Using CRISPR on zygotes is easier for this reason than trying to cure diseases in adulthood. Obviously, CRISPR in adults is a lot trickier, since the daughter cells of one transgenic cell is a much smaller proportion of the entire organism (or to put it another way, most of an adult's tissues have already differentiated). Rather than trying to replace all the non-functional/diseased cells with cells from a CRISPR lineage (which would be really hard to do), researchers have to get more creative.

The first clinical trial using CRISPR in humans started only a few months ago (our current date is March 2017) (check out this article: gene editing ). In this case, the scientists took some of the patient's immune cells out of their body, cultured them in a dish, used CRISPR to change the genes they wanted, and then injected the cells back into the patient. This allowed them to create cells that would be good at targeting cancer, but would not be degraded by the host's own immune system. The scientists did not have to worry about changing all of the patient's immune cells because they were able to generate enough in the lab to fight the disease.

Hereditary diseases in adults (such as Huntington's or cystic fibrosis) will be harder to fight with this technology because thousands of cells in the body would probably have to be reprogrammed. There might be other ways of delivering a gene to cells, such as using a genetically engineered virus (known as gene vector therapy). This could allow a greater number of cells to incorporate the gene of interest. However, there are challenges there as well, such as generating a virus that will target very specific cell types.

The reason CRISPR has been so successful is because it specifically edits only a single gene, not any unrelated ones. Also, no known organism has billions of genes. Humans have about 20,000 genes though their DNA has billions of nucleotide base pairs. So, to clarify, technically CRISPR modifies a small number of base pairs. In fact, what makes a sequence of DNA a gene or not is not that straight forward. But it is true that modifying a single gene, while not modifying the DNA of other genes, could affect the function of the other genes. If you use CRISPR to modify a gene which ultimately activates other genes, then it could have an effect. To actually use CRISPR, unless you’re modifying a fertilized egg, you are probably going to modify more than 1 cell. How many cells are needed for repair a function will depend on the application. A single human cell takes around 12-24 hours to divide so if you don’t start with many cells, it would likely take weeks to months before the faulty function is repaired.

The cell that you modify will be different, but no other cell will be affected. If the cell then divides, then the cells that come from it will also be different, but not the cells that come from other cells. To "repair" a human, say, you would need to alter the genome of every cell in the human's body. This would take so long that you can't effectively do it, which is why genetic engineering is almost always done on the reproductive cells: if you alter the genome of an animal while that animal is a single cell, then that will affect the entire animal because every cell in the animal's body is descended from that single cell.

Now, if the problem you're trying to correct is the lack of a hormone or some other chemical that a cell produces and then secretes, then you don't have to affect every cell, just the cells of the organ producing the chemical, and even then only enough of them to produce enough of the needed chemical.

CRISPR is a really powerful tool for editing DNA but it is extremely difficult to use on humans for a variety of reasons, both scientific and philosophical. Let’s start with the science because, surprisingly enough, it’s the easier part to answer.When you edit with CRISPR, you change a specific portion of the DNA in a single cell. All edits that are made with CRISPR are exclusive to the cell the edits were made. Now, that one cell can go onto proliferate and make many daughter cells that will all carry the CRISPR mutation, but it won’t pass the mutation onto cells near it. For this reason, you cannot just use CRISPR for a person with blue eyes to make them have brown eyes. You may now be thinking, “couldn’t I just CRISPR ever eye cell so that they all have the brown mutation”, and the answer is … maybe? There’s nothing that says that you couldn’t do this besides the fact that CRISPR as an experimental technique isn’t perfect enough to do this. So if you wanted to apply CRISPR to fix genetic diseases, you would need to target every cell that expresses the disease. Realistically, this can only be done when the embryo is only a few cells large. So there’s good news and bad news. We can use CRISPR to fix genetic diseases but only if the disease is found very early in development.

Now that we’ve considered the scientific hurdles of using CRISPR on humans, let’s think about the moral implications. For one, CRISPR is fairly new technique and we do not fully understand it yet. We are pretty sure we understand what happens when we CRISPR single cells, but doing the same to a multicellular organism like a human is a totally different story. What if CRISPR had some devastating effect that impaired or killed the fetus? We don’t know what could happen and it is morally wrong to test. Furthermore many people argue that physically editing human DNA is going too far. They would argue humans shouldn’t have that sort of power to design life. On the other hand, others would argue that if we have a way to eliminate debilitating genetic diseases then we should. I don’t think there is an objective correct answer, but it is something to consider. Thank you for the question, and I hope you take the time to think about the ethical and scientific effects that using CRISPR on people can have.