Hi Joan,
I'm sorry this is so late. I started a post, but it was gobbledy-gook, so I trashed it.
That top onc is kinda right for now, but he's wrong for the future. At the moment, genetic/epigentic medicine is in its infancy. See for a quick rundown on the numbers we're dealing with:
http://geneticgenie.org/blog/2013/01...of-your-genes/
Up above, you have a list of "gene panels." They're looking at specific snps on those panels, but what about the snps they overlook? And what about the combinations (what SNPedia calls "genosets") they haven't considered?
The day is coming that an individual will be able to have her genome sequenced and know exactly what went wrong and maybe even how to fix it. For now, medicine is hampered by a lack of data. Cheap ($1000) genome sequencing will help fix that, and it's in the pipeline.
The bottom line (in my opinion) is that for the moment, these are the tests we have, and they do give us some useful information, so there's no reason not to use them. Your top onc disagrees. But he's fails to consider that the worst thing that happens by using these tests is that the research will advance for the benefit of those who come later.