|The following presentations were taped and edited by the International Association of Physicians in AIDS Care. Some of the questions were not included because they were not audible on the taped recording, or were not pertinent to the topics of the update.|
Norvir Community Update - 10/15/98
Stephen Byrn, PhD
Abbott Laboratories, Abbott Park, Illinois
Charles Jorden Professor
Head, Industrial and Physical Pharmacy Department
West Lafayette, IN
Additional Comments by:
John Bauer, PhD
BYRN: I am going to discuss crystals to give you a better understanding of
polymorphism. But first, I want to emphasize a point that Dr. Sun made a few minutes
ago - that once a polymorph molecule is dissolved in solution, it forgets what it was.
Regardless of whether it was a Form I molecule or a Form II molecule before it
dissolved, once that molecule is dissolved, it has the same pharmacology as other forms of the same molecule.
Science cannot predict when crystallization will occur. All science can tell us that it is possible for crystallization to occur. We do not know why crystallization occurs, and we do not know how it occurs. If we could unlock those scientific secrets, we might be able to prevent crystallization.
Crystallization is a word with different connotations or meanings, depending on how we
use the term. We can use the word crystallization in its broadest connotation to describe
the process by which matter is transformed from a non-crystal substance into a crystal.
We can also use the word in a more narrow connotation to describe a scientific analysis of a
substance to determine how it was transformed into crystals.
A polymorph is the same substance in a different crystalline form or structure. Graphite
and diamond are good examples of polymorphs. Both graphite and diamond are carbon,
but they are carbon with entirely different physical properties. The graphite form of
carbon is a relatively soft substance that can be easily shaped and used in the
manufacture of pencils. The diamond form of carbon is remarkably hard and more
difficult to cut when it is shaped into jewels or industrial drills. This analogy is somewhat
comparable to Form I and Form II ritonavir. Form I ritonavir is much more soluble than
Form II. Therefore, it is easier to formulate and easier to make into a drug. Form II is less
soluble, and more difficult to formulate and make into a drug. That is essentially the
problem that Abbott is currently facing.
I am passing around a
set of three polymorphs - red, orange, and yellow. They are exactly the same molecule,
but they are three different solid forms. These are my favorite examples of polymorphs
because they visually demonstrate the huge difference in physical properties that different
polymorphs can have. Here we have examples of three very different color properties of
the same substance.
Abbott has conducted crystallization studies on ritonavir to try to
understand what caused the agent's polymorphism. Form I ritonavir was the only
known solid form of ritonavir ever seen by scientists until sometime in May or June of
1998. Form II (the crystal form of ritonavir) was found by going through the normal
crystallization prestudies that chemists use to routinely test drugs for quality control.
Chemists use a number of techniques in conducting crystallization studies Here is a slide demonstrating a crystallization technology done by slow cooling. You can identify crystals by their
different shapes. These are aspirin crystals. Here is a set of steroid crystals. When you
find these different shaped crystals, you can conduct an analysis of these crystals to
understand their structure and, very importantly, their solubility. Solubility is the key
problem with Form II ritonavir.
Here is a slide of five different polymorphs of the same drug. All five of these crystals
are different. Each of these polymorphs have different solubility and different physical
properties. These are all different forms of the same drug. Each of these polymorphs
have different physical properties - but they are all the same drug.
The different physical properties of polymorphs may help us
understand why chemists conduct crystallization studies. The first physical property is
solubility. That is the physical property of Form II that is causing the current problem.
The main physical property difference is that Form I ritonavir is about twice as soluble as
Form II ritonavir.
A second physical property difference is chemical reactivity. That is
not a problem that we have with Form II, but that is a problem that occurs. Solubility
and chemical reactivity differences in polymorphs are very important from a regulatory
perspective since they can affect the pharmacology of a drug. Because of these different
physical properties, polymorphs of the same drug can be patented.
QUESTION: When you say that Form I is twice as soluble as Form II, does that imply that if you could take an extra pill or two, you might make up for the difference in solubility?
BYRN: That was a general approximation. I am not certain that I could accurately
comment on the pharmacology of ritonavir. It might be theoretically possible to
counteract the solubility problem with additional drug, but only if the degree of
crystallization was uniform in each batch - which it isn't. Without data on a uniform
degree of crystallization, one cannot predict the amount of compensatory drug to take..
QUESTION: If there is a little bit of crystallization present, does that mean that the
whole dose is ineffective?
BYRN: The amount that was in solution would be effective. The amount that is
crystallized would be ineffective. The problem is that no one can tell in a given dosage
unit how much is Form I and how much is Form II.
QUESTION: Can't you look at the liquid in the bottle?
BYRN: If you see crystals in the liquid you can warm it, shake it, and try to redissolve
them. But in the capsules, we can't see that.
BAUER: There is a significant difference in the amount of crystallization between the
liquid and the semisolid. In the liquid, the crystallization has been so minimal that it
would not affect the drug's efficacy. Even if you had crystals in the liquid. it probably
would be 90 to 95 percent effective. The amount of crystallization in the semisolid
formula was significant and would affect the drug's efficacy. This is why we won't be
using that particular formulation anymore. What we are trying to do in the liquid
formulation is to keep it a temperature where crystallization cannot occur. But even if you got crystallization, it wouldn't be extensive, and what you had would still be a full formulation.
QUESTION: Well, doesn't this relate back to the previous question that if your
temperature is not between 68 and 77 degrees Fahrenheit and you do observe a little
crystallization at 80 or at 62, it may not be relevant?
BAUER: It may not be relevant. What Eugene Sun pointed out was that, because of
the regulatory guidelines by which we have to abide, a pharmaceutical manufacturer
cannot make a statement about a drug's efficacy without backing up that statement with
appropriate data. We have to demonstrate to the FDA that this is the temperature range
at which the drug should be stored, and then produce the data upon which we have
based those recommendations. And that's what we have to put on our label.
QUESTION: Does the risk of crystallization in the capsules increase over time?
BYRN: There is some scientific speculation that the risk would grow with time. But this is a very difficult question to answer. What we are talking about scientifically is the rate of nucleation and crystal growth, which is difficult to predict.
QUESTION : Why is there such a narrow range of temperatures for ideal storage?
BAUER: We had to balance two concerns. We want the drug stored at a higher temperature to reduce the risk of crystal formation. But at higher temperatures, the stability of the drug is going to be shorter. The upper temperature limit on the label was to prevent chemical degradation. The upper temperature limit doesn't have anything to do with the crystallization.
BYRN: Just to restate, the higher the temperature, the more likely you are but
going to have chemical degradation, the less likely you have the crystals. The lower the
temperature, the more likely you have crystals, but less likely that you will have
chemical degradation. So it's a balance.
QUESTION:. With the liquid, you can see if any crystallization is in there. But with the
capsules, you can't.
QUESTION: From what you explained, I can't understand whether crystallization
could have occurred in the: existing capsules that some people may have stockpiled?
BAUER: All the batches that we have released from which any existing stockpiles
that anybody has accumulated anywhere in the world, were all produced prior to the
formation of the Form II crystal. We know that from our analytical results. We know
that it' is good product. The FDA and EMEA know this is good product. And if it
wasn't, we would have pulled it off the market.
QUESTION: So you can guarantee that Form II can't form in existing manufactured
BAUER: I can only tell you that the dissolution profiling analytical results from the product that you have in capsules is acceptable at the bioavailable level, for the way it's tested, from the specs that have been reviewed by the FDA and EMEA.
QUESTION: I believe you. But you had said that
over time, the crystallization can occur already been formed. So, if you're keeping a six month supply, isn't there a risk of crystallization in the capsules?
BAUER: There's another caveat to that. If we could let Steve go ahead and finish
his primer on crystallization, it might help. He has explained that there is a whole variety
of different forms that seem to make crystals. But a substance cannot make a new form
unless it has a model - unless there's something that shows it how to get into that form.
Those are called seed-crystals. We had no seed-crystals in the world for Form II at the time
that those drugs were made. So there' was nothing to show that ritonavir how to make
BAUER: That's why we are saying that with the ritonavir that was on the market
before Form II showed up, there's a very, very, very strong reason to believe that it can
never form Form II.
Different polymorphs can have different solubility because of their different structure. the But
here we are looking at two different polymorphs. There are six on this slide. Four of
them show solubility of around nine. One of them goes up to a high solubility of about
14, but then drops back down to crystallization. And actually, in this particular system,
what is happening is the molecules dissolving, reaching a high concentration and then the
crystallization event occurs. Then it drops down to the lower solubility of these other
forms. So here we have a type, but this is from the literature of a depiction of the
different solubility that different polymorphs can have. And this is in effect, this is not
ritonavir, but this is in effect what can happen with different crystal forms.
I am going to discuss patents on crystal forms about this particular patent here. This
patent is a very interesting one. This drug is beclomethasone diproprionate. It's in
metered dose inhalers for asthma. And Glaxo was making metered dose inhalers and
beclomethasone diproprionate in the 1980s and they noticed crystal growth inside
those inhalers. The crystals got large and when they delivered them to patients, they
were so large that they didn't go into the lungs properly. So they started investigating
the problem and what they found out was that the propellant- the fluorocarbon was
crystallizing with the steroid to form a new crystal form, a solvate. And they were smart
enough to realize, well, that has its own properties. If we made that by itself, then we
could avoid this crystal growth. And so they did and the got a patent and this has been a
very valuable advance, both medically and patent-wise.
I'm going to talk a little bit about ranitidine, because ranitidine is somewhat- is more
similar to the ritonavir than to beclomethasone - because you formed a new compound.
It incorporated the halocarbon end of the crystal. That did not happen in ritonavir.
Glaxo was developing ranitidine. Now, ranitidine's the largest-selling compound in the-
was the largest-selling drug in the world over the years. They made 800 kilograms of
Form I, a large amount of Form I ranitidine.
In a specific batch named 3B13, a specific production batch, a new polymorph showed
up. And from then on, it was impossible for Glaxo to make Form I pure. They could no
longer go back to the original situation. In effect, the world changed when that batch
was made. And that's what we think happened with Ritonavir is the world changed once
Form II showed up and now it's impossible to make Form I.
QUESTION: Is this Zantac?
BYRN: Yes. Ranitidine is Zantac. Glaxo switched from Form I to Form II.
They fixed the problem. Form I and Form II were equivalent in the bioequivalence
study. So they were able to switch from Form I to Form II. Ranitidine hydrochloride is
highly water soluble and that's one of the things that helped them make that switch.
They were able to get a patent on Form II and that patent was litigated and ruled valid.
BYRN:What happened in the ranitidine case is that the
generic companies figured out how to make Form I. So what the patent- at first the
patent was thought to be worth $21 billion because the patent on Form II was seven
years offset from the original patent on Form I. But generic companies figured out how
to make Form I. And so when you go to the store, if you buy Zantac, that's Form II. If
you buy generic ranitidine, that's Form I. These, though, are like I said, highly water
soluble, so it's a different situation. It's a different situation from ritonavir. The point here is that in any drug, a new form can
show up at any time. The luck of Glaxo was that it was bioequivalent. The bad luck for
Abbott was that it wasn't bioequivalent.
What the generic companies realized as a result of this experience was that seeding in
large quantities would allow them to control the process better. So they made a
discovery. They didn't patent it, but they made a discovery that a lot of people,
including myself, wouldn't have believed. But they realized that if they seeded heavily,
they could make Form I reproducibly.
One other comment. If we start talking about Form I and Form II ranitidine - their
solubility is very close. It's nothing like a factor of two. Anything that has a difference
of two solubility is a huge energy difference. And these are much more similar in energy.
QUESTION: Could you explain the seeding business?
BYRN: Seeding is hard to explain. But let's go back to the candy-making
analogy. In making candy, if you poured out your candy and sprinkled sugar on that
candy, it would be much more likely to crystallize than if you kept it inert and didn't
expose it to a model as John called it, to allow it to crystallize. And that is what we
think happens with Form II, because it is so much more stable and also because it is
water insoluble, it seeds. It's seeding crystallizations all over the place. And so as time
goes on, it was observed that there is more and more seeding occurring in a
QUESTION: How do we know there's not going to be Form III?
BYRN: Unfortunately, we don't know that. There is nothing to suggest that a
Form II ritonavir is more likely that a Form II of any other protease inhibitor. But if there
is a Form III, we will do the same thing we are doing now, except that we will be better
quipped to handle it.
QUESTION: Is there any reason to believe that something in the chemistry of ritonavir,
makes it more amenable to polymorphism?
BYRN: No. Abbott made every effort to find Form II and didn't early on. We
actually repeated searching for Form II in a seed-free environment, starting with Form I,
and were unable to find Form II. We have redone polymorph screens to see if there is
Form III out there. We have done everything we can scientifically to make sure that
there will never be a Form III. But science cannot guarantee that
QUESTION: What form goes into the SEC?
BAUER: The SEC is being designed so either form could go into it. We are trying
to do is design the SEC so that every molecule of ritonavir we put in is will still have
significant solubility so it won't make any difference.
QUESTION: Is Abbott also going to be changing or adjusting the liquid formulation to
make sure that you can have a wider zone of comfort in terms of both temperature and
your ability to assure people that there isn't crystallization?
QUESTION: Is polymorphism a problem with any drug?
BYRN: It's a potential problem in any drug.. But the likelihood of crystallization
is higher in soft gelatin capsules or soft elastic capsules than it is if you have a solid. For
example, those red, yellow, orange that I'm passing around, if I put those in a dosage
form where they were semisolid or in solution, the odds of them crystallizing or changing
from red to yellow are higher than if I keep them dry. This is because it's easier to get
one form to change to another in the dry state is called a solid-solid phase transformation.
And that is less likely. But ritonavir is not bioavailable in the solid state in any form.