Significant technological advances in genomics and transcriptomics have helped to provide many new insights in the development of therapies to fight cancer, but they are only part of the equation.
While proteomics as a field is not new, proteomics technologies such as mass spectrometry have only recently achieved the high degree of finesse, reproducibility, and quantitation required to provide integrated insights across every stage of the oncology research pipeline.
Advances in quantitative proteomics have great potential to improve cancer research and treatment by uncovering and validating novel drug targets, helping to understand drug mechanisms of action, and searching for new diagnostic, prognostic, or predictive biomarkers.
Biognosys, a spinoff from the lab of proteomics pioneer Ruedi Aebersold at ETH Zurich, has continued to push the development of technologies and methods in mass spectrometry — the primary tool for proteomics research.
The company has focused on inventing and implementing workflows for data-independent acquisition (DIA) mass spectrometry. This is a deeper, more efficient, reliable, and reproducible quantitative proteomics method compared with conventional technologies.
This next-generation proteomics approach combines highly parallelized data acquisition with advanced algorithms for data processing that can capture all detectable signals in every sample analysis.
Several examples of this innovation were on display at the recent American Association for Cancer Research (AACR) Virtual Annual Meeting, where Biognosys published seven posters highlighting the utility of its next-generation proteomics platform in oncology.
Enabling Early Drug Discovery
Finding new drugs for treating cancer can be a lengthy process. Many drugs fail at some point along the journey from lab to clinic, with target validation being a common stumbling block.
To solve this challenge, ETH Zurich and Biognosys have developed a new technology called Limited Proteolysis (LiP), which allows drug targets to be identified and validated across the human proteome.
The LiP technique differentiates drug-bound proteins from unbound proteins using quantitative mass spectrometry. It provides a precise readout of all the proteins and the sites within them that bind a particular compound, even in complex mixtures of cell extracts, including identifying undesired off-target binding events.
In a new extension of the LiP technique, Biognosys, in collaboration with biotech partner Cedilla, has developed a high-resolution method allowing drug binding sites to be mapped in peptide-level resolution.
A further application also reveals whether the binding process causes any structural changes that affect protein-protein interactions, helping researchers home in on compounds with high affinity and efficacy.
Understanding Drug Mechanisms of Action and Resistance
PARP inhibitors are designed to kill BRCA1/2 deficient tumor cells and have revolutionized treatment for some patients with inherited gene variants. However, varying degrees of treatment response and resistance have been observed with these drugs, particularly in patients with advanced disease.
Researchers based at the University of Cambridge collaborated with Biognosys to investigate PARP-inhibitor resistance using proteomic profiling. They characterized samples taken from a patient-derived xenograft model of BRCA1/2 breast cancer and identified metabolic, structural, and neuronal protein pathways linked to PARP resistance, highlighting novel therapeutic avenues.
Proteomics is also being put to work to solve a similar challenge around the efficacy of cancer immunotherapies. Immune checkpoint inhibitors, such as PD-1 and PD-L1 inhibitors for advanced melanoma, have shown great efficacy in treating some patients. but are only effective in around 20 percent of patients. Yet the reasons for this — and the precise mechanisms of action of immunotherapy drugs on the wide array of immune cell subtypes — are still not well understood.
In collaboration with Crown Bioscience, Biognosys researchers showed that a reduction in the number of CD4+ immune cells improved the efficacy of PD-1 inhibition in syngeneic mouse models of colorectal and liver cancer. The researchers believe this may be due to a linked increase in levels of interferon gamma, a protein that appears to boost the immune system’s ability to fight cancer.
Identifying Actionable Clinical Biomarkers
Researchers from INT-Pascale, a large cancer center in Southern Italy, used Biognosys’ proteomic techniques in a clinical setting to dig deeper into why PD-1 immunotherapy is effective in only some patients.
In samples taken from melanoma patients treated with PD-1 immunotherapy, 25 proteins were associated with treatment response. One in particular — PLEKHA5 — was associated with poor response to treatment and increased risk for brain metastasis, highlighting the utility of this technique for prognostic biomarker discovery.
A further advantage of Biognosys’ next-generation proteomics platform is that it can identify and quantify thousands of proteins in relatively small samples of highly complex tissues or tumors, as well as biofluids such as blood plasma or urine. This now makes it possible to use proteomics to measure and compare protein biomarkers across large-scale, longitudinal clinical studies.
It is well established that detection of circulating tumor biomarkers in the blood can be useful for early cancer detection, supporting diagnosis and monitoring therapeutic response. While these liquid biopsy techniques have tended to focus on genomic and transcriptomic information, proteomic data has the potential to provide deeper understanding of the changes in circulating proteins linked to cancer development and progression.
Until recently, there were technical challenges that limited the reliability and reproducibility of quantitative proteomic analysis on a large scale. DIA mass spectrometry has largely solved these problems and can analyze complex tissue or tumor samples quickly and effectively. The high levels of reproducibility of this technique make it possible to compare multiple longitudinal samples collected over the duration of a clinical trial.
To demonstrate the potential for proteomics in cancer diagnostics, Biognosys tested blood plasma samples taken from patients with non-small cell lung cancer (NSCLC). A panel of 162 proteins appeared to be significantly dysregulated in NSCLC, showing good diagnostic potential, with 25 of the most variable of these proteins linked to the host immune response to the tumor.
Finally, Biognosys has also been working with German precision oncology specialists Indivumed to develop a high-throughput workflow for robust quantitative analysis of proteomes and phospho-proteomes across Indivumed’s biobank of tumor and adjacent healthy tissue samples from thousands of cancer patients.
Quantifying an average of 5,903 proteins and 28,819 phosphopeptides per NSCLC tissue sample revealed a number of clusters in known dysregulated pathways, such as EGFR signaling, as well as revealing new insights into cancer biology.
This information is being used to enrich Indivumed’s IndivuType multi-omics database, to help researchers identify and validate biomarkers or drug targets and aid the design of clinical trials. This work illustrates the suitability of next-generation proteomics for large clinical research projects, making large-scale multi-omics a reality.
Toward Multi-Omic Cancer Research at Scale
Collectively, these research findings presented at the AACR 2020 Annual Meeting demonstrate the depth and breadth of insights that next-generation proteomics can bring to the field of oncology.
While DNA and RNA provide valuable information, a multi-omics approach — including proteomics — is needed to better understand the complexity of the disease, overcome the emergence of resistance to current treatments, and provide new directions for the development of novel therapies in the future.
View and download the posters summarized in this article here.