Tuesday, August 24, 2004

My Infatuation

Since I’ve changed the title of my blog from it’s original I’ve written very little about science, instead focusing mostly on my personal life. I’d like to try to share more of what I read and study, even if the audience is very small. At times I’m sure I get caught up in jargon, and may assume too much background, but I want the practice writing. So below is my first post where I really consider some science. I’ll try to make a science related post periodically.

I have an infatuation. It’s a rather nerdy, scientific one. I am fascinated by proteomics (the study of proteins) technique abbreviated as SELDI-TOF (Surface-Enhanced Laser Desorption-Ionsiation Time-of-Flight for those that are curious or hate acronyms). I was first introduced to it in the summer of 2002, when a researcher (Joe Paulauskis) from Pfizer came to talk to our class about drug discovery.

At first he talked about how drug companies are working at ways to understand what small genetic differences occur in the proteins that metabolize and dispose of drugs in our bodies. The differences are called SNPS (Single Nucleotide Polymorphisms), and they can affect how one person responds to a drug compared to another person. This was all compelling stuff, but what really raised my interest was when he started talking about applications of mass spectrometry in characterizing people’s response to drugs.

In the basics of SELDI-TOF are that one puts a compound to be analyzed on a small chip that has a surface that will bind more strongly to compounds that have a particular characteristic (for example a specific shape or the ability to bind to copper). A laser is used energize the compound and it’s surrounding materials to a point where they float into the air, and because they’ve acquired a magnetic charge in the process, they can then be pushed/pulled down a tube that measures the amount of time it takes for a particle to reach the end of the tube (hence the designation Time-of-Flight). The larger a particle in the compound is, the slower it flies, the longer the time of flight. Knowing this allows scientists to determine the makeup of a compound.

What truly excites me about the application of this tool is in the prediction of disease. Some authors (Petricoin and Liotta) think that by using blood samples from the people with or without certain types of cancer they can predict who has a tumor based on the particle sizes detected in SELDI-TOF. Since tumors secrete proteins and other small molecules into the blood, and the body responds by producing proteins and small molecules of its own, it may be possible to detect these proteins (or a pattern of expression among several) that offers a high sensitivity method for screening people.

In what I consider to be a very promising paper the published in the Lancet (Petricoin et al 2002 ), a method similar to what I described predicted ovarian cancer with a sensitivity of ~100% (it correctly predicted every incidence of ovarian cancer in the test set they used). Other current, accepted methods have a sensitivity of ~35%! The implication of such a powerful screening tool is that we could detect cancer at an earlier, possibly more treatable, stage.

The paper has it’s flaws. It’s statistical methods were weak, almost inappropriate. The paper was widely criticized for that. A paper by E. Diamandis (2004) astutely criticizes some of the features of SELDI-TOF applications to cancer prediction. I’m not entirely persuaded by Diamandis’ arguments against SELDI-TOF though. Some of his strongest statements aren’t necessarily backed by evidence. This is an emerging technology, still very young. We need time to describe it better, but it’s promise is great enough that I think it warrants further study.


Post a Comment

<< Home