Chemistry Reference
In-Depth Information
7.3 TRANSLATIONAL DEVELOPMENT AFTER DISCOVERY
7.3.1 Solid Form Development: Preparing Stable Dispersions
By the time a molecule formally transitions from the discovery to the development phase,
the compound in question has passed animal toxicology studies and shown exposure
in vivo
. The next step is to shape it into at least a provisional solid formulation that will be
not only soluble
—
a parameter that has already been optimized for
—
but also manufac-
turable, dosable, and stably stored downstream.
As noted above, at the end of Section 7.2.2, the recrystallization kinetics of most
SDDs tend to require that some additional steps be taken to stabilize them. Analogously
to the use of polymers in the suspension formulation, certain polymers, when added to
the spray drying solution and thereby incorporated into the dispersion, stabilize it and
markedly slow the transition to the crystalline state in the dispersion and, eventually, in
the solid dosage form. Thus, the next step in this stage of SDD formulation development
is to de
ne a design space for these polymers and other excipients (such as surfactants to
improve the wetting of the material). Since API is often in short supply this early in
development, it is infeasible to empirically test each polymer with the drug. Instead, we
consider which polymers are theoretically expected to mix favorably with the API in
question, such that we can test promising polymers at the bench. Our polymer mixing
calculations are grounded in statistical mechanics, speci
Huggins theory
developed in the 1940s and often used since that time to estimate the free energy of
mixing between solids. Per this theory,
cally the Flory
-
Δ
G
mixing
RT
n
1
ln
Φ
1
n
2
ln
Φ
2
n
1
Φ
2
χ
12
;
χ
12
is the dimensionless interaction parameter, which is a guide to which polymers
are miscible with the desired API at the desired ratio (e.g., 1:1 API:polymer). To deter-
mine the interaction parameter, we use an
a priori
method in which we compare the
solubility parameter of the API with that of various polymers and choose the polymers
that are most closely matched.
Once these polymers are selected, we prepare blends of amorphous drug and
polymer at slightly larger scale. To that end, we again use small-scale spray dryers
where
—
a
“
model that requires a
few hundred mg (but tends to give poor yield, around 50%). Solvent
mini
”
model that requires about 1 g of the material and a
“
nano
”
film casting is
another option. This approach has the bene
it of being automated. However, it entails a
greater risk of recrystallizing the dispersion by mistake, and for a given sample it is
slower than spray drying.
To assess how readily the blended dispersions prepared here will lend themselves, in
practice, to a solid dosage form, focus naturally returns to their solid-state properties.
These properties, which had been temporarily set on the back burner during suspension
formulation development, may affect an SDD
'
s manufacturability, dosability, or
storability.
