Missense mutations are among the most frequent genetic alterations in prostate cancer, but determining whether – and how – they reshape the proteome has remained an open question. Unlike transcriptomic analyses, which only partially explain protein abundance, proteogenomic approaches can directly link mutations to functional protein changes.
In a new study published in Oncology Advances, researchers at the Federal University of Rondônia, Brazil, integrated SILAC-based quantitative mass spectrometry with curated mutation databases to map the impact of missense variants across prostate tumor and healthy tissues. The analysis revealed that proteins carrying mutations often showed significant reductions in abundance, particularly in key metabolic and stress-response pathways, and that variant-enriched proteins clustered in oncogenic processes such as apoptosis suppression and immune modulation. Hotspot analysis flagged recurrent alterations in genes including ACTB, PPIF, and HSPA9, with several mutations predicted to destabilize protein function.
By demonstrating how genetic alterations translate into measurable proteomic shifts, the study strengthens the case for proteogenomics in precision oncology, highlighting new candidate biomarkers and potential therapeutic targets in prostate cancer. We reached out to the study’s corresponding author, Lucas Marques da Cunha, to dig deeper into the study’s motivations, challenges, and translational implications.
Could you explain, in a nutshell, how your proteogenomic approach works?
Our strategy integrates quantitative proteomics using SILAC with mutation databases. Using this technique allows us to directly compare protein expression between healthy and tumor tissues, while the mutation databases (e.g. RefSeq and dbPepVar) reveal which mutated proteins are not only present in the genome, but are also expressed at the proteomic level. By combining these perspectives, we can identify which genetic alterations truly affect protein abundance and cellular pathways in prostate cancer, providing insights that would be missed by genomic or transcriptomic analysis alone.
What motivated your team to focus on the role of missense mutations in shaping the prostate cancer proteome?
Missense mutations are among the most common genetic alterations in prostate cancer, with potential to affect protein stability, function, and abundance. Yet, our understanding of how these mutations translate into measurable proteomic changes remains unclear. The challenge is that not every mutation in the genome produces a functional or abundance effect at the protein level. Distinguishing signal from noise requires a careful integration of genomic and proteomic data, which, until recently, has proved technically difficult. Our work was motivated by the need to close this gap and directly demonstrate how mutations shape the tumor proteome.
What was the greatest analytical challenge you faced during the study – and how did you overcome it?
The greatest challenge came in ensuring correct pairing between mutated and non-mutated peptides. Initially, these datasets seemed misaligned, and it was unclear whether we were comparing equivalent proteins. The breakthrough came when we introduced cross-validation using keys such as raw file and leading razor protein. This allowed us to match reference and variant peptides with high confidence, ensuring we were measuring the true effect of mutations on protein abundance rather than artifacts from data processing.
What kinds of clinical or translational applications might benefit most from this kind of proteogenomic analysis?
I see three main areas of impact. First, in biomarker discovery, where we identified recurrent mutations in genes such as ACTB, PPIF, and HSPA9 that directly influence protein abundance. Second, in therapeutic targeting, as we showed that enzymes like PGK1 and MDH2 are mutated and dysregulated in tumor tissues, highlighting their potential as drug targets. Third, in patient stratification, where integrating mutational and proteomic profiles may help distinguish patients at higher risk of recurrence, resistance, or metastasis – thus supporting more personalized treatment decisions.
What are the next steps for your team? More broadly, what advances still need to happen to make proteogenomic tools a practical part of precision oncology?
At the moment, we are developing a tool that allows any researcher to build a personalized mutation database tailored to their dataset or clinical study, which we believe will democratize access to proteogenomic analysis. In parallel, we are creating a module based on transformer-type artificial intelligence models that can predict the impact of mutations on protein structure and function. This will help us move beyond detection and provide deeper insights into molecular mechanisms.
More broadly, two advances are essential for proteogenomics to become routine in precision oncology. We first need more integrated and accessible computational infrastructures that bring together genomic, proteomic, and clinical data. Second, we need translational validation, ensuring that the variants and proteins identified by these approaches are confirmed in clinical cohorts and trials, ultimately transforming them into actionable biomarkers and therapeutic targets.
Lucas Marques da Cunha has a PhD in Bioinformatics and is an Adjunct Professor in the Department of Computer Science at the Federal University of Rondônia, Brazil
Newsletters
Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.
