The comparison is apt to the point I was making: that it is not the genetic nature of autism that makes it difficult to treat but our ignorance of its mechanism

The comparison is apt to the point I was making: that it is not the genetic nature of autism that makes it difficult to treat but our ignorance of its mechanisms (perhaps also the nature of the mechanisms themselves.) I think that you're confusing the etiology of the thing with the thing itself. Just as there is no reason to suspect that gene therapy would be necessary, there is no reason to suspect that it would be helpful in a person who has already developed autism.