Two articles from a recent issue of nature (here and here) raise the issue of how genetic diversity may affect anti-tumour chemotherapies. A nice overview of the significance of these articles in the grander scale is also provided here.
The two papers mention research that could be included in the nascent field of pharmacogenomics in which scientists study how genetic variation (like the one found in a typical tumour) affects the response to a drug in terms of efficacy or toxicity. And of course it seems that this diversity has a huge and rather negative impact on the efficacy of the drugs. When a particular cancer gene can have its effect duplicated by a mutation in a different gene then a therapy that targets cells with the first type of mutation are effectively selecting for cells with the second kind. This is not necessarily a bad thing but at least should be taken on account before pursuing any option. If selecting for a tumour composed by this second alternative mutation leads to a tumour that is effectively easier to treat then the treatment is a good one. But it could also happen that things go the other way and a drug helps the tumour evolve towards malignancy.
One way to stop tumour (somatic) evolution is to target more than one mutant simultaneously, although this also makes it more likely that there will be side effects that will be felt by healthy cells. A different approach has been announced recently. This approach fights evolution with evolution with a virus that helps to fight brain tumour cells. This is an exciting development that is being currently used by researchers in Yale to treat very serious cases in a type of tumour, brain tumours, in which tumour growth happens very rapidly and in which surgery is very complicated. In any case one is left wondering what will stop the virus to evolve to attach healthier cells once the tumour cells it lived on start being scarce.
I read those Nature papers last week and thought they were excellent. The selection of drug-resistant tumour cells is obviously one of the biggest problems in cancer research. Have you read any of the recent literature on how the rate of brain metastasis seems to be increasing in patients treated with Herceptin? One theory is that there’s something about the evolution of resistance to the drug that predisposes the surviving cells to relocation and / or survival in the brain.
We’re just lucky that we haven’t developed any infectious tumours, like those poor Tasmanian devils. Imagine if the evolved drug resistant cancer cells were able to start the process again in a new host, instead of having to start from scratch in each individual. It would solve the overpopulation problem anyway!
This Cell paper describes another transmissible cancer, canine transmissible venereal tumor (CTVT). It’s a fascinating paper that we had picked for a Journal Club. The cancer originated from a single dog or wolf 200-2500 years ago and can now be found in at least 40 dog breeds from 5 continents.
@Cath — That is a very good point. If tumours could spread from person to person things could get difficult indeed. It is not easy since every cancer in a human starts the process from scratch, with the only tools available initially to healthy cells. Although, like many other evolutionary processes, given enough times and individuals on which to grow, chances are that eventually an infectious breed of tumours might emerge. Of course it could very well be that the scale of time needed is just gigantic.
I also guess that even if such infectious cancer appears in the not improbable remote future AND can invade a substantial number of different hosts (that is, us humans) it should also find an appropriate rate of aggressiveness if it wants to spread to more than a handful individuals and kill them before having the opportunity to infect new hosts.