Patent: 10,004,788

Summary for Patent: 10,004,788

Title:	Treatment of ocular neovascularization using anti-VEGF proteins
Abstract:	The present disclosure provides compositions and methods for the prevention or treatment of ocular neovascularization, such as AMD, in a human subject, by administering subretinally a pharmaceutical composition comprising a pharmaceutically effective amount of a vector comprising a nucleic acid encoding soluble Fms-related tyrosine kinase-1 (sFlt-1) protein to the human subject.
Inventor(s):	Constable; Ian J. (Mosman Park, AU), Rakoczy; P. Elizabeth (Scarborough, AU), Lai; Chooi-May (Waterford, AU), Chalberg, Jr.; Thomas W. (Redwood City, CA)
Assignee:	Avalanche Australia Pty Ltd. (Southbank, AU)
Application Number:	15/851,650
Patent Claims:	see list of patent claims

United States Patent 10,004,788 (US10004788): Critical claim analysis and US patent landscape for AAV-delivered anti-VEGF “functional cure” methods

US Patent 10,004,788 is directed to methods of treating ocular neovascularization (including wet age-related macular degeneration subfoveal CNV) using a single unit dose of an AAV vector encoding an anti-VEGF protein (expressly including aflibercept), administered to eyes that have already received prior intravitreal anti-VEGF. The central technical and legal pivot is that after AAV dosing, the subject does not require rescue anti-VEGF during a defined window (about 180 days to 1 year and in dependent claim sets also 90 days to 1 year), with vector genome dose ranges and rAAV capsid subtypes (AAV1 through AAV12; with narrower embodiments tied to AAV2/AAV8).

Below is a claim-by-claim critical interpretation of what is actually covered, where infringement traction will come from (and where it will fail), and how the patent landscape typically constrains or undermines this type of AAV anti-VEGF “treatment regimen” estate.

---

What does US 10,004,788 claim in practice: the “single-dose AAV after anti-VEGF” regimen?

Plain-English core:
A patient with ocular neovascularization who has already been stabilized with intravitreal anti-VEGF (first injection within ~1-30 days; dose-level sufficient to halt progression at the time of AAV dosing) receives one intravitreal or subretinal injection of an rAAV carrying a nucleic acid encoding an anti-VEGF protein (aflibercept or ranibizumab/anti-VEGF protein), at a single unit dose defined by vector genomes (min 1×10^6 to max 1×10^15; narrower dependent claim ranges 1×10^9 to 3×10^13), and then shows sufficient expression such that no additional “rescue” anti-VEGF is needed for a defined period.

Claim architecture that matters legally

1. Preconditioning requirement (prior anti-VEGF injection): the method is not “AAV alone.” It is a regimen combining prior intravitreal anti-VEGF plus a later AAV dose.
2. Rescue-treatment absence as the differentiator: “does not require rescue treatment … between about 180 days and about one year” (or 90 days to 1 year in dependent claims) is the functional outcome that distinguishes the claimed method from typical gene transfer approaches that still require later injections.
3. Expression durability measured at ≥1 year (claim 1): the method includes “elevated levels” measured at least one year after administration.
4. Single unit dose constraint: “a single unit dose” with specific vector genome ranges. This can be leveraged to argue non-infringement where dosing schedules differ.

Key definitional and proof pressure points

• “Sufficient to prevent progression at time of application”: requires clinical evidence that the prior anti-VEGF injection achieved control by the time of AAV administration.
• “Does not require rescue treatment”: framed as a patient management outcome. Litigation will fight about whether a subject “required” rescue based on physician criteria and protocol thresholds.
• Vector dose and route: dependent claims specify route (sub-retinal injection in claim 21) and intravitreal injection in claim 24.

---

Claim 1: What is the broadest method coverage and where are the infringement bottlenecks?

Claim 1 (broadest): ocular neovascularization; prior intravitreal anti-VEGF; then single unit dose AAV encoding anti-VEGF protein; vector genome range 1×10^6 to 1×10^15; elevated levels at least one year; and no rescue anti-VEGF between ~180 days and ~1 year.

Critical elements and how they constrain infringement

• Recombinant virus scope is not limited to AAV in claim 1 (it recites “recombinant virus” and only later dependent claims specify rAAV). If an accused therapy uses adenovirus, lentivirus, or other vectors, claim 1 is still potentially in play only if that vector is within “recombinant virus” (typically argued to include many viral vectors, but claim construction can narrow based on specification).
• Anti-VEGF protein is generic in claim 1 (not limited to aflibercept).
• Outcome window is central: “does not require rescue treatment … between about 180 days and about one year.” If a competitor’s program uses scheduled follow-up injections or defines “rescue” differently, proof becomes contested.

Most likely infringement scenario

A competitor’s regimen mirrors the same clinical choreography:

1. anti-VEGF stabilization by injection in the 1–30 day range,
2. single AAV dose with a vector genome count falling inside the claimed band,
3. long enough expression to avoid any rescue injection in that 180 day to 1 year window.

Most likely non-infringement scenario

• Multi-dose AAV administrations, split dosing, or retreatment with AAV breaks the “single unit dose” requirement.
• If the competitor’s clinical protocol mandates injections at 3–4 month intervals regardless of “rescue,” they will argue no “rescue treatment” was “required.” Conversely, plaintiff will frame any additional injection as rescue.
• If expression is measured and falls short of the asserted “elevated levels” by one year, defendants can challenge that element.

---

Dependent claims 2–10: How much is actually narrowed to rAAV capsids, vector dose bands, and anti-VEGF protein identity?

rAAV requirement (claim 2)

Claim 2 locks the recombinant virus to rAAV.

Legal effect: narrows any asserted theory for non-AAV vectors and forces focus on AAV capsid selection, dosing, and insertion cassette.

Capsid options (claim 3) vs specific capsids (claim 4)

• Claim 3 lists AAV1, AAV2, AAV2.5, AAV3–AAV12.
• Claim 4 limits to AAV2 and AAV8.

Infringement leverage:
If an accused product uses AAV5 or AAV9, it may evade the AAV2/AAV8 embodiment but could still fall under claim 3 if the asserted claim scope includes those capsids.

Vector genome dose band (claim 5)

• Claim 5: single unit dose comprises ≥1×10^9 and ≤3×10^13 vector genomes.

Infringement leverage:
This is a clean numeric line. If the competitor uses a dose outside that range, it can defeat at least the dependent claim set. Claim 1 has a wider band (up to 1×10^15), but claim 5 can still be used as a “fallback” for narrower infringement theories (or conversely, as a non-infringement argument).

Prior anti-VEGF timing (claim 6)

• Claim 6: prior first intravitreal injection between 1 and 30 days before AAV.

Proof pressure:
Clinically anchored. If a competitor schedules the AAV at 45–60 days after anti-VEGF loading, this element becomes harder to satisfy.

Anti-VEGF identity (claims 7–9)

• Claim 7 aflibercept
• Claim 8 ranibizumab
• Claim 9 anti-VEGF protein is aflibercept

Impact:
If an accused gene encodes another anti-VEGF (non-aflibercept, non-ranibizumab), plaintiffs pivot to claim 1 (generic anti-VEGF protein), unless the specification narrows identity through definitions.

Rescue window variant (claim 10)

• Claim 10: no rescue required between 90 days and 1 year.

Legal effect:
A tighter window increases the evidentiary burden. If competitor protocols show rescue in that 90–180 day zone, claim 10 is vulnerable.

---

Claim 11 and Claim 20: How does the ETDRS visual acuity improvement requirement affect patent enforceability?

• Claim 11: adds measuring BCVA via ETDRS letters and improvement of at least 1 line after administration.
• Claim 20: similar ETDRS improvement language in the subfoveal CNV/wet-AMD method.

Critical issue:
“Method of treatment” claims that include a measurable clinical endpoint can be harder to enforce because infringement can depend on patient outcomes rather than only on product attributes. In litigation, this typically forces plaintiffs into evidence-intensive proof of outcome in the accused population.

Defense leverage:
Competitors can argue variability in outcomes, differences in baseline BCVA, and statistical handling. They can also challenge whether a given trial’s endpoints correspond to “at least one line as measured by ETDRS letters.”

---

Claim 12–20: Subfoveal CNV method and the wet-AMD anchoring

Claim 12 is directed to subfoveal choroidal neovascularization secondary to wet-AMD, with prior anti-VEGF injections within about 30 days.

Structural differences from claim 1

• It narrows to subfoveal CNV and secondary etiology (wet-AMD).
• It requires one or more prior intravitreal anti-VEGF injections within ~30 days (not necessarily “first injection” or the same specific timing language used in claim 6 for claim 1’s dependency tree).
• Outcome is reduction of subfoveal CNV for at least one year, plus no rescue between ~180 days and 1 year.

Claims 13–15 and 16–18: AAV capsid options and dose band

These track the earlier structure:

• capsid group (claims 13 and 14)
• vector genome band (claim 15)
• anti-VEGF identity (aflibercept or ranibizumab; claim 16–18)

Claim 19: rescue window again

Claim 19 provides the 90 days to 1 year no-rescue variant.

---

Claims 21–27: How do route-specific and aflibercept-specific variants narrow the estate?

Claim 21: subretinal route + wet-AMD + AAV2/AAV8

Claim 21 tightens:

• wet-AMD secondary ocular neovascularization,
• prior first intravitreal anti-VEGF within 1–30 days,
• recombinant AAV is selected from AAV2 and AAV8,
• single unit dose administered via sub-retinal injection,
• dose band 1×10^9 to 3×10^13,
• at least 1 year reduction,
• no rescue between 180 days and 1 year.

Legal effect:
Route is an easy non-infringement lever. If an accused program uses purely intravitreal delivery, claim 21 is avoidable. If it uses subretinal, defendants then focus on capsid identity and vector genome dose.

Claim 24: intravitreal route + rAAV encoding aflibercept

Claim 24 tightens further:

• prior first anti-VEGF injection within 1–30 days,
• rAAV encoding aflibercept,
• intravitreal injection route,
• dose band 1×10^9 to 3×10^13,
• no rescue between 180 days and 1 year.

Defense posture:
If a competitor encodes aflibercept but uses different capsids (not limited here to AAV2/AAV8 in claim 24), or uses a different vector dose, claim 24 can still be vulnerable if claim 1 and other dependent claims cover broader capsids. But route and dose remain crisp.

Claim 26: prior aflibercept already halting progression

Claim 26 adds that the eye had received sufficient aflibercept so that progression at the time of AAV dosing has been halted.

Proof leverage:
This is a clinical state requirement, adding evidentiary burden and offering potential non-infringement if baseline progression status differs under competitor protocols.

Claim 27: 90 day no-rescue variant

Claim 27 expands the “no rescue” window to 90 days to 1 year, again increasing evidentiary difficulty.

---

How strong is the patent estate for this concept: what claim features likely survive novelty/obviousness scrutiny?

Even without importing the full prosecution history, the claim set is structured around a few technical differentiators that are typical focal points in anti-VEGF AAV litigation:

1. Regimen sequencing: prior intravitreal anti-VEGF followed by single AAV dosing.
2. Functional treatment outcome: no rescue anti-VEGF during a defined period.
3. Durability marker: elevated protein levels at ≥1 year.
4. Dose band limitation: numeric constraints on vector genomes.
5. Route specificity in narrower claims: intravitreal vs subretinal.
6. Named therapeutic protein: aflibercept (and ranibizumab in some dependent claims).

These features can help distinguish from earlier generic gene-therapy disclosures that did not teach (i) regimen sequencing to avoid injections, (ii) the specific no-rescue time window, and (iii) measurable ≥1-year expression.

However, the estate is also exposed to classic AAV patentability and enforceability pressures:

• Functional outcome claims can be attacked as obvious based on known AAV anti-VEGF expression studies combined with routine clinical design choices.
• Outcome-dependent enforcement increases litigation complexity and settlement pressure because infringement may turn on patient-level endpoints.

---

What patents likely compete in this landscape: how generic and alternative AAV anti-VEGF estates typically overlap?

US10004788’s claim theme sits inside a crowded conceptual space:

• AAV-delivered anti-VEGF proteins (aflibercept-like or ranibizumab-like) for ocular neovascular disease
• one-time or limited-dose gene delivery strategies intended to reduce injection burden
• combination strategies with anti-VEGF loading injections
• route-specific delivery (intravitreal vs subretinal)
• capsid selection and genome dose targeting

From a strategic landscape standpoint, the key commercial risk is not “whether someone can do AAV anti-VEGF,” but whether their clinical regimen and vector parameters can be positioned outside the claimed numeric and timing constraints.

In practice, competitors and generic challengers do not need to avoid “AAV anti-VEGF” in general. They need to avoid at least one required claim element:

• no prior anti-VEGF loading injection in the specified timeframe, or different timing
• different vector dose range outside the dependent claims
• different capsid not asserted under the fallback claims
• different route (subretinal vs intravitreal)
• rescue injections that occur inside the defined no-rescue windows, or physician-defined “rescue” that breaks the element
• different anti-VEGF protein identity if the plaintiff anchors on aflibercept or ranibizumab

---

What generic entry risks exist for US 10,004,788: does this patent block gene therapy commercialization?

Gene therapy does not map to Orange Book-style generic “label copy” entry the way small molecules do. The entry risk is typically via:

• IP design-around (capsid, dose, route, expression cassette, and regimen sequencing)
• license acquisition from patent holders
• Paragraph IV-style litigation analogs if FDA approval relies on pathways that still provoke patent disputes (mechanistically different from Hatch-Waxman)

Because US10004788 is directed to a method of treatment, gene-therapy market entrants still need to avoid practicing the method elements. If they market and treat using the claimed regimen, the patent can be asserted even if their vector is not “identical,” as long as the method steps and parameters fall within claim scope.

---

When does exclusivity “run out” for a US method patent like this?

The expiration timeline for US patents is governed by patent term rules (generally 20 years from the earliest effective non-provisional filing date, subject to adjustments). Without the patent’s filing data and any terminal disclaimer or patent term adjustment record, a precise exclusivity end date cannot be produced from the claim text alone.

---

Key takeaways

• US 10,004,788 is a regimen-and-outcome patent: it is not only about AAV anti-VEGF expression. It is about treating ocular neovascularization after anti-VEGF loading with a single AAV dose that yields no rescue anti-VEGF for a defined window.
• The most infringement-sensitive claim elements are: prior anti-VEGF timing, vector dose count, route (intravitreal vs subretinal in narrower claims), AAV capsid identity in the dependent set, and the “no rescue” window that can depend on physician protocol and patient outcomes.
• The most attackable points are typical for treatment-outcome claims: evidentiary burden of “does not require rescue,” variability in clinical endpoints (ETDRS), and obviousness arguments built from prior art AAV anti-VEGF expression plus standard clinical trial management.
• Design-around strategy is straightforward conceptually: shift route, dose band, timing, capsid selection, expression construct, or rescue/injection protocol behavior so at least one claim element is not met.

---

FAQs

1) Does US 10,004,788 cover AAV gene therapy without prior anti-VEGF injections?

No. The independent claim requires prior intravitreal anti-VEGF such that anti-VEGF is present at a concentration sufficient to prevent progression at the time of AAV administration.

2) Which claim elements most effectively support a non-infringement argument?

Different delivery route (intravitreal vs subretinal in dependent claims), vector genome dose outside the claimed range(s), omission or timing change of prior anti-VEGF loading, and rescue anti-VEGF that occurs within the no-rescue window.

3) Is the patent limited to aflibercept?

No for claim 1, but several dependent claims are expressly tied to aflibercept (including claims 9 and 24–27). Ranibizumab is also explicitly covered in dependent claims.

4) How does the ETDRS requirement change enforcement?

It adds a patient-outcome endpoint (BCVA improves by at least 1 line). That makes infringement proof more outcome-dependent in trials or patient cohorts aligned to the accused regimen.

5) Can a competitor evade by using a different AAV capsid?

Potentially. Claim 4 and claim 21 narrow to AAV2/AAV8, but claim 3 lists broader capsids. Design-around depends on whether plaintiff asserts broader or narrower dependent claims.

---

References

1. United States Patent 10,004,788. (claims provided in prompt).

Details for Patent 10,004,788

Applicant	Tradename	Biologic Ingredient	Dosage Form	BLA	Approval Date	Patent No.	Expiredate
Genentech, Inc.	LUCENTIS	ranibizumab	Injection	125156	June 30, 2006		2037-12-21
Genentech, Inc.	LUCENTIS	ranibizumab	Injection	125156	August 10, 2012		2037-12-21
Genentech, Inc.	LUCENTIS	ranibizumab	Injection	125156	October 13, 2016		2037-12-21
Genentech, Inc.	LUCENTIS	ranibizumab	Injection	125156	March 20, 2018		2037-12-21
Regeneron Pharmaceuticals, Inc.	EYLEA	aflibercept	Injection	125387	November 18, 2011		2037-12-21