A dividing cell doesn't just copy its DNA and split in two on a fixed schedule. It passes through a series of checkpoints first, each one a genuine pause point where the cell verifies specific conditions are met before it's allowed to proceed to the next stage — checking, among other things, that its DNA has been copied without damage and that its chromosomes are correctly attached to the machinery that will pull them apart. A cell that fails one of these checks is normally held in place, given a chance to repair the problem, or in some cases directed to destroy itself rather than divide with damaged genetic material intact.
Three checkpoints, each asking a different question
The cell cycle contains several distinct checkpoints, each interrogating a different aspect of the cell's readiness to proceed. One checkpoint, occurring before DNA replication begins, checks that the cell's DNA is undamaged and that conditions are generally favourable for division. A second checkpoint, after DNA replication, verifies that the copying was completed accurately and checks again for any damage that may have occurred during the process. A third checkpoint, during the stage when chromosomes are being separated, specifically verifies that every chromosome is correctly attached to the spindle apparatus responsible for pulling the two copies apart — a cell won't proceed with separation until every single chromosome reports a proper attachment, since even one chromosome pulled apart incorrectly can leave a daughter cell with the wrong number of chromosomes entirely.
A checkpoint that fails doesn't just cause an error, it removes a safeguard
What makes checkpoint failure so consequential isn't that it directly causes uncontrolled growth by itself, but that it removes the safeguard that would otherwise have caught and corrected a problem before it became permanent. A cell with damaged DNA that's allowed to divide anyway passes that damage on to two daughter cells instead of one, and if the damage happens to affect a gene that itself helps regulate the checkpoints, the daughter cells may have an even weaker ability to catch future problems than the parent cell did. This is a large part of why checkpoint proteins turn up so consistently across different types of cancer, arising from otherwise very different original mutations: cancer isn't usually caused by a single damaged gene directly forcing runaway growth, but by an accumulating loss of the checkpoint safeguards that would normally have caught and stopped that damage at an earlier stage, long before it could compound into something more serious.
What we're still unsure about
The existence of these checkpoints and their basic mechanisms are extremely well established, confirmed through decades of cell biology research using techniques that can directly track individual checkpoint proteins in living cells. What remains a genuinely active area of research is exactly how the many different checkpoint pathways interact and compensate for each other in a real, complex tumour, since a cancer cell rarely loses just one checkpoint safeguard in isolation — it typically loses several over time, in a sequence and combination that can vary considerably between different cancers, which is part of why a treatment that successfully restores checkpoint function in one type of cancer often doesn't transfer straightforwardly to another.
This sits inside Cell Cycle Regulation & Cancer, one of seven topics in Cell Biology, one of six domains in Biology, one of seventeen subjects the app can quiz you on.