The peptide drug market is expected to generate $50B in revenue for companies but the FDA is concerned about the number of impurities that may be introduced in the synthetic process. Advances in chemical peptide synthesis have allowed the production of synthetic peptide drugs to become easier and more cost effective. However, the peptide manufacturing process can result in synthesis-related impurities that can introduce immunogenic epitopes within the amino acid sequence of the peptide resulting in an unexpected and undesired immune response against the drug. For instance, during the process of solid phase synthesis, impurities may arise as the result of incomplete coupling and side reactions. Several classes of impurities related to manufacturing and degradation pathways exist and include: amino acid insertions and deletions, truncations, isomerization, and side chain modifications. Any of these impurities can affect the safety and efficacy of these drugs and it is therefore important to assess any impurities that may be present after synthesis.
T cell- (thymus-) dependent (Td) responses play a critical role in the development of antibody responses to biologic therapeutics. Td responses, by definition, are contingent upon T cell recognition of therapeutic peptide-derived T cell epitopes through the basic processes of antigen processing and presentation. A fundamental requirement for peptides to be considered as T cell epitopes is that they bind to human leukocyte antigen (HLA) molecules. The HLA-peptide interaction involves binding of specific amino acid side-chains to pockets in the HLA molecule binding ‘groove’, an interaction that is well-characterized for many HLA. Briefly, the requirements for HLA binding can be codified in ‘motifs’ or patterns. Based on these characterizations, pattern-matching algorithms, such as the EpiMatrix algorithm, have been developed to screen amino acid sequences for peptides that will bind HLA. EpiMatrix can be used to screen both the drug API sequence and its known peptide-related impurities. When peptide-related impurities are unknown, modifications at each amino acid position in the peptide can be performed in silico, their immunogenicity risk can be predicted, and they can be assigned an impurity risk score. The “What if Machine” (WhIM), performs all possible changes to the natural amino acid sequence of the drug substance and measures their impact on the epitope content of the peptide.
Here we present a retrospective case study of Taspoglutide – a GLP-1 agonist that was under investigation for the treatment of type 2 diabetes, but development was halted during phase III clinical trials due to serious hypersensitivity reactions and GI intolerance. It is suspected that the cause of the observed hypersensitivity is due to the presence of amino acid duplication synthesis side product(s) which gives rise to novel T cell epitopes. HLA typing in allergic patients shows an enrichment of five particular HLA DRB1 alleles. Two of these alleles (DRB1*0701 and DRB1*1104) were shown to be able to bind the impurity rather than the drug. Using the WhIM algorithm, we have evaluated all possible amino acid duplication impurities for the presence of new T cell epitopes at both a population level and an individualized level. Out of these possible impurities, five create putative T cell neo-epitopes for the HLA- DR7 and/or DR11 families, and one or more of these could be contributing to the observed hypersensitivity to Taspoglutide in subjects with DRB1*0701 and/or DRB1*1104.