GMPとは？

プロセスバリデーションとは？EP抜粋含

程が、設定規格・品質特性に適合した中間体・原薬（製品）を製造するた

めに効果的かつ再現性よく機能できることに関する文章による確証です。

プロセスバリデーションには三種類（＋α）あります。今回はその３種類
のプロセスバリデーションを紹介します。

１．予測的バリデーション：開発段階で製造プロセスの危険度を解析する方
　　　　　　　　　　　　法で実施するものです。
　　　　　　　　　　　　3回連続して成功した製造ロットを1つの指標とし
　　　　　　　　　　　　て使用すべきです。

２．同時的バリデーション：英語ではコンカレントバリデーションといいま
　　　　　　　　　　　　す。通常生産を行っている間に実施します。発段
　　　　　　　　　　　　階でのプロセスの基本事項が適正に認識されてい
　　　　　　　　　　　　るときにのみ効果があります。実生産規模で製造
　　　　　　　　　　　　した当初の３バッチについては、できる限り広い
　　　　　　　　　　　　範囲でモニターしなければなりません。このモニ
　　　　　　　　　　　　タリングの評価の結果を踏まえて、その後に行う
　　　　　　　　　　　　工程管理・最終試験の特性・規格を設定します。
　　　　　　　　　　　　同時的バリデーションは、安定性その他の傾向解
　　　　　　　　　　　　析と同時に、その製品の製造を続けている期間中、
　　　　　　　　　　　　継続的に適切な頻度で実施するのが望ましいです。

３．回顧的バリデーション：原料、装置、システム、設備または製造工程での
　　　　　　　　　　　　変更に起因する製品（原薬）の重要な品質に変動が
　　　　　　　　　　　　ないことが十分に確立されている工程に実施します。
　　　　　　　　　　　　工程の恒常性を評価するために、一般的に１０～３
　　　　　　　　　　　　０ロットの連続するデータを検討すべきです。

（EP抜粋）
PROCESS VALIDATION
General
20. The requirements and principles outlined in this chapter are applicable to the
manufacture of pharmaceutical dosage forms. They cover the initial
validation of new processes, subsequent validation of modified processes and
re-validation.
21. Process validation should normally be completed prior to the distribution
and sale of the medicinal product (prospective validation). In exceptional
circumstances, where this is not possible, it may be necessary to validate
processes during routine production (concurrent validation). Processes in
use for some time should also be validated (retrospective validation).
22. Facilities, systems and equipment to be used should have been qualified and
analytical testing methods should be validated. Staff taking part in the validation
work should have been appropriately trained.
23. Facilities, systems, equipment and processes should be periodically evaluated
to verify that they are still operating in a valid manner.

Prospective validation
24. Prospective validation should include, but not be limited to the following:
(a) short description of the process;
(b) summary of the critical processing steps to be investigated;
(c) list of the equipment/facilities to be used (including measuring/
monitoring/recording equipment) together with its calibration status
(d) finished product specifications for release;
(e) list of analytical methods, as appropriate;
(f) proposed in-process controls with acceptance criteria;
(g) additional testing to be carried out, with acceptance criteria and analytical
validation, as appropriate;
(h) sampling plan;
7
(i) methods for recording and evaluating results
(j) functions and responsibilities;
(k) proposed timetable.
25. Using this defined process (including specified components) a series of
batches of the final product may be produced under routine conditions. In
theory the number of process runs carried out and observations made should
be sufficient to allow the normal extent of variation and trends to be established
and to provide sufficient data for evaluation. It is generally considered
acceptable that three consecutive batches/runs within the finally agreed
parameters, would constitute a validation of the process.
26. Batches made for process validation should be the same size as the intended
industrial scale batches.
27. If it is intended that validation batches be sold or supplied, the conditions
under which they are produced should comply fully with the requirements of
Good Manufacturing Practice, including the satisfactory outcome of the validation
exercise, and with the marketing authorisation.

Concurrent validation
28. In exceptional circumstances it may be acceptable not to complete a validation
programme before routine production starts.
29. The decision to carry out concurrent validation must be justified, documented
and approved by authorised personnel.
30. Documentation requirements for concurrent validation are the same as
specified for prospective validation.

Retrospective validation
31. Retrospective validation is only acceptable for well-established processes
and will be inappropriate where there have been recent changes in the composition
of the product, operating procedures or equipment.
32. Validation of such processes should be based on historical data. The steps
involved require the preparation of a specific protocol and the reporting of
the results of the data review, leading to a conclusion and a recommendation.
33. The source of data for this validation should include, but not be limited to
batch processing and packaging records, process control charts, maintenance
8
log books, records of personnel changes, process capability studies, finished
product data, including trend cards and storage stability results.
34. Batches selected for retrospective validation should be representative of all
batches made during the review period, including any batches that failed to
meet specifications, and should be sufficient in number to demonstrate process
consistency. Additional testing of retained samples may be needed to
obtain the necessary amount or type of data to retrospectively validate the
process.
35. For retrospective validation, generally data from ten to thirty consecutive
batches should be examined to assess process consistency, but fewer batches
may be examined if justified.

ジャンル別一覧