University Technology Transfer

The Bayh-Dole Act of 1980 (35 U.S.C. §§ 200–212) revolutionized technology transfer from federal research by allowing universities, small businesses, and nonprofits to elect title to inventions made using federal funding: PRE-BAYH-DOLE PROBLEM: before 1980, the federal government retained title to inventions made with federal research grants; result: as of 1980, the government held title to approximately 28,000 patents and had licensed fewer than 5% of them; most federally funded inventions sat in government files never reaching commercial use; BAYH-DOLE SOLUTION: allowed universities and small businesses to elect title to federally funded inventions; created framework that incentivized universities to build technology transfer offices and actively license inventions; result: Stanford; MIT; Caltech; University of California; Johns Hopkins built licensing programs generating hundreds of millions annually; annual AUTM survey shows ~$3.9 billion in licensing income from US universities (2022); WHEN BAYH-DOLE APPLIES: applies to inventions that were: conceived or first actually reduced to practice using federal funding; made by any employee or student of the university while working under a federal grant; even partial federal funding can trigger Bayh-Dole if the invention used any federal resources; KEY UNIVERSITY OBLIGATIONS WHEN ELECTING TITLE: DISCLOSURE: disclose each invention to the federal funding agency within a reasonable time after the inventor discloses to the university; ELECTION: elect title to the invention (notify funding agency that university wants title) within 2 years of disclosure to the agency or within 60 days of filing a patent application; FILE PATENT: file a patent application within 1 year of electing title (or before a statutory bar); preferably within 10 months of election; DOMESTIC PREFERENCE: any exclusive license must include a requirement for US manufacture of products embodying the invention unless a waiver is obtained; REPORTING: report utilization of each licensed invention annually to the funding agency; GOVERNMENT RIGHTS: government always retains a royalty-free, nonexclusive, irrevocable license to practice (NOT commercialize) any Bayh-Dole patent for or on behalf of the government; MARCH-IN RIGHTS: funding agency can require university to grant licenses to third parties in four specific circumstances — this is the government's most powerful Bayh-Dole right; SMALL BUSINESS PREFERENCE: exclusive licenses should preference US small businesses where practical.

Technology transfer offices (TTOs) at major research universities have developed sophisticated licensing practices that affect faculty inventors, graduate students, and startup companies: THE INVENTION DISCLOSURE PROCESS: faculty or student inventor discovers a potentially patentable innovation; must disclose to TTO (most university IP policies require all inventions made using university resources to be disclosed; failure can breach employment agreement); TTO receives disclosure and evaluates: commercial potential; patentability (prior art search); costs of patent filing vs. expected licensing revenue; whether industry partner exists; TTO DECISION ON FILING: TTO may decline to file (estimated commercialization value too low); TTO elects title and files patent; inventor retains right to publish (typically after patent filed or provisional filed); some TTOs allow inventors to independently commercialize if TTO declines; INVENTOR REVENUE SHARING: most US research universities pay inventors a percentage of net licensing income (after patent costs); TYPICAL STRUCTURES: 1/3-1/3-1/3 model (Stanford; similar institutions): 1/3 to inventor(s); 1/3 to inventor's department; 1/3 to inventor's school/college; MIT model: 1/3 to inventor; remainder split between department/lab and central MIT; University of California: 35% to inventor(s); 15% to department/ORU; 50% to UC campus; HIGH-INCOME EXAMPLES: Stanford Google PageRank (US6,285,999): Sergey Brin and Larry Page assigned invention to Stanford; Stanford licensed to Google; received $336M in Google stock; one of the largest TTO payouts in history; Gatorade (University of Florida 1965): research funded by university; licensing produced hundreds of millions; UF receives 20% of royalties; inventor heirs receive proceeds; Cisplatin; Lyrica; Taxol — major pharmaceutical licensing programs; EXCLUSIVE vs. NONEXCLUSIVE UNIVERSITY LICENSES: EXCLUSIVE: one company gets right to develop; higher upfront + royalty; appropriate when significant development investment required; risk: if company fails to develop, patent value lost; march-in rights and diligence milestones protect against this; NONEXCLUSIVE: multiple companies can license; lower per-license revenue; appropriate for enabling technologies; no development required; SPINOUT vs. ESTABLISHED COMPANY: spinout license (to faculty startup): often lower royalty rates (2-5%); equity stake in startup (1-10%); development milestones; company typically gets exclusive license; license to established company: higher royalty; larger upfront; less risk for university.

The CRISPR patent dispute is the most consequential academic patent dispute in modern history, involving billions of dollars in potential licensing revenue, multiple Nobel Prize winners, and fundamental questions about who invented one of the most transformative biotechnologies ever discovered: THE COMPETING INVENTIONS: UC BERKELEY / JENNIFER DOUDNA: Jennifer Doudna (UC Berkeley) and Emmanuelle Charpentier (then at University of Vienna; later Max Planck) published their foundational CRISPR-Cas9 paper in Science (June 2012) describing: how Cas9 protein can be programmed with guide RNA to cut DNA at a specific location in a test tube; they proved the concept using purified components; patent filing: UC Berkeley filed patent application claiming CRISPR-Cas9 as a genome editing tool (broadly, in any cell type); BROAD INSTITUTE / FENG ZHANG: Feng Zhang at the Broad Institute (MIT/Harvard) filed a patent application in December 2012 claiming CRISPR-Cas9 specifically for editing in eukaryotic cells (including human cells); Zhang paid for expedited examination and received a patent in April 2014; Broad argued: adapting CRISPR for eukaryotic cells required significant inventive steps beyond the Doudna/Charpentier work; the leap from prokaryotic biochemistry to human cell editing was not obvious; THE INTERFERENCE/DERIVATION PROCEEDING: UC Berkeley argued priority: Doudna/Charpentier invented CRISPR editing; Zhang's application was obvious in light of their work; PTAB INTERFERENCE (2016-2017): PTAB conducted an interference proceeding to determine priority; PTAB ruled (February 2017): the two inventions were patentably DISTINCT; the Broad's claims (eukaryotic editing) and UC's claims (general CRISPR editing) did not interfere because one was not obvious from the other; Broad retained its eukaryotic editing patents; APPEAL AND FURTHER PROCEEDINGS: Federal Circuit affirmed PTAB (2018): upheld finding of no interference; UC Berkeley filed new patent applications; UC eventually received patents on CRISPR in various cell types; subsequent PGR proceedings; COMMERCIAL IMPACT: Broad/Zhang licensees: Editas Medicine (NASDAQ:EDIT); Broad licensed multiple biotech companies; UC Berkeley/Charpentier licensees: Caribou Biosciences (Doudna spinout); CRISPR Therapeutics; Intellia Therapeutics; both licensing programs generate significant royalties; NOBEL PRIZE (2020): Jennifer Doudna and Emmanuelle Charpentier received the Nobel Prize in Chemistry for discovering CRISPR-Cas9 — before the patent dispute was fully resolved; Feng Zhang was notably not included; LESSON: invention and patent priority are separate questions; the Nobel Committee and the USPTO/PTAB can reach different conclusions about who deserves credit for a scientific breakthrough.

Licensing technology from a university TTO requires navigating a distinctive process that differs significantly from commercial licensing negotiations: UNDERSTANDING THE TTO'S CONSTRAINTS: TTOs operate under Bayh-Dole obligations that constrain what they can agree to: must retain US manufacture preference in exclusive licenses; must report utilization; must not license in a way that undermines the government's royalty-free license; TTOs are often risk-averse and slow (government-like institution); faculty inventor relationship management (inventor may be founding the spinout or may be at a competing company); TTO staff may not be experienced commercial negotiators; EVALUATING THE TECHNOLOGY PACKAGE: before engaging TTO, assess: what is actually being licensed? (provisional applications vs. issued patents vs. know-how); how many inventors? (multiple inventors = multiple institutions if at different schools); is the technology funded by multiple agencies? (NSF + NIH = multiple sets of Bayh-Dole obligations); is there existing art or prior licenses that limit exclusivity?; NEGOTIATION STRATEGY WITH TTOs: FIELD OF USE LIMITATION: negotiate the narrowest exclusive field of use you actually need; broader fields of use = higher royalties + higher development obligations; leaving TTOs room to license other companies in other fields makes deal more feasible; TERRITORY: US-only often negotiable at lower royalty; worldwide exclusive = significant premium + US manufacture obligation; DILIGENCE MILESTONES: TTOs will require development milestones (FDA filing; commercial launch; minimum sales) to prevent warehouse effect; negotiate realistic timelines based on actual development schedule; ROYALTY STRUCTURES IN UNIVERSITY LICENSES: RUNNING ROYALTY: 1-5% of net sales typical for software/tech; 3-8% for biotech/pharma; UPFRONT LICENSE FEE: $10,000-$100,000 range for typical tech startups; MILESTONE PAYMENTS: IND filing; Phase I; Phase II; Phase III; FDA approval typical in pharma; MINIMUM ANNUAL ROYALTIES: escalating after first commercial sale; EQUITY STAKE: many TTOs now take equity in startup spinouts (1-5% common; some take warrants instead); SUBLICENSE REVENUE SHARING: 20-40% of sublicense income to university; COMMON PITFALLS: failing to negotiate publication rights clearly (faculty want to publish; company wants delay); not addressing background IP ownership (what IP did researcher have before the funded project?); not obtaining reps/warranties about ownership (TTO may discover post-signing that a collaborator has rights); ignoring government march-in risk for high-priced commercial products (NIH has received drug pricing march-in petitions for Ritonavir; Xtandi; COVID antivirals; none granted as of 2024 but political pressure continues).