US 7977357

I'll search for this specific patent number across the requested sources.

Let me get the remaining claim text and check for any 2026 CAFC activity.

US 7,977,357 B2 — Analyst Summary

Scope note / tool limitation up front: I could run targeted web searches but could not directly query the USPTO Patent Center, USPTO full-text search, or the CAFC docket system. Everything below is grounded in the fetched Google Patents record, Espacenet's claim view, and litigation filings retrieved via search. I flag every point where sources disagree or where I lack an authoritative record. I did not find any CAFC docket involving patent 7,977,357 in 2026 (see "Litigation" below) — treat that as a negative search result, not a certified clearance.

1. Bibliographic data

Note the date tension: the patent's stated expiration (2025-01-08) is well before the litigation narrative's frequently cited "April 2027" crystal‑patent expiry, which the court filings attribute to the '800 patent. The '357 and '800 are related but distinct, and their terms are not identical.

2. Abstract (verbatim)

"Polymorphic forms of 3-(4-amino-1-oxo-1,3 dihydro-isoindol-2-yl)-piperidine-2,6-dione are disclosed. Compositions comprising the polymorphic forms, methods of making the polymorphic forms and methods of their use are also disclosed."

3. Independent claims — plain language

Source caveat: the authoritative full text supplied to me is truncated mid‑description (it ends in §5.2.2 on Form B) and therefore does not contain the claims. The claim text below is taken from Espacenet's claim view of US7977357B2, which is a secondary mirror. Claim 14 was cut off in the source I retrieved and I cannot state its full text with confidence.

The '357 has 14 claims. Based on the claim set as retrieved, there are two independent claims:

Claim 1 — the core chemical-entity claim.
The unsolvated (anhydrous) crystalline Form A of 3‑(4‑amino‑1‑oxo‑1,3‑dihydro‑isoindol‑2‑yl)‑piperidine‑2,6‑dione (lenalidomide), characterized by a differential scanning calorimetry (DSC) thermogram having an endotherm at approximately 270 °C.
Plain language: You get the claim by having the particular anhydrous crystal form of lenalidomide that shows the ~270 °C melting endotherm — nothing more is required of you; it is a composition‑of‑matter claim to a specific solid form.

Claim 2 — the "picture" claim.
Unsolvated crystalline Form A of lenalidomide corresponding to the representative X‑ray powder diffraction (XRPD) pattern provided in FIG. 1.
Plain language: Same material, but defined by reference to the drawn XRPD diffractogram in the patent rather than by a DSC number. This is the classic polymorph claim format that made the patent vulnerable to indefiniteness/claim‑scope attacks (see §4).

Claims 3–14 are all dependent, and (per the claim list I retrieved) they are all still directed to unsolvated crystalline Form A — they narrow it by added characterizations rather than claiming different forms or methods:

• 3 — XRPD peaks at ~8, 14.5, and 16 degrees 2θ (a numeric alternative to claim 2's figure reference).
• 4 — adds XRPD peaks at ~17.5, 20.5, 24 and 26 degrees 2θ.
• 5 — corresponding to the DSC thermogram of FIG. 4.
• 6 — TGA curve indicative of an unsolvated material.
• 7 — corresponding to the TGA curve of FIG. 4.
• 8 — corresponding to the IR spectrum of FIG. 2.
• 9 — corresponding to the Raman spectrum of FIG. 3.
• 10 — does not exhibit a significant weight gain from 5% to 95% relative humidity.
• 11 — corresponding to the moisture sorption/desorption isotherm of FIG. 5.
• 12 — corresponding to the solubility curve of FIG. 51.
• 13 — corresponding to the Form A intrinsic dissolution curves of FIGS. 53 and 54.
• 14 — begins "The unsolvated crystalline 3-(4-amino-1-oxo-1,3 dihydro-isoindol-2-yl)-piperidine-2,6-dione of claim 3 or…" — text incomplete in my source; content uncertain.

Important structural point: although the specification discloses eight polymorphs (Forms A–H), the issued claims of the '357 are limited to Form A. The '357 is therefore the "Form A" member of the Celgene polymorph family, alongside the '800 (Form B hemihydrate), '219 and '598 patents. This is stated consistently across the D.N.J. claim-construction briefs ("Form A ('357 patent, claims 1-14)"). I did not find any composition or method-of-treatment claim in the '357 claim set, notwithstanding the abstract's broader language about compositions and methods.

4. Litigation and validity posture (context, not legal advice)

• Asserted in multiple D.N.J. actions against generics: 2:10‑cv‑05197, 2:12‑cv‑04571 (Natco/Watson/Arrow), 2:18‑cv‑08519, 2:18‑cv‑08964, 2:18‑cv‑11518 (Lotus/Alvogen/Natco), 2:19‑cv‑10099, 2:19‑cv‑13994, 2:19‑cv‑14731, 2:20‑cv‑08570, 2:20‑cv‑14389 (Hetero), 2:21‑cv‑00624, 2:21‑cv‑10398, 2:21‑cv‑20459.
• Markman: the '357 was construed together with the '598 and '219; the dispute was whether "Form A" imports all of the specification's assigned characteristics (Natco's position) or is merely "a polymorphic form … distinguishable from other forms" (Celgene's position).
• Antitrust/validity allegations in In re Revlimid and Thalomid Purchaser Antitrust Litigation allege the '357 (with '598 and '219) is invalid as anticipated by Celgene's own U.S. 5,635,517 ('517) patent — mirroring the EPO's 2015 revocation of EP '682, where the EPO found Form A inevitable from the '517 Example 1. These are plaintiffs' allegations, not adjudicated findings.
• Google Patents flags litigation for this family; no Federal Circuit appeal captioned to patent 7,977,357 surfaced in my 2026 docket search.

5. Explicit uncertainties

1. Claims text is from Espacenet, not from the USPTO face copy; claim 14's full text is unknown to me.
2. Filing/grant dates and priority date differ by one day between Google Patents and Unified Patents — I could not resolve to the day.
3. Inventor list is inconsistent between the Google Patents inventor field (3) and the assignment record (5).
4. CAFC 2026: my only relevant hit was an unrelated notice of docketing, Contour IP Holding LLC v. GoPro, Inc., Fed. Cir. No. 2026‑2012 — it does not involve 7,977,357. Absence of a hit is not proof no appeal exists.
5. Expiration (2025-01-08) is Google's machine‑computed adjusted date and is expressly not a legal conclusion.
6. Date discrepancy: your prompt states April 26, 2026; my system date is October 1, 2026. I have not used either to infer post‑2026 events.

I'll research litigation involving US patent 7977357 (the Celgene lenalidomide polymorph patent). Let me run several targeted searches.

I have good initial results. Let me dig into the specific case numbers and details.

I'm building a picture. Let me identify the parties in the additional case numbers listed on the patent record.

Litigation Involving U.S. Patent No. 7,977,357

Important framing note: U.S. 7,977,357 ("Polymorphic forms of 3-(4-amino-1-oxo-1,3 dihydro-isoindol-2-yl)-piperidine-2,6-dione") is a Celgene polymorph ("crystal form") patent covering the unsolvated Form A crystal form of lenalidomide (brand REVLIMID®). It issued July 12, 2011 and is recorded as expired – lifetime (adjusted expiration 2025-01-08).

Two important accuracy caveats before the list:

1. Database disagreement. The Google Patents record for US 7,977,357 carries a "Family has litigation" flag listing 13 D.N.J. case numbers. Because that flag is family-level, it aggregates suits touching any member of the patent family — not necessarily suits in which the '357 patent itself was asserted. By contrast, the Stanford NPE Litigation Database (which maps litigation per patent) lists only 2 cases for patent 7,977,357 specifically. I therefore separate "confirmed asserted" from "family-flagged/unconfirmed" below.
2. I exhausted my search budget before I could independently confirm the parties for every one of the 13 family-flagged case numbers. I flag those I could not verify rather than guess.

A. Suits where U.S. 7,977,357 was confirmed asserted

Note on row 1: The '357 patent issued July 12, 2011, so it could not have been in the original Oct. 8, 2010 Natco complaint (2:10-cv-05197); it entered the Natco litigation later (the May 27, 2014 Markman opinion construed "Form A" in the '357 and '598 patents' claims). The Stanford NPE database nonetheless records 2:12-cv-04571 as a case involving this patent.

Note on row 4: The IIPRD summary confirms the three patents in the Sun settlement but does not give the docket number. Candidate D.N.J. docket numbers in the patent family record that I could not definitively tie to Sun are 2:18-cv-08519, 2:19-cv-13994 and 2:21-cv-10398.

B. Family-flagged litigation listed on the US 7,977,357 record (assertion of '357 itself unconfirmed in my search)

The Google Patents record for this patent lists the following additional D.N.J. actions as family litigation. I could not independently verify, within my search budget, whether the '357 patent itself was asserted in each:

• 2:10-cv-05197 — Celgene Corp. v. Natco Pharma Ltd. (filed Oct. 8, 2010; the foundational REVLIMID ANDA suit; later-filed related actions and amendments)
• 2:18-cv-08519 — parties not confirmed
• 2:18-cv-08964 — Celgene v. Cipla (see §A)
• 2:19-cv-10099 — parties not confirmed
• 2:19-cv-13994 — parties not confirmed
• 2:19-cv-14731 — Celgene Corp. v. Cipla Ltd. (filed July 3, 2019; consolidated 2:17-cv-06163 and 2:18-cv-08964; settled Dec. 2020)
• 2:20-cv-08570 — parties not confirmed
• 2:20-cv-14389 — parties not confirmed
• 2:21-cv-00624 — parties not confirmed
• 2:21-cv-10398 — parties not confirmed
• 2:21-cv-20459 — parties not confirmed

Context for the 2021 wave: Celgene filed 2021 ANDA suits against Hikma Pharmaceuticals USA, Aurobindo Pharma / Aurolife, Torrent Pharmaceuticals, Biocon Pharma, Lupin Ltd., and Alembic Pharmaceuticals — some of the unconfirmed 2:21 docket numbers above likely correspond to these defendants. I did not have enough search budget to map them individually, and I am not asserting a match.

C. Antitrust / "sham litigation" cases referencing the '357 patent

U.S. 7,977,357 features prominently as one of the "Form A crystal patents" at issue in the REVLIMID/THALOMID antitrust litigation, not as a direct infringement claim:

• In re Revlimid & Thalomid Purchaser Antitrust Litigation, D.N.J., 2:19-cv-07532 (and predecessor 2:14-cv-06997) — purchasers/insurers allege Celgene's serial ANDA suits (including those asserting the '357, '219, '598 Form A crystal patents) were sham litigation and that settlements were reverse payments. The court's June 6, 2024 opinion (Dkt. 446) addressed sham-litigation allegations.
• Cigna Corporation v. Celgene Corporation, D.N.J., 2:21-cv-11686 (direct-purchaser opt-out).
• MSP Recovery Claims, Series LLC v. Celgene Corporation, D.N.J., 2:21-cv-20451.
• More recent BMS/Celgene antitrust complaints (2024–2026) also describe the '357 patent in the context of the alleged REVLIMID monopoly scheme (e.g., the New Jersey antitrust complaint describing the "'800 family," crystal patents, and the Form A patents '219/'598/'357).

D. Key claim-construction history (relevant to the '357 patent)

• May 27, 2014 Markman opinion (Judge Susan Wigenton, D.N.J., in the Natco litigation): construed "Form A" as "the lenalidomide crystal form described in the specification as Form A, having all of the characteristics assigned to Form A in the specification" (applies to all claims of the '357 patent and claims 1–4 of the '598 patent), and construed "hemihydrate" as containing approximately half a mole of water per mole of compound.

E. What I could not confirm

• I found no Federal Circuit appeal specifically arising from a judgment on the '357 patent.
• I could not confirm the parties/outcomes for family-flagged dockets 2:18-cv-08519, 2:19-cv-10099, 2:19-cv-13994, 2:20-cv-08570, 2:20-cv-14389, 2:21-cv-00624, 2:21-cv-10398, 2:21-cv-20459. These should be verified against PACER/CourtListener before being cited.
• The Sun Pharma docket number is unconfirmed.

Primary sources consulted: Google Patents US7977357B2 (family-litigation list, status); Stanford NPE Litigation Database (patent/7977357-0); CourtListener dockets 2:18-cv-11518 (Dkt. 40 consent judgment) and 2:10-cv-05197 (Dkt. 366); Justia D.N.J. opinions; IIPRD summary of the Sun settlement; In re Revlimid & Thalomid antitrust complaint filings.

If you need the unconfirmed dockets pinned down, I can run additional targeted PACER/CourtListener lookups per case number.

I'll start by verifying the structured ODP signal against public records.

Let me run targeted searches for any AIA trial activity tied specifically to the '357 patent and its family.

Let me check a few more angles, including the '800 family and any appeals.

PTAB / AIA Trial Analysis — US 7,977,357 ("Polymorphs of lenalidomide," Form A)

Source of the proceedings list: the structured "PTAB proceedings on file" block supplied in this prompt (USPTO Open Data Portal ingest) states no AIA trial proceedings for US 7,977,357. I ran independent web searches to try to falsify that. They turned up nothing — no IPR, PGR, or CBM petition, institution decision, FWD, or settlement on this patent appears in any public source I retrieved. That corroborates the ODP default. Note two limits on my verification, stated up front rather than papered over: (1) I was unable to query PTAB E2E / PTAB APIs directly in this session, and (2) my searches were general-web, not an exhaustive docket crawl. Treat "zero proceedings" as the strongest reading of the evidence available, not as a certified docket pull.

Proceedings overview

Total AIA trial proceedings on US 7,977,357: 0. Breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied.

Bottom-line defensive posture: This is not "the patent survived two IPRs and is hardened" — it is the opposite signal: the '357 patent was never tested at the PTAB at all. Its validity was challenged only in district court and at the EPO (where the counterpart EP '682 was revoked), and Celgene/BMS repeatedly settled rather than litigate the polymorph patents to judgment. The absence of any IPR is itself meaningful — eleven generic ANDA filers were sued on this patent, several of them sophisticated serial IPR filers, and none filed an AIA petition against it. The most likely explanations are timing (the patent expired 2025-01-08, and Hatch-Waxman defendants had strong non-infringement positions after the 2014 Markman ruling, so no one needed the PTAB) and the fact that the '357 was not listed in the Orange Book for Revlimid.

For anyone facing an assertion of the '357 today (2026-10-01), the dispositive facts are not PTAB facts at all: the patent term has run (adjusted expiration 2025-01-08, per the patent's Google Patents legal-status entry and the expiry table in the Revlimid antitrust complaints).

Per-proceeding detail

There are no proceedings to document. Rather than fabricate sections for proceedings that do not exist, here is the adjacent record a defendant actually needs.

N/A — no IPR / PGR / CBM on US 7,977,357

• Type: —
• Filed: —
• Status: No AIA trial proceeding on file (USPTO ODP); verified against public sources 2026-10-01.
• Judge panel: —
• Petition grounds: —
• Institution decision: —
• Final Written Decision: —
• Settlement / termination: —
• Appeal: No PTAB-derived Federal Circuit appeal exists because there was no PTAB trial.
• Defensive value: You cannot build an IPR strategy around a proceeding that was never filed — but you also face no § 315(e) estoppel, because no petitioner ever triggered it. More importantly, IPR is now largely moot: the patent expired 2025-01-08.

Closest PTAB activity in the lenalidomide family — and why it does not involve this patent

The only PTAB outcomes I could identify in the Revlimid patent estate concern the REMS / restricted-distribution patents (the '501 and '720 patents), which the PTAB invalidated on 2018-10-26 according to the consolidated antitrust complaints in In re Revlimid & Thalomid Purchaser Antitrust Litigation (D.N.J. 2:19-cv-07532). Those are a different patent family from the '800 polymorph family (the '357's parent application 10/934,863); do not conflate them. I am relaying this from a complaint allegation and have not independently pulled the FWD.

The challenges that did happen — district court and EPO (not PTAB)

1. D.N.J. Markman, Celgene Corp. v. Natco Pharma Ltd., No. 10-5197 (D.N.J. 2014-05-27) — the court rejected Celgene's construction of "Form A" and adopted Natco's, construing "Form A" as the lenalidomide crystal form described in the specification as Form A, having all of the characteristics assigned to Form A in the specification. The court expressly held Celgene's proposal ("a polymorphic form … that can be distinguished from other forms") would "give no meaning to the term 'Form A'" and "ignore the specific attributes of Form A as defined in the specification." That is a substantially narrowed claim scope for claims 1–14 of the '357. Opinion text: https://storage.courtlistener.com/recap/gov.uscourts.njd.[247596](/patent/247596).252.0.pdf and https://www.courtlistener.com/docket/14599654/446/in-re-revlimid-thalomid-purchaser-antitrust-litigation/
2. EPO revocation of EP '682 (2015-06-24) — the European counterpart claiming Form A was revoked as anticipated by Celgene's own U.S. '517 compound patent, based on Mylan/Teva testing showing Form A inevitably results from Example 1 of the '517. The antitrust plaintiffs expressly argue the same reasoning invalidates the '357, '598 and '219 (Form A) patents. The '357's claims were never adjudicated on that ground in any U.S. tribunal.
3. Settlements, not judgments — Natco/Watson/Arrow (settled 2015-12), Cipla (2020-12-14), Sun (2021), Alvogen/Lotus, plus DRL/Zydus etc., all resolved confidentially with volume-limited licenses rather than a validity trial.

Strategic summary

Claim status. All 14 claims of the '357 (claims 1–14, directed to unsolvated crystalline Form A; claim 1 requires a DSC endotherm at ~270 °C; claim 3 adds XRPD peaks at ~8, 14.5 and 16 °2θ; claim 2 corresponds to the FIG. 1 XRPD pattern) are UNTESTED at the PTAB — none canceled, none sustained. The only authoritative narrowing is judicial, not administrative: the 2014 Markman construction of "Form A." Meanwhile the entire family's commercial runway is gone — the sibling '219 and '598 expired 2024-09-03, and the '357 expired 2025-01-08. Concretely: if a demand letter today cites claims 1–14 of the '357, it is citing an expired, never-Orange-Book-listed, judicially narrowed patent whose European twin was revoked for anticipation. That is a much weaker posture than "the patent survived the PTAB."

Estoppel landscape. § 315(e)(2) estoppel is a blank slate — because no IPR/PGR was ever instituted, no petitioner or privy is estopped from raising any § 102/§ 103 ground, and no ground is foreclosed by the Board. Practically, the question is inverted: rather than asking which grounds remain after estoppel, ask whether an IPR is worth filing at all. For a patent that expired 2025-01-08, the Board rarely has reason to institute, and § 315(b)'s one-year clock has long since run for every defendant served in the 2010–2019 D.N.J. wave. The live defenses are district-court defenses — inherent anticipation by the '517 as found by the EPO, obviousness of a polymorph of a known compound, and the Markman-driven non-infringement position.

Pattern signals. No repeat petitioner (there are no petitioners). No defensive aggregator in the chain — the Unified Patents portal entry for this patent is a litigation-data page, not an IPR record. Celgene/BMS did not litigate any polymorph patent to a validity judgment; it settled with every generic that pushed hard, which the antitrust plaintiffs characterize as pay-for-delay. The patent owner therefore never had occasion to pursue a PTAB appeal on this patent.

Recommended next steps

• If you are a defendant now: lead with expiry. The '357 reached its adjusted expiration on 2025-01-08; there is no prospective infringement exposure for post-expiry conduct. If the demand targets pre-expiry sales, the § 286 six-year damages lookback window has almost entirely closed, and the non-infringement position flowing from the 2014 Markman opinion (Form A requires all characteristics attributed in the specification, a showing Celgene "did not even attempt to make" per the D.N.J. antitrust complaints) is strong. Cite the Markman opinion directly: https://storage.courtlistener.com/recap/gov.uscourts.njd.247596.252.0.pdf
• On invalidity: the best-developed record is the EPO's 2015-06-24 revocation of EP '682 on inherent-anticipation grounds, replicated by Mylan/Teva testing of Example 1 of US 6,635,517. That argument has never been adjudicated against the '357 in the U.S. — no FWD exists to cite, and no estoppel blocks it.
• On IPR: given expiry, filing is likely a poor use of fees, and for defendants served years ago § 315(b) bars it. Do not represent to a client that an IPR remedy is available here without confirming the service dates against § 315(b) and confirming the Board's current practice on expired patents.
• Verification caveat (do this before relying on this memo): run a direct PTAB E2E / PTAB API docket search on the patent number and on its family members ('800, '217, '219, '598) at https://ptacts.uspto.gov/ptabweb/ , and check the Federal Circuit docket for any appeal from a polymorph-patent decision. My conclusion of zero PTAB activity rests on the ODP structured data plus non-exhaustive web search, and I could not complete those two primary-source checks in this session.

I'll research the assignment record before writing anything. Let me run several targeted searches.

Let me check for any post-issuance assignment activity, especially the BMS–Celgene merger.

US 7,977,357 B2 — Ownership Reconstruction

Patent: Polymorphic forms of 3-(4-amino-1-oxo-1,3 dihydro-isoindol-2-yl)-piperidine-2,6-dione (lenalidomide polymorphs)
App. 12/220,336 · filed 2008-07-23 · issued 2011-07-12 · divisional of 10/934,863 (now US 7,465,800), which claims benefit of provisional 60/499,723 filed 2003-09-04.
Status: Expired – Lifetime; adjusted expiration 2025-01-08.

Sourcing caveat up front: I could not retrieve the USPTO Assignment Center reel/frame numbers, execution dates, or correspondent-of-record for this patent's records from the sources reachable to me. The Assignment Center search page itself did not return content in my searches, and no court filing I found quotes the '357 assignment notice the way filings do for sister patents. Everything below is grounded in what I could verify (Google Patents legal events as reproduced in the authoritative patent text, Unified Patents portal, Stanford NPE Litigation Database, FDA/PTAB/pleading documents). Where a reel/frame would normally go, I say "not retrieved" rather than guess. Verify directly at https://assignmentcenter.uspto.gov/ (search "7977357") and https://patents.google.com/patent/[US7977357B2](/patent/US7977357B2)/en#legal-events.

Inventors

Unusual pattern — assignor/inventor mismatch. The recorded assignment on the Google Patents legal events for this file (entry dated 2008-07-23) lists five assignors: CAMERON, LOUISE M.; CHEN, ROGER SHEN-CHU; JAWORSKY, MARKIAN S.; MULLER, GEORGE W.; and SAINDANE, MANOHAR T. Only three of those five — Chen, Jaworsky, Muller — are named inventors on the '357 itself. Cameron and Saindane appear only as assignors. This is consistent with a single group assignment executed across the lenalidomide polymorph family (the '800 and '217 applications) and recorded against the divisional at filing, rather than a bespoke per-patent assignment. It is not an NPE tell, but it means the '357's reel/frame, when located, will likely be shared with sibling applications.

No evidence that any inventor departed Celgene within 12 months of filing; all three named inventors are career Celgene personnel and no inventor-side transfer appears in the chain.

Original assignee

Celgene Corporation, then of 7 Powder Horn Drive, Warren, New Jersey 07059 (later 86 Morris Avenue, Summit, NJ 07901); a Delaware corporation.

• Line of business: brand biopharmaceutical — oncology/hematology, immunology & inflammation. This patent is part of the lenalidomide estate.
• Did they ship a product embodying the claims? Yes — REVLIMID® (lenalidomide), approved by FDA 2005-12-27 (NDA 21-880) and commercially the largest product in the estate (2017 net product sales cited in the House Oversight staff report at ~$8.19B). One nuance worth recording precisely: the '357 claims are drawn to Form A (unsolvated) — see claim 1 ("The unsolvated crystalline Form A … having an endotherm at approximately 270 °C"). Celgene's own specification states that "Form B is the desired polymorph for the active pharmaceutical ingredient (API)" — i.e., the marketed API is the hemihydrate, Form B. So the commercial product is coextensive with the family's Form B claims (e.g., US 7,465,800) rather than literally with the '357's Form A claims. Relatedly, a 2025 plaintiffs' brief table lists the '357-family patent as "NOT Listed in the Orange Book" (that table renders the number as "7,977,357" — I reproduce it literally without correction).
• Assertion posture: Celgene asserted this patent in its own name in Hatch-Waxman/ANDA suits — per the Stanford NPE Litigation Database the '357 appears in Celgene Corp. v. Natco Pharma Ltd. et al., 2:12-cv-04571 (D.N.J., filed 2012-07-20) and Celgene Corp. v. Lotus Pharmaceutical Co. et al., 2:18-cv-11518 (D.N.J., filed 2018-07-10). Unified Patents classifies the plaintiff as "Entity Type: Operating Company, Industry: Medical". Google Patents' legal-events panel lists ~14 related D.N.J. cases against generic filers (Watson/Natco/Arrow, Apotex, Cipla, Sun, Lupin, Hetero, Zydus/Cadila, Alvogen/Lotus, Hikma, Aurobindo/Aurolife/Eugia, etc.).
• Current status: Acquired. Bristol-Myers Squibb completed its $74B acquisition of Celgene on 2019-11-20; per BMS's closing press release, "Celgene became a wholly owned subsidiary of Bristol-Myers Squibb Company." FDA correspondence still describes "Celgene (a wholly-owned subsidiary of BMS)" as the RLD holder, and PTAB filings continue under "Celgene Corporation" as patent owner — I found no evidence of a recorded assignment of this patent to Bristol-Myers Squibb Company, which is unsurprising for a stock-and-cash merger where Celgene survives as the record owner. Cellgene was not in bankruptcy at any point.

Assignment timeline

One recorded assignment link (inventor → Celgene). No post-issuance transfer of any kind appears in the record I could reach.

• Executed date: not retrieved / recorded 2008-07-23 (the divisional's filing date) — Reel not retrieved / Frame not retrieved
  • Conveyance: Assignment (assignment of assignors' interest; "SEE DOCUMENT FOR DETAILS")
  • Assignor: CAMERON, LOUISE M.; CHEN, ROGER SHEN-CHU; JAWORSKY, MARKIAN S.; MULLER, GEORGE W.; SAINDANE, MANOHAR T.
  • Assignee: CELGENE CORPORATION, 7 Powder Horn Drive, Warren, New Jersey 07059
  • Correspondent: not retrieved — cannot assess the repeat-correspondent signal for this chain (see Signal 3).
  • Context: Original prosecution assignment — inventors/group assignors convey to the operating-company employer; filed contemporaneously with the divisional. No fire-sale, securitization, or asserter transfer.

Everything after 2011-07-12 is litigation, not conveyance. The '357 was sued on repeatedly (2012, 2018, and later Celgene-vs-generic campaigns through 2021) by Celgene itself. There is no assignment to any IP-holding LLC, no license record, no security agreement, and no merger-deed recordation of this patent that I could verify. If Assignment Center shows additional entries, that would be news relative to every secondary source I checked.

Correspondent context (other Celgene patents — NOT this patent; listed only to show Celgene's recordation practice): Celgene's assignment filings in this estate have historically been handled by outside litigation counsel rather than a volume assignment-service firm. Examples surfaced in litigation/PTAB records: U.S. 5,635,517 — assignment recorded Reel 8147, Frame 0954 (cited in Celgene's PTE application); U.S. 6,315,720 — assignment recorded Reel 11469, Frame 0882 (recorded 2001-01-05); U.S. 7,968,569 — assignment recorded Reel 014517, Frame 0556 (recorded 2003-09-15); and a Williams/Kaminski→Celgene assignment recorded 2003-04-23 at Reel 013982/0697. In the '720 IPR the 37 CFR 3.73(b) statement was signed by Francis Dominic Cerrito; Celgene's PTAB papers list Jones Day (250 Vesey Street, New York) as counsel. None of these firms/attorneys is asserted here to be the correspondent on the '357 record. Note the deliberate absence of a high-volume NPE-recordation attorney pattern: Celgene's recordations read like counsel-of-record work for a single operating company, not a shell-LLC conveyor.

Timeline diagram

timeline
    title Ownership of US 7977357
    2003 : Provisional filed by Celgene
    2004 : Parent app filed resulting in 7465800
    2008 : Divisional filed as 12 220336
         : Five assignors convey to Celgene
    2011 : Patent issued to Celgene
    2012 : Celgene sues Natco and Watson
    2018 : Celgene sues Lotus and Alvogen
    2019 : Celgene acquired by Bristol Myers Squibb
    2025 : Patent expires

NPE / troll-pattern signals

Aggravating-but-not-NPE context (for completeness, not a score): the U.S. House Oversight Committee's Sept. 2020 staff report and the Viking/Pomalyst and In re Thalomid/Revlimid antitrust complaints characterize Celgene's lenalidomide portfolio as a "patent fortress" built on Orange Book listings and repeated ANDA suits. That is brand-pharma patent-thicket and regulatory-dispute conduct, which is a different phenomenon from NPE assertion. It does not move any of the eight signals above.

Verdict

Operating-company assertion

Celgene Corporation — the original and only assignee of record, per the inventor-to-Celgene assignment recorded 2008-07-23 (reel/frame not retrieved) — is a brand pharmaceutical manufacturer that commercialized lenalidomide as REVLIMID® and sued actual generic competitors under this patent in its own name (Celgene v. Natco/Watson/Arrow, 2:12-cv-04571, filed 2012-07-20; Celgene v. Lotus/Alvogen, 2:18-cv-11518, filed 2018-07-10), with Unified Patents and the Stanford NPE Litigation Database both classifying the plaintiff as an operating/product company. The chain contains no shell-entity transfer, no cascading assignments, no pre-litigation transfer, no bankruptcy sale, no privateering, and no defensive-aggregator endpoint; the patent remains with Celgene as a Bristol-Myers Squibb wholly owned subsidiary following the 2019-11-20 merger, and no post-issuance assignment of this patent exists in any record I could reach. The only real caveat is doctrinal rather than ownership-related: the '357's claims cover Form A (unsolvated), while the marketed API is Form B (hemihydrate) per Celgene's own specification — so product-coextensiveness is stronger for the family's Form B claims than for this patent's claims standing alone. Verify the reel/frame and correspondent fields directly at the USPTO Assignment Center (https://assignmentcenter.uspto.gov/, search patent 7977357); those were the two data points I could not confirm.

I'll research US 7977357B2 and its cited prior art references.

Let me verify the specific claims of US 7,977,357 B2 and its front-page cited references.

Let me pin down the exact front-page "References Cited" list and dates for US 7,977,357 B2.

Scope, sources, and caveats

Important limitations up front. US 7,977,357 B2 has a very long front‑page "References Cited" list (roughly 100 U.S. patent documents, ~20 foreign documents, and dozens of non‑patent items). I was able to confirm the patent's identity, claims, and the identity/nature of the cited references, but I could not retrieve a single authoritative, complete per‑reference citation-and‑date table in one fetch. Below I give exact data where I verified it, describe the remainder by category, and explicitly flag where I do not know a date. I have not fabricated grant dates.

Also note the operating rule: a patent's own "References Cited" list is not an examiner's rejection. Applying § 102 to each item is my analysis, not a record fact.

1. The patent under review (verified)

Claims actually at issue. Per DrugPatentWatch and the Natco Markman record, claims 1–14 are all directed to the unsolvated crystalline Form A of lenalidomide, e.g.:

• Cl. 1: unsolvated crystalline Form A having a DSC endotherm at ~270 °C;
• Cl. 2: Form A corresponding to the XRPD pattern of FIG. 1;
• Cl. 3–4: XRPD peaks at ~8, 14.5, 16 (and 17.5, 20.5, 24, 26) °2θ;
• Cl. 5–14: Form A characterized by the FIG. 4 DSC/TGA, FIG. 2 IR, FIG. 3 Raman, no significant 5–95 %RH weight gain, and FIG. 51 solubility curve.

The Natco court expressly construed "Form A" for "'357 claims 1 through 14" (Casetext, Celgene v. Natco, D.N.J. May 27, 2014).

Consequence for § 102: for any cited reference to anticipate, it must disclose that specific unsolvated crystal form — expressly or inherently, and enablement-supplying — because the claim is to a crystal form, not to the molecule.

2. Legal framework applied

Priority predates the AIA (2003–2004; U.S. filing 2008), so pre‑AIA 35 U.S.C. § 102(a)/(b)/(e) governs. Pre‑AIA § 102(b) critical date ≈ Sep 3–4, 2003 (one year before the earliest U.S. filing/priority). Only '517 (1997) and '230 (2001)-type documents clearly fall before that date.

3. Tier 1 — References that disclose the compound (only realistic § 102 candidates)

These are the only cited items capable of anticipating claims 1–14, and only under an inherency theory (i.e., if practicing the reference's disclosed synthesis/recrystallization necessarily yields Form A).

Real-world corroboration on #1. The EPO, in the counterpart opposition to EP 1 667 682, revoked the patent on June 24, 2015 on the rationale that Form A was anticipated by the '517 patent — i.e., that "following the teachings of the '517 Patent inevitably leads to Form A of lenalidomide" (Revlimid/Thalomid antitrust complaint, ¶ discussing the EPO decision). That is the strongest documented § 102-type attack on the Form A claims.

Caveat. Whether '517/US '357 is actually anticipated in the U.S. is contested and not finally adjudicated as far as my sources show. I flag this as a potential, litigable § 102 ground, not a settled invalidity finding.

4. Tier 2 — Same-family Celgene patents (NOT § 102 prior art)

US 7,465,800 (the parent), US 7,855,217, US 8,058,443, US 8,143,286, US 8,193,219, US 8,431,598, US 9,365,538, together with publications US 2005/0096351, US 2009/0062343, WO 2005/023192 (publ. Mar 17, 2005) and EP 1 667 682 (publ. Jun 14, 2006).

These share inventorship/priority with the '357 and postdate its Sep 2003 priority; they are common-family members, not anticipatory prior art under § 102(a)/(b). They are relevant only to obviousness-type double patenting exposure.

5. Tier 3 — Methods-of-use and angiogenesis patents (not anticipatory)

• D'Amato family — US 5,593,990; 5,629,327; 5,712,291; 6,071,948; 6,114,335; 6,235,756; 6,420,414; 6,469,045; 7,112,602 (1997–2006, dates not individually verified): disclose methods of treating disease/angiogenesis using thalidomide and analogs. They disclose uses, not polymorphs → no § 102 anticipation of claims 1–14; at most § 103 background.
• Andrulis family — US 5,731,325; 6,141,346; Green US 6,518,298: uses/analogs. Same conclusion.

6. Tier 4 — Formulation / delivery-technology patents (not pertinent to polymorph claims)

US 3,536,809; 3,598,123; 3,845,770; 3,916,899; 4,008,719; 4,810,643; 4,999,291; 5,059,595; 5,073,543; 5,120,548; 5,229,496; 5,354,556; 5,385,901; 5,391,485; 5,393,870; 5,528,823; 5,580,755; 5,591,767; 5,639,476; 5,674,533; 5,733,566.

These are drug-delivery/controlled-release carriers (e.g., Theeuwes osmotic pumps, Zaffaroni transdermal, Oshlack oral dosage forms). They appear in the "Referenced Cited" list only as general formulation background. They do not disclose lenalidomide Form A → no § 102 anticipation of any claim. (Individual grant dates not verified in this fetch.)

7. Tier 5 — Foreign patent documents

WO 97/46526; WO 98/03502; WO 01/70275; WO 01/87307; WO 02/26737; WO 02/59106; WO 02/64083; WO 03/86373; WO 03/097052; WO 2004/103274; EP 1 667 682; JP H10‑53576; JP 2001‑503384; and later WO 2010/… items.

• The compound/use PCTs (WO 98/03502, WO 98/54170, WO 02/26737 etc.) could be § 102(a)/(b) art only on the same inherency theory described in Tier 1.
• EP 1 667 682 and the post‑2008 WO 2010/… documents cannot anticipate the '357 claims (published after the Sep 2003 priority) — they are at most background.
• (Exact publication dates not individually verified here.)

8. Tier 6 — Non-patent literature (potential § 102(b) printed publications)

The NPL list around the '800 family includes general polymorphism references that a § 102/§ 103 attack could invoke:

• Haleblian & McCrone, "Pharmaceutical Applications of Polymorphism," J. Pharm. Sci. 58(8):911–929 (1969) — the classic polymorphism review.
• DiMartino et al., "Preparation and physical characterization of forms II and III of paracetamol," J. Thermal Analysis 48:447–458 (1997) — cited in the spec.
• Knapman, "Polymorphic Predictions," Modern Drug Discovery 3(2):53 (2000).
• Bernstein, Polymorphism in Molecular Crystals (2002); Grant, "Theory and Origin of Polymorphism" (1999); Byrn, Solid-State Chemistry of Drugs, 2nd ed. (1999); Brittain (1999/2002); Caira, "Crystalline Polymorphism of Organic Compounds," Top. Curr. Chem. 198:163 (1998); Vippagunta et al., Adv. Drug Deliv. Rev. 48:3–26 (2001).
• Muller et al. (1996, 1998, 1999) thalidomide-analog medicinal-chemistry papers; Corral et al. (1999).

These are general-state-of-the-art / § 103-obviousness references. None discloses lenalidomide Form A → none anticipates claims 1–14 under § 102.

9. Bottom line

1. Only references that disclose the lenalidomide molecule itself are even capable of anticipating the '357 claims, and only on an inherency theory that practicing their disclosure necessarily yields unsolvated crystalline Form A. The leading candidates are US 5,635,517 and US 6,281,230 (plus WO 98/54170 / WO 98/03502 as foreign counterparts) — potentially against all of claims 1–14.
2. This is corroborated by the EPO's June 24, 2015 revocation of EP 1 667 682, which held Form A anticipated by the '517 patent.
3. Every other cited reference (use patents, formulation patents, foreign documents, and the NPL) fails to disclose Form A and therefore cannot anticipate; they are relevant, if at all, only under § 103 or as technology background.
4. The other members of the '800 family (7,465,800; 7,855,217; 8,058,443; 8,143,286; 8,193,219; 8,431,598; 9,365,538) are not § 102 prior art — they share the '357's priority and inventorship; they bear only on double patenting.

Verification / uncertainty statement

• Verified this session: patent identity, appl./filing/grant/priority dates, legal status, and claims 1–14 (DrugPatentWatch; Google Patents; Casetext Celgene v. Natco; Unified Patents).
• Verified dates: US 5,635,517 (Jun 3, 1997); US 6,281,230 (Aug 28, 2001); WO 98/54170 (Dec 1998); WO 2005/023192 (Mar 17, 2005); EP 1 667 682 (Jun 14, 2006).
• Not individually verified: grant/publication dates for the ~90 remaining cited U.S. patents and most foreign/NPL entries; I have marked these rather than guess.
• I could not obtain a single authoritative front‑page reproduction of the '357's "References Cited" list in this session; the list compiled above is drawn from the shared '800‑family citation record and should be confirmed against the '357 front page before being relied on for a filing.

I'll ground this analysis by pulling the actual claim language and the prior-art references cited on the face of this patent.

Let me pull the front-page cited references and full claim set directly.

Let me attempt to retrieve the front-page (56) "References Cited" list for the '357 patent specifically.

Obviousness Analysis — U.S. Patent No. 7,977,357 under 35 U.S.C. § 103

Date of analysis: April 26, 2026 · Patent status per source page: Expired – Lifetime (adjusted expiration 2025‑01‑08) · Assignee: Celgene Corp.

1. What the "Prior Art" section of this page actually supplies

A threshold caveat, stated expressly rather than papered over: the Google Patents page supplied to me contains a "Prior Art" block consisting of (a) Prior art date: 2003‑09‑04, and (b) prior‑art keywords: dione, piperidine, dihydro, oxo, amino. It does not reproduce the front‑page (56) "References Cited" list, and my searches this session did not retrieve that list. I am therefore not asserting that I know the examiner's cited art. What I can ground is the art that the specification itself identifies as prior art in its Background section and in the text — and that is a legitimate "prior art section" of the document.

References on the face of the specification (Section 2, "Background of the Invention"):

The specification also contains a set of admissions that function as prior‑art evidence against its own claims:

• "Polymorphs of a molecule can be obtained by a number of methods known in the art. Such methods include, but are not limited to, melt recrystallization, melt cooling, solvent recrystallization, desolvation, rapid evaporation, rapid cooling, slow cooling, vapor diffusion and sublimation."
• "Polymorphs can be detected, identified, classified and characterized using well‑known techniques such as… DSC, TGA, XRPD, single crystal X‑ray diffractometry, vibrational spectroscopy… IR… Raman…"
• "Form A can be obtained from various solvents, including, but not limited to 1‑butanol, butyl acetate, ethanol, ethyl acetate, methanol, methyl ethyl ketone, and THF." (col. 6)
• The API "has been used in the formulation of API into drug product… Three batches were produced as apparent mixtures of polymorphs in the non‑micronized API."

The title as rendered on the page, "3‑(4‑amino‑1‑oxo‑1, 3 dihydro‑isoindo1‑2‑yl)‑piperidine‑2,6‑dione," contains an artifact (digit "1" for letter "l"); per your strict rule I flag it rather than silently correcting it. The specification body uniformly renders the ring as "isoindol," and the parties in the related litigation agree the compound is lenalidomide.

2. What the claims cover

I could not retrieve the full verbatim claim set for claims 1–14. From the claim‑construction record in Celgene Corp. v. Natco Pharma Ltd. (D.N.J., Judge Wigenton) and Natco's briefs, the following is well‑supported:

• Every asserted claim of the '357 patent contains the term "Form A." The asserted claims are claims 1–14.
• Claim 1: "unsolvated crystalline Form A of 3‑(4‑amino‑1‑oxo‑1,3 dihydro‑isoindol‑2‑yl)‑piperidine‑2,6‑dione, which has a differential scanning calorimetry thermogram having an endotherm at approximately 270 °C."
• Further claims recite an XRPD pattern "comprising peaks at approximately 8, 14.5, 16, 17.5, 20.5, 24, and 26 degrees 2θ"; combinations of the ~270 °C DSC endotherm with XRPD peaks at ~8, 14.5, and 16 °2θ and "a thermogravimetric curve indicative of an unsolvated material."
• The court characterized the '357, '219, and '598 patents collectively as claiming "unsolvated crystal forms of lenalidomide and pharmaceutical compositions containing those forms."
• The court construed "Form A" as "the lenalidomide crystal form described in the specification as Form A, having all of the characteristics assigned to Form A in the specification" (XRPD per Fig. 1, IR/Raman per Figs. 2–3, TGA/DSC per Fig. 4, moisture sorption per Fig. 5).

(Sources: https://storage.courtlistener.com/recap/gov.uscourts.njd.[247596](/patent/247596).252.0.pdf ; http://g.casetext.com/case/celgene-corp-v-natco-pharma-ltd ; https://storage.courtlistener.com/recap/gov.uscourts.njd.247596.249.0.pdf)

Critical framing consequence: every claim is a solid‑state characterization of a molecule that the specification admits was already known. No claim recites a new molecular structure, a new therapeutic use, or a new synthetic route. The entire inventive contribution asserted is the discovery that this known compound crystallizes in a particular unsolvated habit.

3. The § 103 framework and the decisive factual issue

Under Graham v. John Deere Co., 383 U.S. 1 (1966), and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), the inquiry is the scope/content of the prior art, the differences from the claims, PHOSITA level, and secondary considerations.

The pivotal factual question is whether the prior art discloses or inherently produces a crystalline solid form of lenalidomide at all. If it does, the differences collapse to "which of a small number of known solid‑state habits." If it does not, the difference is "the existence of any crystalline form," and the case turns on whether routine polymorph screening was obvious to try with a reasonable expectation of success.

The specification's own admissions cut against the patent here. The applicant states the prior API was "an apparent mixture of polymorphs." A POSA reading the '230 patent (which discloses the compound and its formulation into dosage forms) would necessarily understand that the synthesized, isolated, and formulated material was a solid — and solids of organic small molecules in pharmaceutically relevant purity are presumptively crystalline unless shown otherwise. Under In re Best, 562 F.2d 1252 (CCPA 1977), where the prior art discloses a product/process that necessarily yields the same material, the burden shifts to the applicant to show an unexpected difference. The applicant has not done so for Form A against a characterized prior‑art form.

4. Combinations that would render the claims obvious

Combination A — '517/'230 + DiMartino + Knapman ("routine polymorph screening")

Rationale (KSR‑recognized): Known compound + known technique + known problem + predictable result.

1. '517 and/or '230 disclose the exact molecule and its pharmaceutical utility. The claims add no new molecule.
2. DiMartino (1997) and Knapman (2000) — both cited by the applicant — supply the motivation: polymorphs differ in solubility, stability, flowability, compressibility, and bioavailability, and the "discovery of new polymorphs of a drug can provide a variety of advantages." The specification concedes exactly this.
3. The specification concedes that the methods of obtaining polymorphs (solvent recrystallization, desolvation, evaporation, cooling) and the methods of characterizing them (XRPD, DSC, TGA, IR, Raman) were all "known in the art."
4. Combining: a POSITA seeking an improved manufacturing form of lenalidomide would dissolve/slurry the known compound in the ordinary pharmaceutically acceptable solvents (the specification itself lists ethanol, ethyl acetate, methanol, MEK, THF, butyl acetate, 1‑butanol for Form A) and isolate the crystalline product by routine evaporation, cooling, or slurry conversion. The '357 claims recite the result of that routine work.

Why the motivation is strong, not hindsight‑biased: the motivation existed before the 2003 priority date, is documented in the two references the applicant itself cites, and is echoed in the specification's own "advantages of polymorphs" paragraph. Under KSR, "a finite number of identified, predictable solutions" — here, the small set of crystal habits a pharmaceutical salt/free base can adopt — supports obviousness when a POSITA has good reason to pursue them.

Combination B — '517/'230 + the specification's own slurry/desolvation teachings ("obvious to try")

The specification states Form A "can convert to Form B in aqueous solvent systems," "can convert to Form C in acetone solvent systems," and that Form A is obtained from non‑aqueous solvents. Whether Form A is obtained by slurrying/solvent screening is a process‑parameter question squarely within In re Aller, 220 F.2d 454 (CCPA 1955) (optimization of a known process using known parameters is obvious absent a showing of criticality/unexpected results). No criticality data are presented for Form A.

Combination C — Chief Judge's own family publication as § 102(b)/§ 103 art if priority fails

The '357 is a divisional of application Ser. No. 10/934,863, filed Sep. 3, 2004 (now U.S. Pat. No. 7,465,800), which claims the 60/499,723 provisional of Sep. 4, 2003. The family was published as US 2005/0096351 A1 and as WO 2005/023192, both of which third parties (e.g., EP 2 688 649 B1; US 2017/0107193 A1) identify as disclosing Forms A, B, C, D, E, F, G and H of lenalidomide, including Form A's characterization.

If the '357's Form A claims are not supported by the Sept. 3, 2004 parent as filed (e.g., formulations of the claim limitations introduced by later amendment), priority fails and the family's own March 2005 publications — published more than one year before the July 23, 2008 filing — become § 102(b) art. Because those publications disclose the same Form A characterization, the result is anticipation, which is a fortiori dispositive, and in the alternative renders the claims obvious. I flag this as a conditional theory: the question whether a divisional's claims are entitled to the parent's filing date is a factual one on which I have no prosecution history in this session, and I state it as a hypothesis, not a conclusion.

Combination D — For the composition / dosage‑form / method claims

'230 discloses pharmaceutical compositions and unit dosage forms of the same compound for the same diseases. Where a compound is known and its utility is known, "[t]he combination of a known compound and a known use of that compound" is obvious. A composition comprising Form A need add only the obvious crystal form to the known formulation — and the specification's own Table 12 formulation (pregelatinized corn starch + magnesium stearate, encapsulated in a Dosator machine) is a conventional capsule fill, not an inventive contribution.

5. Claim chart

6. Secondary considerations — and why they are weak on this record

The patent does assert a stability advantage ("Form A… appears to be the most thermodynamically stable anhydrous polymorph"). For that to defeat obviousness it must be:

1. Unexpected relative to the closest prior art, not merely relative to the other new forms disclosed in the same application. Hoffmann‑La Roche v. Apotex is directly on point: a new polymorph of a known compound is prima facie obvious, and the patentee bears the burden of showing unexpected results. The '357 specification reports no comparative stability data against a prior‑art crystalline form of lenalidomide — only against Forms B–H of the same patent family.
2. Not contradicted elsewhere in the specification. The specification states Form A "can convert to Form B in aqueous solvent systems," "can convert to Form C in acetone solvent systems," and "In water systems and in the presence of Form E, Form A tends to convert to Form E." That is a disclosure of instability in the aqueous environments most relevant to oral dosing, undercutting the "unexpected stability" story.
3. Tied to the claimed subject matter. The commercial success of REVLIMID® is attributable to the '517 compound's therapeutic utility, not to the Form A crystal, and the marketed product is not even Form A — the specification states "Form B is the desired polymorph for the active pharmaceutical ingredient (API)." That is a nexus failure on its face.

7. Where a validity defense could plausibly gain traction

I want to be balanced, because the '357 patent survived a decade of ANDA litigation and was the subject of settlements rather than a final invalidity judgment that I can confirm from this session's materials:

• No express prior‑art disclosure of a crystalline lenalidomide form. If the (56) art cited by the examiner consists only of '517 and '230, Celgene's best answer is that neither reference discloses any crystalline form, and inherency requires proof that the prior process necessarily produces Form A (In re Oelrich, 666 F.2d 578 (CCPA 1982)). An "apparent mixture of polymorphs" is not automatically Form A. That is a genuine factual gap in Combination A/B.
• Keipert/"teaching away" argument: the art does not teach which solvent system yields an unsolvated form. But the specification itself lists a broad, conventional set of non‑aqueous solvents — which is the opposite of teaching away; it is a roadmap.
• Claim construction risk cuts both ways. The court ultimately adopted Natco's narrow construction (Form A = the specific characterized polymorph). Narrow construction helps validity but narrows infringement — which is why every generic eventually settled.

8. Bottom line

The claims of US 7,977,357 recite no new molecule, no new use, and no new manufacturing process. On a § 103 analysis they are prima facie obvious over U.S. Pat. Nos. 5,635,517 and 6,281,230 as combined with the polymorphism‑science references the applicant itself cited (DiMartino 1997; Knapman 2000), because the discovery of a crystalline habit of a known drug substance by "methods known in the art" and its identification by "well‑known techniques" is precisely the kind of finite, predictable, routine investigation that KSR holds unpatentable absent unexpected results — and the only unexpected result asserted (stability) is neither measured against the prior art nor consistent with the specification's own conversion data.

9. Confidence and limitations

• High confidence: the identity and citation of U.S. Pat. Nos. 5,635,517 and 6,281,230, the DiMartino and Knapman citations, the prior‑art date (2003‑09‑04) and keywords on the page, the priority chain (60/499,723 → 10/934,863 → 12/220,336), and the substance of the Form A disclosure (col. 6).
• Moderate confidence: the exact wording of all of claims 1–14 (I have claim 1 and several dependent limitations via the D.N.J. Markman record, not the issued claim set), and the '357's inclusion of composition claims.
• Explicitly unknown to me: the front‑page (56) "References Cited" list for the '357 patent, the examiner's stated reasons for allowance, and any IPR/PGR outcome. I did not retrieve these — I am not asserting that none exist. If you can supply the (56) list, I can re‑run Combinations A–C against the actual cited art rather than the specification's own background section.
• Not legal advice. This is a technical obviousness screen, not an opinion of counsel, and the patent is now expired per the source page, so any remaining significance is retrospective (e.g., damages, antitrust pleadings such as those at https://www.hbsslaw.com/sites/default/files/case-downloads/revlimid/2024-11-05-corrected-amended-complaint-redacted.pdf, which allege that the '357, '598, and '219 patents "are also invalid as anticipated or obvious based on the '517 Patent").