Why is cancer so hard to cure?

It seems to happen every other week: Compound X found to “cure cancer” in a cell culture, or in rats, or even in chimps. Chain emails and overzealous reportage combine with some genuine academic incentives to exaggerate, and result in a medical climate in which cancer seems to be one of the most curable diseases in history. And yet, as AIDS and other terrifying diseases slowly give way to the march of modern science, cancer remains intractable and deadly. What makes cancer so different?
The biggest problem, by far, turns out to simply one of language. When we refer to HIV, polio, or the mumps, we’re referring to a fairly narrow spectrum of organisms. There will always be some variation between individual strains, but for the most part a person with polio has polio, and that relative uniformity has all manner of useful implications for treatment. However, cancer is not a disease, but type of disease. In terms of treatment and effect, a tumor in the brain has little in common with one in the lung or on the skin. All they share in their base-level problem: the cells that make up the tumor grow infinitely and without restriction. Beyond that, they are totally different diseases.

This should really say “Fight for the Cures” but that’s harder to explain.

That’s why the phrase “cure for cancer” is nonsense; there will not be any single cure for cancer because there is no single cause of cancer. Even total control over the reproduction of every cell won’t be enough, since the drugs that could exert such control still have to reach the tumor, enter the cells, and begin their action. Liver cancers present unique problems to do with that organ’s filtering of the blood, while leukemias (blood cancers) present their own  challenges. Two similar brain cancers can be caused by entirely different mutations in entirely different genes.
Remember that (with some very specific exceptions) we don’t catch cancer from one another , but develop it on our own each and every time — every cancer patient is the creator of their own disease. Though similar cancers generally arise from similar genetic defects, even slight differences can have powerful implications for treatment.
The unchecked replication of cancer cells is arguably the natural state of all life — this situation where healthy skin cells grow to a certain point then voluntarily stop is a tenuous agreement that allows complex organisms to survive but which works against the short-term goals of each individual cell type. In a healthy person, each cell type agrees to hold back on growth in the short term so the community can survive in the long term. Cancer is when cells break this social contract, and anything that mutates DNA — from radiation to heavy metals — can provide the opportunity to do so.
That’s why genomic analysis and quick, inexpensive sequencing of DNA is so important to cancer research; since every tumor is different, it’s vitally important to be able to characterize each individual type. A complete catalog of cancer-causing mutations (which would be very long and constantly expanding) would allow doctors to tailor a treatment regime to each instance of the disease. In a very real way, every individual needs their own tailor-made “cure for cancer” because every one has (or could have) their own unique type.