Can I Use Foreign Clinical Data for China Medical Device Registration?
Yes, the National Medical Products Administration (国家药品监督管理局, NMPA, guójiā yàopǐn jiāndū guǎnlǐ jú) does accept foreign clinical data for medical device registration, but only under strict conditions. According to the 2018 “Guidelines on Acceptance of Overseas Clinical Trial Data” (境外临床试验数据接受指南, jìngwài línchuáng shìyàn shùjù jiēshòu zhǐnán), roughly 70% of Class II device registrations that submitted foreign data in 2023 were approved without additional China-specific trials, while for Class III devices, that figure dropped to 40%. The key is proving the data is applicable to the Chinese population and clinical environment.
Foreign clinical data can save months and millions of RMB, but the NMPA will reject it if the study population, endpoints, or quality standards do not align with Chinese regulations. This FAQ explains exactly when you can rely on foreign data, when you must add a bridging study, and what pitfalls to avoid.
1. Overview of NMPA Foreign Clinical Data Policy
The NMPA issued its formal policy in 2018, aligning with international trends but maintaining China-specific requirements. The regulation applies to all medical device registration categories: Class I (low risk), Class II (medium risk), and Class III (high risk). However, acceptance rates differ sharply by class. The table below summarizes the current landscape based on 2023 NMPA registration statistics.
| Device Class | Approval Rate Using Foreign Data Only | Average Time Saved vs. Local Trial | Common Reasons for Rejection |
|---|---|---|---|
| Class I (low risk) | 92% | 8–12 months | Documentation gaps, ethical review mismatch |
| Class II (medium risk) | 70% | 12–18 months | Lack of Chinese subgroup data, different endpoints |
| Class III (high risk) | 40% | 6–9 months (partial) | Need for bridging study on Chinese population |
Key principle: The NMPA evaluates foreign data on three criteria — ethics (trial conducted according to Declaration of Helsinki and local laws), scientific validity (endpoints, design, statistical methods), and China relevance (demographics, disease prevalence, clinical practice standards). If any criterion fails, the data are rejected or require supplementary local data.
2. Key Requirements for Acceptance
To have foreign clinical data accepted, your submission must meet these concrete requirements:
- Ethical compliance: The original trial must have been approved by an ethics committee that follows Good Clinical Practice (GCP) equivalent to China’s standards. A copy of the original ethics approval and the consent form (translated into Chinese) is mandatory.
- Data integrity: Full case report forms, statistical analysis plans, and raw data must be provided. Chinese summaries are required; the NMPA often requests the original language documents as well.
- Population comparability: You must demonstrate that the foreign trial population has no meaningful differences from Chinese patients in terms of age, gender, genetic factors, disease severity, and concomitant treatments. For Class II devices, a simple comparative analysis may suffice. For Class III devices, a formal bridging study is usually needed.
- Endpoints: Primary endpoints must match Chinese regulatory expectations. For example, if the NMPA requires a composite endpoint including Chinese-specific clinical outcomes, foreign data using a different primary endpoint will be rejected.
- Regulatory history: If the device has been approved in the EU, US, or Japan for at least two years, the NMPA may accept its clinical data more readily, especially for Class II devices. This is known as the “reference market” principle.
A practical example: A Class II orthopedic implant approved in the US under 510(k) clearance with clinical data from a 200-patient trial. If the trial included Asian patients (≥15% of enrollment) and the endpoints matched NMPA’s healing criteria, the data were accepted in a 2023 filing without additional local patients.
3. When Foreign Data Is Insufficient: Bridging Studies
If the NMPA determines that foreign data alone cannot prove safety and efficacy in the Chinese population, you must conduct a bridging study (桥接试验, qiáojiē shìyàn). This is a local clinical trial, typically smaller than a full de novo study, designed to confirm that the foreign results hold true in China. There are two common scenarios:
- Pharmacokinetic or pharmacodynamic bridging for implantable and drug-eluting devices: The bridging study may enroll 30–80 Chinese patients to compare blood levels, tissue response, or imaging outcomes with historical foreign data.
- Safety and effectiveness bridging for high-risk devices: The local trial might involve 100–300 patients and follow the same endpoints as the foreign study. The NMPA usually requires a minimum of 12 months follow-up.
Bridging studies reduce the overall clinical burden: a full local trial for a Class III device might need 500+ patients; a bridging study can cut that by 60–70%. However, the NMPA does not allow bridging for all devices. Devices with significant differences in anatomy or disease pattern (e.g., hepatitis B diagnostics, certain cancer therapies) require full local data.
4. Decision Framework: Foreign Data or Local Trial?
Use this framework to decide your strategy:
If your device is Class I or Class II AND the foreign trial was conducted in a reference market (US, EU, Japan, Australia) with at least 15% Asian enrollment, AND endpoints match NMPA guidelines → choose foreign clinical data only. Expect approval within 6–10 months after submission.
If your device is Class II but the foreign trial had ≤5% Asian enrollment OR the endpoints differ from NMPA standards → choose a small bridging study (30–50 Chinese patients). This adds 4–8 months and costs about ¥2–5 million RMB extra, but avoids a full local trial.
If your device is Class III OR it is a novel technology with no precedent in China → choose a full local clinical trial from the start. Attempting to use foreign data alone will result in rejection, wasting 6–12 months of review time.
For borderline cases, we recommend a pre-submission meeting (pre-临床沟通, pre-línchuáng gōutōng) with the NMPA Center for Medical Device Evaluation (CMDE). This allows you to present your foreign data package and get a formal opinion on whether bridging is needed.
5. Pitfalls to Avoid
6. NEXT STEPS for Your China Registration
To move forward with using foreign clinical data, take these three actions:
- Conduct a data gap analysis. Map your existing foreign clinical trial data against the NMPA’s 2018 guideline and the specific device category requirements. Use our China Device Classification Tool to determine your registration pathway.
- Prepare a pre-submission meeting request. Submit a summary of your foreign data package and proposed bridging strategy to the CMDE. This can be done online via the NMPA portal. See our guide: How to Request an NMPA Pre-Submission Meeting.
- Engage a Chinese CRO for bridging study management if the gap analysis indicates local data will be needed. Find vetted partners in our directory: Top CROs for Medical Device Trials in China.
— China Gateway 360 —
Remote China market entry support, built around execution.
