Where the technology once centered on detecting more molecules, developers are now asking a different question: Can complex biology be understood earlier, with results that hold up across larger studies and more decision points?
That shift was a throughline in conversations at this year's American Society for Mass Spectrometry (ASMS) conference, according to Aaron Robitaille, Senior Director of Product and Vertical Marketing for mass spectrometry at Thermo Fisher Scientific, who pointed to a broader move away from standalone instruments toward integrated, end-to-end workflows spanning sample preparation through data interpretation.
Complex modalities demand deeper characterization
As therapeutic modalities such as biologics, oligonucleotides, and cell and gene therapies grow more structurally complex, mass spectrometry is being asked to do more than confirm identity. "[Mass spectrometry] is moving from 'Can we confirm what this is?' to 'Can we deeply understand the molecule and its critical attributes?'" Robitaille told DDN.
For biologics, that means greater emphasis on intact and subunit analysis, top-down and middle-down workflows, impurity profiling, and advanced fragmentation techniques capable of resolving highly heterogeneous molecules, such as antibodies, antibody-drug conjugates, and multispecific therapeutics.
Oligonucleotide analysis presents a different challenge, Robitaille said, since small analytical differences between molecules can carry outsized meaning. Workflows for this modality need to minimize sample loss, improve reproducibility, and support confident sequence and impurity analysis.
Reproducibility, not just data volume, is the bottleneck
In proteomics and metabolomics, Robitaille said the central challenge isn't generating more data but trusting it across cohorts, timepoints, and decision points. "The practical question is not simply, 'Can I generate more data?'" he told DDN. "It is, 'Can I trust the data across cohorts, across time, and across decision points?'"
That is pushing the field toward standardized workflows, AI-assisted data interpretation, and closer integration between discovery-scale omics and large-cohort validation studies. In proteomics specifically, dynamic range remains a persistent obstacle, since disease-relevant proteins or modified proteoforms are often present at low abundance and difficult to quantify reproducibly at scale. In metabolomics, the bigger constraint is confident annotation: Detecting a feature is different from identifying it, particularly for low-abundance metabolites in complex matrices like plasma.
What it means for drug developers
For teams running discovery and translational programs, the broader implication is that mass spectrometry infrastructure decisions are increasingly workflow decisions rather than single-instrument decisions.
Reproducibility across sites and studies, not just raw sensitivity, is becoming the standard programs are measured against as biomarker and characterization work scales into larger cohorts and later-stage development.
