第一篇：FDA對原料藥廠檢查流程

FDA對原料藥廠檢查流程

一、概述

“FDA”是美國食品藥物管理局(Food and Drug Administration)的英文縮寫，它是國際醫(yī)療審核權威機構，由美國國會即聯(lián)邦政府授權，專門從事食品與藥品管理的最高執(zhí)法機關。FDA是一個由醫(yī)生、律師、微生物學家、藥理學家、化學家和統(tǒng)計學家等專業(yè)人事組成的致力于保護、促進和提高國民健康的政府衛(wèi)生管制的監(jiān)控機構。就原料藥而言，F(xiàn)DA檢查目的是為了保證從國外進口的原料藥的質量充分符合USP的要求，美國政府規(guī)定外國的藥物生產(chǎn)商向美國出口藥物產(chǎn)品，除了要對該產(chǎn)品的樣品進行質量檢查之外，還要對藥物制造企業(yè)的相關設施進行檢查才能做出批準與否的決定，F(xiàn)DA現(xiàn)場檢查由此而生。FDA檢查主要分為三類：一是批準前的現(xiàn)場檢查(Pre—approval Inspection)，即我們通常說的“FDA檢查”，對新藥和仿制藥品的生產(chǎn)采取的檢查行動；二是定期檢查（Biennial）,對批準后的藥品進行定期的合規(guī)性檢查，通常兩年進行一次；三是基于投訴、召回或不良反應FDA臨時決定進行專門的檢查或監(jiān)督。

FDA檢查的依據(jù)起源于是美國國會1938年頒發(fā)的聯(lián)邦食品、藥品和化妝品法案（常縮寫為FFDCA，F(xiàn)DCA，或FD&C），該法案賦予美國食品藥品監(jiān)督管理局（FDA）監(jiān)督監(jiān)管食品安全、藥品、及化妝品的權力。關于藥品方面，主要是受“食品、藥物及化妝品法案”第501款(a)(2)(b)的管制，即所有藥物的制造、加工和包裝，均要嚴格符合cGMP的要求。GMP制度在聯(lián)邦法規(guī)(code 0f Federal Regulations)中的第210和第211條款中有具體規(guī)定。不過，自發(fā)布以來的GMP主要是為制劑藥而制定的。在它的前言中說明：雖然它不是用于原料藥，但有許多實例說明對原料藥的GMP要求是與第211條款中所制定的要求很近似。因此，F(xiàn)DA就采用第2ll條款作為規(guī)范來對原料藥廠進行檢查。在這點上，F(xiàn)DA對原料藥與制劑藥的要求都是一樣的嚴格，沒有區(qū)別。1997年9月，國際協(xié)調(diào)會議(ICH：InternationalConference 0f Harmonization)公布了專為原料藥制定的GMP草案，更切合原料藥的生產(chǎn)實際。2001年8月，美國健康人類服務部食品藥物管理局藥物評價研究中心和生物制品評價研究中心與國際協(xié)調(diào)會議聯(lián)合發(fā)布了用于活性藥物成分(原料藥)生產(chǎn)的GMP指南：Guidance for Industry Q7A—Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients)，即Q7A GMP。此后，F(xiàn)DA宣布以這個指南文件為原料藥生產(chǎn)的GMP統(tǒng)一標準，并以此對原料藥廠進行符合性檢查。

二．FDA檢查流程

FDA檢查通常由一位檢查官和一位藥物審查化學家或微生物學家到藥廠進行4—5日的檢查，F(xiàn)DA到藥廠后，會和工廠人員先進行一個簡短的見面會，在見面會上FDA會首先說明一下檢查的背景及檢查安排，藥廠由負責人向對方介紹所有出席會議人員，出席人員一般包括總經(jīng)理、質量授權人及各部門的負責人，然后由藥廠對工廠進行一個簡要的介紹，讓檢查官對工廠有一個初步的了解。檢查主要分為兩塊：現(xiàn)場視察與文件檢查。

2.1 現(xiàn)場檢查

原則上，F(xiàn)DA檢查官是按照原料藥的生產(chǎn)順序（物流走向），即從原料接受到成品包裝再到放行的順序來先進行現(xiàn)場檢查的，但根據(jù)環(huán)境條件和檢查官個人的專業(yè)背景、習慣與判斷方式的不同，也可能先檢查文件再看現(xiàn)場或中間穿插著去檢查現(xiàn)場。

檢查官的第一站是倉庫，包括原料、成品及包材倉庫，倉庫的關注點物料的管理，F(xiàn)DA不僅要求進廠的起始物料符合預先建立的質量標準，進行良好的儲存，還需要能夠預防不同物料或不同批次的物料混淆或產(chǎn)生交叉污染的風險。檢查官通常會從以下幾點來評估：首先起始物料入庫前是否有適當檢查流程，是否有入庫臺帳，不同物料之間是否有物理隔離，倉庫的溫濕度是否有監(jiān)控并能夠達到物料所要求的存儲溫度，倉庫是否有防蟲和防鼠的措施(如窗戶或風扇進風口是否裝有紗窗，倉庫內(nèi)是否有滅蚊蠅燈，粘鼠板或電貓)，物料容器的標識(取樣證、合格證或不合格者)，倉庫是否有專門的區(qū)域存放不合格品、退貨、召回產(chǎn)品，標簽發(fā)放與控制等。

2.2 按照生產(chǎn)順序，檢查完倉庫后，檢查官將會去車間看生產(chǎn)，通常要求在檢查期間車間處于動態(tài)生產(chǎn)的狀態(tài)，通常檢查官會對照著產(chǎn)品的生產(chǎn)工藝流程圖，一步一步地了解整個生產(chǎn)過程，以便對GMP的執(zhí)行過程進行深入的檢查。FDA官員會關注每個重點操作崗位SOP是否在現(xiàn)場，原始記錄是否與崗位SOP及實際運行情況一致以評估SOP的可操作性及員工的培訓是否到位。除外，檢查官會評估在整個生產(chǎn)過程中是否有物料污染或混淆的風險，如設備清洗是否足夠、設備及標示牌與管道上的物料標記是否清晰準確，各種設備是否有醒目的編碼便于操作，化學合成工藝中離心機濾袋的清洗及管理是否到位、不同房間及房間與走廊間的壓差是否正常。

2.3 公用系統(tǒng)（純化水、空調(diào)系統(tǒng)、壓縮空氣）

檢查官主要評估公用系統(tǒng)能否有效地運行并滿足產(chǎn)品生產(chǎn)所要求的條件，具體包括公用系統(tǒng)的日常維護保養(yǎng)、年度回顧與驗證、在線監(jiān)控，如純水電導率超標如何處理，是否有報警裝置及在線排放裝置，過濾器更換頻率，在線取樣等），壓縮空氣是否進行水油檢測、空調(diào)系統(tǒng)的過濾器更換等。

2.4 實驗室

現(xiàn)場檢查的重點之一是實驗室，F(xiàn)DA檢查至少會用半天甚至更多的時間檢查實驗室，通常由化學家或微生物學家進行。檢查官到實驗室時首先要了解的是樣品的流向及管理，即樣品接受、存放、分發(fā)及檢驗后的管理，是否有合理的臺帳確保樣品不會被混淆并可以追溯；空白檢驗記錄的管理與控制，其它包括實驗室儀器的合理使用及相應的記錄，如：HPLC設備與色譜柱的使用臺帳，二者是否可以相互追溯并；化學試劑的使用的管理(有效期，啟用日期等)；配制試液的標簽(試液名稱，配制日期，配制人簽字、復標日期及復標人，試液的效期規(guī)定等)和設備與儀器。FDA對設備與儀器的校驗尤其重視，藥廠需要定期進行校驗的常用的測量儀表有溫度計(包括自動記錄溫度計)，溫濕度計，壓差表，液體流速計，空氣流量計，液位計等；計量設備和器具如磅稱，粗天平，普通天平，分析天平以及高精密度天平等；工藝過程控制和質控試驗室用的pH計，滴定管，移液管等；儀器分析用的UV，IR，HPLC和GC等設備。FDA檢察官首先注意的是儀器設備上是否貼有校驗合格的標志，查看有關設備儀器校驗的SOP和使用與校驗的原始記錄。通常，我國的原料藥廠的度量衡儀器都是由藥廠專設的儀器計量室(大多是從屬設備管理部門)負責定期校驗并簽發(fā)校驗合格證。技術性高的儀器設備如UV，IR，HPLC和GC等設備則一般是委托地方政府的計量管理部門進行校驗。凡是發(fā)現(xiàn)重要的儀器設備沒有進行校驗，或提供不出完整的文件記錄說明，F(xiàn)DA檢察官都認為是嚴重的問題，因為未經(jīng)過合理校驗的儀器無法確保其準確性，同樣用該儀器檢測所得數(shù)據(jù)的可靠性也無法保證。

微生物實驗室包括培養(yǎng)基配制及適用性試驗、菌種傳代、工作服、工作鞋的清洗及器皿的滅菌等。

三、文件檢查

在FDA通過現(xiàn)場檢查之后，對工廠的布局及工藝流程有了一個具體的了解之后，便開始回到會議室集中看文件，通常兩個檢查官為分頭進行，化學家或微生物學家會集中檢查QC相關的文件。檢查的文件涵蓋生產(chǎn)系統(tǒng)、質量系統(tǒng)、廠房設施、實驗室等。

3.1 生產(chǎn)系統(tǒng)

檢查官在接到FDA對某個藥廠進行現(xiàn)場檢查的通知后，會從FDA文件管理處調(diào)出該藥廠提交的DMF文件提前準備。到現(xiàn)場后通常會隨機選一批或幾批完整的批記錄，然后對應DMF文件對應著檢查，對關鍵的操作步驟一條一條核對，這就要求提交的DMF文件必須與現(xiàn)場的操作完全一致。在這過程中檢查官會關注對關鍵的操作步驟的控制，涉及偏差時的相關調(diào)查記錄。

3.2.質量體系

質量體系是FDA檢查的核心，檢查會涵蓋關鍵的質量文件(偏差、投訴、OOS、召回、變更、自檢、驗證與確認、供應商管理等)，通常檢查官會隨機選取其中一個投訴或偏差，然后查看相關的調(diào)查記錄及相應的SOP，一方面評估藥廠的SOP是否合理，另一方面評估工廠能否夠按照SOP對質量事件進行徹底的調(diào)查并采取有效的整改與預防措施（CAPA）以防止類似事件再次發(fā)生。

3.3 廠房與設施

這部分檢查官會關注工廠是否有良好的廠房與設施維護計劃并通過檢查相關的記錄來評估工廠能否按計劃對廠房與設施進行很好的日常維護，包括儀器的校準與確認、設備與設施（水系統(tǒng)與空調(diào)系統(tǒng)、壓縮空氣）的3Q確認等。

3.4 實驗室

QC文件檢查通常包括產(chǎn)品的質量標準與檢驗方法確認與驗證，檢測記錄、穩(wěn)定性試驗方案與相關記錄以及其它檢查官在現(xiàn)場檢查時臨時需要的文件。

四、檢查的關鍵點

對任何公司來說，通過FDA驗廠最重要的條件是自己要嚴格執(zhí)行已經(jīng)確立的程序和操作流程以及記錄與數(shù)據(jù)的完整性與真實性，這兩點最能反映工廠的GMP質量管理水平，而質量體系正是FDA檢查的核心所在。這就要求記錄的填寫一定要規(guī)范可讀，不得隨意篡改記錄，寫錯更改需要保證原輸入的可讀性，一旦檢查官對記錄的真實性產(chǎn)生質疑，那對藥廠會非常的不利。另外，接待人員回答提問要有技巧，不清楚的事情切忌馬上回答或者是使用“我記得、好像”之類的詞匯，這樣會給檢查官留下很不專業(yè)的印象，不確定的可以先查文件，幾個人商量定下來之后再回答。

五、總結會

檢查官通常會留出一天時間來做總結，對整個檢查期間的發(fā)現(xiàn)進行匯總，即483表，并現(xiàn)場宣讀483上的每一條發(fā)現(xiàn)，并詢問工廠對各條發(fā)現(xiàn)是否有異議，如果工廠有需要解釋的地方，可以充分利用好此機會。如果FDA檢查官認為解釋有理，一般會對所提的問題進行修改或取消。如果對所發(fā)現(xiàn)的問題無異議，一般工廠代表人或公司總經(jīng)理需要作出表態(tài)表示接受檢查過程中的發(fā)現(xiàn)，然后雙方在483表格上正式簽字。FDA要求藥廠對提出的問題盡快(一般在兩周到一個月，根據(jù)情況而定)做出書面答復，其中要求提供明確的較詳細的整改回復，在規(guī)定時間內(nèi)遞交到FDA地區(qū)辦公室。FDA檢查官在回國后根據(jù)藥廠的整改報告寫出一份非常詳細的檢查報告送交FDA有關主管部門(如新藥評價中心，獸用藥評價中心等)。

六、最后的決定——批準(或不批準)的通知函

按FDA的規(guī)定，F(xiàn)DA檢查官無權對是否批準做出決定。FDA檢查官在檢查報告中非?？陀^地說明一切情況和存在問題，藥廠的態(tài)度和整改措施，對該藥廠是否可以得到批準會提出個人的建議，它對FDA做出批準或不批準的決定是具有關鍵性的影響的。

來源：HPC藥聞藥事/Joanna

第二篇：FDA是如何對原料藥廠進行檢查的（推薦）

(一)FDA是如何對原料藥廠進行檢查的

一、前言

本文作者從1981年起開始涉及原料藥的FDA申請事務，從1992年一直專門從事這項專業(yè)工作至今。其間先后參與或主持了近20個原料藥產(chǎn)品的FDA申請工作，制作歸檔了十幾個DMF文件，與十多位美國代理商或美國終端用戶的GMP符合及FDA申請顧問一起工作，九次參加FDA官員對我國一些制藥企業(yè)進行的GMP符合性現(xiàn)場檢查，其中8個品種通過了FDA的現(xiàn)場檢查，從而積累了一些有益的經(jīng)驗。作者現(xiàn)任北京康利華公司咨詢服務有限公司終身董事和監(jiān)事，仍專門從事FDA的申請工作。本文就FDA對原料藥管制的有關文件的學習了解，結合作者20年來的實際經(jīng)驗，對FDA官員對原料藥廠如何進行現(xiàn)場檢查作了簡要的敘述，希望能給我國廣大原料藥企業(yè)提供有益的參考。1．定義 1．1原料藥

通過化學合成、微生物發(fā)酵、天然物提取分離、酶工程、DNA重組等技術和手段得到的具有藥物活性、符合一定質量標準的物質。原料藥通常只供生產(chǎn)制劑之用，不可直接用于臨床。歐美對原料藥的稱呼有以下幾種：Bulk Pharmaceutical Chemical簡稱BPC?，F(xiàn)常用：Active Pharmaceutical Ingredient簡稱API，或Drug Substance。

生產(chǎn)原料藥的起始物料在生產(chǎn)過程中都要經(jīng)歷明顯的化學變化，然后經(jīng)分離純化制成具有藥物活性、且符合一定質量標準的原料藥。

原料藥又分為無菌(Sterile)和非無菌(Non—sterile)兩種級別。前者常用于生產(chǎn)非消化道給藥的制劑藥，后者常用于生產(chǎn)口服制劑或外用制劑，或再經(jīng)過無菌處理生產(chǎn)非消化道制劑藥。

1．2制劑藥

將原料藥與輔料一起進一步加工，制成的適合臨床應用的各種形式，所得到的產(chǎn)品為制劑藥。歐美常把制劑藥稱作：

Finished Pharmaceutical，F(xiàn)inished Product，Dosage Form，F(xiàn)inished Dosage Form或Drug product。

在20世紀50年代中期，為了保證從國外進口的原料藥的質量充分符合USP的要求，美國政府規(guī)定外國的藥物生產(chǎn)商向美國出口藥物產(chǎn)品，除了要對該產(chǎn)品的樣品進行質量檢查之外，還要對藥物制造企業(yè)的相關設施進行檢查才能做出批準與否的決定。2．FDA對國外原料藥生產(chǎn)商的檢查 2．1檢查的歷史

FDA從1955年開始進行第一次國外檢查(抗生素)。1961年國外檢查達13項，以后繼續(xù)增加。1971年成倍增加，達到80項。此后，在整個70年代里不斷增加。到了80年代及至90年代初，達到了每年檢查160項之多。1993年FDA計劃要進行340項檢查·2000年進行的48項檢查中，批準了28項。在未批準的項目中，有14個是問題很大的。發(fā)出了1l封警告信(2001年發(fā)出9封)。2002年和2003年FDA都派了檢察官來我國檢查一些藥廠。2．2 FDA國外檢查的范圍

檢查地域的面較廣，主要有歐洲的意大利、荷蘭、德國、俄羅斯、波蘭、匈牙利等國，亞洲的中國、日本及印度等國。

只有加拿大，瑞典和瑞士與美國訂有雙邊協(xié)議，美國不派員去檢查，由他們自己的官員進行檢查，F(xiàn)DA認可檢查結果，并彼此交換檢查信息。2．3 FDA國外檢查的情況：

一美國制劑制藥企業(yè)使用的原料藥有70％來自國外。

一接受FDA檢查的企業(yè)有三分之二是生產(chǎn)原料藥的。

一有17％的原料藥制藥企業(yè)和11％的制劑制藥企業(yè)不符合檢查要求。

一外國接受檢查的制藥企業(yè)有六分之一不符合GMP要求。

一FDA向一些檢查結果不能令人滿意的制藥企業(yè)發(fā)出了警告函。

一不合格的企業(yè)的醫(yī)藥產(chǎn)品被禁止進入美國。

一在歐洲擁有良好的符合要求記錄國家是意大利。

一日本是接受FDA檢查制藥企業(yè)最多的國家，占總數(shù)的百分之十六。并有良好的符合要求的記錄。

3．FDA對我國原料藥廠的檢查

自從八十年代初我國的原料藥企業(yè)開始向美國FDA提出申請，接受FDA官員的檢查，迄今，已有一些企業(yè)的生產(chǎn)原料藥的設施通過了檢查，得到FDA的認可，已有為數(shù)可觀的各種原料藥(如四環(huán)素，土霉素，慶大霉素，鏈霉素，金霉素，潔霉素，依維菌素，硫鏈絲菌素等抗生素，甲硝唑，撲熱息痛等合成藥)進入美國市場，占有了一定的份額。

自從八十年代初以來，我國藥廠取得FDA批準出口美國的原料藥主要還是非無菌級的原料藥。

二、FDA對原料藥管制的依據(jù)

原料藥要接受什么法規(guī)管制?簡單說來，就是要接受cGMP的管制。具體地說，就是要接受美國“食品、藥物及化妝品法案”(FDCA)第501款(a)(2)(b)的管制，即所有藥物的制造、加工和包裝，均要嚴格符合cGMP的要求。

GMP制度在聯(lián)邦法規(guī)(code 0f Federal Regulations)中的第210和第211條款中有具體規(guī)定。不過，自發(fā)布以來的GMP主要是為制劑藥而制定的。在它的前言中說明：雖然它不是用于原料藥，但有許多實例說明對原料藥的GMP要求是與第211條款中所制定的要求很近似。因此，F(xiàn)DA就采用第2ll條款作為規(guī)范來對原料藥廠進行檢查。在這點上，F(xiàn)DA對原料藥與制劑藥的要求都是一樣的嚴格，沒有區(qū)別。1997年9月，國際協(xié)調(diào)會議(ICH：International Conference 0f Harmonization)公布了專為原料藥制定的GMP草案，更切合原料藥的生產(chǎn)實際。2001年8月，美國健康人類服務部 食品藥物管理局 藥物評價研究中心 和 生物制品評價研究中心 與國際協(xié)調(diào)會議聯(lián)合發(fā)布了用于活性藥物成分(原料藥)生產(chǎn)的GMP指南：Guidance for Industry Q7A—Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients)，即Q7A GMP。此后，F(xiàn)DA宣布以這個指南文件為原料藥生產(chǎn)的GMP統(tǒng)一標準，并以此對原料藥廠進行符合性檢查。

三、規(guī)章的實施與指導文件 1．法規(guī)的實施

*在美國，每個FDA的管轄區(qū)都有一套原料藥制造商名單，一般對藥物生產(chǎn)廠每兩年要進行一次檢查。由各轄區(qū)安排檢查計劃。FDA檢查官是分地區(qū)的，美國各地區(qū)國內(nèi)檢查官也兼國外檢查官的任務。這樣就保證了國外的藥物生產(chǎn)商也受到國內(nèi)生產(chǎn)商同樣要求的檢查。*進行批準前的現(xiàn)場檢查(Pre—approval inspection)，即我們通常說的“FDA檢查”)一對新藥和仿制藥品的生產(chǎn)采取的檢查行動。

*在有投訴、藥源性傷亡或疾病發(fā)生時，F(xiàn)DA要進行專門的檢查或監(jiān)督。2．指導文件

FOAL公布了各種指導文件，其中有專供FDA檢查官使用的關于原料藥生產(chǎn)的“原料藥生產(chǎn)檢查指南”【The Guide for Inspection 0f Bulk Pharmaceutical Chemicals】。

四、FDA檢查官檢查的方式和重點 1．檢查方式

FDA對我國原料藥廠進行檢查，在6年前，每次是派一位檢查官到藥廠進行2—3日的檢查。此后，改為每次派兩位檢查官(確切地說，是一位檢查官和一位藥物審查化學家)到藥 2 廠進行4—5日的檢查，比過去的檢查大為強化。其中：

*預備會議 約半日 *現(xiàn)場檢查 約1.5-2日 *文件檢查 約1-1.5日 總結會議 約1日

FDA檢查官對藥廠一般采用如下程序進行檢查： 1．1預備性會談

首先FDA官員作自我介紹，向我方出示FDA檢查官身份證。我方負責人向對方介紹所有出席會議人員，出席人員一般包括：

*主談人(1名)：由公司總經(jīng)理(或廠長)，或負責生產(chǎn)技術的副總經(jīng)理(或副廠長)，或總工程師擔任，根據(jù)各藥廠的具體情況而定。

*出席人員：總工程師，生產(chǎn)(或技術)部門負責人，申請產(chǎn)品的車間主任，車間工程師，QA及QC負責人，設備管理部門負責人，倉庫主任，我方FDA申請顧問(一般都應兼任翻譯)，英語翻譯(2名)和聯(lián)絡員(1—2名，負責及時提交供檢查的文件和與接受檢查的各部門聯(lián)系工作?？捎缮a(chǎn)技術管理部門人員或市場銷售部門的外貿(mào)工作人員擔任，最好熟悉英語)。*美方有美國代理人(有時還有美方FDA申請顧問)也出席會議和參與整個檢查過程。1．2現(xiàn)場視察

包括(固體和液體)原材料，包裝材料，標簽和成品倉庫；生產(chǎn)車間，用水供應設施(主要是純化水制備車間)，質控實驗室(QC)，以及FDA官員即興提出要視察的任何地方(例如，三廢處理設施，藥廠自行加工的農(nóng)產(chǎn)品原料的車間或無菌區(qū)的空氣過濾系統(tǒng)等等)。FDA檢查官一般不要求視察與申請產(chǎn)品無關的地方。1．3文件檢查

a．DMF(藥物主文件)一FDA檢查官在接受檢查任務后從文件管理辦公室領取一份DMF文件，事先仔細閱讀，熟悉接受檢查的制藥企業(yè)的全面情況。FDA官員到達藥廠現(xiàn)場后，在整個的檢查過程里不斷地把DMF中所描述的情況和現(xiàn)場的實際情況進行核對。所以，DMF與現(xiàn)場實際的符合性或一致性是非常重要的。如果發(fā)現(xiàn)對不上號的地方太多，而且不能做出令人信服的解釋，將會構成嚴重的問題。

b．重點檢查的文件有：抽查近期的一個批的全套生產(chǎn)原始記錄；有關生產(chǎn)工藝，質量控制，設備清洗，環(huán)境衛(wèi)生，無菌區(qū)或潔凈區(qū)的清潔和消毒等方面的標準操作規(guī)程(SOP)，工藝驗證及設備的無菌度或清潔度的驗證方案和記錄；穩(wěn)定性試驗的方案和記錄以及FDA檢查官提出要查閱的任何文件。1．4會談與總結：檢查結束時與企業(yè)各有關負責人員會談向企業(yè)發(fā)給FDA意見表483(FDA FORM 483)其中列出企業(yè)存在的問題。限期做出改進和答復。

1．5 FDA檢查官向FDA總部提交檢查結果的詳細報告，由主管部門做出批準與否的決定。

2．現(xiàn)場檢查的重點

原則上，F(xiàn)DA檢查官是按照原料藥的生產(chǎn)順序，即從原料到成品包裝及出廠的順序來進行檢查的。但根據(jù)環(huán)境條件和檢查官個人的專業(yè)背景，習慣與判斷方式的不同而從任何一個工序開始進行檢查。以下是檢查的重點： i．原材料(倉庫管理)ii．生產(chǎn)工藝的執(zhí)行：生產(chǎn)操作和工藝驗證。iii．廠房、設備和設施 iv．設備的清洗 v．批成品的混合

vi．包裝及貼簽 vii．生產(chǎn)用水 viii質控實驗室

*儀器配備和校驗方法 *工藝過程的分析與控制 *試劑和試液的配制 *穩(wěn)定性試驗和留樣制度

2．1 原材料(倉庫管理)原材料是泛指從精細化學品到植物粗品及動物制品等用于生產(chǎn)原料藥的所有物料，其中包括所用到的各種溶媒及輔助化學品。

FDA檢查官最關心的是原材料的控制系統(tǒng)一倉庫管理和質量控制。在大多數(shù)情況下都是先到倉庫去檢查原材料的管理情況，同時也檢查成品，包裝原材料和標簽的倉儲管理情況。FDA檢查官要求進廠的原材料保證能符合要求的質量標準，并能良好地儲存。這類控制系統(tǒng)一般包含一些臨時隔離待驗系統(tǒng)采用物理方法的分隔及識別標志。

FDA檢查官對原材料的堆放要求符合“先進先出’’(First in and first out)的原則，對原材料包裝容器上的貼簽(諸如取樣證，合格證及不合格證等)尤其重視。在固體原材料方面，抗生素生產(chǎn)的發(fā)酵用的原材料中的農(nóng)產(chǎn)品原材料，例如黃豆粉，用粗麻布口袋包裝，很難貼好標簽，其它如裝淀粉的棉布口袋和裝化學品的塑料編織袋，由于重量大，堆放高，搬動不便，在取樣檢驗合格后，很難做到每袋都貼上合格證。FDA檢查官對此是講情理的，例如在一次檢查中對一批數(shù)量達2000個布袋的發(fā)酵原材料，不要求每袋都貼有待檢驗，合格或不合格的標簽。但要求堆垛外圍的袋上都貼有標簽。

對倉庫管理，有的FDA檢查官著重檢查管理的方法。在接收原材料時，對每一批原材料的條件，標簽批號及有關信息的記錄進行檢查。在多數(shù)情況下，至少應有一項目測或化學鑒別實驗。通常還附有一個合格的生產(chǎn)廠家的分析合格證(COA)，說明該原材料的質量規(guī)格均能符合標準的要求。對一些很關鍵的原材料和中間體則要求有全面的分析報告。

在液體原材料方面，有的FDA檢查官對用于藥物分離提取和純化精制的溶媒原材料特別注意，因為在結晶過程和精制階段，如果溶媒有污染，則成品會受到污染。對于桶裝溶媒，更注意的是是否嚴格管理而不會混用。我國化學工業(yè)和制藥業(yè)有個共同的缺點，就是對某些種類的溶媒或農(nóng)產(chǎn)品原料(如玉米漿液)是利用其它原料用過的舊桶，而沒有把原有的文字標記除去。鋼桶容器由于沒采取良好的方法，貼簽非常不牢，常有脫落。對用槽車裝運

溶媒也常有混用的現(xiàn)象，對此，F(xiàn)DA強調(diào)在取樣檢驗時，要從輸液管大量取樣，而不是像有些企業(yè)那樣，只從槽車的艙口取樣100毫升進行檢驗。

FDA檢查官要求企業(yè)對原材料供應商制定有整套的合格驗證制度，根據(jù)供應商的分析合格證書和附有的鑒別試驗來接收原材料。對原材料的批號的編制也有具體的要求。有時檢查官會要求查看入庫臺帳。

FDA檢查官對倉庫的建筑結構和通風照明等條件沒有具體的規(guī)定。一般說來，我國的一些老舊的未經(jīng)改造的制藥企業(yè)的倉庫條件都較差，對此，他們只要求整個內(nèi)部環(huán)境清潔整齊，沒有空調(diào)設備的要求有通風設備(一般是采用排風扇)。室內(nèi)有溫度和濕度計，定時觀察溫濕度，要求有記錄。要求有有效的防蟲和防鼠的措施(如窗戶或風扇進風口裝有紗窗，倉庫內(nèi)有滅蚊蠅燈，粘鼠板或電貓，并要求有記錄)。經(jīng)驗證明，企業(yè)改建或新建的倉庫，特別是比較現(xiàn)代化的倉庫結合較完善的管理能給予FDA檢查官以良好的印象。2．2生產(chǎn)工藝規(guī)程的執(zhí)行 2．2．1生產(chǎn)操作

FDA檢查官一般不對生產(chǎn)工藝的先進或落后，高效率或低效率的水平進行評價。由于FDA 4 檢查官不是全能的技術專家，不可能什么都懂，所以他們往往在檢查本人不太熟悉的生產(chǎn)工藝過程時，更多的是向企業(yè)了解和弄清整個工藝過程。通常要求藥廠出示該產(chǎn)品的生產(chǎn)工藝規(guī)程，以便對GMP的執(zhí)行過程進行深入的檢查。FDA官員要查看每個重點操作崗位上的原始記錄并對現(xiàn)場生產(chǎn)的實際運行情況核對。

由于原料藥的生產(chǎn)工藝范圍極廣，很難說FDA檢查官主要檢查哪一個步驟。對于合成藥物，通常是把注意力集中在生產(chǎn)的起著活性作用的關鍵的中間體的第一次反應步驟上。對于非合成藥物(如抗生素)則重點放在藥物的分離和提取的第一步上。但FDA檢查官通常從頭到尾都要進行檢查。特別是對生產(chǎn)開始的配料和后期的凈化過程，干燥和包裝等都很關注。2．2．2工藝驗證

FDA重視對生產(chǎn)工藝的驗證。在八十年代中期時FDA還沒有對藥物生產(chǎn)工藝的驗證進行檢查。直到1987年FDA才決定要對生產(chǎn)工藝的驗證進行檢查，并于1987年5月發(fā)布了一個有關原料藥的生產(chǎn)工藝驗證指南的文件：[工藝驗證的總的原則指南](Guideline 0n General Principles 0f Process Validation)。

此后，F(xiàn)DA對工藝驗證極為重視。

FDA認為，對一種已采用多年的生產(chǎn)工藝，企業(yè)應行一次回顧性的驗證(Retrospective validation)。生產(chǎn)工藝即使通過一次驗證，也不等于自動地永久性地驗證合格。如工藝變動，還要重做驗證?，F(xiàn)在，制藥企業(yè)都把工藝驗證方案和當前三個連續(xù)批的驗證結果作為一個獨立的章節(jié)編入向FDA申報的DMF文件中以供審查。2．3廠房，設備和設施

FDA對生產(chǎn)車間的廠房，設備和設施的先進性不作評論，但重視廠房的清潔，通風和防蟲措施(如所有窗戶均有窗紗)。對潔凈區(qū)(100，000或100，000加100級)則要求廠房密閉性良好。室內(nèi)外保持內(nèi)高外低的壓差。對設備管理，其中尤其重視設備(如發(fā)酵罐，反應儲罐，高壓滅菌釜，各種物料及壓縮空氣，工藝用水，蒸汽等的運送管線等)上應有的醒目的編碼號以保證操作工不會發(fā)生誤操作。FDA官員非常重視設備上的生產(chǎn)運行狀態(tài)標記牌和管線上的物料標記。

FDA官員進入生產(chǎn)車間，首先要檢查的是配料間的設備和物料管理，其管理的原則要求是和倉庫管理的要求一樣的，這在我國許多原料藥廠往往是個弱項。個別藥廠(特別是合成藥廠)，甚至沒有配料間，配料就在反應器旁邊進行，而且，往往比較凌亂，這是FDA所不能接受的。

其它檢查的要點還有：發(fā)酵法生產(chǎn)的種子制備室(其中包括無菌室，搖瓶間等)，無菌室，杯碟法生物效價測定間，生產(chǎn)工藝控制間，屬于手動控制的，重點檢查儀表的校驗和原始記錄的準確性；屬于半自動控制的，檢查自控儀表的校驗周期和記錄；屬于電腦系統(tǒng)全自動控制的，要對該系統(tǒng)的準確性的驗證提供充分的說明。

2．4設備的清洗

FDA檢查官對所有的生產(chǎn)關鍵設備，特別是反應器(如發(fā)酵罐，化學合成反應罐等)，容器(如中間體的轉運容器，儲罐，儲槽等)，分離設備(如液體過濾器，氣體過濾器，離心分離器，振動篩等)，結晶罐，干燥器和混合器等的清洗周期，清洗方法和清洗效果的驗證都很關注。原則上是要防止不同批的殘余物料和外來雜質對藥物產(chǎn)品造成交叉污染或感染。

除了在現(xiàn)場一般地觀察設備和容器的清洗情況外(一般無法在現(xiàn)場看到完整的清洗情況)FDA檢查官重點是檢查設備清洗的SOP，清洗驗證方法的文件和清洗的記錄，對于用同一套設備來進行周期性的交叉生產(chǎn)多種產(chǎn)品的生產(chǎn)設施，則要求更為嚴格。主要檢查清洗規(guī)程是否制定得完善，還要詢問這些清洗方法是否經(jīng)過驗證。

對于用水、洗滌劑或溶媒作為清洗劑清洗設備的效果，過去多是目測清洗液的澄清程度，或用TLC法測定洗液中殘留物的含量。從而做出殘留物是否洗凈的結論。近年來，許多藥廠 5 廣泛采用HPLC儀器分析的手段來測定洗液的殘留量。所以，現(xiàn)在不是有無殘留物存在的問題，而是殘留量的多少的問題。應根據(jù)具體情況，制定容許殘留量的標準。2．5批成品的混合

制劑生產(chǎn)商通常要求原料藥生產(chǎn)商的生產(chǎn)批量越大越好。他們認為，對兩個500Kg 批量的產(chǎn)品進行制劑生產(chǎn)要比1000 Kg批量的產(chǎn)品的檢驗耗費更多的人力和物力。

通常，來自一個結晶周期的或其它提煉周期的批量的產(chǎn)品被認為具有良好的均一性。

批混合的數(shù)量取決于混合器的能力。如果FDA官員發(fā)現(xiàn)原料藥生產(chǎn)商出廠產(chǎn)品的批量大于混合器的實際能力，則該藥廠就不會被FDA所接受，得不到批準，進口就會被擱置。FDA禁止把不合格的產(chǎn)品與合格的產(chǎn)品進行混合以取得合格品。

擁有足夠大的混合器的藥廠還必須向FDA檢查官顯示它的混合工藝確能產(chǎn)生均勻的批量產(chǎn)品，必要時要出示混合器的均一性的驗證記錄。但要注意，由于多數(shù)情況下成品是在混合器里同時進行干燥的，混合時間不能太長，太長會造成藥品的降解。2．6包裝及貼簽

FDA認為對原料藥的包裝和貼簽的控制應與對制劑藥的要求一樣地嚴格。包裝和貼簽的操作人員應該訓練有素并且可靠。如果，標簽是由電腦控制的，則該控制系統(tǒng)應經(jīng)過驗證。雖然，原料藥的貼簽發(fā)生錯誤造成的危害與制劑藥相比沒有那么大的危害，但會損害原料藥廠的聲譽。因此對原料藥成品包裝容器進行最后的控制檢查是十分必要的。2．7生產(chǎn)用水 2．7．1飲用水

用于抗生素原料藥的發(fā)酵階段或合成工藝的早期階段不需要高化學純度的條件下，采用從深井或地表的飲用水是可接受的。只要水質符合規(guī)定的標準。在美國，城市自來水必須符合聯(lián)邦環(huán)保局的要求?？蓮淖詠硭畯S取得合格的飲用水。當然，原料藥廠不可能去驗證自來水廠。FDA希望藥廠能提供定期檢查水質是否符合質量標準的資料。我國的大多數(shù)藥廠是使用城市自來水，一般都能從城市自來水廠取得定期的水質檢驗報告。FDA對水質的關注的重點是用水的微生物含量狀況，特別是致病菌，如大腸桿菌的含量。2．7．2純化水

純化水也廣泛用于原料藥生產(chǎn)。FDA重視對純化水系統(tǒng)的檢查。由于脫離子器(裝有離子交換樹脂的離子交換柱)，超微過濾和反滲析系統(tǒng)都存在著細菌滋長的問題，對這些系統(tǒng)應進行驗證和控制?？刂频姆椒òㄖ贫ㄋ|標準，當含菌量超標時采取的措施，完善的維修規(guī)程(如離子交換樹脂的再生和滅菌)，對用水的化學物和細菌含量制定標準并定期進行檢查。如果，水是用在工藝過程的較后的階段，如最后清洗濾餅，或原料藥是從水溶液中結晶析出，則對水質的要求要比對一般的純化水為高。這對于將要作為注射劑的原料藥廠來說是特別重要的。

FDA還認為大多數(shù)的純化水及注射用水系統(tǒng)，包括反滲析和超微過濾系統(tǒng)都存在著產(chǎn)生內(nèi)毒素的可能。如果要求原料藥成品無熱源或無菌，或是要用于制備注射劑產(chǎn)品，則應對工藝用水定期檢驗其內(nèi)毒素的含量。只做終點檢驗是不夠的。原料藥廠還應對注射用水處理系統(tǒng)以控制內(nèi)毒素的滋生。

FDA對用于生產(chǎn)獸用藥，非無菌口服制劑，局部外用藥如油膏劑，乳劑等制劑的原料藥，對用水質量的仍要進行嚴格的檢查。2．8質量控制實驗室(QC Laboratory)2．8．1檢查的重點中之重點一質量控制試驗室 FDA檢查官在為時4—5天的檢查中，其中之一的藥物審查化學家至少有3／5的時間用于對QC實驗室的檢查。

FDA對QC實驗室的清潔程度，化學分析室，微生物分析室，種子室(包括無菌室的無菌狀 6 況)，儀器分析室，穩(wěn)定性試驗和留樣室等都很重視。2．8．2儀器配備和檢驗方法

從FDA立場看，在原料藥廠中沒有比質量控制試驗室的管理更重要的了。事實上，在有些方面，它們對于原料藥廠比對制劑廠更為重要。制劑廠一般的典型作法是只對進廠的批量的原料藥按藥典的規(guī)定標準進行檢驗。在多數(shù)情況下，不足以充分檢查出可能存在的污染物。這些污染物因不同的藥廠及采用不同的生產(chǎn)工藝而不同。所謂污染物包括合成過程中未經(jīng)化學反應的起始物料，中間體，立體異構體，殘留溶媒，交叉污染物，微生物污染物及從工藝設備上帶下來的碎屑，通?？瓷先ナ呛谏陌唿c。原料藥廠的試驗室最了解污染物的種類和存在狀況，并保證這些污染物不會超過規(guī)定含量而送入生產(chǎn)制劑的制藥廠。如果送人了制藥廠，就說明這些污染物未被發(fā)現(xiàn)。

FDA檢察官對原料藥廠生產(chǎn)進行視察的一個基本項目是弄清該藥廠的試驗室是否擁有該廠的DMF或USP對該產(chǎn)品所要求的足夠的儀器設備。近年來，F(xiàn)DA對中國原料藥藥廠的檢查，每次是派兩名官員，其中一人是FDA有關部門的化學家(有時是美籍華人)，他的任務是對藥廠的儀器使用方法及檢驗方法做出評價。對USP的實驗方法，只要藥廠是遵照USP的規(guī)程進行檢驗，F(xiàn)DA不要求藥廠對分析方法進行驗證。

但在這種情況下，藥廠還是要做USP規(guī)定的有關驗證，例如，在采用HPLC法時要進行HPLC設備適用性的驗證?？偟恼f來，因為原料藥是單一的純物質，對分析方法的驗證并不具有制劑藥所具有的同樣的問題。制劑藥常在藥物活性物質之外含有幾種不同的物質。有些物質會干擾分析的方法。不同的制劑藥廠有不同的添加物質。因此同一種藥物的USP分析方法或申請書上所采用的分析方法不能適用于所有的制劑藥。通常原料藥產(chǎn)品就沒有這個問題。

2．8．3工藝過程的分析控制

我國的原料藥廠的工藝過程分析一般是在生產(chǎn)該產(chǎn)品的車間里的試驗室里進行的。有些藥廠的工藝過程控制中要用HPLC設備進行分析的項目則是由車間取樣送交質控試驗室分析的。FDA認為，工藝過程分析可以在生產(chǎn)車間進行，也可以在質控實驗室進行。但FDA要求這些分析應由稱職的人員和采用完整的分析儀器來進行，并有完整的分析結果的記錄．

在有些藥物產(chǎn)品的生產(chǎn)工藝里，可能被帶入成品的不純物是在工藝過程中取樣而不是從成品取樣進行檢查的。在這種情況下，較為恰當?shù)霓k法是由質控試驗室來做分析試驗。對于一個一批又一批地接續(xù)生產(chǎn)同一個中間體，然后又生產(chǎn)另一個中間體的藥廠，則較為恰當?shù)淖龇ㄈ允怯绍囬g控制試驗室在放行到下一步工藝之前進行檢驗。FDA強調(diào)，不論由何單位進行檢驗。QC部門都應對工藝過程的每一個重要的步驟的物料的批準合格或不合格負責。2．8．4化學藥品和試劑，試液的配制

FDA檢查官注重檢查化學品的使用的管理(有效期，開始使用日期等)；成品和原材料的樣品的登記和管理；配制試液的標簽(注明試液名稱，配制日期，配制人簽字等)；配制標準操作規(guī)程及原始記錄等。還有微生物試驗的培養(yǎng)基配置，滅菌記錄等。2．8．5穩(wěn)定性試驗和留樣制度

FDA要求原料藥，特別是抗生素原料藥必須規(guī)定有效期(或復驗期)，并應編制有書面的穩(wěn)定性試驗方案和試驗記錄?，F(xiàn)在，我國大多數(shù)原料藥廠都為原料藥產(chǎn)品制定了穩(wěn)定性試驗的制度并付諸實施。

FDA官員要查看加速穩(wěn)定性試驗和長期穩(wěn)定性試驗，作穩(wěn)定性試驗的樣品的包裝和儲存的方式(它應模擬市場方式)，穩(wěn)定性試驗的溫度和濕度的條件，測定的方法和原始記錄。

我國原料藥廠一般都在QC實驗室里設有留樣間。有的藥廠是把長期穩(wěn)定性試驗的樣品也放在留樣間里。FDA檢查官要檢查留樣的方式和記錄。2．9設備和儀器的校驗

FDA對藥廠的設備和儀器的校驗非常重視。

FDA認為如果一個原料藥廠對工藝設備儀器和試驗室設備儀器不進行校驗，則說明該藥廠對它的操作失控，因此是不符合要求的。何況，沒有一套正規(guī)的校驗控制系統(tǒng)，該藥廠就不能對它的生產(chǎn)工藝做出符合要求的驗證。凡是用于工藝驗證的儀器本身就應該是經(jīng)過校驗和驗證的。結果是該藥廠對該原料藥的申請有得不到批準的危險。

藥廠需要定期進行校驗的常用的測量儀表有溫度計(包括自動記錄溫度計)，溫濕度計，壓力計(包括自動記錄壓力計)，液體流速計，空氣流量計，液位計等；計量設備和器具如 地磅，磅稱，粗天平，普通天平，分析天平以及高精密度天平等；工藝過程控制和質控試驗室用的pH計，滴定管，移液管等；儀器分析用的UV，IR，HPLC和GC等設備。FDA檢察官首先注意的是儀器設備上是否貼有校驗合格的標志，查看有關設備儀器校驗的SOP和使用與校驗的原始記錄。通常，我國的原料藥廠的度量衡儀器都是由藥廠專設的儀器計量室(大多是從屬設備管理部門)負責定期校驗并簽發(fā)校驗合格證。技術性高的儀器設備如UV，IR，HPLC和GC等設備則一般是委托地方政府的計量管理部門進行校驗。凡是發(fā)現(xiàn)重要的儀器設備沒有進行校驗，或提供不出完整的文件記錄說明，F(xiàn)DA檢察官都認為是嚴重的問題。2．10電腦控制系統(tǒng)的驗證

在發(fā)達國家的原料藥廠里對生產(chǎn)的啟動，監(jiān)測，調(diào)節(jié)等生產(chǎn)工藝步驟都已普遍采用電腦控制系統(tǒng)來實現(xiàn)自動化操作。我國也有少數(shù)原料藥廠采用了這種先進的手段。因此，電腦系統(tǒng)的工作準確性即可靠性是非常重要的。要保證不發(fā)生未經(jīng)認可的或疏忽大意造成的變化，并能提供充分的工藝信息，F(xiàn)DA要求對電腦系統(tǒng)應進行驗證以保證其運行的準確性及可靠性。

對電腦系統(tǒng)的硬件和軟件都要進行以下五個方面的考慮：

ix．確定電腦系統(tǒng)要執(zhí)行的任務，并將此任務配置人電腦系統(tǒng)。

x．鑒定電腦系統(tǒng)的操作限定值，并使其與標準操作法所規(guī)定的限定值一致。Xi．測試整個電腦系統(tǒng)。

Xii．做好文件記錄，并建立起監(jiān)測各種變化及重新驗證的系統(tǒng)。

FDA認為電腦系統(tǒng)主要應由作為終端用戶的藥廠負責進行驗證，并要保證它應用于原料藥生產(chǎn)操作的適用性。但許多實際的測試工作和文件記錄還是要由電腦系統(tǒng)供應商來做。電腦系統(tǒng)的驗證記錄數(shù)據(jù)應由終端用戶提供。FDA檢查官要檢查這些記錄數(shù)據(jù)。FDA為電腦控制系統(tǒng)的驗證提供了以下一些文件指南： xiv．“藥物生產(chǎn)采用電腦控制系統(tǒng)的檢查指南”

(Guide t0 Inspection 0f Computerized System in Drag Processing)1983年2月公布。

XV．“符合性政策指南”中的7132a。07、08、ll、12及15章節(jié)(Compliance Policy Guide)1987年公布。Xvi．“軟件發(fā)展活動的技術報告”

(Technical Report on Software Development Activities)1987年公布。3．會談形式的檢查

一般在進行約占總時間1／3的現(xiàn)場檢查后(有時是會談和現(xiàn)場檢查交叉進行)FDA檢查官與藥廠的有關負責人員會談。會談檢查的重點和需要提供的文件如下： 3．1批生產(chǎn)紀錄

FDA檢察官要隨意抽查一個當前生產(chǎn)批的全套原始記錄，并以該批記錄作為判斷藥廠的GMP管理水平的重要依據(jù)。FDA要求原料藥廠生產(chǎn)原料藥制定有完整的工藝規(guī)程和操作規(guī)程。FDA檢查官主要是通過批記錄來了解藥廠的工藝規(guī)程及操作規(guī)程的執(zhí)行情況和對每一個生產(chǎn)步驟里的工藝參數(shù)和原料投人的控制情況進行追溯。

FDA十分重視在每一個步驟的記錄上由操作者和監(jiān)督人的簽字的制度。認真檢查原始記 8 錄填寫得是否清晰和正確。對任何不規(guī)范的數(shù)據(jù)的修改，一發(fā)現(xiàn)，就要作為問題記入483表。

3．2成品質量標準和檢驗方法

如果該申請產(chǎn)品原來就是采用USP當前版的標準和方法來進行檢驗，一般提供數(shù)批產(chǎn)品檢驗報告單(COA)和根據(jù)USP制定的質量檢驗的SOP文件給FDA檢查官進行檢查。如果原來是采用其它的藥典標準如中國藥典(CP)，英國藥典(BP)或歐洲藥典(EP)，則要提供有關檢驗方法和USP的對比驗證資料和數(shù)據(jù)。由于我國的原料藥廠的產(chǎn)品大量出口到世界各地，其質量標準大都是根據(jù)客戶的要求而定，在許多情況下，并不是采用USP的標準和方法，或者是只采用USP的部分標準，或相同的標準但檢驗方法不同。FDA的原則是：希望申請方最好采用USP標準和方法，這有便于美方進行質量控制，F(xiàn)DA不規(guī)定一定要采用USP標準和方法。但要求提供充分的對比驗證資料來說明其它的標準和方法能滿足USP對產(chǎn)品質量的要求。

為了避免不必要的麻煩，我國藥廠對申請出口美國的產(chǎn)品盡量采用USP當前版的標準和方法。

3．3驗證文件

FDA要了解藥廠的有關驗證文件如下：

*工藝驗證一工藝驗證的方案和驗證記錄數(shù)據(jù)

*設備清洗的驗證一重點設備(如發(fā)酵罐或反應器)的清洗驗證方案和驗證記錄數(shù)據(jù)。*消毒(滅菌)的驗證一重點是發(fā)酵罐或反應器，高壓滅菌釜，無菌室，無菌區(qū)等的 消毒驗證方案和驗證記錄數(shù)據(jù)。

*電腦制動控制系統(tǒng)的驗證方案和驗證記錄數(shù)據(jù)。

*其它方面，如質控試驗室的重點儀器分析設備(HPLC，GC，UV，IR等)的適合性試驗，純水制備系統(tǒng)的驗證，發(fā)酵生產(chǎn)的空氣過濾(超微)過濾器和無菌區(qū)的空調(diào)系統(tǒng)的空氣過濾器的除菌效果驗證，成品干燥混合器的混合均勻度的驗證。3．4標準操作規(guī)程(SOP)FDA認為符合GMP要求的藥廠應該制定有完整的SOP(以及SMP)文件。FDA可能只是重點檢查幾份與問題有關的文件，但藥廠主動出示全套SOP及SMP的文件的目錄清單(英文譯文或中英對照)供FDA檢查官審閱也有助于對藥廠的管理水平的了解。

凡是在會談中與上述的FDA檢查所涉及的事項有關的SOP，一般FDA都要求我方提供SOP文件當場檢查。某些重要的SOP如藥廠職工的技術培訓和GMP培訓，藥廠的對用戶的投訴的處理，對不合格品的重加工的規(guī)程，藥廠的自查制度等一般都已編人DMF(大多數(shù)是SOP的形式，英文譯文或中英文對照)可供FDA檢查官隨時查閱。在檢查中涉及的某些行政管理上的問題，還要提供有關的標準管理規(guī)程(SMP)文件。

五、總結會一發(fā)給483表

在全部檢查結束后，F(xiàn)DA檢查官用半天時間單方面對檢查中發(fā)現(xiàn)的問題進行匯總，分析和研究。根據(jù)檢查結果填寫“FDA483表”，該表是FDA印制的表格式的文件。

從理論上說，如果一個問題都沒有發(fā)現(xiàn)，F(xiàn)DA就不必發(fā)給此表，并預示著下一步的正式批準沒問題了。這種情況是罕見的，只有極個別的制劑藥廠在兩年一度的復查時有過。通常FDA會提出不少的問題(作者所接觸過的場合一般是10—20條問題)。問題提得越多，固然意味著藥廠的缺點較多，獲得批準的難度加大。但FDA主要看問題的大小和性質，一般說來，不能有嚴重的大問題，“致命性的問題”(Lethal problem)或“不可接受的問題”(unacceptable problem)。即使只有一個重大的問題也可能會導致通不過。

“FDA檢查無小事”(這是作者的體會)。大的問題固然是大事，例如，沒有做過工藝驗證，沒有制定SOP或消毒滅菌不徹底造成交叉感染，特別是某些文件被FDA發(fā)現(xiàn)弄虛作假，與實 9 際對不上號等等。但有些看起來似小事，卻成為嚴重的問題，如對原始記錄的不規(guī)范的改動(涂改)，沒按照規(guī)定放置清潔工具，在配料間原料包裝袋上的合格證脫落等。某藥廠在向FDA檢查官出示的工藝規(guī)程文件里的幾個參數(shù)發(fā)現(xiàn)被手寫涂改，盡管藥廠作了種種解釋并承認錯誤，但FDA檢查官認為這是不可接受的問題，從而導致檢查未能通過的惡果。

FDA檢查官寫好483表后召集我方人員開會。先把483表中列出的觀察結果和問題，逐條逐字讀給我方人員聽。然后進行討論，我方可以就不能接受的提法提出解釋(看法，解說或辯解)，如果FDA檢查官認為解釋有理，一般會對所提的問題進行修改或取消。在雙方取得一致意見后，我方做出接受觀察結果的表態(tài)。FDA檢查官在該表上簽字。立即復印該文件，一份交給藥廠，另一份他們回國后送交FDA主管部門。

FDA要求藥廠對提出的問題盡快(一般在兩周內(nèi)，至多不超過一個月，根據(jù)情況而定)做出書面答復，其中要求提供明確的較詳細的改進措施和解決問題的期限，及早報送FDA地區(qū)辦公室。FDA檢查官在回國后根據(jù)藥廠的改進報告寫出一份非常詳細的檢查報告送交FDA有關主管部門(如新藥評價中心，獸用藥評價中心等)。

六、最后的決定——批準(或不批準)的通知函

按FDA的規(guī)定，F(xiàn)DA檢查官無權對是否批準做出決定。FDA檢查官在給主管的報告中非?？陀^地說明一切情況和存在問題，藥廠的態(tài)度和改正的措施，對該藥廠是否可以得到批準會提出個人的建議。它對主管做出批準或不批準的決定是具有關鍵性的影響的。

在檢查結束后FDA檢查官不會向藥廠表示生產(chǎn)設施是否合乎FDA要求和取得最后批準的可能性。只有極個別的例子，一位資深的年過古稀的FDA檢查官在結束檢查時，主動伸出手來與藥廠主管握手并說該藥廠是一個令人滿意的藥廠。表示祝賀。但那是20年前的事情(我國第二個申請得到批準的藥廠)。近年來，F(xiàn)DA檢查官至多是在藥廠的美方代理人私下要求下，間接透露483表中沒有不可接受的問題。提交報告后的3—6個月，F(xiàn)DA有關主管部門做出最后批準(或不批準)的決定，并給出書面通知函。

FDA要求得到批準的藥廠在批準后每年一次向FDA提出報告，說明DMF中的任何可能發(fā)生的變動(如重要設備和廠房，原材料，工藝路線，工藝參數(shù)等的變動)，如果沒有變動也要向FDA報告無變動。

一般在接到批準通知函后，得到批準的原料藥產(chǎn)品就可以出口到美國，美國海關給予放行。

FDA規(guī)定，凡是第一次檢查沒有得到批準的藥廠，要兩年后才能再次提出申請檢查。FDA規(guī)定，得到批準的藥廠每兩年要接受一次復查，復查的程序和初查是一樣的。如果復查沒有通過就要取消資格，又要等兩年后才能申請檢查。

第三篇：美國FDA關于制劑藥廠cGMP的檢查指南-93-10

Dosage Form Drug Manufacturers cGMPs(10/93)GUIDE TO INSPECTIONS OF DOSAGE FORM DRUG MANUFACTURER'S-CGMPR'S

Note: This document is reference material for investigators and other FDA personnel.The document does not bind FDA, and does no confer any rights, privileges, benefits, or immunities for or on any person(s).I.INTRODUCTION

This document is intended to be a general guide to inspections of drug manufacturers to determine their compliance with the drug CGMPR's.This guide should be used with instructions in the IOM, other drug inspection guides, and compliance programs.A list of the inspection guides is referenced in Chapter 10 of the IOM.Some of these guides are:

o Guide to Inspections of Bulk Pharmaceutical Chemicals.o Guide to Inspections of High Purity Water Systems.o Guide to Inspections of Pharmaceutical Quality Control Laboratories.o Guide to Inspections of Microbiological Pharmaceutical Quality Control Laboratories.o Guide to Inspections of Lyophilization of Parenterals.o Guide to Inspections of Validation of Cleaning Processes.o Guide to Inspections of Computerized Systems in Drug Processing.o Guideline on General Principles of Process Validation.II.CURRENT GOOD

MANUFACTURING PRACTICE

REGULATIONS

Prescription vs.Non-prescription

All drugs must be manufactured in accordance with the current good manufacturing practice regulations otherwise they are considered to be adulterated within the meaning of the FD&C Act, Section 501(a)(2)(B).Records relating to prescription drugs must be readily available for review in accordance with Sec.704(a)(1)(B)of the FD&C Act.If the product is an OTC drug which is covered by an NDA or ANDA, FDA may review, copy and verify the records under Sec.505(k)(2)of the FD&C Act.However, if the product is an OTC drug for which there is no application filed with FDA, the firm is not legally required to show these records to the investigator during an inspection being conducted under Section 704 of the FD&C Act.Nonetheless, all manufacturers of prescription and OTC drugs must comply with the drug CGMPR requirements, including those involving records.The investigator should review these records as part of the inspection in determining the firm's compliance with the CGMP regulations.On rare occasions, a firm may refuse to allow review of OTC records stating they are not legally required to.While the firm may be under no legal obligation to permit review of such records, this does not relieve the firm of its statutory requirement to comply with the good manufacturing practices under section 501(a)(2)(B)of the Food Drug and Cosmetic Act, including the requirements for maintaining records.If a firm refuses review of OTC records, the investigator should determine by other inspectional means the extent of the firm's compliance with CGMPR's.Inspectional observations and findings that CGMPR's are not being followed are to be cited on a List of Inspectional Observations, FDA-483, for both prescription and non-prescription drugs.Organization and Personnel [21 CFR 211 Subpart B]

The firm must have a quality control department that has the responsibility and authority as described in the referenced CFR.The quality control department must maintain its independence from the production department, and its responsibilities must be in writing.Obtain the name, title and individual responsibilities of corporate officers and other key employees as indicated in the IOM.In the drug industry, an employee's education and training for their position has a significant impact on the production of a quality product.Report whether the firm has a formalized training program, and describe the type of training received.The training received by an employee should be documented.Quality control must do product annual review on each drug manufactured, and have written annual review procedures.Review these reports in detail.This report will quickly let you know if the manufacturing process is under control.The report should provide a summary all lots that failed in-process or finished product testing, and other critical factors.Investigate any failures.Quality control must validate the manufacturing process for each drug manufactured.Review and evaluate this data.Buildings and Facilities [21 CFR 211 Subpart C]

Review the construction, size, and location of plant in relation to surroundings.There must be adequate lighting, ventilation, screening, and proper physical barriers for all operations including dust, temperature, humidity, and bacteriological controls.There must be adequate blueprints which describe the high purity water, HEPA, and compressed air systems.The site must have adequate locker, toilet, and hand washing facilities.The firm must provide adequate space for the placement of equipment and materials to prevent mix-ups in the following operations:

o receiving, sampling, and storage of raw materials;

o manufacturing or processing;

o packaging and labeling;

o storage for containers, packaging materials, labeling, and finished products;

o production and control laboratories.Equipment [21 CFR 211 Subpart D]

Review the design, capacity, construction, and location of equipment used in the manufacturing, processing, packaging, labeling, and laboratories.Describe the manufacturing equipment including brief descriptions of operating principles.Consider the use of photographs, flow charts, and diagrams to supplement written descriptions.New equipment must be properly installed, and operate as designed.Determine if the equipment change would require FDA pre-approval and/or revalidation of the manufacturing process.The equipment must be cleaned before use according to written procedures.The cleaning must be documented and validated.The equipment should not adversely effect the identity, strength, quality, or purity of the drug.The material used to manufacture the equipment must not react with the drug.Also, lubricants or coolants must not contaminate the drug.The equipment should be constructed and located to ease cleaning, adjustments, and maintenance.Also, it should prevent contamination from other or previous manufacturing operations.Equipment must be identified as to its cleaning status and content.The cleaning and maintenance of the equipment are usually documented in a log book maintained in the immediate area.Determine if the equipment is of suitable capacity and accuracy for use in measuring, weighing, or mixing operations.If the equipment requires calibration, they must have a written procedure for calibrating the equipment and document the calibration.Components and Product Containers [21 CFR 211 Subpart E]

Inspect the warehouse and determine how components, drug product containers, and closures are received, identified, stored, handled, sampled, tested, and approved or rejected.They must have written procedures which describe how these operations are done.Challenge the system to decide if it is functioning correctly.If the handling and storage of components are computer controlled, the program must be validated.The receiving records must provide traceability to the component manufacturer and supplier.The receiving records for components should contain the name of the component, manufacturer, supplier if different from the manufacturer, and carrier.In addition, it should include the receiving date, manufacturer's lot number, quantity received, and control number assigned by the firm.Check sanitary conditions in the storage area, stock rotation practices, retest dates, and special storage conditions(protection from light, moisture, temperature, air, etc.).Inspect glandular and botanical components for insect infestation.Components or finished product adulterated by rodents, insects, or chemicals must be documented and submitted for seizure.Collect the evidence even if the firm plans to voluntarily destroy the product.Be alert for components, colors, and food additives that may be new drug substances, appear to have no use in the plant or appear to be from an unknown supplier.Check the colors against the Color Additives Status List in the IOM Determine if the color is approved for its intended use, and required statements are declared on the drug label.Components might be received at more than one location.Components must be handled in accordance with the drug CGMP's including components used in the research and development lab.Determine how components are identified after receipt and quarantined until released.Components must be identified so the status(quarantine, approved, or rejected)is known.Review the criteria for removing components from quarantine and challenge the system.Determine what records are maintained in the storage area to document the movement of components to other areas, and how rejected components handled.The component container has an identification code affixed to it.This unique code provides traceability from the component manufacturer to its use in the finished product.Review the sampling and testing procedures for components, and the process by which approved materials are released for use.Decide if these practices are adequate and followed.Determine the validity, and accuracy of the firm's inventory system for drug components, containers, closures and labeling.Challenge the component inventory records by weighing a lot and comparing the results against the quantity remaining on the inventory record.Significant discrepancies in these records should be investigated.Evaluate the following to determine whether the firm has shown that the containers and closures are compatible with the product, will provide adequate protection for the drug against deterioration or contamination, are not additive or absorptive, and are suitable for use:

o Specifications for containers, closures, cotton filler, and desiccant, etc.o What tests or checks are made(cracks, glass particles, durability of material, metal particles in ointment tubes, compliance with compendium specifications, etc.).o Cleaning procedures and how containers are stored.o Handling of preprinted containers.Are these controlled as labeling, or as containers? The firm must review the labeling for accuracy.Production and Process Controls [21 CFR Subpart F]

1.Critical Manufacturing Steps [21 CFR 211.101]

Each critical step in the manufacturing process shall be done by a responsible individual and checked by a second responsible individual.If such steps in the processing are controlled by automatic mechanical or electronic equipment, its performance should be verified.Critical manufacturing steps include the selection, weighing, measuring and identifying of components, and addition of components during processing.It includes the recording of deviations from the batch record, mixing time and testing of in-process material, and the determination of actual yield and percent of theoretical yield.These manufacturing steps are documented when done, and not before or after the fact.2.Equipment Identification [21 CFR 211.105]

All containers and equipment used in to manufacture a drug should be labeled at all times.The label should identify the contents of the container or equipment including the batch number, and stage of processing.Previous identification labels should be removed.The batch should be handled and stored to prevent mixups or contamination.3.In-Line and Bulk Testing [21 CFR 211.110]

To ensure the uniformity and integrity of products, there shall be adequate in-process controls, such as checking the weights and disintegration time of tablets, the fill of liquids, the adequacy of mixing, the homogeneity of suspensions, and the clarity of solutions.Determine if in-process test equipment is on site and the specified tests are done.Be alert for prerecording of test results such as tablet weight determinations.The bulk drug is usually held in quarantine until all tests are completed before it is released to the packaging and labeling department.However, the testing might be done after packaging.product.4.Actual Yield [21 CFR 211.103]

Determine if personnel check the actual against the theoretical yield of each batch of drug manufactured.In the event of any significant unexplained discrepancies, determine if there is a procedure to prevent distribution of the batch in question, and related batches.5.Personnel Habits

Observe the work habits of plant personnel.Determine:

Their attitudes and actions involving the jobs they perform.(Careless, lackadaisical, disgruntled, etc.).Their dress.(Clean dresses, coats, shirts and pants, head coverings, etc.If proper equipment is used for a given job or whether short cuts are taken(i.e.use of hands and arms to mix or empty trays of drug components).If there are significant written or verbal language barriers that could affect their job performance.Tablet and Capsule Products

Become familiar with the type of equipment and its location in the tableting operation.The equipment may include rotary tableting machines, coating and polishing pans, punches and dies, etc.The equipment should be constructed and located to facilitate maintenance and cleaning at the end of each batch or at suitable intervals in the case of a continuous batch operation.If possible, observe the cleaning and determine if the cleaning procedure is followed.The ingredients in a tablet are the active ingredient, binders, disintegrators, bases, and lubricants.The binder is added to the batch to keep the tablet together.Excess binder will make the tablet too hard for use.The disintegrator is used to help disintegration of the tablet after administration.The base should be an inert substance which is compatible with the active ingredient and is added to provide size and weight.The lubricant helps in the flow of granulated material, prevents adhesion of the tablet material to the surface of punches and dies, and helps in tablet ejection from the machine.Tablets and capsules are susceptible to airborne contamination because of the manipulation of large quantities of dry ingredients.To prevent cross-contamination in the tableting department, pay close attention to the maintenance, cleaning, and location of equipment, and the storage of granulations and tablets.To prevent cross-contamination, the mixing, granulation, drying and/or tableting operation should be segregated in enclosed areas with its own air handling system.Determine what precautions are taken to prevent cross-contamination.When cross-contamination is suspect, investigate the problem and collect in-line samples(INV)and official samples of the suspect product.Determine what temperature, humidity, and dust collecting controls are used by the firm in manufacturing operations.Lack of temperature and humidity controls can affect the quality of the tablet.Observe the actual operation of the equipment and determine whether powders or granulations are processed according to the firm's specifications.The mixing process must be validated.The drying ovens should have their own air handling system which will prevent cross-contamination.Does the firm record drying time/temperature and maintain recording charts including loss on drying test results? Review the in-line tests performed by production and/or quality control.Some in-process tests are tablet weight, thickness, hardness, disintegration , and friability.Evaluate the disposition of in-process samples.Capsules may be either hard, or soft type.They are filled with powder, beads, or liquid by machine.The manufacturing operation of powders for capsules should follow the same practice as for tablets.Determine manufacturing controls used, in-line testing, and basis for evaluating test results for the filling operations.Sterile Products

Typically, a sterile drug contains no viable microorganisms and is non-pyrogenic.Drugs for intravenous injection, irrigation, and as ophthalmic preparations, etc., meet this criteria.In addition, other dosage forms might be labeled as sterile.For instance, an ointment applied to a puncture wound or skin abrasion.Parenteral drugs must be non-pyrogenic, because the presence of pyrogens can cause a febrile reaction in human beings.Pyrogens are the products of the growth of microorganisms.Therefore, any condition that permits bacterial growth should be avoided in the manufacturing process.Pyrogens may develop in water located in stills, storage tanks, dead legs, and piping, or from surface contamination of containers, closures, or other equipment.Parenterals may also contain chemical contaminants that will produce a pyretic response in humans or animals although there are no pyrogens present.There are many excellent reference materials which should be reviewed before the inspection.Some of these are the “Guideline on Sterile Drug Products Produced by Aseptic Processing,” and chapter 84 on pyrogens in the Remington's Pharmaceutical Sciences.Determine and evaluate the procedures used to minimize the hazard of contamination with microorganisms and particulates of sterile drugs.o Personnel

Review the training program to ensure that personnel performing production and control procedures have experience and training commensurate with their intended duties.It is important that personnel be trained in aseptic procedures.The employees must be properly gowned and use good aseptic techniques.o Buildings

The non-sterile preparation areas for sterile drugs should be controlled.Refer to Subpart C of the proposed CGMPR's for LVP's;however, deviations from these proposed regulations are not necessarily deviations from the CGMPR's.Evaluate the air cleanliness classification of the area.For guidance in this area, review Federal Standard #209E entitled “Airborne Particulate Cleanliness Classes in Cleanrooms and Clean Zones.” Observe the formulation practices or procedures used in the preparation areas.Be alert for routes of contamination.Determine how the firm minimizes traffic and unnecessary activity in the preparation area.Determine if filling rooms and other aseptic areas are constructed to eliminate possible areas for microbiological/particulate contamination.For instance, dust-collecting ledges, porous surfaces, etc.Determine how aseptic areas are cleaned and maintained.1.Air

Air supplied to the non-sterile preparation or formulation area for manufacturing solutions prior to sterilization should be filtered as necessary to control particulates.Air being supplied to product exposure areas where sterile drugs are processed and handled should be high efficiency particulate air(HEPA)filtered under positive pressure.Review the firm's system for HEPA filters, determine if they are certified and/or Dioctyl Phthalate(DOP)tested and frequency of testing.Review the compressed air system and determine if it is filtered at the point of use to control particulates.Diagrams of the HEPA filtered and compressed air systems should be reviewed and evaluated.2.Environmental Controls

Specifications for viable and non-viable particulates must be established.Specifications for viable particulates must include provisions for both air and surface sampling of aseptic processing areas and equipment.Review the firm's environmental control program, specifications, and test data.Determine if the firm follows its procedure for reviewing out-of-limit test results.Also, determine if review of environmental test data is included as a part of the firm's release procedures.Note: In the preparation of media for environmental air and surface sampling, suitable inactivating agents should be added.For example, the addition of penicillinase to media used for monitoring sterile penicillin operations and cephalosporin products.o Equipment

Determine how the equipment operates including the cleaning and maintenance practices.Determine how equipment used in the filling room is sterilized, and if the sterilization cycle has been validated.Determine the practice of re-sterilizing equipment if sterility has been compromised.Determine the type of filters used.Determine the purpose of the filters, how they are assembled, cleaned, and inspected for damage.Determine if a microbial retentive filter, and integrity testing is required.o Water for Injection

Water used in the production of sterile drugs must be controlled to assure that it meets U.S.P.specifications.Review the firm's water for injection production, storage, and delivery system.Determine that the stills, filters, storage tanks, and pipes are installed and operated in a manner that will not contaminate the water.Evaluate the firm's procedures and specifications that assure the quality of the water for injection.As reference material, review the “FDA Guide to Inspecteons of High Purity Water Systems” before initiating an inspection.o Containers and Closures

Determine how containers and closures are handled and stored.Decide if the cleaning, sterilization, and depyrogenization are adequate, and have been validated.o Sterilization

1.Methods

Determine what method of sterilization is used.A good source of reference material on validation of various sterilization processes is the Parenteral Drug Association Technical Reports.For instance, Technical Report #1 covers “Validation of Steam Sterilization Cycles.” Review and evaluate the validation data whatever the method employed.If steam under pressure is used, an essential control is a mercury thermometer and a recording thermometer installed in the exhaust line.The time required to heat the center of the largest container to the desired temperature must be known.Steam must expel all air from the sterilizer chamber to eliminate cold spots.The drain lines should be connected to the sewer by means of an air break to prevent back siphoning.The use of paper layers or liners and other practices which might block the flow of steam should be avoided.Charts of time, temperature, and pressure should be filed for each sterilizer load.If sterile filtration is used, determine the firm's criteria for selecting the filter and the frequency of changing.Review the filter validation data.Determine if the firm knows the bioburden of the drug, and examine their procedures for filter integrity testing.Filters might not be changed after each batch is sterilized.Determine if there is data to justify the integrity of the filters for the time used and that “grow through” has not occurred.If ethylene oxide sterilization is used, determine what tests are made for residues and degradation.Review the ETO sterilization cycle including preconditioning of the product, ETO concentration, gas exposure time, chamber and product temperature, and chamber humidity.2.Indicators

Determine the type of indicator used to assure sterility.Such as, lag thermometers, peak controls, Steam Klox, test cultures, biological indicators, etc.Caution: When spore test strips are used to test the effectiveness of ethylene oxide sterilization, be aware that refrigeration may cause condensation on removal to room temperature.Moisture on the strips converts the spore to the more susceptible vegetative forms of the organism which may affect the reliability of the sterilization test.The spore strips should not be stored where they could be exposed to low levels of ethylene oxide.If biological indicators are used, review the current U.S.P.on sterilization and biological indicators.In some cases, testing biological indicators may become all or part of the sterility testing.Biological indicators are of two forms, each of which incorporates a viable culture of a single species of microorganism.In one form, the culture is added to representative units of the lot to be sterilized or to a simulated product that offers no less resistance to sterilization than the product to be sterilized.The second form is used when the first form is not practical as in the case of solids.In the second form, the culture is added to disks or strips of filter paper, or metal, glass, or plastic beads.During the inspection of a firm which relies on biological indicators, review background data complied by the firm to include:

o Surveys of the types and numbers of organisms in the product before sterilization.o Data on the resistance of the organism to the specific sterilization process.o Data used for selecting the most resistant organism and its form(spore or vegetative cell).o Studies of the stability and resistance of the selected organism to the specific sterilization process.o Studies on the recovery of the organism used to inoculate the product.o If a simulated product or surface similar to the solid product is used, validation of the simulation or similarity.The simulated product or similar surface must not affect the recovery of the numbers of indicator organisms applied.o Validation of the number of organisms used to inoculate the product, simulated product, or similar surface, to include stability of the inoculum during the sterilization process.Since qualified personnel are crucial to the selection and application of these indicators, review their qualifications including experience dealing with the process, expected contaminants, testing of resistance of organisms, and technique.Review the firm's instructions regarding use, control and testing, of the biological indicator by product including a description of the method used to demonstrate presence or absence of viable indicator in or on the product.Review the data used to support the use of the indicator each time it is used.Include the counts of the inoculum used;recovery data to control the method used to demonstrate the sterilization of the indicator organism;counts on unprocessed, inoculated material to

indicate the stability of the inoculum for the process time;and

results of sterility testing specifically designed to demonstrate the presence or absence of the indicator organism for each batch or filling operation.In using indicators, you must assure yourself that the organisms are handled so they don't contaminate the drug manufacturing area and product.3.Filled Containers

Evaluate how the filled vials or ampules leave the filling room.Is the capping or sealing done in the sterile fill area? If not, how is sterility maintained until capped?

Review the tests done on finished vials, ampules, or other containers, to assure proper fill and seal.For instance, leak and torque tests.Review examinations made for particulcte contamination.You can quickly check for suspected particulate matter by using a polariscope.Employees doing visual examinations on line must be properly trained.If particle counts are done by machine, this operation must be validated.4.Personnel Practices

Check how the employees sterilize and operate the equipment used in the filling area.Observe filling room personnel practices.Are the employees properly dressed in sterile gowns, masks, caps, and shoe coverings? Observe and evaluate the gowning procedures, and determine if good aseptic technique is maintained in the dressing and filling rooms.Check on the practices after lunch and other absences.Is fresh sterile garb supplied, or are soiled garments reused?

Determine if the dressing room is next to the filling area and how employees and supplies enter the sterile area.o Laboratory Controls

For guidance on how to inspect micro and chemistry labs, review the “FDA Guide to Inspections of Pharmaceutical Quality Control Laboratories” and “FDA Guide to Inspections of Microbiological Pharmaceutical Quality Control Laboratories.”

1.Retesting for Sterility See the USP for guidance on sterility testing.Sterility retesting is acceptable provided the cause of the initial non-sterility is known, and thereby invalidates the original results.It cannot be assumed that the initial sterility test failure is a false positive.This conclusion must be justified by sufficient documented investigation.Additionally, spotty or low level contamination may not be identified by repeated sampling and testing.Review sterility test failures and determine the incidence, procedures for handling, and final disposition of the batches involved.2.Retesting for Pyrogens

As with sterility, pyrogen retesting can be performed provided it is known that the test system was compromised.It cannot be assumed that the failure is a false positive without documented justification.Review any initial pyrogen test failures and determine the firm's justification for retesting.3.Particulate Matter Testing

Particulate matter consists of extraneous, mobile, undissolved substances, other than gas bubbles, unintentionally present in parenteral solutions.Cleanliness specifications or levels of non-viable particulate contamination must be established.Limits are usually based on the history of the process.The particulate matter test procedure and limits for LVP's in the U.S.P.can be used as a general guideline.However, the levels of particulate contamination in sterile powders are generally greater than in LVP's.LVP solutions are filtered during the filling operation.However, sterile powders, except powders lyophilized in vials, cannot include filtration as a part of the filling operation.Considerable particulate contamination is also present in sterile powders which are spray dried due to charring during the process.Review the particulate matter test procedure and release criteria.Review production and control records of any batches for which complaints of particulate matter have been received.o Production Records

Production records should be similar to those for other dosage forms.Critical steps, such as integrity testing of filters, should be signed and dated by a second responsible person.Review production records to ensure that directions for significant manufacturing steps are included and reflect a complete history of production.Ointments, Liquids, and Lotions

Major factors in the preparation of these drugs are the selection of raw materials, manufacturing practices, equipment, controls, and laboratory testing.Following the basic drug inspection fundamentals, fully evaluate the production procedures.In addition, evaluate specific information regarding:

o The selection and compatibility of ingredients.o Whether the drug is a homogeneous preparation free of extraneous matter.o The possibility of decomposition, separation, or crystallization of ingredients.o The adequacy of ultimate containers to hold and dispense contents.o Procedure for cleaning the containers before filling.o Maintenance of homogeneity during manufacturing and filling operations.The most common problem associated with the production of these dosage forms is microbiological contamination caused by faulty design and/or control of purified water systems.During inspections, evaluate the adequacy of the water system.Review and evaluate the micro/chemistry test results on the routine monitoring of the water system including validation of the water system.Review any microbiological tests done on the finished drug including in-process testing.Some of these drugs have preservatives added which protect them from microbial contamination.The preservatives are used primarily in multiple-dose containers to inhibit the growth of microorganisms introduced inadvertently during or after manufacturing.Evaluate the adequacy of preservative system.Preservative effectiveness testing for these products should be reviewed.For additional information, review the “Antimicrobial Preservatives-Effectiveness” section of the U.S.P..Equipment employed for manufacturing topical drugs is sometimes difficult to clean.This is especially true for those which contain insoluble active ingredients, such as the sulfa drugs.The firm's equipment cleaning procedures including cleaning validation data should be reviewed and evaluated.Packaging and Labeling [21 CFR Subpart G]

Packaging and labeling operations must be controlled so only those drugs which meet the specifications established in the master formula records are distributed.Review in detail the packaging and labeling operations to decide if the system will prevent drug and label mix-ups.Approximately 25% of all drug recalls originate in this area.Evaluate what controls or procedures the firm has to provide positive assurance that all labels are correct.Determine if packaging and labeling operations include:

o Adequate physical separation of labeling and packaging operations from manufacturing process.o Review of:

1.Label copy before delivery to the printer.2.Printer's copy.3.Whether firm's representative inspects the printer.4.Whether or not gang printing is prohibited.5.Whether labels are checked against the master label before released to stock.Determine who is responsible for label review prior to release of the labels to production.Also, whether the labels are identical to the labeling specified in the batch production records.o Separate storage of each label(including package inserts)to avoid mixups.o Inventory of label stocks.Determine if the printer's count is accepted or if labels are counted upon receipt.o Designation of one individual to be responsible for storage and issuance of all labels.o Receipt by the packaging and labeling department of a batch record, or other record, showing the quantity of labels needed for a batch.Determine if the batch record is retained by the packaging supervisor or accompanies the labels to the actual packaging and labeling line.o Adequate controls of the quantities of labeling issued, used, and returned.Determine if excess labels are accounted for and if excess labels bearing specific control codes, and obsolete or changed labels are destroyed.o Inspection of the facilities before labeling to ensure that all previously used labeling and drugs have been removed.o Assurance that batch identification is maintained during packaging.o Control procedures to follow if a significant unexplained discrepancy occurs between quantity of drug packaged and the quantity of labeling issued.o Segregated facilities for labeling one batch of the drug at a time.If this is not practiced, determine what steps are taken to prevent mix-ups.o Methods for checking similar type labels of different drugs or potencies to prevent mixing.o Quarantine of finished packaged products to permit adequate examination or testing of a representative sample to safeguard against errors.Also, to prevent distribution of any batch until all specified tests have been met.o An individual who makes the final decision that the drug should go to the warehouse, or the shipping department.o Utilization of any outside firms, such as contract packers, and what controls are exercised over such operations.Special attention should be devoted to firms using “rolls” of pressure sensitive labels.Investigators have found instances where:

o Paper chips cut from label backing to help running the labels through a coder interfered with the code printer causing digits in the lot number to be blocked out.o Some rolls contained spliced sections resulting in label changes in the roll.o Some labels shifted on the roll when the labels were printed resulting in omitting required information.The use of cut labels can cause a significant problem and should be evaluated in detail.Most firms are replacing their cut labels with roll labels.Review prescription drugs for which full disclosure information may be lacking.If such products are found, submit labels and other labeling as exhibits with the EIR See 21 CFR 201.56 for the recommended sequence in which full disclosure information should be presented.Review labels of OTC products for warnings required by 21 CFR 369.A control code must be used to identify the finished product with a lot, or control number that permits determination of the complete history of the manufacture and control of the batch.Determine:

o The complete key(breakdown)to the code.o Whether the batch number is the same as the control number on the finished package.If not, determine how the finished package control number relates, and how it is used to find the identity of the original batch.Beginning August 3, 1994 the following new requirements will become effective:

o Use of gang-printed labels will be prohibited unless they are adequately differentiated by size, shape or color.(211.122(f))o If cut labels are used one of the following special control procedures shall be used(211.122(g)):

(1)Dedication of packaging lines.(2)Use of electronic or electromechanical equipment to conduct a 100-percent examination of finished product.(3)Use of visual inspection to examine 100-percent of the finished product for hand applied labeling.The visual examination will be conducted by one person and independently verified by a second person.o Labeling reconciliation required by 211.125 is waived for cut or roll labeling if a 100-percent examination is performed according to 211.22(g)(2).Holding and Distribution [21 CFR subpart H]

Check the finished product storage and shipping areas for sanitary condition, stock rotation, and special storage conditions needed for specific drugs.Evaluate any drugs that have been rejected, or are on hold for other than routine reasons.Laboratory Controls [21 CFR Subpart I]

Laboratory controls should include adequate specifications and test procedures to assure that components, in-process and finished products conform to appropriate standards of identity, strength, quality, and purity.In order to permit proper evaluation of the firm's laboratory controls, determine:

o Whether the firm has established a master file of specifications for all raw materials used in drug manufacture.This master file should include sampling procedures, sample size, number of containers to be sampled, manner in which samples will be identified, tests to be performed, and retest dates for components subject to deterioration.o The firm's policies about protocols of assay.These reports are often furnished by raw material suppliers;however, the manufacturer is responsible for verifying the validity of the protocols by periodically performing their own complete testing and routinely conducting identity tests on all raw materials received.o Laboratory procedure for releasing raw materials, finished bulk drugs or packaged drugs from quarantine.Determine who is responsible for this decision.Raw material specifications should include approved suppliers.For NDA or ANDA drugs, the approved suppliers listed in their specifications should be the same as those approved in the NDA or ANDA.o If the laboratory is staffed and equipped to do all raw material, in-process, and finished product testing that is claimed.o Whether drug preparations are tested during processing.If so, determine what type of tests are made and whether a representative sample is obtained from various stages of processing.o Specifications and description of laboratory testing procedures for finished products.o Procedures for checking the identity and strength of all active ingredients including pyrogen and sterility testing, if applicable.o If the laboratory conducts pyrogen tests, safety tests, or bioassays;determine the number of laboratory animals and if they are adequately fed and housed.Determine what care is provided on weekends and holidays.o Sterility testing procedures.Entries should be permanently recorded and show all results, both positive and negative.Examine representative samples being tested and their records.When checking the sterility testing procedures, determine:

1.Physical conditions of testing room.The facility used to conduct sterility testing should be similar to those used for manufacturing products.2.Laboratory procedures for handling sterile sample.3.Use of ultra-violet lights.4.Number of units tested per batch.5.Procedure for identifying test media with specific batches.6.Test media's ability to support growth of organisms.7.Length of incubation period.8.Procedure for diluting products to offset the effects of bacteriostatic agents.o Pyrogen testing procedures

Determine if animals involved in positive pyrogen tests are withdrawn from use for the required period.If the L.A.L.Test is used, review the FDA “Guideline on Validation of the Limulus Amebocyte Lysate Test ***.”

o If any tests are made by outside laboratories, report the names of the laboratories and the tests they perform.Determine what precautions the firm takes to insure that the laboratories' work is bona fide.o Methods used to check the reliability, accuracy, and precision of laboratory test procedures and instrumentation.o How final acceptance or rejection of raw materials, intermediates, and finished products is determined.Review recent rejections and disposition of affected items.o The provisions for complete records of all data concerning laboratory tests performed, including dates and endorsements of individuals performing the tests, and traceability.o For components and finished product, the reserve sample program and procedures should be evaluated.Challenge the system and determine if the samples are maintained and can be retrieved.The storage container must maintain the integrity of the product.o Whether stability tests are performed on:

1.The drug product in the container and closure system in which marketed.2.Solutions prepared as directed in the labeling at the time of dispensing.Determine if expiration dates, based on appropriate stability studies, are placed on labels.o If penicillin and non-penicillin products are manufactured on the same premises, whether non-penicillin products are tested for penicillin contamination.Obtain copies of laboratory records, batch records, and any other documents that show errors or other deficiencies.Control Records [21 CFR Subpart J]

1.Master Production and Control Records [21 CFR 211.186]

The various master production and control records are important because all phases of production and control are governed by them.Master records, if erroneous, may adversely affect the product.These records must be prepared according to the drug CGMPR's outlined in 21 CFR 211.186.These records might not be in one location, but should be readily available for review.2.Batch Production and Control Records [21 CFR 211.188]

The batch production and control records must document each significant step in the manufacture, labeling, packaging, and control of specific batches of drugs.21 CFR 211.188 provides the basic information the batch records must provide.A complete production and control record may consist of several separate records which should be readily available to the investigator.Routinely check the batch record calculations against the master formula record.Give special attention to those products on which there have been complaints.Be alert for transcription errors from the master formula record to the batch record.Be alert for transcription or photocopying errors involving misinterpretation of symbols, abbreviations, and decimal points, etc.It is important that batch production records be specific in terms of equipment(v-blender vs.ribbon blender)and processing times(mixing time and speed).The equipment should have its own unique identification number.The manufacturing process for these products must be standardized, controlled, and validated.3.Distribution [21 CFR 211.196]

Complete distribution records should be maintained per 21 CFR 211.196.Be alert for suspicious shipments of products subject to abuse or which have been targeted for high priority investigation by the agency.These include steroids, counterfeits, diverted drugs(i.e.;physician samples, clinical packs, etc.).Determine and evaluate if the firm checks on the authenticity of orders received.What references are used, e.g.current editions of the AMA Directory, Hays Directory, etc.4.Complaint Files [21 CFR 211.198] CFR 211.198 requires that records of all written and oral complaints be maintained.Although FDA has no authority to require a drug firm, except for prescription drugs, to open its complaint files, attempt to review the firm's files.The complaint files should be readily available for review.Do a follow-up investigation on all applicable consumer complaints in the firm's district factory jacket.Review and evaluate the firm's procedures for handling complaints.Determine if all complaints are handled as complaints and not inappropriately excluded.Review the complaints and determine if they were fully investigated.Evaluate the firm's conclusions of the investigation, and determine if appropriate corrective action was taken.Determine if the product should be recalled, or warrant a comprehensive investigation by FDA

Returned Drug Products [21 CFR Subpart K]

Returned drugs often serve as an indication that products may have decomposed during storage, are being recalled or discontinued.Determine how returned drug items are handled.For example, are they quarantined, destroyed after credit, or returned to storage?

If an abnormally large amount of a specific drug item is on hand, determine why.Check if returned drug items are examined in the laboratory, and who makes the ultimate decision as to the use of the returned drugs.Note: Dumping salvage drugs in the trash is a potentially dangerous practice.Advise management to properly dispose of the drugs to preclude salvage.Drugs should be disposed of in accordance with E.P.A.regulations.

第四篇：FDA工廠檢查心得

FDA工廠檢查心得

FDA工廠檢查是美國FDA對醫(yī)療器械生產(chǎn)現(xiàn)場的調(diào)查。檢查對象是所有在美國境內(nèi)銷售醫(yī)療器械的制造商，包括美國內(nèi)和國外的制造商。由美國FDA派出檢查官員到醫(yī)療器械實際生產(chǎn)場所，執(zhí)行對工廠檢查，以確定被檢查的工廠是否符合QSR法規(guī)的要求。

在這里對FDA的工廠檢查談一點感受。

一、企業(yè)必須按FDA QSR的要求編制質量手冊和相關的程序文件及第三層次文件，要充分了解和理解法規(guī), 制定既符合要求又實用的體系文件。最最重要的是要切實地執(zhí)行這些文件，整理相關的歷史記錄和資料，工作必須嚴謹，不得浮于表面。做一項工作，首先要問自己為什么做？怎么做？誰來做？什么時間做（包括完成的時間）？

二、FDA檢查重點：評審文件；按QSIT方法——基于7個子系統(tǒng)

4個主要子系統(tǒng)（管理、設計、糾正預防、生產(chǎn)過程）；3個支持子系統(tǒng)（文件、物料、生產(chǎn)工具和設備控制）；FDA檢查工作時會以點帶面，抓住一點，可在一個問題上幾個來回，也可能檢查整個公司的質量管理體系。

三、在FDA官員來工廠檢查前公司內(nèi)部應進行多次的核查和確認；對任何公司來說，通過FDA驗廠最重要的條件是自己要嚴格執(zhí)行已經(jīng)確立的程序和作業(yè)流程，文件編制合理性及可操作性當然重要，檢查官員對有文件不執(zhí)行最反感。這樣的事情肯定會上Form483。

四、FDA對工藝過程特別是特殊過程中的一些關鍵步驟的操作條件、方法及設備進行的驗證(Validation)非常重視。但他們更注重過程控制，應該有的作業(yè)指導書一定要準備好；至少要有懂行的人陪同，萬一缺少文件，這時就*懂行的人員臨場發(fā)揮，只要能說出你們實際操作、控制的方法，能說服檢查官，一般也過得去，觀察項可能會有，但不太會作為不符合項，這一點他們到不死板。

五、FDA非常重視對生產(chǎn)記錄的檢查，對原材料的入庫、檢驗及發(fā)放、生產(chǎn)工藝過程的控制、成品的質量檢驗以及各項重點項目的驗證等均要求有完整的原始記錄及整套的批記錄，F(xiàn)DA官員在工廠檢查要任意取樣抽查批記錄，批記錄的真實性與完整性能具體體現(xiàn)工廠的GMP管理的水平。

六、FDA檢查官對不合格品和顧客投訴的控制及處理方法、過程及相關的記錄非常關注。這方面的準備務必充分，來不得半點馬虎，在這里利嘴巧舌不管用，必須得有真實的證據(jù)。

七、接待人員回答提問要有技巧，不清楚的事情切忌馬上回答，可以先查文件，幾個人商量定下來再回答。更不要“靈機一動”，以為自己應付得了，接下來的問題可能會讓你目瞪口呆。我曾看到過報道： FDA檢查官在工廠檢查時問：你們的文件看起來怎么很新啊，回答人員靈機一動：哦，我們的文件均用塑料文件袋保管的。檢查官也不再說什么了，可到了最后一天，檢查官突然說：我在你們工廠檢查了4天，可我從沒有看見你們遞給我的文件使用塑料文件袋保管的。到了這時你還有什么話說？

八、檢查結果：什么都沒有得到，這是最好的情況，但這恐怕不太可能。能接到無批評的483表（無不合格項，只有觀察項）已經(jīng)是相當不錯；到于做得不夠好的企業(yè)也許會接到有批評的483表，這就危險啦，它可能導致：警告信（Warning letter）、自動滯留（Automatic Detention）、QSR扣留、（QSR Hold）撤回（Recall）、直至永久不得進入美國市場。

所以，凡是接美國FDA通知要來工廠的企業(yè)務請注意，需要全公司員工的非常重視，全員動員，全力準備，將可能出現(xiàn)的不符合減到最少，才能避免接到警告信

第五篇：Peter Baker 和FDA 檢查

Peter Baker 和FDA 檢查---沒啥秘密的揭秘

2016-11-03

Peter Baker，一個外國人，不遠萬里，來到東方，早期以志愿者身份暫居廣西柳州，習得基本漢語，兼而了解炎黃子孫的秉性乃至思維習慣。后返美，在藥廠實驗室操練ＨＰＬＣ，據(jù)稱對Ａｇｉｌｅｎｔ之Ｃｈｅｍｓｔａｔｉｏｎ了如指掌，猜測其亦頓悟實驗室儀器數(shù)據(jù)處理的后門和漏洞。加入ＦＤＡ，后派駐印度，頻發(fā)警告信，把印度各大藥廠攪得雞飛狗跳，把數(shù)據(jù)完整性問題（現(xiàn)曰數(shù)據(jù)可靠性）提到了前所未有的高度?，F(xiàn)派駐中國北京辦事處，每年執(zhí)行約２０個檢查，不少國內(nèi)藥廠亦頻頻中招，Peter B所到之處不免人心惶惶。細數(shù)Peter B的厲害之處在于: 精通實驗室操作流程，也深諳數(shù)據(jù)“處理之道”，思維敏捷，檢查過程直接“粗暴”，不留情面；還有，通中文。

現(xiàn)就Peter Baker采用的檢查流程和可能的應對之策，作一簡要匯總。由于信息來源有限，難免掛一漏萬，權作參考吧。

檢查通知：

由于是北京辦公室派出的檢查員，一般通知期較短。多數(shù)是周末通知（周四、周五），下周一即開始檢查。若確定是Peter Baker，除了接機接站，不必費心思接風，安排游玩等等接待活動。該美國公務員廉潔奉公，不與公司餐聚，不接受禮物，不要求額外接待，所以好好準備業(yè)務檢查就好。

檢查準備：

若有一點時間，二三天也很寶貴呀！能做什么準備呢？電子數(shù)據(jù)真有問題，靠臨時處理，早干嘛去了？不建議再做處理了，時間短，處理急，反而弄巧成拙。熟悉下目前的流程、數(shù)據(jù)情況，特別是缺陷情況非常重要，因為要準備解釋呀，或者招供呀！

若時間還比較寬裕，目前FDA檢查周期還是規(guī)律，所以一般工廠能預感FDA要來，那么建議整個自查報告，數(shù)據(jù)完整性專項自查吧。面對Peter Baker沒有啥抹不開臉面的，勇敢自查，自爆短處，自揭傷疤。這樣做，有利有弊。好處，下面詳解，壞處是，風聞Peter Baker曾將自查結果直接寫進了483缺陷中。但個未經(jīng)授權許可，嚴禁轉載 人判斷還是自查報告寫得有瑕疵，不深刻，遮遮掩掩，整改措施也不到位，一句話，打鐵還需自己硬。

檢查過程：

Peter Baker的檢查，看時間吧，若有五天，那么P老兄會過一下實驗室以外的系統(tǒng)；若時間短，譬如四天，乃至三天，那么，棄其他不顧直奔實驗室而去，也是情理之中，確實也發(fā)生過的。舉個極端例子：某次，暫定四天實則執(zhí)行了三天的FDA檢查，Peter 檢查大員（僅Peter一位檢查員），在換名片，介紹彼此，亮出FDA檢查工作證后，便打斷工廠意欲開場PPT介紹，PPT可以打出來，帶回去看，咱們直接去QC開始工作吧。面面相覷哦，大家可以體會其中的感受吧！

談談見識過的Peter正常的檢查流程。某次FDA檢查（二位檢查員，Peter是當仁不讓的主角嘍）。開幕會結束，言明去現(xiàn)場，在辦公樓前，了解方向位置后，指向車庫，先從門衛(wèi)車庫開始，然后基本溜邊而行，鉆廠區(qū)圍墻，清潔工具存放間，廢棄物存放間，鍋爐房，外圍的廁所，任何不明、沒有標識的房間，統(tǒng)統(tǒng)打開，看啥？你懂的。。文件、物料嘛！

某次現(xiàn)場巡查，工廠聲明某建筑物幾間房是同場地兄弟公司的地盤，Peter B先和陪同管理層核實: 非工廠管轄，沒有工廠任何文件、物料，不在工廠體系，有否書面協(xié)議證據(jù)，如何標識，如何分割。似乎工廠的表述沒有漏洞。轉過臉Peter B 堅決要求開門，一番磨蹭拖延。。開門后，文件。。物料。。Peter Baker的小數(shù)碼相機“咔咔，咔咔”的同時，大家。。在數(shù)“草泥馬”啦！然后，PeterB嚴斥工廠不誠實，并申明要中斷檢查。Peter B 的檢查之道還有欲擒故縱哦。

基本的判斷，Peter Baker走其他現(xiàn)場，除了暗查不該有的物料，文件外，主要在于了解實驗室相關的樣品取樣、中控、結果報告的流程，更好地檢查實驗室數(shù)據(jù)問題。

下面說說Peter B 的興奮點---實驗室檢查。首先，Peter B一般會事先詢問，實驗室有沒有研發(fā)樣品檢測，小試樣品檢測等一切不必嚴格GMP條件下完成的操作。若承認有，下一個問題，如何區(qū)分? 如批號、文件名、操作記錄、樣品登記，等等。有時還會問，實驗室是否還有專有的實驗室“偏差”系統(tǒng)（針對有工廠專設實驗室異常處理程序，游離于偏差系統(tǒng)之外）。這時，一定要如實回答啊！這時不實說，按Peter B的描述，就上綱上線到欺騙美國政府，有點狠哦！也見過Peter B直接問，工廠自己有沒有發(fā)現(xiàn)不合規(guī)的數(shù)據(jù)處理操作，這時，自查報告可以適時登場，據(jù)個人的實際體驗，凡是Peter B查出的數(shù)據(jù)問題，若自查報告中有體現(xiàn)，則他都沒有寫進483，還算有職業(yè)道德。

走實驗室現(xiàn)場，一般常被Peter Baker深究的問題還有：

? ? ? ? ? 天平稱量的打印管理，紫外儀器的審計追蹤功能，紅外的使用前校準的記錄和操作，樣品的保存（特別是有小樣、研發(fā)樣品納入GMP體系時），GMP放行實驗室和研發(fā)實驗室的關系（可能會特地去研發(fā)實驗室，看看系統(tǒng)里的數(shù)據(jù)），儀器數(shù)量和實驗員數(shù)量（親見，4臺GC，3個操作分析員，被Peter B 寫上483，實驗室人員配備不充分），主管的職責，只要是在數(shù)據(jù)、電子數(shù)據(jù)的審核方面（親見，說實驗室經(jīng)理有上機復核的職責，則讓經(jīng)理演示上機，結果以經(jīng)理操作不熟練，上483的）

走實驗室現(xiàn)場也不會是個拖延時間的手段，因為往往沒有等到走完整個實驗室，Peter B 就已經(jīng)迫不及待地坐在了電腦前，選擇的對象，可能是最舊的那臺儀器。坐定，一般就是一天哦。先問操作者，再問管理員權限，確認管理員權限可以查看任何數(shù)據(jù)。然后，請以管理員權限登入，以保證可以審核所有數(shù)據(jù)信息，開始了最激動人心的檢查。

一般流程是：

問，操作系統(tǒng)工作站，數(shù)據(jù)存哪個盤符，哪個文件夾，如何命名，如何歸檔，如何備份。演示操作路徑。

看，設備使用記錄，選取個時間段，可能是近三年，每年選取幾周，逐條上機核查。

核，書面的原始檢驗記錄和數(shù)據(jù)。檢驗批記錄，大家一定費心準備過的啦。調(diào)，備份數(shù)據(jù)。備份數(shù)據(jù)的恢復和完整性，大家自己心里掂掂分量啦。查，系統(tǒng)中數(shù)據(jù)和上述文件、記錄、數(shù)據(jù)的一致性。

重點來了。。一般，初始，Peter檢查員會請管理員權限人員操作，但不一會兒，便會親自上陣，更具體的程序大致是：

?

? ? 任意挑一天開始，工作站有操作，但設備使用記錄沒有登記的操作和數(shù)據(jù)，會優(yōu)先關注先查。若設備使用記錄登記在案，則核對是否登記信息完整（如產(chǎn)品、批號、人員）。即設備使用記錄的信息和電腦中數(shù)據(jù)的一致性。任何一項電腦操作系統(tǒng)（審計追蹤或日志）中有操作痕跡，但沒有登入設備使用記錄的，都會細問。譬如，系統(tǒng)中有積分處理的記錄，但沒有相應的人員登入記錄，便順藤摸瓜，先把當事分析員或操作者叫來，細問數(shù)據(jù)處理細節(jié)。這里提示大家，設備使用記錄一定填寫完整呀！包括不入賬、不開報告的研發(fā)樣品、小試樣品（也該有記錄啊，因為進入了GMP體系了），重復測試、重新處理這事就不贅述該不該記錄了哈！

? 查序列信息和報告數(shù)據(jù)，序列安排了7針，報告記錄了6針? 那一針為什么不記錄報告？SOP規(guī)定了嗎？分析員你當時是啥考慮？序列中斷，有偏差調(diào)查嗎？調(diào)查結果拿來看看? ? 按照序列的提示，逐針核對審計追蹤，先看進針順序，后看每針間隔（基本一致，偶爾的序列中斷或停頓都可能問理由和解釋），再看存儲路徑，還看有么有異常出錯信息。

? 按照打印圖譜和審計追蹤提示的路徑信息（不止一次發(fā)現(xiàn)有工廠二者顯示的路徑不一致，可想而知想干嘛呢），去對應路徑查看圖譜文件，先點數(shù)（文件少了，又想干嘛呢!），后看文件名，再看文件夾中有沒有近似文件名（特別是批號相同的）。

? 在工作站調(diào)看圖譜，積分合理與否在放大縮小指指畫畫中比對進行，在上列路徑中發(fā)現(xiàn)的任何有興趣的圖譜也一并調(diào)出審看。

再看審計追蹤記錄，是否有相同批號的檢測，被發(fā)現(xiàn)沒有原因的復測可是低級錯誤哦。一般，序列運行前的任何單針進樣都值得審問，更別提相同批號乃至序列被發(fā)現(xiàn)重復進樣，還沒有調(diào)查和評估。?

? 有時，PeterBaker會要求把備份數(shù)據(jù)恢復，和工作站中的數(shù)據(jù)、紙質打印的數(shù)據(jù)一并核對，看備份的完整性(怎么完整的數(shù)據(jù)備份，大家看指南哈)

坦白說，我寫煩了。還有未盡信息，也到此為止吧！

總結三句：打鐵還靠自身硬，平日里合規(guī)，十個Peter 來也無妨。若有問題，數(shù)據(jù)完整性的專項自查、評估和整改報告很重要。誠信很重要，檢查時的合作坦誠，至少對Peter有點效果。