Oil Droplet Navigates Maze

[Oniya]

http://news.sciencemag.org/sciencenow/2010/01/13-04.html

So, how do you get the drugs to the tumor?  You teach it to run a maze.

[Muninn]

That is amazing! Science is flippin' awesome.

[DarklingAlice]

Full citation:
<a href=http://pubs.acs.org/doi/abs/10.1021/ja9076793>Maze Solving by Chemotactic Droplets</a>
Istvn Lagzi, Siowling Soh, Paul J. Wesson, Kevin P. Browne and Bartosz A. Grzybowski
Department of Chemical and Biological Engineering and Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208-3113
J. Am. Chem. Soc., 2010, 132 (4), pp 1198–1199

To start off with, this is cool. I in no way deny that it is a visually impressive demonstration of an elegant combination of statistical and chemical principles. They should be applauded for their work and I am certain that this will have interesting applications. However, cancer therapy is not one of them. Not without vast alterations to the basic principles of this experiment. Here is why:

1) The claim about cancer and acidity is deceptive. The PH of cancerous cells fluctuates wildly based on the type of tissue they originated in, whether angiogenisis has occurred, whether they are using oxphos or have switched to the Warburg effect (aerobic glycolysis), the rate of growth, etc. While it is generally lower than that of average human tissue there are acidic hotspots in the body (mostly associated with metabolic and excretory systems), as well as varying PH levels in other foreign invaders. The introduction of any other source of acid would render the method outlined in the article untenable as it would skew the path of the oil droplet wildly (which seems quite likely to be dangerous, the reason that it is so important to target cancer drugs precisely is that they tend to be nasty).  In addition your body PH level can vary based on exercise due to anaerobic glycolysis and a number of other factors, meaning that the relative strength of the pull would fluctuate as well. Also the internal PH of a tumor is usually identical to that of normal cells, it is the external PH that is lower (cancer cells tend to dump metabolic byproducts into their surroundings because they would lose viability if they became too acidic internally) meaning that even if this did work the process would only get the drugs into the general area of the tumor and it would require a secondary mechanism to 'land' as it were and infiltrate the cells.

2) The procedure outlined in the paper is a very simple 2d limited random walk with only one element to skew the preference towards any particular path. Within the body you have two situations A) the drug molecule is small enough to either locally or generally diffuse, in which case the entire nature of the problem changes as you add a third dimension and the interplay numerous diffusion factors aside from PH; or B) the molecule is large enough that it must travel in the circulatory system, in which case physical motions, currents, etc. will overpower the PH attraction. Further the experiment as it is carried out only deals with motion through a homogeneous medium which is too idealized. Basically this model is far, far too limited to predict what would occur within a living body. To put it another way, to claim that this one source of attraction will let drug molecules find cancer cells is rather like building a device that lets you echo-locate in room filled with statues, and saying that because of that you can navigate a rave with it. You can't. There is too much noise.

It's a cool experiment, a great example of statistical mechanics, but to say that it has applicability to cancer is to ignore a great deal of biology for the sake of optimism.