Definitely strongly agree with delivery being the most significant hurdle. Many problems arise from difficulty delivering gene therapies via viral vectors, most

Definitely strongly agree with delivery being the most significant hurdle. Many problems arise from difficulty delivering gene therapies via viral vectors, most notably treatment efficacy and treatment durability, which should never really be a problem because you're replacing the genes. There's also the much-feared off-target effects, which can and do rapidly and gruesomely kill people in the gene therapy clinical trials-- an eminently solvable problem given more research into the correct epitope targeting, I think. Most of the current viral delivery vectors are shitty in a multitude of ways, but the kinks are being rapidly worked out for AAVs and lentiviruses. Crispr/Cas9 genome engineering is also a huge leap forward, as you mentioned. It's important to note that some groups were having luck with gene therapy even before Crispr, though-- imagine what they can do now, barely two years later. The door to de novo synthetic biology has been kicked open.