The ICH guidelines are undoubtedly of value, but are dedicated to specific topics rather than reflecting a comprehensive picture of the whole area. For small drug regulatory authorities it is vital not to get lost in details and to focus on priorities. The ICH guidelines are therefore not vital to our regulatory structure and they should be evaluated at national level with regard to their relevance and implemented according to the needs of the country.
One important aim for my own country is to bring legislation in line with that of the European Union. For this reason, the ICH guidelines are considered important. However, we are often not in a position to comment on the voluminous draft ICH texts with intricate technical details which we receive given the limited resources of our regulatory authority. On the other hand it is true that the ICH documents contain technical and scientific information which is valuable in improving the professional knowledge of regulators
In considering the ICH safety guidelines, it is apparent that they are of practical value and, for example in the case of carcinogenicity studies, will allow the regulator to evaluate whether studies should be required in the approval process. More specifically, we have found that the guideline on the assessment of systemic exposure is too general in its advice. Only limited practical value can be accorded to the guidance for repeated dose tissue distribution studies although it may have educational value as it covers a very specific area of expertise. The guidelines on the detection of toxicity to reproduction for medicinal products and toxicity to male fertility together form valuable material. But for the time being, this value is educational rather than regulatory, and the document would seem to have more importance for the pharmaceutical industry. Concerning wider dissemination, I would like to know whether the guidelines under discussion could be made available on the Internet and whether a server group could be established for comments and feedback?
* * *