(600e) Purity Control Strategy in Crystallization Based on in-Silico Classification of Impurity Retention Mechanisms | AIChE

(600e) Purity Control Strategy in Crystallization Based on in-Silico Classification of Impurity Retention Mechanisms

Authors 

Linehan, B. - Presenter, Boehringer Ingelheim Pharmaceuticals Inc.
One of the most important requirements for the manufacture and release of pharmaceutical
drugs is impurity control. Purification of the active pharmaceutical ingredient (API) is primarily
achieved through crystallization. Impurities found in the final API can be introduced during each
step of the synthesis and may consist of unreacted starting materials, degradation products,
side products, inorganics, metals, residual solvent, etc. Some of these impurities may exhibit
both acute and chronic toxic effects and must be controlled at very low levels, e.g., mutagenic
impurities. To ensure the safety of potential patients, and to produce a therapy that can be
implemented over an extended period, these types of impurities need to be effectively
rejected.
Given the importance of impurity rejection, it is extremely valuable to understand the impurity
retention mechanisms present in each step of a synthesis process, especially steps in which
crystallization is used to isolate and purify intermediate species or the final API. Recent
literature has postulated that the formation of solid solutions of impurity with API and
coprecipitating impurities are the most prevalent impurity retention mechanisms in
crystallization1. Both of these mechanisms can be modeled using appropriate in-silico tools.
This presentation offers an in-silico strategy, using commercially available molecular modeling
and solubility prediction to assess the risk of retention for known and potential impurities in
crystallization. For this specific model, Materials Studio software is used to incorporate the
impurities to be evaluated into a supercell generated from the crystal structure of the target
API or intermediate. After incorporation of the impurity, followed by MD calculation and lattice
energy determination, the difference between the inter-potential energy of the system before
and after impurity incorporation is calculated. This number, described in the presentation as
Disruption Energy (DE), is then compared with physical descriptors, like solvus or normalized
solubility increase, generated from characterization of known API-Impurity solid solutions. This
model is then used to generate a retention risk score (RRS) for each impurity. After an RRS for
each potential and known impurity is generated, this information will be used to inform process
development for each step of the synthesis and will contribute to the overall impurity control
strategy, which adds a proactive component to the typically reactive approach to impurity
control in API manufacturing.
1. Nordstrom, Sirota, Hartmanshenn, Kwok, Paolello, Li, Abeyta, Bramante, Madrigal, Behre, Capellades, Org. Proc.
Res. & Dev., 2023