(588ai) Optimization Approaches for Improved Peptide Synthesis Including Downstream Separation

Synthetic peptides are increasing in popularity in the pharmaceutical industry with many products hitting the market in recent years. With the increased demand for synthetic peptide drugs, an efficient way to manufacture them is needed. The primary method for manufacturing is Solid-Phase Peptide Synthesis, a highly complex process where the synthetic peptide is built by adding each amino acid in sequence. However, the resulting peptide chains that are formed are often riddled in impurities requiring difficult and time-consuming downstream separation processes.

We present our computational decision-making framework that aims to aid process scientists by identifying the best operating conditions for this complex manufacturing process. In our previous work¹, we developed a mechanistic model of the SPPS process via the postulated reaction mechanisms, and we utilized that model in the context of process simulation. Our model tracks all impurities that are formed to accurately calculate purity and yield at the end of the build. Our model also provides an estimate for the post-build total cost and throughput. We used this process simulation capability as the basis for our derivative-free optimization approach to determine the best operating conditions under objectives such as maximizing purity, minimizing cost, or maximizing throughput.

In this talk, we focus on our more recent work adding a chromatographic separation model into our computational framework. We use the impurity tracking result from our existing SPPS model as input to the multi-phase separation model, where we estimate the amount of each impurity that can be removed in addition to the time required, leading to post-separation purity, yield, cost, and throughput calculations. This advancement allows us to understand how different operating conditions for the SPPS process may lead to a more challenging separation, therefore incurring more costs to reach final purity standards. Overall, this updated framework allows us to study non-intuitive optimal reaction conditions that lead to a better separation process. For example, longer reaction time in each cycle may seem sub-optimal when solely optimizing the SPPS process based on the additional cost acquired; however, when combining the SPPS and separation models together, longer reaction time may result in separating fewer single-amino acid deletion impurities that dramatically reduces the cost and effort for separation such that it is an optimal operating condition for the whole process. Therefore, our updated results determine optimal operating conditions for the combined SPPS and separation process while giving process scientists insight into the effort required to reach final purity standards for synthetic peptides.

(1) Walsh, M. E.; McClary Groh, J.; Berglund, M.; Wang, J.; Agrawal, P.; Smith, C.; Viswanath, S.; Laird, C. D.; Gounaris, C. E. Simulation-Based Optimization of the Solid-Phase Peptide Synthesis Process. 2025, Forthcoming