US 12104185

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

Let me get the specific claim text and PTAB/litigation status for this exact patent.

Let me retrieve the exact independent claim language from the patent.

US Patent 12,104,185 B2 — Summary

Source note: The authoritative full patent text (Google Patents, https://patents.google.com/patent/[US12104185](/patent/US12104185)/en) was supplied and I have treated it as controlling over search results. Where search results differ in minor ways, I flag it.

---

1. Bibliographic data

Uncertainty flags: A third-party database (Unified Patents portal) lists the priority date as 2011‑12‑29, application date 2024‑03‑07, grant date 2024‑09‑30, and expiration 2032‑12‑27 — each one day earlier than the patent text. This appears to be a time-zone/reporting artifact; I have used the patent text figures. The chain of continuation/divisional parents recited in the specification is long and partly internally repetitive (the "Related Applications" section restates several of the same relationships); I have reported the operative chain, not every restatement.

2. Abstract

"Modified PH20 hyaluronidase polypeptides, including modified polypeptides that exhibit increased stability and/or increased activity, are provided. Also provided are compositions and formulations and uses thereof."

3. Field / technology

The patent concerns modified human PH20 hyaluronidase (sperm adhesion molecule 1 / SPAM1) polypeptides. PH20 degrades hyaluronan (hyaluronic acid) in the extracellular matrix; because hyaluronan is a principal component of the interstitial barrier, PH20 acts as a "spreading factor" that increases tissue permeability and the dispersion/delivery of co-administered therapeutic agents. The work described generates a large site-saturation mutagenesis library of soluble PH20 (using SEQ ID NO:3, the 447‑residue soluble rHuPH20, as reference numbering) and identifies variants with altered activity, stability (e.g. resistance to phenolic preservatives, elevated temperature, low salt), or inactivity.

---

4. Independent claims — plain-language overview

The patent has claims 1–11, and claim 1 is the only independent claim; claims 2–11 all depend from claim 1. (Note: Halozyme statutorily disclaimed claims 5 and 6 during the PGR, leaving 1–4 and 7–11, per PTAB/party filings.)

Claim 1 (the sole independent claim) — a method of use claim

"A method for increasing delivery of a therapeutic agent to a subject," comprising administering to a subject a pharmaceutical composition comprising:

(a) a therapeutic agent; and

(b) a modified PH20 polypeptide comprising an amino acid sequence, wherein:

• (i) at least 95% of the residues of the modified PH20 sequence are identical to the residues of an amino acid sequence selected from SEQ ID NOs: 3 and 32–66, when aligned to maximize identical residues;
• (ii) the modified PH20 sequence comprises an amino acid modification at a position corresponding to position 320 with reference to SEQ ID NO: 3; and
• (iii) the modification at position 320 is a replacement selected from the group consisting of H, K, R and S.

Plain-language: This is not a claim to the variant protein itself; it is a method claim covering the co-administration of any therapeutic agent with a particular class of modified PH20 hyaluronidase enzymes in order to enhance delivery of that agent. The claimed enzyme must (a) closely match — at least 95% identity — one of 36 specified soluble PH20 sequences, and (b) carry a substitution of histidine, lysine, arginine, or serine for the wild-type aspartic acid at position 320 (i.e., D320H, D320K, D320R, or D320S). In practice the 95% identity floor corresponds to roughly 16–22 additional tolerated substitutions within those sequences (a point the PGR petitioner emphasized).

Dependent claims (no additional independent claims)

Bottom line: The entire enforceable scope of this patent rests on one independent method claim requiring (i) ≥95% identity to certain soluble PH20 sequences and (ii) a D320H/K/R/S substitution, used together with a therapeutic agent to increase its delivery.

---

5. Post-grant and litigation posture (as of the search date)

PTAB — PGR2025-00039 (Merck Sharp & Dohme LLC v. Halozyme, Inc.)

• Petition filed March 28, 2025; instituted October 2, 2025; status Pending – Instituted per Unified Patents' PTAB case list. Google Patents' family page likewise lists "PTAB case PGR2025-00039 filed (Pending - Instituted)."
• Petitioner: Merck Sharp & Dohme LLC (the "Unified Patents PTAB Data" label on the Google Patents page is the data source attribution, not the petitioner).
• Challenged claims 1–11 on § 112(a) written description and enablement (broad genus of modified PH20 polypeptides) and obviousness. Per reporting on the institution decisions, the Board agreed the evidence favored the petitioner on the § 112 grounds but found the petitioner had not shown a likelihood of prevailing on obviousness (Venable BiologicsHQ, Oct. 10, 2025, https://biologicshq.com/ptab-grants-institution-of-three-pgrs-related-to-mercks-keytruda-qlex/).
• Halozyme statutorily disclaimed claims 5 and 6 of the '185 patent before institution.
• The '185 patent is one of 14 Halozyme patents Mercko challenged in PGRs filed Nov. 2024–Jun. 2025.

District court — Halozyme, Inc. v. Merck Sharp & Dohme Corp., No. 2:25-cv-03179 (D.N.J.)

• Google Patents' family page lists a "US case filed in New Jersey District Court" with the case URL above; PTAB filings identify this as the co-pending litigation in which the '185 patent is asserted.
• Background (from press coverage): Halozyme sued MSD/Merck in April 2024 over subcutaneously administered Keytruda® (pembrolizumab) with berahyaluronidase alfa; Halozyme alleges the modified PH20 in SC Keytruda® infringes its MDASE patents. Halozyme seeks damages and an injunction.

CAFC / Federal Circuit 2026 dockets — not found.
I could not locate any Court of Appeals for the Federal Circuit 2026 docket or opinion involving US 12,104,185. Searches for this patent number in CAFC 2026 materials returned only unrelated decisions (e.g., GoTV Streaming v. Netflix, Linfo IP v. Trustpilot, Constellation Designs v. LG, Arendi v. Oath), none of which involve this patent. Because the PTAB PGR was only instituted in October 2025 and the district court case is in early stages (with a projected trial around 2030 per petitioner briefing), a Federal Circuit appeal at this time would be unusual. I therefore state explicitly: I have no authoritative evidence of a 2026 CAFC appeal for this patent, and cannot confirm one exists.

---

6. What I cannot verify / open items

• I could not retrieve the PTAB's full institution decision text or a final written decision for PGR2025-00039; the outcome of the § 112 challenge is unresolved.
• The precise status of the D.N.J. litigation (motions, claim construction, any stay) beyond the case number and parties is not established by the sources I found.
• No CAFC docket was identified for this patent (see above).
• Some third-party databases report the priority/filing/grant dates one day earlier than the patent text; I have flagged this rather than reconciled it.

I'll search for litigation records on this specific patent number.

Let me confirm the district court case details and check for any other proceedings.

Let me check the current status of both proceedings and look for any additional cases.

Bottom line

US Patent 12,104,185 ("PH20 polypeptide variants, formulations and uses thereof," Halozyme, Inc.) is asserted in one US district court action and is the subject of one USPTO post-grant review. Both are still pending; no final judgment or final written decision on the '185 patent has been reported. Both are between Halozyme and Merck Sharp & Dohme (MSD).

I found no other US litigation naming US 12,104,185 — no additional district court suits, no Hatch-Waxman cases, and no CAFC appeal specific to this patent.

---

1. District court litigation

Nature and scope: The complaint is styled "Complaint for Patent Infringement and Declaratory Judgment of Patent Infringement." Halozyme asserts 15 patents — US 10,865,400; 11,041,149; 11,066,656; 11,952,600; 12,018,298; 12,037,618; 12,049,652; 12,054,758; 12,077,791; 12,091,692; 12,104,185; 12,110,520; 12,152,262; 12,195,773; and 12,264,345 — all described as belonging to a single patent family covering Halozyme's "MDASE™" modified-PH20 technology. The accused product is Merck's SC KEYTRUDA® (subcutaneous pembrolizumab; subsequently approved as KEYTRUDA QLEX™), which contains berahyaluronidase alfa (Alteogen's ALT-B4), a modified PH20. Halozyme seeks damages and an injunction.

Procedural history located:

• 04/24/2025 — Complaint filed (ECF No. 1).
• 05/13/2025 — Merck's counsel advised that "Merck Sharp & Dohme Corp. does not exist" (merged into Merck Sharp & Dohme LLC in spring 2022) and that service was ineffective; parties negotiated a substitution stipulation.
• Stipulation and Order substituting Merck Sharp & Dohme LLC as defendant, acknowledging service, and giving Merck 60 days from entry to respond.
• 07/14/2025 — Merck corporate disclosure statement (ECF No. 19).
• 08/22–08/25/2025 — Pro hac vice admissions (incl. David A. Nelson, Nancy L. Zhang, Zachariah B. Summers, Lauren Martin, Anastasia M. Fernands, Elliot Choi).
• 08/27/2025 — Local Civil Rule 16.1 letter order; Initial Scheduling Conference set for 10/20/2025.
• 10/10–10/14/2025 — Joint request for extension; letter order granting extension (ECF Nos. 32–33).

Sources: Law.com Radar docket card; Justia docket; stipulation text (PTACTS/petition exhibits); complaint as filed. The Google Patents page for US 12,104,185 links the same case via Unified Patents: https://portal.unifiedpatents.com/litigation/New%20Jersey%20District%20Court/case/2%3A25-cv-03179

---

2. PTAB post-grant review

Notes and nuances:

• The Google Patents "litigation" box labels this entry with "Petitioner: 'Unified Patents PTAB Data'" — that is provenance metadata for the data feed, not the identity of the petitioner. Every primary document (petition, declarations, exhibits) names Merck Sharp & Dohme LLC as petitioner and Halozyme Inc. as patent owner.
• Grounds appear directed primarily at written description and enablement (and, in part, obviousness), attacking the breadth of claims covering large genera of modified PH20 polypeptides.
• Halozyme filed a statutory disclaimer of claims 5–6 of the '185 patent, receipt dated June 13, 2025 (PGR exhibit EX2003). This narrows the claims at issue but does not end the proceeding.
• I could not confirm whether a Final Written Decision has issued for PGR2025-00039. The statutory deadline would fall roughly one year from institution (i.e., around October 2026), and the latest secondary-source table I located still listed the proceeding as merely "Instituted 1 Oct 2025."

Sources: Unified Patents PTAB case list; GreyB IPVerse PGR2025-00039; Docket Alarm exhibits; EX2003 disclaimer filing.

---

3. Related proceedings — same technology/family, but NOT US 12,104,185

These are frequently reported alongside the '185 patent but do not involve US 12,104,185:

• Merck's other PGRs against Halozyme's PH20 family: PGR2025-00003 (US 11,952,600), -00004 (US 12,018,298), -00006, -00009 (US 12,123,035), -0017 (US 12,110,520), -0024 (US 12,060,590), -0030 (US 12,054,758), -0033 (US 12,049,652), -0042 (US 12,037,618), plus petitions against US 12,091,692, 12,077,791, 12,195,773 and 12,264,345. All 14 petitions were reported as instituted. Reported Final Written Decisions finding the challenged claims unpatentable include US 11,952,600 (May 12, 2026), US 12,152,262 (May 2026), US 12,018,298 (June 8, 2026) and US 12,110,520 (September 1, 2026). None of these is the '185 patent.
• Germany: Munich Regional Court granted Halozyme a preliminary injunction against sales of MSD's KEYTRUDA SC™ (announced December 4, 2025) — European counterpart patents, not the US '185.
• United Kingdom: Halozyme v. MSD proceedings in the High Court (Patents Court) concerning EP 622 and EP 347 (modified PH20 claims) — e.g., [2026] EWHC 121 (pleading/statement of case on infringement) and [2026] EWHC 1838 (Mayne Pharma disclosure). Again, EP family members rather than US 12,104,185.

---

Caveats

• Date anomaly: My searches returned sources dated after the April 26, 2026 date you gave me (through early September 2026). I have reported them as found rather than discarding them; if the operative date is April 26, 2026, then the May–September 2026 Final Written Decisions on the other patents would be post-dated, but none of them concerns the '185 patent, so this does not change the outcome for your target patent.
• I did not locate any CAFC appeal, any second US district court action, or any ITC investigation specifically involving US 12,104,185.
• I located no litigation in which US 12,104,185 was asserted against any party other than Merck/MSD.
• Docket statuses in D.N.J. move quickly; the PGR status should be verified directly at PTAB (PGR2025-00039) before relying on it, since a Final Written Decision could issue around October 2026.

I'll start by gathering what's publicly known about the PTAB proceeding flagged in the patent record itself.

Good — the patent's own record and web sources confirm an active PGR. Let me drill into the institution decision, panel, and any related filings.

Let me confirm the current procedural posture, the exact panel, and whether any appeal or FWD exists.

PTAB Proceedings — US 12,104,185 ("PH20 polypeptide variants, formulations and uses thereof")

Data-integrity flag before anything else. The "PTAB proceedings on file" block states the USPTO ODP API returns no AIA trial proceedings for US 12,104,185 and that the default should be "no PTAB activity on file." That is wrong. The very Google Patents record hosting that block separately lists PTAB case PGR2025-00039 filed (Pending - Instituted) on this patent's litigation panel, with a live link to the Unified Patents PTAB docket, and a real PGR petition is on file at the Office. This is almost certainly an ODP indexing/ingest artifact (ODP's AIA-proceedings endpoint routinely lags and often keys off the application number rather than the patent number). I am treating PGR2025-00039 as the canonical proceeding and disregarding the "no activity" default, because the primary record contradicts it. Every proceeding identifier below is taken from the record, not invented.

---

Proceedings overview

There is one AIA trial proceeding on US 12,104,185: PGR2025-00039, instituted on 2025-10-01 and still pending (trial instituted; no Final Written Decision on file as of my sources). Breakdown: 1 active/instituted · 0 claims invalidated by the Board · 0 claims sustained by the Board · 0 settled · 0 institution denials. Separately, claims 5 and 6 were canceled by Halozyme's own statutory disclaimer (2025-06-13) — a patent-owner kill, not a Board kill, and one that a defendant should still bank.

Bottom-line defensive posture: This is not a "hardened patent." The Board instituted on § 112(a) written description and enablement — the same genus-over-disclosure theory that has already knocked out the challenged claims of four to five sibling MDASE patents in the same Merck campaign. Halozyme has already abandoned claims 5-6 for free. But two cautions: (i) no FWD has issued yet in -00039, so there is no claim-level cancellation to point at and no estoppel has attached; and (ii) the Board refused institution on obviousness, so § 103 is not on the table from this proceeding. Net: a defendant today has a public, Board-endorsed § 112 attack and a credibility problem for the patent owner — but until the FWD lands (statutory deadline ~2026-10-01), the patent's claims 1-4 and 7-11 remain live and enforceable.

---

PGR2025-00039 — Merck Sharp & Dohme LLC v. Halozyme, Inc.

• Type: Post-Grant Review (35 U.S.C. §§ 321–329) — the only AIA trial on this patent. No IPR, no CBM (the patent issued 2024-10-01, years post-AIA, so CBM is unavailable).
• Filed: 2025-03-28 (well within the § 321(c) nine-month window; the patent issued 2024-10-01).
• Status: Pending - Instituted (Unified Patents PTAB case list and the patent's own Google Patents litigation block) / Trial Instituted (Docket Alarm, ipverse). Plain English: institution granted, trial underway, decision not yet issued. Institution date is reported as 2025-10-01 (Pearce IP; ipverse); Unified Patents lists 2025-10-02 — expect a one-day database discrepancy, not two institutions.
• Judge panel: Reported as APJs Jeffrey N. Fredman, Susan L. C. Mitchell, Cynthia M. Hardman, and Michael A. Valek (Docket Alarm docket; Unified Patents' "Panel Judge: Michael A. Valek" case list includes PGR2025-00039). Hedge: the Board's consolidated 2026-03-03 order covering all fourteen PGRs expressly warns that "the listed judges are paneled on various of the listed cases. This is not a decision by an expanded panel." Confirm the exact three-judge panel for -00039 on PTAB E2E (https://e2e.uspto.gov/).
• Petition grounds — challenged claims 1-11:
  • Ground 1 — § 112(a) written description: the claims cover a genus of modified PH20 polypeptides estimated at 10⁴⁹ to 10⁶⁶ distinct molecules; the specification's examples are singly-substituted and non-representative, and the disclosure is a "research plan."
  • Ground 2 — § 112(a) enablement: no disclosed working example of a multiply-substituted, enzymatically active PH20 that increases delivery; the disclosed method is a prophetic iterative "make-and-test" process (Idenix/Amgen line).
  • Ground 3 — § 103 obviousness: claims 1-11 obvious over Halozyme's own U.S. 7,767,429 (the "'429 patent") in view of Chao (2007, Biochemistry) — specifically the single D320K species. This ground was NOT instituted.
  • Petitioner also had to establish PGR eligibility (i.e., that the claims get no § 120 benefit to the pre-2013-AIAs '731 application because of the § 112(a) defects), which the Board accepted.
• Institution decision: Partially instituted — 2025-10-01. The Board instituted on Grounds 1 and 2 (§ 112(a)) and declined to institute on Ground 3 (§ 103). In one or two sentences: the Board found the evidence better supported Petitioner that the claims encompass a vast genus of modified PH20 polypeptides without representative disclosure, and reasonably interpreted claim 1's "modified PH20 polypeptide" as requiring hyaluronidase activity (rejecting Halozyme's purely-structural construction) — but held Merck had not shown it more likely than not that the claims were obvious, because Merck never articulated why a skilled artisan would substitute at position 320 other than through hindsight. The Board also rejected Halozyme's PGR-eligibility challenge, its claim-construction argument, and its discretionary-denial positions (Fintiv / § 325(d) / Director-referral points). Institution decision (public PDF); practitioner summary: Venable BiologicsHQ.
• Final Written Decision: None issued as of 2026-09-26 (per my sources). The § 326(a)(11) one-year clock runs from the 2025-10-01 institution, putting the FWD deadline at ~2026-10-01, extendable up to six months for good cause. I found no claim-level holding in -00039 to report, and I will not manufacture one. (For calibration on timing, the sibling PGR2025-00017 on the '520 patent was instituted 2025-09-08, heard 2026-07-23, and decided 2026-09-01.)
• Settlement / termination: No settlement. But note a live termination risk: Halozyme moved to terminate these proceedings for Petitioner's alleged failure to name Merck & Co., Inc. ("MCI") as a real party-in-interest. The Board heard argument 2026-03-02 and, in an order entered 2026-03-03, found material facts still disputed (MCI's ability to control Petitioner; the role of shared general counsel) and authorized Petitioner to file a declaration on the RPI facts. I could not confirm a ruling on that motion. The fact that FWDs continued to issue in the sibling proceedings through May–September 2026 indicates trial was not stayed pending the RPI motion.
• Appeal: None. No FWD, therefore nothing appealable, and no Federal Circuit docket exists for this patent.
• Defensive value: The institution decision is the single most valuable public document against this patent — it is a Board finding, on the merits, that a § 112(a) attack on the claimed genus is more likely than not to succeed, and it includes an adverse claim construction (the "modified PH20 polypeptide" must be active), which simultaneously narrows the claims for non-infringement purposes. But because no FWD has issued, there is no cancellation to cite and no § 325(e)(2) estoppel, so a defendant must build its own record rather than lean on the patent owner's loss.

---

Strategic summary

Claim status on US 12,104,185. The patent has 11 claims. Claims 5-6 are CANCELED — not by the Board, but by Halozyme's statutory disclaimer under 37 C.F.R. § 1.321(a), filed 2025-06-13 in application 18/599,428 (see EX2003 in PGR2025-00039; corroborated by Venable and Pearce IP). A disclaimed claim is treated as if it never existed — the patent owner cannot recover past or future damages on claims 5-6, and any demand letter citing them is meritless. Claims 1-4 and 7-11 are INSTITUTED-BUT-UNTESTED — they are the claims in trial, and they remain presumptively valid and enforceable until a FWD issues. No claim has been sustained (or canceled) by the Board in -00039. Note that the disclaimer means the Board's trial covers claims 1-4 and 7-11 only; the Board cannot and does not reach 5-6. The broader campaign context matters: in the same 14-PGR campaign, the Board has invalidated the challenged claims of several sibling MDASE patents (Petitioner-reported: '600 via PGR2025-00003 on 2026-05-12; '262 via PGR2025-00006; '298 via PGR2025-00004 on 2026-06-08; '520 via PGR2025-00017 on 2026-09-01), and Alteogen — Merck's ALT-B4 partner — publicly counted five MDASE invalidations by 2026-09-02. Those are other patents, not '185; do not conflate them, but they are strong signal on how the Board reads the common specification.

Estoppel landscape. Because no FWD has issued, no § 325(e)(2) estoppel has attached to anyone. Once the FWD issues, Merck (and its privies/RPIs) will be barred from asserting in the D.N.J. litigation (or the ITC) that the claims are invalid on any ground it raised or reasonably could have raised in the PGR — which sweeps in the § 112(a) written description and enablement grounds and, arguably, the non-instituted obviousness ground, since it was actually raised in the petition. For any other defendant (a non-petitioner, non-privy), estoppel does not apply at all — you are free to run the full § 112(a) case, the '429+Chao § 103 case, and any art Merck did not plead. Two practical caveats: (i) if the RPI motion were granted and the proceedings terminated, the institution decisions and estoppel would evaporate — a material risk to any defendant building a § 112 strategy on -00039's institution decision; and (ii) non-mutual issue preclusion against Halozyme could become available in the D.N.J. case once a FWD is affirmed on appeal, independent of § 325(e)(2).

Pattern signals. The challenger is not a defensive aggregator. Merck Sharp & Dohme LLC — a large operating company, represented by Jeffrey P. Kushan / Sidley Austin LLP — is defending its own commercial product (subcutaneous pembrolizumab, KEYTRUDA QLEX™, using Alteogen's ALT-B4), and Merck filed between November 2024 and June 2025 against 14 Halozyme MDASE patents, all 14 now instituted. That is a coordinated competitor-vs-competitor campaign mirrored by Halozyme's own infringement suit, Halozyme, Inc. v. Merck Sharp & Dohme Corp., No. 2:25-cv-03179 (D.N.J.) (docket), in which the '185 patent is asserted. Halozyme's PTAB posture has been aggressive-but-losing: it pressed PGR-eligibility, claim-construction, Fintiv, § 325(d), and RPI-termination arguments; the Director declined discretionary denial and referred institution to the panels, and the Board has repeatedly rejected those threshold arguments. There is also parallel UK litigation in the Patents Court (a 2026 EWHC decision on Halozyme's Mayne Pharma disclosure order recounts MSD attacking an EP counterpart with the same "modified PH20 polypeptide / increased hyaluronidase activity" claim structure). No Unified Patents or other aggregator petition is in this chain — the Unified Patents URL in the record is merely the litigation-data source.

---

Recommended next steps

1. Pull the institution decision itself, today. Public copy · PTAB E2E: https://e2e.uspto.gov/. The § 112(a) reasoning (genus size 10⁴⁹–10⁶⁶; non-representative singly-substituted examples; prophetic "make-and-test" plan; no working multiply-substituted active) is a ready-made invalidity narrative, and the claim construction — claim 1's "modified PH20 polypeptide" requires hyaluronidase activity — is immediately useful for non-infringement contentions.
2. Diary the FWD. Statutory deadline ~2026-10-01 (§ 326(a)(11), extendable to ~2027-04-01 for good cause). Watch PTAB E2E for the FWD and for any order on Halozyme's pending RPI motion to terminate (argued 2026-03-02; unresolved on the record I reviewed). An RPI-based termination would erase the institution decision and the estoppel that would otherwise flow from it — that is the single biggest swing factor for anyone relying on this proceeding.
3. Do not cite claims 5-6 in any analysis of your exposure. They were disclaimed 2025-06-13 and cannot support liability. If a demand letter from Halozyme relies on claims 5-6, that is a strong signal of boilerplate.
4. If you are a non-petitioner defendant: you are outside § 325(e)(2), so run both tracks — the § 112(a) written description/enablement challenge (Board-endorsed at institution) and the § 103 case (the '429 patent + Chao / D320K), which the Board declined to institute and which therefore has no adverse Board precedent against it. Consider whether the pending FWD will make your invalidity case redundant and whether a stay is now supportable.
5. If you are Merck or a privy: track the estoppel clock. The moment the FWD issues, your § 112 and § 103-that-was-raised positions become unavailable in D.N.J. Plan your district-court case around art and grounds outside the petition.
6. Absence of a second proceeding is itself a data point. No IPR or additional PGR against the '185 patent has been filed (checked as of 2026-09-26). The PGR window closed ~2025-07-01; an IPR remains available to any non-estopped party, and no one has filed one. If you need a fallback invalidity vehicle and you can't join an existing challenge, that lane is still open.

I'll research this patent's assignment record and litigation posture before writing.

Good — the litigation picture is clear. Let me verify the corporate-name history and check for any security-interest or other recorded assignments.

Let me make one targeted attempt at the recorded reel/frame data and confirm the litigation docket.

Assignment & Ownership Analysis — US 12,104,185 B2

"PH20 polypeptide variants, formulations and uses thereof" · App. 18/599,428 · filed 2024-03-08 · granted 2024-10-01 · anticipated expiry 2032-12-28/29 · priority 2011-12-30 (prov. 61/796,208, 2011-12-30; prov. 61/631,313, 2012-11-01)

Source note / limitation up front: The USPTO Assignment Center (assignmentcenter.uspto.gov) was not directly queryable in this session. The recorded-assignment events below are taken from the Google Patents legal-events feed for US12104185B2 (https://patents.google.com/patent/US12104185/en), which mirrors the same underlying USPTO assignment records. Reel/frame numbers and the correspondent of record were not exposed by any source I could retrieve, and I will not invent them. That gap is flagged as such in the relevant sections rather than filled.

---

Inventors

Pattern observations:

• The invention was made c. 2011 (priority 2011-12-30), but the application on file (18/599,428) was filed 2024-03-08 — a late continuation. These are not "new" inventors in 2024; they are the original PH20 engineering team carried forward through a 13-year continuation chain (13/694,731 → … → 18/599,428).
• No mass-departure signal. Three of four inventors are San Diego–based and consistent with continuous Halozyme employment across the family. Shepard is the outlier (Eugene, OR; founder/former officer) and is the one inventor who assigned directly to a different Halozyme entity than the other three (see timeline) — an artifact of a separate inventor agreement, not a fire-sale tell.
• Shepard's exact separation date from Halozyme is not verified by any source retrieved; PGR exhibit EX2005 (LinkedIn profiles of all four inventors) exists in PGR2025-00039 but I have not inspected its contents. Do not read a departure date into this.

---

Original assignee

Halozyme, Inc. (San Diego, California) — named as assignee on the issued patent and as the party filing the 2025 statutory disclaimer (EX2003 in PGR2025-00039 lists "HALOZYME, INC." as patentee with "100%" ownership interest).

• Corporate structure: Halozyme, Inc. is the wholly owned operating subsidiary of Halozyme Therapeutics, Inc. (Delaware; NASDAQ: HALO; ~500 employees; HQ San Diego). Per the company's 10-K, "the Company" refers to "Halozyme Therapeutics, Inc., its wholly owned subsidiary, Halozyme, Inc. and Halozyme Inc.'s wholly owned subsidiaries, Halozyme Holdings Ltd., Halozyme Royalty LLC, Halozyme Switzerland GmbH and Halozyme Switzerland Holdings GmbH." The '185 patent sits in the IP-holding operating subsidiary, not in a separately named licensing vehicle.
• Line of business: Biopharmaceutical — recombinant human hyaluronidase (rHuPH20) drug-delivery platform (ENHANZE®) and the MDASE modified-PH20 portfolio. Revenue from royalties on partner products (Daratumumab/hyaluronidase, Phesgo, Herceptin Hylecta, Tecentriq Hybreza, Ocrevus Zunovo, Vyvgart Hytrulo, Opdivo Qvantig, etc.) plus its own marketed hyaluronidase product (HYLENEX).
• Product embodying the family: Yes for the platform — Halozyme commercializes a PH20-based drug-delivery technology and licenses it to >10 pharma partners. Nuance: the specific asserted claim of the '185 patent covers modified PH20s bearing a substitution at position 320 (H/K/R/S). Halozyme's own marketed enzyme is wild-type-derived rHuPH20; the accused product (Merck's SC Keytruda / berahyaluronidase alfa) is the alleged practice of the position-320 claims. So Halozyme is an operating company asserting platform patents against a competitor's product, not an NPE asserting against end users.
• Status: Operating, profitable, NASDAQ-listed, active litigant. Secured lender (Bank of America, 2022 credit agreement; predecessor Oxford Finance facility) holds a security interest in substantially all assets — including IP — but I found no evidence of any USPTO-recorded lien, security agreement, or release against this patent. Treat as unrecorded/UCC-only unless Assignment Center shows otherwise.

---

Assignment timeline

All recorded entries below are the three reassignment events of 2024-08-21 surfaced on US12104185B2. They are intra-family recordations. Execution dates and reel/frame numbers are not shown in the retrieved feed — recording date only.

• Executed n/a / recorded 2024-08-21 — Reel not retrievable in this session

  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
  • Assignor: CONNOR, ROBERT JAMES; WEI, GE; ZHAO, Qiping (joint inventors)
  • Assignee: HALOZYME THERAPEUTICS, INC.
  • Correspondent: not exposed by the source retrieved — the Assignment Center correspondent field would need to be read directly. Cannot assess recurrence. (Separately, the 2025 statutory disclaimer in 18/599,428 bears the typed name Robert Smyth and was filed by Kalpesh Upadhye; that is a prosecution filing, not a recorded-assignment correspondent, and must not be conflated.)
  • Context: Original inventor-to-company assignment for the 2024 continuation; catch-up recordation ~5 months after filing, ~6 weeks before grant.

• Executed n/a / recorded 2024-08-21 — Reel not retrievable in this session

  • Conveyance: Assignment
  • Assignor: SHEPARD, MICHAEL H.
  • Assignee: HALOZYME, INC.
  • Correspondent: not exposed by the source retrieved.
  • Context: Original inventor-to-company assignment — Shepard's rights go straight to the operating subsidiary, consistent with a founder/former-officer inventor agreement distinct from the employee agreements of the other three.

• Executed n/a / recorded 2024-08-21 — Reel not retrievable in this session

  • Conveyance: Assignment
  • Assignor: HALOZYME THERAPEUTICS, INC.
  • Assignee: HALOZYME, INC.
  • Correspondent: not exposed by the source retrieved.
  • Context: Internal corporate reorganization — parent-to-subsidiary consolidation of title so that 100% of the patent sits in Halozyme, Inc. (corroborated by the June 2025 disclaimer identifying Halozyme, Inc. as 100% owner).

Net effect of the chain: four inventors → (three to Halozyme Therapeutics, Inc.; one to Halozyme, Inc.) → Halozyme Therapeutics, Inc.'s interest → Halozyme, Inc. = sole 100% owner. No third party, no LLC, no licensing vehicle ever appears in the recorded chain.

If Assignment Center, queried directly, returns no post-2024-08-21 records beyond these three, that is itself the finding: the original assignee still owns the patent outright.

---

Timeline diagram

timeline
    title Ownership of US 12104185
    1998 : Halozyme founded in San Diego
    2011 : PH20 MDASE invention disclosed
    2012 : Parent application 13 694731 filed
    2016 : First MDASE patent issues as 9447401
    2024 : Continuation 18 599428 filed in March
         : Patent US 12104185 issues 1 Oct
         : Inventor assignments recorded 21 Aug
    2025 : Halozyme sues Merck in New Jersey
         : Merck files PGR2025-00039
         : Patentee disclaims claims 5 and 6

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. Every recorded assignee is named "Halozyme" — the operating subsidiary of a NASDAQ-listed biopharma with ~500 employees and marketed partner products. No "IP / Licensing / Holdings / Ventures" vehicle appears anywhere in the chain. The corporate family does contain Halozyme Royalty LLC and Halozyme Holdings Ltd., but neither is a recorded assignee of the '185 patent — naming alone is not a finding, and here there is no transfer to them at all.

2. Known asserter in the chain — NOT PRESENT. No assignee matches any public NPE list (Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Spangenberg entities, etc.). To the contrary, Unified Patents' own PTAB case list classifies the owner as "Halozyme Inc — Operating Company" (https://portal.unifiedpatents.com/ptab/caselist — PGR2025-00039, patent 12104185, instituted 2025-10-02). Note carefully: Halozyme is the patent owner being challenged, not the petitioner — the petitioner is Merck Sharp & Dohme LLC, designated a "Large Operating Company." A patent owner suing a bigger competitor is the inverse of the NPE pattern.

3. Repeat correspondent across the chain — UNASSESSABLE / NO FINDING. The correspondent of record was not exposed by any source I could retrieve, so recurrence cannot be tested. This is the single evidentiary gap in this analysis; it should be closed by reading the correspondent field directly in Assignment Center for each of the three 2024-08-21 entries. Absent that read, no correspondent-based signal can be asserted either way.

4. Cascading transfers — NOT PRESENT. Three recordings share a single date (2024-08-21) and are all intra-family: two inventor→company grants plus one parent→subsidiary consolidation. There is no chain of unrelated LLCs, no serial re-assignment in <24 months, no shared registered-agent address pattern. The ownership interval between core events is measured in years (2012 → 2024), not months.

5. Pre-litigation transfer — NOT PRESENT. The only 2024 recording (2024-08-21) sits ~8 months before the earliest reported infringement complaint (Halozyme v. Merck, D.N.J. 2:25-cv-03179, filed on/about 2025-04-24), i.e. outside the 6-month window, and it is an inventor→company recordation, not a transfer to an asserting third party. The suit is brought by the original assignee's own subsidiary, so there was no standing to manufacture.

6. Bankruptcy fire-sale — NOT PRESENT. Halozyme Therapeutics, Inc. is a going concern with royalty revenue and an active license/assertion program; no Chapter 7/11, no patent auction, no reorganized-estate seller anywhere in the record.

7. Privateering — NOT PRESENT. No operating company transferred the patent to a separate assertion entity. Halozyme, Inc. — a real operating company — is asserting in its own name, and its parent filed the complaint. The disclosure posture (licensing offers to Merck in March 2025, then suit) is ordinary platform monetization, not privateering through a proxy.

8. Defensive aggregator — NOT PRESENT. The chain terminates at Halozyme, Inc. (asserting), not at RPX, AST, LOT, Unified Patents, or OIN. However, note the neutralization risk in the opposite direction: the patent is under an instituted PGR (PGR2025-00039, instituted 2025-10-02) on § 112 written-description/enablement grounds, and Halozyme statutorily disclaimed claims 5–6 on 2025-06-13 (EX2003, App. 18/599,428). Eight-plus related Halozyme patents are under parallel Merck PGRs, several already instituted. The asset's value is being attacked, not aggregated.

---

Verdict

Operating-company assertion.

Three recorded assignments — all dated 2024-08-21 and all intra-family — move 100% of US 12,104,185 from the four named inventors through Halozyme Therapeutics, Inc. to Halozyme, Inc., the wholly owned operating subsidiary of a NASDAQ-listed biopharma (~500 employees; ENHANZE®/MDASE platform; marketed hyaluronidase products and 10+ pharma royalty partners). No shell LLC, no licensing vehicle, no listed NPE, no cascading re-assignment, and no pre-suit transfer appears anywhere in the chain; Unified Patents classifies the owner as an "Operating Company," and the assertion (D.N.J. 2:25-cv-03179, on/about 2025-04-24) is a platform-holder suing a far larger competitor. The one signal that would require care — repeat assignment correspondent — cannot be scored because the reel/frame and correspondent fields were not retrievable in this session; verifying them in Assignment Center is the only outstanding item, and it would not change the verdict given signals 1, 2, 4, 6, 7, and 8 are all negative on the record above.

Verify at: https://assignmentcenter.uspto.gov/ (search patent 12,104,185) · mirrored feed: https://patents.google.com/patent/US12104185/en · litigation/PTAB overlay: https://portal.unifiedpatents.com/patents/patent/US-12104185-B2

I'll search for the specific patent and its cited prior art references.

Let me get the actual "References Cited" list for the '185 patent and the PGR petition details.

Let me retrieve the actual face-of-patent "References Cited" list and the Google Patents citation data.

Prior-Art Analysis — U.S. Patent No. 12,104,185 B2

1. Identification (interpreted literally — no identifiers auto-corrected)

Sources: https://patents.google.com/patent/US12104185/en ; https://portal.unifiedpatents.com/patents/patent/US-12104185-B2 ; https://ai-lab.exparte.com/case/[ptab](/ptab)/PGR2025-00039/doc/1001

Retrieval caveat (stated honestly): I could not pull the complete verbatim "(56) References Cited" block from the USPTO full-text server or Google Patents full text with the tools available. The list below is reconstructed from (a) the '185 patent's own face as reproduced in PGR2025-00039 Exhibit 1001, (b) the Exhibit-1001 page image index at Docket Alarm, and (c) the identical specification's reference list as it appears on sibling family members (US 11,066,656, US 10,328,130, US 9,677,062, US 8,765,685). The list is therefore substantially complete but I cannot certify it as exhaustive or verbatim. Items I could not independently verify are marked "(not verified)".

---

2. What "prior art" means here — a threshold point that changes the answer

The '185 patent is an AIA-first-to-file case with a Dec. 30, 2011 priority claim. Two consequences:

1. A reference must predate the effective filing date to be § 102 art. If the priority claim is perfected (Halozyme's position), everything published after Dec. 30, 2011 is not § 102(a)(1) art. Merck expressly attacked this — its petition argued the challenged claims are "not entitled to benefit to any pre-March 13, 2013 application" (Pet. § V). If Merck's priority challenge succeeded, the effective date would move later and far more art (including Halozyme's own WO 2013/102144) would become available.
2. Most of the (56) list predates 2011 comfortably and is § 102(a)(1) art.

Bottom line up front: No single reference on the face of the '185 patent discloses every element of claim 1 (a position-320 D→H/K/R/S replacement plus ≥95% identity to SEQ ID NOS:3 and 32–66). On the record I can verify, nothing in the (56) list anticipates claims 1–11 under § 102; the references function as § 102 background/§ 103 art and as support for the formulation disclosure. The PTAB agreed in part — it instituted on §§ 112 only and found Petitioner had not shown a reasonable likelihood of success on obviousness. This is a § 102 rebuttal, not a § 102 hit.

---

3. Category A — References most relevant to the PH20/variant subject matter

These are the only (56) references that could conceivably reach the claims. Each is analyzed claim-by-claim below.

Additional PH20/hyaluronidase-family art cited in the same (56) lineage (all Halozyme/Bookbinder–Frost publications): US 2005/0260186 A1; US 2006/0104968 A1; US 2009/0123367 A1; US 2009/0181013 A1; US 2009/0181032 A1; US 2009/0214505 A1; US 2009/0253175 A1; US 2009/0304665 A1; US 2009/0311237 A1; US 2010/0003238 A1; US 2010/0015698 A1; US 2010/0184845 A1; US 2011/0152359 A1; US 2011/0212074 A1; US 2012/0093770 A1; US 2012/0148555 A1; US 2012/0171153 A1; US 2012/0251517 A1; US 2012/0251620 A1; US 2012/0294830 A1; US 2013/0058893 A1; US 2014/0135682 A1; US 2014/0199282 A1; US 2014/0248237 A1 (dates as printed: 11/2004–3/2020 window, Bookbinder/Frost). These teach rHuPH20/sHASEGP production, PEGylation, formulations and uses. None discloses the position-320 replacements; none anticipates claims 1–11.

---

4. Category B — Formulation / delivery / conjugation art on the (56) face

These are cited to support the specification's formulation and dosage-form disclosure (sustained-release, PEG conjugates, preservatives, insulin co-formulation). They are categorically incapable of anticipating claims 1–11, all of which require a position-320-modified PH20 polypeptide. Representative entries:

• Osmotic/sustained-release dosage forms: US 3,598,123 (Zaffaroni, 8/1971); US 3,630,200 (Higuchi, 12/1971); US 3,710,795 (Higuchi et al., 1/1973); US 3,845,770 (Theeuwes et al., 11/1974); US 3,916,899 (Theeuwes et al., 11/1975); US 4,008,719 (Theeuwes et al., 2/1977); US 5,059,595 (Le Grazie, 10/1991); US 5,120,548 (McClelland et al., 6/1992); US 5,639,476 (Oshlack et al., 6/1997); US 5,674,533 (Santus et al., 10/1997)
• Packaging/blister art: US 5,033,252 / US 5,052,558 (Carter, 7/1991; 10/1991); US 5,323,907 (Kalvelage, 6/1994)
• Aerosol/steroid delivery: US 4,044,126 / US 4,414,209 (Cook et al., 8/1977; 11/1983); US 4,364,923 (Cook et al., 12/1982)
• PEG–protein conjugation: US 4,002,531 (Royer, 1/1977); US 4,179,337 (Davis et al., 12/1979); US 5,122,614 / US 5,324,844 (Zalipsky, 6/1992; 6/1994); US 5,446,090 (Harris, 8/1995); US 5,612,460 (Zalipsky, 3/1997); US 5,643,575 (Martinez et al., 7/1997); US 5,672,662 (Harris et al., 9/1997)
• Liposomes/microspheres/controlled release: US 4,769,027 (Baker et al., 9/1988); US 4,980,286 (Morgan et al., 12/1990); US 5,073,543 (Marshall et al., 12/1991); US 5,591,767 (Mohr et al., 1/1997)
• Cloning/expression vectors: US 4,952,496 (Studier et al., 8/1990)
• Insulin chemistry: US 3,536,809 (Applezweig, 10/1970)

§ 102 verdict for Category B: none anticipates any of claims 1–11. Every independent claim recites the modified PH20 element; these references contain no PH20 variant disclosure.

---

5. Category C — Applicant's own family (not (56) art; relevant to priority)

The following are listed under the '185 patent's cross-reference/continuity data, not as prior art: US 9,447,401; US 10,865,400; US 11,066,656; US 12,091,692; US 12,264,345; US 11,952,600; US 12,018,298; US 12,123,035; US 12,110,520; US 12,054,758; US 12,060,590; US 12,049,652; US 12,037,618; US 12,077,791; US 12,195,773; US 12,091,384; US 12,104,184 and applications 13/694,731; 17/327,568; 18/338,189; 18/759,577; 18/922,889; 18/069,651; 18/340,786; 19/071,005; 19/071,055; 19/075,092; 19/071,264; 19/071,345; PCT/US2012/072182.

These are § 102(a)(2)/§ 102(b)(2)(C) territory (commonly owned, same inventive entity — Wei, Shepard, Zhao, Connor). They cannot be § 102(a)(1) art against the '185 patent, but they are the primary § 112 written-description/enablement battleground, and Merck filed PGRs against essentially the whole set. Note the correspondence mapping in PGR2025-00039 EX1015 and the redline in EX1045, which identify three substantive specification edits between the '731 application and the '185 patent: removal of positions 282, 298, 431 from the "mutations to avoid" list; substitution of I083K with I208K in Table 9; and removal of L and W from position 288 in Table 3. See https://www.docketalarm.com/cases/PTAB/PGR2025-00039/Merck_Sharp_%26_Dohme_LLC/docs/03-28-2025-Petitioner/Exhibit-1068-Declaration_of_Jeffrey_P_Kushan.pdf

---

6. The substantive prior-art challenge actually made — and how it fared

Case: Merck Sharp & Dohme LLC v. Halozyme Inc., PGR2025-00039 (U.S. Patent 12,104,185), petition filed March 28, 2025; instituted Oct. 1–2, 2025.

• Ground 1 (instituted? NO): Claims 1–11 obvious over U.S. 7,767,429 in view of Chao et al. (2007). Merck's theory was that the claims encompass the single species D320K PH20₁₋₄₄₇, and that the '429 patent's statement that "single amino acid substitutions in non-essential regions … do not substantially alter biological activity" plus Chao's alignment made D320K obvious. Examiner-side and PTAB-side: the Board found Petitioner had not shown it was more likely than not that the claims are obvious, noting Merck offered no reason why a POSA would modify position 320, and that declarant Park conceded he was "asked by counsel to report [his] conclusions with respect to position 320" (EX1004 ¶103).
• Grounds 2–3 (instituted? YES): insufficient written description (35 U.S.C. § 112(a)) and lack of enablement for the broad, structurally-defined genus (Petitioner estimated 10⁴⁹–10⁶⁶ sequences).

Authority: https://biologicshq.com/wp-content/uploads/2025/10/PTAB-PGR2025-00039-Institution-Decision.pdf ; https://www.lexology.com/library/detail.aspx?g=12e547c3-5cea-4251-837a-e07ff5fa2a19

Additional prior-art exhibits mattering only to § 103/§ 112, not § 102: Steipe (2004); Chica (2005); Bordoli (2008); Green (2010); Branden & Tooze (1999); Gmachl (1993); Markovic-Housley (2000); Hardy (2004); Pomering (2002); Baba (2002); Primakoff (1997); Tung; KEGG "Hyaluronidase (human recombinant)"; plus prov. apps 61/631,313 and 61/796,208.

---

7. Direct answer to the question asked

"Which reference potentially anticipates which claim under § 102?" On the verified record:

Most relevant prior art overall, ranked:

1. U.S. 7,767,429 B2 — closest art; § 103 primary reference; not anticipatory.
2. WO 2010/077297 A2 — modified hyaluronidase art; § 103.
3. Arming et al. (1997) — the only reference that actually maps PH-20 residues by mutagenesis; still no position-320 disclosure.
4. Chao et al. (2007) — structural reference used for the "non-essential region" argument; § 103.
5. Frost (2007) / Bookbinder (2006) — disclose the claimed use (rHuPH20 enabling SC delivery) but not the claimed variant; § 103/§ 112 background.

If you want a genuine § 102 attack, the realistic candidate is Halozyme's own WO 2013/102144 A2 (published July 4, 2013) — the PCT counterpart of the '185 family — which is § 102(a)(1) art only if the Dec. 30, 2011 priority claim is defeated (the argument Merck actually advanced in Pet. § V, "None Are Entitled to Benefit to Any Pre-March 13, 2013 Application"). Note also the printed priority inconsistency across databases: Google Patents shows 2011-12-30, while Unified Patents shows 2011-12-29; I am not auto-correcting either, and I flag it because a one-day discrepancy around a provisional filing date is the kind of thing that matters to a § 102 date computation.

Confidence statement: High confidence on the patent identification, claims, dates, PGR outcome, and the '429/Chao/WO 2010/077297/Arming/Frost/Bookbinder relevance. Lower confidence on the completeness and verbatim accuracy of the (56) list, because I could not retrieve the full face-of-patent text directly; the reconstruction is drawn from PGR Exhibit 1001's index and sibling-family spec pages, and I have marked the items I could not independently verify.

I'll research the prior art cited on this patent and the parties' obviousness contentions before building the analysis.

I have the core obviousness grounds. Let me dig into the specific rationale and the reference set.

Let me pin down the remaining prior-art references and any art directly addressing position 320.

Obviousness Analysis — U.S. Patent No. 12,104,185 B2 (35 U.S.C. § 103)

Analyst note on sources. The Google Patents page supplied to me reproduces the front matter, the bibliographic/classification tables, and only the prior-art keywords ("position corresponding," "polypeptide," "amino acid," "modified," "seq") and the advertised prior-art date (2011‑12‑30). The page's "References Cited" list is truncated in the supplied text, so I cannot quote the patent's own IDS verbatim. To avoid fabricating citations, I have grounded the reference set below in (i) the reference list and grounds actually relied on by the petitioner in the co-pending post-grant review of this exact patent (PGR2025‑00039, Merck Sharp & Dohme LLC v. Halozyme, Inc.), and (ii) the patent family's own prior documents. Where a legal conclusion is contestable I say so explicitly. Per the operating rules, the patent text controls over search results; the date discrepancy flagged in the earlier summary is carried forward.

---

1. The claim that must be invalidated

Claim 1 (the sole independent claim) is a method of use claim:

A method for increasing delivery of a therapeutic agent to a subject, comprising administering a pharmaceutical composition comprising (a) a therapeutic agent and (b) a modified PH20 polypeptide that (i) is ≥95% identical to SEQ ID NO:3 or SEQ ID NOs:32–66, (ii) has an amino acid modification at a position corresponding to position 320 of SEQ ID NO:3, and (iii) where the position‑320 replacement is H, K, R or S.

Two features of the claim scope drive the whole §103 analysis and are easy to lose sight of:

1. Claim 1 requires no improvement. It does not require increased hyaluronidase activity, increased stability, or resistance to any denaturant. The only claims that did — claims 5 and 6 (increased activity; ≥120% activity) — were statutorily disclaimed by Halozyme before institution. The surviving claim is therefore satisfied by a variant that merely retains activity sufficient to act as a spreading agent.
2. Claim 1 covers single‑substitution species. A single D320K change in the 447‑residue SEQ ID NO:3 is >99.7% identical to SEQ ID NO:3. The claim is thus infringed by — and obvious if — one such species is obvious. Merck's petition zeroed in on D320K PH20₁₋₄₄₇, which is also the closest analogue to the modified PH20 in the accused/asserted product context (berahyaluronidase alfa; and cf. the D320K-type enzyme discussed in the EPO/D.N.J. litigation materials).

---

2. Person of ordinary skill and the governing legal framework

Merck's petition defines the POSITA as having an undergraduate degree plus a Ph.D. and post‑doctoral experience in protein structure/function (chemistry, biochemistry, biophysics), with familiarity with sequence searching/alignment, protein modeling, recombinant production of modified proteins, and biological assays. Halozyme disputes the omission of a hyaluronidase-specific requirement, but the dispute is not outcome‑determinative for §103.

The analysis below uses the KSR/MPEP rationales to supply the required reason to combine:

• (A) express or implicit suggestion in the prior art;
• (B) a "known technique used to improve similar devices in the same way";
• (C) "obvious to try" — a finite, predictable set of solutions;
• (D) design incentives and market/regulatory pressure;
• (E) teachings, suggestions or motivations in the references themselves, considered as a whole with the POSITA's background knowledge.

Critically, §103 also permits an "obvious species" theory: where a claim covers a genus that includes a species the prior art would have made obvious, the genus claim fails. In re Petering; In re Baird.

---

3. The reference set

The critical temporal point: every one of these references predates December 30, 2011 (the advertised prior-art date on the face of the '185 patent). The '429 patent and WO 2010/077297 are also commonly owned by Halozyme, which is relevant to the "same field / same problem / developed by the same entity" line of motivation reasoning — and, under §103, the inventor's own prior work is prior art like any other.

---

4. The primary combination — '429 patent + Chao (+ the POSITA's knowledge)

4.1 What each reference supplies

The '429 patent supplies every element of the method except the specific residue and substitution:

• The enzyme: soluble PH20₁₋₄₄₇ — the same molecule as SEQ ID NO:3, which is one of the claim's two identity reference sequences.
• The method: co-administration of PH20 with a therapeutic agent to increase spreading/delivery; expressly including antibodies, other proteins, and small-molecule drugs; including subcutaneous administration.
• The variant-making instruction: modified PH20 polypeptides containing one or more amino acid substitutions are disclosed and claimed; and the specification affirmatively states that "single amino acid substitutions in non-essential regions of a polypeptide do not substantially alter biological activity." That sentence is both a motivation (a POSITA is told such variants are useful and safe to make) and a reasonable-expectation teaching (they will retain activity).

Chao supplies the roadmap for locating a non-essential position:

• Chao's experimentally determined HYAL1 structure plus the five-human-hyaluronidase alignment identifies conserved/essential positions (the catalytic acidic and tyrosine residues, the conserved cysteines, the conserved glycosylation asparagines) and, by contrast, the variable/non-conserved positions that a POSITA uses as candidates for substitution.
• Chao places the region around PH20 position 320 within the α8 helix, flanked by essential residues (petition identifies positions 316 and 327 as bounding essential residues C316 and L327). Position 320 sits between them — i.e., in the non-essential interval. This is exactly the sort of "structural feature" analysis a POSITA performs when implementing the '429 patent's instruction.
• Chao's alignment further shows that the residue corresponding to PH20 D320 varies among homologous hyaluronidases — i.e., the position tolerates substitution.

The POSITA's own knowledge (and the multi-sequence alignment analysis) supplies the specific choice, K:

• A multiple-sequence alignment of hyaluronidase sequences homologous to PH20 published as of Dec. 2011 (petition's Park declaration used a set of 88 sequences) shows lysine is the most prevalent residue at the position corresponding to PH20 320 (petition: 51 of 88 sequences), and lysine also occurs at the corresponding position in three other human hyaluronidases.
• Hecht adds two structural reasons: (i) lysine has high α‑helical propensity and position 320 lies in the α8 helix; (ii) lysine is hydrophilic, like the aspartate it replaces.

4.2 Motivation to combine (rationales)

1. Express suggestion in the '429 patent (rationale A). The '429 patent is not merely analogous art; it is the same inventors' and same assignee's disclosure of the very method claimed (co‑administering soluble PH20 with a therapeutic agent to increase delivery). It affirmatively directs the artisan to make single substitutions in non-essential regions. A reference that teaches the claimed method and tells the artisan how to make variants of the enzyme used in that method supplies a strong reason to do so.
2. Same field, same problem, predictable solution (rationales B and C). Chao and the hyaluronidase reviews are in the same field (human hyaluronidase structure/function). Applying a structure/sequence alignment to select a tolerated substitution is a "known technique used to improve similar devices in the same way." Selecting one residue out of a finite, ranked candidate set flagged by conservation analysis is the archetype of a "predictable" solution with a reasonable expectation of success.
3. Design incentive (rationale D). As of 2011, rHuPH20 was being commercialized as an "enabling platform" for SC delivery (Frost 2007; Bookbinder 2006), with known practical problems — immunogenicity relative to bovine/ovine hyaluronidases, sensitivity to phenolic preservatives and elevated temperature, and low-salt instability. The '429 patent itself motivates modification (PEGylation, sequence variants). A POSITA therefore had a concrete reason to generate and test sequence variants of PH20₁₋₄₄₇ while preserving activity.
4. The '429 patent's own claim scope as an admission. The '429 patent claimed modified PH20 polypeptides with one or more substitutions, pharmaceutical compositions thereof, and methods of increasing delivery — without exemplifying a substituted active mutant. Halozyme obtained that breadth from the Office on the representation that such substitutions retain activity. Merck expressly argues Halozyme "cannot contradict its past representations to the PTO." This is a §103 "applicant's own admissions" point as much as an estoppel point.

4.3 Reasonable expectation of success

The expectation need only be reasonable, not certain (and it need only support retained activity, because claims 5–6 were disclaimed):

• Sequence distance is trivial. One substitution in 447 residues. The '429 patent's own statement that single substitutions in non-essential regions "do not substantially alter biological activity" maps directly onto this degree of change.
• Conservation data. Lysine is observed at the corresponding position in dozens of naturally occurring homologs and in three other human hyaluronidases. Evolution's own tolerance is evidence of structural tolerance.
• Structural data. Position 320 is in a surface-involved helix, not in the catalytic site; the catalytic acidic residues and tyrosines identified by Zhang/Chao are elsewhere. Chao's α8 assignment plus helix-propensity reasoning (Steipe; Pace, cited in the Hecht declaration) supports tolerance.
• Empirical precedent. Arming (1997) had already shown that single substitutions in PH-20 yield activity-retaining mutants, and Zhang (2009) had shown that single substitutions in Hyal1 can be made and assayed.

4.4 Dependent claims

Because claims 2–11 depend from claim 1 and add either a mechanism of action or an administration route already disclosed, the §103 case against them collapses into the case against claim 1.

---

5. Alternative / reinforcing combinations

Even if a tribunal finds the '429 + Chao combination insufficient on its own, the following alternative combinations independently supply the missing elements and can be pleaded as separate grounds:

• Ground B — '429 + WO 2010/077297 + Stern/Zhang (essential-residue mapping). WO 2010/077297 (Halozyme, published July 2010, commonly owned) expressly discloses amino-acid-substituted human esPH20 polypeptides with retained or increased activity. That is a per se suggestion that the soluble PH20 backbone tolerates substitution, reinforcing the expectation of success. Stern and Zhang then mark out where activity cannot be touched (the catalytic acids/tyrosines), leaving positions like 320 as the natural design space. Rationale: express suggestion + exclusion-based design logic.
• Ground C — '429 + Frost 2007 + Bookbinder 2006 + Arming 1997. Frost and Bookbinder establish the method (rHuPH20-enabled SC/interstitial delivery of co-administered therapeutics), and Arming establishes the technique of single-substitution mutagenesis of PH‑20. This combination supplies the "increasing delivery" and "therapeutic agent" elements while treating the specific substitution as a routine optimization. Rationale: known technique + design incentive + obvious-to-try.
• Ground D — '429 in view of Chao in further view of Steipe (2004) and Chica (2005). These references supply the consensus/consensus-engineering methodology by which a POSITA converts a multi-sequence alignment into a ranked list of tolerated substitutions — directly answering the criticism that the specific residue choice was hindsight-driven. Rationale: reasoned application of a known engineering methodology to a known target.

---

6. The patent owner's counterarguments — and how strong they are

I would be doing this analysis a disservice not to state the weaknesses of the petitioner's case, particularly because the PTAB instituted on the §112 grounds but found the petitioner had not shown it was more likely than not to prevail on obviousness (Venable BiologicsHQ, Oct. 10, 2025, https://biologicshq.com/ptab-grants-institution-of-three-pgrs-related-to-mercks-keytruda-qlex/). Halozyme's POPR arguments (PGR2025‑00039, Paper 17) are, in substance:

1. The prior art does not disclose the limitation. Neither the '429 patent nor Chao mentions any mutation at PH20 position 320, much less D320H/K/R/S. Halozyme cites PAR Pharmaceutical v. TWi, 773 F.3d 1186 (Fed. Cir. 2014) for the proposition that a §103 analysis requires "determining that claimed elements are present in the prior art."
   • Rebuttal: The record does not need an express disclosure of the species if the genus is rendered obvious and the species is a predictable member. The absence of an express recitation is not dispositive where the '429 patent affirmatively directs single substitutions in non-essential regions and the position is objectively identifiable from Chao. Whether that suffices is the crux.
2. No reason to make the modification at all. Because the '429 patent teaches that such substitutions produce the same activity and same utility, a POSITA would have had no incentive — "why expend resources to arrive at something the art says does nothing new?" Halozyme then argues the motivation to make an activity-neutral variant is absent, citing Metalcraft of Mayville v. Toro, 848 F.3d 1358 (Fed. Cir. 2017) and Virtek v. Assembly, 97 F.4th 882 (Fed. Cir. 2025).
   • Rebuttal: A "same activity / same utility" variant is precisely what claim 1 covers. The patent owner's argument proves too much: the claim requires no benefit over wild-type PH20, so the absence of an improvement motivation cannot defeat obviousness. Moreover, the '429 patent's own claims to substituted PH20 and its PEGylation/immunogenicity discussion supply an independent engineering incentive (e.g., reduced immunogenicity, as WO 2010/077297 teaches).
3. No reasonable expectation of success. Halozyme characterizes the '429 patent's statement as general guidance of the Medichem v. Rolabo / In re Stepan type, and stresses that this guidance is not even about D→K, which is a charge reversal (acidic → basic), not a conservative substitution.
   • Rebuttal: The expectation is not built solely on the '429 patent's general sentence; it is built on the conservation data (K observed at the corresponding position in 51/88 homologs and 3 human hyaluronidases) and the structural placement in α8. That is art-founded, satisfying In re Vaeck's requirement that the expectation be "founded in the prior art." The charge-reversal objection is real but weakened by the "hydrophilic-for-hydrophilic" and helix-propensity points, and by the fact that the claim's express alternatives include S (a conservative, polar substitution) as well as H/K/R.
4. Improper reliance on a litigant-generated MSA. Halozyme notes that Dr. Park's 88-sequence alignment was assembled at counsel's request with the knowledge that position 320 was the target (Park conceding he "was asked by counsel to report [his] conclusions with respect to position 320"), and that the alignment is dominated by non-PH20 hyaluronidases that differ in function and substrate.
   • Rebuttal: The underlying sequences were publicly available in Dec. 2011; the alignment merely quantifies what the art already contained. But this remains the petitioner's most vulnerable flank, because it invites a hindsight characterization.
5. §325(d) / Advanced Bionics. Halozyme argues the '429 patent was in the IDS and Chao is cumulative to Stern, Zhang and Arming, which the Examiner considered — so the same art/arguments were before the Office.
   • This is a discretionary argument, not a merits one; the Board's institution shows it did not carry the day at the threshold.

Net assessment. On the papers as they exist, the §103 ground is stronger than the Board credited at institution — chiefly because claims 5–6 (which required increased activity) were disclaimed, so the "no unexpected benefit / no motivation to improve" defense loses much of its force. But it is not a foregone conclusion. A reviewing tribunal must accept that (i) the specific residue choice was reached by an MSA performable but not performed in the prior art, and (ii) every one of the ~30 discrete analytical steps the petitioner's declarants enumerate could be characterized as an impermissible post-hoc roadmap. The strongest framing for the petitioner is the obvious-species framing with an emphasis on the single-substitution, activity-neutral nature of the claim.

---

7. The priority-date question — a distinct and potentially decisive route

This is the single most important strategic point and it interacts directly with §103.

Merck's PGR position is that no claim of the '185 patent is entitled to the Dec. 2011 provisionals or to the Dec. 28, 2012 filing date of the '731 application, because the '731 application added subject matter (new examples, new tested mutants) and does not provide §112(a) support; the Board found the evidence better supported the petitioner. If that conclusion is sustained, the '185 patent's effective filing date becomes March 8, 2024 (its actual filing) — or at best 2012‑12‑28 if the §112 finding is reversed.

Why that matters for §103: with a 2024 effective filing date, a substantial body of Halozyme's own intervening disclosures becomes available art, including:

• WO 2013/102144 A2 (published July 3, 2013) — the PCT publication of this very family, which discloses the modified PH20 polypeptides and the position‑320 variants. As a printed publication it is §102(a)(1) art, and the common-ownership exception in §102(b)(2)(C) does not reach §102(a)(1) art. It is more than one year before 2024, so the §102(b)(1) inventor-disclosure grace period does not apply either.
• US 2013/0302275 A1 and U.S. Pat. No. 9,447,401 — these are §102(a)(2) documents and would ordinarily be excepted as commonly owned under §102(b)(2)(C), so I flag them as legally weaker prior art than the WO publication and would not lead with them.

If WO 2013/102144 is properly §102(a)(1) art, the obviousness case becomes far easier — and, if that publication discloses D320K with the recited identity, the analysis shades into anticipation under §103's close cousin, §102. I flag this as conditional and provisional: I have not independently verified the full text of WO 2013/102144's disclosure of the four specific position‑320 residues, nor the ownership record as of the relevant dates, and the §112 priority question is unresolved on the merits.

---

8. Secondary considerations

There is no evidence of record establishing nexus between any asserted secondary consideration and the surviving claim scope:

• Claims 5–6 (which would have required increased activity) were disclaimed — so any evidence of "unexpected activity" is not commensurate with the claims actually in force. In re Kao; Rambus.
• Any stability/preservative-resistance benefit would need to be tied to the full genus of ≥95%-identity variants, which the petitioner's §112 arguments already assert is not representative.
• Halozyme's own litigation positions (including the EPO proceeding on EP 3 130 347, where its Notice of Experiments compares a reference-standard SEQ ID NO:3 enzyme with MSD's ALT‑B4 enzyme; see [2026] EWHC 1838 (Pat)) concern an activity comparison that is not an element of claim 1 as it now stands.

---

9. Conclusion

Claims 1–4 and 7–11 are, in my assessment, more likely than not obvious under §103 in view of:

• Primary ground: US 7,767,429 in view of Chao et al. (2007) and the knowledge of a POSITA — including the multi-sequence-alignment and structural evidence placing position 320 in a non-essential α8-helix interval with lysine well tolerated there — rendering the D320K PH20₁₋₄₄₇ species obvious, which brings the genus of claim 1 within §103;
• Reinforcing grounds: the addition of Frost (2007) and Bookbinder et al. (2006) for the "increasing delivery / subcutaneous / antibody / small-molecule" elements; WO 2010/077297 for the express teaching of amino-acid-substituted soluble PH20; Stern (2006), Jedrzejas (2005) and Zhang (2009) for the essential-residue exclusions; Arming (1997) for single-substitution mutagenesis of PH‑20; and Steipe (2004) / Chica (2005) for the consensus-engineering methodology; and
• A conditionally stronger route: if the claims are denied their 2011/2012 priority dates, WO 2013/102144 A2 (published July 3, 2013) discloses the position‑320 variants themselves and supplies §102(a)(1) art that no common-ownership exception reaches.

The principal obstacles the petitioner must overcome are (i) the complete absence of any express position‑320 mutation in the '429 patent or Chao, (ii) the hindsight challenge to the litigant-generated 88-sequence alignment, and (iii) the Federal Circuit's insistence (Medichem, In re Stepan, Virtek, Metalcraft) that a reason to make the specific modification — not merely the capability of making it — be founded in the prior art. Those obstacles are materially reduced by the disclaimer of claims 5–6, because the surviving claim requires only retained, not improved, enzymatic activity.

Explicit uncertainties

• I could not retrieve the full PTAB institution decision text or a final written decision in PGR2025‑00039; the Board's statement that the petitioner had not shown a likelihood of prevailing on obviousness (as reported) is adverse to the conclusion above and should be weighted accordingly.
• I could not verify the '185 patent's own "References Cited" list because the supplied Google Patents text is truncated after the prior-art keywords; the reference set above is drawn from the PGR record and family documents, not from the patent's IDS.
• I have not verified the disclosure content of WO 2013/102144 as to the four specific position‑320 residues, nor the ownership record underpinning the §102(b)(2)(C) analysis.