Yes, this is very much a treatment of killing off immune cells and sneaking in a set of replacements before the body generates its own. So that way the replacements are the ones you want, not the ones you'd otherwise get.

In autoimmunity the problem is that the immune system is misconfigured in some way - so culling existing immune cells and replacing with fresh ones stands a good chance of getting rid of the cell-based misconfiguration. That is true whether or not the misconfiguration is understood; it is a good shortcut past lack of knowledge.

This has worked for other autoimmune diseases, but is sufficiently drastic that the research community has abandoned it in favor of even marginally effective treatments for conditions where those exist. For example there were promising results in the burn-and-rebuild approach for rheumatoid arthritis a way back, but then biologics arrived offering control and less pain for some patients, and that line of research died out.

It will probably return with the advent of more targeted therapies that are not as harsh as chemotherapy, ways to selectively destroy only immune cells with specified surface chemistry. This sort of thing is underway in the cancer research community, and the technologies produced will have very broad application: safe and painless removal of immune cells, fat cells, senescent cells - anything we don't want and can be distinguished. This is a big deal but still in the works; the issue isn't destroying the cells, but rather coming up with sufficiently discriminating targeting systems.