The research, led by Cédric Poyet of Stadtspital Triemli in Zurich and Marie Lork of the University Hospital of Zurich, together with colleagues from Kantonsspital Baden, Luzerner Kantonsspital and University Hospital Zurich, set out to identify molecular correlates of overall survival in muscle-invasive bladder cancer, often abbreviated MIBC. The team analyzed tumor DNA extracted from cystectomy specimens, the surgical samples obtained when the bladder is removed, from 32 patients treated at Swiss centers. The study received ethical approval from the Cantonal Ethics Committee Zurich and was conducted in accordance with the Declaration of Helsinki.

To characterize the genomic landscape of each tumor, the investigators used the FoundationOne CDx comprehensive genomic profiling platform, a targeted next-generation sequencing assay capable of detecting substitutions, insertions and deletions, copy number alterations and selected genomic instability markers across hundreds of cancer-related genes. Patients were then divided into two comparison groups based on a hard clinical endpoint: a favorable outcome group of 14 patients who survived at least 60 months after surgery, and a poor outcome group of 18 patients who survived fewer than 60 months. This dichotomy allowed the researchers to ask a simple but clinically vital question: which genomic features separate long-term survivors from those who die earlier of their disease?

Across the entire cohort, the sequencing effort uncovered 279 pathogenic or likely pathogenic mutations distributed across 88 genes. The most frequently altered genes were familiar names in bladder cancer biology: TP53, the guardian-of-the-genome tumor suppressor whose inactivation is a near-universal event in this disease; PIK3CA, a signaling kinase driving PI3K pathway activation; KDM6A, a histone demethylase involved in chromatin regulation; and FGFR3, a receptor tyrosine kinase that is a well-established oncogenic driver and drug target in urothelial carcinoma. Perhaps surprisingly, the distributions of these frequent alterations were similar between the favorable and poor outcome groups, indicating that the presence or absence of these canonical mutations alone does not explain the dramatic survival differences observed in the clinic.

The team next turned to the standard quantitative indicators of genomic instability that have been proposed as prognostic and predictive biomarkers in many tumor types. Tumor mutational burden, or TMB, reflects the total number of somatic mutations carried by a tumor and is widely used as a proxy for responsiveness to immune checkpoint inhibitors. Microsatellite instability, or MSI, marks defects in DNA mismatch repair and carries prognostic and predictive significance in colorectal and several other cancers. In this MIBC cohort, however, both metrics were comparable between the long-term survivors and the poor outcome group, and neither proved prognostically informative. The finding is a caution against assuming that biomarkers validated in other cancers will translate directly to bladder cancer.

The decisive signal emerged from a different layer of genomic complexity: copy number alterations. Tumors from the poor outcome group exhibited a significantly higher frequency and burden of gene amplifications, events in which segments of DNA containing particular genes are copied multiple times, often massively, driving overexpression of the encoded proteins. Crucially, these amplifications frequently involved known oncogenes and co-amplification hotspots, regions of the genome where neighboring growth-promoting genes are gained together in a single event. In other words, patients whose tumors carried a heavy load of oncogene amplifications were disproportionately represented among those who died within five years of cystectomy.

The biological logic behind this observation is compelling. While point mutations typically disable tumor suppressors or alter the function of a single protein, amplifications act as gene dosage escalators, flooding tumor cells with growth factor receptors, signaling kinases and cell cycle accelerators. High-level amplification of oncogenes can simultaneously promote proliferation, survival under therapeutic stress and metastatic competence. Moreover, co-amplification events can deliver several oncogenic payloads at once, creating tumors that are intrinsically more aggressive and harder to eradicate with a single targeted agent. The Swiss findings suggest that this dosage-driven mode of tumor evolution may be a hallmark of the most lethal forms of MIBC.

The results also carry therapeutic implications. Amplified oncogenes are, in principle, druggable targets. FGFR inhibitors are already approved for metastatic urothelial carcinoma in tumors with FGFR alterations, and agents directed against amplified receptor kinases and downstream signaling nodes are in clinical development across many cancer types. If the association between amplification burden and poor survival is confirmed, comprehensive copy number profiling at the time of cystectomy could help identify patients who warrant intensified treatment, such as perioperative systemic therapy escalation, enrollment in targeted therapy trials or closer surveillance for recurrence. Conversely, the lack of prognostic value for TMB and MSI in this cohort suggests that these markers should not be relied upon in isolation for outcome prediction in MIBC.

The authors are careful to frame the study as exploratory, and the caveats are substantial. The cohort comprised only 32 patients, divided into groups of 14 and 18, a sample size that limits statistical power and leaves open the possibility of confounding by clinical factors such as stage, nodal status and treatment sequence, which the abstract does not address in detail. The use of a targeted panel, while broad, does not capture the full spectrum of structural variants and noncoding alterations that whole-genome sequencing would reveal. The authors explicitly call for validation in larger cohorts to determine whether oncogene amplifications can serve as robust prognostic markers and to explore their potential as therapeutic targets. It is also worth noting that Roche funded the genomic testing through the FoundationOne CDx platform but had no role in study design, data analysis, interpretation or manuscript writing, apart from being granted the opportunity to review the manuscript prior to submission.

Even with these limitations, the study adds an important dimension to the ongoing effort to bring precision oncology to bladder cancer. The field has long focused on the mutational catalog of urothelial carcinoma, one of the most heavily mutated of all common tumors, yet this work suggests that the architecture of copy number gains may carry at least as much prognostic weight as the mutation list itself. For patients facing cystectomy, a procedure with significant morbidity and a five-year survival that remains unsatisfactory for many, any molecular signal that reliably separates indolent from lethal disease is valuable. If larger studies confirm that oncogene amplification burden predicts poor overall survival, clinicians may one day sequence not just for mutations but for the sheer number of oncogene copies a tumor carries, using that information to triage patients toward more aggressive, and hopefully more effective, treatment strategies from the moment of diagnosis.

Subject of Research: Genomic profiling of oncogene amplifications as prognostic markers of overall survival in muscle-invasive bladder cancer

Article Title: Oncogene-driven genomic profiles are linked to poor overall survival in muscle-invasive bladder cancer (MIBC)

Article References: Poyet, C., Franzen, A. S., Bieri, U., Kaufmann, E., Eberli, D., Schmid, M., Zoche, M., Moch, H., & Lork, M. (2026). Oncogene-driven genomic profiles are linked to poor overall survival in muscle-invasive bladder cancer (MIBC). Journal of Cancer Research and Clinical Oncology. https://doi.org/10.1007/s00432-026-06626-2

Image Credits: AI Generated

DOI: 10.1007/s00432-026-06626-2

Keywords: muscle-invasive bladder cancer, oncogene amplifications, genomic profiling, tumor mutational burden, microsatellite instability, TP53, FGFR3, PIK3CA, copy number alterations, prognostic biomarkers, cystectomy, FoundationOne CDx

The research emerged from the Copenhagen Prospective Personalized Oncology trial, known as CoPPO, an ongoing precision oncology study in which patients with advanced cancer undergo molecular profiling to guide therapy selection. In this setting, metastatic biopsies are routinely processed using the Qiagen AllPrep system, a commercial kit that simultaneously extracts DNA and RNA from a single sample. During that extraction, proteins are separated from the nucleic acids and end up in a flowthrough fraction that is normally discarded. The team, led by Filip Mundt and including renowned proteomics pioneer Matthias Mann, systematically asked whether this residual protein fraction could serve as a reliable starting material for mass spectrometry-based proteomics, and the answer turned out to be a resounding yes.

The technical challenge was considerable. Protein recovered from extraction flowthroughs is diluted in chemical buffers designed for nucleic acid work, and it must be recovered, concentrated, and cleaned up before it can be digested into peptides and analyzed by liquid chromatography coupled to mass spectrometry. The researchers developed and optimized procedures for each of these steps, including protein precipitation, enzymatic digestion, and peptide cleanup, and they evaluated multiple mass spectrometry acquisition strategies to determine which would deliver the best performance from this unconventional source material. Both data-dependent acquisition, which favors depth of identification, and data-independent acquisition, which favors quantitative consistency across samples, were tested in the context of real clinical biopsy specimens rather than idealized reference material.

The results were striking. Using optimized single-shot workflows, the team consistently generated deep proteomes from the AllPrep-derived material, quantifying more than 10,000 protein groups in data-independent acquisition analyses. That depth of coverage, from material that would otherwise have been discarded, places the quality of these measurements on par with dedicated proteomics workflows that consume precious additional tissue. Perhaps even more importantly for clinical applications, the proteomes retained tissue-associated biological signatures, meaning the protein profiles faithfully reflected the biology of the tumors from which the biopsies were taken rather than artifacts introduced by the extraction process.

One of the most clinically significant findings concerns compatibility with existing genomic diagnostics. The researchers found that the proteomes derived from the flowthrough material quantified protein products corresponding to approximately 71 percent of the genes represented on the FoundationOne CDx panel, a widely used clinical genomic profiling assay. This overlap means that the protein data can be directly interpreted alongside the genomic results that oncologists already rely on, providing a functional layer of information on top of the genetic alterations detected in the same specimen. Where genomics reveals which genes are mutated or amplified, proteomics reveals whether those changes are actually reflected in the abundance of the proteins they encode, and whether downstream signaling pathways are active.

Beyond total protein abundance, the team extended the workflow to phosphoproteomics, the analysis of phosphorylation marks that act as molecular switches on proteins and report on the activity of signaling pathways. Phosphoproteomic analyses of the recovered material identified more than 10,000 phosphorylation sites, including many associated with key oncogenic signaling pathways driven by AKT1, BRAF, and EGFR, proteins that sit at the heart of some of the most commonly targeted pathways in modern cancer therapy. This is particularly valuable in precision oncology, where a tumor may harbor a mutation in a druggable gene, yet the downstream pathway may or may not be activated. Phosphoproteomics offers a direct readout of that functional state, information that genomic sequencing alone cannot provide.

Practical considerations for clinical deployment were also addressed head-on. The researchers assessed sample stability during long-term storage and found that protein integrity was preserved in samples kept at minus 80 degrees Celsius for up to five years. This finding has immediate operational implications, because it means that flowthrough fractions recovered during routine DNA and RNA extraction can simply be archived in existing freezers, creating a retrospective proteomic resource from biopsies processed years earlier. Hospitals would not need to change their current extraction protocols or acquire new tissue; they would only need to stop discarding the protein fraction and store it instead.

Scalability, often the Achilles heel of clinical proteomics, was another focus of the study. Mass spectrometry throughput has historically been a bottleneck for large-scale clinical studies, but the team implemented short liquid chromatography gradients that enabled high-throughput analysis of up to 60 proteomes per day without compromising data quality. At that rate, a single instrument could theoretically process thousands of patient samples per year, bringing proteomic profiling within reach of the kind of scale that genomic sequencing pipelines already achieve routinely. The analytical reproducibility of the workflow was also evaluated, confirming that measurements were consistent enough to support the comparisons that clinical decision-making would require.

The implications for precision oncology are substantial. Current molecular tumor boards typically work from genomic and transcriptomic data, which describe the blueprint and the transcriptional activity of a tumor but say little about the actual effector molecules, the proteins, that carry out cellular processes and respond to drugs. Proteins are, as the authors note, the active effectors of cellular signaling, and integrating proteomic and phosphoproteomic measurements with genomic data has become an important objective in translational oncology. By demonstrating that this integration can be achieved from a single biopsy already being processed in routine clinical workflows, the study removes one of the biggest practical obstacles: the need for extra tissue that patients often cannot spare and that clinics often cannot obtain.

The Copenhagen team frames the work as a framework for integrated proteogenomic studies in precision oncology, and the phrase is apt. What they have built is not merely a new laboratory technique but a bridge between the molecular diagnostics that hospitals already perform and the deeper functional characterization that proteomics can add. As mass spectrometry instruments become faster and more sensitive, and as workflows like this one demonstrate that clinical material streams can feed them without added burden on patients, the vision of truly multi-layered molecular profiling, spanning genome, transcriptome, proteome, and phosphoproteome from a single needle biopsy, moves closer to routine reality. For patients with advanced cancer, whose treatment options often hinge on the quality and completeness of the molecular information available, that could mean better-matched therapies chosen from a far richer picture of what their tumor is actually doing.

Subject of Research: Recovery of protein from DNA/RNA extraction flowthroughs for scalable clinical proteomics and integrated proteogenomic profiling of cancer biopsies

Article Title: Scalable clinical proteomics from DNA/RNA extraction flowthroughs enables integrated proteogenomic profiling from a single cancer biopsy

Article References: Mundt, F., Bach Nielsen, A., Wang, J., Kerzel Duel, J., Westmose Yde, C., Amnitzbøll Eriksen, M., Lassen, U., Cilius Nielsen, F., Rohrberg, K., & Mann, M. (2026). Scalable clinical proteomics from DNA/RNA extraction flowthroughs enables integrated proteogenomic profiling from a single cancer biopsy. Clinical Proteomics. https://doi.org/10.1186/s12014-026-09632-1

Image Credits: AI Generated

DOI: 10.1186/s12014-026-09632-1

Keywords: proteomics, precision oncology, mass spectrometry, phosphoproteomics, cancer biopsy, proteogenomics, DNA/RNA extraction, clinical proteomics, CoPPO trial, FoundationOne CDx, data-independent acquisition, signaling pathways