While creating opportunities, the Regulations present significant compliance challenges: What constitutes “sufficient” non-clinical data? How can companies verify that partner sites meet qualification requirements? What happens when sites are downgraded or key researchers leave? How should intellectual property rights for improved technologies be structured? How can sites’ publication needs be balanced with patent protection? What disputes arise from “jointly owned” data provisions? How can informed consent achieve substantive understanding beyond formal compliance? How should causal relationships be determined for adverse events? How should dual regulatory requirements for medical devices be coordinated? How should civilian and military approval systems be integrated?

Drawing on biomedical compliance experience, this article systematically examines these operational challenges, providing practical guidance for companies.

I. Filing Material Standards

Articles 15-16 require clinical research sites to file with the National Health Commission within 5 working days after passing academic and ethical review. Filing materials include: sponsor and site information, researcher credentials, non-clinical research reports, clinical protocols, risk mitigation plans, review opinions, informed consent templates, and funding documentation. This seemingly straightforward requirement raises practical challenges.

Determining “sufficient” non-clinical data is often the primary hurdle. Article 9 requires only that “safety and effectiveness have been proven through non-clinical research”, but proof standards depend on case-specific factors, including technology type, mechanism of action, and potential risks. Corporate teams design experiments scientifically, but regulators may focus on different risk points. Expectations vary across technologies regarding animal species, observation periods, dose ranges, and toxicology depth. More novel technologies with uncertain risks face more cautious regulatory scrutiny. How can companies accurately anticipate these standards?

Protocol design presents ethical sensitivity challenges. Inclusion criteria, withdrawal standards, and risk controls involve both scientific rationality and ethical legitimacy. Broad inclusion criteria may be questioned as expedient; vague risk controls may suggest inadequate contingency planning. These details can trigger repeated revisions and filing delays. How can companies balance scientific rigor with ethical propriety?

Informed consent form design requires similar attention. Excessive technical terminology prevents patient understanding; while oversimplification omits critical risk information. How can complex principles and risks be explained across educational backgrounds? How can content meet ethical requirements while remaining accessible? This balance requires refinement that companies often lack experience in achieving.

Companies should consider pre-submission expert reviews involving professionals with both technical and regulatory expertise to evaluate materials from a regulatory perspective. Pre-communication with regulatory authorities before formal submission can clarify project-specific requirements. Professionals familiar with assessment standards and review priorities help identify risk points and improve materials, avoiding repeated corrections and delays. For highly innovative technologies with uncertain risks, designing communication strategies, obtaining pre-filing guidance, and reducing post-filing supplementation requests require support that combines technical understanding and regulatory experience. Earlier recognition enables greater process control.

II. Research Site Qualification Management

The Regulations establish strict thresholds for clinical research sites: tertiary grade-A status, compliant academic and ethics committees, appropriate professional qualifications, facilities, equipment, management systems, personnel, research capabilities, quality assurance systems, and stable funding. These requirements exclude many sites, but even qualifying hospitals may harbor substantive risks.

Qualifications are not static. After signing and initiating research with a tertiary grade-A hospital, that site may be downgraded during a national evaluation. Companies then face immediate questions: Must the research terminate? How are enrolled subjects handled? Are invested funds recoverable? While evaluations are periodic and downgrades uncommon, they remain possible.

More subtle risks involve personnel turnover. Article 11(3) requires sites to have “professional technical personnel and research capabilities appropriate to the proposed research”. If core team leaders depart with key members, the hospital may retain tertiary grade-A status and facilities yet lack “appropriate research capability”. When experienced personnel leave while equipment remains, does the site still qualify? For technically demanding, personnel-dependent projects, core team loss substantially weakens research capability, yet this change is less visible than downgrades and may go undetected.

Companies should include qualification continuity clauses and dynamic assessment mechanisms in collaboration agreements, stipulating responses to qualification changes. How should clauses be worded for enforceability and operability? How should triggering thresholds be set? How should breach liabilities be allocated? These require customization based on project characteristics and negotiating positions. More importantly, companies should establish ongoing assessment mechanisms—conducting not only due diligence on signing but also continuous monitoring of qualification status, team stability, and research capability throughout the research period, promptly identifying and addressing risks. For long-cycle, high-investment projects, continuous monitoring is essential.

III. Intellectual Property Rights

While the Regulations do not directly address intellectual property attribution, they require written agreements clarifying parties’ rights and obligations. Clinical research may generate new technical solutions, data, improved methods, or discoveries. Result ownership should be among the most critical agreement provisions. IP disputes typically stem from unclear advance stipulations or insufficient foresight of possible results.

(I) Technology Boundaries

When companies provide core technology and sites discover improvements during research, improvement patent ownership may spark controversy. If agreements stipulate only that “original technology belongs to company, research data is shared” without addressing improved technology, divergence can easily arises.

The key lies in defining the boundaries of “improvement”. Should parameter optimization be considered part of the original technology? Should entirely new mechanisms be considered independent inventions? Generic terms like “original technology” and “derivative technology” often defy specific application. Another controversy involves contribution assessment. Companies may believe improvements are impossible without their core technology; sites may believe improvements result from their teams’ exploration. Such disputes often cause relationship breakdowns, delayed patent applications, missed optimal timing, or competitor preemption.

Agreements need detailed result boundary definitions: What constitutes “original technology”, including specific content and parameters? How is “improved technology” defined—including parameter optimization, process improvements, and new applications? When does “jointly developed technology” arise? How is contribution assessed? These definitions require customization based on technical characteristics and negotiating positions, deep technical understanding, and clear legal translation. Negotiations must balance interests, designing allocation mechanisms that protect corporate core technology while incentivizing site R&D, avoiding breakdowns or execution difficulties from one-sided terms.

(II) Publication Timing

During research, site teams often seek rapid publication for academic recognition, while companies worry that premature disclosure affects patent novelty. If researchers present key technical details at international conferences and companies learn of them afterwards, patent novelty may be lost. Sites may believe presentations don’t constitute “publication” and agreements don’t prohibit academic exchange, but corporate losses have occurred.

This exposes information disclosure management gaps. Agreements typically require “mutual consent for significant results publication”, but what constitutes “significant”? Do improved technical solutions qualify? Newly discovered biomarkers? Oral conference presentations? Poster displays? Without advance stipulation, post-hoc definition proves difficult. Another timing conflict involves patent application decision processes. Corporate internal assessment of improvement patentability may require weeks or months, but site researchers may urgently seek publication, viewing corporate delays as restricting academic freedom, even threatening independent publication, forcing hasty, low-quality applications.

Agreements should clarify that all disclosure forms require prior approval, including papers, presentations, posters, media interviews, and social media. Patent application timelines can be stipulated: companies should decide within reasonable periods (e.g., 60 days) after receiving notice of a patentable result whether to apply; if declining, sites may independently apply or publish. Publication approval processes should be established: proposed external disclosures require advance submission to the other party for review, with opinions due within specified periods (e.g., 15 working days), with silence deemed consent.

These processes must balance interests. Companies cannot indefinitely prevent site publication; and sites cannot publish regardless of corporate interests. Designing mechanisms that protect commercial interests while respecting academic rights requires a deep understanding of patent law, contract law, and publishing norms. Companies should establish rapid response mechanisms to quickly evaluate and decide when sites request publication, avoiding conflicts caused by inefficient processes or lost patent opportunities due to slow decisions.

(III) Data Rights

Research-generated data proves crucial for corporate R&D and financing, yet ownership and usage rights division may raise significant controversy. If agreements stipulate “data jointly owned”, post-research companies may seek data use for next-generation development and investor presentations, while sites seek data for new project applications and other corporate collaborations—both believing they have usage rights, yet the other’s usage potentially harming their interests. When sites provide jointly owned data to corporate competitors for cooperative research, commercial interests face direct threats.

How is “joint ownership” interpreted legally? Property law divides joint ownership into ownership by shares and common ownership—which applies to data rights? Can property rules apply to data? If applicable, does the use of jointly owned data require unanimous co-owner consent? If either party can veto the other’s use, does this prevent effective utilization? In specific scenarios, companies may wish to use data for future FDA submissions, but sites worry that data leakage affects their IP and refuse consent. Companies may believe regulatory submission constitutes legitimate use beyond site restrictions, but such disagreements may delay international registration.

Data usage also involves subject privacy. If companies use real trial case data and images (even de-identified) during financing presentations to demonstrate efficacy, subjects may self-identify through certain features and complain, believing companies used their information for commercial promotion without consent. Companies may argue that data was de-identified with site authorization, but subjects may believe consent forms specified scientific research use only, not commercial promotion. Has the company exceeded the consent scope? This involves the application and interpretation of personal information protection law.

Agreements need clear data right distinctions. Regarding ownership, clarify who owns physical carriers and electronic files—typically, sites keep original medical records. Regarding usage rights, distinguish purposes and separately stipulate whether consent is required: product development and regulatory submission typically grant companies independent usage with site notification; academic publication requires bilateral negotiation of authorship and content; commercial promotion requires bilateral written consent; third-party cooperation requires prior approval and agreement review. Regarding disposal rights, data transfer or licensing to third parties typically requires unanimous co-owner consent.

Data usage approval mechanisms and dispute resolution procedures merit consideration. For example, establish data management committees with bilateral representation responsible for approving major usage applications. For data usage disputes, first attempt negotiation, then independent expert committee mediation if needed, then arbitration or litigation. These institutional designs require refined application of property law, contract law, personal information protection law, and IP law principles. “Data joint ownership” appears simple but proves legally complex, requiring in-depth interest analysis and solution design that accommodates all parties’ rights, avoiding irreconcilable conflicts arising from unclear stipulations.

IV. Informed Consent Standards

Article 19 requires written informed consent from subjects, information disclosure in easily understandable ways, and prohibits consent through deception, coercion, or inducement. But how is “easily understandable” defined? If researchers spend adequate time explaining technical principles, risks, and alternatives, and patients sign consent forms, but subsequent regulatory follow-up reveals vague technical understanding and fuzzy adverse reaction probability recall, does this consent meet requirements? If patients didn’t fully understand when signing, is the consent valid?

While companies don’t directly implement informed consent, they face the dilemma of ensuring that site-obtained consent meets “easily understandable” requirements. If sites must test subject understanding post-explanation, asking key questions like “What are the main risks?” and “What rights do you have if serious adverse reactions occur?”, considerable proportions may answer incorrectly. But excluding these subjects may raise new ethical controversies: excluded patients may be lower-education populations urgently needing new treatment options. yet losing opportunities through failed understanding tests—constituting another unfairness form?

Informed consent timing and environment affect validity. If sites obtain consent when patients are critically ill and emotionally distressed, families may question the subsequent “voluntariness”. Consent content dynamism also deserves attention. If no serious adverse events have been internationally reported at research initiation, but similar foreign research reports deaths mid-study, does the company obligate sites to inform all subjects of this new information? Is re-obtaining consent necessary?

Companies should stipulate informed consent implementation standards in agreements, requiring sites to establish multi-level processes. Written materials should be distributed in advance with adequate reading time; oral explanations should use plain language and visual aids like diagrams and videos; dedicated Q&A time should encourage questions with patient answers; cooling-off periods should allow subjects to take materials home for family discussion before deciding; entire processes should be recorded, including explanation content, subject questions, and provided answers, proving consent adequacy. Companies should establish supervision mechanisms for site consent implementation, regularly spot-checking consent records and conducting subject follow-ups when necessary to ensure quality.

Informed consent process complexity requires adjustment based on technical risks and subject population characteristics. High-risk technologies like gene editing require stricter processes; vulnerable populations like children and psychiatric patients require special protections. How can consent quality be balanced with recruitment timelines? Companies should explore with sites, combining ethics expert input, designing processes meeting compliance requirements with operability, ensuring site frontline personnel receive adequate training in communication skills and ethical awareness, preventing consent formalism.

V. Adverse Event Liability

Article 27 stipulates that treatment costs for research-caused subject health damage shall be borne by sponsors, or by research sites if caused by their fault, and encourages commercial insurance for subject protection. But how is “caused by research” judged? When subjects experience health problems with complex underlying conditions, causal determination may prove extremely difficult. For example, cell therapy research subjects experience abnormal liver function—patients have hepatitis B history, but liver enzymes elevated post-cell infusion. Companies may believe hepatitis B virus replication caused it; sites may believe cell infusion triggered immune responses. Parties may dispute endlessly over causation.

Many medical causal relationships defy clear determination, particularly with complex patient conditions. Even when establishing adverse event determination committees with independent medical experts, expert opinions may diverge: some believe “definitely related”, others “possibly related”, others “possibly unrelated”. What determination standard applies? Majority rule or the most unfavorable to the company? Liability limitation periods also pose questions. If subjects develop malignant tumors years after gene therapy, suspected gene editing-related, but interim lifestyle, environmental exposures, and other diseases may have contributed, how can tumor-research causation be determined? Does corporate responsibility still apply?

More pressing issues involve potential treatment delays from liability disputes. When patients experience serious adverse reactions requiring emergency treatment, if companies and sites disagree on cost bearing, sites may postpone expensive treatments while awaiting corporate payment confirmation—such delays may worsen conditions. The Regulations intend “subjects first”—treatment regardless of liability determination—but unclear cost bearing may cause treatment delays, the most critical situation to avoid.

Companies can stipulate adverse event causal determination procedures and standards in agreements, establishing independent medical expert committees with clear composition, procedures, and opinion legal effect. More importantly, stipulate corporate cost advancement during determination periods, with ultimate liability adjusted based on results, ensuring subjects receive immediate treatment. Companies are advised to purchase adequate clinical trial liability insurance for subjects—insurance not only disperses corporate risk but also ensures subjects receive timely treatment regardless of liability determination.

But how to select appropriate insurance products? Determine insurance amounts? Define coverage scope? This requires assessing research risk levels, possible treatment costs, insurance market supply, and other factors, selecting optimal cost-effective solutions. Agreements should also stipulate insurance claim initiation procedures and cost advancement mechanisms, ensuring rapid treatment and claim process initiation when adverse events occur, rather than endless liability disputes. This institutional design requires a deep understanding of medical risks, insurance practice, and emergency response to balance subject protection with reasonable responsibility allocation.

VI. Medical Device Coordination

Article 21 stipulates that research involving medical devices must also comply with medical device supervision provisions. This seemingly simple clause may raise regulatory pathway complexity. When biomedical technology combines with medical devices, companies may face dual regulation: both Regulations filing and Medical Device Supervision and Management Regulations compliance.

The key lies in defining “involving medical devices”. If cell therapy uses specific cell separation equipment or culture devices, does this constitute “involving medical devices”? Some biomedical technology product attributes lie between “technology” and “device”, with uncertain regulatory characterization.

When research involves medical devices, companies must simultaneously meet two regulatory systems. In preparing filing material, not only Regulations-required non-clinical research reports but also device-related testing reports and product technical requirements may be needed. During research implementation, both Regulations provisions on subject protection and ethical review plus device clinical trial quality management standards must be followed. This dual requirement may increase compliance costs and operational complexity.

Companies should clarify product or technology regulatory attributes at the project design stages. If uncertain, written consultations can be submitted to both the National Health Commission and National Medical Products Administration, describing technology or product characteristics, mechanisms, usage methods, requesting regulatory pathway clarification. Agreements should also clearly stipulate cost sharing when regulatory pathway adjustments create additional compliance costs. For situations truly requiring dual regulatory compliance, companies should establish coordination mechanisms to ensure compliance work integration, avoiding risks from neglecting one aspect.

VII. Military Medical Site Considerations

Article 56 stipulates that military medical sites conducting biomedical new technology clinical research and translational applications shall implement procedures according to military regulations and file with the National Health Commission. This clause recognizes military system particularity but may cause operational confusion.

Military medical sites often possess strong medical technology and research capabilities, making them ideal partners. But the military system has unique management and approval processes potentially differing from civilian systems. When companies cooperate with military sites, they must both comply with military regulations and file with the Health Commission—how should these requirements be coordinated?

Military regulations may involve confidentiality reviews, military ethics committee reviews, and military research management approvals. These procedures’ specific requirements, approval standards, and timelines may differ from civilian systems. For example, military site ethics committees may not be included in civilian health commission filing systems—when submitting filing materials, how can companies prove ethical review validity? Military confidentiality reviews may require special handling of certain research materials—will this affect complete filing material submissions?

Time coordination also poses practical issues. If military approval processes are lengthy while the Regulations require filing within 5 working days after ethical review passes, how should companies grasp timing? File after military approval completion or after obtaining military ethical review opinions? Inconsistent timing may cause procedural problems.

When cooperating with military sites, companies should fully understand military system special requirements at the project initiation. Pre-communication with military site research management departments and ethics committees can clarify approval processes and timing. Agreements should clearly stipulate military approval responsibility attribution, time arrangements, and delay handling. When submitting materials to the Health Commission, filing instructions should clearly explain military site particularity, attaching military approval certification documents when necessary, avoiding filing obstruction from procedural differences.

For companies unfamiliar with military systems and lacking military cooperation experience, such project complexity may exceed expectations. How can cooperation plans be designed to meet dual civilian-military requirements? How to appropriately explain military system particularity in filing materials? How to coordinate civilian-military timing and procedural requirements? These issues require advisory experts with deep understanding of dual regulatory systems to provide feasible solutions.

Conclusion

On May 1, 2026, the new Regulations take effect—preparation windows are fleeting. Establishing comprehensive compliance systems and selecting professional advisors constitute not only necessary regulatory responses but also strategic choices for establishing core competitiveness in the highly regulated biopharmaceutical industry. Only by integrating compliance management into every operational aspect can companies balance innovation with regulation, proceeding steadily where opportunities and challenges coexist.