重庆莱美药业因数据完整性获FDA警告和进口禁令

今年4月19日， 美国食品和药物管理局发布了66-40项下针对重庆莱美药业股份有限公司两个厂区的进口禁令（因GMP检查不符合而不经过实物检查直接在海关扣留）。也就是说自该进口警报发出之日起至整改确认期间，来自上述厂区的制剂或原料药将不被美国接受。到目前为止，事发已两个月，莱美药业一直没对此事进行公开披露，该事件对公司业绩造成的影响也不得而知。

关于FDA对重庆莱美药业的警告信正文部分内容如下：

Dear Mr. An Lin:
+ {5 w* j* I+ d9 S' n; @" ~% Z

The U.S. Food and Drug Administration (FDA) inspected your Chongqing Lummy Pharmaceutical Co., Ltd. facilities, Chayuan at No. 8 Rose Road, Nan’an District Chongqing, and Changshou at No. 2 the 4th Branch Road Hua’nan Road, Chongqing, from March 14–16, 2016.

美国FDA于2016年3月14-16日检查了你们位于重庆的莱美药业生产工厂。

This warning letter summarizes significant deviations from current good manufacturing practice (CGMP) for active pharmaceutical ingredients (API).

本警告信总结了原料药生产严重违反CGMP的行为。

Because your methods, facilities, or controls for manufacturing, processing, packing, or holding do not conform to CGMP, your API are adulterated within the meaning of section 501(a)(2)(B) of the Federal Food, Drug, and Cosmetic Act (FD&C Act), 21 U.S.C. 351(a)(2)(B).

由于你们的原料药生产、加工、包装或保存的方法、场所或控制不符合CGMP要求，你们的原料药根据FDCA被认为是掺假药品。

We reviewed your firm’s March 31, 2016, response in detail and acknowledge receipt of your subsequent responses.

我们详细审核了你们公司2016年3月31日及之后的回复。

Our investigator observed specific deviations including, but not limited to, the following.

我们的调查人员发现的具体问题包括但不仅限于以下：

1.   Failure to prevent unauthorized access or changes to data and failure to provide adequate controls to prevent manipulation and omission of data.

未能防止未经授权的登录或更改数据，未能提供足够的控制来防止数据篡改和删节。

During the inspection, FDA’s investigator discovered a lack of basic laboratory controls to prevent changes to and deletions from your firm’s electronically-stored data. Your firm relied on incomplete and falsified records to evaluate the quality of your drugs and to determine whether your drugs conformed with established specifications and standards.

在检查期间，FDA调查人员发现化验室缺乏基本控制来防止从你们公司的电子存贮数据中更改和删除数据。你们公司依赖不完整的和假造的记录来评估你们药品的质量，决定你们的药品是否符合既定质量标准。

Our investigator found that your firm failed to prevent data manipulation on multiple computerized analytical systems. Your firm re-tested samples without justification and deleted raw analytical data from computerized systems. You are responsible for determining the causes of these deviations, for preventing their recurrence, and for preventing other deviations from CGMP.

我们的调查人员发现你们公司未能防止多个计算机化分析系统上的数据篡改。你们公司重新测试了样品，但没有论证，从计算机化系统中删除了原始分析数据。你们有责任确定这些问题的原因，以防止其再次发生，防止其它的CGMP违规行为。

a.    Our investigator’s review of the audit trail for the residual solvent stability testing indicated that an analyst manipulated your computerized gas chromatography (GC) system to falsify residual solvent stability results for multiple batches of (b)(4) API distributed to the U.S.

我们的调查人员审核了残留溶剂稳定性测试的审计追踪，审核发现一名化验员篡改了计算机化气相色谱（GC）系统以伪造多个批次销售至美国的某原料药的残留溶剂稳定性结果。

For example, on March 4, 2016, your analyst set the GC personal computer (PC) clock back to make it appear as if testing had been done seven months earlier – on August 3, 2015. The analyst then performed five different injections to produce falsified results for long term stability 25C/65% RH 12 month time-point residual solvent testing for finished API lot (b)(4). The analyst deleted the first four backdated results and reported only the results of the fifth and final injection as passing in the quality control data package. Your quality unit relied on this incomplete data package to evaluate the quality of this lot of API and determine whether it was within specification. Our investigator observed that long-term stability results for at least five other lots of (b)(4) API were falsified using this technique of setting back the clock on the GC personal computer and then performing multiple injections until favorable results were obtained. Your firm failed to prevent analysts’ access to manipulate and delete laboratory data. In addition, your laboratory equipment lacked software controls to assure data integrity.

例如，在2016年3月4日，你们化验室倒置了GC电脑时钟，使得检验时间看起来像是7个月之前的---2015年8月3日。化验员然后进了5针样品，生成假的结果作为某原料药批号在12个月时间点长期稳定性25度/65%RH的检验结果。化验室删除了前面4针倒置日期的结果，只报告了第5针合格的结果，并将该结果放在质量控制数据包里。你们的质量部门依赖于这样不完整的数据包来评估了此批原料药的质量，决定了其是否符合质量标准。我们调查人员发现至少还有其它5批原料药长期稳定性试验结果是使用同样的倒置GC电脑时钟日期，然后重复进样直到获得想要的结果的手段来伪造的。你们公司未能禁止化验室篡改和删除化验室数据的权限。另外，你们化验室设备缺乏软件控制来保证数据完整性。

b.    Our investigator’s review of the audit trails for the high performance liquid chromatography (HPLC) system indicated that, just prior to the completion of certain stability analyses for (b)(4) API, analysts routinely aborted the ongoing tests to prevent your HPLC system from recording some assay and impurities test data.

我们的调查人员审核了HPLC系统的审计追踪，审核发现在某原料药稳定性分析完成之前，化验员常常中止正在进行的测试，防止HPLC系统记录一些含量和杂质检测数据。

Your HPLC system, controlled by Chemstation software, was configured to automatically delete the results if a test was aborted prior to completion. Our investigator’s review of the system’s limited audit trails indicated that when an analyst aborted assay and impurities tests, the partial results from the aborted tests were automatically deleted from your computerized HPLC system’s records.

你们的HPLC系统，由CHEMSTATION软件控制，其参数设置允许在测试完成之前中止运行并自动删除结果。我们的调查人员审核了系统内有限的审计追踪，发现当化验室中止含量和杂质检测时，从被中止的检测中获得的部分结果会自动从你们的计算机化HPLC系统记录中删除。

For example, our investigator reviewed a data file audit trail that showed that during impurities analysis of an 18-month stability sample of (b)(4) crude batch (b)(4), your analyst aborted the injection before the test was complete, set the HPLC PC clock back, and then repeated the injection. Your analyst only reported the results of the second injection in the quality control data package. This test, for which your computerized system did not retain original data about the quality of your (b)(4) crude, had been performed as part of a stability study your firm executed in response to FDA’s previous inspection in July, 2013. Our investigator observed the same technique for data manipulation and deletion in multiple other impurities analyses for (b)(4).

例如，我们的调查人员审核了一份数据文件审计追踪，发现在某产品批18个月的稳定性样品杂质分析中，你们化验员在检测完成之前中止了进样，倒置了HPLC电脑时间，然后重复进样。你们化验员在质量控制数据包中只报告了第2次进样的结果。这个检测是针对你们给上次FDA在2013年7月检查回复中承诺要实施的稳定性试验的一部分，而你们的计算机化系统并没有保存某粗品的质量原始数据。我们的调查人员发现在某产品其它杂质分析中也使用了相同的数据篡改和删除方法。

When our investigator asked your staff about these instances of falsification and manipulation, your quality control manager stated that your firm “forgot” to perform stability testing and therefore created falsified results for each missed time point by manipulating the controlling PC clock.

当我们的调查人员询问你们的员工关于伪造和篡改数据的事实时，你们质量控制经理声称你们公司“忘记了”实施稳定性测试，因此通过篡改电脑时钟的方式给每个时间点做了假的结果。

In your response to the FDA-483, you stated that you would upgrade your GC and HPLC software, and revise standard operating procedures (SOPs) for handling analytical data and train your staff on these revised SOPs. You also indicated that you would retrospectively review analytical data, and if data manipulation was identified, conduct a risk assessment to determine whether “re-testing actions are required.”

在你们给FDA483的回复中，你们声称你们会升级你们的GC和HPLC软件，修订处理分析数据的标准操作程序（SOP），培训你们的员工学习新SOP。你们还说你们会回顾分析数据，如果发现数据篡改，将会进行风险评估来决定是否需要“复测”。

Your response is inadequate because you have not demonstrated how the software upgrades, SOP revisions, and training will correct the broad data manipulation and deletion problems observed at your facility and to prevent their recurrence. Your quality unit must review all pertinent analytical data when making decisions about the quality of your drugs and when evaluating their conformance to established specifications. When your electronic systems permit the falsification and manipulation of data to obscure or delete test results, the quality unit is presented with incomplete and inaccurate information about the quality of your drugs. Your response does not demonstrate how your proposed software upgrades, revised procedures, or training will prevent the deletion of data or how your quality unit ensures that the records relied upon for batch release and other quality review decisions are complete and accurate

你们的回复是不充分的，因为你们没有说明升级软件、修订SOP以及培训如何纠正在你们工厂发现的如此广泛的数据篡改和删除，并防止其再次发生。在做出关于你们药品质量的决策时，在评估你们的药品是符合既定质量标准时，你们的质量部门必须审核所有相关分析数据，当你们的电子系统允许对数据进行篡改和删除时，掩盖或删除检验 结果时，质量部门给出的是你们药品不完整不准确的质量信息。你们的回复并没有说明你们如何升级软件、如何修订程序以及如何培训你们的员工来防止数据删除，你们的质量部门如何保证赖以放行批次的记录和做出其它质量审核决策的数据是完整准确的。

) e  G9 z7 V$ M, n) \2 }

2.    Failure to document manufacturing operations at the time they are performed.

未能在生产进行时同步记录生产操作

During the inspection, our investigator reviewed 20 executed batch manufacturing records and found that most of them contained similar or identical entries that could not be adequately explained. For example, our investigator examined batch records for (b)(4) different batches of (b)(4) API manufactured between January and February 2015. All (b)(4)batch records indicated that certain process steps or measurements had transpired at exactly the same time for each different batch. When our investigator asked your production supervisor to explain why the time stamps were identical on these records, the production supervisor stated that the full manufacturing process takes (b)(4) to complete, and that all batch records are kept in the production area until (b)(4) lots are completed. The production supervisor stated that the operators most likely did not record the actions at the time they were performed but rather completed batch records in groups.

在检查期间，我们的调查人员审核了20批已填写的生产记录，发现其中大部分录入数据是相似的或相同的，无法给出充分的解释。例如，我们调查人员检查了某原料药在2015年1-2月期间若干不同批次生产批记录。所有不同批次生产记录显示具体的工艺步骤或措施都是在完全相同的时间执行的。当我们的调查人员询问你们生产主管要求解释为什么这些记录的时间戳是相同的时候，生产主管声称全部生产工艺要花多少时间，所有批记录都保存在生产区域，直到所有批次完成。生产主管声称操作人员很可能并没有在执行活动当时记录，而是一起填的。

In your response to the FDA-483, you indicated that you would not release any new (b)(4) API to the U.S. market until “FDA deems our facility acceptable.” You also indicated that you had reviewed all manufacturing data and found “some batches have the same falsification” and committed to revising your batch manufacturing record template and SOPs, and retraining your staff.

在你们给FDA483表的回复中，你们说你们不会放行任何新的某原料药至美国市场直到“FDA认为工厂是可以接受的”。你们还说你们已经审核了所有生产数据，发现“有些批准具有相同的伪造情况”并承诺要修订你们的批生产模板和SOP，并培训你们的员工。

Your response was inadequate. Neither revised templates and procedures nor retraining your staff alone can prevent operators from continuing to falsify batch manufacturing records.

你们的回复是不充分的。修订后的模板和程序以及你们员工的培训均不能防止操作人员继续伪造批生产记录。

1 Y/ M5 q& q% U( Y1 E

FDA查的很严，有理有据~

受益了，上面很多东西对我们企业是共性，期待改变！

还是FDA牛！