I was reading Siddharth Mukherjee's book "The Emperor of All Maladies: A Biography of Cancer" and started thinking about currently used cancer therapies. First, a little background on cancer. Cancer is uncontrolled growth of your body's own cells caused by a single or set of mutations in the cell's DNA. In normal cells, before a cell divides, it needs to pass a checkpoint process where certain proteins determine if the cell is in a viable state (eg. functioning normally without any serious mutations). There are many complex pathways involved in this process and we're still understanding this process. If the cell isn't in a viable state, the cell will commit suicide (apoptosis). Cancer occurs when this checkpoint process fails and rather than the cell committing suicide, it continues to divide. As I mentioned, there may be more than one mutation involved for a cell to become cancerous, so it may take many generations before the cell acquires the right set of mutations to make it malignant.
Most of our strategies for fighting cancer, whether it's chemotherapy or radiation, involve introducing further mutations in the cell so that it's DNA is so screwed up that it can't divide anymore (some drugs interfere with cell division and other things to stop it from dividing). But in general, we try to screw up the cell further so that the machinery for cell division doesn't work. The therapies are usually toxic to other cells so there is collateral damage and we localize treatments to minimize the effects of friendly fire.
Now, instead of doing this, why not try to fix the malignant cells by expressing normal versions of checkpoint genes in them. Once these genes are in the cell, their protein products should cause malignant cells to fail the checkpoint at cell division and thus die. I know many of you are reading this and are thinking "OK, this is impossible with our current repertoire of biotechnology tools". And you're probably right, this isn't easy and may well be impossible now. But suppose we find a way to do this. In theory, the therapy can be administered at a systemic level allowing us to wipe malignant cells from a specific cell line anywhere in the body with minimal collateral damage (assuming the normal checkpoint genes or payload method doesn't damage normal cells).
The issues. First, we don't have a complete picture of all of the genes involved in the checkpoint process so we don't know what mutations need "fixing". Secondly, how do we construct the gene payload? We need to ensure that we include the appropriate transcriptional elements in addition to the genes (if we're using cDNA or maybe use mRNA so we only need to worry about translation -- but then the question is whether these checkpoint genes should be expressed in normal cells that receive the payload?). Also, we need to ensure that our DNA payload does not integrate into the genome through recombination events as we could damage the host cell's DNA. Next, how do we deliver the DNA payload (eg. viral vector, other methods)? There are probably heaps of other issues preventing us from doing this but maybe one day, we'll have the technology to do this.
My point with this rant is that maybe we're approaching cancer the wrong way. Instead of trying to further damage cancerous cells, let's try to fix them.
Thoughts of a dreamer (or crazy person ;)).
