Summary

Synthetic peptides, such as GLP-1 receptor agonists, are manufactured by solid-phase peptide synthesis (SPPS), a process that produces a characteristic family of related impurities: deletions, truncations, isomers and adducts. Detecting and quantifying these species, often at levels of 0.1% or below, is a core requirement of peptide drug characterization, and regulatory expectations are evolving. In July 2026, FDA published revised draft product-specific guidances (PSGs) for several generic peptide products, including liraglutide, and withdrew its May 2021 synthetic peptide guidance.¹ The revised draft liraglutide PSG recommends that product related impurities above 0.1% of the drug substance be reported and those above 0.5% be identified.² This application note describes how the Byos intact mass workflow was adapted into a dedicated workflow for synthetic peptide impurity analysis. Liraglutide and its impurities were analyzed by LC-UV-MS, and the data were processed in Byos to identify the main peak and multiple impurities. Impurities were quantified by UV peak purity down to 0.01% of the main peak, a level ten times lower than the 0.1% reporting threshold.

Key capabilities highlighted:

• Support for unnatural and modified amino acids through user-defined building blocks
• A clipping tool for identifying clipped and truncated impurities
• An MS2 reader for sequence confirmation by fragment annotation
• A customizable biopolymer model with user-defined elemental composition
• Flexible UV and/or MS (TIC) trace integration
• A preloaded report template for UV-MS peak purity

Introduction

Peptide therapeutics are increasingly complex. Many contain unnatural amino acids (UAAs), side-chain modifications such as lipidation, or unusual elements introduced for stability or potency. Liraglutide is a GLP-1 analogue acylated at a lysine side chain with a palmitoyl-glutamic acid moiety, which extends its half-life.

The impurity profile of such a molecule is typically assessed by LC-UV, with MS used for identification. Two practical challenges arise:

1. Software must understand the molecule. Standard protein-oriented tools assume the 20 natural amino acids and a protein-like elemental model, so synthetic residues or unusual peptides are poorly handled.
2. UV quantitation is complicated by co-elution. Multiple peptidoforms often elute within the same integrated UV peak, so UV area alone overstates the main species or misassigns the impurity levels.

The Byos synthetic peptide impurity workflow was built from the existing intact mass workflow to address both.

Workflow Features

The intact mass workflow already provides deconvolution, mass-based identification against a sequence list, support for MS2 fragmentation and integrated reporting. The following features were added or adapted for synthetic peptide impurity analysis.

1. Support for UAAs through the building blocks table

The building blocks table supports ASCII symbols as residue identifiers. The user assigns a symbol and enters the chemical formula of the new or modified amino acid. That symbol can then be used directly in the input sequence.

Figure 1. Modified Building Blocks Table. In addition to the 26 letters of the alphabet, 25 Ascii symbols have been added to accommodate a large number of UAAs. *Average mass values have been changed to monoisotopic.

This allows the following:

• Support for UAAs
• Chemically modified residues, including side-chain lipidation
• Residues with unusual elemental content

2. Customizable biopolymer model

Synthetic peptides can deviate from the “averagine” isotope model for two broad reasons: the elemental composition is unusual, or the isotopes themselves have been deliberately or accidentally altered.

The biopolymer model can be customized to the peptide being analyzed by entering the chemical formula of the synthetic peptide. This ensures the correct monoisotopic peak is labelled, even when the peptide contains unusual elements such as fluorine or selenium, whose isotope patterns differ from the typical C, H, N, O and S distribution assumed for natural proteins.

Because deconvolution and isotope assignment depend on a realistic isotope model, this step is important for accurate mass assignment of modified peptides.

Figure 2. Biopolymer model options. If a synthetic peptide contains unusual elements, the user can enter the chemical formula to ensure the correct monoisotopic mass is labelled.

3. Clipping tool for truncated impurities

The clipping tool enumerates and searches for clipped forms of the parent sequence (from either N or C termini of the peptide), so these impurities are identified without having to list every possible fragment manually.

Figure 3. Clipping tool. This tool can be configured to clip at a specific site or at every possible site from either the N or C terminus.

4. MS2 reader for sequence confirmation

The workflow includes an MS2 reader that annotates the MS2 fragment spectrum. When a candidate impurity is identified by intact mass, its MS2 spectrum can be annotated against the proposed sequence.

5. Flexible trace peak integration

The trace peak integration parameters allow the user to integrate the areas of UV and/or MS (TIC) traces. Impurity profiling can therefore be carried out regardless of the trace type, whether the laboratory's method of record is UV-based or MS-based.

6. Preloaded peak purity report template

The report template contains a preloaded template for calculating peak purity.

Relative quantitation of peptides and impurities is commonly measured by LC-UV trace peak integration. Often, multiple peptidoforms elute within the same integrated UV peak, complicating UV quantitation. The Byos UV-MS Report tab uses the deconvoluted MS intensity ratio to approximately quantitate the multiple peptides observed in the same UV peak. The result appears in the UV-MS column of the Report tab.

Experimental

Liraglutide (a synthetic GLP-1 analogue) and its impurities was analyzed by LC-UV-MSMS and the resulting raw file was processed using the custom Synthetic Peptide Impurity Analysis Workflow in Byos.

Figure 4. Structure of Liraglutide: A GLP-1 agonist consisting of 31 amino acids with a modified lysine residue containing a palmitoyl group (C16) plus glutamic acid.

Workflow Set-up

Step 1: The raw file was imported into the Samples table and the Traces table was used to integrate the UV trace (see Figure 5).

Step 2: In the Sequences and masses tab, the liraglutide sequence along with an in silico generated list of single-residue deletions, was supplied as a FASTA file. 

Step 3: The lysine residue was modified in the building blocks table to contain the side-chain modification (see Figure 6).

Step 4: The peptides were deconvoluted, with deconvolution parameters adapted for peptides (Figure 7) using appropriate mass and m/z ranges for a ~3800Da peptide.

Figure 5. Sample tab showing the trace peak integration options. Any associated trace (MS,UV or FLR) can be integrated using the options listed in the Traces table.

Figure 6. Building blocks table showing the modified lysine residue. The chemical formula of the modified (plus the glutamic acid andthe palmitoyl group was added in place of the original lysine formula.

Figure 7. Advanced Deconvolution Settings. The settings selected were optimized for peptide deconvolution and should be suitable for most synthetic peptides.

Results and Discussion

The Byos project consists of two parts 1) The Inspection View and 2) The Report.

1. Byos Inspection View

The Byos project contains a fully integrated UV chromatogram with automatic identification of the main peak and several impurities, isomers and adducts. Deletion impurities from the FASTA list, as well as truncated forms identified with the clipping tool, are matched by deconvoluted intact mass.

Figure 8. Byos Inspection view consists of: A) Project Table, showing all raw files and traces used in the project, B) Trace Peaks Table showing all integrated peaks, C) Trace Plot, displaying the integrated UV trace plot with editable integration boundaries, D) Masses Table, showing all deconvoluted masses, complete with identifications, for the selected trace peak, E) Annotated MS2 spectrum, F) MS1 plot displaying the summed raw data across the selected peak and G)Deconvoluted Mass Plot.

The Byos inspection view is used to view the results and validate the data. The user can click each peak sequentially and choose whether to unassign any of the deconvoluted masses based on evidence like charge state distribution and alternative match criteria.

MS2 fragment spectra were annotated with the MS2 reader to confirm the sequence  This is possible for both the sequence of the full product, deletions and truncations. Figure 9 shows the MS2 spectrum for the clipped species (3-31).

Figure 9. MS2 Spectrum Annotation. The intact workflow utilizes the MS2 spectra in a top-down manner and displays the labelled MS2 scans for the selected deconvoluted, matched peptide.

2. Byos Report

Every Byos workflow includes a pre-loaded, fit-for-purpose report template. The synthetic peptide workflow provides one built specifically for UV peak purity assessment. As with all Byos reports, the template is fully customizable to meet individual user requirements.

As is typical in impurity analysis, several peptidoforms were found to co-elute within single integrated UV peaks. To address this, the deconvoluted MS intensity ratio in the UV-MS column of the Report tab was used to apportion each UV peak area among the species present. Following validation, the UV-MS tab was used to filter impurities down to 0.1% of the main product, in line with the FDA reporting threshold.

Figure 9 shows a screenshot of the UV-MS reporting tab for peaks 1-5, including the MS ratio and the UV-MS %. Product-related impurities (including adducts) were quantified down to 0.01% of the main product. FDA guidance recommends that impurities above 0.5% be identified, and the workflow identified all impurities above this level except for one 3802 Da component (at 0.7%), which remained unidentified.

Figure 10. UV-MS tab of the UV Peak Purity Report. The MS intensity ratio is used to apportion UV peak area among co-eluting species allowing each impurity to be quantified by UV.

Table 1 is taken directly from the Byos report and was filtered to list all impurities (excluding adducts) over 0.5% of the main product.

Table 1. List of deconvoluted identified components exceeding 0.5% of the drug substance.

Conclusion

Revised FDA draft guidance recommends that active ingredient-related impurities in generic peptide products be reported above 0.1% and identified above 0.5% of the drug substance.² Adapting the Byos intact mass workflow for synthetic peptides makes it practical to meet these expectations for modified and unusual peptides in a single environment:

• Building blocks with elemental compositions make UAAs and modified residues, such as the lipidated lysine of liraglutide, to be fully integrated into the sequence.
• A customizable biopolymer model ensures the correct monoisotopic peak is assigned, including for peptides containing elements such as fluorine or selenium.
• The clipping tool and MS2 reader support identification and sequence confirmation of truncated impurities.
• Flexible UV and/or TIC integration and the preloaded UV-MS peak purity report resolve co-eluting species within UV peaks. This allows quantitation of impurities down to 0.01% of the main peak, ten times below the 0.1% reporting threshold.

The workflow also maps directly onto the tiered structure of the guidance. For unknown species (such as the 3802Da impurity) the Byos Peptide workflow, which contains the Byonic search engine, can be used for more in-depth MS2-level characterization. This strategy is a two-step approach. In step one, the intact mass workflow determines the levels of all impurities relative to the main peak, which supports reporting against the 0.1% threshold. In step two, the masses and retention times of impurities above a chosen threshold are carried forward into a targeted MS2 search using Byonic. Setting that threshold at the 0.5% identification level focuses MS2 characterization on the species that require it, while the intact mass results still document the lower-level impurities.

References

1. U.S. Food and Drug Administration. FDA Publishes Revised Draft Product-Specific Guidances for Certain Generic Peptide Products. July 28, 2026. https://www.fda.gov/drugs/drug-alerts-and-statements/fda-publishes-revised-draft-product-specific-guidances-certain-generic-peptide-products
2. U.S. Food and Drug Administration. Draft Guidance on Liraglutide (Recommended May 2024; Revised Dec 2025, Jul 2026). Product-Specific Guidance, NDA 206321. https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_206321.pdf

The Synthetic Peptide Impurity Profiling Workflow described here is based on the Intact module and is available to users licensing this module. If you are an existing customer and would like a copy of this workflow contact us at support@proteinmetrics.com.