Monday, December 27, 2010

Asking the Right Question

I am rereading Steven Rosenberg's book from 1992, The Transformed Cell, on using the human immune system to attack cancer. We had the blizzard so I thought I'd go thru the stacks and clean out things, usually do this once a year, and I ran across this one. Rosenberg is a true intellect in cancer research and I have been following his work for a few decades. The book is worth a read, again, and there are few books that I do this with. As an aside I have read Norbert Wiener's autobiography about five times, and still remains something that I would go back to.

Now to Rosenberg, the issue is clearly one of asking the right question. He tells the time when he started focusing on IL-2 that his wife, always great to have a helpful spouse, always good to put a plug in for mine as well, how his wife after his discussion with her turned to him and said, "It's five ifs and a then..." Brilliant, I probably did not think so twenty years ago but I have used a variant of this many times in the past twenty years. How many ifs is a part of the question. In a start up company we can deal with two or three ifs, and in research we can deal with five. The key is asking the questions getting to the ifs. This is almost a Wittgenstein type approach.

Now to asking the right question. I have been looking at clinical data, I don't do any myself, other than plants, and I saw the following. There is a putative causal relationship between prostatic intraepithelial neoplasia, PIN, and prostate cancer, PCa. Namely PIN was always considered a precursor, almost. Every paper you read says so. But, and that is an important but, there are more than a handful of times when a patient goes from PIN to totally benign, and no PCa.

Why? If the process is genetic, then a change has occurred, a methylation and possible change in PTEN, but  what happened. Did the cells go thru apoptosis, was the immune system activated, where did these cells go? Was there some exogenous factor which caused the change. Can we create a few ifs as did Mrs. Rosenberg to model what happened and then try and determine if it did. Then can we reverse engineer it? Always goo to read the words of great people from time to time, helps clarify one's thoughts. And Rosenberg has done some great work.