PHARMACEUTICALS specifications and ensure the material can be reliably controlled.
However, if any of these impurities affect the impurity profile of the API, the compound is not a suitable starting material, due to the impact on product quality. In most cases, commercially available chemicals used in synthesis do not require separate justification as starting materials.
However, if one of these chemicals introduces impurities or structural features that influence API quality, it must be treated like an RSM. This means its route of synthesis must be defined and its specifications justified based on fate and purge data that show how impurities behave during the process.
Chemistry: Optimising reactions & parameters
A reliable manufacturing process depends on the consistency and robustness of each chemical transformation step. Understanding how each reaction behaves under real manufacturing conditions is essential to maintaining quality and reproducibility.
A good starting point for assessing process dependability is a review of previous batch records. These records provide insight into how the process has performed in practice and can help identify any variability, inefficiencies or recurring issues.
Ideally, these batches will be those used for the initial registration filing, as these will have been performed using the synthetic route planned for future manufacture and on a similar scale, at the site planned for commercial manufacturing. If issues are found, this evaluation may lead to additional development work, such as refining reaction conditions to reduce impurity formation or adding purification steps to strengthen process control.
Onyx carries out process optimisation for commercial manufacturing using a Design of Experiments( DoE) approach. This structured method systematically evaluates multiple process factors, such as temperature, solvent volume and reagent quantity, to understand their effect on product quality and process performance. By doing so, DoE identifies the CPPs that have the greatest influence on reaction success.
From this analysis, a proven acceptable range is established for each key parameter. These ranges define the safe operating window within which the desired CQAs of the product can be consistently achieved. Operating within this defined space allows for minor process adjustments without risking product quality or compliance.
Work-up & isolation: Managing process risks
While DoE typically focuses on optimisation of the reaction conditions, the reaction itself is only a small part of the overall manufacturing process. The work-up and isolation procedures, where intermediates and the final API are separated, purified and recovered, are also critical to controlling product quality. Therefore, it is important to identify any additional critical parameters in the entire manufacturing process.
To do this, Onyx uses failure modes and effects analysis( FMEA), which examines the impact of different types of failure( e. g., equipment failure, human error) on pre-defined quality characteristics. Every step of the process is evaluated to identify where failures could occur and how severe their consequences might be. Each potential failure mode is scored according to three key criteria:
• Severity: The impact of the failure on product quality / yield or on the overall safety of the process
• Probability: The likelihood of the failure occurring
• Detection: How readily the failure is identified These scores are then combined to give the overall risk level for this failure mode. Table 1 gives an example.
As many failure modes may not have been encountered previously in the manufacturing process, accurately assessing the severity may require additional experiments to be performed to confirm the outcome. If the risk level is sufficiently high, additional measures or controls will be put in place in order to reduce the risk to an acceptable level.
By systematically identifying and addressing risks, the FMEA approach enhances control across the manufacturing process. This leads to greater process robustness, reproducibility and confidence that every batch will meet the same quality standards.
Table 1 – Example of FMEA
MAR / APR 2026 SPECCHEMONLINE. COM
21