ANOTHER SILVER BULLET TARNISHED **

 On our Mexico trip

She loved little kids of all sizes    

I would have liked to believe that, if only we knew all about a person’s genome, we could predict his or her lifetime cancer risk, thus enabling us to initiate prevention early enough to chop off  most  cancers at their roots.  Of course, I knew it wouldn’t be that simple – I am gradually realizing that nothing in this business is simple  - but I had hope.  A paper in Genomics, by N. J. Roberts and five others, published in May of this year, has dashed cold water on my youthful enthusiasm. 

These guys used registries* of identical twins (“monozygotic”, if you want to appear learned), and also dizygotic twins (we call them fraternal twins.)  Monozygotic twins have precisely the same genome at birth; dyzogitic twins don’t.  The bottom line in this research, as I get it, is this:  The relative “heritability” of a disease represents the frequency with which both identical twins get a given disease, as compared to the same frequency for fraternal twins.  For instance:  If every time an identical twin got a disease her sibling did too, but the same did not hold for  fraternal twins, then the disease is heritable.  (Actually, what we are talking about here is a propensity to develop the disease; things other than genotype obviously are involved.)  Then they did a lot of fancy mathematics, and came to the discouraging (but predictable?) conclusion that, for most diseases, sequencing genomes at birth would be of minimal value.  In the case of the disease most interesting to me – ovarian cancer – only about 12% of people with the disease would have tested positive using their whole-genome algorithm;  alternatively,  a positive (bad) test result indicates only that the patient is  about 10% more likely to develop the disease than the general population.  And so, another silver bullet is tarnished. 

This is not to say that genome sequencing is useless.    Far from it.  There are specific genes that, when mutated, confer a high probability of contracting a specific disease – the BRCA1/2 genes for ovarian and breast cancer, for instance.  The whole genome studies these guys did suggest that they might be useful  for Alzheimer’s disease, or male coronary disease, or even Type 1 diabetes, but hardly so- and prohibitively expensive - otherwise. 

*Mostly from the Scandinavian countries, which apparently accumulate vast storage bins filled with neatly ordered data.   Now we know what they do during those dark winter months, other than write mystery stories and create more Scandinavians.

** Ah, come on, guys!  Won't SOMEBODY read my blog of 9/1/12?  It is embarrassing to see the big goose egg in the column headed "page views".  You don't need to read it, just bring it up so the zero will disappear.

Thanks.