What causes cancer? I am going to attempt to give the shortest, best, detailed description of cancer ever written. ;-) Ready?! Here goes…

Cancer is a breakdown in the integrity of the signaling pathways in a group of cells. Signaling pathways are collections of molecules that interact, working together in a sequence, that lead to execution of different routines in your genetic program, to control various cellular functions (e.g., cell division). Computer scientists can think of “signaling proteins” as procedure calls (there is a hierarchical structure to them, just as there is in a computer program).

Metaphorically, cancer is a problem with a cell's accelerator and brake. The “accelerator” is the set of signaling pathways that switch on cell growth (cell cycle progression). The “brake” is the set of signaling pathways that instruct a cell to self-destruct for the good of other cells (this is called “apoptosis”).

There are two types of signaling proteins in cancer pathways: tumor suppressors and oncogenes (these are normally called proto-oncogenes when they are functional, but I digress).

1. The distal impact a normal TUMORE SUPPRESSOR is to INHIBIT cell GROWTH. (Distal impact means the net effect, as opposed to the proximal, or immediate, impact on the signaling pathway).

2. The distal impact of a normal ONCOGENE is to PROMOTE cell GROWTH.

There are two types of mutations in pathways:

1. loss of function mutations turn a node in a pathway OFF, and

2. gain of function mutations result in the signal always being ON.

A Loss-of-Function mutation in a tumor suppressor reduces the ability of a cell to inhibit growth (the brake gets broken). A Gain-of-Function in an oncogene reduces the ability of a cell to stop promoting growth (the accelerator gets locked on). These two types of mutations, as they accumulate in a line of cells, lead normal cells down an evolutionary path toward cancer.

There can be thousands of mutations in any line of cancer cells, but usually there are only a dozen or so “driver” mutations. Driver mutations are the key mutations pushing the accelerator down (forcing cell cycle progression), and preventing the brake from being applied (blocking the cell’s ability to recognize it is too damaged, that its time is up and it needs to commit suicide).

Interestingly, any “cancer” is genetically very heterogeneous. As a growing ball of tumor cells loses the ability to spot the growing numbers of errors in the genome and get rid of them, different sub-populations in that ball emerge and follow different evolutionary paths. When you treat a “cancer” and “it” comes back, the main reason that is frequently very bad news is because you have killed off the “weak” cancers (you might say you have many cancers in that mass), making room for the really tough bastards to feed off the previously more limited food supply.

I think you can see where this is going.

The future of cancer treatment is identifying which of the more common driver mutations you have (by sequencing your “cancers”), and producing a “cocktail” of drugs that treat each of these mutations and drive “them” into remission.

I am skipping over some of the other pathways that contribute to cancer (e.g., metastasis and angiogenesis) but I think you get the general idea.