Cancer: When Cell-cycle Control Fails

The G₁/S checkpoint asks three questions before the cell commits to copying its DNA: is the cell large enough, is the DNA undamaged, and are growth signals present. The G₂/M checkpoint asks whether replication finished without errors and whether resources suffice for mitosis. In M phase the spindle assembly checkpoint holds division until every chromosome is correctly attached to the spindle.

The molecular basis of this control became visible when Robert Weinberg's laboratory at MIT isolated a mutated ras oncogene from bladder cancer cells in 1982, proving that a cancer cell is a normal cell carrying specific, identifiable mutations. Varmus and Bishop had already shown that proto-oncogenes are normal genes, and later sequencing projects catalogued thousands of driver mutations across cancer types.

Three oncogenes are worth memorising. RAS is mutated in roughly 30% of all cancers: a single amino-acid change locks the RAS protein in its active, GTP-bound state, so it signals division whether or not a growth factor is present. HER2 is amplified in about 20% of breast cancers, flooding the cell membrane with growth receptors. MYC is an amplified transcription factor that switches division programs on.

A common error is writing that oncogenes cause cancer by producing too much protein. The real problem is regulation, not quantity: the mutated protein is constitutively active, permanently switched on regardless of normal signals. A RAS protein stuck in its active form drives division continuously, and the normal allele cannot override it, which is exactly why one mutant copy is enough.

TP53 is mutated in roughly half of all human cancers, making it the most commonly altered gene in cancer biology. RB1, the retinoblastoma gene, blocks entry into S phase until conditions are right. BRCA1 and BRCA2 repair double-strand DNA breaks, APC restrains cell growth in the colon, and CDKN2A encodes p16, a checkpoint protein lost in many melanomas. Each brake has a distinct job.

Retinoblastoma is the cleanest evidence for Knudson's two hits. Children with the inherited form carry one faulty RB1 allele in every cell and develop multiple tumours in both eyes, early in life. Children with the sporadic form need both hits by chance in the same cell, so they develop a single tumour in one eye, usually later. One gene, two mutation patterns, one prediction confirmed.

Human papillomavirus strains 16 and 18 cause about 70% of cervical cancers. The virus inserts its DNA into the host genome and makes two oncoproteins. E6 binds p53 and targets it for destruction, so damaged cells lose their repair-and-apoptosis safeguard. E7 inactivates the RB1 tumour suppressor, removing the brake on entry into S phase. One infection disables two of the cell's most important brakes.

This mechanism is exactly why vaccination prevents the cancer. Australia introduced free Gardasil vaccination against HPV 6, 11, 16 and 18 for girls in 2007 and extended it to boys in 2013. Precancerous cervical lesions have fallen sharply, and Australia is projected to be the first country to eliminate cervical cancer as a public health problem, defined as fewer than 4 cases per 100,000 women per year.

Australia has the highest skin cancer rates in the world. About 2 in 3 Australians will be diagnosed with some form of skin cancer by age 70, including roughly 16,000 new melanomas and more than 800,000 non-melanoma skin cancers each year. The cause is ultraviolet radiation: UVB drives C-to-T mutations in tumour suppressor genes such as CDKN2A and TP53 in skin cells.

Melanoma shows the multi-hit model within a single disease. An activating BRAF mutation (V600E, an oncogene found in about half of melanomas) is often the first driver, frequently after intense intermittent sun exposure. Later hits in the CDKN2A and PTEN tumour suppressors disable further checkpoints until a cell can invade and metastasise. Each hit removes another layer of control.

Knowing the mechanism changed treatment. BRAF inhibitor drugs such as vemurafenib and dabrafenib target the mutant BRAF protein directly, while immune checkpoint inhibitors such as pembrolizumab and nivolumab block the PD-1 pathway that melanoma cells use to hide from immune attack. Five-year survival for metastatic melanoma rose from under 10% in 2010 to about 50% by 2023.

Each metastasis step needs specific molecular equipment. Local invasion requires losing the adhesion protein E-cadherin and gaining proteases such as matrix metalloproteinases that digest the surrounding tissue matrix. Entering a vessel demands surviving without attachment, because normal detached cells self-destruct through a process called anoikis. Cancer cells that resist anoikis can survive the journey.

Circulation acts as a filter: most roaming cancer cells are destroyed by shear forces or immune cells, and the survivors often hide inside platelet clusters. At a distant capillary the cell arrests, squeezes out of the vessel and colonises new tissue, inducing angiogenesis, the growth of new blood vessels, to feed a secondary tumour. The whole process is inefficient, but rare successes are lethal.

Where cells land follows blood flow and lymphatic drainage. Bowel cancer commonly seeds the liver via the portal circulation, lung cancer spreads to the brain and adrenal glands, and breast cancer reaches bone, liver, lung and brain. Once multiple sites are involved, treatment must address the whole body at once, which is why metastatic disease is rarely curable.