How Cancer Spreads (Metastasis)

A tumour that stays in one spot is often curable with surgery. What kills is metastasis — when cancer cells escape and set up new colonies in distant organs. Metastasis causes roughly 90% of cancer deaths. Press play and watch the brutal five-step journey: cells at the tumour's edge lose their grip, chew through the surrounding matrix, grow their own blood vessels, squeeze into the bloodstream, ride the circulation where almost all of them die, and — very rarely — a lone survivor lands far away and grows a new tumour.

Try this: watch Into the bloodstream and count how many cancer cells burst apart mid-vessel — then switch on 💉 Anti-angiogenesis drug and see the tumour's new vessels wither and starve.

Diagram is illustrative — not to scale.
1 · Invadelose grip · dig out 2 · Angiogenesisgrow new vessels (VEGF) 3 · Intravasatesqueeze into the blood 4 · Circulatemost cells die here 5 · Colonisea metastasis grows Epithelium (normal cells) basement membrane Primary tumour (where it started) extracellular matrix Blood vessel circulation — a hostile ride (most cells die) intravasation Distant organ (e.g. the liver) extravasation & new colony Sentinel lymph node (sampled first to check for spread) lymphatic vessel the metastatic cascade travels one way → most cells fail; a few survive

Live metastasis readout

Metastatic colonies formed
0
a new tumour in a distant organ
Circulating tumour cells
0
in the bloodstream right now (CTCs)
Angiogenesis drive (VEGF)
0% — signal telling vessels to sprout
Matrix (ECM) integrity
100% — enzymes dissolve it during invasion
Cells surviving the trip (model)
— in reality far fewer than 1 in 10,000

What's happening

A local tumour divides but stays behind the basement membrane. Pick a scenario to watch it try to escape…
cancer cell red blood cell VEGF signal MMP enzyme

Real numbers: metastasis causes about 90% of cancer deaths; tumours can't grow past roughly 1–2 mm without new blood vessels (VEGF-driven angiogenesis); and in experimental models fewer than 0.01% of cells that enter the circulation ever form a metastasis. The moving cells, colony counts and the “survival” percentage in this diagram are an illustrative model tuned so you can see the events — not measured cell counts.


The Science in Plain Language

1. Metastasis — the part of cancer that actually kills

Here is the fact that reframes everything: a tumour sitting in one place is usually the treatable kind. Surgeons can cut out a lump; radiation can burn one spot. What makes cancer lethal is metastasis — the same tumour cells breaking away and building new colonies in the liver, lungs, bones or brain. By most estimates, about 90% of cancer deaths are caused by metastasis, not by the original tumour. That single number is why almost everything in cancer medicine — screening, staging, the words “caught it early” — is really about one question: has it spread yet?

2. Losing their grip (the epithelial-to-mesenchymal switch)

Healthy epithelial cells are glued to their neighbours like tiles in a floor, held together by an adhesion protein called E-cadherin (the gene is CDH1). A cell that stays stuck can't wander. In invasion, cells at the tumour's edge switch off E-cadherin and take on a loose, crawling, migratory personality — biologists call this the epithelial-to-mesenchymal transition (EMT). It's the same toolkit an embryo uses to build itself; cancer hijacks it. Once a cell lets go of its neighbours and grows a “foot,” it can move — the first thing you see when you switch to Invasion & angiogenesis.

3. Digging out (enzymes that dissolve the matrix)

Between and beneath your cells is the extracellular matrix (ECM) — a mesh of collagen and other proteins, with a tough sheet called the basement membrane separating tissue compartments. To invade, a cancer cell secretes matrix-degrading enzymes, chiefly matrix metalloproteinases such as MMP-2 and MMP-9, that snip type-IV collagen and open a tunnel. In the animation, the yellow sparks are those enzymes and the Matrix integrity meter drops as the mesh is chewed apart. Crossing the basement membrane is the moment a growth officially becomes invasive rather than “in situ.”

4. Building a supply line (angiogenesis and VEGF)

A tumour is greedy for oxygen and can't grow beyond roughly 1–2 mm on diffusion alone — the classic threshold described by researcher Judah Folkman. So it cheats: starved, low-oxygen tumour cells release VEGF (vascular endothelial growth factor), a chemical shout that makes nearby blood vessels sprout new branches straight toward the tumour. This is angiogenesis. Watch the green VEGF dots stream out and a new vessel grow inward. Those hastily built tumour vessels are leaky and disorganised — which, conveniently for the cancer, also makes them easy for tumour cells to slip into.

5. The deadly commute (intravasation and circulation)

To spread, a cell must intravasate — squeeze through a vessel wall into the blood or lymph — then survive the ride. The bloodstream is a brutal place for a lone tumour cell: it's battered by shear forces, has lost its anchoring (which normally triggers a self-destruct program called anoikis), and is hunted by immune cells, especially natural-killer cells. The result is staggering inefficiency. In experimental models, fewer than 0.01% — well under 1 in 10,000 — of the cells that enter the circulation go on to seed a metastasis. In the animation you'll see most cancer cells burst and vanish mid-vessel; that carnage is the honest picture. Clinically, the survivors can be detected as circulating tumour cells (CTCs) in a blood draw (a “liquid biopsy”), and higher CTC counts track with worse prognosis.

6. Seed and soil (why cancers spread to predictable organs)

A survivor eventually arrests in a small vessel far away, extravasates out into the tissue, and tries to grow. Whether it can depends on the neighbourhood. Back in 1889, Stephen Paget proposed the “seed and soil” idea: a tumour cell (the seed) only thrives where the local environment (the soil) suits it. That's why metastasis isn't random — it follows organ tropism: breast and prostate cancers love bone, colon cancer tends to seed the liver (its blood drains there first), and several cancers favour the lungs. Importantly, the new tumour keeps the identity of the original: breast cancer that spreads to bone is still breast cancer in the bone, not bone cancer — which is exactly why it's treated with breast-cancer drugs.

7. Dormancy — the sleeper cells

Landing somewhere is not the same as growing there. Most disseminated cells stall as single cells or tiny clusters and go dormant — alive but not dividing, invisible on scans, held in check by the immune system and an unfriendly niche. They can stay asleep for years or even decades. That is why a breast-cancer survivor can relapse eight years later: a sleeper cell finally woke up and found its footing. Understanding and keeping these cells asleep is one of the most active frontiers in cancer research.

8. The lymph nodes and the sentinel-node biopsy

Many solid tumours spread through the lymphatic system before they reach the blood, and the first node that drains a tumour — the sentinel lymph node — is usually the first place cells land. That's the logic behind the sentinel lymph node biopsy: surgeons inject a dye or tracer at the tumour, find the sentinel node it drains to, and remove just that one (or few) to check under the microscope. If it's clear, the deeper nodes almost always are too, which spares patients the swelling and nerve damage of removing every node. It's a beautiful example of using the body's own plumbing to read how far a cancer has travelled. (See the companion lymphatic-system visualization.)

9. What this means for treatment — and one big myth

The cascade is a map of where to intervene. Early detection matters most because a tumour removed before it has invaded and seeded is often cured — screening (mammograms, colonoscopies, skin checks) is aimed squarely at catching cancer before Step 5. Anti-angiogenesis drugs such as bevacizumab (an antibody that blocks VEGF) try to starve the tumour's blood supply — toggle the drug in the animation and watch the new vessels wither; in real patients the benefit is real but usually modest, which is itself an honest lesson about how tough this disease is. And the myth: many people fear that a needle biopsy or surgery will “make the cancer spread everywhere.” The risk of a biopsy seeding a tumour is real but extremely small, and the information a biopsy provides is essential to treating the cancer correctly. Delaying diagnosis out of that fear does far more harm than the biopsy ever could.

↑ Back to the animation

Connections