15 - Process Validation: Ensuring Consistent Quality (S15E1)

Okay, so you know how we just like automatically
expect things to work, right? Like I grab my
allergy meds I take one, I assume the next one
is going to be just as effective. You buy a bottle
of ibuprofen and you kind of trust that each
pill will be just like the last one. Right. You
don't really think about the fact that there's
this whole world of manufacturing that goes into
ensuring that consistency. And it's not just
like luck. Right. It's actually a very systematic
and rigorous process. Yeah. And that's what we're
going to delve into today. Absolutely. For our
deep dive, we're going to unpack this whole world
of process. validation, which is basically how
manufacturers prove that their processes are
going to reliably pump out products that meet
all those quality standards. Exactly. It's all
about building in quality from the very beginning.
And we've got a bunch of sources, like we've
got some stuff from the FDA, some industry guidelines,
some real world examples. Yeah, and even some
historical data from past episodes. Right. Exactly.
So our mission today for you, our listener, is
to really grasp how these manufacturing processes
are constantly proven to deliver that quality
that we take for granted. We want to go beyond
just the what. We want to understand the how.
So we're not just going to skim the surface.
We're diving deep into the specifics. It's going
beyond just trusting that it works to really
understanding why it works. Exactly. So first
off, let's talk about perspective validation.
What exactly is that all about? So perspective
validation is basically taking that proactive
forward -thinking approach. You're trying to
establish documented evidence that a system,
a manufacturing system, is going to function
exactly as it's supposed to. Oh. And this is
all based on pre -planned testing before you
even start regular production. Got it. So you're
not like waiting for things to go wrong in the
real world. You're trying to catch them beforehand.
Exactly. It's like a really, really thorough
test run, kind of like a dress rehearsal, but
with a lot more scientific rigor. Yeah. OK, so
when would they typically use this prospective
validation approach? So this is typically done
when you have a brand new process, something
that's never been done before. Or if you have
an existing process, but you're making some major
changes to it. OK. Maybe you're changing a key
ingredient or upgrading a piece of equipment.
You need to make sure that this change isn't
going to mess things up. That makes sense. So
it's like proving it's going to work before you
go live with it. Precisely. and a really crucial
part of this whole perspective validation process
is the validation protocol. Okay, what exactly
is that? Think of the validation protocol as
like the instruction manual, the super detailed
instruction manual for the whole testing process.
Okay. It outlines every single step like really
meticulously. So what kind of stuff is in this
protocol? Well, for starters, it's going to define
the predefined acceptance criteria. Like what
does that mean? So these are the specific standards,
the benchmarks, that the process absolutely has
to meet in order to be considered validated.
OK, so it's like pass or fail. It's going to
hit these markers, or it's back to the drawing
board. Exactly. And then the protocol is also
going to specify the exact test methods that
they're going to use. OK, so you can't just make
up some tests on the fly. These have to be legit,
right? Right, they need to be. scientifically
valid, and there are a few key organizations
that provide these scientifically valid methods,
especially for pharmaceuticals, like the USP.
The USP. The United States Pharmacopeia. They're
like kind of the gold standard. They set those
standards for like drug quality, and they give
you scientifically sound ways to actually test
and see if a drug meets those standards. You've
also got AOAC International, which is more for
like food and beverages. Got it. They give you
those validated methods. And then you've got
the FCC, the Food Chemicals Codex for certain
ingredients. OK. And if you're worried about
bacteria, you've got the BAM. the Bacteriological
Analytical Manual. Oh, wow, okay. And even the
EPA, the Environmental Protection Agency, comes
into play if environmental testing is something
that's relevant to the product. And the EAS Consulting
Group video transcript specifically mentions
subpart J of 21 CFR 111, which talks about the
importance of these scientifically valid methods,
Section 320 and all that. Okay, so basically
you gotta make sure you're using the right tools
for the job. And the protocol spells all that
out. Exactly. And then once you have that plan
with all the success criteria and the right tests,
what happens next? Well, then you actually do
the test runs. Right. And during these runs,
you're collecting tons of data, documenting every
single thing to see if the process can consistently
produce products that meet those quality standards.
So it's like a real world simulation kind of.
Right. It's like, OK, let's see if this thing
actually works the way we think it does. And
you're looking for that consistency batch after
batch. And then you put all that data together
into a what a report. Exactly. A validation report.
Yeah. And this report summarizes all the findings
from those test runs. And most importantly, it
says, did this process pass or fail? OK. Did
it meet those predefined criteria that we set
out in the protocol? And if it passes, Boom.
You're good to go. You have documented evidence
that this new or changed process is actually
reliable. Awesome. So that's prospective validation,
right? Yeah. Proving it will work before you
even really start. But what happens if you've
got a process that's already up and running?
Right. Maybe you don't have all that fancy documentation
from the get -go. Yeah. Well, in that case, you
might use concurrent validation. Concurrent.
Concurrent validation is basically, you know,
you're validating While you're producing. Oh,
okay. So it's like a real -time validation. Exactly.
It's happening at the same time Yeah, and this
might be necessary if like you don't have enough
historical data to do a prospective validation.
Got it Maybe this process has been running for
a while, but no one really documented it properly.
So instead of those pre -planned test runs, you're
actually using your everyday production batches
to gather your validation data. OK. So you're
essentially killing two birds with one stone.
Exactly. You're making product and validating
at the same time. Smart. But how do you actually
make sure it's being done properly if you're
like in the thick of things? Well, that's where
in -process monitoring and testing come in. Okay.
And as we talked about in like season two, in
-process controls are those checks and balances
that happen while a product is being made. Right.
So you're not just waiting for the final product
to be done, you're constantly checking along
the way. So it's like built -in quality checks
at every stage. Exactly. And this is crucial
for concurrent validation. because you're using
those real world batches to prove that your process
is reliable. So are you like testing every single
batch then? Essentially, yes. And because of
that, documentation becomes even more critical.
Right. You need to keep meticulous records of
everything that's happening with each batch.
OK, so lots of paperwork. Yeah, it can be a bit
overwhelming. Yeah. But it's necessary because
you're building your validation evidence from
those actual production runs. I see. And what
happens to all that data, all that documentation
from those routine batches? Well, all that data
is then analyzed very, very carefully. OK. You're
looking for those trends and patterns to see
if the process is consistently meeting those
predefined quality attributes. So it's not just
about one good batch. It's got to be consistent
across the board. Exactly. You're looking for
that reliability over time, even under normal
operating conditions. It sounds like this concurrent
validation approach requires a really tight ship.
Yeah, it definitely does. You need a strong system
for process control and monitoring. You have
to trust that your systems are sensitive enough
to catch anything that might be going wrong.
Got it. So we've talked about proving a process
will work before it starts. perspective and proving
it works while it's running concurrent. But what
about those processes that have been around forever?
Maybe they were set up before all these fancy
validation methods were even a thing. Yeah, well,
that's where retrospective validation comes in.
Retrospective validation is all about looking
back at historical data. OK, so you're digging
into the archives. Exactly. You're looking at
all those old production records, testing results,
everything you can find to see if the process
has been reliable over time. Oh, interesting.
It's like building a case for the reliability
of a process based on its track record. So what
kind of historical records are we talking about
here? Well, batch records are really important.
Okay. The EAS consulting group video transcript
and our own season two transcript both mentioned
how crucial those are. They tell you exactly
what happened during each production run. You're
also looking at testing results, both from the
finished product and any in -process tests that
they did. Stability data is important, too, to
see how the product's quality holds up over time,
which we also discussed in season two. And I'm
guessing it's not just about finding the good
stuff. You also have to look for like the times
when things went wrong, right? Oh, absolutely.
You're looking for those complaint records, which
ICHQ914 mentions to see if customers have reported
any recurring issues. Right. And you're also
looking for deviation reports, which ICHQ914
and the Miltani Biotech video transcript both
talked about. OK. These tell you about any unplanned
events that happened during production and how
they handle those situations. So it's like a
full investigation into the history of the process.
Exactly. You want to get the complete picture.
And then you do like a trend analysis of all
that data. Right. Which both the season two and
EAS consulting group transcripts mentioned along
with subpart J, section 315 of 21 CFR 111 from
that EAS transcript, which emphasizes the importance
of this historical quality review. So for those
who might not be familiar, ICH is the International
Council for Harmonization. Right. They developed
those guidelines for pharmaceutical quality.
Got it. And with trend analysis, you're looking
for consistency. You want to see that the process
has been performing within those acceptable limits
over a long period. So it's not just about a
few good years. It's got to be a proven track
record. Exactly. But what if the data is all
over the place? Or what if there haven't been
many changes to the process so you don't have
many data points to compare? Well, that's a really
good point. Retrospective validation isn't always
appropriate. OK. If the data is messy or if there's
not enough of it, you can't really rely on this
method. Got it. So it's not a magic bullet. No,
it's not. You got to have that solid foundation
of data to make it work. OK, so we've talked
about the three main validation methods, perspective,
concurrent, and retrospective. Now let's get
into the nitty gritty of the documentation. OK.
We've mentioned validation protocols. And the
EAS consulting group video transcript really
stressed the importance of written procedure.
Yeah, absolutely crucial. So let's break it down.
What exactly goes into a validation protocol?
Well, first and foremost, you need to define
the objectives. OK, so what are you trying to
prove? Exactly. What's the goal of this whole
validation study? Then you need to describe the
process itself in detail. OK. What are the steps
involved? What equipment and materials are you
using? And that should also include equipment
qualification, DQ, IQ, OQ, and PQ. Right, which
we talked about earlier and which is also referenced
in Syed Nts Hader's book, as well as the EAS
Consulting Group and Season 2 transcripts. Right.
For those who might not remember, that's design
qualification, installation qualification, operational
qualification, and performance qualification.
Yep. So you got to make sure the equipment itself
is up to snuff. Exactly. And then the protocol
also needs to specify the test parameters. OK,
so like, what exactly are you measuring? Right.
And more importantly, what are those acceptance
criteria? What are the limits? And then going
back to those scientifically valid test methods,
you need to describe exactly how you're going
to measure those parameters. OK. And then there's
the sampling plan. OK, so how many samples are
you taking and from where? Right. And how often
are you taking them? And this might involve referencing
those established procedures like 21 CFR 211
.84, which talks about sampling testing for drug
products. So lots of details to iron out. Yeah
the protocol is all about being specific. Yeah
so you've got your objectives, your process description,
your equipment specs, your tests, your sampling
plan. What else goes in this protocol? Well,
you also need to describe the statistical methods
you'll be using to analyze the data. OK. So things
like statistical quality control criteria, which
21 CFR 211 .165 mentions. You also need to define
roles and responsibilities. So who's doing what?
Exactly. And then finally, you outline the format
for the validation report. So the protocol is
like the blueprint, and the report's the final
product. Right. So what goes into that validation
report? Well, first, you've got a summary of
the whole study. Then you present the detailed
results of all the tests that you ran Yeah, and
if there were any deviations from the protocol
you have to document those and explain what happened
Okay, so like if something went wrong or didn't
go as planned exactly and then you analyze all
the data Comparing it to those acceptance criteria
that we talked about. Okay, and then comes the
big conclusion Did the process pass or fail?
Yeah, is it validated and then you might also
include some recommendations for ongoing monitoring
Got it. So the protocol sets the stage, the tests
are performed, and the report delivers the verdict.
Exactly. Now, we've talked a lot about, like,
procedures and consistency, but what about anticipating
potential problems? Right. That's where risk
assessment comes in, right? Yeah, absolutely.
The Season 7 handbook and all those season materials
on quality risk management really hammered that
home. Right. So how does risk assessment fit
into this whole validation puzzle? Well, as the
Season 7 handbook puts it, Risk assessment is
all about asking that question, what could possibly
go wrong? You're identifying those potential
failure points in the process. And then once
you know what could go wrong, you can figure
out how critical each step is. Okay, so not all
steps are created equal in terms of risk. Exactly.
And by focusing on those high -risk areas, you
can really streamline your validation efforts.
Right, because you're putting your resources
where they matter most. The Season 2 transcript
also touched on that with those risk -based approaches
to equipment management. Exactly. And risk assessment
also helps you determine those acceptance criteria.
Okay. If a step is high -risk, you might need
to set tighter limits on what's considered acceptable.
Got it. And it also lets you figure out how much
sampling and testing you need to do. OK. High
risk areas might need more frequent checks or
more thorough testing. So are there like specific
tools or methods that they use for risk assessment?
Yeah, there are a bunch of different tools and
techniques, but one concept that's often used
is like hazard analysis and critical control
points or HACCP. HACCP, okay. Yeah, the PECB
video transcript mentioned that, mainly for food
safety, but the principles are pretty similar.
Okay. It's about systematically identifying those
potential hazards in the process and then putting
controls in place to minimize those risks. So
it's about being proactive, not just reactive.
Exactly. Okay, so you've done your risk assessment,
you've validated your process, are you done?
Oh no, not at all. Validation isn't a one -time
thing. You've got to keep an eye on things. So
there's ongoing monitoring? Absolutely. Continuous
monitoring is essential to make sure that process
stays validated over time. Okay, so what are
they monitoring? They're monitoring those key
process parameters. Like what? Things like temperature,
pressure, mixing times, flow rates, pH levels.
Okay, so anything that could affect the quality
of the final product. Exactly. And 21 CFR 110
.80 or B2 specifically talks about controlling
those factors in food manufacturing. So if they
see something out of whack, that's a red flag,
right? Yeah, definitely. That could mean you
need to revalidate the process. Revalidate, OK.
And there are other things that can trigger revalidation
too. Like what? Like if you change your raw materials
or components as 21 CFR 211 .84 outlines, you
might need to revalidate. OK. Or if you make
major changes to your equipment, like if you
do a big repair or even just move a piece of
equipment, the EAS consulting group video transcript
mentioned that. Right, because even a small change
in the environment could throw things off. Exactly.
Or if you change the manufacturing process itself.
OK. Or if you make changes to the facility or
the surrounding environment. So basically anything
that could affect the process. Right. and technological
upgrades, too. OK. Like if you implement a new
software system that ties back to 21 CFR Part
11 on electronic records and signatures, which
the EAS transcript talked about. And what about
like if they see something weird in the data?
Right. Like an unexpected trend. Absolutely.
The season two transcript mentioned that specifically
with data analysis under subpart of relevant
regulations, those unexpected trends or deviations
can also be a sign that you need to revalidate.
And there's a whole system for managing these
changes called engineering change control, which
Syed Mitiaz Haider's book discusses. So basically
revalidation is like a checkup to make sure everything's
still working as it should. Exactly. It's about
maintaining that validated state even as things
change around you. Okay, so that was a lot of
information, but I think we covered a lot of
ground. We talked about the three main validation
methods, the documentation involved, the role
of risk assessment, and the importance of ongoing
monitoring and revalidation. It really makes
you appreciate all the work that goes into making
sure the things we use every day are safe and
effective. It really does. It's a whole system
designed to build quality into every step of
the process. Exactly. So thinking about all this,
What do you think are the biggest challenges
for manufacturers trying to maintain these validated
processes, especially with all the crazy technological
advancements happening these days? It seems like
it'd be a constant battle to stay ahead of the
curve. It definitely is. I mean, technology is
evolving so quickly. Yeah. And that can make
it. tough to keep your processes up to date.
And then you've got all the regulatory requirements,
which can be complex and constantly changing.
And then of course you have to balance all of
that with the need to be efficient and cost effective.
So it's a delicate balancing act. Yeah, it definitely
sounds like a lot to juggle. Well, I think that's
a great place to wrap things up for today. Thanks
for joining us for this deep dive into the world
of process validation. It was a pleasure. And
for you, our listener, if you're interested in
learning more about specific validation methods
or the regulatory landscape, be sure to check
out the resources mentioned in this episode.
And as always, feel free to reach out with any
questions or comments. Thanks for listening.
Thanks.