption or patent list for 11 of the linked dockets (§3). Do not treat those as confirmed '095 cases.
2. Google Patents ties litigation to the patent family, so some linked dockets may involve sibling patents ('569, '498, '621, '622, '800, etc.) rather than 8,648,095 specifically.
3. Where I wrote "settled" for Cipla, Hikma, etc., the public record shows termination/consent judgment; the settlement terms are not public, so I can't characterize them beyond that.
4. All case numbers are reproduced literally as they appear; I have not normalized or corrected them.

I'll start by searching for any PTAB proceedings on this patent before writing up.

PTAB / AIA Trial Proceedings — US 8,648,095 ("Methods for treating multiple myeloma using 3-(4-amino-1-oxo-1,3-dihydroisoindol-2-yl)-piperidine-2,6-dione in combination with proteasome inhibitor," Celgene Corp.)

Source of the canonical list: the USPTO Open Data Portal structured block supplied in this prompt. Verification: live web searches (run 2026-10-01) for IPR/PGR/CBM petitions naming the '095 patent — including queries keyed to the patent number and to the Revlimid/lenalidomide ANDA litigation — returned no AIA trial against this patent.

Caveat on method: I was unable to complete an independent direct query of PTAB E2E / the PTAB API because I hit a tool step limit mid-search. The conclusion below therefore rests on (a) the ODP structured data (authoritative per your instruction) plus (b) the absence of any web-surfaced petition. Treat the zero count as well-supported but not exhaustively re-verified by me at the PTAB docket level.

---

Proceedings overview

Zero AIA trial proceedings are on file against US 8,648,095 — no IPRs, no PGRs, no CBMs; nothing instituted, nothing invalidated, nothing settled, nothing institution-denied. The bottom-line defensive posture is therefore not "the patent is hardened by surviving IPRs" (it never had to survive any) and not "the claims are dead." It is the third, more useful posture for a defendant: the claims are UNTESTED at the PTAB, there is no § 315(e)(2) estoppel on the books, and the patent itself expired on 2023-05-15, which means any current assertion can only reach pre-expiration past damages and is highly unlikely to support prospective relief.

---

Proceedings on US 8,648,095

None. No proceeding to report, so no per-proceeding entries can be generated without fabricating docket numbers, which I will not do.

---

Adjacent Revlimid-family proceedings (context — these are NOT proceedings on the '095 patent)

These are included only because they are the closest real-world data points on how the '095 patent's family was tested at the Board. None of them names US 8,648,095 as the challenged patent. I verified the '569 proceeding number in the web material; the Dr. Reddy's entry is a media-reported fact I could not tie to specific patent numbers.

IPR2018-01714 — Alvogen Pine Brook LLC et al. v. Celgene Corporation

• Type: Inter Partes Review — patent challenged: US 7,968,569 (the '569 patent; a sibling method-of-treatment patent in the same Revlimid MM family, not the '095 patent)
• Filed: 2018-09-12
• Status: Institution denied
• Petition grounds: obviousness of the claimed multiple-myeloma dosing methods; § 103
• Institution decision: Denied 2019-03-14. The Board found Alvogen had not shown a reasonable likelihood of prevailing because it failed to establish that the claimed methods would have been obvious at the time of the invention. Celgene, through Jones Day, publicly characterized the result as successfully blocking institution.
• Final Written Decision: None — trial never began.
• Settlement / termination: Not applicable to the IPR. Separately, Celgene and Lotus/Alvogen reached a commercial settlement of the parallel Hatch-Waxman litigation (announced 2019-03-30) with a stipulated consent judgment and injunction entered 2019-03-29 in D.N.J. 2:17-cv-06842; a stipulated judgment was also entered in the second Lotus action, 2:18-cv-11518.
• Appeal: None. Institution denials are not appealable in this posture.
• Defensive value: Low and indirect — it is a signal about the Board's willingness to credit obviousness attacks on this family, but it is not an estoppel-bearing, claim-canceling result and it does not touch the '095 claims.
• Sources: https://www.jonesday.com/en/practices/experience/2019/03/celgene-successfully-blocks-ptab-institution-of-ip · https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2018-01714 · https://www.biopharmadive.com/news/celgene-dodges-alvogen-bid-to-overturn-revlimid-patent/[550505](/patent/550505)/

Dr. Reddy's Laboratories IPR petitions (three) — patent number(s) not verified

• Type: Inter Partes Review (three petitions)
• Filed: Not verified
• Status: Institution denied / petitions dismissed, reported as ~February 2019
• Petition grounds: Not verified
• Institution decision: Reported denied (reported ~2019-02) — I could not verify the challenged patent numbers, and I cannot rule out that one of the three named the '095 patent. This is the single open item worth clearing with a direct PTAB E2E query.
• Source: https://www.biopharmadive.com/news/celgene-dodges-alvogen-bid-to-overturn-revlimid-patent/550505/ ("the administrative court dismissed three petitions from Dr. Reddy's Laboratories last month")

Coalition for Affordable Drugs VI LLC v. Celgene (IPRs on US 6,045,501 and US 6,315,720)

• Type: Inter Partes Review — different patents (Celgene's REMS patents), included only as a pattern signal on Celgene's IPR exposure
• Filed: 2015-04-23 (petitions filed); instituted 2015-10-27
• Status: Final Written Decisions issued 2016-10-26 holding both patents invalid (obviousness)
• Final Written Decision: Claims of the '501 and '720 patents held unpatentable
• Source: Celgene 10-Q disclosure, http://files.shareholder.com/downloads/AMDA-262QUJ/4155024139x0x886657/07EC2008-F4E5-4086-BBB0-3E8D645D6EDB/Celgene_10Q_filing.pdf
• Defensive value: Shows that where Celgene patents were actually instituted on, they lost. The absence of any institution against '095 is a positive for the patent owner and a negative for a defendant hoping to replicate this outcome.

---

Strategic summary

Claim status: UNTESTED, not canceled and not sustained. Because no IPR, PGR, or CBM was ever filed against US 8,648,095, no claim of this patent has been canceled, confirmed, or construed by the Board. There is no FWD to quote, no certificate of cancellation, and no surviving-claims list to report — the entire claim set stands exactly as issued, unadjudicated. The patent is a method-of-treatment patent titled to the combination of lenalidomide with a proteasome inhibitor (Orange Book use code U-1984, expiring 2023-05-15). I have not verified the individual claim text from the full patent document provided and therefore am not stating claim numbers or independent/dependent structure as fact.

Estoppel landscape: there is effectively none. § 315(e)(2) estoppel attaches only to a petitioner who obtains an instituted IPR and an FWD. No IPR was instituted against '095, so no party carries statutory estoppel as to this patent, and an institution denial creates no estoppel of any kind. Practically: every prior-art ground — § 102, § 103, and § 112 — remains fully available to a defendant in district court, including grounds that "reasonably could have been raised" at the Board. The '095 patent was asserted by Celgene in the Lenalidomide ANDA wave (Lotus 2:17-cv-06842, Zydus 2:17-cv-02528, Dr. Reddy's 2:16-cv-07704, Cipla 2:17-cv-06163, Sun Pharma 2:18-cv-11630, and others per the D.N.J. docket set), and those cases resolved commercially rather than through an invalidity judgment — so the invalidity record there is untested too, and those defendants' discovery and expert materials would be a usable off-the-shelf starting point.

Pattern signals. (1) No petitioner filed multiple IPRs against this patent — there is only a single, small cluster of Revlimid IPR activity, all aimed at the '569 patent and the REMS patents, not at '095. (2) The '095 patent was the subject of the first worldwide family litigation flag on the Google Patents record (family 32314544), i.e., it was asserted broadly in Hatch-Waxman litigation, yet the generics chose to petition only against the '569 and REMS patents. (3) Celgene did not have to defend any PTAB appeal on this patent because it never lost an institution here; its aggressive Board posture is confined to the '501/'720 appeals context. (4) No defensive aggregator appears in the chain — there is no Unified Patents filing or similar on this family in anything surfaced.

---

Recommended next steps

1. If you are a defendant and a demand letter or complaint cites US 8,648,095, note first that the patent expired on 2023-05-15 (Google Patents "Anticipated expiration"; Orange Book expiration 2023-05-15). There is no live patent right to enjoin. Exposure, if any, is limited to pre-expiration infringement and is bounded by the six-year damages lookback of 35 U.S.C. § 286 — so any demand for forward-looking royalties, ongoing-activity damages, or a permanent injunction is facially overstated for the period after 2023-05-15.
2. There is no PTAB disposition to link or quote. I will not manufacture one: no IPR was filed, so there is no Final Written Decision for this patent on PTAB E2E, https://ptacts.uspto.gov/, and no CAFC opinion. Confirm independently via the PTAB E2E search page and the PTAB API for patent number 8648095 before taking a position in a filing.
3. Clear the one open item: run a targeted PTAB E2E search for the three Dr. Reddy's Laboratories Revlimid IPR petitions dismissed around 2019-02 to confirm none named the '095 patent. Media coverage does not identify the patent numbers. If one did, it was an institution denial, which changes nothing about claim validity or estoppel but is worth knowing before you represent to anyone that the patent has "never been challenged."
4. Because there is no estoppel, you have the full prior-art menu. Build invalidity under §§ 102/103/112 without restriction. The obviousness theories actually deployed against the sibling '569 patent (including the Corral/Kaplan IMiD literature, the '230 patent, and the thalidomide-plus-dexamethasone MM prior art identified in the Alvogen and DRL petitions) are transferable analytical starting points for the combination claims of '095, though the proteasome-inhibitor limitation is an added element your art will need to reach.
5. The absence of PTAB activity is itself a signal, and here it cuts both ways. Well-asserted life-science patents usually attract at least one IPR; that the generic challengers took only the '569 and REMS patents to the Board suggests either that '095 was bundled into settlements before it became the gating patent, or that challengers viewed the proteasome-inhibitor combination claims as comparatively harder to attack. Either way, do not read "no IPRs" as "weak patent" — read it as "no free invalidity judgment, and no estoppel saving you the work."
6. Parallel-track check: if there is current litigation, pull the D.N.J. docket set (2:16-cv-07704; 2:17-cv-02528; 2:17-cv-06163; 2:17-cv-06842; 2:18-cv-11518; 2:18-cv-11630 and the later 2020–2022 cases on the Google Patents litigation list) for admissions, claim constructions, and expert reports touching '095, and check CourtListener, https://www.courtlistener.com/, for any appealable judgment referencing the patent.

Confidence note: I am highly confident there is no AIA trial instituted against US 8,648,095. I am moderately confident there is no petition at all, based on the ODP block plus negative search results, with the residual uncertainty being the unverified Dr. Reddy's petition numbers noted in step 3.

I'll research the assignment record and litigation history before reconstructing the chain.

Good — I've found Celgene's assignment-recording pattern for sibling patents. Let me pin down the specific reel/frame for the '095 family and confirm no post-Celgene transfers.

US 8,648,095 B2 — Ownership & Assignment Reconstruction

Patent: US 8,648,095 B2 ("Methods for treating multiple myeloma using 3-(4-amino-1-oxo-1,3-dihydroisoindol-2-yl)-piperidine-2,6-dione in combination with proteasome inhibitor")
App. No.: 13/488,888 · Filed: 2012-06-05 · Issued: 2014-02-11 · Earliest priority: 2002-05-17 (prov. 60/380,842; also 60/424,600, 2002-11-06)
Continuity: 13/488,888 is a continuation of 12/640,702 (now US 8,198,306), itself a continuation of 10/438,213 (now US 7,968,569).

Sourcing caveat (read first): I was able to read the Google Patents legal‑events record and cross‑reference litigation/SEC material, but I could not complete a live query of the USPTO Assignment Center for this patent within this session. Where I give a reel/frame, it is a record for a sibling patent in the same Celgene–Zeldis assignment pattern and is labeled as such. I do not attribute a reel/frame to the '095 itself that I did not actually observe. Nothing below is invented; gaps are marked "not retrieved."

---

Inventors

Unusual‑pattern check — negative. The classic red flag ("all inventors leave the original assignee within 12 months of filing, prefiguring a fire‑sale") is not present. There is one inventor, he was a senior Celgene employee when the 2012 continuation was filed, and he remained at Celgene for years afterward. The related names appearing in this family (e.g., Bruce A. Williams) are Celgene employees on the sibling distribution/distribution‑method patents, not departing assignors here.

Structural note worth flagging: the '095 is a late continuation — filed 2012 off a May 2002 priority, ~10 years after the earliest priority — issued in 2014, and then asserted against generics. That is life‑cycle‑management "continuation practice" by an operating company; it is a prosecution/portfolio signal, not an ownership signal.

---

Original assignee

• Celgene Corporation, 86 Morris Avenue, Summit, NJ 07901 (Delaware corporation). Named as assignee on the face of the printed patent (Summit, NJ); Google Patents lists the current assignee as "Celgene Corp."
• Product embodying the claims: Yes. Celgene marketed REVLIMID® (lenalidomide), the compound recited in the claims. The '095 claims a method of treating multiple myeloma with lenalidomide in combination with a proteasome inhibitor. The patent was carried in the FDA Orange Book against lenalidomide capsules (patent code U‑1984; listed expiration 2023‑05‑15 per the Orange Book 41st Annual edition), i.e., Celgene itself treated this as a listed, product‑covering method patent.
• Primary line of business: branded biopharmaceuticals (hematology/oncology; Revlimid, Pomalyst, Thalomid, Otezla at the time).
• Current status: Acquired — not dissolved, not in bankruptcy. Bristol‑Myers Squibb completed the $74B acquisition of Celgene on 2019‑11‑20; Celgene "becomes a fully controlled subsidiary" of BMS (BMS press release, 2019‑11‑20). Celgene's shares ceased trading at close that day.
• Patent status: Google Patents shows "Expired – Fee Related," with an anticipated expiration of 2023‑05‑15.

---

Assignment timeline

Records found for the '095 itself: one (1) recorded transfer — inventor → Celgene. No post‑issuance assignments surfaced.

• 2012‑06‑05 (executed) / recorded 2012‑06‑05 — Reel/Frame not retrieved (Google Patents legal events records the event but not the reel/frame; the Assignment Center query was not completed in this session)
  • Conveyance: Assignment of assignor's interest ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: Zeldis, Jerome B.
  • Assignee: Celgene Corporation, 86 Morris Avenue, Summit, New Jersey 07901
  • Correspondent: not retrieved. (For transparency: the prosecution attorney of record on the sibling '740 patent is Jones Day, per the printed patent — that is prosecution counsel, not the assignment correspondent, and I am not treating it as the recording attorney.)
  • Context: inventor‑to‑employer assignment taken contemporaneously with filing of the continuation; a standing Celgene housekeeping filing, not an acquisition or transfer to a third party.

Sibling‑patent records in the same Celgene–Zeldis recording pattern (context only — these are NOT '095 records):

• 2003‑09‑03 (executed) / 2008‑08‑19 recorded — Reel 021461/0407 (3 pages)
  • Conveyance: Assignment of assignor's interest
  • Assignor: Zeldis, Jerome B. · Assignee: Celgene Corporation, 86 Morris Avenue, Summit, NJ 07901
  • Context: same inventor→employer pattern, recorded against app. 12/229,074 (US 8,198,262). Belated recording ≈5 years after execution — typical of a bulk housekeeping filing, not a transaction.
• Recorded 2011‑03‑24 — Reel 026014/0756
  • Assignor: inventor Zeldis · Assignee: Celgene Corporation
  • Context: cited in Celgene's IPR2018‑01507 mandatory notice as the chain giving Celgene "all right, title, and interest" in sibling US 8,404,717. This is the template for the '095's chain; I did not observe the '095's own reel/frame.

Not found (i.e., a finding in itself):

• No assignment to Bristol‑Myers Squibb recorded against the '095 following the 2019‑11‑20 merger. That is consistent with an all‑stock/cash merger by operation of law in which Celgene survived as a wholly‑owned BMS subsidiary — title did not move to a new record owner, and Celgene therefore remained the record assignee/patent owner of record.
• No transfer to any IP‑holding, licensing, or special‑purpose entity. Zero.

Because the Assignment Center returned only the inventor→Celgene link, I stop the chain there per the protocol; the remaining sections analyze that record.

---

Timeline diagram

timeline
    title Ownership of US 8648095
    2002 : Priority provisional filed by Celgene
    2012 : Continuation application filed
         : Inventor assigns rights to Celgene
    2014 : Patent US 8648095 issues
         : Celgene sues generic ANDA filers
    2019 : BMS acquires Celgene
    2023 : Patent term ends

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
No assignment to an IP/Licensing/Holdings/Ventures entity exists. The only recorded transfer is Zeldis → Celgene Corporation, an operating pharmaceutical manufacturer at 86 Morris Avenue, Summit, NJ — a corporate headquarters, not a registered‑agent drop. No single‑member Delaware/Texas LLC appears anywhere in the chain.

2. Known asserter in the chain — NOT PRESENT.
Assignee is Celgene (now a BMS subsidiary). Celgene does not appear on the Acacia / Marathon / IV / IPNav / Wi‑LAN / Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock lists. To the contrary, the Stanford NPE Litigation Database classifies Celgene as "Patenting/Product Company" with NPE Status "Practicing Entity" (e.g., Celgene Corp. v. Zydus Pharmaceuticals (USA) Inc., 2:17‑cv‑02528, D.N.J.). That is the exact inverse of this signal.

3. Repeat correspondent across the chain — UNCLEAR / NOT RETRIEVABLE.
I could not retrieve the correspondent of record for any of these recordings in this session (neither for the '095 nor for the sibling reel entries above). I will not guess. Note only that Celgene's filings in this family were a single assignor → single assignee pattern repeated many times across the portfolio, which implies a repeat filer — but without the recorded attorney name I cannot call it, and I decline to.

4. Cascading transfers — NOT PRESENT.
Zero consecutive post‑issuance assignments. The chain is exactly one link long (inventor → employer), with no LLC hop, no shared correspondent address issue, and no common‑principals pattern to evaluate.

5. Pre-litigation transfer — NOT PRESENT.
The assignment (2012‑06‑05) is dated ~2 years before the earliest suit naming the '095 (Celgene's 2:14‑cv‑3126 action, 2014 — the patent issued 2014‑02‑11, so it could not have been asserted earlier) and ~9 years after Celgene's first Revlimid ANDA suit (2:10‑cv‑05197, 2010). Nothing here was arranged within 6 months of a filing to manufacture standing or venue.

6. Bankruptcy fire-sale — NOT PRESENT.
Celgene's exit was a $74B strategic acquisition by Bristol‑Myers Squibb (announced 2019‑01‑03; closed 2019‑11‑20), with a CVR and a forced Otezla divestiture to Amgen to clear FTC review. No Chapter 7/11, no patent auction.

7. Privateering — NOT PRESENT.
Celgene litigated the '095 in its own name against its own competitors. The asserting party (Stanford/Lex Machina: "Patent Asserter: Celgene Corporation — Product company") is the same entity that owned and practiced the patent. No NPE asserting on Celgene's behalf appears.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at Celgene/BMS.

---

Verdict

Operating-company assertion.

Celgene — the assignee under the single recorded inventor→employer assignment (Zeldis to Celgene, 2012‑06‑05, assignee address 86 Morris Avenue, Summit, NJ) — is the same entity that manufactures REVLIMID® (lenalidomide), the compound its claims cover, listed the patent in the Orange Book (U‑1984), and enforced it in its own name against actual generic competitors in Hatch‑Waxman ANDA suits in D.N.J. (e.g., 2:14‑cv‑3126; 2:17‑cv‑02528). All eight NPE signals are not present (one is merely unretrievable: correspondent identity), and the chain contains no assignment to any licensing vehicle; the only subsequent event is BMS's 2019‑11‑20 acquisition, in which Celgene survived as a wholly owned operating subsidiary. This is textbook operating‑company assertion, not NPE activity.

Verify at:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ and https://assignment.uspto.gov/patent/index.html (search patent number 8648095)
• Google Patents legal events — https://patents.google.com/patent/US8648095/en#legalEvents
• Stanford NPE Litigation Database — https://npe.law.stanford.edu/case/[188700](/patent/188700) (Celgene as "Practicing Entity")

Open items I did not close: (a) the '095's own assignment reel/frame; (b) the correspondent of record on that recording; (c) whether a merger‑related assignment to BMS was ever recorded against this patent (none surfaced, consistent with merger by operation of law). A direct Assignment Center pull on "8648095" resolves all three in one query.

I'll research the prior-art citations for US 8,648,095 and try to recover the claim text so I can map references to specific claims.

Let me pull the actual cited-references tables and try to recover verbatim claim text.

Scope note before the analysis

I ran USPTO/Google Patents/FreePatentsOnline/Justia queries against US 8,648,095 (application 13/488,888). Two important retrieval limitations:

1. I could not retrieve the examiner's PTO-892 "References Cited" table or the granted claim set verbatim from the USPTO PatentCenter/PAIR in this session. The Google Patents snippet exposes the bibliographic "prior art keywords" and the cited-by table only partially.
2. I did recover the pre-grant publication claim set (US 2012/0244214 A1, the same application 13/488,888, published 2012-09-27). That is probative for the granted claims but not identical to them — the application was amended and an RCE was filed 2013-09-06, with allowance 2013-12-27. Flagging this as a partial correction to the earlier section's caveat #1: I now have publication claim text, not granted claim text.

Everything below is labeled by confidence. I have not invented citations or dates.

---

1. The claim set I can actually work from

From the pre-grant publication for the same application (Justia, https://patents.justia.com/patent/20120244214):

22. A method of treating multiple myeloma, which comprises cyclically administering to a patient having multiple myeloma: (a) about 5 to about 25 mg per day of a compound having the formula [lenalidomide] … or a pharmaceutically acceptable salt or stereoisomer thereof for 21 consecutive days followed by seven consecutive days of rest from administration of said compound in a 28 day cycle, and (b) a therapeutically effective amount of a proteasome inhibitor.

Dependents 23–45 add: MM subtype (newly diagnosed / smoldering / refractory / relapsed / relapsed-refractory), salt/stereoisomer, enantiomerically pure R or S, additional active agent (dexamethasone, melphalan, doxorubicin, vincristine, prednisone, cyclophosphamide, biaxin), autologous SCT, capsule/tablet, 5/10/15/20/25 mg doses, dexamethasone 40 mg dosing schedules, etc.

Claims 1–21 were listed by Justia but their text was not returned in my retrieval. I do not know what claims 1–21 recite, and I do not know the final granted claim 1. Given that published claim 22 carries the structural elements (cyclic 21/7 lenalidomide dosing + proteasome inhibitor + MM indication), it is the most likely lineage ancestor of granted claim 1, but that is an inference, not a verified quote.

This matters to your §102 question: a reference only anticipates if it discloses every element arranged as in the claim. The proteasome-inhibitor element is the discriminating limitation. A reference that teaches lenalidomide for MM but no proteasome inhibitor cannot anticipate claim 22's lineage. That framing drives the analysis below.

---

2. References cited in the specification (applicant/family disclosures)

The authoritative full text supplied for '095 expressly incorporates these by reference. These are the "cited" references I can state with high confidence as appearing in the patent.

* Dates marked "not verified this session" are from my training data. I did not confirm them against USPTO records in this session, so treat them as provisional. The 2001-08-28 issuance of '230 and the 2001-10-05 / 2002-04-12 filing dates are confirmed from text retrieved in this session.

---

3. Family / related patents surfaced in the FPO related-cases listing

From https://www.freepatentsonline.com/[8648095](/patent/8648095).html — I want to be careful about labeling: these appeared in the related-cases block, not necessarily in the examiner's cited-art table. I am reporting them as family/sibling art, not as examiner citations.

---

4. The real §102 vulnerability — and it is not a cited patent

The most substantive §102/§112 question for '095 is priority, not the cited references:

• The asserted priority date is 2002-05-17 (Google Patents; Unified Patents says 2002-05-16 — unresolved, consistent with the earlier section's flagged conflict).
• If the 2002 priority application did not disclose a proteasome inhibitor for MM, the proteasome-inhibitor limitation is entitled only to the later effective filing date, which would expose the claim to intervening art. Conversely, if it did disclose it, then US 2010/0093683 also shares that priority and is not art at all.
• I cannot resolve this without the 2002 priority specification text (USSN 10/411,665 / the provisional it claims), which I did not retrieve.

The second vulnerability is the preamble/efficacy construction: the Celgene v. Lotus claim-construction briefing (D.N.J.) argued that "A method of treating multiple myeloma" imports a requirement of efficacy against MM (courtlistener recap gov.uscourts.njd.348812.251.0). That was in the pomalidomide MM-patent case, but the same intrinsic reasoning applies to the '095 preamble.

---

5. Bottom line on "which claims does each reference anticipate"

No applicant-cited or family reference in my retrieved set appears to be a valid §102 anticipation of the claim 22 lineage. The realistic challenge would be §103 over '569 (cyclic lenalidomide + dexamethasone for MM) in view of a bortezomib-in-combination teaching, plus a priority challenge on the PI limitation.

---

6. Explicit uncertainties (carried forward and updated)

1. Granted claim 1 text is still unverified. I now have published claims 22–45, not granted claims. Published claims 1–21 were not retrievable.
2. The examiner's PTO-892 list is unverified. I could not pull the "List of References cited by applicant and considered by examiner" filed 2014-01-16 (Docket Alarm timeline for 13/488,888). Any examiner-cited art beyond the specification citations is therefore not in this report.
3. Priority-date discrepancy (2002-05-17 vs 2002-05-16) remains unresolved.
4. Dates marked "not verified this session" are from training data and should be checked in USPTO PatentCenter before being relied on.
5. The §102 conclusions are directional, not final — they assume the granted claims retain the proteasome-inhibitor limitation implied by the title and the published claim 22. If granted claim 1 dropped that limitation, the anticipation analysis changes materially and '230/'569 become far more dangerous.

If you can supply the PTO-892 and the granted claim 1 text (or the 2002 priority specification), I can convert items 1, 2, and 5 above from directional to definitive.

I'll research the claim text and the prior-art references cited on the patent's page before analyzing obviousness.

Excellent — I found an IPR reference. Let me pin down the claim text and the prior art actually at issue.

Obviousness Analysis — US 8,648,095 under 35 U.S.C. § 103

0. Scope limits I have to state before the analysis

Two things I could not retrieve, which materially limit what follows:

1. Verbatim claim text. The patent text supplied to me stops inside the detailed description, and my searches did not return the claims of '095. The closest primary evidence I found is a claim-construction brief in D.N.J. that opens a claim with "A method of treating multiple myeloma, which comprises administering to a patient having multiple myeloma, and which patient…" (https://storage.courtlistener.com/recap/gov.uscourts.njd.[348812](/patent/348812)/gov.uscourts.njd.348812.251.0.pdf). I cannot confirm that is '095 claim 1 rather than a sibling patent's claim. Everything below is therefore framed by limitations, not by quoted claim language.

2. The "Prior Art" section of the page. The Google Patents record for US 8,648,095 exposes only "Prior art keywords: administered, compound, multiple myeloma, patients, cancer" — a machine keyword extraction, not a reference list — and the FreePatentsOnline "US Patent References" list is populated almost entirely with same-inventor / same-family Celgene documents (e.g., US 7,968,569; US 8,188,118; US 2012/0035145; US 2010/0093683 — all Zeldis). Those are not usable § 103 art against '095 (pre-AIA § 103(c) disqualifies commonly-owned § 102(e)/(f)/(g) art; and same-inventor documents are not § 102(a)/(b) art at all). A § 103 case built solely from that tab would be defective. So I have reconstructed the actual art-of-record for this family from the prosecution/IPR record, and I flag which references are verified vs. mine.

Also flagging a live contradiction between two fetches of the same page: the supplied text says status "Expired – Fee Related"; a later search snippet of the same URL says "Expired - Lifetime." Unresolved.

---

1. The claim limitations that matter

Reconstructed from the title, the specification, and the parent publication US 2010/0093683 (application 12/640,702, published 2010-04-15), the independent claim appears to require:

Critical observation: the specification text I have contains no data for the L3 combination. It reports only in vitro [³H]-thymidine data comparing REVIMID™ to thalidomide in MM cell lines (FIG. 1). The MM combination disclosure in the spec is a laundry list ("dexamethasone, zoledronic acid, palmitronate, GM-CSF, biaxin, vinblastine, melphalan, busulphan, cyclophosphamide, IFN, … arsenic trioxide, PEG INTRON-A, vincristine"), and no proteasome inhibitor appears in that list. That asymmetry is itself a § 103 vulnerability: the claim reads on a combination the specification never exemplified.

---

2. The threshold issue: effective filing date

The family chain is: 60/380,842 (2002-05-17) → 60/424,600 (2002-11-06) → 10/438,213 (2003-05-15, issued as US 7,968,569) → 12/640,702 → US 2010/0093683 → 13/488,888 → '095 (tree confirmed in the Cigna antitrust complaint, https://business.cch.com/ald/TheCignaGroupvCelgeneCorporationComplaint.pdf). Google Patents says prior art date 2002-05-17; Unified Patents says 2002-05-16.

This is the pivot for § 103, because the art universe changes completely across it:

• If 2002-05-17 / 2002-11-06 controls: the § 103 combination must be built from pre-May-2002 (equivalently pre-Nov-2002) art. Bortezomib's MM data (Richardson, NEJM 2003; the Phase II presentation at ASH December 2002) and the May 2003 VELCADE approval are not available. The case then rests almost entirely on Hideshima 2001 (below) plus IMiD art.
• If a later date controls: the case becomes much easier. Note that in IPR2018-01714 (Alvogen v. Celgene), Petitioner argued the sibling '569 claims were "not entitled to claim priority to at least the '842 provisional" and that the effective filing date was no earlier than November 6, 2002, because the provisional listed the relevant agent "once… among hundreds of other" options (https://gaeflexstaging-dot-docketupdate.appspot.com/cases/[PTAB](/ptab)/IPR2018-01714/Alvogen_Pine_Brook_LLC_v._Celgene_Corporation/09-12-2018-Petitioner/Petition-1-Petition_for_Inter_Partes_Review_of_US_Patent_No_7,968,569/). That is the exact template a '095 challenger would use if the proteasome-inhibitor limitation were first supportable only in a later member of the chain.

I cannot resolve which date the claims get. Both scenarios below are therefore presented.

---

3. The art of record in this family (verified)

The single most useful data point for a § 103 attack is the actual examiner rejections in the parent application 10/438,213, captured in the December 7, 2010 interview summary that Alvogen reproduced as Exhibit 2008 in IPR2018-01714:

"attorneys for Applicant explained that Kyle et al. (Semin. Oncol., 2001), Davies et al. (Blood, 2001) and Corral et al. (Ann. Rheum. Dis., 1999), alone or in combination with other cited references, fail to suggest the claimed methods…" (https://gaeflexstaging-dot-docketupdate.appspot.com/cases/PTAB/IPR2018-01714/Alvogen_Pine_Brook_LLC_v._Celgene_Corporation/12-18-2018-Patent_Owner/Exhibit-2008-Ex_2008_File_History_Excerpt/)

So the Office's own shortlist for lenalidomide for MM was Kyle 2001 + Davies 2001 + Corral 1999, and Celgene overcame it by (a) narrowing to specific doses/dosing and (b) arguing unexpected results, FDA approval, and commercial success.

Additional verified art (from the IPR2018-01714 exhibit list, Exs. 1004–1017): Muller et al., Bioorg. Med. Chem. Lett. 9:1625 (1999); Corral & Kaplan (1999); Singhal et al., NEJM 341:1565 (1999) (thalidomide anti-tumor activity in refractory MM); Hideshima et al., Blood 96:2943-50 (2000) (thalidomide and its analogs overcome drug resistance of human MM cells to conventional therapy); Rajkumar & Kyle, JCO 19(16):3593 (Aug. 2001); Celgene press releases of May 8, 2001 and Aug. 28, 2001.

From my own knowledge, not verified this session (flagging explicitly): Hideshima et al., "The proteasome inhibitor PS-341 inhibits growth, induces apoptosis, and overcomes drug resistance in human multiple myeloma cells," Cancer Res. 61:3071-3076 (2001), and US 5,780,454 (Adams et al., Millennium — boronic-acid/ester proteasome inhibitors, the genus including bortezomib). Verify both before relying on them — they are the load-bearing references for the § 103 case at a 2002 date.

---

4. Ground 1 (primary): the core three-reference combination

Ground 1: '230/'471 (Muller) or Corral & Kaplan 1999, in view of Hideshima 2000 (Blood 96:2943), in view of Hideshima 2001 (Cancer Res 61:3071) [or US 5,780,454].

Mapping:

Motivation to combine, element by element:

1. Same disease, same patient population. Hideshima 2000 and Hideshima 2001 are both directed at MM, and both are framed around the clinical problem that MM becomes refractory to dexamethasone, melphalan and doxorubicin. A POSA treating relapsed/refractory MM in 2002 was already practicing multi-agent regimens (VAD; melphalan/prednisone; thalidomide + dexamethasone). Combining an IMiD with a second agent was the established treatment paradigm, not a departure from it — and the '095 specification itself says so ("cocktails").
2. Different, non-overlapping mechanisms. Lenalidomide's documented activities (TNF-α inhibition, T-cell co-stimulation, anti-angiogenesis) versus proteasome inhibition (stabilizing IκB → blocking NF-κB, inhibiting IL-6 signaling). A POSA combines agents with distinct mechanisms precisely to avoid cross-resistance and to add efficacy. Notably, both mechanisms converge on NF-κB/IL-6, giving an affirmative reason to expect additive or better-than-additive effect.
3. Non-overlapping dose-limiting toxicities. IMiDs: myelosuppression/teratogenicity/thrombosis. Bortezomib: peripheral neuropathy/thrombocytopenia. Combination on the basis of non-overlapping toxicity is a classic, expressly accepted rationale.
4. Both were shown to overcome resistance and to potentiate dexamethasone (Hideshima 2000 for IMiDs; Hideshima 2001 for PS-341, which reports the PS-341 + dexamethasone combination). A POSA whose backbone therapy was dexamethasone had a direct reason to add each, and therefore both.
5. KSR. Where the claim is "administer A in combination with B," A and B are each known to be useful for the same indication, and there is a finite, identified class of proteasome inhibitors with one clear lead (bortezomib), the claim is the paradigmatic "combination of known elements according to known methods, yielding no more than expected results." A challenge pleaded in the alternative would add KSR's "obvious to try" rationale: a small, identified set of proteasome inhibitors, one of which was already in MM trials.

---

5. Ground 2 (2002 date): the same combination framed the way the Office framed it

Ground 2: Kyle 2001 + Davies 2001 + Corral 1999 (the examiner's actual '569 rejections), further in view of Hideshima 2001 or US 5,780,454.

This has the advantage of being the combination the Office already found prima facie obvious for the analogous lenalidomide-for-MM claim, which is powerful evidence of what a POSA would have found obvious. Celgene's rebuttal in the family — that the art "fails to suggest the claimed methods, which require the use of [lenalidomide] and dexamethasone for treating multiple myeloma in specific amounts and specific dosing regimens" — is a rebuttal about dexamethasone and dosing, not about the proteasome-inhibitor combination. If '095 claim 1 adds only "in combination with a proteasome inhibitor" (and no 21/28-day dosing, cf. '569 claim 13), the differentiation Celgene relied on simply is not in the '095 claim, and the examiner's prior combination plus a proteasome-inhibitor reference carries the day.

---

6. Ground 3 (if the effective date is later than 2002)

If the proteasome-inhibitor limitation is entitled only to the 2009 filing date of 12/640,702 (or later), the following become § 102(a)/(b) art and the obviousness case is substantially stronger:

• Richardson et al., NEJM 348:2609-2617 (2003) (VELCADE in relapsed/refractory MM) — bortezomib demonstrated single-agent activity in the exact indication;
• REVLIMID® (lenalidomide) FDA approval, December 2005, for MM (in combination with dexamethasone) — two approved MM agents;
• The 2005–2009 "RVD"-type reports of lenalidomide + bortezomib ± dexamethasone (e.g., Richardson et al., Blood 2010), which would be contemporaneous art against a 2009 date.

At that date, "combine the two approved agents for the same disease" is close to per se obvious, and the only escape for the patentee would be an unexpected-results showing not attributable to additivity (see § 8).

---

7. Why the genus framing of L3 helps the challenger

If L3 reads "a proteasome inhibitor" (generic, as the title suggests: "in combination with proteasome inhibitor," not "bortezomib"), then:

• No selection is required from any disclosed genus; a reference disclosing any proteasome inhibitor for cancer (e.g., the boronic-acid genus) meets L3, and
• Broad genus claims unsupported by data are more easily obvious and more exposed on written description (the spec here provides no combination data at all). A challenger would plead that the specification's silence as to the proteasome-inhibitor combination shows the limitation was a hindsight-stitched genus, not an enabled invention.

---

8. What the patentee's rebuttals would be, and how they hold up

---

9. Procedural posture (relevant to risk)

• I found no IPR against '095. The only verified family IPR is IPR2018-01714, Alvogen Pine Brook LLC v. Celgene Corp., on US 7,968,569 — institution denied 2019-03-14 (Petitioner's Refund Request, Paper 8), with the Board's § 314(a)/inefficiency analysis resting heavily on the parallel D.N.J. Hatch-Waxman cases.
• '095 was asserted in the lenalidomide ANDA wave (Lotus/Alvogen listed '095 among 16 asserted patents per the TIPO disclosure summary), and all identified cases ended in consent judgments/settlements — so no merits ruling on '095 validity exists.
• Practical consequence: the obviousness question above has never been decided by any tribunal, in either direction. The '569 denial does not tell us anything about '095's claims.

---

10. Bottom line

The strongest, cleanest § 103 case is a three-reference combination: (1) lenalidomide's own disclosure and IMiD-for-hematologic-malignancy art (Muller 1999 / Corral & Kaplan 1999 / US 6,281,230 / US 6,316,471 / Kyle 2001 / Davies 2001 / Hideshima 2000), (2) MM-specific art, and (3) a proteasome-inhibitor-for-MM or -for-cancer reference (Hideshima 2001 / US 5,780,454), with motivation supplied by (i) identical indication, (ii) convergent NF-κB/IL-6 mechanisms, (iii) non-overlapping toxicities, (iv) the field's established multi-agent paradigm, and (v) KSR's combination/"obvious to try" rationales. The decisive variable is the effective filing date: at a true 2002-05-17/2002-11-06 date the case depends almost entirely on Hideshima 2001 and the boronic-acid genus patent, whereas at a 2009-type date the case is close to overwhelming.

Explicit uncertainties (do not treat as resolved):

1. Claim 1's verbatim text is unverified; the limitation framework is inferred.
2. Priority date conflicts across sources (2002-05-17 vs. 2002-05-16), and the entitlement of the L3 limitation to that date is untested for '095.
3. Hideshima 2001 (Cancer Res 61:3071) and US 5,780,454 were not verified in this session — they are from my own knowledge and are load-bearing; verify before use.
4. The Google Patents "Prior Art" tab is keyword-based and its cited-document list is dominated by non-art family members; it should not be used as the § 103 reference set.
5. Status discrepancy: "Expired – Fee Related" (page text) vs. "Expired - Lifetime" (search snippet) for the same URL.
6. No IPR, no CAFC appeal, and no merits validity judgment on '095 were located — a negative finding, not proof of absence.