Thursday, February 24, 2011

Some Observations on the New Genomics for Cancer Screening

 

There has been a flurry recently about many entities trying to promote gene testing to ascertain cancer growth and staging. The classic one is the determining of whether a detected prostate cancer is indolent or not. A recent one describes a test by some Dana Farber scientists who are targeting certain genes. The article from Xconomy in Boston states:

...studied the roles of specific genes—including cyclin D1, SPP1, PTEN, and SMAD4—in both mouse models of prostate cancer and tissue samples from human patients with the disease, finding that the four genes were predictors of lethal tumors that are hardwired to spread to other organs in the body

Now there are several concerns that one may have about such a statement:

1. PTEN is lost after the cell has lost its ability to control the Akt pathway. Namely it is lost late in the game.

2. Cancer Stem Cell theory may apply to the prostate and thus one should be looking at stem cells only since that cell controls the entire process.

3. One should, nay must, understand the pathways involved. Not just the products.

4. Every cancer is most likely a little different. Some pathways are blocked and lost and some are altered.

5. The epigenetic factors of miRNA and methylation also add a dimension of complexity to the pathways as aberrant factors.

6. Sampling can be done or should be done on a cell by cell basis, not just on the gross tumor. By doing tumor sampling we get too much noise.

7. After understanding the pathways, we should examine the tumor cells and then from that do a system identification to determine the specific pathway changes and then from that ascertain the aggressiveness of the tumor.

In conclusion, we are perhaps 5-10 years too early to deliver reliable tests.