Loading
Measuring the proteins in a tumour, which is what drugs actually hit, rather than the genes that encode them.
Proteomics measures the proteins and phosphorylation sites in a tumour directly, using liquid chromatography tandem mass spectrometry with isobaric labelling to quantify thousands of proteins and phosphosites at once. Mass-spectrometry proteomics (CPTAC) reveals pathway activity and ADC target abundance that transcript levels do not predict, since RNA and protein levels often disagree. Quantitative IHC and mass-spec assays of HER2 and TROP2 aim to improve ADC patient selection by measuring the actual antigen the drug binds rather than a proxy. Throughput and standardisation remain the barriers to routine clinical use, and it is still open whether protein-level selection will outperform established IHC scoring in trials. For a newcomer, proteomics measures what drugs actually hit, the proteins, rather than the genes that encode them.
Liquid chromatography tandem mass spectrometry with isobaric labelling quantifies proteins and phosphosites.
Exact Sciences' multi-cancer blood test, sold since 2025 as a lab test that has not been through FDA approval; it says whether a cancer signal is present but not where, so a positive result leads to scans.
The Johns Hopkins blood test that first showed one draw could find several cancers and point to the organ; its approach passed to Thrive Earlier Detection and then to Exact Sciences, whose Cancerguard test followed.
A German blood test that measures how blood proteins change shape; the only studies are small, run with the inventor, and do not say how often it finds early cancers.
An experimental blood test that reads a handful of proteins, with different panels for men and women, to find many early cancers; so far it has been tried only on a few hundred stored samples.
SeekIn's low-cost protein blood test for 14 cancers, designed for countries that cannot afford DNA sequencing; it finds about six in ten cancers but flags about one in twelve people without cancer.
A low-cost blood test from 20/20 BioLabs that measures a panel of protein tumour markers and is sold in the United States as a laboratory test without FDA review.
A German blood test sold through a private health insurance add-on, which claims to signal cancer of any type; its published evidence is small studies and case reports.
SeekIn's multi-omic blood test that adds DNA sequencing to its protein test; in published studies it found about six to seven in ten cancers and about four in ten at stage I.
Supports the design choice of most current tests: methylation first, with proteins as an add-on rather than mutations.
The current honest statement of where early detection stands: no blood test is ready, and the research is about finding the threshold at which to operate.
It is the only peer-reviewed stage-by-stage figure for the Cancerguard approach, and it describes a development classifier at 98.5% specificity, not the launched test, which Exact Sciences says runs at 97.4% specificity.
Consistent across sites and platforms, with the same limit as before: specificity too low for population screening without a second test.
One of few multi-cancer tests with a stage I figure above one in three, from a case-control set; the prospective figures rest on few cancers.
Very high figures from a small discovery set in which the same samples helped choose the proteins; figures like these usually fall when a test meets a larger, independent population.
Cheap and runs on hospital analysers, but at about 93% specificity a screening programme would generate many false alarms for each cancer found.
Proof of principle in three cancers from a small case-control set; it is not a screening result.
Query for this technology: (TITLE:"proteomics" OR ABSTRACT:"proteomics" OR TITLE:"proteogenomic" OR ABSTRACT:"proteogenomic" OR TITLE:"phosphoproteomics" OR ABSTRACT:"phosphoproteomics") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about Proteomics & phosphoproteomics, not a curated reading list.
Shares 20/20 BioLabs, Novelna, A large-scale, multi-centre validation study of an AI-empowered blood-based test for multi-cancer early detection, A panel of seven protein tumour markers for effective and affordable multi-cancer early detection by artificial intelligence: a large-scale and multicentre case-control study.
Shares Biodesix, Nonagen Bioscience, Carcimun test, OneTest for Cancer.
Shares Elucidate Bio, Find the parts of a tumour the drug never reaches, A global rapid tissue donation network for metastatic disease, Spatial-omics-guided treatment selection.
Shares Performance of a multi-biomarker class, multi-cancer early detection (MCED) blood test in a prospectively collected cohort, CancerSEEK, Cancerguard, Early detection roadmap: organ screening → blood tests for many cancers.
Open-source projects that implement or serve this technology, from OnCo's own catalogue: licence and last activity as the repository reported them on the day of the fetch. Listing is not endorsement; check the licence before reuse and the validation before clinical use.
A Python package that streams CPTAC proteogenomic cancer datasets into pandas.