扫一扫 添加小助手
服务热线
13818320332
扫一扫 关注我们
11.8 Process Analytical Technology
PAT is defined by the U.S. FDA to be “a system for designing, analyzing, and controlling manufacturing through timely measurements (i.e., during processing) of critical quality and performance attributes of raw and in-process materials and processes, with the goal of ensuring final product quality.” (22)
The U.S. FDA further notes that “the term ‘analytical’ in PAT is viewed broadly to include chemical, physical, microbiological, mathematical, and risk analysis conducted in an integrated manner.” Much has been published about PAT in general and about PAT in many processes; the reader should consult those references for general background on PAT. However, there are limited publications about PAT in cleaning processes and cleaning validation (23-26) as compared to PAT for other manufacturing operations. The emphasis for PAT here is for the use of a feedback loop from the analytical measurement to control a cleaning process or cleaning process step. It should be noted that consistent with PAT principles, the timely measurement could be in-line, on-line or at-line.
“Timely measurements” have long been used in cleaning processes to assist in the design of rinse cycle times in automated CIP systems, including those in the biotechnology industry. For example,a common practice in the design of the rinsing process has been to measure conductivity of thefinal rinse as a function of rinse time. Conductivity is a useful parameter for this determination,since cleaning in biotechnology manufacturing usually involves highly alkaline and/or acidic cleaningagents, which possess significant conductivity (in addition to the conductivity of the manufacturedproduct and/or its degradants). If evaluated over several cleaning process runs in the design phase, aminimum time to consistently complete the rinsing process can be effectively determined. A safetyfactor (additional time) may be included as part of this determination. While such a study in thedesign phase would be appropriate for a PAT application, unless it combines the timely measurementwith a feedback mechanism to control the cleaning process during commercial cleaning processes, itwould not be considered PAT. As described in this paragraph, the purpose of the timely measurementis not to control the rinsing process, but to assist in selecting a fixed rinse time.
11.8 过程分析技术
美国FDA将过程分析技术定义为一个系统，即作为生产过程的分析及控制，依据生产过程中的即时检测(也就是过程中检测)，关键质量参数的控制，原材料和中间产品质量的控制及生产过程，确保最终产品质量达到认可质量标准的程序。
美国FDA进一步指出过程分析技术中“技术”一词广义上被视为集合了化学、物理、微生物、数学及风险分析的综合分析方法。对于PAT有很多基础性资料及不同工艺下PAT的应用，读者可以通过阅读相关资料了解PAT的整体背景。然而，相对于生产过程，关于清洁程序和清洁验证的PAT资料却很少。本文强调的是，PAT是通过分析技术的回馈机制控制清洁程序或某一步骤的方法。需要说明的是与PAT技术的原则相一致，即时测量可以是在线，原位或就地测量。
“即时测量”在清洁程序中的应用已经很久，用于协助设计自动在线清洁系统的淋洗周期，包括在生物技术领域的应用。例如，一个常见的应用为设计淋洗程序时通过测量最终淋洗水的电导率考察淋洗时间。电导率在此处是一个有用的参数，因为生物技术生产后的清洗通常会使用强酸强碱性清洗剂，清洗中会产生高电导率（包括产品本身和/或其降解产物产生的电导率）。如果在设计阶段考察若干组清洁程序的运行，可以有效的确定清洁程序需要的最少时间。并且测定时可以考虑引入安全系数（额外的时间）。然而只有在正式清洁程序中加入即时测量的反馈机制以控制清洁程序，此时在设计阶段的研究才是过程分析技术的应用，否则不能被视为过程分析技术。本段的例子中，即时测量的目的不是为了控制清洁程序，而是用于确定固定的淋洗时间。
11.8.1 PAT for Cleaning Process Control
The more relevant use of PAT for cleaning processes is the use of a timely measurement to define the completion of a cleaning process step. In this case, the achievement of a certain analytical measurement is a controlling mechanism for completion of that step. In the situation referred to previously about measuring conductivity online, if it is possible to determine that the achievement of a certain conductivity correlates with the end of the rinsing process, conductivity could be employed in a PAT approach. That is, the rinse time is not fixed, but could be variable depending on the time needed to achieve that predetermined conductivity value. In addition, consistent with PAT principles, it would be expected that the achievement of that conductivity value would be within a defined time window. The U.S. FDA PAT guidance states “Within the PAT framework, a process end point is not a fixed time; rather it is the achievement of the desired material attributes. This, however, does not mean that process time is not considered. A range of acceptable process times (process window) is likely to be achieved during the manufacturing phase and should be evaluated, and considerations for addressing significant deviations from acceptable process times should be developed.”(22)
For example, achievement of a desired conductivity in a very short time could be due to insufficient cleaning solution in the cleaning process. Achievement of the desired conductivity in a very long time may be the result of a clogged spray device. In both cases, a final conductivity is recorded and a final rinse time is recorded. However, in the traditional approach time is the step-controlling parameter, and conductivity is the monitoring parameter. In a PAT approach, conductivity could be the step-controlling parameter, and time would be the monitoring parameter.
清洁程序中和PAT最相关的应用是利用即时测量以确定清洁程序的完成。此应用下，确定的某种分析方法是步骤完成的控制机制。之前提到的在线测定电导率，如果可以将电导率与淋洗程序的结束建立关联，电导率就可能应用于PAT方法。此时，淋洗时间不是固定的，而是会随到达预设的电导率时使用的时间而不断变化。另外，与PAT原则保持一致，到达预设电导率的时间应该有一个确定的范围。美国FDA的PAT指南指出“PAT框架下，工艺结束点没有固定的时间；而是通过获得预期的参数值确定。但这并不意味着可以不考虑工艺时间。生产中应有可接受的工艺时间范围（工艺窗口）并加以评估，并应制定对于严重偏离可接受的工艺时间时的相关规程。
例如，很短时间内就到达预定电导率可能是由于清洁程序缺乏足够的清洁剂引起的。很长时间内才到达预定电导率可能是由于喷雾装置堵塞造成的。以上两种情况下，最终电导率和最终淋洗时间都应被。在传统方法中，时间是作为步骤控制参数而电导率只是监控参数，但在PAT方法中，电导率也可作为步骤控制参数，而时间则作为监控参数。
Sometimes there is an inappropriate objection to the use of PAT in this way, because it seems to violate the cleaning validation principle of not cleaning until clean (or testing until it’s clean). However, one of the features of PAT is that traditional rules of what is done for validation may not apply. As noted in the U.S. FDA’s PAT guidance, “Systems that promote greater product and process understanding can provide a high assurance of quality on every batch and provide alternative, effective mechanisms to demonstrate validation (per 21 CFR 211.100(a), i.e., production and process controls are designed to ensure quality). In a PAT framework, validation can be demonstrated through continuous quality control whereby the process is continually monitored, evaluated, and adjusted using validated inprocess measurements, tests, controls, and process end points.” (22)
对于此种PAT的使用方式有时会有不恰当的反对意见，认为这样做似乎违反了清洁验证的原则，也就是清洁直到干净为止（另一种说法是测试直到干净为止）。然而，PAT的一个特性就是清洁验证的传统理念可能并不适用于PAT。美国FDA的PAT指南指出，“通过系统提升产品及对工艺的理解可以对每批产品提供高度的质量保证并提供了另一种有效的方式达到验证的目的（根据21 CFR 211.100(a), 也就是生产及工艺控制被设计用于保障质量）。在PAT框架下，验证可通过连续的质量控制，即使用验证过的手段、检测、控制及工艺结束点以对工艺进行持续的监控、评估及调整”。
While this example of conductivity as a timely measurement to control the rinse process has been used, there are at least theoretically other opportunities for timely measurement to assist in the cleaning process design. For example, timely TOC measurements during the washing step may be indicative of the minimum time needed to complete the washing step (before rinsing is initiated). By this, it is meant that as proteinaceous soils are removed from the equipment surfaces in the washing step, it would be expected that the TOC in the wash solution would increase and then level off at a time when no more soil is removed (that is, the wash step is complete).
本例说明了电导率可作为即时测量参数控制淋洗程序，理论上其他参数也可应用于清洁程序的设计中。比如，冲洗阶段即时TOC的测量可能对于确定完成冲洗步骤的最短时间有指导意义（开始淋洗前）。它可以显示蛋白质类污垢在冲洗阶段被从设备表面去除，因为冲洗水中TOC的值会先升高，当没有更多的污垢冲洗出来时TOC值则会降低（表明冲洗完成）。
11.8.2 PAT Measurement Tools for Biotechnology Cleaning Processes
生物技术清洁程序的PAT测试工具
Currently, the most common tools with potential PAT application in biotechnology cleaning processes are conductivity and TOC, because these can be measured online in the cleaning or rinse solution. Surfaces techniques, such as NIR for surfaces, may not be practical for timely control, because such techniques involve measuring for residues after the cleaning process is completed, not during the cleaning process.
目前，生物技术领域的清洁程序最常见的PAT应用是电导率及TOC，因为对清洗或淋洗溶液可以在线测量。表面分析技术，如近红外可能并不适用于即时控制，因为这类技术需要在清洗结束后检测残留而不是在清洗过程中检测。
Conductivity sensors are readily available for in-line measurements and have been widely used for in-line monitoring (but not necessarily for control). Online TOC does not involve an in-line sensor, but rather a “sipper tube” which diverts a stream from the process piping to the online instrument (U.S. FDA calls this “on-line in a diverted stream”). One concern about the use of TOC in this way is the delay between taking the sample and the output of the actual measurement. Another concern is that if the instrument is continually taking and measuring samples during a cleaning process, earlier samples with high TOC values may carry over to the following sample and cause a false high reading. Of course, if the process is performed until the desired TOC value is achieved, there is an assurance that the process is adequate, because that possible carryover situation reflects a worst case.
在线电导传感器很容易获得而且已经广泛用于在线检测（并不一定用于控制）。在线TOC并不具备在线传感器，而是通过管路将液体从工艺管路转移至在线仪器（美国FDA称之为“转移流在线”）。这种方式需要考虑取样和实际测试之间的延迟。另外还需考虑在清洗过程中如果仪器连续取样测试，早期高含量TOC的样品可能对之后的样品产生残留效应而造成检测结果偏高。当然，如果清洗一直持续进行，直至达到预期的TOC值，就可保证整个程序是适当的，因为可能的残留代表了最差条件。
11.8.3 Additional Considerations for PAT
PAT的其他考虑
It should be noted that in the conductivity example described in Section 11.8.1, all aspects of traditional cleaning validation are not avoided. If conductivity were the measure of a residual cleaning agent, and if only sampling rinse water were acceptable for determining residues of a cleaning agent, a PAT approach of measuring conductivity asa rinse step control parameter would also provide assurance that the cleaning agent was adequately removed for each and every cleaning process. However, it would not address issues of residues of the active and/or bioburden. Those residues would have to be measured in the traditional manner, unless a timely measurement of those residues could be utilized.
必须指出的是，在11.8.1节提到的电导率的例子，传统清洁验证的各个方面都不能避免。如果电导率用量测试清洁剂的残留，而且仅有淋洗水用于测定清洁剂的残留，通过PAT技术测试电导率作为淋洗步骤的控制参数也可提供每一清洁程序中清洁剂是否清洗完全的保证。然而，这不适用于活性物质和/或生物负载的去除。这些残留需使用传统方式测量，除非出现新可针对此类残留的即时测量方式。
It should be clarified that rapid and/or online methods by themselves do not necessarily constitute PAT. As discussed previously, online conductivity can be a routine monitoring tool in a cleaning process step without controlling a process step. Online TOC (other than during the design phase) is not the use of PAT,unless the achievement of a certain analytical measurement of TOC determines and/or controls the completion of a cleaning process step. The same is the case with rapid microbiological methods. Rapid methods may enable one to obtain lab data faster, but unless those measurements determine and/or control the end of a process step, they are just rapid monitoring tools, not PAT tools (although they have the potential to bePAT tools).
需要澄清的是，快速和/或在线检测技术本身并不是PAT。如前所述，在线电导率可以作为清洁程序的常规监测工具而并不控制程序步骤。在线TOC（除了在设计阶段）并不是PAT，除非TOC的测定可以控制清洁程序的完成。这同样适用于快速微生物检测法。快速的方法可以加速获得实验室数据，但只要测定不能控制工艺步骤，它们就仅仅是快速检测工具，而不能称之为PAT工具（虽然它们有成为PAT工具的可能）。
The examples given illustrate the use of PAT for process design and for process step completion. In an ideal world, PAT would be used for real-time release of cleaned equipment and would be used instead of cleaning validation. However, at this time the tools to utilize PAT to confirm that equipment surfaces are appropriately clean (measuring removal of active, cleaning agent, bioburden and endotoxin in the case of biotechnology manufacturing) have not been adequately developed to enable real-time release for cleaning biotechnology equipment.
以上的例子描述了在工艺设计及工艺步骤完成时应用PAT。理想情况下，PAT可用于清洁设备实时放行从而取代清洁验证。但现实状况是目前使用PAT以确认设备表面的清洁（在生物技术生产中测量活性物质，清洁剂，生物负载及内毒素）还不足以支持生物技术设备清洁的实时放行。
11.9 Product Changeover
产品转换
Much biotechnology manufacturing involves campaigning. In a campaign, the same product is made again and again. However, typically between each batch in a campaign, validated cleaning is performed.At the end of a campaign, some additional steps may be taken to prepare the equipment for the subsequent campaign of a different product. This extra precaution typically involves performing an additional cycle of the same validated cleaning process used for cleaning between batches in a campaign. Because of concern about possible migration of residues (particularly product active) into gasket materials, or more accurately, into the interstices between gaskets and stainless steel surfaces, changeover of soft parts such as gaskets may also be done after the initial cleaning. During changing of soft parts, a more comprehensive visual examination of the equipment surfaces is made. Following reassembly of the equipment, the validated cleaning process is repeated. Routine monitoring of both the initial cleaning and the final cleaning is performed as is normally done. Some companies also might perform a specific analytical test (such as an ELISA procedure) as an extra check for the previous active protein in the final rinse water of the second cleaning. It should be recognized, however, that the likelihood of any native protein surviving one cleaning process, much less two cleaning processes, is very remote.
很多生物技术生产涉及连续生产。连续生产时，相同的产品不断被生产。然而，通常在连续生产时批次间会进行已验证过的清洁。连续生产结束后，可能会有额外的措施来为接下来不同产品的生产做准备。额外的措施通常包括将批次间已验证过的清洁方法重复运行一次。由于残留（特别是活性产物）可能会迁移至垫片处，确切的说是垫片和不锈钢表面之间的缝隙中，首次清洗后可能需要更换软性部件，如垫片。更换软性部件时，应对设备表面进行更全面的目视检查。设备重新组装后，应重复已验证的清洁规程。对首次及末次清洗的常规监测应像往常一样进行。一些公司可能还会进行特殊的测试（如ELISA）作为对二次清洗时最终淋洗水中活性蛋白的额外检查。需要认识到的是，在一次清洗后残留下来的天然蛋白在二次清洗后仍然残留的可能性是微乎其微的。
An alternative is not to change out those soft parts based on data showing no migration of residues into interstices between gaskets and stainless steel surfaces or analysis based on potential carryover. Such data can be based on studies on commercial equipment, on scale-up equipment, and/or in a laboratory simulation. In such cases, one validated cleaning cycle is used both between batches of one campaign and for a campaign changeover.
另一种办法是不更换软性部件，但需要数据支持残留不会进入垫片和不锈钢表面之间的缝隙或基于对潜在残留的分析评估。数据可以基于对生产设备的研究，等比例大规模设备和/或通过实验室模拟的方式。此种情况下，连续生产中批次之间的清洗和不同品种生产间的清洗采用同一种清洗方法。
Certain equipment is generally dedicated to a given product. This includes chromatography resins and ultrafilters. Cleaning may be done on these items within a campaign; however, at the end of the campaign the resins and ultrafilters are cleaned, but typically are not used for campaigns involving different products.
某些设备通常专门用于某种产品，如色谱树脂和超滤。清洗可能在连续生产后进行；然而，连续生产结束后树脂和超滤被清洗后，通常不会用于不同产品的连续生产。
文章来源：允咨生物GMP学苑
本网站刊载的所有内容，包括文字、图片、音频、视频、软件等，如非标注为“原创”，则相关版权归原作者所有，如原作者不愿意在本网站刊登相关内容，请及时通知本站，我们将第一时间予以删除。
分享精彩资讯