MEMORANDUM
PRIVILEGED AND CONFIDENTIAL ATTORNEY WORK PRODUCT
This memorandum is protected by the attorney-client privilege and/or constitutes attorney work product prepared in anticipation of litigation or in connection with the provision of legal advice. It is intended solely for the use of the addressees identified below and should not be disclosed to any third party without the prior written consent of Calderwood & Harkness LLP.
FROM: Christopher Aiken, Lead Associate, and Amy Whitfield, Working Associate, Calderwood & Harkness LLP
TO: Joel Strauss, Responsible Partner, Calderwood & Harkness LLP; Marcus D. Pruitt, J.D., General Counsel, Thornwick Medical Research Institute, Inc.
DATE: February 18, 2026
RE: Analysis of intellectual Property Ownership and Publication Rights — Clinical Trial Agreement for Cobalt Ridge Health IT-3 Trial (ThornValve MV-200 System) between Thornwick Medical Research Institute, Inc. and Hargrove University Medical Center, Inc.
MATTER NO.: 1029-00001
Right-click to update Table of Contents
I. Executive Summary
The purpose of this memorandum is to analyze the contested intellectual property ownership and publication rights provisions in the proposed Clinical Trial Agreement (the "CTA") for the Cobalt Ridge Health IT-3 Trial and to recommend compromise positions for the upcoming negotiation session with Pemberton, Yates & Caldwell LLP ("PY&C"), outside counsel to Hargrove University Medical Center, Inc. ("HUMC"), currently scheduled for February 21, 2026.
Calderwood & Harkness LLP represents Thornwick Medical Research Institute, Inc. ("TMRI"), a Massachusetts Section 501(c)(3) nonprofit research institution, in its capacity as sponsor of the Cobalt Ridge Health IT-3 Trial. Cobalt Ridge Health IT-3 is a prospective, single-arm, open-label feasibility study of the ThornValve MV-200 System for transcatheter mitral valve repair, to be conducted at HUMC under the direction of Julian M. Castañeda, M.D., FACC, as Principal Investigator. The total value of the CTA is $6,000,000, funded entirely through TMRI's internal cardiovascular device development program allocation (Grant Award No. TMRI-CV-2025-018). The trial is registered on ClinicalTrials.gov under Identifier NCT06198234 and is being conducted under an Investigational Device Exemption issued by the U.S. Food and Drug Administration (IDE No. G260012).
TMRI's first draft of the CTA was circulated to PY&C on January 28, 2026. HUMC's first redline was received on February 7, 2026, transmitted by Rebecca S. Norcross, Partner at PY&C. A second redline reflecting additional revisions was received on February 19, 2026. A negotiation call specifically addressing the IP and publication terms is scheduled for February 21, 2026.
The parties are substantially aligned on financial terms, regulatory compliance obligations, trial procedures, subject enrollment criteria, and safety monitoring provisions. However, the parties remain in active dispute over three clusters of provisions that are the subject of this memorandum:
(1) Ownership of Trial Data. TMRI asserts sole ownership of all clinical, safety, imaging, and biomarker data generated in the performance of the Cobalt Ridge Health IT-3 Protocol. HUMC seeks institutional ownership of all data generated at its site, offering TMRI only a non-exclusive license for regulatory and research purposes.
(2) Foreground IP Licensing and Jointly Owned IP. TMRI seeks an exclusive, royalty-free, perpetual license to all foreground intellectual property conceived by HUMC or Principal Investigator personnel in the course of performing the Protocol. HUMC rejects both the exclusivity and royalty-free elements and proposes a non-exclusive license only. The parties also disagree on the right of first negotiation for jointly conceived inventions.
(3) Publication Review Periods and Editorial Control. TMRI proposes a 45-day initial review period with a 60-day extension for patent filing (105 days total). HUMC proposes a 30-day review period with a single 30-day extension (60 days total). The parties further disagree on the scope of TMRI's redaction rights and the allocation of editorial control over scientific content.
Summary of Recommendations. Based on the analysis set forth below, we recommend that TMRI adopt the following negotiation posture for the February 21 call:
• Trial Data Ownership: TMRI should maintain its position on sole ownership of Trial Data. This position is well-supported by the regulatory framework governing IDE-held devices, industry practice in investigator-initiated trials funded by nonprofit sponsors, and the need for consistency with the data ownership structures established in the Cobalt Ridge Health IT-1 and Cobalt Ridge Health IT-2 trials. As a meaningful concession, TMRI should offer HUMC and the Principal Investigator a non-exclusive, royalty-free, perpetual license to use Trial Data for academic, non-commercial research, teaching, institutional compliance, and scholarly publications, subject to the CTA's publication review provisions.
• Foreground IP Licensing: TMRI should maintain its position on exclusive licensing for foreground IP conceived by HUMC/PI personnel but offer to modify the royalty structure from "royalty-free" to "reasonable royalty upon commercial exploitation," with royalty terms to be negotiated in good faith at the time of commercialization and benchmarked to comparable academic institution license agreements. PI academic use rights should be expressly preserved as a retained carve-out.
• Right of First Negotiation: TMRI should retain the right of first negotiation for jointly owned IP but reduce the exclusive negotiation period from 120 days to 90 days and introduce a materiality threshold limiting the right to inventions directly related to the ThornValve MV-200 System or successor devices.
• Publication Review Period: TMRI should offer a compromise of 30 days for initial review plus a single 45-day extension (75 days total), available only upon written notice identifying patentable subject matter and confirming that patent application preparation is actively underway.
• Editorial Control: TMRI should concede that PI retains final editorial control over scientific content, methodology, and conclusions, while preserving TMRI's right to request removal of specifically identified confidential commercial information. TMRI should not seek the ability to suppress clinical data or scientific conclusions.
The target execution date for the CTA remains March 2, 2026. Timely resolution of the IP and publication terms is critical to support TMRI's IDE supplement filing target of March 15, 2026.
II. Background and Parties
A. Thornwick Medical Research Institute, Inc. (Sponsor)
Thornwick Medical Research Institute, Inc. ("TMRI") is a Massachusetts nonprofit corporation organized and existing under the laws of the Commonwealth of Massachusetts and recognized as a tax-exempt organization under Section 501(c)(3) of the Internal Revenue Code. TMRI's principal offices are located at 280 Longmeadow Drive, Cambridge, Massachusetts 02139. Its Employer Identification Number is 04-3871256.
TMRI was founded in 1987 and has established itself as a leading research institution specializing in structural heart disease therapies and cardiovascular medical device development. Its leadership team includes Dr. Ansel R. Varnham, M.D., Ph.D., who serves as President and Chief Executive Officer; Dr. Elaine K. Sutherland, M.D., who serves as Vice President of Clinical Development; and Marcus D. Pruitt, J.D., who serves as General Counsel. TMRI's annual research budget for fiscal year 2025 was approximately $74 million, drawn from a combination of internal reserves, institutional endowment income, licensing revenues, and non-federal grants.
The ThornValve MV-200 System is TMRI's lead investigational device. The MV-200 System is a transcatheter mitral valve repair device designed for patients with moderate-to-severe mitral regurgitation who are at prohibitive or high risk for conventional open-heart surgery. The device is currently under an Investigational Device Exemption from the U.S. Food and Drug Administration, IDE No. G260012, which authorizes TMRI to conduct clinical investigations of the device in human subjects. TMRI holds the IDE and bears full regulatory responsibility for the investigational program.
The Cobalt Ridge Health IT-3 Trial is the third in a series of investigator-initiated feasibility studies evaluating the ThornValve MV-200 System. Cobalt Ridge Health IT-1 was conducted at Massachusetts General Hospital (completed December 2023; 20 subjects) and Cobalt Ridge Health IT-2 was conducted at the Cleveland Clinic (completed September 2025; 30 subjects). Data from these prior trials demonstrated acceptable safety and preliminary efficacy, supporting expansion to a third feasibility trial site and providing the evidentiary basis for an anticipated future pivotal trial submission. Consistent data ownership by TMRI across the Cobalt Ridge Health IT program is essential for a coherent regulatory strategy.
The Cobalt Ridge Health IT-3 Trial is funded entirely through TMRI's internal resources under Grant Award No. TMRI-CV-2025-018, an allocation from TMRI's cardiovascular device development program. No federal funding—whether through grants, cooperative agreements, contracts, or any other mechanism—supports the Cobalt Ridge Health IT-3 Trial. This funding structure has significant implications for the Bayh-Dole Act analysis set forth in Section V of this memorandum.
B. Hargrove University Medical Center, Inc. (Institution)
Hargrove University Medical Center, Inc. ("HUMC") is a Connecticut nonprofit corporation with its principal address at 1400 Sterling Heights Boulevard, New Haven, Connecticut 06510. Its Employer Identification Number is 06-2194738. HUMC is a 726-bed academic medical center affiliated with Hargrove University, a private research university also located in New Haven. HUMC's President and Chief Executive Officer is Dr. Priya Anand-Kessler, M.D., MBA, and its Senior Vice President for Research Administration is Leonard B. Ostrowski, Ph.D.
HUMC's Structural Heart Center is among the highest-volume transcatheter valve intervention programs in the northeastern United States, performing approximately 650 transcatheter valve procedures annually. The Center's clinical expertise and patient volume were key factors in TMRI's selection of HUMC as the Cobalt Ridge Health IT-3 Trial site. HUMC's affiliation with Hargrove University creates a dual institutional structure—HUMC faculty members hold concurrent appointments at Hargrove University School of Medicine and are subject to the university's intellectual property policies, faculty handbook, and publication norms. This dual structure is directly relevant to the IP ownership and publication rights issues discussed in this memorandum.
HUMC is represented in the CTA negotiation by Pemberton, Yates & Caldwell LLP ("PY&C"), a nationally recognized law firm with offices in New Haven and New York. The lead negotiating partner at PY&C is Rebecca S. Norcross, Partner, who has extensive experience in academic medical center research contracting. We anticipate that Ms. Norcross will lead HUMC's negotiation team on the February 21 call.
C. Principal Investigator and Sub-Investigator
The Principal Investigator for the Cobalt Ridge Health IT-3 Trial is Dr. Julian M. Castañeda, M.D., FACC. Dr. Castañeda is Professor of Cardiovascular Medicine at Hargrove University School of Medicine and Director of the Structural Heart Interventions Program at HUMC. He has served as Principal Investigator on nine prior device trials and has authored or co-authored over 140 peer-reviewed publications in the fields of structural heart disease and interventional cardiology. Dr. Castañeda is widely regarded as a leading opinion leader in the transcatheter mitral valve intervention field, and his participation is important both for trial execution and for the credibility of the resulting data within the scientific and regulatory communities.
The Sub-Investigator is Dr. Hannah L. Forsberg, M.D., who serves as Associate Professor of Cardiovascular Medicine at Hargrove University School of Medicine and Associate Director of Structural Heart Interventions at HUMC. Dr. Forsberg has been designated as Co-PI designate and will perform study-related clinical procedures alongside Dr. Castañeda.
Both Dr. Castañeda and Dr. Forsberg are employees of HUMC and hold faculty appointments at Hargrove University. This dual employment creates overlapping institutional obligations relevant to intellectual property assignment. Inventions conceived by Dr. Castañeda or Dr. Forsberg in the course of their work on the Cobalt Ridge Health IT-3 Trial may be subject to claims of ownership by both HUMC (as their employer) and Hargrove University (under its faculty intellectual property policy). The CTA must be structured to account for these potentially competing claims, and the exclusive license approach advocated in this memorandum is designed in part to navigate this complexity by allowing HUMC/PI to retain nominal ownership of foreground IP while granting TMRI the commercial rights it requires.
D. CRO and DSMB
Data management, biostatistical analysis, and safety monitoring for the Cobalt Ridge Health IT-3 Trial are being performed by Meridian Clinical Sciences, Inc. ("Meridian"), a Delaware corporation with offices at 900 Weston Parkway, Suite 400, Cary, North Carolina 27513. Meridian has been separately engaged by TMRI under a master services agreement and is not a party to the CTA. The primary point of contact at Meridian is Victoria Reeves, Vice President of Cardiovascular Programs. Meridian's data management platform will serve as the central repository for Trial Data, and TMRI's recommended compromise on data access (discussed in Section IV.D below) contemplates that PI access to Trial Data will be provided through the Meridian platform.
The Data Safety Monitoring Board (DSMB) for the Cobalt Ridge Health IT-3 Trial was constituted in November 2025 under a separate charter. The DSMB Chair is Dr. Garrett P. Lindholm, M.D., Ph.D., Professor of Biostatistics and Cardiology at Ellsworth University in Chicago, Illinois. The DSMB operates independently of both TMRI and HUMC and reports directly to the TMRI Institutional Review Board and the FDA. The DSMB charter and composition are not at issue in the current CTA negotiation.
E. Trial Parameters
The full title of the trial is: cadence-3: A Prospective, Single-Arm, Open-Label Feasibility Study of the ThornValve MV-200 System for Transcatheter Mitral Valve Repair in Patients with Moderate-to-Severe Mitral Regurgitation. The trial is registered on ClinicalTrials.gov under Identifier NCT06198234 and is being conducted under IDE No. G260012.
Key trial parameters are as follows:
• Planned enrollment: 45 subjects
• Study design: Single-arm, open-label, single-site feasibility study
• Study site: HUMC Structural Heart Center, New Haven, CT
• Follow-up period: 24 months per subject
• Estimated study duration: 36 months from first enrollment
• IRB approval: December 18, 2025 (HUMC IRB Protocol No. HUMC-IRB-2025-0847)
• Estimated first patient enrollment: April 7, 2026
• Estimated last enrollment: June 30, 2027
• Estimated study completion: June 30, 2029
• IDE supplement filing target: March 15, 2026
• Target CTA execution date: March 2, 2026
The timeline underscores the urgency of resolving the outstanding IP and publication provisions. The IDE supplement filing target of March 15, 2026 is only 13 days after the target CTA execution date, and TMRI's regulatory counsel has advised that the executed CTA (or, at minimum, a substantially final draft) must be referenced in the IDE supplement submission.
F. Financial Overview
The total value of the CTA is $6,000,000, structured as follows:
Per-subject payments are structured in three installments: 40% ($38,000) payable upon enrollment, 30% ($28,500) payable upon completion of the 12-month follow-up visit, and 30% ($28,500) payable upon completion of the 24-month follow-up visit. The screen failure allowance provides $6,111.11 per screen failure, capped at $275,000 (approximately 45 screen failures).
All funding is sourced from TMRI Grant Award No. TMRI-CV-2025-018, an allocation from TMRI's cardiovascular device development program. As discussed in Section V below, the absence of federal funding is significant for the Bayh-Dole Act analysis and strengthens TMRI's position on IP ownership and licensing terms.
The $200,000 publication and manuscript preparation fund is not in dispute and reflects TMRI's commitment to supporting the dissemination of cadence-3 results through peer-reviewed publications and conference presentations. The existence of this fund is relevant to the publication rights analysis in Section VIII below, as it demonstrates that TMRI's interest is in controlling the timing and patent-protection aspects of publication, not in suppressing publication altogether.
III. Summary of Contested IP and Publication Rights Provisions
This section provides a detailed comparison of the IP and publication rights provisions as set forth in TMRI's current draft CTA (January 28, 2026) and HUMC's redlines (February 7, 2026, and February 19, 2026, transmitted by PY&C). The parties' positions are organized by subject matter to facilitate analysis in the sections that follow.
A. TMRI's Position (Current Draft — January 28, 2026)
1. Background IP
TMRI's draft provides that TMRI retains sole and exclusive ownership of all Background IP, which is defined in Article 1 of the CTA to include the ThornValve MV-200 System design, specifications, manufacturing processes, proprietary algorithms, engineering data, preclinical study results, and all other pre-existing intellectual property owned by or licensed to TMRI prior to the Effective Date of the CTA or developed by TMRI independently of the cadence-3 Protocol. No license to Background IP is granted to HUMC or PI except a limited, non-exclusive, non-transferable, non-sublicensable license to use the ThornValve MV-200 System and related documentation solely as necessary for the performance of the Protocol at the study site.
2. Trial Data Ownership
Under TMRI's draft, all Trial Data is defined broadly to encompass clinical data, safety data, adverse event reports, imaging data (including echocardiographic and angiographic images), biomarker results, patient-reported outcome measures, source documents, case report forms, and any other data generated in the performance of the CADENCE-3 Protocol. TMRI's draft provides that all Trial Data is owned solely and exclusively by TMRI.
TMRI's rationale for sole data ownership rests on three pillars: (i) TMRI is funding the trial entirely through its own internal resources, without any contribution from federal agencies, HUMC, or third-party funders; (ii) Trial Data is essential for TMRI's pending IDE supplement filing, future pivotal trial design, and eventual PMA submission to the FDA; and (iii) as the IDE holder, TMRI bears statutory and regulatory responsibility for the accuracy, integrity, and availability of all investigational data under 21 C.F.R. Part 812.
HUMC and PI receive, under TMRI's draft, a limited, non-exclusive, non-transferable license to access Trial Data solely for safety monitoring and regulatory compliance purposes during the study period. This license terminates upon the earlier of study completion or termination of the CTA.
3. Foreground IP
TMRI's draft addresses foreground intellectual property—that is, inventions, discoveries, improvements, or innovations conceived or first reduced to practice in the performance of the cadence-3 Protocol—through a three-tier allocation:
(a) Solely Conceived by TMRI Personnel. Inventions solely conceived by employees, agents, or contractors of TMRI are solely owned by TMRI, with no license or rights granted to HUMC or PI.
(b) Solely Conceived by HUMC/PI Personnel. Inventions solely conceived by employees, agents, or contractors of HUMC or by PI are solely owned by HUMC or PI (as applicable), subject to an exclusive, royalty-free, perpetual, worldwide license granted to TMRI for regulatory submissions, commercialization, sublicensing, and all other uses. Under this formulation, HUMC/PI would retain legal title to the invention but would have no practical ability to commercialize it independently, as the exclusive license would preclude HUMC/PI from granting any competing license to a third party.
(c) Jointly Conceived Inventions. Inventions jointly conceived by TMRI personnel and HUMC/PI personnel are jointly owned. TMRI receives a right of first negotiation for an exclusive commercial license, as described below.
4. Right of First Negotiation
For any jointly owned foreground IP, TMRI's draft provides a 120-day exclusive negotiation period during which the parties will negotiate in good faith the terms of an exclusive commercial license to TMRI. During the 120-day period, HUMC may not negotiate with or solicit offers from third parties for commercial licensing of the jointly owned IP. If the parties fail to reach agreement within the 120-day period, HUMC is free to license the jointly owned IP to third parties, but TMRI retains a non-exclusive, royalty-free license for regulatory purposes (including IDE supplement filings, PMA applications, and any required post-market reporting).
B. HUMC's Position (Per PY&C Redlines — February 7 and February 19, 2026)
1. Background IP
There is no material dispute regarding Background IP. HUMC's redlines accept TMRI's sole ownership of Background IP and the limited license for Protocol performance. PY&C's comments on this provision were confined to minor definitional clarifications, including a request to add "and related regulatory filings and correspondence" to the exclusion of TMRI Background IP from HUMC's access rights. We do not anticipate difficulty resolving these minor points and do not address them further in this memorandum.
2. Trial Data Ownership
HUMC's redlines represent a fundamental departure from TMRI's draft on data ownership. HUMC seeks ownership of all Trial Data generated at HUMC's site, including clinical data, safety data, imaging data, biomarker results, and all other data produced by HUMC personnel or using HUMC resources in the course of performing the cadence-3 Protocol. Under HUMC's proposed formulation, TMRI would receive a non-exclusive, royalty-free license to use Trial Data for regulatory and research purposes, but would not own the data.
HUMC's rationale, as articulated in PY&C's cover letter accompanying the February 7 redline, rests on the following arguments:
(a) As the institution where the data is generated and where the PI and study staff perform all clinical procedures, HUMC asserts institutional rights over site-generated data under its standard research policies.
(b) Hargrove University's faculty handbook (applicable to Dr. Castañeda and Dr. Forsberg as Hargrove faculty members) contains provisions that may require institutional ownership or, at minimum, institutional rights to data generated by faculty in the course of their university-related research activities.
(c) The PI retains the right to use Trial Data for academic, non-commercial research and teaching purposes without the need for TMRI's consent. HUMC views this right as inherent in the PI's role as an academic investigator and as essential to the PI's academic freedom.
In its February 19 second redline, PY&C reiterated these positions and added a new provision (Section 9.2(d) of the redline) providing that HUMC may use Trial Data for internal quality assurance, patient safety monitoring, accreditation, and institutional compliance purposes without TMRI approval.
3. Foreground IP
HUMC's redlines challenge both the exclusivity and the royalty-free elements of TMRI's proposed foreground IP licensing structure.
With respect to inventions solely conceived by HUMC or PI personnel, HUMC rejects the exclusive license and offers only a non-exclusive, royalty-free license to TMRI for regulatory submissions and internal research. HUMC's position is that a non-exclusive license adequately protects TMRI's regulatory needs while preserving HUMC's and PI's ability to independently commercialize PI-generated innovations or license them to third parties.
HUMC further rejects the royalty-free treatment of HUMC/PI-conceived foreground IP. In PY&C's February 7 cover letter, Ms. Norcross stated that if TMRI seeks commercial exploitation of any invention conceived by HUMC personnel or the PI, HUMC insists that licensing terms be negotiated at arm's length and include a reasonable royalty reflecting the fair market value of the invention.
With respect to jointly conceived inventions, HUMC insists that any licensing to third parties require the mutual written consent of both parties, rejecting TMRI's unilateral right of first negotiation. HUMC's position is that joint ownership should confer equal governance rights, and that neither party should be able to license or encumber the jointly owned IP without the other's affirmative agreement.
4. Right of First Negotiation
HUMC's redlines reject TMRI's proposed right of first negotiation in its entirety. In its place, HUMC proposes a general good-faith negotiation obligation under which the parties would discuss in good faith the potential terms of a commercial license if either party identifies a commercial opportunity for jointly owned IP. HUMC's proposed replacement contains no exclusivity commitment, no defined negotiation period, and no restriction on HUMC's ability to simultaneously negotiate with third parties.
C. Publication Rights — Summary of Positions
The following table summarizes the parties' competing positions on the key publication rights provisions:
Both parties acknowledge the importance of timely publication of cadence-3 results. The dispute centers on the duration of prepublication review, the scope of TMRI's ability to redact or modify manuscript content, and the degree of editorial authority retained by the PI over the final published product.
IV. Analysis of Trial Data Ownership
A. Legal Framework
1. Contractual Freedom
Under both Massachusetts and Connecticut law, intellectual property ownership in research collaborations is primarily governed by contract. There is no default statutory rule in either jurisdiction that automatically assigns data ownership to a trial site or to a study sponsor in the absence of an agreement between the parties.
Massachusetts courts have long recognized the principle that parties to a research collaboration are free to allocate intellectual property rights—including rights to data, inventions, and other work product—by contract, subject to the general requirements of contract formation (offer, acceptance, consideration, capacity, and lawful purpose). See, e.g., Boulanger v. Dunkin' Donuts Inc., 442 Mass. 635, 639 (2004) (reaffirming that contractual provisions governing IP rights are generally enforceable when supported by adequate consideration); Jet Spray Cooler, Inc. v. Crampton, 361 Mass. 835 (1972) (enforcing restrictive covenants and IP assignment clauses in employment and research contexts). The $6,000,000 in consideration flowing from TMRI to HUMC under the CTA provides more than adequate consideration to support an IP assignment or ownership provision.
Connecticut law similarly recognizes broad contractual freedom in the allocation of IP rights. Connecticut courts have upheld IP assignment provisions in research and development agreements, provided they are clearly expressed and supported by consideration. See Robert S. Weiss & Assocs., Inc. v. Wiederlight, 208 Conn. 525, 536-37 (1988) (enforcing contractual IP assignment clauses); see also Conn. Gen. Stat. § 42a-2-204 (recognizing formation of contracts by any manner sufficient to show agreement). There is no Connecticut statute that restricts the assignment or licensing of data rights by contract in the clinical trial context.
Accordingly, the ownership of Trial Data in the cadence-3 Trial will be determined by the terms of the CTA, not by any default rule of law. Both parties have the legal capacity to agree to either TMRI sole ownership or HUMC ownership, and the enforceability of the agreed-upon provision will not be in question so long as the CTA is properly executed.
2. FDA Regulatory Considerations
The regulatory framework governing investigational devices provides strong support for TMRI's position on Trial Data ownership. Under 21 C.F.R. Part 812, which governs the conduct of clinical investigations of devices, the sponsor of an IDE-approved investigation bears primary responsibility for the investigation and is subject to extensive regulatory obligations regarding the data generated in the course of the study.
Specifically, 21 C.F.R. § 812.40 requires the sponsor to ensure the proper monitoring of the investigation, including the accuracy and completeness of records and reports. Section 812.140(a) requires the sponsor to maintain complete and accurate records of the investigation, including all correspondence, data, and adverse event information. Section 812.150 requires the sponsor to submit various reports to the FDA, including IDE supplements, progress reports, and final reports, all of which must be supported by the underlying Trial Data.
Under 21 C.F.R. Part 11, which governs electronic records and electronic signatures, the sponsor must maintain adequate controls over electronic data systems to ensure data integrity, audit trails, and the ability to reproduce data for regulatory inspection.
TMRI, as the IDE holder (IDE No. G260012), must retain control over Trial Data to fulfill these regulatory obligations. The obligation to submit IDE supplements, annual reports, and any future Premarket Approval Application (PMA) requires that TMRI have unrestricted, unconditional access to all Trial Data, including the ability to analyze, compile, and submit the data to the FDA without the need for prior approval from any third party. If HUMC owns the Trial Data and TMRI possesses only a non-exclusive license, TMRI's ability to meet its regulatory obligations could be jeopardized in the event of a dispute between the parties—for example, if HUMC were to withhold access to data, impose conditions on its use, or delay provision of data in connection with a regulatory submission.
This concern is not hypothetical. The FDA has emphasized in guidance documents that sponsors of IDE investigations must maintain control over investigational data and that delegation of data management to third parties (whether CROs, clinical sites, or other entities) does not relieve the sponsor of its ultimate responsibility for data integrity and availability. See FDA Guidance for Industry and FDA Staff, Responsibilities of Sponsors of Clinical Investigations of Devices (2019). Any arrangement that places data ownership in the hands of the clinical site and limits the sponsor to a mere license creates a structural risk of the type the FDA's regulatory framework is designed to prevent.
We therefore recommend that TMRI maintain its position on sole data ownership. At an absolute minimum, TMRI must have unrestricted, irrevocable, and unconditional access to all Trial Data for regulatory purposes, an outcome that is most cleanly and reliably achieved through sole ownership rather than through a license that may be subject to dispute, revocation, or conditionality.
3. Industry Practice
Industry practice supports TMRI's position. In investigator-initiated trials funded by a nonprofit sponsor, data ownership by the sponsor is the predominant contractual arrangement, particularly where the sponsor is the IDE holder and intends to use the data for future pivotal trial support and eventual regulatory submission.
A review of our internal database of comparable clinical trial agreements (see Attachment B) reveals that in approximately 78% of investigator-initiated device trial CTAs in which the sponsor is the IDE holder and the trial is funded without federal support, the CTA provides for sole data ownership by the sponsor. The remaining 22% of CTAs in our database provide for shared ownership or site ownership with a broad license to the sponsor; however, in the majority of those cases, the clinical site is contributing its own funding or the trial involves a multi-site design in which data governance is shared among participating institutions.
HUMC's position that site-generated data belongs to the institution is more commonly seen in (i) industry-sponsored multi-site trials in which the pharmaceutical or device company serves as the sponsor but the clinical sites retain certain data rights as a condition of participation, and (ii) federally funded research governed by the Bayh-Dole Act or the Common Rule, in which institutional ownership of data is tied to federal funding requirements. Neither of these scenarios applies to the cadence-3 Trial, which is a single-site, non-federally-funded feasibility study sponsored by a nonprofit research institution.
B. TMRI's Interests and Risk Assessment
TMRI's need for sole data ownership is driven by four overlapping considerations:
First, TMRI requires sole ownership for its pending IDE supplement filing. The IDE supplement referencing the cadence-3 Protocol and the CTA must be filed with the FDA by March 15, 2026. TMRI must be able to represent to the FDA that it has unrestricted authority over the data that will be generated in the trial.
Second, TMRI will use cadence-3 data, together with data from cadence-1 and Cobalt Ridge Health IT-2, to design a future pivotal trial of the ThornValve MV-200 System. The pivotal trial design will depend on the feasibility data, and TMRI must be able to analyze, pool, and present the data without restrictions or the need for third-party consent.
Third, TMRI anticipates filing a PMA application with the FDA for the ThornValve MV-200 System, incorporating data from the entire cadence program. The PMA submission will require that TMRI have clear, documented ownership of all supporting data, and the FDA will expect TMRI to certify its authority over the data included in the application.
Fourth, TMRI may seek to out-license the ThornValve MV-200 System to a commercial partner. Any potential licensee or acquirer will conduct due diligence on TMRI's data rights, and ambiguity in data ownership—particularly the possibility that a clinical site could restrict TMRI's use of data—would represent a material risk that could reduce the value of a licensing transaction or prevent it from closing.
If HUMC owns the data and TMRI possesses only a non-exclusive license, TMRI faces the risk of delay (if HUMC does not promptly provide data for regulatory submissions), the risk of leverage (if HUMC uses its data ownership as a bargaining chip in future disputes), and the risk of loss of control over regulatory strategy (if HUMC's ownership rights allow it to independently submit or publish data in a manner inconsistent with TMRI's regulatory timeline).
Shared ownership of Trial Data would create similar, though somewhat attenuated, risks. Joint ownership introduces ambiguity regarding which party may submit data to the FDA, publish results, or grant third-party access, and would require the development of a detailed data governance framework that would add complexity and delay to the CTA negotiation.
C. HUMC's Interests and Legitimate Concerns
While we recommend that TMRI maintain its sole ownership position, we recognize that HUMC has legitimate interests that should be addressed through appropriate licensing and access provisions.
HUMC has a legitimate interest in ensuring that the PI and HUMC faculty can use Trial Data for academic purposes, including scholarly publications, conference presentations, and teaching. The PI's academic career depends on his ability to publish and present the results of clinical trials in which he serves as principal investigator. If TMRI's sole ownership is structured in a way that effectively prevents or unreasonably delays publication, HUMC's willingness to participate in the trial—and the willingness of other academic medical centers to participate in future TMRI trials—may be diminished.
HUMC also has legitimate institutional interests in data access for internal quality assurance, patient safety monitoring, accreditation compliance (e.g., Joint Commission requirements), and institutional research compliance obligations. These are routine administrative needs that do not conflict with TMRI's ownership of the data and can be accommodated through an appropriate license grant.
Finally, the PI's stature in the field—140 peer-reviewed publications, nine prior device trials as PI, and a leadership role in the structural heart disease community—means that any perceived restriction on the PI's academic freedom could have reputational consequences for TMRI within the academic and clinical community. A reputation for suppressing investigator publications or restricting academic data use would make it more difficult for TMRI to recruit leading investigators for future trials.
D. Recommended Compromise on Trial Data Ownership
Based on the foregoing analysis, we recommend the following compromise position for the February 21 negotiation:
(1) TMRI maintains sole and exclusive ownership of all Trial Data, as defined in TMRI's January 28 draft.
(2) HUMC and PI receive a non-exclusive, royalty-free, perpetual license to use Trial Data for the following purposes:
(a) Academic, non-commercial research conducted by PI, Sub-Investigator, or other HUMC faculty members under the direction or supervision of PI;
(b) Teaching, including the use of Trial Data (in de-identified form) in lectures, seminars, and training programs at HUMC and Hargrove University;
(c) Internal quality assurance, patient safety monitoring, accreditation compliance, and institutional regulatory compliance; and
(d) Scholarly publications and conference presentations, subject to the publication review provisions of the CTA (as discussed in Section VIII below).
(3) The license is non-transferable: HUMC and PI may not grant sublicenses or share raw Trial Data with third-party institutions, investigators, or entities without TMRI's prior written consent. De-identified, aggregated data included in published manuscripts or conference presentations is not subject to this restriction (as such data will already have been reviewed through the publication review process).
(4) TMRI commits to providing PI with reasonable and timely access to Trial Data in usable electronic formats through the Meridian Clinical Sciences data management platform. Access will be provided on a rolling basis as data is collected, cleaned, and locked, and PI will have access to the final locked database upon study completion.
This compromise preserves TMRI's essential ownership rights while addressing HUMC's legitimate academic and institutional interests. It provides the PI with the access and use rights necessary for his academic work while maintaining TMRI's control over the data for regulatory and commercial purposes.
V. Bayh-Dole Act Applicability Analysis
A. Statutory Framework
The Bayh-Dole Act, codified at 35 U.S.C. §§ 200-212, and its implementing regulations at 37 C.F.R. Part 401, establish the legal framework governing ownership of inventions arising from federally funded research. The Act was enacted in 1980 to promote the utilization of inventions arising from federally funded research by allowing universities, small businesses, and nonprofit organizations to elect title to "subject inventions"—defined as inventions conceived or first actually reduced to practice in the performance of work under a "funding agreement" with a federal agency. 35 U.S.C. § 201(e).
Under Bayh-Dole, when a contractor (such as a university or nonprofit) elects title to a subject invention, the federal government retains certain rights, including a non-exclusive, nontransferable, irrevocable, paid-up license to practice the invention throughout the world on behalf of the United States, 35 U.S.C. § 202(c)(4), and "march-in rights" permitting the federal agency to require licensing of the invention to third parties under specified circumstances, 35 U.S.C. § 203.
The critical threshold for Bayh-Dole applicability is the presence of federal funding. The Act applies only when the research in question is funded "in whole or in part" by a federal agency under a "funding agreement," defined as "any contract, grant, or cooperative agreement entered into between any Federal agency . . . and any contractor for the performance of experimental, developmental, or research work funded in whole or in part by the Federal Government." 35 U.S.C. § 201(b). If no federal funding is involved, Bayh-Dole does not apply, and the parties are free to allocate intellectual property rights by private contract without regard to the Act's requirements.
B. Application to the cadence-3 Trial
The Cobalt Ridge Health IT-3 Trial is funded entirely through TMRI's internal resources under Grant Award No. TMRI-CV-2025-018, an allocation from TMRI's cardiovascular device development program. This allocation derives from TMRI's general operating budget, which includes institutional endowment income, licensing revenues from TMRI's patent portfolio, and non-federal grants and donations. The TMRI-CV-2025-018 allocation is not a pass-through of federal funds and is not subject to any federal funding agreement.
No federal grant, cooperative agreement, or contract funds the cadence-3 Trial. Neither the National Institutes of Health (NIH), the National Science Foundation (NSF), the Department of Defense (DoD), nor any other federal agency has provided funding for the trial. The $6,000,000 in CTA payments to HUMC is sourced entirely from TMRI's internal allocation.
The Investigational Device Exemption issued by the FDA (IDE No. G260012) is a regulatory authorization permitting TMRI to conduct clinical investigations of the ThornValve MV-200 System in human subjects. The IDE is not a funding mechanism and does not constitute a "funding agreement" within the meaning of 35 U.S.C. § 201(b). Receipt of an IDE does not trigger Bayh-Dole obligations.
TMRI's annual research budget of approximately $74 million includes funds from various sources. However, the specific allocation funding the cadence-3 Trial—TMRI-CV-2025-018—is an internal, non-federal allocation. Even if other portions of TMRI's budget include federal funding (e.g., NIH grants supporting other research programs), commingling of funds at the institutional level does not, standing alone, render the cadence-3 Trial "federally funded" for Bayh-Dole purposes. Bayh-Dole applies to inventions arising from specific federally funded research projects, not to all inventions produced by an institution that happens to receive some federal funding. See Stanford University v. Roche Molecular Systems, Inc., 563 U.S. 776, 786-87 (2011) (clarifying that Bayh-Dole vests ownership in the contractor only for inventions arising from federally funded work).
C. Conclusion on Bayh-Dole Applicability
Based on the foregoing analysis, we conclude that the Bayh-Dole Act has limited applicability to the cadence-3 Trial. The Act does not apply to inventions arising from the CTA because the trial is not funded by a federal agency.
Caveat: There is a potential complication if the PI's or Sub-Investigator's salary at HUMC is partially supported by federal awards—for example, NIH training grants, NIH career development awards (K-series), NIH research project grants (R-series), or other federal funding mechanisms—and if a portion of the PI's or Sub-Investigator's effort on the cadence-3 Trial is charged to or otherwise supported by such federal awards. In that scenario, there is a theoretical argument that Bayh-Dole could reach certain inventions conceived by the PI in the course of effort that is partially supported by federal funds, even if the cadence-3 Trial itself is not federally funded.
This argument has not been definitively resolved by the courts and is fact-specific. The strength of the argument depends on factors such as whether the PI's effort on the cadence-3 Trial is separately accounted for, whether the PI's federal salary support is general (covering all of the PI's research activities) or specific (covering only particular projects), and whether the invention in question was conceived in the course of the federally supported effort or the cadence-3 work.
We recommend that TMRI raise this issue during the February 21 negotiation and request HUMC to confirm that no federal funding supports any portion of the PI's or Sub-Investigator's effort on the CADENCE-3 Trial. If HUMC confirms the absence of federal funding involvement, that confirmation should be memorialized as a representation in the CTA, providing TMRI with contractual assurance that the Bayh-Dole framework does not apply.
If it is confirmed that no federal funding is involved, TMRI can contract freely for IP ownership and licensing without Bayh-Dole constraints. The parties' rights will be governed entirely by the CTA's IP provisions, and neither party will be able to invoke Bayh-Dole as a basis for overriding the contractual allocation of IP rights.
Even if Bayh-Dole were applicable—for example, due to ancillary federal salary support for the PI—its impact on the CTA negotiation would be limited. Bayh-Dole applies to "subject inventions," defined as inventions "conceived or first actually reduced to practice in the performance of work under a funding agreement." 35 U.S.C. § 201(e). Clinical data generated in the course of a trial is generally not considered an "invention" under the Act, as data does not meet the statutory definition of a patentable invention (requiring novelty, utility, and non-obviousness). Accordingly, Bayh-Dole would primarily affect the ownership and licensing of patentable inventions conceived by the PI or Sub-Investigator in the course of the trial, not the ownership of Trial Data. TMRI's sole ownership of Trial Data would remain unaffected by Bayh-Dole even if the Act were found to apply to certain inventions arising from the trial.
D. HUMC's Potential Bayh-Dole Argument
PY&C has not explicitly raised the Bayh-Dole Act in either its February 7 or February 19 redlines. However, we anticipate that HUMC may invoke Hargrove University's intellectual property policy during the February 21 negotiation. University IP policies at major research institutions frequently mirror Bayh-Dole principles and assert institutional ownership over inventions by faculty, regardless of whether the specific research project is federally funded. Hargrove University's publicly available IP policy (excerpted in Attachment D) provides that the university retains ownership of inventions "conceived or first reduced to practice by university faculty, staff, or students using university resources, facilities, or in the course of their university-related activities," with limited exceptions for inventions made entirely on the faculty member's own time without university resources.
If PY&C invokes this policy, we recommend that TMRI take the following approach:
(a) Acknowledge that the CTA must be consistent with HUMC's institutional IP policies and that TMRI does not seek to override Hargrove University's faculty IP policy.
(b) Clarify that Bayh-Dole itself does not mandate HUMC's requested ownership structure for this privately funded trial. Bayh-Dole is a federal statute that applies to federally funded research; Hargrove University's IP policy is a private institutional policy that may or may not apply depending on the specific circumstances of the PI's employment and research activities.
(c) Propose the exclusive license structure discussed in Section VI.E below as a solution that is compatible with Hargrove University's IP policy (HUMC/PI retains ownership; TMRI receives an exclusive license) while preserving TMRI's commercial interests.
This approach avoids a confrontation over the applicability of Bayh-Dole—which is unnecessary given the absence of federal funding—while acknowledging HUMC's legitimate institutional policy concerns and steering the negotiation toward the practical question of license terms.
VI. Analysis of Foreground IP Ownership and Licensing
A. Massachusetts IP Assignment Law
Massachusetts law provides a generally favorable framework for IP assignment and licensing. The Commonwealth's principal employee IP protection statute, Mass. Gen. Laws ch. 149, § 19A, restricts certain employer obligations requiring employees to assign inventions made entirely on the employee's own time, without use of the employer's equipment, supplies, facilities, or trade secret information, unless the invention relates to the employer's business or the employee's work. Section 19A is designed to protect employees from overbroad IP assignment obligations in the employment context.
However, § 19A is inapplicable to the CADENCE-3 CTA for two reasons. First, the statute governs employer-employee relationships, and the CTA is a contract between two institutional entities (TMRI and HUMC), not an employment agreement. TMRI is not the employer of Dr. Castañeda, Dr. Forsberg, or any HUMC personnel. Second, even if the statute were applicable by analogy, the inventions at issue—those conceived in the performance of the CADENCE-3 Protocol using TMRI-funded resources and the TMRI-owned ThornValve MV-200 System—would fall squarely within the exception for inventions relating to the employer's (or, in this case, the sponsor's) business.
Under Massachusetts contract law more generally, IP assignment and licensing terms are enforceable when supported by adequate consideration, clearly expressed, and not unconscionable or contrary to public policy. See Analogic Corp. v. Data Translation, Inc., 358 Mass. 791 (1971); Augat, Inc. v. Aegis, Inc., 409 Mass. 165, 172 (1991). The $6,000,000 in CTA consideration and the mutual covenants exchanged by the parties provide ample consideration to support TMRI's requested IP provisions, including the exclusive license for foreground IP.
B. Connecticut IP Assignment Law
Connecticut does not have a statute equivalent to Massachusetts § 19A restricting IP assignments in the employment or research context. Connecticut courts have generally enforced contractual IP assignment and licensing provisions in accordance with the parties' agreement, subject to standard contract law principles. See Robert S. Weiss & Assocs., Inc. v. Wiederlight, 208 Conn. 525, 536-37 (1988); Advanced Financial Solutions, Inc. v. Brade, 2007 WL 4462754 (Conn. Super. Ct. 2007).
The more significant issue under Connecticut law is the potential application of Hargrove University's faculty IP policy. As discussed in Section V.D above, Hargrove University's IP policy asserts institutional ownership over inventions created by faculty using university resources or within the scope of university-related activities. If Dr. Castañeda or Dr. Forsberg conceives an invention in the course of the CADENCE-3 Trial, Hargrove University's IP policy may require that the university (or HUMC, as the university's medical center affiliate) retain ownership of the invention. The CTA cannot override a pre-existing obligation between the PI and the university, and any attempt to do so would create a conflict that could render the CTA provision unenforceable as against the university.
Accordingly, the CTA must be structured to be compatible with HUMC's existing IP policies. If HUMC's IP policy requires institutional ownership of faculty inventions, the exclusive license approach is the practical maximum that TMRI can achieve with respect to PI-conceived foreground IP. Under this approach, HUMC/PI retains legal ownership of the invention (consistent with Hargrove University's IP policy), and TMRI obtains the exclusive commercial rights it needs through a license agreement. This structure avoids any conflict with the university's IP policy while providing TMRI with substantially all of the commercial benefit of ownership.
C. Evaluation of TMRI's Position: Exclusive Royalty-Free License
TMRI's current draft requests an exclusive, royalty-free, perpetual, worldwide license to all foreground IP solely conceived by HUMC or PI personnel in the performance of the CADENCE-3 Protocol. The license would cover regulatory submissions, commercialization, manufacturing, sublicensing, and all other uses.
Strengths of TMRI's position:
• The exclusive license provides TMRI with maximum commercial flexibility, including the ability to sublicense to a commercial partner without the need for HUMC/PI consent.
• An exclusive license is essential for TMRI's regulatory submissions under IDE G260012. If PI conceives an invention that is embodied in or necessary for the ThornValve MV-200 System, TMRI must have the right to practice and exploit that invention in the regulatory context without risk of infringement claims or licensing disputes.
• TMRI is the entity funding the trial at a cost of $6,000,000 and providing the investigational device. Any inventions conceived in the course of the trial will necessarily build upon TMRI's Background IP and the ThornValve MV-200 System. TMRI's contribution to the inventive process—through funding, device provision, protocol design, and regulatory sponsorship—justifies an exclusive license.
Risks of TMRI's position:
• HUMC has strongly objected to both the "royalty-free" element and the "exclusive" nature of the license. PY&C's cover letters indicate that HUMC views the royalty-free element as inequitable and the exclusivity as unduly restrictive of PI's academic and commercial freedom.
• The "royalty-free" element is the most vulnerable aspect of TMRI's position. In arm's-length negotiations between a nonprofit sponsor and an academic medical center, insisting on royalty-free treatment of PI-conceived inventions—which may have independent commercial value—is unusual and may be perceived as overreaching. Academic medical centers and universities typically expect to share in the commercial upside of inventions generated by their faculty, and HUMC's position on this point reflects well-established institutional norms in the academic research community.
• If TMRI maintains the royalty-free position, it risks a prolonged negotiation that could delay CTA execution beyond the March 2 target date and jeopardize the March 15 IDE supplement filing.
D. Evaluation of HUMC's Position: Non-Exclusive License Only
HUMC proposes granting TMRI only a non-exclusive, royalty-free license for regulatory submissions and internal research, while retaining HUMC's and PI's freedom to license foreground IP to third parties, including TMRI's competitors.
This position is unacceptable from TMRI's perspective for three reasons:
(1) A non-exclusive license would allow HUMC/PI to license foreground IP to TMRI's competitors, potentially enabling a competing company to develop a product based on an invention derived from the ThornValve MV-200 System. This would undermine TMRI's competitive position and commercial strategy.
(2) Any PI-conceived invention arising from the CADENCE-3 Trial is likely to be closely related to, or derivative of, the ThornValve MV-200 System. The PI's inventive contribution will build upon TMRI's Background IP, and the resulting foreground IP will have value primarily because of its relationship to the TMRI platform. Allowing HUMC/PI to license such inventions to third parties without TMRI's control would effectively allow competitors to benefit from TMRI's investment in the ThornValve program.
(3) TMRI needs freedom to operate for its PMA submission and any subsequent commercial licensing. If HUMC retains the ability to grant competing licenses, TMRI cannot guarantee to the FDA or to a potential commercial partner that it has the unencumbered right to practice all IP necessary for the ThornValve MV-200 System.
E. Recommended Compromise on Foreground IP
We recommend the following compromise position, which is designed to address HUMC's legitimate concerns while preserving TMRI's essential commercial interests:
Solely Conceived HUMC/PI IP:
HUMC and/or PI retain ownership of inventions solely conceived by HUMC or PI personnel in the performance of the cadence-3 Protocol. TMRI receives an exclusive license to such inventions for all uses related to the ThornValve MV-200 System, including regulatory submissions (IDE supplements, PMA applications, post-market reports), manufacturing, commercialization, and sublicensing. The license is worldwide, perpetual, and irrevocable.
The key modification from TMRI's current draft is the introduction of a reasonable royalty payable upon commercial exploitation. Rather than insisting on a royalty-free license, TMRI would agree that if and when TMRI (or a sublicensee) generates commercial revenue from a product incorporating a HUMC/PI-conceived invention, TMRI will pay HUMC a reasonable royalty. The royalty terms would be negotiated in good faith at the time of commercialization, benchmarked to comparable academic institution license agreements in the medical device field. This approach defers the royalty negotiation to the point at which the commercial value of the invention can be assessed, rather than requiring the parties to negotiate royalty terms now, when the nature and value of any future inventions are unknown.
The exclusive license would not apply to HUMC/PI's use of the foreground IP for academic, non-commercial research and teaching purposes (see Section VI.F below).
Solely Conceived TMRI IP:
TMRI retains sole ownership of inventions solely conceived by TMRI personnel. No change from TMRI's current draft. No license or rights are granted to HUMC or PI with respect to TMRI sole inventions.
Jointly Conceived IP:
Joint ownership, with each party having the right to practice and use the jointly conceived invention for internal, non-commercial, academic, and regulatory purposes without accounting to the other party. This is a partial concession to HUMC, which objected to any restriction on HUMC's ability to use jointly owned IP for non-commercial purposes.
For commercial licensing to third parties, mutual written consent of both parties is required. This is a concession from TMRI's current draft, which would allow TMRI to proceed unilaterally under the right of first negotiation. We recommend this concession because it addresses HUMC's core objection regarding governance of jointly owned IP while still providing TMRI with a practical veto over third-party licensing that could harm TMRI's competitive interests.