19.2 — The Hallmarks, and Why Each Is a Target

Every hallmark of cancer is a capability the cell had to acquire — which means every one is something the cell depends on, and therefore something that can be taken away.

That is the logic of targeted therapy, and it is why the last twenty years have produced drugs that would have been inconceivable when chemotherapy was the only option.

Sustained growth signalling

Normal cells divide only when instructed. Cancer cells generate their own instruction, or leave the receiver switched on.

How: producing their own growth factor; over-producing the receptor; or a mutation that activates the signalling pathway regardless of any signal.

And each has a matching drug.

HER2 over-expression in around 15 to 20 percent of breast cancers — trastuzumab, an antibody blocking the receptor (Chapter 1.8). It turned the worst-prognosis subtype into one of the more treatable ones.

EGFR mutations in lung cancer — oral inhibitors producing response rates around 70 percent, against around 30 percent for chemotherapy in the same patients.

BCR-ABL fusion in chronic myeloid leukaemia — imatinib, designed to fit the fused protein's active site (Chapter 2.7). Median survival went from around five years to near-normal life expectancy.

BRAF V600E mutation in melanoma and others — specific inhibitors producing rapid, dramatic responses.

And the pattern is consistent: find the driver, block it, and the tumour that depends on it collapses. The limitation is equally consistent: resistance emerges, because the tumour is an evolving population (Chapter 19.1).

Insensitivity to anti-growth signals

Tumour suppressors disabled.

p53 — mutated or disabled in more than half of all human cancers (Chapter 1.7). Its loss removes the checkpoint that would have detected the damage and ordered the cell to die.

Which is also why p53-deficient tumours resist chemotherapy and radiotherapy — both work by damaging DNA and relying on p53 to trigger apoptosis.

RB — the retinoblastoma gene, controlling entry into the cell cycle. CDK4/6 inhibitors act on the same pathway and are now standard in hormone-receptor-positive breast cancer.

Evading apoptosis

The cell refuses the order to die (Chapter 1.8).

Follicular lymphoma is the clearest example — over-production of Bcl-2 keeps the mitochondrial membrane shut, so the death signal never fires. The cells do not divide unusually fast; they simply do not die.

And venetoclax was designed to fit Bcl-2's binding pocket and release the block, with substantial effect in several leukaemias. A drug derived directly from the mechanism.

Limitless replication

Normal cells count their divisions through telomere shortening and stop at the Hayflick limit (Chapter 1.7).

Around 85 to 90 percent of cancers reactivate telomerase, which rebuilds the ends and removes the counter.

Telomerase inhibitors have proved difficult — partly because the effect takes many divisions to appear, and partly because stem cells also need it.

Angiogenesis

A tumour cannot grow beyond 1 to 2 millimetres without its own blood supply, because diffusion cannot reach further.

So it releases signals — principally VEGF — that recruit new vessels.

And the vessels it builds are abnormal: leaky, tortuous, and poorly organised, which is why tumours are frequently hypoxic despite being vascular, and why drug delivery into them is inefficient.

Anti-VEGF drugs — bevacizumab and others — produce modest benefit in several cancers, less than the elegance of the idea suggested.

And the same drug class, used in the eye, transformed wet macular degeneration (Chapter 11.11) — one of the clearer examples of a cancer drug finding a better use elsewhere.

Altered metabolism

The Warburg effect — tumours running glycolysis even with oxygen available.

It is inefficient in ATP terms (Chapter 1.6) and it generates the building blocks a rapidly dividing cell needs, which appears to be the point.

It is what makes PET scanning work (Chapter 16.6).

And it is why the "sugar feeds cancer" claim is a misreading (Chapter 19.1) — the observation is real; the dietary conclusion does not follow.

Metabolic targets are an active area — IDH inhibitors in some leukaemias and gliomas are the clearest success so far.

Evading immune destruction

And this is the hallmark that produced the largest therapeutic change.

Your immune system removes abnormal cells continuously (Chapter 13.2), and the evidence is direct: immunosuppressed transplant recipients have dramatically increased rates of specific cancers, particularly virus-driven ones (Chapter 13.8).

So a tumour that becomes clinically apparent has escaped.

How it escapes:

Reducing MHC class I display, so killer T cells cannot inspect it — and natural killer cells exist to close exactly that loophole (Chapter 13.1).

Recruiting regulatory T cells and suppressive macrophages.

And — the exploitable one — expressing checkpoint ligands.

T cells carry "off switches" — PD-1 and CTLA-4 — which exist to prevent autoimmunity. Tumours express the molecules that trigger them.

So the T cell arrives, recognises the tumour, and is switched off.

Checkpoint inhibitors block those switches, releasing a response that was already present.

James Allison and Tasuku Honjo shared the 2018 Nobel Prize for the work.

And the results in some cancers have been extraordinary — Chapter 19.7.

Genome instability

The enabling characteristic — it generates the mutations that produce all the others.

Caused by defects in DNA repair (Chapter 2.5).

And it is exploitable in two ways.

Synthetic lethality. BRCA-mutant cells cannot repair double-strand breaks by homologous recombination, so they depend on a backup pathway — and PARP inhibitors block the backup (Chapter 2.5). The tumour dies; normal cells, which still have one working BRCA copy, survive.

Immunogenicity. Tumours with defective mismatch repair accumulate enormous numbers of mutations, so they display large numbers of abnormal proteins — which makes them unusually visible to the immune system, and unusually responsive to checkpoint inhibitors (Chapter 2.2).

Which is why mismatch repair status is now tested routinely: a defect that makes the cancer worse also makes it treatable.

Tumour-promoting inflammation

Chronic inflammation supplies growth factors, blood vessels and mutagenic reactive species (Chapter 16.3).

And several cancers have inflammation as the direct route: hepatitis to liver cancer, H. pylori gastritis to gastric cancer, inflammatory bowel disease to colorectal cancer, and asbestos to mesothelioma.

Which is why treating the inflammation prevents the cancer — eradicating H. pylori, curing hepatitis C (Chapter 17.10), and controlling inflammatory bowel disease.

The tumour microenvironment

A tumour is not a pure mass of cancer cells.

It contains fibroblasts, immune cells, blood vessels and extracellular matrix, and a substantial proportion of the cells in a solid tumour are not malignant at all.

And the surrounding cells are recruited to help. Cancer-associated fibroblasts supply growth factors; suppressive immune cells protect the tumour; and the dense matrix in pancreatic cancer physically impedes drug delivery, which is one reason that cancer is so hard to treat.

Which reframes treatment: targeting the environment rather than only the cell.

Heterogeneity

And this is the practical obstacle to everything above.

A single tumour is not genetically uniform. Different regions carry different mutations, because the tumour evolved as a branching tree rather than a single line (Chapter 3.4).

Which has three consequences.

A biopsy samples one branch, and may miss the mutation that matters.

Targeted therapy kills the cells carrying the target and selects for those that do not — which is exactly why resistance is near-universal with single-agent targeted treatment.

And metastases can differ genetically from the primary, and from each other.

Liquid biopsy — sequencing tumour DNA circulating in blood — partly addresses this, by sampling shed DNA from all sites at once, and it is increasingly used to track resistance mutations as they emerge.

What this framework delivered

Before it, chemotherapy attacked division in general and hit every dividing cell (Chapter 1.7).

After it, treatment could ask which capability this particular tumour depends on — and block that.

Which is why cancer treatment is now preceded by molecular testing, and why two patients with histologically identical tumours may receive completely different drugs.

And why the classification is shifting from where a cancer started to what drives it. Some drugs are now approved for any tumour carrying a specific mutation, regardless of organ — the first genuinely tissue-agnostic approvals, and a real conceptual change.

What the next page fixes

Chapter 19.3 covers the capability that makes cancer lethal. Localised cancer is usually curable. Around 90 percent of cancer deaths are caused by metastasis, and what determines whether and where it happens is a genuinely interesting question.