Exposing Hidden Layers of Cancer Biology With Proteomics

At this year's American Association for Cancer Research annual meeting, the conversation around cancer research felt noticeably different compared to the last time I attended. Genes and mutations were still very much present—and of course remain central to how the field thinks about disease—but they no longer seemed to define the full scope of discussion. 

There was a stronger focus on what is actually happening inside cells in real time, under true biological conditions, driven by molecules that do the action. Proteomics, despite how far the field has come, is still massively underleveraged relative to genomics, but it is increasingly part of the conversation about where cancer biology needs to go next. 

“Genomics looks at the mutational landscape, meaning you have this level of abstraction between what is expressed at the transcript level and what proteins are actually present, or which protein pieces the immune system monitors on the cell surface to distinguish cancer from normal cells,” said Dr. Daniel Hornburg, vice president of biomarkers and precision medicine at Bruker.  

A single gene does not produce a single protein. Instead, one gene can give rise to many different proteoforms (sequence variants and post-translational modifications). These different molecular forms can change how a protein behaves and interacts, ultimately shaping cellular activities.

As Hornburg put it, “For us as a proteomics community, the opportunity and a responsibility is to engage more closely, continue to radically simplify access, and clearly communicate how different proteomics modalities can dissect cancer biology today in practice.”

It seems a static map of mutations can only take us so far. So, where can proteomics have the biggest impact?  

Technology Networks spoke with Hornburg to explore that question and the key areas where proteomics is starting to reshape cancer biology in practice. 

A system-level view of drug action 

Traditional drug discovery has often been a narrow pursuit: find a target, design a molecule, test whether it binds—simple, right? But just because this has been the dominant approach for a long time doesn’t necessarily mean it’s the best way forward.

What if we broaden the field of view—rather than peeking through a keyhole, you want to swing the door wide open so that the entire room is in view. That’s where chemoproteomics comes in.

Instead of focusing on a single interaction, chemoproteomic techniques allow researchers to profile the effects of a compound across thousands of proteins at once. 

“You have a compound library, and you want to understand which of those compounds—some already FDA [US Food and Drug Administration] approved—help you to down-regulate a certain protein that could be problematic in a disease,” explained Hornburg. “That’s protein degradation.” 

While it is obviously important to understand which compounds modulate the target protein, it’s equally important to know their effects on everything else they touch along the way. You can achieve this with mass spectrometry (MS), whereby cell lines are treated in parallel, and thousands of proteins are tracked across the system, showing which compounds drive downregulation of specific proteins. 

“Importantly, you also see all the off-target effects and what else happens in the cell,” he reiterated. 

The need for speed 

But newer approaches based on ion mobility are collapsing timelines. Hornburg noted that one of their collaborators, Dr. Jacob Geri at Weill Cornell, has spearheaded an approach using trapped ion mobility. This in-gas phase separation technique replaces liquid phase pull-down approaches to identify which proteins are active and can be engaged with chemical compounds.  

“The liquid phase takes time, requires quite a bit of input material, and it can be messy. With trapped ion mobility, you can monitor target engagement by observing how relevant protein fragments travel. They ‘fly’ differently because they have a different shape in the gas phase,” said Hornburg. The result is separation in the mass spectrometer in milliseconds, rather than hours or minutes. 

Beyond its sensitivity and speed, MS offers another key advantage: it creates a complete digital record of the enrichment process. Researchers can track all ions as they pass through the instrument, with every event captured in real time. This is very different from traditional liquid-phase reactions in a test tube, where the underlying processes cannot be observed directly.

Circulating Immune Proteins in Liver Cancer Research

This webinar explores how circulating immune-related proteins can improve our understanding of hepatocellular and gallbladder cancer risk across diverse populations with viral hepatitis.
View Webinar

Advertisement

Looking beyond mutations 

I asked Hornburg why it is important to look beyond what is mutated when it comes to tumor neoantigens, and what proteomics approaches can reveal that genomics can’t. 

He explained that while genomics can identify mutations that may produce neoantigens, it cannot show which of those neoantigens are actually presented on the cell surface for recognition by the immune system.
 

MS has become sensitive and accurate enough to directly measure this cancer‒immune interface.
 
“We can now directly measure what’s on the surface of a cancer cell, thousands of molecules at a time. We cleave off these HLA [human leukocyte antigen] complexes and see what’s really sitting there,” Hornburg explained. 

This goes beyond genetically encoded changes, too. It also captures modified peptides that come about after translation. “We can also do de novo sequencing and look for new antigens in the space of PTMs (post-translational modifications), like phosphorylation or glycosylation, which is a space completely opaque to genomics,” he added. 

Hornburg noted that it’s even possible to “do this in the absence of the cancer cell.” 

“Matthias Mann published a paper on capturing HLA directly from plasma using an IMBAS [Immunopeptidomics by Biotinylated Antibodies and Streptavidin] workflow, pulling out potential neoantigens directly from the circulation,” he said.  

You take a blood sample and examine the immunopeptide repertoire circulating in the blood, which could reveal cancer-associated neoantigens, as well as other inflammatory signatures, even before the primary tumor or tissue context is identified. 

The drug graveyard, revisited 

For all the excitement around new technologies, there is also growing recognition that some past failures in drug development may not have been true failures at all. Instead, they may simply reflect an incomplete view of biology at the time. 

Scaling Genome Editing From Discovery to Manufacturing

In this webinar, our expert speaker will discuss how non-viral electroporation-based genome editing supports applications ranging from CRISPR screening to complex cell engineering.
View Webinar

Advertisement

That opens an underexplored question around existing compounds and how they are evaluated. Hornburg stressed that “drug repurposing at a proteoform level is a landscape for biomedically relevant insights that is largely untapped.” 

The challenge is that proteins are not single, static entities. They exist as multiple proteoforms—shaped by sequence variation, post-translational modifications, and cellular context—which can alter their function and interactions. 

“If you take all the proteoforms and variants into account,” he added, “we are talking about millions of different molecules.” 
 
This complexity also changes how researchers think about drug action. As Hornburg noted, if researchers can identify specific signaling pathways that a tumor relies on for survival, it could be possible to “target relevant proteoforms involved in critical tumor survival mechanisms. Such intervention may become very specific to the tumor and less disruptive to normal cellular processes. This requires a system- and proteoform-level understanding of biology as well as the biophysical engagement between small molecules and proteoforms.”  

Hornburg emphasized that it is often the specific proteoform, rather than the protein itself, that matters most. A signaling protein can exist in multiple states, and modifications such as phosphorylation can create proteoforms with different functions. A compound that appears ineffective against a protein in general may be highly relevant when viewed through the lens of specific proteoforms. 

Perhaps some discarded compounds may still have potential in the clinic... 

Cancer as a spatial disease 

This complexity is partly driven by intratumor heterogeneity, which comes about through multiple mechanisms that can be broadly grouped into intrinsic and extrinsic factors.  
 
Intrinsic factors refer to properties within the cancer cell that impact its disease phenotype (i.e., genomic alterations, epigenetic changes, and transcriptomic or proteomic variation). Extrinsic factors come from the tumor microenvironment—the surrounding ecosystem of immune cells, stromal cells, extracellular matrix, and secreted signaling molecules that influence how cancer cells behave. 

“Spatial context matters a lot,” Hornburg said. “Understanding how immune signaling and gradients of cytokines are distributed across the tumor requires spatial investigation.” 

New tools are making that investigation possible. Researchers can now isolate tiny regions of tissue and measure thousands of proteins simultaneously, providing insight into how biology varies across different areas of a tumor. 

“You can cut out small pieces—just a few cells—to create a functional cluster, and quantify thousands of proteins in minutes.” 
 
A similar approach can be performed even faster using matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS). This allows researchers to analyze small molecules and lipids by essentially scanning across a tumor section with a laser, mapping which lipids and metabolites are present and active in different regions. In the same context, targeted proteomics can also be applied to build a spatial view of metabolic state across the tumor. 

“When combined with deep proteomic profiling, you get a much more complete picture of the system. You are not only capturing what is happening inside cells, but also what is being secreted and how the extracellular matrix is behaving—areas that are largely invisible to genomics,” said Hornburg.  

What emerges from our conversation is the growing centrality of proteomics in capturing the complexity of cancer as it unfolds in real time. The next step is making this complexity usable—turning richer biological insight into clearer clinical decisions.