Now all that being said, wierd certainly has a point that allowing severe genetic defects to spread through the population via treating the symptoms so that those who suffer from them can lead a relatively normal life and pass those defects on, could certainly have disastrous consequences down the road. Possibly leading to large die-offs of humans, or even human extinction if it goes on long enough. And everyone needs to decide for themselves whether or not they believe that potential future harm to the species/society outweighs the immediate needs of the individual. I'm personally of the opinion that the best course of action is "Treat the individual (to the extent that they desire treatment, and resources to treat them are available), but acknowledge the long-term problem that can create, and devote as many resources as feasible to figuring out a way to mitigate or prevent the long term issue".
I hope this is all coherent, because it's nearly 4am and I haven't slept yet.
Well, it is not coherent in so far that this is the gene-engineering thread and not the generic-medcine thread. If you alter the human dna, why should you do so only for somatic cells and not germline? The only exception would be cancer which (usually?) is only somatic.
Another argument by weird was that we may not want to lose some currently unwanted traits from the gene pool (like sickle-cell trait that gives resistance against malaria) as we do not know which advantage they may have in the future. This is a valid point but I think does not correlate to much with the most advanced progress in genetic engeneering in single base exchanges via nucleases like zink-fingers, TALEN and (most recently) the CRISPR-Cas system. In these systems (you could already use them to make test-tube babies resistant against HIV via altering the gene for CCR5, to name something usefull) only a certain base is exchanged (the process in theory is reversable). So, if the SNP is harmfull, it would either die out or be replaced. In neither of these cases it stays in existance, so there is no real loss in the gene pool. It might be a different for deseases that only manifest in higher ages, but then again, you just treat yourself and not your children and there is only no loss in the gene pool.
Futhermore, since the thread started with it: Since genetic engeneering is basically a software update to the body, the usage of it is pretty cheap, it is the development where money is burnt. Therefore the upper-class problem only exists in a world where people restict technology to much (see the current discussion about Hepatitis C treatment) and slightly different obvoiusly race specific research (not neccessary targeted but people in europe have different inheritate diseases than asians). None of this is genetic engineering specific.
What really could become a problem (as always) is weaponizing. Again, not specific to this field but
gene drives could become horrifing (and/or a blessing). Also note that we are not really talking about future anymore. If there was a sudden fashion of having blue eyes or blond hair in china (or anywhere somewhat developed else) now, in ~half a year I think someone could develop an (unapproved) CRISPR to change the color permanently.
Note: I am talking about genetic engineering, not tissue engineering which is a totally different field.