WFI obtained by non-distillation methods - What are the Next Steps?

The European Pharmacopoeia's revised monograph 169 on water for injection (WFI) will become effective by the 1st of April 2017. For the first time in Europe, a monograph will allow the production of WFI with other methods than distillation. According to the text applying as of April, all methods equivalent to distillation will be officially allowed. Reverse osmosis is only mentioned as an exemplary technique. In a previous draft, it was yet another. The European monograph also differs from the Japanese Pharmacopoeia for example in which reverse osmosis is still mentioned explicitly.

But there are still unanswered questions, e.g. how to deal with HPW systems (highly purified water) which have already been producing water with WFI quality. Many other points - mainly GMP aspects - remain unclear. Yet, it is not the duty of a pharmacopoeia to set GMP requirements. Now although there have been expectations about it, Annex 1 (currently undergoing revision) will only provide little information on WFI obtained by non-distillation methods. The only hope for more details about GMP specifications remains the Question & Answer document of the European Medicines Agency (EMA). Even though the document isn't at the top of the GMP documents hierarchy, it is likely to become one of the leading GMP guidances for the cold production of WFI. The competent expert group of inspectors (IWG) met in the 7th calendar week (2017) to discuss the revision of the first draft version of this document. Generally, content output can be expected after 4 weeks - let's say at the beginning or mid of March.

So far, it is clear that the competent GMP regulatory authority has to be informed when the use of WFI by membrane is planned. When a third country is involved (like India, China or the USA), the competent body is the GMP authority of the respective importer. It is also clear that sanitisation in systems with cold-generated WFI will be playing a much greater role. The present draft of the Q&A paper addresses a routine steam sanitisation of the storage and distribution system, whereby hot-storage systems can actually be considered to be self-sanitising. What about cold storage in permanently ozonised systems? According to the draft, ozone isn't sufficient. It will be interesting to see how the final version of the paper will deal with that.

Also the controls of feed water will become much more important. Although they don't play a really important role in distillation, impurities in the feed water of membrane processes are highly relevant. The control of conductivity and TOC will be more important too. Here, the TOC can also be considered as an indicator of the system's microbiologial contamination risk. According to the present draft, the TOC limits should be statistically justified. The use of 50% of the specification as action limit is thus invalid.

In technical terms, because of the elimination of the phase transition from liquid to steam, the membrane of the new system will become the critical installation section. Regarding this, it is also important to know that a hot sanitation is always recommended after installation. On the one hand, the system suppliers cannot guarantee the sterility of their membranes. On the other hand, the membranes first gain their complete operational capability through tempering. Another interesting point is that the membranes used e.g. in the reverse osmosis grow old. Their selectivity (cut off) changes over the years. Yet, there exist no specifications concerning this. The membranes should also be checked after a change - analogous to a measuring instrument which has to be calibrated once more before being scrapped to demonstrate its correct functioning in the last interval to the end. Moreover, an online integrity test of the membrane would be ideal - which is however in technical terms not available yet.

For many users who have been operating HPW systems for years, the question naturally arises of how the transition to WFI production should run. As already mentioned, the GMP regulatory authorities have to be informed. In terms of GMP, experts agree that the treatment/purification part of the system mustn't undergo a new PQ phase. Nothing changes regarding the manufacture of water. This is not the case regarding the storage and distribution system. Here, qualification activities should demonstrate that the cold produced WFI is as safe as the distilled one. For this purpose, the activities described close to maintenance, monitoring and sanitisation will be necessary. More details regarding this are expected in the coming version of the Q&A document.

Another interesting question is how long will HPW remain as a monograph in the European Pharmacopoeia? In the pharmaceutical industry, it has in fact no application except for cleaning purposes before the final rinse with WFI. But it is different in the medical devices industry. The outcome thus remains exciting.