109 – Post-Marketing Surveillance (S8E4)

We all want to believe that when a drug gets
approved, it's safe and effective, right? I mean,
you see the commercials, you hear about the breakthroughs,
and you figure all the hard work is done. Yeah,
you'd think so, wouldn't you? But the reality
is, a drug's journey doesn't really end when
it hits the market. It's like, you know, you
think of something like thalidomide, initially
prescribed for morning sickness, but then...
Devastating birth defects. Exactly. And it just...
really highlights that getting a drug approved
is just the beginning of truly understanding
its impact. It's like the first chapter. It really
is. And that's where post marketing surveillance
or PMS comes into play. A PMS. Yeah. It might
sound a little odd, but think of it as the ongoing
story of a drug, you know, once it's out in the
real world. Right. Clinical trials are crucial,
of course. They give us so much valuable data,
but they occur in these, you know, very controlled
environments, often with specific groups of patients.
PMS is where we see how a drug really behaves
in the wild, so to speak. How in the real world.
Yeah, with a much broader and more diverse population.
People with all sorts of different health conditions,
taking various other medications. All the complexities
of real life. Exactly. So for this deep dive,
we're not going back to square one, you know,
the whole approval process. No. We're focusing
on what happens after a drug. gets that green
light, that stamp of approval. We've got some
fascinating information here about how drugs
are tracked in the real world. And our mission
today is to understand how those systems and
methods actually work to monitor safety and effectiveness
for you, the listener. Because once a drug is
available, that monitoring is crucial. It is,
and I think it comes down to recognizing that
the initial testing phase, the clinical trials
have their limitations. They do. They're incredibly
important, but... But they can only tell us so
much, right? Exactly. For starters, the number
of people in these trials is a fraction of the
potential users once a drug is on the market.
You could have millions of people using it. Millions,
yeah, compared to a few thousand maybe in a trial.
And the participants in a trial are usually selected
very carefully based on certain criteria. So
they might not represent the full range of patients
who will eventually be using the drug. It makes
sense. Age, ethnicity, other existing health
conditions, all those factors could be different
in the broader population. Yeah. and they could
all influence how a drug affects someone. And
not to mention the other medications people might
be taking. Oh, that's a huge factor. So many
potential interactions that you just can't fully
predict in a controlled trial. It's like test
driving a car on a perfectly smooth track. Yeah.
It's only when you take it out on real roads
with potholes and traffic that you see how it
truly performs. That's a great analogy. And thinking
about it, it's not just about the human population.
Even those preclinical studies, the ones done
before human testing. Right. Often using animal
models. Right. Like they often use what are called
mouse xenographs, which are basically models
where human tumors are implanted in mice. To
see if a drug can shrink them, for example. Exactly.
But I remember reading that these models have
their limitations, too. They do. One of the biggest
issues is that the mice used often have suppressed
immune systems. Oh, right. And that's to prevent
them from rejecting the human tumor. Right, right.
But it also means they're not fully representative
of how a drug might interact with a normal functioning
human immune system. That's a big deal, especially
for cancer treatments, right? Because so many
new treatments are targeting the immune system.
Absolutely. Immunotherapy relies on a healthy
immune response. And you also have to consider
the tumor environment itself. The environment.
Yeah, like the surrounding cells and tissues.
In these mouse models, it can be quite different
from the naturally occurring environment in a
human body. Oh, I see. And that can impact things
like angiogenesis, which is the formation of
new blood vessels, a key target for many cancer
drugs. So what looks really promising in the
lab? based on these models. Might not translate
perfectly to the much more complex reality of
human biology. Right. And then there's the whole
issue of rare side effects. Oh yeah, those can
be tricky to catch in trials. If something only
happens in, say, one out of 10 ,000 people. You're
unlikely to see it in a trial with a few hundred
or even a few thousand participants. Exactly.
So you could have a drug out there being used
widely. And then a rare side effect starts popping
up that just wasn't detected in the initial testing.
And even if the trials are large? There's still
the time factor. Some problems might take months,
years even, to develop. Right. Well beyond the
time frame of most clinical trials, PMS provides
that long -term perspective, which is so important.
It really is. So we've established the why behind
PMS. The why? Yeah, why it's so crucial. Now
let's delve into the how. To how? Yeah, the systems
and methods that make this ongoing safety check
possible. Okay, so we know why we need to keep
tabs on drugs after approval. How do we actually
do that? There are several key systems in place,
one of the most important being drug registries.
Drug registries? Yeah. Essentially, they're organized
databases that collect specific information about
patients taking particular medications. So it's
like a dedicated record -keeping system for each
drug? In a way, yeah. What kind of information
do they actually collect? Is it just noting whether
someone experienced a side effect? Oh, it's much
more detailed than that. They typically gather
a whole range of data. Patient demographics,
for example. Things like age and gender. Exactly.
Then you have their diagnosis. That's a specific
condition they're being treated for. And of course,
the treatment detail. The drug name. The dosage.
Yeah. How long they've been taking it. And crucially,
the outcomes. How well the drug is working and
whether they experienced any adverse events.
So it's about building a really comprehensive
picture of how a drug is performing across different
types of patients. That's the idea. That's got
to be incredibly helpful in identifying patterns,
right? Absolutely. By systematically collecting
all this data, registries can help pinpoint trends
in a drug's safety and effectiveness. For instance,
a specific side effect might be more common in
older patients or in those with certain pre -existing
conditions. It allows you to see those nuances.
Yeah, and that can inform prescribing practices
and help tailor treatments more effectively.
And I imagine there's a whole other wealth of
information in those existing health care databases.
You're right. Those are massive. Things like
electronic health records, insurance claims data.
It's like a gold mine of real -world patient
data. It really is. And analyzing these large
databases is another really important aspect
of post -marketing surveillance. But how do you
even begin to find potential drug -related issues
in all that data? It must be like looking for
a needle in a haystack. It can be, but that's
where sophisticated data analysis techniques
come in. Data analysis. Yeah. Researchers can
apply statistical methods and computational tools
to these datasets to search for associations
between a drug's use and various health outcomes.
I see. The scale of data available in these systems
is immense, so it allows for the detection of
even very rare signals that might have been missed
in smaller studies. It's almost like having this
massive net that can catch even the smallest
hints of a problem. That's a good way to think
about it. So we've talked about these big data
sources, registries, healthcare databases. How
does the information actually get into these
systems in the first place? Several ways. One
of the most basic being spontaneous reporting.
Spontaneous reporting. Yeah. Essentially, healthcare
professionals and even patients themselves can
report any suspected adverse events. Directly.
Yeah. They can submit reports to regulatory agencies
like the FDA. So if a doctor notices a patient
experiencing a strange side effect after starting
a new medication, they can just file a report.
Exactly. And that system helps capture a wide
range of real -world experiences. It can be an
early warning system for potential safety issues.
That makes sense. But it's all based on people
noticing and reporting, right? So there's a chance
some things get missed. There is. That's one
of the limitations of spontaneous reporting.
You're relying on people to voluntarily submit
these reports. And there's no guarantee everyone
will. Right. And even when a report is submitted,
proving a direct causal link between the drug
and the reported event can be tricky. Right.
Correlation doesn't equal causation. Exactly.
Just because something happens after taking a
drug doesn't automatically mean the drug caused
it. So what are some more proactive methods for
gathering data? Well, that's where active surveillance
systems come in. Active surveillance? Yeah. They
involve more deliberate efforts to collect data
on drug safety and effectiveness. OK. For example,
conducting targeted studies specifically designed
to look for certain side effects or surveying
patients who are taking a specific drug. So instead
of waiting for reports to trickle in, researchers
are actively seeking out the information. Got
it. And what about this idea of data mining and
signal detection? Those are really important
too. They involve using statistical and computational
methods to analyze those large data sets we talked
about, the healthcare databases, the spontaneous
reports. The goal is to identify potential safety
signals. Safety signals? Yeah, things like unexpected
patterns or a higher than expected rate of certain
events in patients taking a specific drug. It's
like using algorithms to sift through mountains
of data and flag anything that looks statistically
unusual. It sounds incredibly powerful. It is.
And it allows for a much more comprehensive and
nuanced analysis than relying on individual case
reports alone. So you can pick up on subtle trends
that might otherwise go unnoticed. Exactly. Now,
I also remember hearing about risk evaluation
and mitigation strategies or... RIMs. RIMs, yeah.
Those are specific programs that regulatory agencies
can require for certain drugs. For all drugs?
No, not all of them. Just the ones that have
known or potential risks. It's like an extra
layer of safety monitoring. What kinds of things
do these strategies involve? They can involve
a whole range of things, depending on the specific
drug and its associated risks. For instance,
requiring special training for the doctors who
are going to prescribe the drug. So they're fully
informed about the potential issues. Yeah. Or
restricting the drugs used to certain health
care settings where patients can be monitored
more closely. OK. In some cases, a REMES program
might mandate the creation of a patient registry
specifically for that drug. So it's a more focused
and intensive approach for drugs with a higher
risk profile. Exactly. It's about managing those
risks as effectively as possible. So we've got
all these systems and methods for gathering data
and analyzing it. What's the role of the regulatory
agencies like the FDA in this whole process?
They're absolutely central. The FDA and other
agencies around the world are constantly monitoring
all this post -market data. From all those sources
we've been talking about. Yep. The registries,
the databases, the spontaneous reports. And they
have the authority to take action if they identify
safety concerns. Action, like what can they actually
do? They have a few options, depending on the
severity of the concern. They can issue safety
warnings to health care professionals and the
public. Like an alert about a potential issue.
Exactly. They can also require changes to the
drugs label. So including new safety information
or limitations on its use. Right, they can even
restrict a drug's use to certain patient populations.
Like only allowing it for people who meet specific
criteria. Yeah, maybe limiting it to those who
haven't responded to other treatments, for instance.
And in the most serious cases where the risks
outweigh the benefits, they can actually withdraw
a drug from the market entirely. So they can
pull it off the shelves. They can. It's not a
decision they take lightly, but it's an important
safeguard. It really emphasizes that this whole
approval process isn't a one -time thing. It's
not. It's an ongoing evaluation. They're constantly
reassessing whether the benefits of a drug still
outweigh the risks as new data comes in. Exactly.
They have to stay on top of the situation to
ensure patient safety. And it's not just drugs.
Even after a medical device gets cleared or approved,
the FDA can request additional clinical data.
Yeah, if concerns come up later on. It's a good
reminder that there's this continuous scrutiny
of medical products. even after they're out in
the market. Definitely. And it highlights the
importance of staying informed as a patient.
This all makes me think about drug interactions,
something we touched on earlier. Oh, that's a
huge area of concern in post -marketing surveillance.
Because it's hard to predict all the potential
interactions in those controlled clinical trial
settings. Right, especially when you have people
taking multiple medications. And that's common,
especially as people get older. It is. They might
be taking medication for high blood pressure,
cholesterol, diabetes, arthritis. It all adds
up. And any one of those drugs could potentially
interact with a new medication in a way that
wasn't anticipated. Do you have an example of
a drug interaction that really came to light
after a drug was already on the market? Oh, absolutely.
There's a classic case involving certain HIV
protease inhibitors like ritonavir and an antifungal
medication called ketoconazole. Okay. After these
drugs were being used in clinical practice, it
was discovered that ketoconazole is a very potent
inhibitor of CYP3A4. CYP3A4. It's a liver enzyme
that plays a key role in metabolizing, breaking
down many different drugs. Got it. And it turns
out that HIV protease inhibitors are metabolized
by this enzyme. Okay, I see where this is going.
So if someone's taking ketoconazole at the same
time as a protease inhibitor? The ketoconazole
can inhibit the CYP3A4 enzyme. which can lead
to much higher levels of the protease inhibitor
in the bloodstream. Because it's not being broken
down as efficiently. Exactly. And that increased
concentration can increase the risk of side effects.
This particular interaction wasn't fully understood
until these drugs were being used together more
widely. It really highlights why this ongoing
monitoring is so crucial. It is. It's not just
about looking at a single drug in isolation.
It's about understanding how it behaves in the
context of all the other things going on in a
patient's body. The complexities of real life.
Exactly. Now, we've been hearing a lot about
artificial intelligence in healthcare these days.
While our research for this deep dive didn't
focus specifically on AIs, role in PMS, it does
make you wonder how it might be used in the future.
It's a really interesting question. We're already
seeing AI being used to analyze large data sets
in various areas of health care. Yeah, and it
seems like a natural fit for PMS with all this
data to sift through. It does. Imagine AI algorithms
scanning those massive healthcare databases,
the spontaneous reporting systems, looking for
subtle patterns and potential safety signals.
It could potentially identify issues much faster
and more efficiently than humans could. That's
the idea. It could be a real game changer in
terms of drug safety. It's pretty exciting to
think about. It is. It's a rapidly evolving field,
so it'll be interesting to see how AI shapes
the future of post -marketing surveillance. Well,
this has been a fascinating look into a part
of the drug development process that doesn't
always get a lot of attention. It's often overlooked,
but it's absolutely critical. Post -marketing
surveillance is this continuous dynamic process.
Relying on a whole network of systems and methods.
Registries, databases, spontaneous reports, targeted
studies, all these things working together. And
at the heart of it all are those regulatory agencies,
vigilantly monitoring the data and ready to take
action to protect public health. It's reassuring
to know that the system is in place. It is. It's
all designed to make sure that the medications
we all rely on are as safe and effective as possible.
And it makes you think, with all this data being
generated every day, what other innovative ways
might we find to use this information to make
medicines even better? It's an exciting area
of exploration for sure. The potential is enormous.
It really is. And that's something for all of
us to keep in mind, that the quest for better,
safer medicines is an ongoing journey. It definitely
is. We're always learning, always striving to
improve. And that's a good thing for all of us.
Absolutely. Gives me hope for the future of health
care. Me too.