WEBVTT
00:00:16.960 --> 00:00:21.120
Hello and welcome back to Better Biopharma, the official podcast of Bioprocess Online.
00:00:21.280 --> 00:00:24.640
I'm your host, Tyler Manichello, and on this episode, I'm excited to be joined by Dr.
00:00:24.719 --> 00:00:28.960
Tatiana Matviva, an experienced professional in the cell and gene therapy manufacturing space.
00:00:29.199 --> 00:00:32.320
Tatiana works in Boston and is an expert in compliance and quality.
00:00:32.479 --> 00:00:35.280
Tatiana, thank you so much for joining me on Better Biopharma.
00:00:35.679 --> 00:00:36.320
It's my pleasure.
00:00:36.399 --> 00:00:37.520
Thanks for having me, Tyler.
00:00:37.840 --> 00:00:38.320
Of course.
00:00:38.479 --> 00:00:39.280
It's my pleasure.
00:00:39.439 --> 00:00:57.520
And uh to frame up today's conversation, I wanted you to give our audience a little bit of background about yourself and your experience in the GMP space and specifically tell me how you think how you how you approach and think about quality and compliance as it relates to cell and gene therapy manufacturing.
00:00:58.159 --> 00:00:58.399
Sure.
00:00:58.560 --> 00:00:58.960
Yeah.
00:00:59.119 --> 00:01:10.879
Um, so cell and gene therapy manufacturing is of course more complicated than um the production of traditional um pharmaceuticals.
00:01:11.040 --> 00:01:14.079
It's been around for a limited time.
00:01:14.400 --> 00:01:29.359
And so we are still kind of seeing in real time that the regulatory framework around those cross seas is um always moving, it's changing, it's constantly being perfected.
00:01:29.519 --> 00:01:55.280
Um, and in the good manufacturing practice space, that regulatory framework around quality um is particularly relevant and just not quite as well followed as one might want to think, um, at least judging by um compliance issues discovered or published by the FDA and so on.
00:01:55.359 --> 00:02:10.319
And so I really am a strong proponent for quality by design, um, for progressive improvement of process, product training, and the entire infrastructure that really makes up GMP, right?
00:02:10.800 --> 00:02:30.240
So when we think about GMP, um, we're never really talking about no matter what phase of production we're in, we're never talking about a single clean room and some protocol, and um, you know, you you go inside and you cook whatever it is you're making.
00:02:30.479 --> 00:02:33.280
Uh it is much more than that.
00:02:33.439 --> 00:03:05.680
And it is an entire system, um, infrastructural system, which specifies quality criteria and um follows compliance and ensures that um there are controls every critical step of the way so that the product that's being made is safe and effective, and most importantly, that it meets the quality criteria that guarantee um patient safety.
00:03:06.000 --> 00:03:06.560
Yeah.
00:03:06.879 --> 00:03:07.360
Thank you.
00:03:07.520 --> 00:03:08.800
Yeah, I appreciate that.
00:03:09.039 --> 00:03:14.400
And when we spoke earlier, you told me this term quality unit independence.
00:03:14.639 --> 00:03:18.159
And I want to hear a little more about that, if you wouldn't mind telling our audience.
00:03:18.319 --> 00:03:21.039
What do you mean when you say quality unit independence?
00:03:21.120 --> 00:03:24.719
And what does that specifically mean in context of GMP?
00:03:26.560 --> 00:03:27.039
Sure.
00:03:27.199 --> 00:03:34.560
Yeah, so um again, GMP is a very kind of all-encompassing big term.
00:03:35.520 --> 00:04:08.560
Um, and it involves things like operations and manufacturing and environmental control and quality control and um, you know, various assays and how you um obtain your raw materials, all of these things and many more um are required for good manufacturing practice because together they ensure not only regulatory compliance by itself, which is important, but also that you have continuous quality um oversight.
00:04:08.879 --> 00:04:32.000
And what that allows you to do is really to capture any flaws or any vulnerabilities in any part of your process and design around those vulnerabilities controls that essentially increase the odds, if not always guarantee, that quality throughout the life cycle is preserved.
00:04:32.240 --> 00:04:59.839
And if you imagine any phase, phase one, two, three of manufacturing or even post-market, um you have a manufacturing um operations facility, and then you also have your supply chain, and you have um, you know, your validation of all of your facilities and your equipment and your process, and you have your training for all qualified personnel and so on.
00:05:00.399 --> 00:05:20.000
And ultimately, you are making a therapy or or a drug um that is supposed to go into a patient and uh keep them safe and help them with whatever indication um it's about.
00:05:21.040 --> 00:05:50.319
Well, uh it's very important that during various phases of this whole process um there can be oversight about whether everything is done according to um approved written procedures, whether there have been any incidents or deviations from a set process, whether all quality control um assays have been correctly performed and quality attributes met.
00:05:50.800 --> 00:06:13.120
Um, all of those things guarantee at various points in a process that the therapy you are uh sending to the patient or putting up on the market is designed and produced according to the highest quality and safety standards that are enforced federally but also internationally.
00:06:13.839 --> 00:06:16.800
And as you can imagine, a lot goes into production, right?
00:06:16.879 --> 00:06:18.319
There's so much at stake.
00:06:18.560 --> 00:06:21.600
It's um very expensive very often.
00:06:21.759 --> 00:06:31.120
Um, a lot, again, a lot of infrastructure has gone into setting up a place that can, in fact, manufacture according to good manufacturing practices.
00:06:31.199 --> 00:06:35.600
Um, a lot of very highly qualified expert personnel is required.
00:06:35.839 --> 00:06:42.319
A lot of special conditions for um the production of the therapy are required.
00:06:42.480 --> 00:06:57.680
And so you want to be able to make your drug product or your therapy and put it out in the market so that it, you know, it's commercially available, that's related to profits, it's also related to patient well-being, and so on.
00:06:57.839 --> 00:07:05.439
And so the most important component I think of of all of this is your quality unit.
00:07:05.680 --> 00:07:26.800
And the quality unit, as contemplated by the FDA, by other international organizations, is a unit that must be independent from the rest of the managerial aspect and operational aspects of a manufacturing operation.
00:07:27.040 --> 00:07:27.600
Why?
00:07:27.920 --> 00:07:44.959
Because you want to be able to have a highly qualified, skilled unit whose only concern is not to agree with management, it's not to push something out into the world that does not qualify.
00:07:45.120 --> 00:08:00.639
Um, their only concern has to be whether the data they're presented with, the product they're presented with, the process records, whether all of those are actually meeting pre-specified criteria.
00:08:00.879 --> 00:08:09.439
And then and only then to sign off and say, yes, this is a well-made compliant product.
00:08:09.680 --> 00:08:35.919
And when your quality unit is not truly independent, what happens is that because of pressures from maybe management or for whatever reason, your quality unit um has priorities other than singularly assessing, objectively assessing the quality and compliance of your um production.
00:08:36.320 --> 00:08:38.240
And that's very dangerous, right?
00:08:38.639 --> 00:08:48.399
And you never want to only on paper have a separate quality unit that does not depend on um the approval of management.
00:08:48.559 --> 00:09:02.320
You never want to only on paper have a quality unit that is not subject to pressures, or there that is not overridden, or in any other way non-authoritative with respect to the job they have.
00:09:02.559 --> 00:09:33.600
And so independence of your quality unit is about as important as anything else in your process, if not more, because that is the safeguard against um pushing a therapy or a product or a process or a system of processes that ultimately is acting not with the patient's best interest at heart, but perhaps with other motives of production.
00:09:33.919 --> 00:09:36.720
And this is why independence is absolutely imperative.
00:09:36.799 --> 00:09:41.279
And we've seen recently what happens when independence is doubted.
00:09:42.159 --> 00:09:43.600
Yeah, thank you, Tatiana.
00:09:43.679 --> 00:09:57.840
Yeah, I mean, I completely agree with this concept of kind of uh separation of church church and state, if you will, of like quality should remain independent, and uh you certainly can't take the teeth, it needs to have its teeth, right?
00:09:57.919 --> 00:10:04.639
As a as a kind of governing waypoint or or uh lighthouse almost, if you will.
00:10:04.879 --> 00:10:12.240
Um, and I think that kind of ties into this this notion of uh a quality hierarchy, I think you described it when we when we spoke before.
00:10:12.399 --> 00:10:23.600
Um, do you see this kind of in in organization as like if we have an independent quality unit that's sitting at the top of this this pyramid almost of of a quality hierarchy?
00:10:23.679 --> 00:10:37.279
And would you mind explaining that a little further and and what happens when this hierarchy maybe isn't respected, or when you try to take the teeth out of your quality team and keep them from actually prioritizing product quality and thus patient safety?
00:10:37.600 --> 00:10:38.320
Yeah, absolutely.
00:10:38.639 --> 00:10:42.240
That's a very dangerous um thing to do.
00:10:42.399 --> 00:10:47.600
And you know, sometimes it happens uh without any malice or intent.
00:10:47.759 --> 00:10:56.720
Um, I certainly don't think that people aim to breach quality or to disable a quality unit.
00:10:56.960 --> 00:10:58.799
Nobody sets out to do that, right?
00:10:58.960 --> 00:10:59.279
Right.
00:10:59.440 --> 00:11:10.639
Um, but for various pressures, uh, maybe even knowledge gaps, sometimes authority and responsibility misalign, right?
00:11:10.720 --> 00:11:16.399
And so you have a quality unit and it's um apparently separate.
00:11:16.559 --> 00:11:39.759
However, um the quality unit's decision regarding any part of the process they're responsible for is overridden, or perhaps there's something more urgent, and so something is overlooked, or perhaps there is legitimate pressure from um upper management for a trillion possible reasons and so on.
00:11:40.000 --> 00:11:51.039
And there is also occasionally the misunderstanding of what the quality unit really should be relative to other functions in a manufacturing environment, right?
00:11:51.600 --> 00:12:04.720
Management um has a lot of responsibility, and ultimately uh they are accountable for everything that happens um in an organization that they're in charge of.
00:12:04.799 --> 00:12:21.840
Um, and quality is a part of that, but only in the sense that management has an obligation to um to run with the judgment of a good and independent quality unit.
00:12:22.159 --> 00:12:28.000
They are not authorized to um say, oh no, you know, ignore that.
00:12:28.159 --> 00:12:29.759
We're just gonna continue.
00:12:30.000 --> 00:12:43.679
Um they cannot um at any point in practice consider themselves to be um to have the final word when it comes to quality and quality attributes.
00:12:43.919 --> 00:12:48.720
And when that happens, um, the consequences are very serious.
00:12:48.879 --> 00:12:57.120
And they're serious not only because, you know, during an inspection, if that ever transpires, that's a very big problem.
00:12:57.679 --> 00:12:59.840
It's a problem of credibility.
00:13:00.080 --> 00:13:18.639
Um, it is much more than a singular violation because not only does it capture violation of some regulatory uh standard, but it also demonstrates that there is a great degree of subjectivity in the way quality oversight works at that organization, right?
00:13:19.360 --> 00:13:25.759
And um that's a very big red flag if you're the FDA, and rightfully so.
00:13:26.080 --> 00:13:54.480
And so when uh when that happens, um it's important for an organization, I think, to really uh craft a strategy, a response that doesn't just say something like, you know what, yes, we acknowledged that this or that violation was committed, and we're gonna rewrite our SOPs and we're gonna retrain our people.
00:13:54.799 --> 00:13:58.320
And we're this isn't this is grossly inadequate, right?
00:13:58.480 --> 00:14:17.120
Because it doesn't address what transpires to be a cultural problem, a problem of authority, a problem of management understanding where they need to step back, um, a problem of essentially having a quality unit that is dysfunctional or doesn't work at all.
00:14:17.440 --> 00:14:40.480
And so it's important that everybody understands why quality is independent, and that that independence means not only the ability and obligation to conduct independent analysis and issue decisions, but that those decisions are imperative to follow.
00:14:40.960 --> 00:14:54.080
And if that if that's broken, then the organization is broken and the manufacturing culture and quality culture are not working the way they're supposed to.
00:14:55.440 --> 00:14:57.279
Yeah, thanks, thanks, Tatiana.
00:14:57.360 --> 00:14:59.759
I I agree, and I think you make a great point there.
00:14:59.919 --> 00:15:19.279
Like nobody means to uh compromise quality and quality oversight in manufacturing, but it's it's kind of a slippery slope of when you let I don't want to say conflicting interests, because ultimately I think everybody would agree that patient safety and efficacious product is of the utmost importance.
00:15:19.440 --> 00:15:27.679
But you know, when you when you kind of let the noise creep in and distract from rigid quality oversight, I think it it can happen.
00:15:27.759 --> 00:15:34.960
And you alluded to this earlier, some incidents of serious quality violations we're seeing at some pretty well-established manufacturing sites.
00:15:35.039 --> 00:15:40.080
And I won't name any names, but I'm curious to hear you talk a little bit more about that.
00:15:40.240 --> 00:15:50.159
Um, first of all, would you say it is accurate that there has been an increase in these kinds of incidents of uh non-compliance at top manufacturing spots?
00:15:50.240 --> 00:15:52.240
And and are you surprised by this?
00:15:53.519 --> 00:15:56.639
Yeah, you know, we've all seen them.
00:15:56.799 --> 00:16:01.679
Um, some of them are pretty egregious um and and truly unfortunate.
00:16:01.759 --> 00:16:36.159
Um, and I think they speak exactly to that loss of credibility because of uh a culture of misunderstanding the way uh quality should work, um, and misaligning responsibility and authority when it comes to the quality unit, where you know um they are responsible for quality oversight, but de facto they're not being given perhaps the authority to exercise that responsibility correctly.
00:16:36.799 --> 00:16:41.759
And um I will say, yeah, some of some of these are very egregious.
00:16:41.840 --> 00:16:45.039
Um and would you say is the most egregious one that comes to mind?
00:16:45.200 --> 00:16:46.159
Not to interrupt.
00:16:46.639 --> 00:16:54.399
Oh, you know, there were there were several recent ones that were just to me, and I'm sure to too many, um, quite shocking.
00:16:54.559 --> 00:17:26.319
And they were not only about environmental findings, so contamination and um, you know, growth of mold and and all those terrible, terrible things that are hard to imagine for many who understand what GMP must be, but um then you know there's some the build on top of that, and then there was some um data integrity breaches, there was falsification of data, there was hiding or replacing documents that weren't previously there.
00:17:26.640 --> 00:17:40.480
Um there was um very obvious um gap in knowledge between regulatory practices, written procedures, and what was really happening.
00:17:40.799 --> 00:17:52.240
And ultimately there were some admissions that, in fact, there was some data substitution and and uh forging, forging of documents, and that that's about as serious as you can get, right?
00:17:52.559 --> 00:18:49.839
So, not only is it a quality breach in the manufacturing environment per se, with contamination and poor environmental controls and various such violations, which are serious enough on their own, but when you add to the knowledge of those breaches, um a very considerable and demonstrable um data integrity breach, now we're talking about not only a problem, but also a mentality or a culture in which knowingly we take that problem and we acknowledge that it's there and we don't take care of it, but we are also going to kind of mask that whole situation and um and try to um disguise it or be dishonest about it.
00:18:50.000 --> 00:18:51.839
That's very serious, right?
00:18:52.000 --> 00:19:09.359
So now the problem is that it's very difficult for anybody um who's an independent um investigator or observer or auditor, even, to really know when they're looking through records of an organization like that.
00:19:09.920 --> 00:19:15.680
And those records indicate that some part of a process um has gone really well.
00:19:16.000 --> 00:19:18.559
There's no violation, nothing's going on.
00:19:18.799 --> 00:19:20.640
That's no longer believable, right?
00:19:20.720 --> 00:19:36.240
It's very difficult to trust that the scope of very egregious violations is limited because there is no evidence of accountability and there's no evidence of um really understanding the scope of the problem.
00:19:36.400 --> 00:19:38.079
There's evidence to the contrary, right?
00:19:38.160 --> 00:19:47.839
There is evidence that um there's been a little bit of deception, perhaps, that you create data that wasn't there, that you correct data that was there.
00:19:48.079 --> 00:19:57.759
And once that happens, defining the scope of risk to patients is very difficult to uh do.
00:19:58.640 --> 00:20:18.880
And so, you know, that's when you see not only recommendations for, say, an independent um consultant to be hired to really assess everything because there's no belief that if management is given the task to assess the scope of the problem, they will do it, um, or they will do it without bias anyway.
00:20:19.039 --> 00:20:24.240
So that's when those recommendations for independent consultants come in.
00:20:24.559 --> 00:20:29.680
And um that's a signal that that's a signal that there's loss of credibility.
00:20:29.839 --> 00:20:40.960
Um, and I will say we also see more and more some of the responses that are published to such serious letters findings by the FDA.
00:20:41.119 --> 00:20:46.880
And the responses often demonstrate a little bit of lack of insight, right?
00:20:47.119 --> 00:21:08.400
So um they kind of respond with um uh to a specific localized problem that's been mentioned, or uh with band-aid treatments, kind of, oh well, we designed a new SOP and now we're gonna retrain everyone, or oh yeah, there is mold, we're gonna remove the mold.
00:21:08.880 --> 00:21:13.039
Those are inadequate, and and very often the FDA will say that, right?
00:21:13.200 --> 00:21:20.400
They will say um this, they they will describe and acknowledge the response, and then they will say it's inadequate and demonstrate why.
00:21:20.640 --> 00:21:24.720
And the why is because the scope is not acknowledged.
00:21:24.799 --> 00:21:28.400
So you can't just, you know, spot treat those things.
00:21:28.559 --> 00:21:39.440
You have to do a thorough retrospective analysis of the full scope, not where problems happened, but where they could have happened, right?
00:21:39.599 --> 00:21:42.079
So you need to really go deep.
00:21:42.400 --> 00:21:48.880
Anywhere where there presumably can be a risk, you need to um you really go and investigate.
00:21:49.200 --> 00:21:55.359
And you don't get to say, well, presumably there could be risk someplace else.
00:21:56.160 --> 00:21:59.680
Um, but we can confirm that there was.
00:22:00.400 --> 00:22:02.000
You're now in a very different place.
00:22:02.160 --> 00:22:11.920
Actually, the job is unless you can confirm with objective evidence that there wasn't risk, you have to assume there was.
00:22:12.240 --> 00:22:19.119
So, you know, it's just important in those responses, nobody's looking for a quick solution, that there isn't one.
00:22:19.279 --> 00:22:22.559
And I think inspectors understand that.
00:22:22.799 --> 00:22:31.440
Um, they're they are looking for accountability, they're looking for depth of analysis, they're looking for true remediation measures.
00:22:32.079 --> 00:22:44.960
And um, unless they see them, um, they see a reasonable plan that's not only risk-based, but it's also very well justified, um it's not going to be enough.
00:22:45.200 --> 00:22:57.920
So the first and most important thing when you have such um serious um problems with quality and data integrity, um, the most important thing is to demonstrate accountability.
00:22:58.160 --> 00:23:04.960
And sometimes what that means is causing the company or the organization significant discomfort, right?
00:23:05.039 --> 00:23:37.039
Because effectively what they have to say is we're gonna put the brakes, we're gonna stop operations until we have implemented a very thorough remediation because we want to demonstrate that until we can reliably produce high-quality therapeutics, we're not going to allow our production interests to jeopardize um people on the market who are simply seeking to buy therapies they believe are safe.
00:23:37.279 --> 00:23:39.759
And so um it's a big deal.
00:23:39.920 --> 00:23:45.599
Um, yes, they are very egregious, um, and we've all seen them.
00:23:45.680 --> 00:23:47.119
That that's not a secret.
00:23:47.359 --> 00:24:08.319
Um I hope, you know, no matter what, I'm I'm fairly young in this field, and I want to say that I hope it's not much worse than it's always been, but it feels like uh there are some pretty significant findings that uh leave us all a little bit surprised.
00:24:09.039 --> 00:24:10.480
Yeah, thank you, Tatiana.
00:24:10.799 --> 00:24:21.680
In any case, it's it's crazy that in in 2026 we're seeing these kind of uh infractions um amongst, like you know, you said established names.
00:24:21.839 --> 00:24:29.039
It's kind of it's worrisome to say the least for the state of of the industry as well as just for for patient safety.
00:24:29.119 --> 00:24:33.200
It's kind of it's insane to me that it it's even happening.
00:24:33.440 --> 00:24:47.759
But I'm curious if out of all these 483 letters and and violations, if is there a particular kind of common theme that you've seen across them or a specific type of infraction that seems the most common uh among among those you've seen?
00:24:48.559 --> 00:24:49.119
I think so.
00:24:49.279 --> 00:24:52.000
And and it's kind of twofold.
00:24:52.079 --> 00:25:06.559
Um, and uh someone might disagree with me and they're welcome to, but I think um a 483 uh letter is always quality related in some way.
00:25:06.799 --> 00:25:29.680
Um and uh there transpire to be uh two parts that I the two themes, I suppose, that keep recurring that I keep seeing um across these multiple letters uh of various seriousness, um, but there is a common thread, and it is one um a dysfunctional quality unit.
00:25:29.920 --> 00:25:31.680
You have to ask why, right?
00:25:31.839 --> 00:26:19.359
So it it the the answer is not gonna be the same for every place, but for some reason, whether it's contamination, whether it's quality control, whether it is um incomplete validation procedures or incomplete or inadequate SOPs, or it's maybe um, you know, data breaches or dependence um of the quality unit on um management, um, which is not ideal, it there is a lack of sufficient authority and preparedness in the quality unit, which means that the quality unit is not functioning effectively.
00:26:19.599 --> 00:26:22.400
And if it's not, that's it.
00:26:22.640 --> 00:26:24.400
It that really is it.
00:26:24.640 --> 00:26:32.079
If you have a non-functional quality unit, which you're responsible solely for the authorization of, right?
00:26:32.240 --> 00:26:48.000
If I'm management with respect to a quality unit, my job is to make sure that quality and everybody else knows that what quality says goes, not the other way around, right?
00:26:49.039 --> 00:26:55.279
And as soon as that's not true, problems will inevitably emerge.
00:26:55.599 --> 00:26:56.400
So that's one.
00:26:56.480 --> 00:27:04.640
And the other one is this um kind of perplexing dropping up the ball.
00:27:04.799 --> 00:27:22.240
Um, and we've seen that a little bit, we've seen that in Europe now, we've seen some of it in the United States, we've seen some in Asia, but this kind of uh facility level, very um obvious environmental breach, right?
00:27:22.480 --> 00:27:38.880
In which even if you have uh the best documentation and the best procedures otherwise, the environment in which you manufacture has become commonly unsafe.
00:27:40.000 --> 00:27:50.400
And uh there is evidence that some of these breaches in environmental control have been noted and they've been noted before, and then they've been noted again.
00:27:50.880 --> 00:28:14.160
And uh it's hard to fathom at each of those findings what the response was and whether there was a systemic response and whether there was any follow-through or follow-up with whether it was implemented completely and correctly and effectively, and whether there's a plan forward for you know continuous oversight.
00:28:14.640 --> 00:28:39.599
But it is kind of shocking that in this day and age, as you said, you can have multiple documented violations of basic safety in manufacturing, and yet they continue until you know there is a very serious notice, and perhaps now you really are going to have to stop operations, there's loss of credibility, there's loss of reputation.
00:28:39.839 --> 00:28:42.880
Um it's it's very serious.
00:28:43.279 --> 00:29:11.119
So between compromised quality authority and a kind of an inability to really give an in-depth strategic long-term tested response to violations, I think those two are uh they can be lethal, they can be uh damaging beyond repair to an organization, but also who suffers in the end, right?
00:29:11.279 --> 00:29:17.599
Um, it's it's always patients, it's always people who uh seek a therapeutic.
00:29:17.759 --> 00:29:25.440
And you know, I don't when you buy um anything from a pharmacy, do you ever think that it could be dangerous to you?
00:29:25.599 --> 00:29:26.799
Probably not, right?
00:29:26.880 --> 00:29:33.039
We all trust that medicines and therapies are gonna be good for us.
00:29:33.279 --> 00:29:47.440
And so when the manufacturer digresses from that practice or forgets that biggest of all responsibilities, I think that's um that's just a tragedy, really.
00:29:47.920 --> 00:29:49.279
Yeah, I completely agree.
00:29:49.440 --> 00:29:52.559
Especially, again, it's 2026 for A.
00:29:52.960 --> 00:29:54.559
That's insane in and of itself.
00:29:54.720 --> 00:30:09.039
But especially like I, you know, in my lifetime, I don't think I've ever seen public trust be so thin in institutional power and protection and just uh just institutions and science and medicine in general.
00:30:09.119 --> 00:30:22.240
And so, like, absolutely should there should not be any any actions going on at established manufacturing sites, no less, that could in any way damage the fabric of that already thin veil of trust.
00:30:22.400 --> 00:30:24.640
And um, it is a tragedy indeed, I think.
00:30:25.039 --> 00:30:27.119
I think that's a very important point you're making, right?
00:30:27.200 --> 00:30:44.480
Because one of the side damages, right, if we just zoom out away from the specific finding in a 483, um one of the horrible side effects of these uh violations is that ultimately that affects the entire field, right?
00:30:44.559 --> 00:31:00.000
It affects the the public trust, it affects belief that um our our practices and our processes are the way they should be, and that patient safety is really the biggest priority as it is and it should be.
00:31:00.240 --> 00:31:02.000
And that hurts everybody, right?
00:31:02.079 --> 00:31:05.839
That that hurts other manufacturers who are doing a good job.
00:31:06.079 --> 00:31:12.160
Um it hurts public trust, it hurts um patients the most.
00:31:12.480 --> 00:31:26.000
And so uh there are so many unfortunate consequences of that level of neglect for um good manufacturing practice and just compliance and safety.
00:31:26.160 --> 00:31:31.920
Um and and that's that's not a small side effect of those issues.
00:31:32.400 --> 00:31:34.480
No, yeah, I mean perception is everything, right?
00:31:34.559 --> 00:31:39.759
And it doesn't certainly isn't helping the perception that needs to be corrected in the first place.
00:31:40.559 --> 00:31:41.359
I agree.
00:31:41.599 --> 00:31:58.880
Um you know, we we've talked about kind of what what dominoes need to fall, so to speak, and to to allow for quality control to slip and um when when quality isn't held independently and at the top of the hierarchy.
00:31:59.039 --> 00:32:16.079
Um so we know we know kind of how this happens abstractly, but I'm curious to hear from you more on how do you you mentioned before the idea of quality by design, which isn't a new concept, but how do you think organizations should integrate this philosophy into their operations from the beginning?
00:32:16.319 --> 00:32:34.240
And more importantly, you know, in your experience, how do you find success actually upholding that that that idea of quality by design and coordinating myriad of teams and systems that are involved in biomanufacturing, especially I think cell therapies might correct me if I'm wrong.
00:32:34.319 --> 00:32:48.640
I feel like cell and gene therapies have the highest demand for product quality because they're just there's so much there's so much there that could potentially be uh ruinous to the yeah, there's a lot at stake, no doubt about it.
00:32:48.799 --> 00:32:52.799
Um and cell and gene therapy production is not a simple process.
00:32:52.960 --> 00:32:55.359
Um it's actually highly complicated.
00:32:55.519 --> 00:33:02.480
It's also occasionally variable depending on what specifically you're making and what indications that therapy is for.
00:33:02.559 --> 00:33:04.960
And so you're right, the stakes are very high.
00:33:05.119 --> 00:33:11.359
Um, and so quality by design becomes even more imperative if that's possible, right?
00:33:11.519 --> 00:33:14.799
And so just generally quality by design.
00:33:15.200 --> 00:33:30.720
Um, if you take the idea that once upon a time, when producing therapies or pharmaceuticals or drugs, um, you know, you you make them, make them, make them, and at the end you have a quality control test and it's pass or fail.
00:33:30.960 --> 00:33:34.319
Um, that's that's really not a very good system, right?
00:33:34.720 --> 00:33:51.759
Um, you are hoping that at the end uh you get a pass one way or another, but you're not really thoughtfully designing your process and your product with quality integrated already.
00:33:52.000 --> 00:33:55.519
And really, quality by design is intended to remedy that, right?
00:33:55.680 --> 00:34:05.920
So you have whatever your product is, um, you start with something we all know is called the quality target uh product profile.
00:34:06.079 --> 00:34:15.039
And so you kind of you just define what this what the critical quality attributes of your product need to be, right?
00:34:15.679 --> 00:34:53.599
How it gets to be effective, safe, what that means, what what are the the critical basically components or properties um that you want to be seeing in your product, and then you design the processes um and um import the materials for that production by defining then the critical material um attributes and properties, and then you go to um things like critical process parameters.
00:34:53.760 --> 00:35:23.519
So, what parts of the process must be very thoroughly controlled, well understood, um, and you you define a system in which you connect uh critical process parameters with critical material attributes, with um CQAs, so that you know that you're designing a production system specifically with the uh product quality um attributes in mind, right?
00:35:24.079 --> 00:35:34.159
So there are many things in GMP um that are very difficult to constantly control at the same level of resolution, right?
00:35:34.239 --> 00:35:37.760
There are million systems, there are million things that go on.
00:35:38.079 --> 00:36:07.199
And so um instead of trying to distribute all resources uniformly to those myriad things that could influence a process, it it is just more rational and more scientifically sound to sort of grade where the biggest risks are and really focus on those high risk and then lower risk and then lowest risk spaces.
00:36:07.360 --> 00:36:14.559
Um, and when you do that, um you you assess risk with respect to what you want in the end, right?
00:36:14.719 --> 00:36:22.960
To the critical quality attribute, your drug, and you focus resources there, which is sensible because those matter the most.
00:36:23.039 --> 00:36:27.199
It's economical, it is reasonable because it's very well justified.
00:36:27.280 --> 00:36:30.079
You can rationally demonstrate why those are important.
00:36:30.559 --> 00:36:50.719
And so you design a manufacturing operation which has taken into account and defined well what the settings must be so that you end up with the best and safest compliant product that you can make.
00:36:51.199 --> 00:36:52.719
That's quality by design.
00:36:52.880 --> 00:37:10.719
Basically, you integrate controls that enhance and make much more probable the most critical components in manufacturing within an umbrella of compliance and minimize risks that you want to, of course, always be minimizing.
00:37:11.199 --> 00:37:13.920
That process is always iterative.
00:37:14.079 --> 00:37:15.920
You will uh learn as you go.
00:37:16.079 --> 00:37:21.920
I I don't think that in GMP it's possible to come up with a final perfect system.
00:37:22.159 --> 00:37:25.760
There is always going to be a problem we didn't predict.
00:37:26.000 --> 00:37:44.320
Um, but what we can do is have a solid enough framework that can capture the problem, that can define it, that has a very clear path of how that problem will be handled from a quality and compliance perspective.
00:37:44.559 --> 00:37:52.639
Um, and that knowledge must be very, very clear and well understood by everyone on the floor, right?
00:37:52.800 --> 00:38:12.960
Whether it's your uh process optimization team or your management or your quality control people or your manufacturing operators, every one of them must have the answer ready to a question like what would happen if this deviation from your protocol happens?
00:38:14.159 --> 00:38:15.039
What would you do?
00:38:15.199 --> 00:38:16.400
When would you announce it?
00:38:16.559 --> 00:38:17.920
How would you escalate it?
00:38:18.079 --> 00:38:20.000
What would be the next step of control?
00:38:20.400 --> 00:38:21.039
And so on.
00:38:21.119 --> 00:38:21.920
Do you continue?
00:38:22.159 --> 00:38:24.480
Those questions are absolutely critical.
00:38:24.800 --> 00:38:34.960
And everyone in a manufacturing operation must know clearly and confidently what they would do, what they should do.
00:38:35.199 --> 00:38:36.800
That's good training.
00:38:37.280 --> 00:38:52.880
So when things slip, is where you start to see um a little bit of uncertainty, or or you know, people who can't remember when they were last trained, or they're not sure if they want to say the thing they think is right because there is a manager next to them.
00:38:53.039 --> 00:38:54.960
That's where problems start to happen.
00:38:55.199 --> 00:39:02.159
So quality by design is exactly what everybody should be doing.
00:39:02.320 --> 00:39:12.639
It's a very rational framework, it's a very resource economical framework, and it's a safety optimization framework.
00:39:12.880 --> 00:39:14.559
So we have to have it.
00:39:14.639 --> 00:39:35.280
Um, and I think it's important to really spend the time up front before a single phase or as multiple phases are being hypothesized or planned for, and spend sufficient time on determining all of those components of quality by design.
00:39:35.519 --> 00:39:41.039
Um, you know, sit down and imagine all the crazy problems and terrible things that can happen.
00:39:41.360 --> 00:40:14.320
Be imaginative about everything that can go wrong, and then rank them and assess them and understand how they contribute to the various elements of quality by design, be those your um critical quality attributes of the drug or your materials um or process, whatever those are, define them, define them again, risk rank them, understand where you're vulnerable, design controls and test those controls, right?
00:40:14.880 --> 00:40:24.800
Improve them, and you do that over several iterations, and now you really have a system that works in the direction of optimizing quality.
00:40:25.440 --> 00:40:36.639
And the rest is what people will choose to do, in the sense that you do ultimately have to have uh an empowered quality unit.
00:40:36.880 --> 00:40:46.079
You do have to have management with humility to say there are problems and quality found them, and so now we're gonna follow whatever they say.
00:40:46.480 --> 00:41:06.880
Um it's a lot, and there's a lot at stake, and there's a lot that um is invested in continuous operations, and uh it's certainly understandable why disruptions are sometimes unpleasant, sometimes terribly destructive, but uh quality first, right?
00:41:07.119 --> 00:41:23.920
And if that's not the mana of any production operation, sooner or later, problems will begin to emerge more systemically, and um then it would be a little too late.
00:41:24.239 --> 00:41:44.880
So it's it's worth an investment on the front end to design processes, to design practices, and to empower the people who are supposed to carry them out um in the direction of optimizing state and quality so that patients really can get what they need safer and sooner.
00:41:45.599 --> 00:41:45.920
Yeah.
00:41:46.159 --> 00:41:47.360
Thank you, Tatiana.
00:41:47.519 --> 00:41:48.400
I I agree.
00:41:48.559 --> 00:41:51.280
I appreciate your your thoughts and wisdom there.
00:41:51.519 --> 00:41:55.119
Um and I want to ask, and I want to I want to be respectful of your time.
00:41:55.199 --> 00:42:05.199
I just want to ask if there's any single piece of advice or one idea you'd like to impart to our audience, if there's one thing that you want them to take away from this conversation, what would that be?
00:42:06.639 --> 00:42:13.199
It would really be to, you know, uh it always depends on who you're talking to.
00:42:13.280 --> 00:42:47.840
Um, but whether you're talking to management or you're talking to quality or you're talking to operations, I think it's really important to understand why um quality oversight is so important and what that means, and to put it in a larger framework of um certainly the self-interest of the company or production organization in terms of again, um, speed to market and reliability and patient trust and regulators' trust.
00:42:48.000 --> 00:43:07.840
But also just to remember that these are rules that are there for very good reason, and that a single slip into overriding the right decision, um, the single gesture of disempowering um your quality unit is a bad seed.
00:43:08.000 --> 00:43:19.840
Um, and it is the start of a convention which tells people around you, whether you've said it out loud or not, that the rules are going to be bent in certain ways.
00:43:20.000 --> 00:43:26.960
Um, and so what your team will understand is that oh, here are the priorities, right?
00:43:27.119 --> 00:43:35.039
Quality is supposed to be here, but actually it's second to this other priority that was used to override a quality decision.
00:43:35.280 --> 00:43:42.239
So just remember the people are on the line, uh, patients are on the line, and I don't think there is anything more important than that.
00:43:42.480 --> 00:43:45.280
Um, I think public trust is important.
00:43:45.679 --> 00:44:07.679
And so if it's it's entirely possible to uh restore faith in something as important as therbic production, um, if we really um agree that we will protect the principles that we have committed to um to uphold.
00:44:07.760 --> 00:44:10.719
And that may be a little idealistic, but I actually don't believe that.
00:44:10.880 --> 00:44:12.719
I think it's completely possible.
00:44:13.199 --> 00:44:13.360
Yeah.
00:44:13.519 --> 00:44:14.800
Thank you, Tatiana.
00:44:14.960 --> 00:44:18.000
That's a great, great piece of advice, and I completely agree.
00:44:18.239 --> 00:44:26.000
And now, lastly, I'm gonna ask you the question I ask every guest on the show, and that is big or small, Tatiana, how do you think we can better biopharma?
00:44:26.880 --> 00:44:29.679
Um, that's a wow, that's a big question.
00:44:29.840 --> 00:44:37.440
Um I think have more conversations like these and make them uh normal and common.
00:44:37.599 --> 00:44:45.440
Um, have more conversation in the direction of quality improvement, make that a mainstream mentality if it's not already.
00:44:45.679 --> 00:44:49.599
Um, and uh be truthful and transparent.
00:44:49.840 --> 00:44:54.719
So, you know, there are two things that can compromise what we do.
00:44:54.800 --> 00:45:13.440
And one is by loss of patient trust, and one is by um overriding um the roles of people whose job is to stop us even when it's um it's not what we want, right?
00:45:13.760 --> 00:45:35.199
And so have more conversations in the direction of enhancing quality, in the direction of integrity, in the direction of empowering junior people to carry out principles of quality when confronted with other demands, right?
00:45:35.440 --> 00:45:53.599
Um, I think that the most powerful thing is always to have people lower down in the production chain or hierarchy who know that they can say stop, they can say no to a process because they've noticed something and rightfully brought it up.
00:45:53.840 --> 00:45:58.960
Empower those people, empower people in your organization to be transparent.
00:45:59.280 --> 00:46:08.239
Um, thank them when they bring up quality um incidents or or any kind of process problem.
00:46:08.880 --> 00:46:14.400
Um and uh promote them when they do that because that's not easy to do.
00:46:15.039 --> 00:46:15.599
Yeah.
00:46:15.840 --> 00:46:17.119
That's great, Satiana.
00:46:17.199 --> 00:46:17.920
Thank you so much.
00:46:18.079 --> 00:46:26.559
Um thank you for being here, and thank you so much to our audience for tuning in for this episode of Better Bioforma, the official podcast of Bioprocess Online.
00:46:26.639 --> 00:46:29.039
I'm your host, Tyler Minachello, and I'll see you next time.
00:00:16.960 --> 00:00:21.120
Hello and welcome back to Better Biopharma, the official podcast of Bioprocess Online.
00:00:21.280 --> 00:00:24.640
I'm your host, Tyler Manichello, and on this episode, I'm excited to be joined by Dr.
00:00:24.719 --> 00:00:28.960
Tatiana Matviva, an experienced professional in the cell and gene therapy manufacturing space.
00:00:29.199 --> 00:00:32.320
Tatiana works in Boston and is an expert in compliance and quality.
00:00:32.479 --> 00:00:35.280
Tatiana, thank you so much for joining me on Better Biopharma.
00:00:35.679 --> 00:00:36.320
It's my pleasure.
00:00:36.399 --> 00:00:37.520
Thanks for having me, Tyler.
00:00:37.840 --> 00:00:38.320
Of course.
00:00:38.479 --> 00:00:39.280
It's my pleasure.
00:00:39.439 --> 00:00:57.520
And uh to frame up today's conversation, I wanted you to give our audience a little bit of background about yourself and your experience in the GMP space and specifically tell me how you think how you how you approach and think about quality and compliance as it relates to cell and gene therapy manufacturing.
00:00:58.159 --> 00:00:58.399
Sure.
00:00:58.560 --> 00:00:58.960
Yeah.
00:00:59.119 --> 00:01:10.879
Um, so cell and gene therapy manufacturing is of course more complicated than um the production of traditional um pharmaceuticals.
00:01:11.040 --> 00:01:14.079
It's been around for a limited time.
00:01:14.400 --> 00:01:29.359
And so we are still kind of seeing in real time that the regulatory framework around those cross seas is um always moving, it's changing, it's constantly being perfected.
00:01:29.519 --> 00:01:55.280
Um, and in the good manufacturing practice space, that regulatory framework around quality um is particularly relevant and just not quite as well followed as one might want to think, um, at least judging by um compliance issues discovered or published by the FDA and so on.
00:01:55.359 --> 00:02:10.319
And so I really am a strong proponent for quality by design, um, for progressive improvement of process, product training, and the entire infrastructure that really makes up GMP, right?
00:02:10.800 --> 00:02:30.240
So when we think about GMP, um, we're never really talking about no matter what phase of production we're in, we're never talking about a single clean room and some protocol, and um, you know, you you go inside and you cook whatever it is you're making.
00:02:30.479 --> 00:02:33.280
Uh it is much more than that.
00:02:33.439 --> 00:03:05.680
And it is an entire system, um, infrastructural system, which specifies quality criteria and um follows compliance and ensures that um there are controls every critical step of the way so that the product that's being made is safe and effective, and most importantly, that it meets the quality criteria that guarantee um patient safety.
00:03:06.000 --> 00:03:06.560
Yeah.
00:03:06.879 --> 00:03:07.360
Thank you.
00:03:07.520 --> 00:03:08.800
Yeah, I appreciate that.
00:03:09.039 --> 00:03:14.400
And when we spoke earlier, you told me this term quality unit independence.
00:03:14.639 --> 00:03:18.159
And I want to hear a little more about that, if you wouldn't mind telling our audience.
00:03:18.319 --> 00:03:21.039
What do you mean when you say quality unit independence?
00:03:21.120 --> 00:03:24.719
And what does that specifically mean in context of GMP?
00:03:26.560 --> 00:03:27.039
Sure.
00:03:27.199 --> 00:03:34.560
Yeah, so um again, GMP is a very kind of all-encompassing big term.
00:03:35.520 --> 00:04:08.560
Um, and it involves things like operations and manufacturing and environmental control and quality control and um, you know, various assays and how you um obtain your raw materials, all of these things and many more um are required for good manufacturing practice because together they ensure not only regulatory compliance by itself, which is important, but also that you have continuous quality um oversight.
00:04:08.879 --> 00:04:32.000
And what that allows you to do is really to capture any flaws or any vulnerabilities in any part of your process and design around those vulnerabilities controls that essentially increase the odds, if not always guarantee, that quality throughout the life cycle is preserved.
00:04:32.240 --> 00:04:59.839
And if you imagine any phase, phase one, two, three of manufacturing or even post-market, um you have a manufacturing um operations facility, and then you also have your supply chain, and you have um, you know, your validation of all of your facilities and your equipment and your process, and you have your training for all qualified personnel and so on.
00:05:00.399 --> 00:05:20.000
And ultimately, you are making a therapy or or a drug um that is supposed to go into a patient and uh keep them safe and help them with whatever indication um it's about.
00:05:21.040 --> 00:05:50.319
Well, uh it's very important that during various phases of this whole process um there can be oversight about whether everything is done according to um approved written procedures, whether there have been any incidents or deviations from a set process, whether all quality control um assays have been correctly performed and quality attributes met.
00:05:50.800 --> 00:06:13.120
Um, all of those things guarantee at various points in a process that the therapy you are uh sending to the patient or putting up on the market is designed and produced according to the highest quality and safety standards that are enforced federally but also internationally.
00:06:13.839 --> 00:06:16.800
And as you can imagine, a lot goes into production, right?
00:06:16.879 --> 00:06:18.319
There's so much at stake.
00:06:18.560 --> 00:06:21.600
It's um very expensive very often.
00:06:21.759 --> 00:06:31.120
Um, a lot, again, a lot of infrastructure has gone into setting up a place that can, in fact, manufacture according to good manufacturing practices.
00:06:31.199 --> 00:06:35.600
Um, a lot of very highly qualified expert personnel is required.
00:06:35.839 --> 00:06:42.319
A lot of special conditions for um the production of the therapy are required.
00:06:42.480 --> 00:06:57.680
And so you want to be able to make your drug product or your therapy and put it out in the market so that it, you know, it's commercially available, that's related to profits, it's also related to patient well-being, and so on.
00:06:57.839 --> 00:07:05.439
And so the most important component I think of of all of this is your quality unit.
00:07:05.680 --> 00:07:26.800
And the quality unit, as contemplated by the FDA, by other international organizations, is a unit that must be independent from the rest of the managerial aspect and operational aspects of a manufacturing operation.
00:07:27.040 --> 00:07:27.600
Why?
00:07:27.920 --> 00:07:44.959
Because you want to be able to have a highly qualified, skilled unit whose only concern is not to agree with management, it's not to push something out into the world that does not qualify.
00:07:45.120 --> 00:08:00.639
Um, their only concern has to be whether the data they're presented with, the product they're presented with, the process records, whether all of those are actually meeting pre-specified criteria.
00:08:00.879 --> 00:08:09.439
And then and only then to sign off and say, yes, this is a well-made compliant product.
00:08:09.680 --> 00:08:35.919
And when your quality unit is not truly independent, what happens is that because of pressures from maybe management or for whatever reason, your quality unit um has priorities other than singularly assessing, objectively assessing the quality and compliance of your um production.
00:08:36.320 --> 00:08:38.240
And that's very dangerous, right?
00:08:38.639 --> 00:08:48.399
And you never want to only on paper have a separate quality unit that does not depend on um the approval of management.
00:08:48.559 --> 00:09:02.320
You never want to only on paper have a quality unit that is not subject to pressures, or there that is not overridden, or in any other way non-authoritative with respect to the job they have.
00:09:02.559 --> 00:09:33.600
And so independence of your quality unit is about as important as anything else in your process, if not more, because that is the safeguard against um pushing a therapy or a product or a process or a system of processes that ultimately is acting not with the patient's best interest at heart, but perhaps with other motives of production.
00:09:33.919 --> 00:09:36.720
And this is why independence is absolutely imperative.
00:09:36.799 --> 00:09:41.279
And we've seen recently what happens when independence is doubted.
00:09:42.159 --> 00:09:43.600
Yeah, thank you, Tatiana.
00:09:43.679 --> 00:09:57.840
Yeah, I mean, I completely agree with this concept of kind of uh separation of church church and state, if you will, of like quality should remain independent, and uh you certainly can't take the teeth, it needs to have its teeth, right?
00:09:57.919 --> 00:10:04.639
As a as a kind of governing waypoint or or uh lighthouse almost, if you will.
00:10:04.879 --> 00:10:12.240
Um, and I think that kind of ties into this this notion of uh a quality hierarchy, I think you described it when we when we spoke before.
00:10:12.399 --> 00:10:23.600
Um, do you see this kind of in in organization as like if we have an independent quality unit that's sitting at the top of this this pyramid almost of of a quality hierarchy?
00:10:23.679 --> 00:10:37.279
And would you mind explaining that a little further and and what happens when this hierarchy maybe isn't respected, or when you try to take the teeth out of your quality team and keep them from actually prioritizing product quality and thus patient safety?
00:10:37.600 --> 00:10:38.320
Yeah, absolutely.
00:10:38.639 --> 00:10:42.240
That's a very dangerous um thing to do.
00:10:42.399 --> 00:10:47.600
And you know, sometimes it happens uh without any malice or intent.
00:10:47.759 --> 00:10:56.720
Um, I certainly don't think that people aim to breach quality or to disable a quality unit.
00:10:56.960 --> 00:10:58.799
Nobody sets out to do that, right?
00:10:58.960 --> 00:10:59.279
Right.
00:10:59.440 --> 00:11:10.639
Um, but for various pressures, uh, maybe even knowledge gaps, sometimes authority and responsibility misalign, right?
00:11:10.720 --> 00:11:16.399
And so you have a quality unit and it's um apparently separate.
00:11:16.559 --> 00:11:39.759
However, um the quality unit's decision regarding any part of the process they're responsible for is overridden, or perhaps there's something more urgent, and so something is overlooked, or perhaps there is legitimate pressure from um upper management for a trillion possible reasons and so on.
00:11:40.000 --> 00:11:51.039
And there is also occasionally the misunderstanding of what the quality unit really should be relative to other functions in a manufacturing environment, right?
00:11:51.600 --> 00:12:04.720
Management um has a lot of responsibility, and ultimately uh they are accountable for everything that happens um in an organization that they're in charge of.
00:12:04.799 --> 00:12:21.840
Um, and quality is a part of that, but only in the sense that management has an obligation to um to run with the judgment of a good and independent quality unit.
00:12:22.159 --> 00:12:28.000
They are not authorized to um say, oh no, you know, ignore that.
00:12:28.159 --> 00:12:29.759
We're just gonna continue.
00:12:30.000 --> 00:12:43.679
Um they cannot um at any point in practice consider themselves to be um to have the final word when it comes to quality and quality attributes.
00:12:43.919 --> 00:12:48.720
And when that happens, um, the consequences are very serious.
00:12:48.879 --> 00:12:57.120
And they're serious not only because, you know, during an inspection, if that ever transpires, that's a very big problem.
00:12:57.679 --> 00:12:59.840
It's a problem of credibility.
00:13:00.080 --> 00:13:18.639
Um, it is much more than a singular violation because not only does it capture violation of some regulatory uh standard, but it also demonstrates that there is a great degree of subjectivity in the way quality oversight works at that organization, right?
00:13:19.360 --> 00:13:25.759
And um that's a very big red flag if you're the FDA, and rightfully so.
00:13:26.080 --> 00:13:54.480
And so when uh when that happens, um it's important for an organization, I think, to really uh craft a strategy, a response that doesn't just say something like, you know what, yes, we acknowledged that this or that violation was committed, and we're gonna rewrite our SOPs and we're gonna retrain our people.
00:13:54.799 --> 00:13:58.320
And we're this isn't this is grossly inadequate, right?
00:13:58.480 --> 00:14:17.120
Because it doesn't address what transpires to be a cultural problem, a problem of authority, a problem of management understanding where they need to step back, um, a problem of essentially having a quality unit that is dysfunctional or doesn't work at all.
00:14:17.440 --> 00:14:40.480
And so it's important that everybody understands why quality is independent, and that that independence means not only the ability and obligation to conduct independent analysis and issue decisions, but that those decisions are imperative to follow.
00:14:40.960 --> 00:14:54.080
And if that if that's broken, then the organization is broken and the manufacturing culture and quality culture are not working the way they're supposed to.
00:14:55.440 --> 00:14:57.279
Yeah, thanks, thanks, Tatiana.
00:14:57.360 --> 00:14:59.759
I I agree, and I think you make a great point there.
00:14:59.919 --> 00:15:19.279
Like nobody means to uh compromise quality and quality oversight in manufacturing, but it's it's kind of a slippery slope of when you let I don't want to say conflicting interests, because ultimately I think everybody would agree that patient safety and efficacious product is of the utmost importance.
00:15:19.440 --> 00:15:27.679
But you know, when you when you kind of let the noise creep in and distract from rigid quality oversight, I think it it can happen.
00:15:27.759 --> 00:15:34.960
And you alluded to this earlier, some incidents of serious quality violations we're seeing at some pretty well-established manufacturing sites.
00:15:35.039 --> 00:15:40.080
And I won't name any names, but I'm curious to hear you talk a little bit more about that.
00:15:40.240 --> 00:15:50.159
Um, first of all, would you say it is accurate that there has been an increase in these kinds of incidents of uh non-compliance at top manufacturing spots?
00:15:50.240 --> 00:15:52.240
And and are you surprised by this?
00:15:53.519 --> 00:15:56.639
Yeah, you know, we've all seen them.
00:15:56.799 --> 00:16:01.679
Um, some of them are pretty egregious um and and truly unfortunate.
00:16:01.759 --> 00:16:36.159
Um, and I think they speak exactly to that loss of credibility because of uh a culture of misunderstanding the way uh quality should work, um, and misaligning responsibility and authority when it comes to the quality unit, where you know um they are responsible for quality oversight, but de facto they're not being given perhaps the authority to exercise that responsibility correctly.
00:16:36.799 --> 00:16:41.759
And um I will say, yeah, some of some of these are very egregious.
00:16:41.840 --> 00:16:45.039
Um and would you say is the most egregious one that comes to mind?
00:16:45.200 --> 00:16:46.159
Not to interrupt.
00:16:46.639 --> 00:16:54.399
Oh, you know, there were there were several recent ones that were just to me, and I'm sure to too many, um, quite shocking.
00:16:54.559 --> 00:17:26.319
And they were not only about environmental findings, so contamination and um, you know, growth of mold and and all those terrible, terrible things that are hard to imagine for many who understand what GMP must be, but um then you know there's some the build on top of that, and then there was some um data integrity breaches, there was falsification of data, there was hiding or replacing documents that weren't previously there.
00:17:26.640 --> 00:17:40.480
Um there was um very obvious um gap in knowledge between regulatory practices, written procedures, and what was really happening.
00:17:40.799 --> 00:17:52.240
And ultimately there were some admissions that, in fact, there was some data substitution and and uh forging, forging of documents, and that that's about as serious as you can get, right?
00:17:52.559 --> 00:18:49.839
So, not only is it a quality breach in the manufacturing environment per se, with contamination and poor environmental controls and various such violations, which are serious enough on their own, but when you add to the knowledge of those breaches, um a very considerable and demonstrable um data integrity breach, now we're talking about not only a problem, but also a mentality or a culture in which knowingly we take that problem and we acknowledge that it's there and we don't take care of it, but we are also going to kind of mask that whole situation and um and try to um disguise it or be dishonest about it.
00:18:50.000 --> 00:18:51.839
That's very serious, right?
00:18:52.000 --> 00:19:09.359
So now the problem is that it's very difficult for anybody um who's an independent um investigator or observer or auditor, even, to really know when they're looking through records of an organization like that.
00:19:09.920 --> 00:19:15.680
And those records indicate that some part of a process um has gone really well.
00:19:16.000 --> 00:19:18.559
There's no violation, nothing's going on.
00:19:18.799 --> 00:19:20.640
That's no longer believable, right?
00:19:20.720 --> 00:19:36.240
It's very difficult to trust that the scope of very egregious violations is limited because there is no evidence of accountability and there's no evidence of um really understanding the scope of the problem.
00:19:36.400 --> 00:19:38.079
There's evidence to the contrary, right?
00:19:38.160 --> 00:19:47.839
There is evidence that um there's been a little bit of deception, perhaps, that you create data that wasn't there, that you correct data that was there.
00:19:48.079 --> 00:19:57.759
And once that happens, defining the scope of risk to patients is very difficult to uh do.
00:19:58.640 --> 00:20:18.880
And so, you know, that's when you see not only recommendations for, say, an independent um consultant to be hired to really assess everything because there's no belief that if management is given the task to assess the scope of the problem, they will do it, um, or they will do it without bias anyway.
00:20:19.039 --> 00:20:24.240
So that's when those recommendations for independent consultants come in.
00:20:24.559 --> 00:20:29.680
And um that's a signal that that's a signal that there's loss of credibility.
00:20:29.839 --> 00:20:40.960
Um, and I will say we also see more and more some of the responses that are published to such serious letters findings by the FDA.
00:20:41.119 --> 00:20:46.880
And the responses often demonstrate a little bit of lack of insight, right?
00:20:47.119 --> 00:21:08.400
So um they kind of respond with um uh to a specific localized problem that's been mentioned, or uh with band-aid treatments, kind of, oh well, we designed a new SOP and now we're gonna retrain everyone, or oh yeah, there is mold, we're gonna remove the mold.
00:21:08.880 --> 00:21:13.039
Those are inadequate, and and very often the FDA will say that, right?
00:21:13.200 --> 00:21:20.400
They will say um this, they they will describe and acknowledge the response, and then they will say it's inadequate and demonstrate why.
00:21:20.640 --> 00:21:24.720
And the why is because the scope is not acknowledged.
00:21:24.799 --> 00:21:28.400
So you can't just, you know, spot treat those things.
00:21:28.559 --> 00:21:39.440
You have to do a thorough retrospective analysis of the full scope, not where problems happened, but where they could have happened, right?
00:21:39.599 --> 00:21:42.079
So you need to really go deep.
00:21:42.400 --> 00:21:48.880
Anywhere where there presumably can be a risk, you need to um you really go and investigate.
00:21:49.200 --> 00:21:55.359
And you don't get to say, well, presumably there could be risk someplace else.
00:21:56.160 --> 00:21:59.680
Um, but we can confirm that there was.
00:22:00.400 --> 00:22:02.000
You're now in a very different place.
00:22:02.160 --> 00:22:11.920
Actually, the job is unless you can confirm with objective evidence that there wasn't risk, you have to assume there was.
00:22:12.240 --> 00:22:19.119
So, you know, it's just important in those responses, nobody's looking for a quick solution, that there isn't one.
00:22:19.279 --> 00:22:22.559
And I think inspectors understand that.
00:22:22.799 --> 00:22:31.440
Um, they're they are looking for accountability, they're looking for depth of analysis, they're looking for true remediation measures.
00:22:32.079 --> 00:22:44.960
And um, unless they see them, um, they see a reasonable plan that's not only risk-based, but it's also very well justified, um it's not going to be enough.
00:22:45.200 --> 00:22:57.920
So the first and most important thing when you have such um serious um problems with quality and data integrity, um, the most important thing is to demonstrate accountability.
00:22:58.160 --> 00:23:04.960
And sometimes what that means is causing the company or the organization significant discomfort, right?
00:23:05.039 --> 00:23:37.039
Because effectively what they have to say is we're gonna put the brakes, we're gonna stop operations until we have implemented a very thorough remediation because we want to demonstrate that until we can reliably produce high-quality therapeutics, we're not going to allow our production interests to jeopardize um people on the market who are simply seeking to buy therapies they believe are safe.
00:23:37.279 --> 00:23:39.759
And so um it's a big deal.
00:23:39.920 --> 00:23:45.599
Um, yes, they are very egregious, um, and we've all seen them.
00:23:45.680 --> 00:23:47.119
That that's not a secret.
00:23:47.359 --> 00:24:08.319
Um I hope, you know, no matter what, I'm I'm fairly young in this field, and I want to say that I hope it's not much worse than it's always been, but it feels like uh there are some pretty significant findings that uh leave us all a little bit surprised.
00:24:09.039 --> 00:24:10.480
Yeah, thank you, Tatiana.
00:24:10.799 --> 00:24:21.680
In any case, it's it's crazy that in in 2026 we're seeing these kind of uh infractions um amongst, like you know, you said established names.
00:24:21.839 --> 00:24:29.039
It's kind of it's worrisome to say the least for the state of of the industry as well as just for for patient safety.
00:24:29.119 --> 00:24:33.200
It's kind of it's insane to me that it it's even happening.
00:24:33.440 --> 00:24:47.759
But I'm curious if out of all these 483 letters and and violations, if is there a particular kind of common theme that you've seen across them or a specific type of infraction that seems the most common uh among among those you've seen?
00:24:48.559 --> 00:24:49.119
I think so.
00:24:49.279 --> 00:24:52.000
And and it's kind of twofold.
00:24:52.079 --> 00:25:06.559
Um, and uh someone might disagree with me and they're welcome to, but I think um a 483 uh letter is always quality related in some way.
00:25:06.799 --> 00:25:29.680
Um and uh there transpire to be uh two parts that I the two themes, I suppose, that keep recurring that I keep seeing um across these multiple letters uh of various seriousness, um, but there is a common thread, and it is one um a dysfunctional quality unit.
00:25:29.920 --> 00:25:31.680
You have to ask why, right?
00:25:31.839 --> 00:26:19.359
So it it the the answer is not gonna be the same for every place, but for some reason, whether it's contamination, whether it's quality control, whether it is um incomplete validation procedures or incomplete or inadequate SOPs, or it's maybe um, you know, data breaches or dependence um of the quality unit on um management, um, which is not ideal, it there is a lack of sufficient authority and preparedness in the quality unit, which means that the quality unit is not functioning effectively.
00:26:19.599 --> 00:26:22.400
And if it's not, that's it.
00:26:22.640 --> 00:26:24.400
It that really is it.
00:26:24.640 --> 00:26:32.079
If you have a non-functional quality unit, which you're responsible solely for the authorization of, right?
00:26:32.240 --> 00:26:48.000
If I'm management with respect to a quality unit, my job is to make sure that quality and everybody else knows that what quality says goes, not the other way around, right?
00:26:49.039 --> 00:26:55.279
And as soon as that's not true, problems will inevitably emerge.
00:26:55.599 --> 00:26:56.400
So that's one.
00:26:56.480 --> 00:27:04.640
And the other one is this um kind of perplexing dropping up the ball.
00:27:04.799 --> 00:27:22.240
Um, and we've seen that a little bit, we've seen that in Europe now, we've seen some of it in the United States, we've seen some in Asia, but this kind of uh facility level, very um obvious environmental breach, right?
00:27:22.480 --> 00:27:38.880
In which even if you have uh the best documentation and the best procedures otherwise, the environment in which you manufacture has become commonly unsafe.
00:27:40.000 --> 00:27:50.400
And uh there is evidence that some of these breaches in environmental control have been noted and they've been noted before, and then they've been noted again.
00:27:50.880 --> 00:28:14.160
And uh it's hard to fathom at each of those findings what the response was and whether there was a systemic response and whether there was any follow-through or follow-up with whether it was implemented completely and correctly and effectively, and whether there's a plan forward for you know continuous oversight.
00:28:14.640 --> 00:28:39.599
But it is kind of shocking that in this day and age, as you said, you can have multiple documented violations of basic safety in manufacturing, and yet they continue until you know there is a very serious notice, and perhaps now you really are going to have to stop operations, there's loss of credibility, there's loss of reputation.
00:28:39.839 --> 00:28:42.880
Um it's it's very serious.
00:28:43.279 --> 00:29:11.119
So between compromised quality authority and a kind of an inability to really give an in-depth strategic long-term tested response to violations, I think those two are uh they can be lethal, they can be uh damaging beyond repair to an organization, but also who suffers in the end, right?
00:29:11.279 --> 00:29:17.599
Um, it's it's always patients, it's always people who uh seek a therapeutic.
00:29:17.759 --> 00:29:25.440
And you know, I don't when you buy um anything from a pharmacy, do you ever think that it could be dangerous to you?
00:29:25.599 --> 00:29:26.799
Probably not, right?
00:29:26.880 --> 00:29:33.039
We all trust that medicines and therapies are gonna be good for us.
00:29:33.279 --> 00:29:47.440
And so when the manufacturer digresses from that practice or forgets that biggest of all responsibilities, I think that's um that's just a tragedy, really.
00:29:47.920 --> 00:29:49.279
Yeah, I completely agree.
00:29:49.440 --> 00:29:52.559
Especially, again, it's 2026 for A.
00:29:52.960 --> 00:29:54.559
That's insane in and of itself.
00:29:54.720 --> 00:30:09.039
But especially like I, you know, in my lifetime, I don't think I've ever seen public trust be so thin in institutional power and protection and just uh just institutions and science and medicine in general.
00:30:09.119 --> 00:30:22.240
And so, like, absolutely should there should not be any any actions going on at established manufacturing sites, no less, that could in any way damage the fabric of that already thin veil of trust.
00:30:22.400 --> 00:30:24.640
And um, it is a tragedy indeed, I think.
00:30:25.039 --> 00:30:27.119
I think that's a very important point you're making, right?
00:30:27.200 --> 00:30:44.480
Because one of the side damages, right, if we just zoom out away from the specific finding in a 483, um one of the horrible side effects of these uh violations is that ultimately that affects the entire field, right?
00:30:44.559 --> 00:31:00.000
It affects the the public trust, it affects belief that um our our practices and our processes are the way they should be, and that patient safety is really the biggest priority as it is and it should be.
00:31:00.240 --> 00:31:02.000
And that hurts everybody, right?
00:31:02.079 --> 00:31:05.839
That that hurts other manufacturers who are doing a good job.
00:31:06.079 --> 00:31:12.160
Um it hurts public trust, it hurts um patients the most.
00:31:12.480 --> 00:31:26.000
And so uh there are so many unfortunate consequences of that level of neglect for um good manufacturing practice and just compliance and safety.
00:31:26.160 --> 00:31:31.920
Um and and that's that's not a small side effect of those issues.
00:31:32.400 --> 00:31:34.480
No, yeah, I mean perception is everything, right?
00:31:34.559 --> 00:31:39.759
And it doesn't certainly isn't helping the perception that needs to be corrected in the first place.
00:31:40.559 --> 00:31:41.359
I agree.
00:31:41.599 --> 00:31:58.880
Um you know, we we've talked about kind of what what dominoes need to fall, so to speak, and to to allow for quality control to slip and um when when quality isn't held independently and at the top of the hierarchy.
00:31:59.039 --> 00:32:16.079
Um so we know we know kind of how this happens abstractly, but I'm curious to hear from you more on how do you you mentioned before the idea of quality by design, which isn't a new concept, but how do you think organizations should integrate this philosophy into their operations from the beginning?
00:32:16.319 --> 00:32:34.240
And more importantly, you know, in your experience, how do you find success actually upholding that that that idea of quality by design and coordinating myriad of teams and systems that are involved in biomanufacturing, especially I think cell therapies might correct me if I'm wrong.
00:32:34.319 --> 00:32:48.640
I feel like cell and gene therapies have the highest demand for product quality because they're just there's so much there's so much there that could potentially be uh ruinous to the yeah, there's a lot at stake, no doubt about it.
00:32:48.799 --> 00:32:52.799
Um and cell and gene therapy production is not a simple process.
00:32:52.960 --> 00:32:55.359
Um it's actually highly complicated.
00:32:55.519 --> 00:33:02.480
It's also occasionally variable depending on what specifically you're making and what indications that therapy is for.
00:33:02.559 --> 00:33:04.960
And so you're right, the stakes are very high.
00:33:05.119 --> 00:33:11.359
Um, and so quality by design becomes even more imperative if that's possible, right?
00:33:11.519 --> 00:33:14.799
And so just generally quality by design.
00:33:15.200 --> 00:33:30.720
Um, if you take the idea that once upon a time, when producing therapies or pharmaceuticals or drugs, um, you know, you you make them, make them, make them, and at the end you have a quality control test and it's pass or fail.
00:33:30.960 --> 00:33:34.319
Um, that's that's really not a very good system, right?
00:33:34.720 --> 00:33:51.759
Um, you are hoping that at the end uh you get a pass one way or another, but you're not really thoughtfully designing your process and your product with quality integrated already.
00:33:52.000 --> 00:33:55.519
And really, quality by design is intended to remedy that, right?
00:33:55.680 --> 00:34:05.920
So you have whatever your product is, um, you start with something we all know is called the quality target uh product profile.
00:34:06.079 --> 00:34:15.039
And so you kind of you just define what this what the critical quality attributes of your product need to be, right?
00:34:15.679 --> 00:34:53.599
How it gets to be effective, safe, what that means, what what are the the critical basically components or properties um that you want to be seeing in your product, and then you design the processes um and um import the materials for that production by defining then the critical material um attributes and properties, and then you go to um things like critical process parameters.
00:34:53.760 --> 00:35:23.519
So, what parts of the process must be very thoroughly controlled, well understood, um, and you you define a system in which you connect uh critical process parameters with critical material attributes, with um CQAs, so that you know that you're designing a production system specifically with the uh product quality um attributes in mind, right?
00:35:24.079 --> 00:35:34.159
So there are many things in GMP um that are very difficult to constantly control at the same level of resolution, right?
00:35:34.239 --> 00:35:37.760
There are million systems, there are million things that go on.
00:35:38.079 --> 00:36:07.199
And so um instead of trying to distribute all resources uniformly to those myriad things that could influence a process, it it is just more rational and more scientifically sound to sort of grade where the biggest risks are and really focus on those high risk and then lower risk and then lowest risk spaces.
00:36:07.360 --> 00:36:14.559
Um, and when you do that, um you you assess risk with respect to what you want in the end, right?
00:36:14.719 --> 00:36:22.960
To the critical quality attribute, your drug, and you focus resources there, which is sensible because those matter the most.
00:36:23.039 --> 00:36:27.199
It's economical, it is reasonable because it's very well justified.
00:36:27.280 --> 00:36:30.079
You can rationally demonstrate why those are important.
00:36:30.559 --> 00:36:50.719
And so you design a manufacturing operation which has taken into account and defined well what the settings must be so that you end up with the best and safest compliant product that you can make.
00:36:51.199 --> 00:36:52.719
That's quality by design.
00:36:52.880 --> 00:37:10.719
Basically, you integrate controls that enhance and make much more probable the most critical components in manufacturing within an umbrella of compliance and minimize risks that you want to, of course, always be minimizing.
00:37:11.199 --> 00:37:13.920
That process is always iterative.
00:37:14.079 --> 00:37:15.920
You will uh learn as you go.
00:37:16.079 --> 00:37:21.920
I I don't think that in GMP it's possible to come up with a final perfect system.
00:37:22.159 --> 00:37:25.760
There is always going to be a problem we didn't predict.
00:37:26.000 --> 00:37:44.320
Um, but what we can do is have a solid enough framework that can capture the problem, that can define it, that has a very clear path of how that problem will be handled from a quality and compliance perspective.
00:37:44.559 --> 00:37:52.639
Um, and that knowledge must be very, very clear and well understood by everyone on the floor, right?
00:37:52.800 --> 00:38:12.960
Whether it's your uh process optimization team or your management or your quality control people or your manufacturing operators, every one of them must have the answer ready to a question like what would happen if this deviation from your protocol happens?
00:38:14.159 --> 00:38:15.039
What would you do?
00:38:15.199 --> 00:38:16.400
When would you announce it?
00:38:16.559 --> 00:38:17.920
How would you escalate it?
00:38:18.079 --> 00:38:20.000
What would be the next step of control?
00:38:20.400 --> 00:38:21.039
And so on.
00:38:21.119 --> 00:38:21.920
Do you continue?
00:38:22.159 --> 00:38:24.480
Those questions are absolutely critical.
00:38:24.800 --> 00:38:34.960
And everyone in a manufacturing operation must know clearly and confidently what they would do, what they should do.
00:38:35.199 --> 00:38:36.800
That's good training.
00:38:37.280 --> 00:38:52.880
So when things slip, is where you start to see um a little bit of uncertainty, or or you know, people who can't remember when they were last trained, or they're not sure if they want to say the thing they think is right because there is a manager next to them.
00:38:53.039 --> 00:38:54.960
That's where problems start to happen.
00:38:55.199 --> 00:39:02.159
So quality by design is exactly what everybody should be doing.
00:39:02.320 --> 00:39:12.639
It's a very rational framework, it's a very resource economical framework, and it's a safety optimization framework.
00:39:12.880 --> 00:39:14.559
So we have to have it.
00:39:14.639 --> 00:39:35.280
Um, and I think it's important to really spend the time up front before a single phase or as multiple phases are being hypothesized or planned for, and spend sufficient time on determining all of those components of quality by design.
00:39:35.519 --> 00:39:41.039
Um, you know, sit down and imagine all the crazy problems and terrible things that can happen.
00:39:41.360 --> 00:40:14.320
Be imaginative about everything that can go wrong, and then rank them and assess them and understand how they contribute to the various elements of quality by design, be those your um critical quality attributes of the drug or your materials um or process, whatever those are, define them, define them again, risk rank them, understand where you're vulnerable, design controls and test those controls, right?
00:40:14.880 --> 00:40:24.800
Improve them, and you do that over several iterations, and now you really have a system that works in the direction of optimizing quality.
00:40:25.440 --> 00:40:36.639
And the rest is what people will choose to do, in the sense that you do ultimately have to have uh an empowered quality unit.
00:40:36.880 --> 00:40:46.079
You do have to have management with humility to say there are problems and quality found them, and so now we're gonna follow whatever they say.
00:40:46.480 --> 00:41:06.880
Um it's a lot, and there's a lot at stake, and there's a lot that um is invested in continuous operations, and uh it's certainly understandable why disruptions are sometimes unpleasant, sometimes terribly destructive, but uh quality first, right?
00:41:07.119 --> 00:41:23.920
And if that's not the mana of any production operation, sooner or later, problems will begin to emerge more systemically, and um then it would be a little too late.
00:41:24.239 --> 00:41:44.880
So it's it's worth an investment on the front end to design processes, to design practices, and to empower the people who are supposed to carry them out um in the direction of optimizing state and quality so that patients really can get what they need safer and sooner.
00:41:45.599 --> 00:41:45.920
Yeah.
00:41:46.159 --> 00:41:47.360
Thank you, Tatiana.
00:41:47.519 --> 00:41:48.400
I I agree.
00:41:48.559 --> 00:41:51.280
I appreciate your your thoughts and wisdom there.
00:41:51.519 --> 00:41:55.119
Um and I want to ask, and I want to I want to be respectful of your time.
00:41:55.199 --> 00:42:05.199
I just want to ask if there's any single piece of advice or one idea you'd like to impart to our audience, if there's one thing that you want them to take away from this conversation, what would that be?
00:42:06.639 --> 00:42:13.199
It would really be to, you know, uh it always depends on who you're talking to.
00:42:13.280 --> 00:42:47.840
Um, but whether you're talking to management or you're talking to quality or you're talking to operations, I think it's really important to understand why um quality oversight is so important and what that means, and to put it in a larger framework of um certainly the self-interest of the company or production organization in terms of again, um, speed to market and reliability and patient trust and regulators' trust.
00:42:48.000 --> 00:43:07.840
But also just to remember that these are rules that are there for very good reason, and that a single slip into overriding the right decision, um, the single gesture of disempowering um your quality unit is a bad seed.
00:43:08.000 --> 00:43:19.840
Um, and it is the start of a convention which tells people around you, whether you've said it out loud or not, that the rules are going to be bent in certain ways.
00:43:20.000 --> 00:43:26.960
Um, and so what your team will understand is that oh, here are the priorities, right?
00:43:27.119 --> 00:43:35.039
Quality is supposed to be here, but actually it's second to this other priority that was used to override a quality decision.
00:43:35.280 --> 00:43:42.239
So just remember the people are on the line, uh, patients are on the line, and I don't think there is anything more important than that.
00:43:42.480 --> 00:43:45.280
Um, I think public trust is important.
00:43:45.679 --> 00:44:07.679
And so if it's it's entirely possible to uh restore faith in something as important as therbic production, um, if we really um agree that we will protect the principles that we have committed to um to uphold.
00:44:07.760 --> 00:44:10.719
And that may be a little idealistic, but I actually don't believe that.
00:44:10.880 --> 00:44:12.719
I think it's completely possible.
00:44:13.199 --> 00:44:13.360
Yeah.
00:44:13.519 --> 00:44:14.800
Thank you, Tatiana.
00:44:14.960 --> 00:44:18.000
That's a great, great piece of advice, and I completely agree.
00:44:18.239 --> 00:44:26.000
And now, lastly, I'm gonna ask you the question I ask every guest on the show, and that is big or small, Tatiana, how do you think we can better biopharma?
00:44:26.880 --> 00:44:29.679
Um, that's a wow, that's a big question.
00:44:29.840 --> 00:44:37.440
Um I think have more conversations like these and make them uh normal and common.
00:44:37.599 --> 00:44:45.440
Um, have more conversation in the direction of quality improvement, make that a mainstream mentality if it's not already.
00:44:45.679 --> 00:44:49.599
Um, and uh be truthful and transparent.
00:44:49.840 --> 00:44:54.719
So, you know, there are two things that can compromise what we do.
00:44:54.800 --> 00:45:13.440
And one is by loss of patient trust, and one is by um overriding um the roles of people whose job is to stop us even when it's um it's not what we want, right?
00:45:13.760 --> 00:45:35.199
And so have more conversations in the direction of enhancing quality, in the direction of integrity, in the direction of empowering junior people to carry out principles of quality when confronted with other demands, right?
00:45:35.440 --> 00:45:53.599
Um, I think that the most powerful thing is always to have people lower down in the production chain or hierarchy who know that they can say stop, they can say no to a process because they've noticed something and rightfully brought it up.
00:45:53.840 --> 00:45:58.960
Empower those people, empower people in your organization to be transparent.
00:45:59.280 --> 00:46:08.239
Um, thank them when they bring up quality um incidents or or any kind of process problem.
00:46:08.880 --> 00:46:14.400
Um and uh promote them when they do that because that's not easy to do.
00:46:15.039 --> 00:46:15.599
Yeah.
00:46:15.840 --> 00:46:17.119
That's great, Satiana.
00:46:17.199 --> 00:46:17.920
Thank you so much.
00:46:18.079 --> 00:46:26.559
Um thank you for being here, and thank you so much to our audience for tuning in for this episode of Better Bioforma, the official podcast of Bioprocess Online.
00:46:26.639 --> 00:46:29.039
I'm your host, Tyler Minachello, and I'll see you next time.
