Translating peptide evidence into immunotherapy targets: Orbitrap Tribrid Apex MultiOmics MS delivers high-confidence spectral annotation

Significance of the topic

Immunopeptidomics defines the repertoire of HLA-presented peptides that drive T cell recognition and underpin immunotherapy decisions such as neoantigen vaccines and adoptive T cell therapies. High-confidence sequence annotation is essential because low-abundance peptides, isobaric residues, post‑translational modifications (PTMs) and noncanonical sequences can introduce false positives that misdirect target selection and downstream development. Improving fragmentation strategies and workflows to increase sequence coverage and verification of candidate neoantigens directly improves translational confidence and target de‑risking.

Objectives and overview of the study

The work evaluated the Orbitrap Tribrid Apex MultiOmics mass spectrometer using electron‑transfer/higher‑energy collision dissociation (EThcD) versus higher‑energy collisional dissociation (HCD) for HLA Class I peptide analysis. The study had two main aims: (1) compare sequence coverage and fragment information produced by EThcD relative to HCD in data‑dependent acquisition (DDA) runs, and (2) implement and demonstrate Neo‑SCOAD (Neoantigen Simultaneous Confirmation and Discovery) — a combined DDA–parallel reaction monitoring (PRM) workflow that targets a panel of neoantigens (26 HLA Class I peptides) while still performing discovery in a single injection to increase annotation confidence for selected targets.

Methods and experimental design

• Biological material and sample prep: HCT116 cells were cultured ± 10 ng/mL IFN‑γ overnight. Cell pellets (10 million cells for immunoprecipitation; peptide load equivalent to 1 million cells on column) were lysed with Mem‑PER buffer. HLA Class I complexes were immunoaffinity captured using W6/32 antibody coupled to Protein A/G magnetic beads via the Thermo Scientific Pierce MHC Class I Antibody Coupling Kit on a KingFisher Apex system. Bound peptides were eluted with 1% TFA, dried and resuspended in 0.1% formic acid prior to nanoLC–MS/MS.
• LC and MS conditions: Peptides were separated on a Vanquish Neo UHPLC using an IonOpticks Aurora Ultimate 25×75 XT C18 column, 72‑minute total run, EASY‑Spray ion source. Mass spectrometry used the Thermo Scientific Orbitrap Tribrid Apex MultiOmics MS in DDA mode (n=3 replicates) comparing EThcD and HCD fragmentation. Neo‑SCOAD combined DDA discovery scans with targeted PRM scans for 26 peptides in the same injection.
• Data processing: Spectra were analyzed with PEAKS Studio 13 including PEAKS DeepNovo for de novo-assisted database searching against the UniProt human proteome (no‑enzyme search). Targeted PRM data were processed in Skyline. Motif and binding predictions for 9‑mer peptides used Seq2Logo and NetMHCpan 4.1.

Instrumention used

• Orbitrap Tribrid Apex MultiOmics MS (Thermo Fisher Scientific)
• Vanquish Neo UHPLC system (Thermo Fisher Scientific)
• IonOpticks Aurora Ultimate 25×75 XT C18 UHPLC column
• EASY‑Spray ion source (Thermo Fisher Scientific)
• KingFisher Apex Purification System (Thermo Fisher Scientific)
• Pierce MHC Class I Antibody Coupling Kit; W6/32 antibody; Protein A/G magnetic agarose beads
• Software: PEAKS Studio 13 with DeepNovo, Skyline, Seq2Logo, NetMHCpan 4.1

Main results and discussion

• Improved sequence coverage with EThcD: EThcD fragmentation produced more diagnostic fragment ions per backbone cleavage site than HCD, increasing confident amino‑acid assignments (CAA) across 9‑mer HLA peptides. The proportion of peptides with >80% sequence coverage approximately doubled with EThcD versus HCD.
• Identification counts: Across triplicate DDA experiments, EThcD yielded higher peptide counts relative to HCD (reported averages in the work indicate ~7.2k peptides for EThcD versus ~5.8k for HCD under the experimental conditions), demonstrating enhanced spectral annotation for the immunopeptidome.
• Complementary identification performance: Although EThcD improved sequence confidence and depth of fragment information, both fragmentation strategies contributed uniquely to overall identifications; some peptides were preferentially observed by HCD, indicating complementarity rather than absolute replacement.
• Neo‑SCOAD tradeoffs and benefits: Incorporating EThcD into the Neo‑SCOAD DDA–PRM workflow enabled simultaneous targeted confirmation and discovery. Targeted scans increased annotation confidence for the 26 selected neoantigens, but sharing instrument cycle time with PRM reduced raw discovery identification rates compared to an unmodified DDA run. Thus Neo‑SCOAD prioritizes confident validation of predefined targets while retaining discovery capability.
• Allelic and motif consistency: Allele mapping and motif analyses of identified 9‑mers showed no systematic differences between HCD and EThcD datasets, indicating the fragmentation choice did not bias allele assignment or canonical binding motifs.

Benefits and practical applications

• Enhanced neoantigen confirmation: EThcD provides richer fragmentation useful for resolving isobaric residues and localizing modifications or sequence variants, increasing confidence for neoantigen candidates used in vaccine design or T cell receptor development.
• Single‑run confirmation + discovery: Neo‑SCOAD offers a practical workflow for labs that need to both validate specific targets and explore the broader immunopeptidome without multiple separate acquisitions, streamlining sample use and throughput.
• Clinical and translational impact: Better sequence annotation reduces false positives in target nomination, aiding target de‑risking, regulatory documentation, and informing downstream biological validation and immunotherapy design.

Limitations and practical considerations

• Cycle time tradeoff: Combining targeted PRM with discovery DDA reduces time available for untargeted MS2 acquisition and can lower discovery depth; experimental design must balance sensitivity for targets versus breadth of discovery.
• Instrument and method complexity: EThcD requires instruments capable of ETD/EThcD and optimized parameters; reproducible implementation demands careful optimization and software capable of leveraging mixed fragmentation data.
• Sample input: Immunopeptidomics remains sample‑intensive; results here were obtained from immunoprecipitations equivalent to millions of cells and may be challenging for scarce clinical specimens without enrichment or amplification strategies.

Future trends and potential applications

• Integration with real‑time decision making: On‑instrument decision algorithms could dynamically allocate cycles between discovery and targeted scans based on precursor priority and realtime identification, maximizing both breadth and confirmation confidence.
• Advanced computational pipelines: Machine learning‑assisted de novo sequencing and spectrally informed rescoring will further improve confidence in noncanonical or modified peptide assignments from mixed fragmentation spectra.
• Expanded PTM and variant characterization: EThcD’s improved fragmentation is well suited to localize PTMs and sequence variants; future workflows may routinely integrate proteogenomic databases for personalized neoantigen discovery in clinical samples.
• Standardization for clinical use: Harmonized sample prep, instrument settings and data analysis pipelines will be important to translate enhanced fragmentation strategies into regulated clinical assays for immunotherapy target selection.

Conclusion

Use of EThcD on the Orbitrap Tribrid Apex MultiOmics MS improves fragment coverage and sequence confidence for HLA Class I peptides compared with HCD, increasing the proportion of highly covered peptide identifications and providing richer spectral information to resolve ambiguous assignments. The Neo‑SCOAD workflow demonstrates a practical combined targeted‑discovery approach that boosts confidence in nominated neoantigens while still enabling broader immunopeptidome profiling, with a tradeoff in discovery depth due to shared cycle time. This approach strengthens translational immunopeptidomics by delivering higher‑confidence spectral annotation for immunotherapy target nomination.

References

1. Thomsen M.C.F., Nielsen M. Nucleic Acids Research. 2012;40(W1):W281‑W287.
2. Reynisson B., et al. Nucleic Acids Research. 2020;48(W1):W449‑W454.