78 – Process Optimization & QbD Principles (S6E3)

Okay, so picture this. You've got a product coming
off the manufacturing line, right? And it looks
absolutely perfect. It meets every single specification,
but then there's a tiny change. Maybe the raw
materials are a little different or someone tweaks
machine setting just slightly and boom, the next
batch is off. It just makes you think, is it
even possible to make a manufacturing process
so stable that it can handle those little hiccups,
those inevitable changes, and still pump out
high quality products every single time? You've
just hit on the fundamental question behind quality
by design, or QBD as it's commonly known. You
see, it's really a whole different way of thinking
about pharmaceutical manufacturing. The focus
is on building that inherent stability into the
process from the get -go, making sure that it's
not just reproducible under perfect conditions,
but truly robust in the face of real -world variability.
And that's exactly what we're going to unpack
today. It sounds like it's all about making the
process as bulletproof as possible. And for our
listeners who are already deep into the world
of drug development, we've got a ton of great
material to dive into. But for this particular
deep dive, we're zeroing in on the QBD aspects.
How do you actually create those rock solid,
dependable manufacturing processes that you're
talking about? Our mission today is to break
down the core principles of QBD. We'll explore
how manufacturers go about identifying those
critical process parameters, the ones that can
make or break product quality. And then we'll
look at risk -based design and how it moves from
being a theoretical concept to an actual practical
tool for designing processes that are way less
likely to veer off course. And to really bring
this to life, we'll be using real -world examples
from the OPRD literature. Okay, so let's jump
right in. What exactly does quality by design
mean? Is it more than just checking off a list
of good practices? That's a great question, and
you're right, it's a lot more than just a checklist.
QBD is really a systematic approach that starts
long before you even think about manufacturing.
You begin with very clearly defined objectives.
What are we trying to achieve with this drug,
and what level of quality is non -negotiable
for patient safety? So you start with the end
in mind. Exactly. From there, it's all about
gaining a deep scientific understanding of both
the drug product itself, particularly its critical
quality attributes or CQAs, and of course the
entire process used to manufacture it. The key
here is that quality is not something you test
for at the end. It's baked into every single
step right from the beginning based on sound
scientific principles and a very proactive assessment
of potential risks. So it's kind of like designing
a building to withstand an earthquake rather
than just hope hoping for the best and inspecting
it after the fact. I love that analogy. That's
precisely it. And this proactive way of thinking
is what leads to those robust and reproducible
processes we talked about. So now the question
is, why are these two things, robustness and
reproducibility, so crucial in pharmaceutical
manufacturing? Well, when we say a process is
robust, we're talking about its ability to withstand
those little variations that always pop up during
manufacturing. Slight changes in temperature,
maybe the raw materials from different suppliers
are slightly different, those kinds of things.
But the final product quality doesn't suffer.
and then reproducibility, that's about being
able to churn out the same high -quality product
time after time, batch after batch, regardless
of the scale of production or even where in the
world it's being manufactured. And for our listeners,
this is about understanding that a well -designed,
robust process is the foundation of consistent
medicine quality and supply. You nailed it. And
that consistency is what directly impacts patient
safety and how well the drug actually works.
So let's dig a little deeper into how manufacturers
figure out what those critical process parameters
or CPPs really are. What are they exactly and
how do you go about identifying the really important
ones? As far as I understand it, CPPs are the
key variables in the manufacturing process. The
ones that have a direct impact on the CQAs of
the final drug product. So things like the temperature
at a specific stage in the reaction, the mixing
speed, how long a drying step takes, those are
all potential CPPs, right? You got it. And understanding
how those CPPs are linked to the CQA, that's
like the holy grail of QBD, you absolutely need
to pinpoint which process factors and within
what range of operation have a significant effect
on the quality of the drug. That knowledge is
the bedrock of a well -controlled process. Sounds
pretty complex. So how do manufacturers actually
determine which parameters are the critical ones?
It can't be as simple as just listing out every
single step in variable, right? Oh, absolutely
not. It requires a systematic and very scientific
approach. Typically, identifying CPPs involves
a mix of existing knowledge about the process,
along with some good old -fashioned experimental
studies. Scientists might use techniques like
design of experiments or DOE to deliberately
change multiple process parameters at the same
time, and then they analyze the effects on the
CQAs. So they're deliberately introducing variability.
Exactly. This allows them to use statistical
analysis to determine which factors are the heavy
hitters, the ones that really influence the quality
attributes. It's all about building a data -driven
understanding of what's going on in the process.
And this is where it gets really interesting,
because one of the OPRD papers you shared, OPRD2012d2
.pdf, had a perfect example of this in action.
They were scaling up a reaction, and at a small
lab scale, everything was smooth sailing. Right.
But when they moved to a much larger 100 -liter
reactor, they ran into this gelling problem at
a certain stage, something they'd never encountered
before. That's a classic example of how scaling
up can suddenly reveal CPPs that you didn't even
know existed. In that particular case, it was
the longer time it took to reach and maintain
the target temperature in that bigger vessel,
along with potential differences and mixing at
that scale. Those turned out to be the culprits
behind the unexpected gelling issue. And I remember
they ended up with a decent yield for that batch,
but it was only through some manual intervention.
They had to basically babysit the process. Exactly,
and that highlighted a gap in their understanding
of the process. Even though they managed to save
that batch, the gelling was a clear sign that
they hadn't identified all the key factors that
controlled the process. It was a wake -up call
to investigate further and really understand
what was critical for preventing that gelling
from happening consistently, especially at those
larger scales. So even though they got a decent
yield that time, it wasn't a reliable or robust
process. It was a bit of a fluke. Precisely.
That unexpected event was what prompted them
to explore different ways to run the reaction,
like using continuous flow conditions. The idea
there was that the more efficient heat transfer
and mixing in a continuous flow reactor could
potentially solve the temperature control and
hold time issues that were causing the gelling
problem in the larger batches. This is a great
example of using process understanding to design
a more robust and reliable system. And this brings
us right to the concept of risk -based design,
doesn't it? Because that gelling problem they
had at scale, that was a real risk to their ability
to consistently manufacture the drug. Absolutely.
Risk -based design is a cornerstone of QBD. It
involves systematically identifying potential
sources of variability in your manufacturing
process and then figuring out how likely they
are to occur and how bad the impact would be
if they did. By conducting thorough risk assessments,
manufacturers can figure out which CPPs need
the tightest control strategies and the most
in -depth understanding during development. So
it's like a pre -mortem for the manufacturing
process. What could go wrong and how bad would
it be? You got it. And this proactive approach
is all about preventing problems before they
even have a chance to show up. And it makes a
lot of sense because in pharmaceuticals the stakes
are high. They are indeed. And this risk -based
approach is what ultimately leads to the establishment
of the design space. Think of the design space
as a scientifically defined safe zone for your
process. It lays out the acceptable ranges for
your critical input variables and parameters,
the ones that have been proven to consistently
deliver a high quality product. So it's like
having a set of boundaries for your process.
And if you stay within those boundaries, you
know you're going to be okay. Exactly. And here's
the really powerful thing about the design space.
If you operate within those established boundaries
during manufacturing, it's not considered a change
to the approved process. This gives me manufactures
a lot of flexibility and encourages them to continuously
improve their processes without having to constantly
go back for regulatory approval for every minor
tweak. That sounds like a huge win in terms of
efficiency and being able to adapt to new information
or improvements. It absolutely is. And going
back to that OPRD example we talked about earlier,
OPRD2012D2 .pdf, the whole experience with the
gelling issue at scale that directly informed
how they approached risk -based design for future
batches of that product. They learned the hard
way that temperature and hold time were much
riskier parameters at larger scales than they
had initially thought based on their small -scale
experiments. And that wasn't the only lesson.
than they learned from that OPRD paper. They
also ran into some trouble during the methyl
ester hydrolysis step. I think they were using
standard alkaline hydrolysis conditions and ended
up with some unwanted side reactions that affected
the purity of their product. That's right. The
hydrolysis of the methyl ester turned out to
be another critical step where the initial seemingly
straightforward approach didn't quite translate
well to the larger scale. They discovered that
those standard alkaline conditions were causing
ring opening and epimerization which led to the
formation of impurities that they didn't want.
This really highlighted the fact that the reaction
conditions things like the choice of base, the
temperature, and how long the reaction ran for
those were all critical process parameters that
had a big impact not only on the yield of the
product, but also on its purity. They ended up
having to investigate exactly who hit the nail
on the head. And it wasn't just about those main
reactions either. There were also challenges
during the scale up of the patent procedure specifically
related to the sulfonyl chloride hydrolysis.
It seems like the amount of water used as a regent
and the longer heating times needed at a larger
scale led to the formation of a sulfonic acid
byproduct. So even something as seemingly simple
as the amount of water became a critical factor.
That's right. The conditions that worked beautifully
at a smaller laboratory scale around 13 grams
didn't provide the same level of control when
they scaled the reaction up. They observed a
clear link between the increased amount of water
relative to the starting material at larger scales,
the longer heating times needed to get the reaction
to finish, and the formation of that unwanted
byproduct. This just goes to show that even parameters
you might think are well defined can behave differently
at different scales, which is why thorough investigation
and control throughout development are so important.
It really highlights the fact that what works
perfectly in a small flask in the lab doesn't
always translate smoothly to industrial scale
production. You often uncover new challenges
and realize that parameters you might have considered
insignificant can become absolutely critical
when you're dealing with much larger quantities.
Couldn't have said it better myself. And the
OPRD literature is a goldmine of these real -world
examples that really bring the concepts of QBD
to life. It's not just theoretical jargon. It's
about solving real manufacturing problems and
creating more reliable and efficient processes
that ultimately benefit patients. It definitely
makes it much more relatable and understandable.
Absolutely. And speaking of real -world examples,
there's a great one in the OPRD literature about
the production of an R alcohol described in OPRD2014
.bodypdf. When they were scaling up their solvent
extraction process, they ran into a major problem
with emulsions that just wouldn't separate. Emulsions,
those are the bane of so many chemists' existence.
I can only imagine the frustration. Tell me about
it. Anyway, through their investigations, they
started to suspect that residual protein from
the fermentation broth was acting as an emulsifying
agent. So to fix this and make their extraction
process more robust, they introduced an ultrafiltration
step to remove those proteins before extraction.
This is a perfect example of how an unexpected
problem during scale -up led to a deeper understanding
of what was going on and the implementation of
a targeted risk -mitigating solution that was
completely in line with QBD principles. So the
emotion formation wasn't even on their radar
as a major risk at the smaller scale. But when
they scaled up, it became a real threat to their
ability to consistently and efficiently extract
the product. And their solution, adding that
ultrafiltration step that was pure QBD thinking.
identify the problem, understand the root cause,
and implement a control strategy to minimize
the risk and improve the robustness of the whole
process. Exactly, and in that same OPRD paper,
2014g .pdf, they also talk about their work on
optimizing the catalyst loading and reaction
conditions for producing an optically active
intermediate. It wasn't just about getting the
desired chemical transformation to happen, it
was about fine -tuning the reaction parameters
to get the highest possible yield and the desired
optical purity and doing it in the most efficient
and reproducible way possible. That proactive
optimization of key parameters is another major
element of the QBD philosophy, designing for
optimal performance and consistent quality right
from the start. It's so helpful to see how these
real -world examples from the OTRD literature
really illustrate the practical side of QBD.
It shows that it's not just an abstract concept.
It's about tackling real -world manufacturing
challenges and building better, more reliable
processes that ultimately benefit patients. Absolutely.
And while we focused on these specific examples
from OPRD, it's important to remember that the
principles of QBD are tightly connected to the
broader regulatory landscape that governs pharmaceutical
development and manufacturing. So it's not just
a nice to have. It's woven into the fabric of
how drugs are developed and regulated. Exactly.
For instance, the emphasis on thoroughly understanding
and controlling your process parameters. and
how they impact product quality that aligns perfectly
with the regulatory requirements for having well
-defined specifications and robust analytical
methods. It's all outlined in guidelines like
21 CFR 314 .50D3. You simply can't ensure quality
without being able to accurately measure and
control those critical attributes. And that proactive
mindset of preventing contamination and ensuring
consistent product quality through robust process
design that's completely in line with the principles
of Good Manufacturing Practice or GMP, which
are detailed in regulations like 21 CFR Part
211. A well -designed process that's informed
by QBD is the foundation of a clean and controlled
manufacturing environment, which, as our excipient
development source also points out, is essential
for preventing contamination in facilities. Exactly.
And the fundamental principle behind it all,
understanding and controlling process variables.
So you consistently produce a product that meets
its predetermined quality attributes that's also
at the heart of process validation, which is
a critical aspect discussed in the Pharmaceutical
Master Validation Plan source. You need to be
able to prove through data and scientific understanding
that your designed process consistently performs
as intended under different operating conditions.
QBD provides the scientific basis for a successful
validation program. So to sum it all up, we've
really explored the essence of quality by design.
It's about intentionally building quality into
every step of the manufacturing process, from
the initial design to the final product. It involves
deeply understanding the product and how it's
made rigorously identifying those critical process
parameters and using a risk -based approach to
design a manufacturing system that's both robust
and consistently reproducible. It goes beyond
just checking off boxes. It's a fundamental commitment
to quality in every aspect of the process. I
completely agree. And by grasping these principles
our listeners gain a much deeper understanding
of what goes into developing and manufacturing
pharmaceutical products. They can now appreciate
that ensuring the quality of a medicine is about
so much more than just the final test results.
It's embedded in the very DNA of the manufacturing
process. It really makes you think about the
future of pharmaceutical manufacturing. With
increasingly complex drug formulations and constantly
evolving technologies, how will the principles
of QBD adapt and evolve? How can we leverage
advancements like continuous manufacturing, which
we touched on earlier, and the incredible potential
of advanced analytical techniques to take quality
by design to the next level and ensure that everyone
has access to high -quality medicines for years
to come? It's a fascinating and critically important
area to consider. I couldn't agree more. It's
an exciting time to be involved in this field.
Well, on that note, thank you so much for joining
us for this deep dive into quality by design.
It's been a pleasure exploring these concepts
with you. And to our listeners, we hope this
is giving you a deeper understanding of the complexities
and the incredible importance of ensuring quality
in pharmaceutical manufacturing. It's been a
pleasure being here. Thank you for having me.
Until next time, stay curious and keep exploring.