|
Ironically, one of the contributors to and the owner of cancerfocus is a Doctorate in Immunology. He found me over six years ago and wanted me as the moderator. I've been involved with internet cancer research for seventeen years and have studied cell function analysis over the last twelve years. Of course cell function analysis holds no boundaries across the various cancer types and helps in being a moderator of a cancer information website that deals with all the various types. It's been some of the best learning experience in my life, other than studying Laver Curves in the '70s.
As you know, microarrays examine what genes are expressed in cancer cells. It is mainly used for screening/gene discovery work. You screen thousands of genes to discover an association and then you focus in on only a few hundred or so for more careful study by some other method like RT-PCR (real time polymerase chain reaction). But all the gene mutation or amplification studies can tell us is whether or not the cells are potentially susceptible to this mechanism of attack. It doesn't tell you if one drug is better or worse than some other drug which may target this.
It is whether the capacity to judge phenotypes will be easily achieved at the genotype level.
Genotype analysis measures gene expression by microarray. The microarray test looks for genes in the tumor that are associated with treatment options. It is tumor analysis coupled with clinical literature search, which tries to match therapies to patient-specific biomarker information to generate a treatment approach. They never actually test the tumor specimen against any drug agents. So based on this analysis, they identify "potential" therapies for patients and their treating physicians to discuss.
Phenotype analysis takes the tumor with the surrounding tissue (intact and live) and then puts drug agents on it to see which actually kill the cancer cells. It measures biological signals rather than DNA indicators. It "actually" measures the response of the tumor cells to drug exposure, the integrated effect of the drugs on the whole cell, measuring the interaction of the entire genome. No matter which genes are being affected, phenotype analysis is measuring them through the surrogate of measuring if the cell is alive or dead.
The former is testing for mutations, while the latter is testing for drugs. System's biology suggests that the simple knowledge of a gene's presence or absence does not confer a biological behavior. Biology is not linear.
The idea of simply finding a mutation and then picking an appropriately targeted drug seems like a nice idea. However, not every key that looks like it will fit a lock will actually turn it. There are numerous common mutations in various tumor types, but they don't know that all those mutations are going to turn out to be relevant, as many of them are essentially bystanders.
Why don't all the mutation positive patients respond and why do some mutation negative patients respond? Cancer biology is complex. Molecular biologists can only seek and identify that which they know a priori. There are numerous mutations, insertions and deletions.
A gene mutation, deletion, translocation or amplification could disrupt many cell functions, leading to drug resistance, or could inactivate or damage the doors through which a drug enters a cell.
Cancer arises not only from gene mutations but also from small fragments of RNA that can up- or down-regulate normal genes in abnormal ways. The fact that normal genes can function abnormally under the control of these small RNA sequences is just one more example of the genotype-phenotype dichotomy that cannot be adequately examined on static contemporary genomic platforms.
BTW. Speaking of clinical trials, you know the "gold standard" for all diagnostic tests are clinical correlations. However, some private laboratory oncologists are working (again) on randomized clinical trials, but this time a good three-armed clinical trial: physicians' choice (or typically called trial-and-error) vs molecular profiling vs functional profiling.
Back in 2011, in the first head-to-head clinical trial comparing gene expression patterns (molecular profiling) with personalized cancer cytometric testing (functional profiling or chemosensitivity testing), personalized cancer cytometrics was found to be substantially more accurate (Arienti et al. Journal of Translational Medicine 2011, 9:94).
It is hoped that something like the Arienti, et al study would be proposed at one of the GOG meetings. It was and was accepted. The process of funding has been ongoing. Personally, after twenty years, I would like to see this "battle-of-the-bands" three-armed trial. Let's put it all out there and see what's best!
|