Transcription: Validation: Tried and True Success Lessons From the Front Lines Expert Panel Discussion – Live Free Webcast
The following transcription is taken from The Windshire Group’s webcast “Validation: Tried and True Success Lessons From the Front Lines Expert Panel Discussion”:
Terri Melvin: Welcome everyone to today’s Webinar, Validation: Tried and True Success Lessons From the Front Lines Expert Panel Discussion. I am Terri Melvin and I will be today’s moderator. We have with us today two expert panelists. They will first tell us a little bit about themselves and then address some talking points in their particular areas of expertise. Then we will dive into your questions live. I will now turn it over to our first panelist, Dr. James Blackwell.
Dr. James Blackwell: Thanks Terri, and thanks to everyone in our audience for joining us. I am James Blackwell, president and principal consultant for The Windshire Group. Our hope today is to make new industry friends and acquaintances and to serve as an information resource for organizations by providing answers to some of our most pressing questions around this very important and timely topic. Just a couple of house rules before we get started. Please do not ask anything that contains confidential information. Also, for today’s webinar, we are not offering any specific advice for anyone’s particular situation and we are not responsible in any way for any use of the advise or opinions we express. After self-introductions, we will share a few of our personal thoughts about validation and then we will get started with questions. I will then close our expert panel webinar with some concluding thoughts.
Prior to becoming a consultant, I held several senior technical positions within industry development and product companies. I have been an industry consultant for more than 12 years, assisting with CMC technical aspects for products in all major therapeutic classes at all stages of the product life cycle. More recently, due to the growth in this area, we have been doing significantly more work in the quality system area as well. Danielle has significant experience working in validation as well, and I would like to now introduce Danielle DeLucy for herself introduction.
Danielle DeLucy: Thank you, Dr. Blackwell. I am Danielle DeLucy. I have been in the industry now for about 18 years. I had most of my training early on in the microbiology lab and then other quality systems dealing specifically with sterility assurance where we do a lot of validation in aseptic processing areas. Then I had a lot of validation experience on the other side once protocols and reports are issued for review, looking at validation and deviations, et cetera. So hopefully today you will be able to learn a lot of different topics and techniques and different things that we have been through with our validation experience.
Terri: Thank you, Danielle. We are going straight to your slide and talking points. Here we go.
Danielle: Validation, in general, is certainly not one of the things that I have actually been involved in with regard to actually performing the validation. I have had a lot of experience on the other side as I said in my introduction about how we handle validation from a project perspective and from the quality side. When we talk about process validation, we are doing a lot of evaluation of data, collecting the data, looking at all of the different design stages throughout production, and really establishing that scientific evidence to make sure that our process is consistently delivering those quality products that are needed by the market. Three stages of validation are what we are talking about today, stage one, stage two and stage three. We talk about process design, process qualification, and then most important continued process verification.
Looking at our processes in the beginning and looking at all of that knowledge that we are gaining through developing and scaling of activities and then qualify that process. And that is where you get a lot of interaction between the validation engineers, project managers and the quality group to make sure that that process qualification is compliant. And then once all that is done, we want to make sure that your process is actually remaining in a state of control. So that is where we go to the continued process verification where you get your quality folks involved, taking a look, maybe doing some spot checks and making sure that the validation is still going as you intended it to. When you do have a properly installed, well-maintained and properly cleaned equipment, obviously you have more faith in those products. Your products are going to meet more of your intended specifications, knowing all of the different critical parameters, critical processing parameters, critical quality attributes of your equipment, of your process and of your instruments. All of that is key to making sure that your process is controlled systemically, making sure that the products that you are releasing are the same each time and we want to make sure of that. And that is why validation is so, so important. It is one of the most important things that a health authority will look at when they come into your area. Validation protocols and reports are always a very highly requested document. They pour over them. They look at a lot of the different deviations and change controls that are associated with them. So, they are kind of their own little microcosm of documents that usually they bring an expert with them to take a look at all of that information because it is such a big load of information for them to look at. And it does also show a little peek into how they handle their quality system. So, it is a nice way to show that your quality culture is up to par. So those are a few points to consider and kind of set the tone for what we are going to talk about. I will hand it over to Dr. Blackwell for some of his talking points as well.
James: Thanks. I do not have a slide but I would like to share some of my experiences for consideration for people who are thinking about validation. Validation provides documented evidence to support your claims around the robustness and the quality of your product. As such, it is an extremely serious exercise that gets very close scrutiny by the authorities and for good reason. And so, when you are thinking about your validation, you should really be very thoughtful about how you construct your validation protocols, your plans, and how you execute those protocols. When there are deviations from that, they need to be documented correctly and justified appropriately, and things that do not meet the acceptance criteria obviously will need to be investigated, but then constructing the appropriate rationale around that to support the validation and to make sure you can still claim that your process is validated.
It needs to be carefully handled. I have had a client situation where the FDA came in and looked at what had happened during validation and then asked them, “I do not think your processes is validated”, and that is not a good position to be in. And it is not just about the technical aspects. If you have a batch that has 20 significant deviations that occur during execution of that, you have to really question if you are in control and if you can stand up and say that your process is validated. There has also been a transition to the historical view of what process validation meant to the three phases of validation that Danielle presented and this has created some confusion within the industry. I think now most organizations have shifted it towards this three-phase view that really continues on after your process qualification and into continued process verification. And again, that is an area where companies I think struggle a bit to get up to speed because do not have the appropriate software solutions and so forth in place and monitoring to be able to handle the requirements of continued process verification. So, with that, let us get started Terri.
Terri: Okay. Excellent! We are now open for questions. Thank you panelists for those talking points. Simply type your questions into the question box. Our goal today is to answer any and all questions. However, if we do not get to your questions, we will answer your questions after the webinar concludes. Now for our first question: Danielle, I think this goes back to what you mentioned in your slide, What is the difference between validation and qualification?
Danielle: Validation and qualification – I have seen those words used and kind of thrown around in different arenas. When you are qualifying something, I guess you are saying that it has the quality attributes to work. So, when I think of qualification, I think of when I am qualifying my operators to do something, I am qualifying an operator to be gowned trained. When I am validating something, I am saying that “yes, I can”. This piece of equipment, this process, this instrument can repeatedly give me the same result over and over each time. So, they are a little bit different in my opinion. They do get mixed up quite a bit. I would say that the validation, again, is more for the process, making sure that it is repeatable and that qualification, I guess in my opinion, would be that it does meet your quality standards So maybe Dr. Blackwell can elaborate on that a little bit more.
James: Yes, I think there are two major areas where I have seen that distinction to be important. I think qualification on the equipment side tends to be something more focused on the systems and equipment and things to make sure that they are in a state that will support a validation and the process on an ongoing basis obviously. So, in a sense, validation is the documented evidence that I discussed before, whereas qualification is more around documenting your system supporting the validation are appropriate. The other area we are seeing a lot of confusion is the difference between the EU and the US and how the authorities in the companies in those two countries use the term for analytical validation and qualification. The Europeans will tend to just speak in terms of validation, but describe it as fit for purpose, which can vary during the lifecycle of the product from early stage to late stage, whereas Americans will tend to use the word qualification for a method that is not completely validated, but something that is being used for earlier stages of development and early clinical trials and so that can create some confusion, especially when Americans are talking to Europeans. So those are my thoughts on it.
Terri: Great, we have another question here. This is pretty specific. They’ve heard that three runs for validation is no longer sufficient for FDA or needs to be rationalized, is that true?
James: I will start on that one. You know that is interesting especially when the QbD and design space talk was really rolling through the industry. There was a lot of confusion and buzz out there around this issue and I recall someone talking that had so called experts saying, “Well, if you do all this QbD and design specs right, you should be able to validate your process with one run”. Well, I have not seen anyone do that. If anyone has seen that happen and be successful with that, I would love to hear about it, send us an email or give us a call. But I have asked around the industry about practices in this area and where FDA is on this. And I have not seen, I am sure companies would come in and rationalize and justify different things in their filings, but I have not sensed a big shift away from a three-run validation within the industry. And I even asked a client recently that had submitted a package to the FDA and they did three runs. I said, “Well, you know, did FDA challenge that in any way in terms of your rationale or justification for three runs?” And they said “No”. So again, I am very open to hearing other’s experience on that, but that has been my experience.
Danielle: I agree Terri. In my experience, I have not really heard of anything that kind of shies away from the three-run rule I guess you could say. Most of the time that is usually satisfactory for FDA or any of the other health authorities in the event that perhaps they may question your procedure or maybe your validating design, In that respect they may want to have an additional view. So that may be some of the reason as to why they would ask for some additional runs.
Terri: Thank you. I think that answered the question. I have another one here. How do you handle deviations during process validation?
Danielle: I could start with that one. I have been dealing with that in my career for a while. Most of the time, deviations are handled much the same way for validation protocols and reports. They are for regular GMP manufacturing. The only thing that really is different is if you are in an IQ or OQ type of a situation. You are not releasing product at that point. It might just be protocol generation error or perhaps maybe you just have to deviate from the protocol because the equipment deferred or something like that. Sometimes they are not as major or critical as the deviation would be in a full-fledged validated production.
In my experience, I am utilizing the same deviation system. If you are using a software system or the same procedure to log one to write up what happened, to do a pretty good root cause analysis and then correction would always be the kind of the skeletal elements of a deviation, but you may not have any product to take a look at. When you get into the process qualification or process validation at the very end and you collected well, that is when I would handle it much the same way as a normal deviation when the process is fully validated. So, there are a little bit of give and take there with the protocol deviations. Some people in industry or some firms that I have visited actually have a validation deviation system that is kind of separate from the regular GMP deviation system for deviations, but you do not necessarily have to do that. If you do commingle them in your deviation system, you just have to make sure that it is specific, if it is in validation mode, or if it is in regular production mode.
James: Can I add to that, Terri? Sort of to build off what Danielle just said, I think your validation masterplan should clearly define how those are going to be handled. I have seen different practices. Some people have thrown everything into their normal deviation system, others have defined what will be handled within the validation system itself, and then what level does it then get bumped over into the normal quality system for deviations. So, I personally like to see that well defined, and then things that are not serious be handled within the validation system itself. That makes it a little cleaner, but certainly anything with a potential for impact to patient safety, it is probably best to put those kinds of things into your normal deviation and campus system.
I would like to answer that question coming from another angle, and this goes to how you conduct your investigations. We have done other webinars on this topic, but it is especially important during validation. If you have something, it does not mean acceptance criteria to do appropriate investigations, do you understand what is known as true root-cause? The true root cause is the thing that is actionable that you can address to prevent recurrence of that events and that needs to be done with a scientific rationale and justification. That is needed so that you understand the issue and can address it appropriately within the validation.
Terri: Thank you. Dr. Blackwell, I have a question I think pointed to you because you did mention the validation masterplan. The question is: If we have adequate validation protocols, do we have to have a validation masterplan? Is it required?
James: I would say yes. I mean I am trying to think if I have seen it in the regulations. Danielle may be able speak to that. Frankly, I do not think I have seen a situation where people did not have a validation masterplan. I guess the question I would ask is why you would you not want to have a validation masterplan? I would think of it that way. So, it is not an onerous exercise. It gives an overview of what is going to be done. And then it describes those protocols. Again, I am talking about the overall validation of let us say the new process or product something like that. I am not saying it is required for every validation situation, but I think the question is probably directed to it, a new process, a new product. I would say within today’s GMP environment, in that situation, I would say yes.
Danielle: Yeah and I agreed Dr. Blackwell. I have been in quite a few regulatory inspections in my 18-year career and it is one question that usually gets asked. Usually the validation expert on the health authority team do ask for a VMP or validation masterplan. It is kind of a nice way for them to follow along with your schedule and make sure that you have proper validation activities schedule so that all of your different systems and equipment get validated or revalidated. And just as an internal document, I think it is helpful. It keeps you on task. A lot of validation engineers that I have worked with actually do like having one because it kind of keeps their workload scheduled for them and they can plan resourcing appropriately if they know all the different validation products and pieces of equipment and instruments, processes, et cetera, that they are responsible for that year.
Terri: Thank you. I think that did answer that question. I have one here that is interesting. What is the “process validation”? What is considered the process validation?
James: I guess I will start with that. To magnify the term, it is your manufacturing process. It should be described in a process flow diagram that shows your controls and acceptance criteria for that process, so that would be what you are validating in terms of your process. It is how you are making your product.
I do not know what else to say about that one, but it goes to the methods supporting it as well. Your process validation, your analytical methods are an important part of your controls, so that is an extension of what I consider process validation and then the equipment that is supporting it as well. So maybe that is where the confusion is coming from. The term can mean I think different levels of validation, but at the end of the day all those things need to be validated to appropriately support a process validation and so you could think of it holistically as all those activities. Danielle, what are your thoughts on that?
Danielle: I think you have covered it, not much more to say on process validation. I think you did a good job of defining it, so I do not have anything else to add. I think you did a good job.
Terri: Okay, great. I am glad you mentioned equipment Dr. Blackwell, because here is a follow-up question to that. What is the difference between calibration and the validation?
James: Calibration is an ongoing activity that shows that you have systems on an ongoing basis that are appropriate and are fit for purpose and can support the needs of your qualification activities. So, it is an ongoing exercise and it is necessary in the support of validation activities. When you are making measurements, you have confidence that your measurements are accurate and precise to the extent that they need to be to meet the needs of your process and your controls of that process.
Terri: Thank you. I have a general question here. In your experience, do you have any tips for best practices for validation?
James: Yes! I think I could probably name a number. I think one that really kind of strikes me. I have seen a lot of problems on this in organizations and that is people will conduct the validation exercise where folks will write all this stuff up and then they throw it over the fence and use that analogy to people to execute it. That is really a poor way to do it. The people who are going to be executing this need to be fully informed and trained on what they are going to be doing because there is often differences between the validation exercise and what you do on an ongoing basis. Sampling is a primary example where you are taking additional samples for process characterization as part of your protocol.
If there is poor communications and poor training, and not almost day to day monitoring of those activities, it can create lots of problems and then you did not have a bunch of investigation and stuff you have to write up and it can seriously impact the validation status. So that’s something I advise organizations to handle that appropriately to manage the validation exercises. I think developing the lifecycle view now is important that Danielle spoke to for the three different phases and thinking of it as a lifecycle exercise, and so it really starts near the beginning of the product lifecycle and so documentation practices for validation and that context start back then so making sure that you are capturing the appropriate data, storing that data somewhere so it is retrievable later and documenting things appropriately as you go along makes when you get to the process qualification stage inherently easier if you thought about that and the excuse that, “oh, we’re a small company, we’re busy, know blah blah blah.”
Well, if you do not have time now or you are going to have more time later and do you want to spend more time and money late, fixing the fact that you do not want to think about this now and do things that are relatively easy to do that will save you lots of time and effort later? So that is not an excuse. If you do not have the internal resources, then you should reach out to folks that can help you put in simple common systems that will help you with those types of activities that will support you during your process qualification activities as well. In terms of best practices, we have talked about some for process qualification. I think best practices for a continued process verification is to find a good a software solution and many of those you can implement earlier in the product lifecycle. They are not terribly expensive these days. There are some very good value solutions that you can use again to monitor your process and get your process data in a place that makes it readily available and usable for your technical people and quality people as well. So again, the excuses are getting lower and lower as time goes on, which is a very good thing and put together a good program around process monitoring because the benefits are not just pleasing the regulatory authorities, it makes good business sense if you understand your process can make process improvements, can perform investigations quickly and get to true root-cause when things are trending proactively to be able to react to that. It just makes common sense. I mean, people that make soap and potato chips have historically done a much better job over this than our industry does and that justifies common sense.
Danielle: I just like to add a little to the best practice that might be reviewed. I have seen places that are really falling down as they transfer from validation over into full-fledged production. There have been many times that I have been contacted on the quality side that there have been some critical deviations where the validation engineers did not really inform the shop floor, the manufacturers or employees as to what was going on, and there have been some really major and critical deviations that have occurred because of that. So, some of the best practices I have seen is really having that coworking between both shop floor, the operators, the validation engineers who are responsible for the validation protocols and reports. Again, as Dr. Blackwell mentioned we cannot stress enough how those validation engineers should be really knowledgeable of that process. A lot of times, the manufacturing operators will be their primary trainers and kind of the subject matter expert kind of following along with the process in the event that they have any questions or issues and then again, make sure that you are updating your documentation as well.
If this is more of a revalidation type of an activity or you are creating new procedures, make sure your standard operating procedures are there and you are developing them along the way or you are red lining the already existing ones along the way. That always seems to be a good practice. Something can fall through the tracks if you are not actively monitoring it and documenting it. So, those are two areas really where I have seen some major and even some critical deviations occur and really honing in on those two things, making those a little bit better, making sure that your validation study is complete and really kind of holistic, not only in the validation world but you are also making sure that the transfer goes smoothly into the manufacturing area.
Terri: Right. I think that answered the question. I have another one here: What would be considered a strong risk management plan for process validations?
James: I will start with that. It varies at different stages of the product lifecycle, earlier on you do not know much about your process to the extent that you need to do risk assessments. That will be a different risk assessment then you do later in your process and the tools that you use will be different. Earlier on, it is going to be based more on sort of scientific fundamentals, engineering fundamentals and rationale, the literature and that sort of thing. As you get more data, then you can refine those risk assessments and it is a very useful aspect to determine what controls you need so that you focus your efforts on things that are most risk to patient.
So again, taking the patient centric view and then conducting your exercise from that standpoint can really help create a more robust process and identify potential failure points. It is really understanding your process and then doing an appropriate assessment that can be very valuable. I have seen organizations use this to very good effect and also to justify certain things that they are doing especially, for example from a multiproduct type of processing activity and then looking at those types of risks and the controls that they have. So, again a great question, but in an area that can really support an organization and also to focus on validation where it needs to be as well, so that kind of relates to where the controls are. So where those controls are does the things that obviously get incorporated into your validation plan. So, it is good to start earlier as opposed to late. It should not be an exercise done to check a box, but really done as part of preparation for your validation activities and can be incorporated to support your overall approach to validation.
Danielle: I think James covered most of that, Terri.
Terri: Okay. Excellent. I have a pretty general question here and I think it speaks to your expertise on validation: What are common mistakes during validation execution and what should be done about them that you have seen in your experience?
James: Just to go back to something I said earlier, I think one of the mistakes is that communication issue and not having everyone on the same page and not assuming because we write this, new instruction for the example that the operator on the floor has got to know how to take it. There needs to be an effort put into it to make sure that it is reviewed with the people executing it, that they understand what it means and because many of the samples that caused problems are ones that are outside of their normal routine. With one client in particular, I reviewed the history of their validation execution and it was just every protocol they did had this issue and eventually they realized it and they took steps to address it, which, common sense things do pay off.
So what I advise organizations who do not have a lot of history is to think about that, learn from others and be a little more thoughtful and think through what has to happen and making sure that everyone understands it and is on the same page. It is really that simple, not easy to do, easier said than done, but on the other hand, it is not that difficult either with some planning. I think again, I kind of alluded to this earlier, being very careful how you construct your validation exercise and your acceptance criteria. You can really paint yourself into a corner if you tighten things too much, if you have more than you need that sort of thing and it can create a huge, huge problem for yourself, so you have to be very thoughtful about that. I think another big practice is really taking those deviation investigations and handling those appropriately and making sure when someone reviews them later that they understand the effort that you put into it, your rationale and your approach. Those were some of my thoughts on that. Anything to add Danielle?
Danielle: I would say I guess some things that I have seen where people have fallen down in execution. I think James had touched on, especially with the acceptance criteria. At one firm where I was actually the quality rep for the validation product. We were having a preapproval inspection for one of our new manufacturing facilities, and as James had said, you do not really want to paint yourself into a corner with acceptance criteria. Apparently, some of our validation engineers had done that and requested a deviation. After they got their results, they tried to petition quality to expand that acceptance criteria. Obviously, our quality group did not approve that. You really need to make sure that your acceptance criteria makes sense, that it is not too stringent, that you take into consideration all the different anomalies and crazy things that can really happen during a validation run. I mean we see it every day in regular production and sometimes validation engineers and the teams think that everything is going to run smoothly, but you really kind of have to anticipate any anomaly or kind of nonconformance or deviation that may happen and really try to set that acceptance criteria appropriately from an execution point of view, making sure that people are, again, we cannot stress enough training on that process, that they understand the process enough to set those acceptance criteria correctly is really key from what I have gained in my experience.
James: Terri, I’d like to amplify a little bit. Actually, we have proposed one with a major organization for their validation quality system, and one of the approaches that we recommended for that project is to have a master SOP that combines specifications, SOP’s, IQ, OQ into a master SOP, and then associate attachments. So, this really simplifies the traditional approach, and then when you are executing your revalidations or requalification or whatever you want to call it, it can really simplify that approach because it reduces multiple protocols, the time impact of listing components and things and then multiple acceptance criteria across multiple protocols and inconsistencies and things like that. So, part of this is really simplifying your overall validation system and making it easier for your organization to use.
Terri: Thank you. I have a question here about the difference between IQ, OQ and PQ. Can you talk more about that?
James: Well, I think those terms you can easily google the differences between those. Where I have seen some confusion come in is what PQ is used for. I think from an equipment standpoint, it is performance qualification and then I have seen other people use that for process qualification as well, but they are related in the sense that often to the performance qualification is conducted in conjunction with a process qualification because it is qualification. It is done as part of actual execution of your process and using the equipment and systems live, if you will, while you are performing the process. So obviously, it starts with installation qualification that really has to do with documenting what you have and making sure that the system is installed and has the supporting systems for it, that it needs to perform correctly and then operational qualifications and showing that the equipment can perform over the range of its operation that encompasses what your process needs, and then showing that those systems are performing as expected based on the installation qualification.
Terri: Thank you, great. Oh, we are going back to validation versus qualification. What is the difference between method validation and method qualification?
James: I touched on this earlier. I think from the sort of the US centric type of view, qualification of a method would be one that is not fully validated, which is allowed by the guidance. So, I typically see that it is used for method that shows it is fit for purpose, and again it depends on the purpose. I mean, there are some methods that need to be fully validated typically, for example, something like potency because that gets directly at patient’s safety and what you are claiming for the product even in early stages of the clinic. But there are other methods, for example, process related methods where you do not have to do the full method of validation, but you want to qualify to show that it is scientifically appropriate for that stage of process characterization.
Terri: Danielle, do you have anything to add to the method validation question?
Danielle: No. I think James is covering everything very well today.
Terri: It looks like we have answered all the questions. We are going to wrap it up here because we respect everyone’s time. Again, if you think of questions later, there is our contact information on the slide. Just shoot us a note or give us a call, and we will answer your questions. I will now turn it over to Dr. Blackwell for some closing remarks.
James: Thanks everyone for attending. I enjoyed it and we hope you came away today with some useful insights and greater understanding of validation. If anyone has any specific consulting needs on this or another topic, please reach out using our contact information on the slide. We would love to hear from you. We look forward to sharing our expertise with you in the future. Bye everyone.
Terri: Bye everyone. Thank you for attending.
For a free download of the entire data integrity webcast recording – click here