OncoFirm™ / Cancer Atlas 2026 / Cancer biomarkers explained
Cancer Atlas 2026 · Biomarker fundamentals
Cancer Biomarkers Explained: Categories, Tumor Markers, Standards and Validation (2026)
A cancer biomarker is something measurable, in blood, tissue or an image, that informs risk, diagnosis, outlook or treatment. This guide explains the seven FDA-NIH biomarker categories, how serum tumor markers differ from molecular biomarkers, and what each classic marker is and is not recommended for. It also covers laboratory pitfalls and the evidence a new biomarker needs.
At a glance
Topic
What cancer biomarkers are, how they are grouped and validated
Framework
FDA-NIH BEST Resource: seven categories defined by use
Serum markers
Mainly for monitoring and recurrence; none is recommended for screening the general population
Molecular markers
Mainly predictive: they select treatment, often through a companion diagnostic
OncoFirm™ status
CEA and PSA assays in development; research use only; not cleared or approved by the FDA
On this page
Key numbers
Cancer biomarkers in three numbers
7biomarker categories, defined by use, in the FDA-NIH BEST Resource
74%of patients had PSA results more than 20% apart on two WHO-calibrated assays in one comparison
8 of 61FDA Biomarker Qualification Program projects had a fully qualified biomarker by 1 July 2025
Sources: FDA-NIH BEST Resource; Filella, Tumor Biology 2026; Collins et al., Therapeutic Innovation and Regulatory Science.
Definitions
What counts as a cancer biomarker
A biomarker is a defined characteristic measured as a sign of normal biology, of a disease process, or of a response to an exposure or treatment. The definition comes from the BEST Resource (Biomarkers, EndpointS, and other Tools), a glossary from the U.S. Food and Drug Administration (FDA) and the National Institutes of Health (NIH). A biomarker is not a measure of how a person feels, functions or survives.
The National Cancer Institute (NCI) uses the older term tumor marker: anything present in or made by cancer cells, or by other cells in response to cancer or to some benign (noncancerous) conditions, that gives information about a cancer. Benign conditions raise many markers too.
This page opens the OncoFirm™ Cancer Atlas 2026. Companion chapters cover tumor-associated antigens, multi-biomarker analysis and emerging diagnostic technologies.
Categories
The seven BEST biomarker categories
BEST sorts biomarkers by the job they do, not by what they are made of.
The seven biomarker categories of the FDA-NIH BEST Resource, each with a cancer-related example from the BEST chapters.
One molecule can do several jobs. In BEST’s own examples, PSA (prostate-specific antigen) is a monitoring biomarker, and a rising PSA during follow-up is a prognostic biomarker. Inherited BRCA1/2 variants appear as a risk biomarker, as a prognostic biomarker in women with breast cancer, and as a predictive biomarker for PARP inhibitor drugs in ovarian cancer. The category depends on the question asked, which BEST calls the context of use.
Two terms are often confused. A prognostic biomarker estimates the likelihood of an event, such as recurrence, in someone who already has the disease. A predictive biomarker identifies people more likely than others to benefit from, or be harmed by, a specific treatment.
Two families
Serum tumor markers and molecular biomarkers
NCI separates two families. Traditional tumor markers are proteins or other substances made in higher amounts by cancer cells than by normal cells. Genomic markers are tumor gene mutations, patterns of gene expression and other changes in tumor DNA, found in tumor tissue or in tumor fragments shed into body fluids.
| Feature | Serum tumor markers | Molecular biomarkers |
|---|---|---|
| Measured | A circulating protein, hormone or enzyme | Tumor DNA, RNA or protein expression |
| Sample | Blood | Tumor tissue; sometimes plasma |
| Method | Immunoassay (antibody-based test) | Sequencing, PCR or tissue staining |
| Main role | Monitoring treatment and recurrence | Selecting treatment (predictive) |
| Examples | CEA, PSA, AFP, CA-125, CA 19-9 | HER2, EGFR, BRCA1/2, MSI, NTRK fusion |
The split is a rule of thumb: KRAS can be tested in tumor or in plasma. See our liquid biopsy guide.
Classic markers
Classic serum markers: what each is for, and what it is not for
Guidelines endorse these markers mainly in people already diagnosed: to follow treatment and watch for recurrence.
| Marker | What it is | Used for | Not recommended for |
|---|---|---|---|
| CEA | Oncofetal glycoprotein | Follow-up after colorectal cancer surgery; monitoring advanced disease | Screening |
| PSA | Protease made by the prostate | Aid to diagnosis; response; recurrence | Screening at 70 or older (USPSTF grade D) |
| AFP | Oncofetal protein | With ultrasound, surveillance in people at high risk of liver cancer; germ cell tumors | General screening; liver metastases |
| CA-125 | Epitope on the MUC16 mucin | Pelvic mass work-up; response; recurrence | Screening women without symptoms (USPSTF grade D) |
| CA 19-9 | Sialyl-Lewis a carbohydrate | Monitoring pancreatic cancer treatment | Screening or early diagnosis |
| CA 15-3, CA 27.29 | MUC1 mucin epitopes | Breast cancer treatment response and recurrence (NCI) | Screening; routine recurrence testing is disputed |
Screening. No protein tumor marker is recommended for cancer screening in the general population. PSA screening is an individual decision for men aged 55 to 69 (USPSTF grade C, 2018; an update was in progress on 1 October 2026). AFP is used with ultrasound only in people at high risk of liver cancer. CA-125 screening found more early-stage ovarian cancers in the UKCTOCS trial but did not reduce deaths.
Laboratory guidelines explain the caution. No serum marker in current use is specific for cancer, levels are rarely raised in early disease, and a value inside the reference range does not rule cancer out. Benign causes are common: CEA can rise with hepatitis, cirrhosis and smoking, and PSA with benign prostate enlargement and prostatitis.
Analytical issues
Why the same sample can give different numbers
- Reference standards. The World Health Organization (WHO) keeps reference materials for calibrating assays to a common unit. For CEA it is the 1st International Reference Preparation, code 73/601, set up in the mid-1970s and still the only one. For total PSA it is the 2nd International Standard, 17/100 (2018). For AFP it is the 2nd International Standard, 22/216 (2023), which replaced the exhausted 72/225.
- Assay-to-assay differences. A shared standard has not made results interchangeable. A study of four major CEA platforms on 94 patient samples found no consistency between them. For PSA, a 2026 paper reported that two WHO-calibrated assays differed by more than 20% in 74% of patients.
- Hook effect. At very high concentrations some immunoassays read falsely low. Laboratories guard against this by testing at more than one dilution.
- Interference. Some people carry antibodies that react with the animal antibodies used in a test (heterophile antibodies), giving falsely high or low results. Timing matters too: PSA rises after a prostate biopsy or a urinary catheter, and saliva contamination can raise CEA and CA 19-9.
So serial monitoring should stay on one method, and reports should name the assay. Trends matter more than single values: in a European study of 91 healthy adults sampled weekly for 10 weeks, every tumor marker showed marked individuality, so a person’s own baseline tells more than a population range. Being traceable to a WHO standard is not the same as being equivalent to another assay. For background see what is biochemical diagnostics and fluorescent lateral flow versus traditional testing.
Predictive testing
Companion diagnostics and tumor-agnostic approvals
A companion diagnostic is a test that gives information essential for the safe and effective use of a matching drug. In practice it measures a predictive biomarker. An independent tally counted 185 FDA companion diagnostic approval actions from 1998 to 2024: 64 new assays and 121 extensions. Most new assays used PCR (33%), tissue antibody staining (23%) or next-generation sequencing (22%).
Some drug approvals follow the biomarker, not the organ. A June 2025 review listed nine FDA tumor-agnostic indications built on six biomarkers: MSI-H/dMMR (faulty DNA mismatch repair), high tumor mutational burden, NTRK fusions, BRAF V600E, RET fusions and HER2 IHC 3+. The first was pembrolizumab for MSI-H/dMMR tumors in May 2017. Approvals after mid-2025 are not counted here.
Validation
How a biomarker is validated, and why most candidates fail
- Analytical validity. Does the test measure the marker accurately and precisely?
- Clinical validity. Does the result acceptably identify, measure or predict the clinical state of interest in the people the test is meant for?
- Clinical utility. Does using the test lead to a net improvement in health outcomes, or to information that usefully changes care?
Study design decides whether the answers can be trusted. The PRoBE design (prospective specimen collection, retrospective blinded evaluation) collects samples from people who represent the intended users, before outcomes are known. The REMARK guideline covers how prognostic marker studies should be reported.
Diamandis argued in 2012 that very few, if any, new circulating cancer biomarkers had reached the clinic in 30 years. The most common reason, he wrote, is a true finding that does not change a clinical decision; others are false discoveries from flawed sample handling or analysis and, rarely, fraud. Regulatory qualification is slow as well. By 1 July 2025 the FDA Biomarker Qualification Program had accepted 61 projects and fully qualified 8 biomarkers. Qualification is separate from clearing or approving a test.
Validity without utility is common. CA-125 screening shifted ovarian cancers to earlier stages in UKCTOCS without reducing deaths. Our early detection evidence hub explains these endpoints.
Outlook
What to watch
- Multi-cancer early detection (MCED). As of 1 October 2026 no MCED test is FDA approved. An FDA advisory panel backed the Galleri test on 23 September 2026 and a decision is pending. The NHS-Galleri trial missed its primary endpoint.
- PSA guidance and standardization. The USPSTF prostate recommendation was being updated on 1 October 2026. Decision thresholds stay tied to the assay behind them.
- Quantitative testing near the patient. Reader-based formats, described in how lateral flow assays work, face the same three hurdles.
OncoFirm™ roadmap
Where OncoFirm™ fits
OncoFirm™ is developing quantitative fluorescent lateral flow assays for established serum markers. OncoFirm™ makes no screening or multi-cancer claims. Its assays, reader and software are in development, are not cleared or approved by the FDA or any other regulatory authority, are for research use only and are not for sale.
CEA assay (in development)
Designed range 1 to 100 ng/mL, with planned traceability to WHO 73/601. See the CEA guide.
PSA assay (in development)
Designed range 0.5 to 50 ng/mL, with planned traceability to the WHO 2nd International Standard 17/100. See the PSA guide.
Analytical validity first
Method comparison, precision, hook-effect and interference studies come before any clinical claim. The platform has a development target of a result in 20 minutes or less.
Pipeline and concepts
AFP is a future pipeline candidate. Thomsen-Friedenreich (TF) antigen assay concepts are concept-stage.
OncoFirm™ assays, reader and software are in development, have not been cleared or approved by the FDA or any other regulatory authority, are for research use only and are not for sale. Designed ranges and result times are development targets, not validated performance claims.
FAQ
Frequently asked questions
What is a cancer biomarker?
A cancer biomarker is a measurable characteristic, such as a protein in blood, a gene change in a tumor or an imaging finding, that indicates normal biology, disease or response to treatment. The FDA-NIH BEST Resource groups biomarkers into seven categories by use: susceptibility or risk, diagnostic, monitoring, prognostic, predictive, response and safety.
What is the difference between a prognostic and a predictive biomarker?
A prognostic biomarker estimates the likelihood of an event such as recurrence in someone who has the disease. A predictive biomarker identifies people who are more likely to benefit from, or be harmed by, a specific treatment.
Can a tumor marker blood test screen for cancer?
No protein tumor marker is recommended for cancer screening in the general population. PSA screening is an individual decision for men aged 55 to 69 (USPSTF, 2018), and AFP is used with ultrasound only in people at high risk of liver cancer. CA-125 screening did not reduce ovarian cancer deaths in the UKCTOCS trial.
Why do tumor marker results differ between laboratories?
Assays from different manufacturers can give different numbers for the same sample, even when traceable to a WHO standard. In one comparison, two WHO-calibrated PSA assays differed by more than 20% in 74% of patients. Serial monitoring should stay on one method.
Does OncoFirm™ sell a cancer biomarker test?
No. OncoFirm™ CEA and PSA assays, the reader and the software are in development, are not cleared or approved by the FDA or any other regulatory authority, are for research use only and are not for sale.
Cancer Atlas 2026
More in the Atlas
Related OncoFirm™ guides
Sources
References
- FDA-NIH Biomarker Working Group. BEST (Biomarkers, EndpointS, and other Tools) Resource: glossary and category chapters. NCBI Bookshelf; glossary updated 16 January 2025. www.ncbi.nlm.nih.gov/books/NBK338448/
- National Cancer Institute. Tumor Markers fact sheet. Reviewed 7 December 2023. www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-fact-sheet
- National Cancer Institute. Tumor Marker Tests in Common Use. Reviewed 7 December 2023. www.cancer.gov/about-cancer/diagnosis-staging/diagnosis/tumor-markers-list
- Duffy MJ, McGing P. Guidelines for the Use of Tumour Markers, 5th edition. Association of Clinical Biochemists in Ireland; 2018. acbi.ie/wp-content/uploads/2022/12/1644913336-1602832758-Tumour-markers-5th.pdf
- Duffy MJ. Tumor markers in clinical practice: a review. Medical Principles and Practice. 2013;22:4. karger.com/mpp/article/22/1/4/203271/Tumor-Markers-in-Clinical-Practice-A-Review
- Locker et al. ASCO 2006 update on tumor markers in gastrointestinal cancer. Journal of Clinical Oncology. 2006. ascopubs.org/doi/10.1200/JCO.2006.08.2644
- US Preventive Services Task Force. Prostate Cancer: Screening. 8 May 2018. www.uspreventiveservicestaskforce.org/uspstf/recommendation/prostate-cancer-screening
- US Preventive Services Task Force. Ovarian Cancer: Screening. 13 February 2018. www.uspreventiveservicestaskforce.org/uspstf/recommendation/ovarian-cancer-screening
- Menon et al. UKCTOCS long-term mortality results. The Lancet. 2021;397(10290):2182-2193. www.thelancet.com/journals/lancet/article/PIIS0140-6736(21)00731-5/fulltext
- NIBSC. Carcinoembryonic Antigen (CEA), Human, 1st International Reference Preparation, code 73/601. nibsc.org/documents/ifu/73-601.pdf
- NIBSC. Prostate Specific Antigen (human), total, 2nd WHO International Standard, code 17/100. nibsc.org/documents/ifu/17-100.pdf
- NIBSC. Alpha-Fetoprotein, human, 2nd WHO International Standard, code 22/216. nibsc.org/documents/ifu/22-216.pdf
- Zhang et al. Comparison of CEA results across four immunoassay platforms. Clinica Chimica Acta. 2016;454:15-19. www.sciencedirect.com/science/article/abs/pii/S0009898115300991
- Filella X. Article on PSA assay standardization. Tumor Biology. July 2026. journals.sagepub.com/doi/10.1177/14230380261466593
- StatPearls. Chapter on tumor markers. NCBI Bookshelf NBK597378; updated 8 June 2026. www.ncbi.nlm.nih.gov/sites/books/NBK597378/
- Coşkun et al. Within- and between-subject biological variation data for tumor markers (EuBIVAS). Clinical Chemistry and Laboratory Medicine. 2022;60(4):543-552. discovery.dundee.ac.uk/en/publications/within-and-between-subject-biological-variation-data-for-tumor-ma/
- Shajarizadeh, Fazeli. Poster: tally of FDA companion diagnostic approval actions, 1998 to 2024. ISPOR; data cut-off 1 September 2025. www.ispor.org/docs/default-source/cti-meeting-21463-documents/c77019ae-3d66-4adf-af50-cd107ba644ee.pdf?sfvrsn=a1d5719b_0
- Tissue-agnostic biomarkers in solid tumors. Cancer and Metastasis Reviews. 27 June 2025. link.springer.com/article/10.1007/s10555-025-10274-2
- Pepe, Feng, Janes, Bossuyt, Potter. The PRoBE study design for biomarker evaluation. Journal of the National Cancer Institute. 2008;100:1432-1438. academic.oup.com/jnci/article/100/20/1432/900265
- McShane LM, et al. Reporting recommendations for tumor marker prognostic studies (REMARK). Journal of the National Cancer Institute. 2005;97(16):1180-1184. pubmed.ncbi.nlm.nih.gov/16106022/
- Diamandis EP. The failure of protein cancer biomarkers to reach the clinic: why, and what can be done to address the problem? BMC Medicine. 2012;10:87. bmcmedicine.biomedcentral.com/articles/10.1186/1741-7015-10-87
- Collins, Allen, Andrews, Navarro-Serer, Stewart. Hurry Up and Wait (analysis of the FDA Biomarker Qualification Program). Therapeutic Innovation and Regulatory Science. Online 26 October 2025;60(1):302-308. link.springer.com/article/10.1007/s43441-025-00889-6
- FDA advisory panel recommends approval of GRAIL Galleri multi-cancer blood test. STAT. 23 September 2026. www.statnews.com/2026/09/23/fda-advisory-panel-recommends-approval-grail-galleri-multi-cancer-blood-test/
- Annual Galleri screening reduced stage IV cancer diagnoses but missed primary endpoint in first randomized MCED trial. The ASCO Post. June 2026. ascopost.com/news/june-2026/annual-galleri-screening-reduced-stage-iv-cancer-diagnoses-but-missed-primary-endpoint-in-first-randomized-mced-trial/
About this article. Sourced from the FDA-NIH BEST Resource, NCI, USPSTF, WHO reference materials and peer-reviewed papers; some guideline positions were read through reviews. Part of the OncoFirm™ Cancer Atlas 2026, written by the OncoFirm™ Scientific Team from the sources linked above and reflecting public information as of 1 October 2026. It is for education and is not medical advice. OncoFirm™ has no affiliation with the companies or tests named. OncoFirm™ assays are in development, have not been cleared or approved by the FDA and are not available for sale.
Collaborate on the next edition
Clinicians, laboratories, researchers and industry partners can contribute data, corrections or validation studies. Collaboration inquiry · Investor inquiry