The lyophilized drugs in the pharmaceutical industry cannot afford to compromise on cleanliness. Lyophilizer contamination results in wasted effort and resources, or worse, creates health and safety issues that can compromise a company’s reputation and customer confidence.

 

制药行业中的冻干药物不能在清洁度上妥协。冻干机污染会导致精力和资源浪费，更糟的是，还会造成健康和安全问题，从而损害公司的声誉和客户信心 。

 

Detecting and quantifying trace residue concentrations of active pharmaceutical ingredients (API), excipients, and washes during lyophilizer cleaning validation procedures is one of the single largest costs associated with the manufacture of pharmaceutical freeze dried drugs. 

 

在冻干机清洁验证程序中检测和量化活性药物成分( API )、赋形剂和清洗剂的痕量残留浓度是与制药冻干药物生产相关的最大单项成本之一。

 

The lyophilizer cleaning validation is a very laborious process and combined with the idle time of the manufacturing equipment is a very expensive part of pharmaceutical manufacturing cost.

 

冻干机清洁验证是一个非常费力的过程，加上生产设备的闲置时间，是药品生产成本中非常昂贵的一部分。

 

The FDA clearly outlines its expectations within the lyophilizer cleaning validation under Inspections Guides. To satisfy the enumerated expectations, there can be many different approaches to cleaning verification and/or validation, based on the most efficient, cost-effective and practical approach.

 

FDA 在《检查指南》中明确概述了其对冻干机清洁验证的期望。为了满足所列举的期望，基于最有效、最具成本效益和最实用的方法，可以有许多不同的清洁验证和/或确认方法。

 

There are 3 kinds of surfaces in the Pharmaceutical Industry: 

 

制药行业中有 3 种类型的表面：

 

● 直接产品接触：直接接触制成品（或作为最终产品一部分的子产品），且极可能转移残留物和污染物。

 

Direct Product Contact: Directly contact manufactured product (or sub products that then are part of the final product) and for which there is a very high likelihood of transfer of residues and contaminants. 

 

● 间接产品接触：靠近敞口的产品，并且残留物和污染物有很小或中等可能性转移到产品中，通常通过载体（操作员、气流）。

 

Indirect Product Contact: It is in close proximity to open product and where there is a remote or moderate likelihood of transfer of residues and contaminants to product, usually by a vector (Operator or Airflow).

 

● 非产品接触：不直接接触制成品，也不靠近敞口的产品。不可能转移残留物和污染物。    

 

Non Product Contact: Does not directly contact manufactured product and is not in close proximity to open product. Is unfeasible the transfer of residues and contaminants.

 

1.风险分析（FMECA）

 

Risk Analysis (FMECA)

 

The lyophilizer has indirect product contact surfaces (see figure 1): Trays, where the vials are deposited for lyophilized (which allow their transfer and loading) and shelves (internal part of the equipment), which supports the trays:

 

冻干机具有间接产品接触表面（见图1）：托盘、用于存放待冻干的小瓶（允许其转移和装载）和板层（设备的内部部分），用于支撑托盘：

 

FIGURE 1  图1

 

2.一些定义

 

Some definitions

 

Failure mode: the ways, or modes, in which something might fail. Failures are any errors or defects, especially ones that affect the customer, and can be potential or actual.

 

故障模式：某事物可能出现故障的方式或模式。故障是指任何错误或缺陷，尤其是影响客户的错误或缺陷，可能是潜在的，也可能是实际的。

 

Potential failure mode: It is each possible failure mode without being necessary for the fault to actually occur. It usually answers questions such as: - In what way is it conceived, that the product or process could fail? - How could the component fail to meet specifications?

 

潜在故障模式：这是每种可能的故障模式，但故障不一定会发生。它通常回答以下问题：-产品或工艺可能以何种方式发生故障？-组件如何无法满足规范的要求？

 

Potential failure causes: All causes that are assignable to each failure mode.

 

潜在故障原因：可分配给每种故障模式的所有原因。

 

3.潜在故障模式分析

 

Potential Failure Modes Analysis

 

Potential Failure Mode (N°1) = API residues from a previously manufactured product (A), spilled on the trays and without effective cleaning, can be transferred indirectly to the next product vials to be lyophilized (B).    

 

潜在故障模式 (N°1) = 先前生产的产品 (A) 中的 API 残留物洒在托盘上且未经有效清理，可能会间接转移到下一个要冻干的产品小瓶中 (B)。

 

Potential Failure Mode (Nº2) = Microorganisms, endotoxins or detergent, residual products of a previous cleaning, can be indirectly transferred to the vials of the next product to be lyophilized (B).

 

潜在故障模式 (Nº2) = 微生物、内毒素或清洁剂、之前清洁的残留产品可以间接转移到下一个要冻干的产品的小瓶中 (B)。

 

By evaluating the Potential Failure Modes, we can find the following potential contaminants on the manufactured products and their origin:

 

通过评估潜在故障模式，我们可以发现生产的产品中存在以下潜在污染物及其来源：

CONTAMINANTS 污染物	ORIGIN 来源
Chemical Traces 化学痕迹	One or several vials spill their contents on the trays due to the lyophilizer loading process, therefore remnants of APl + Excipients are spilled on the trays of the equipment before beginning the cycle.  由于冻干机装载过程，一个或多个小瓶的内容物溢出到托盘上，因此在开始循环之前，APl+赋形剂的残留物溢出到设备的托盘上。
	One or several vials are broken during the lyophilization cycle and spill their contents on the trays.  在冻干过程中，一个或多个小瓶破裂，将内容物洒在托盘上。
Microorganisms 微生物	Equipment without sterilization or with ineffective sterilization, Wet trays and camera, with moisture traces. Non-Sterile air inlet into the chamber.  未消毒或消毒无效的设备，湿托盘和相机，有水分痕迹。进入腔室的非无菌空气入口。

 

Potential Fault Cause (N°1) = Airborne residues transfer from the previous product, due to vacuum pulled in the lyophilizer, air currents may dislodge residues on surfaces, those residues may become airborne and those airborne residues may deposit into the vials.

潜在故障原因 (N°1) = 空气中的残留物从先前的产品转移而来，由于冻干机中抽真空，气流可能会将表面的残留物冲走，这些残留物可能会飘散在空气中，而这些空气中的残留物可能会沉积在小瓶中。

 

Potential Fault Cause (N°2) = Airborne residues transfer due to repressurization of the lyophilizer chamber, air currents may dislodge residues on surfaces, those residues may become airborne and those airborne residues may deposit into the vials.

 

潜在故障原因 (N°2) = 由于冻干机腔体重新加压导致空气中的残留物转移，气流可能会将表面的残留物冲走，这些残留物可能会飘散在空气中，而这些空气中的残留物可能会沉积在小瓶中。

 

Potential Fault Cause (N°3) = Transfer due to lyophilizer shelves movement (vial sealing stage), residues on the bottom of the surfaces may dislodge, drop and fall into the vials.

 

潜在故障原因（N°3） = 由于冻干机搁架移动（小瓶密封阶段）而导致的转移，表面底部的残留物可能会脱落、掉落并落入小瓶中。

 

NOTE 1: We can establish here that, the relevance of the Failure Modes is not given as a product (B) that can pick up residues of a product (A) from the shelves, but as a residue that can be transferred indirectly to the following vial product (B)

 

注 1 ：我们可以在此确定，故障模式的相关性不在于产品 (B) 可以从货架上拾取产品 (A) 的残留物，而在于残留物可以间接转移到下一个小瓶产品 (B)

 

NOTE 2: As assumed for cleanrooms, it is understood that microorganisms are not "free floating", they require a means of transport to reach the product, this transport is the residues airborne particles.    

 

注 2 ：正如对洁净室的假设一样，微生物并不是“自由漂浮”的，它们需要一种运输方式才能到达产品，这种运输方式就是残留的空气传播颗粒。

 

4.潜在故障模式可能性评估-工程视角

 

Potential Failure Modes Likelihood Assessment - Engineering perspective

 

From the lyophilization process, operation and engineering of the equipment we could ponder as very low or almost nonexistent the likelihood of these failure modes, due to:

 

从冻干过程、设备运行和工程来看，我们可以认为这些故障模式发生的可能性非常低或几乎不存在，原因是：

 

4.1 The partial stoppering of the vials does not create an easy and / or direct pathway for residues streams and / or contaminants to enter the vials. See figure 2:

 

4.1 药瓶部分密闭，不会为残留物和/或污染物进入药瓶提供方便和/或直接的通道。参见图 2：

 

FIGURE 2  图2

 

4.2 During the high vacuum phase (figure 3), in the lyophilization process, any residues and contaminants that become airborne are likely to be pulled in the direction of the evacuation port (butterfly valve - connection between camera and condenser). At the same time the air in the vials is being pulled upward and out in the same direction as the air in the chamber, reducing the likelihood of contaminated airborne entering the vials.

 

4.2 在冻干过程中的高真空阶段（图3），任何残留物和污染物在空气中都可能被拉向抽真空口（蝶阀 - 相机和冷凝器之间的连接）。同时，小瓶中的空气被向上拉出，方向与室内空气相同，从而降低了受污染的空气进入小瓶的可能性。

 

4.3 During chamber repressurization, any loose residues on the chamber surfaces, would have already become airborne and expelled through the vacuum port, furthermore, as the vials become stoppered the likelihood of transfer into vials becomes close to zero.

 

4.3 在腔室复压期间，腔室表面的任何松散残留物都已经随空气传播并通过真空口排出，此外，随着小瓶被塞住，转移到小瓶中的可能性变得接近于零。

 

4.4 During vial sealing phase (figure 3), the shelves are moving and any loose residue and / or contaminant that may be dislodged in this process is likely to only fall straight down, which is not a direct pathway into the partial stoppered vials.

 

4.4 在药瓶密封阶段（图3），架子在移动，任何松散的残留物和/或污染物可能会在此过程中脱落，但很可能只会垂直落下，而不会直接落入部分密闭的小瓶中。    

 

FIGURE 3  图3

 

4.5 Physical glass particle contamination, due to broken or cracked vials also has a very low likelihood of becoming airborne. Those particles and broken vials are eliminated by hand during chamber unload and with water during washing cycles. We can additionally determine that the "visually clean" criterion will be sufficient.

 

4.5 由于小瓶破损或破裂而产生的物理玻璃颗粒污染也极不可能通过空气传播。这些颗粒和破损的小瓶在腔室卸载时用手清除、在清洗过程中用水清除。我们还可以确定“目视清洁”标准就足够了。

 

5.潜在故障模式可能性评估-工艺视角

 

Potential Failure Modes Likelihood Assessment - Process Perspective

 

5.1 In contrast to direct product contact surfaces equipments, for which "adherent or tenacious" residues are more difficult to clean, for lyophilizers "loosely adherent" residues represent the worst case, and ironically it is those molecules which can readily be removed by flowing water. Therefore, the likelihood of having residues of a previous product is almost zero.

 

5.1 与直接接触产品表面的设备相比，其“粘附或顽固”残留物更难清除，而对于冻干机来说，“松散粘附”的残留物代表了最差情况，具有讽刺意味的是，这些分子很容易被流水去除。因此，保留上一产品残留物的可能性几乎为零。

 

5.2 The lyophilized products are generally designed to be reconstituted by water alone for patients use. Therefore, from the QbD, it is known that these product residues can be removed with water alone during the cleaning process. Moreover, most modern lyophilizers with automatic CIP (Clean in Place) cycles operate with WFI water alone, without adding any type of detergent or cleaning solution.

 

5.2  冻干产品通常设计为仅用水复溶以供患者使用。因此，从QbD可知，这些产品残留物可以在清洁过程中仅用水去除。此外，大多数具有自动CIP（在线清洗）循环的现代冻干机仅使用WFI水进行操作，而无需添加任何类型的洗涤剂或清洁溶液。

 

6.GMP关键性评估

 

GMP Criticality Assessment

 

For this assessment we will establish:

 

对于本次评估，我们将建立：

 

● 损害的严重程度 (S) = 1 至 5（无患者损害 - 严重/灾难性患者损害）

 

Severity of Damage (S) = 1 to 5 (No patient damage - Serious / Catastrophic patient damage)

 

● 发生的可能性 (P) = 1 至 5（不太可能发生 - 经常发生）

 

Occurrence Likelihood (P) = 1 to 5 (Unlikely to occur - Frequently occurs)

 

● 关键程度 (CR) = S x P

 

Criticality Level (CR) = S x P

 

● CR1 = 高关键性 – 建立风险缓解行动和关键控制点是必要且重要的。     

 

CR1 = High Criticality - It is necessary and vital to establish risk mitigation actions and critical control points.

 

● CR2 = 中关键性 - 根据获得的RPN（风险优先级数）、行动和关键控制点需求进行评估。

 

CR2 = Medium Criticality - Evaluate according to the RPN (risk priority number) obtained, actions and critical control points needs.

 

● CR3 = 低关键性 – 无需采取任何行动，风险可接受且可控。

 

CR3 = Low Criticality - It is not necessary to establish actions, the risk is acceptable and controlled.

 

7.风险优先数（RPN）计算

 

Risk Priority Number (RPN) Calculation

 

For this assessment we will establish:

 

对于本次评估，我们将建立：

 

ŸDetection (D) = 1 a 5 (100% detectable - Non detectable)

 

检测 (D) = 1 至 5（100% 可检测 - 不可检测）

 

ŸRisk Priority Number (RPN) = S x P x D

 

风险优先级数（RPN）= S x P x D    

 

 

 

8.评估结论

 

Assessment Conclusionn

 

The GMP criticality for each Failure Mode is low (CR3), therefore the risks are controlled and would not require major mitigation and control efforts. Although the severity (for the patient) is very high, the likelihood that they actually occur is insignificant (we have already justified it in above pages from the engineering and pharmaceutical point of view).

 

每种故障模式的GMP关键性都是低(CR3)，因此风险是可控的，不需要采取重大的缓解和控制措施。尽管严重程度（对于患者而言）非常高，但实际发生的可能性却微不足道（我们已经在上文中从工程和制药的角度证明了这一点）。

 

Then, why perform cleaning validation in these type of equipment?

 

那么，为什么要对这类设备进行清洁验证呢？

 

In addition to the usual justification that ... "it is a regulatory requirement" (since cGMP standards and related guidelines deepen about the importance of cleaning and sterilizing freeze dryers before each load to ensure an aseptic process, free from microbiological and particle contamination), and given the above analysis we can provide that:

 

除了通常的理由...“这是一项监管要求”（因为cGMP标准和相关指南强调了每次装载前清洁和灭菌冻干机的重要性，以确保无菌过程，不受微生物和颗粒污染），并且根据上述分析，我们可以提供：

 

● 根据PRN计算和“检测”系统评估，我们获得了不同的优先级值，我们必须（根据我们特定的风险分析SOP）规划预防措施和/或工艺改进，以确保检测到这些潜在的故障模式和/或将其最小化。    

 

According to PRN calculation and "detection" systems assessment, we have obtained different priorities values for which we must (according to our particular risk analysis SOP) plan preventive actions and / or process improvements to ensure that these potential failure modes will be detected and / or minimized.

 

● 清洁验证是那些“预防性”措施之一，因为我们认为“当前控制”措施不够充分，因为成品取样和质量控制不具有代表性，没有使用经过验证的技术“寻找”其他产品的化学污染痕迹水平。

 

Cleaning Validation is one of those "preventive" actions since we have scored the "current control" actions as insufficient due to non-representative finished product sampling and quality controls that do not "look for" chemical contamination trace levels of other products with validated techniques.

 

● 由于冻干机的设计（被认为是无菌加工的延伸），操作员必须手动干预装载、卸载和收集破损的小瓶（增加了微生物污染的可能性），因此从微生物的角度来说，清洁验证是相关的。

 

Cleaning validation is relevant from the microbiological point of view, due to the lyophilizer design (considered an extension of aseptic processing) since operators must intervene by hand in the loading, unloading and collection of broken vials (increasing the probability of microbiological contamination).

 

● 我们不仅要重视冻干机搁板的灭菌，还要重视小瓶装载托盘的灭菌（通常，该过程在高压灭菌器中进行）。

 

We must give importance, not only to the lyophilizer shelves sterilization, but also to the vial loading trays sterilization (usually this process is carried out in an autoclave).

 

● 我们假设冻干机存在微生物污染的可能性，因此必须对其进行灭菌；我们还确定微生物不是自由漂浮的，需要通过运输方式（任何种类的颗粒）才能通过空气传播有效地污染产品；然后清洁设备以消除这些颗粒并“验证清洁”似乎是合理的逻辑。

 

We have assumed that there is a likelihood of lyophilizer microbiological contamination, and therefore we must sterilize it; We also have established that microorganisms are not free floating and require a means of transport (any kind of particles) to effectively contaminate a product through an airborne; Then cleaning equipment to eliminate these particles and "Validate Cleaning" seems like reasonable logic.

 

● 我们必须进行FMECA评估，特别是针对我们的冻干机技术，了解其运行方式和自身的工程设计，以便充分的评估风险。 

 

We must perform an FMECA assessment in particular, for our lyophilizers technology, knowing how they operate and its own engineering, in order to adequately score the risks.