Mack Powers | July 28, 2021
Interview Transcript:
Mack Powers (00:07):
We’re here with Mitchell Wheeler today. Mitchell, it’s great to be talking to you. Mitchell is a senior quality consultant with PharmEng Technology, and today we’re going to be talking about laboratory data integrity compliance, and some of the current challenges. And Mitchell, as we’ve been talking about, there are quite a few challenges today when it comes to companies meeting GxP requirements for data integrity.
Mitchell Wheeler (00:32):
Yes, there are. And thank you for having me, and I’m thankful to be here. Yes, it’s a very meaty subject and one that not just U.S. regulators but regulators around the world are starting to look at for companies.
Mack Powers (00:48):
Well, that’s great. So, it’s a broad topic and we want to focus on what those challenges are. And you’ve identified to me in previous conversations, what some of those real challenges are. But before we get into that, help us to get a picture of what we’re talking about. What is data? And where do we find data when it comes to a laboratory?
Mitchell Wheeler (01:08):
Well, data takes many forms. When they perform tests and data’s generated, then the old school way of considering data is your printout is your raw data.
Mack Powers (01:22):
Okay.
Mitchell Wheeler (01:23):
I grew up in the industry, over 30 years, and that’s what I understood before. But many of those instruments generate an electronic copy, and that is the true copy. And many companies don’t treat that as the original document. That’s the crux of data integrity. And then the quest to make sure we control, manage and get our arms around how we handle the versions of this stuff.
Mack Powers (01:52):
Makes sense. So the true copy now is considered the electronic version, versus 30 years ago when you started. It was certainly paper at the time. Can data be both a hard copy and electronic, and still be considered data?
Mitchell Wheeler (02:05):
Well, the regulations allow you to establish procedures and follow them. That’s why regulations are written. And if you’re doing that, then you’re meeting the letter of the law.
Mitchell Wheeler (02:18):
However, FDA stayed the part 11 regulations back in the early 2000s. And it set the stage for some confusion around, “Is this being enforced, or is it not being enforced?” Now that their guidance document is out, they’re starting to write 43’s on gross violations of the standard. And that is maybe not looking at audit trails, not looking at things. We can talk about some of those things later on.
Mack Powers (02:49):
Sure.
Mitchell Wheeler (02:52):
But if you’re not assuring that that data is true, accurate, concise, complete, and accurate in its entirety, then we are not taking on the true spirit of data integrity.
Mack Powers (03:10):
Yeah, that makes sense. And what you’ve said to me also is virtually everything now, every process within a laboratory, is producing some type of data. So the scope is very broad and it’s something that every company is really specifically having to address to make sure that they stay compliant.
Mitchell Wheeler (03:30):
The challenge is, too, is that many companies are making the effort to assure that equipment is calibrated and qualified, and vendors are in and out of companies a lot.
Mitchell Wheeler (03:48):
However, those companies … It’s been my experience with the people that I’ve worked with, that the vendors will not attest to the part 11 compliance for that company because they state that they’re not a part of their quality system.
Mack Powers (04:09):
Okay.
Mitchell Wheeler (04:10):
It’s the company’s responsibility to assure that it has a program to insure data integrity. They’re not responsible for their data integrity.
Mack Powers (04:19):
Yeah, that makes sense. So we’ve talked about the data integrity as a whole, but one of the things we mentioned earlier, we wanted to talk about some very specific challenges that companies are facing currently. And especially since COVID-19, where you mentioned things have changed a little bit. And when you go into a company now and are asked to do a data integrity evaluation, it’s a little different now. So let’s start with the challenge that we were talking about today around that topic. What are you seeing today as a challenge?
Mitchell Wheeler (04:49):
So that type of challenge forces a remote review. And through that remote review, you do accomplish a few things.
Mitchell Wheeler (04:58):
One, you assure that the quality system established is producing the documentation that it’s supposed to. And it’s a static review of what is, from a records standpoint. Then there has to be, because of things, there are people on site that have access to the equipment and the SMEs and they’re scheduled, because not everybody is on site all the time. And they’re still phasing back in to having a scheduled time and get to the SMEs to explain the equipment, “Show me this, show me that.”
Mitchell Wheeler (05:39):
And then they’re able to further demonstrate, “Are you really following this procedure? Well, during the record review, we actually determined that you who have this procedure. Did you follow it? Did you do it? Where? Show me.” And it becomes a more of a show me exercise. That’s the most valuable exercise you have from a data integrity standpoint, in my view.
Mack Powers (06:09):
So it’s somewhat of a hybrid model, I think you’ve called it, where you do some work remotely. But as you just expressed, nothing really does beast being there onsite with the customer, educating and walking through and guiding them through what an audit might feel like, and just confirming where data is.
Mitchell Wheeler (06:28):
Right. That’s what FDA would do. “Show me how you performed this test. Tell me where this goes. Show me, where’d you get that data, and who confirmed it.” And they follow the whole trail of how it was approved, how it was generated, where it was logged, and if there’s a paper part, and then there’s an electronic part, they follow that trail. And that’s where 43’s get written. Notices of violation, depending on the company.
Mack Powers (07:01):
Sure. Well, so you also mentioned earlier the discussion about what the definition of a true copy is. You and I were talking the other day about how this is also somewhat of a challenge to the industry. So explain more to us about that challenge, surrounding the true copying.
Mitchell Wheeler (07:19):
The true copy, and then the MHRA definition, which is the best place that it’s defined, is the metadata related to the electronic generation of that data.
Mitchell Wheeler (07:32):
So an example of an HPLC, you run an HPLC test. It generates a report based on what it sees from the test. You set up the test, you run the test, it generates a report. That is the true copy, that first report. Now, if you run the test again, it’s going to generate another copy, and then another copy, and another copy. Okay. Now, if you print it out and you put it in the log book, “Which one did you print out?” is the question. Did you print out the third one, which might have better results than the first test?
Mitchell Wheeler (08:10):
So there are a lot of games that can be played with it. There could be, depending on the integrity of the company. I see a lot of high integrity people in this business, and they just don’t realize that, “Hey, this is the way the regulators think when they look at things.” So the system has to be built so that you distinguish that true copy as the original, and you identify and have your system set up to control and manage the electronic copies as such.
Mack Powers (08:47):
So it sounds-
Mitchell Wheeler (08:48):
I thought that was clear.
Mack Powers (08:49):
It was clear. It sounds like you did do a couple of things. One is just somewhat educational, to really help people realize exactly how the FDA is going to view these types of issues. But also, then the other is just setting up procedures and processes to make sure that a company is able to manage that effectively, because that could be a challenge in and of itself, as well. I think, right?
Mitchell Wheeler (09:10):
Absolutely. And it can get to be, depending on the type of tests, one thing that in some of the systems that I set up are for prevention. You set up procedures and training to make sure that people are educated and they have the systems, the knowledge, and the training to do what they need to do.
Mitchell Wheeler (09:42):
Then you have the discovery of things, and you go and you do your evaluations and assessment, and you go in and you look at, “What’s actually being done? And how are the systems working? How is the training working? Maybe you need to retrain and change the procedures.” But that takes monitoring. And then there’s the response, and that’s the corrective action. If you have an OOS, a deviation, or a Kappa, which when if you identify a systemic issue across different test platforms, then you correct that. And then you go forward and then you challenge it again and make sure you’ve done the right thing.
Mack Powers (10:25):
So I’m sure companies will hire you to come in and help them set up from scratch, a data integrity program for their laboratory. So if someone were to call you in to do that, where would you start? And what are some of the key things that you’re going to be helping them to understand and implement?
Mitchell Wheeler (10:42):
Well, based on the things that we’ve just talked about, having a good, sound, conceptual discussion about what is data integrity to them, what is important, we’ll do a risk assessment on the types of tests, what means more to them regarding data. “Are there release tests? Is there stability? What type of tests were being performed? Are they important for release of their products? Are they important for the expiration dating of the product?” Those types of things need to be understood.
Mitchell Wheeler (11:21):
Then use that sound set of procedures, and you do training. And then the biggest thing about the training is to allow time to answer questions. This is new, this is a paradigm shift for a lot of these senior scientific experts who are performing tests, especially in the RND area, because they don’t see … They haven’t seen FDA, but FDA is starting to dig into those RND areas more, a theme that is a trend across the industry.
Mitchell Wheeler (11:53):
And it’s nothing about them being RND people. It’s the way that the regulators and auditors are starting to look at things, clinical trials. A lot of RND people are needed to, because of their talent, to look at things related to a protein characterization standpoint, for example. The ability for expiration dating, I mentioned before.
Mitchell Wheeler (12:18):
There are other things that are important, related to method development, and those things that are absolutely critical that we make sure data integrity is tight for. So those procedures, and then the monitoring of things, has to be done now periodically at a frequency that the group can handle. Everybody’s squeezed for resources, and now we have to be sensitive to that. But still, your program needs to work, so that’s on a case by case basis for companies, too.
Mitchell Wheeler (12:55):
And then the kappa piece, once you see things upon discovery and investigate thoroughly and implement corrective actions. Hard-wire them so that it don’t come undone, they can’t just come done easy. It’s something that deep thought discussion among the groups and collaboration with the teams is important to make sure that things are a holistically implemented.
Mack Powers (13:22):
Yeah. Your insight today has been very helpful. It helps us to get an idea of the way the FDA might see it, and some of the insights that you provided to us about the guidance documents and specific details that need to be thought about in terms of how you would comply. This has been very helpful, and we really appreciate your time today, Mitchell, and thanks for sharing with us.
Mitchell Wheeler (13:43):
Thank you very much, Mack, and I appreciate you having me.