In these days of global clinical trials, most regulators have provisions to accept international clinical data in support of regulatory submissions. Typically the acceptability of foreign data is decided on two separate and distinct criteria:
- Are the foreign standards of GCP acceptable?
- Are the foreign data collected on equivalent populations and in equivalent settings so that they are scientifically valid in relation to the local population?
Demonstrating comparable GCP is relatively straightforward. Most sophisticated jurisdictions will have GCP regulations derived from the Helsinki Convention and apply international standards (typically ISO 14155 for devices or ICH GMP for medicines). It then becomes a matter of compliance with the local GCP regulations as well as ensuring that the conduct, management and monitoring of the trial sites is consistent across all of the sites irrespective of location. While this is not a trivial task it is feasible.
But isn’t it different over there?
The second part is where it gets interesting. There are so many variables which affect comparability of populations and clinical setting. Are the foreign populations of equivalent demographics, are there different prevalences of potentially confounding clinical conditions? Are there cultural differences which may impact safety or effectiveness? Are there significant differences in the practice of medicine? Are there genetic differences? This last one is increasingly important in these days of personalised medicine. Most clinical reviewers will not be closely familiar with the overseas settings and it’s up to the manufacturer to demonstrate equivalence by ruling out potential areas of significant difference. That can be tantamount to attempting to prove a negative.
In the US, there is long standing practice of acceptance of overseas data. However this has been done conservatively and in most cases in conjunction with US data. The general approach has been to require trials in both US and international sites and to assess if there are significant differences between the US and foreign data sets.
New FDA draft guidance for international device clinical studies seeks to clarify US practice – but does it? The guidance essentially reiterates the above two points- that data must have equivalent or better GCP for protection of the human subjects and the data must be applicable to the US clinical setting.
Performance vs. Effectiveness
In a thinly veiled poke at the European/GHTF regulatory model, the FDA guidance points out that some regulatory jurisdictions assess safety and performance, whereas FDA requires consideration of effectiveness. Performance is essentially an engineering measurement, whereas effectiveness is a clinical judgement. Even so, Europe and other GHTF based regulatory systems all require clinical evidence as part of regulatory review and the differences between the European and US approaches to clinical evaluation are disappearing fast as Europe continues to beef up expectations with regard to clinical evaluations. It will be interesting to see if European clinical studies especially become more acceptable to FDA in the future.
The draft includes a range of examples which illustrate how the US FDA assesses international clinical data. This much is very helpful. But those looking for definitive statements on acceptability of international data will be disappointed. It’s clear that the agency will continue to be conservative and there’s a strong recommendation to consult with FDA prior to embarking on international studies. In other words – it’s still very much case by case.
And what about the rest of the world. All IMDRF jurisdictions and many Asian regulators are able to accept international clinical data in support of regulatory filings. Most repeat the two caveats about GCP and clinical applicability of the foreign data. Again – it’s advisable to talk to the agency first.
One of the more interesting areas is in the area of personalised medicine – where there is very real potential for local differences in population genetics to impact on effectiveness of companion diagnostic testing. One of the more thoughtful approaches here can be found in Japan. The PMDA Companion Diagnostics project has resulted in detailed guidance which includes consideration of the likely impact of genetic difference.
Nonetheless – caution still prevails. If you plan on using foreign data to support a regulatory submission, You are well advised to consult the agency first.
Confused about international data? Let’s talk… we are experienced in development of clinical evaluation strategy and in conduct of pre-submission consultations in the US, Australia and in Asian jurisdictions including China.