Early Marker

Your body knows first. Your blood holds the story.

We train AI models to read the response your body mounts to cancer and complex disease, written in a single tube of blood. Detection moves earlier.

A tiny oxblood tumor sits at the center of a much larger field of simplified immune and stromal forms whose white echoes spread across the page.

The body is not the background. It is the measurement.

A tumor begins in one place, but its biology does not stay there. It reshapes nearby tissue, changes cells across the body, and leaves a coordinated RNA signal in blood.

  • Tactile relief of one new blood vessel connected to a tumor while two sprouts continue growing toward it from an open parent vessel.

    01 ⁄ 06

    New vessels form

    A growing tumor demands blood. It signals new vessels into being, wiring its own biology straight into the circulation.

  • Tactile relief pairing quiet and reprogrammed T-cell, myeloid, and dendritic immune states.

    02 ⁄ 06

    Immunity changes state

    The immune system reads the tumor and shifts — T cells, myeloid and dendritic cells change program. Circulating immune RNA carries it.

  • Wide tactile relief of three fibroblasts reorganizing and aligning collagen across an extracellular matrix field.

    03 ⁄ 06

    The matrix remodels

    Cells around the tumor rebuild collagen and connective tissue. The enzymes that cut and stiffen the matrix release their signal into blood.

  • Wide tactile relief of neighboring cell groups relaying fuel from a glycolytic state into mitochondria-rich metabolism.

    04 ⁄ 06

    Metabolism reshapes

    Low oxygen and fast growth rewire how tissues handle glucose, lipids, and cholesterol — pressure felt far beyond the tumor itself.

  • Wide tactile relief of fibroblast, immune, and stromal responder cells releasing distinct messages across an endothelial boundary into flowing blood.

    05 ⁄ 06

    Messages enter circulation

    Cells around the tumor release growth factors, cytokines, lipids, and vesicles into blood — carrying the surrounding tissue’s response through circulation.

  • Wide tactile relief of a continuous field of hepatocytes coordinating changes in protein, lipid, and clotting-factor production.

    06 ⁄ 06

    The liver responds

    The liver retunes the proteins, lipoproteins, and clotting factors it releases into plasma — a distant organ answering a local tumor.

The tumor is local. Its effects are not.

Circulating cells carry RNA — a running record of their state and the signals reaching them. Early Marker reads the coordinated pattern across these programs at once, not any single marker. That pattern is what moves when cancer is present.

How it works

One blood draw. Thousands of messages. One pattern.

One platform turns a single draw into a cancer readout: formulate fresh blood into an RNA sample that concentrates the disease-relevant signal, read across the transcriptome, and train a model to find the coordinated pattern.

  1. Fresh blood, in full
  2. Formulate the sample
  3. Read total RNA
  4. Learn the signature
  5. Detect the signal

Fresh blood, in full

Watercolor study of one capped tube of fresh whole blood, the starting sample.

One fresh tube of whole blood is the starting material, before its components and signals begin to shift.

Formulate the sample

Watercolor comparison flowing from a raw analyte field into a representative rebalanced mixed sample.

Our method selects which blood-derived material to read and prepares its RNA so the disease-relevant transcripts are concentrated rather than buried.

Read total RNA

Abstract watercolor study of one formulated sample unfolding from left to right into many transcriptomic measurement lanes.

Total RNA is prepared and quantified across the whole transcriptome, turning the sample into a molecular readout.

Learn the signature

Watercolor study of seven RNA streams passing left to right through a four-layer network of painted nodes, with a few weighted paths picked out in oxblood, converging on six distinct outcome shapes.

A neural network trains on these readouts to learn the coordinated RNA signature of each cancer.

Detect the signal

Abstract watercolor field with a distributed oxblood ridge running left to right through the detected coordinated signal.

The trained model finds that signature in a new blood sample and produces the cancer readout.

Why it matters

Found early, cancer is a different disease.

The earlier a kidney tumor is found, the more often treatment is one kidney-sparing operation rather than years of systemic therapy.

Stamped-watercolor map of a kidney with one small tumor confined to the upper cortex. Early-stage tumor

93.6%

five-year survival · localized

Small and kidney-confined, up to 4 cm. Most are removed by kidney-sparing surgery, and the smallest can simply be watched.

Stamped-watercolor map of a kidney with a large regional tumor compressing and distorting the upper organ.

77.6%

five-year survival · regional

Larger, or reaching the renal fat or a vein. Usually the whole kidney comes out, and high-risk disease can add a year of immunotherapy.

Stamped-watercolor map of a kidney with locally invasive tumor, renal-vein extension, and separated metastatic deposits.

20.3%

five-year survival · distant

In the vena cava, nodes, or distant organs. Surgery turns complex or selective, and treatment becomes drug combinations across multiple lines.

Pipeline

One platform. Many diseases.

Swipe across the landscape →

Cancer burden and distant-stage survival landscape Major solid cancers are positioned by share of United States cancer diagnoses on the horizontal axis and five-year relative survival after distant-stage diagnosis on the vertical axis. Bubble area represents estimated new cases in 2026. Kidney and lung are lead programs; pancreas, breast and prostate are expansion programs; every cancer type remains on the roadmap. Thyroid: 2.1% of new cases; 45,240 estimated cases; 48.3% distant-stage five-year relative survival. Thyroid Melanoma: 5.3% of new cases; 112,000 estimated cases; 34.0% distant-stage five-year relative survival. Melanoma Ovarian: 1.0% of new cases; 21,010 estimated cases; 31.5% distant-stage five-year relative survival. Ovarian Uterine: 3.2% of new cases; 68,270 estimated cases; 19.9% distant-stage five-year relative survival. Uterine Colorectal: 7.5% of new cases; 158,850 estimated cases; 16.9% distant-stage five-year relative survival. Colorectal Bladder: 4.0% of new cases; 84,530 estimated cases; 9.6% distant-stage five-year relative survival. Bladder Stomach: 1.5% of new cases; 31,510 estimated cases; 8.1% distant-stage five-year relative survival. Stomach Esophageal: 1.1% of new cases; 22,530 estimated cases; 5.3% distant-stage five-year relative survival. Esophageal Liver and intrahepatic bile duct: 2.0% of new cases; 42,340 estimated cases; 3.6% distant-stage five-year relative survival. Liver Prostate: 15.8% of new cases; 333,830 estimated cases; 40.1% distant-stage five-year relative survival. Expansion program. Prostate 40.1% Female breast: 15.2% of new cases; 321,910 estimated cases; 33.8% distant-stage five-year relative survival. Expansion program. Breast 33.8% Pancreatic: 3.2% of new cases; 67,530 estimated cases; 3.4% distant-stage five-year relative survival. Expansion program. Pancreas 3.4% Kidney and renal pelvis: 3.8% of new cases; 80,450 estimated cases; 20.3% distant-stage five-year relative survival. Lead program. Kidney 20.3% Lung and bronchus: 10.8% of new cases; 229,410 estimated cases; 10.5% distant-stage five-year relative survival. Lead program. Lung 10.5%
Cancer burden and distant-stage survival landscape The same landscape arranged for a tall screen: major solid cancers are positioned by five-year relative survival after distant-stage diagnosis on the horizontal axis and share of United States cancer diagnoses on the vertical axis. Bubble area represents estimated new cases in 2026. Kidney and lung are lead programs; pancreas, breast and prostate are expansion programs; every cancer type remains on the roadmap. Thyroid: 2.1% of new cases; 45,240 estimated cases; 48.3% distant-stage five-year relative survival. Thyroid Melanoma: 5.3% of new cases; 112,000 estimated cases; 34.0% distant-stage five-year relative survival. Melanoma Ovarian: 1.0% of new cases; 21,010 estimated cases; 31.5% distant-stage five-year relative survival. Ovarian Uterine: 3.2% of new cases; 68,270 estimated cases; 19.9% distant-stage five-year relative survival. Uterine Colorectal: 7.5% of new cases; 158,850 estimated cases; 16.9% distant-stage five-year relative survival. Colorectal Bladder: 4.0% of new cases; 84,530 estimated cases; 9.6% distant-stage five-year relative survival. Bladder Stomach: 1.5% of new cases; 31,510 estimated cases; 8.1% distant-stage five-year relative survival. Stomach Esophageal: 1.1% of new cases; 22,530 estimated cases; 5.3% distant-stage five-year relative survival. Esophageal Liver and intrahepatic bile duct: 2.0% of new cases; 42,340 estimated cases; 3.6% distant-stage five-year relative survival. Liver Prostate: 15.8% of new cases; 333,830 estimated cases; 40.1% distant-stage five-year relative survival. Expansion program. Prostate 40.1% Female breast: 15.2% of new cases; 321,910 estimated cases; 33.8% distant-stage five-year relative survival. Expansion program. Breast 33.8% Pancreatic: 3.2% of new cases; 67,530 estimated cases; 3.4% distant-stage five-year relative survival. Expansion program. Pancreas 3.4% Kidney and renal pelvis: 3.8% of new cases; 80,450 estimated cases; 20.3% distant-stage five-year relative survival. Lead program. Kidney 20.3% Lung and bronchus: 10.8% of new cases; 229,410 estimated cases; 10.5% distant-stage five-year relative survival. Lead program. Lung 10.5%

Start with the hardest outcomes. Scale to the broadest burden.

  • Lead Kidney · Lung
  • Expanding Pancreas · Breast · Prostate
  • Roadmap Every cancer type

Major solid tumors with comparable staging shown. Sources: NCI SEER, 5-year relative survival, 2016–2022; 2026 U.S. case estimates.

The company

Born from research at UT Southwestern.

Early Marker builds one platform that reads the body’s response to disease in blood.

We bring molecular biology, patient data, and machine learning under one roof, aimed at a single idea: the body’s response to disease is written in blood, and reading it can move detection earlier — in cancer first, and beyond.

Meet the team ↗

Collaborators · partners · team

Build it with us.