This is a thing right now, it seems. Lizzie is making a game about breast cancer; Ole Peder is making one about bipolar syndrome; I ran one about growing up in an environment of alcohol and drug abuse. Sanne Harder Flamant made one about confronting a suicidal parent.
For me, the major challenge seems to be how to - and, indeed, whether to - make these games dramatically/gamewise interesting for others, while at the same time preserving the autobiographical elements. There's a whole spectrum of possibilities there; at one end of the spectrum, you can have games that are inspired by issues the designer knows personally, but filtered and dramatized and made into a generalizable game experience. (For instance, my game could have been about different families with substance abuse, different roles one could take, comparing and contrasting them to show the dynamics and how they affect the families). At the other end, you can have tightly scripted games, directed by the designer, that run through autobiographical scenes in detail (which was pretty much what I ended up doing with my game - "this scene is about me trying to score drugs from X and Y, the year is Z, we're in city C, these guys will act like A, my bike was of brand B, go!")
My designer instinct, habit or preconception says that the last way of doing it is Wrong, because you're supposed to give players freedom, and make games accessible, etc. I like free & accessible games, both as a player and a designer. However, I'm not so sure when I think about it twice. I think there can be a lot of value in having a directed interactive experience, walking in someone else's shoes for a while. As a player, I could find this interesting. I ask myself, of course: "But is that really a role-playing game?" I answer myself, as usual: "Who the hell cares! Irrelevant question."
However, as a game designer, I wasn't happy with the tightly-scripted game I ran. It felt like a cop-out. It wasn't like designing a game; more like running a play or something, I don't know. After doing it, it felt like I had failed somehow in living up to my own expectations - it was supposed to have a lot of player input, meta-techniques etc, but didn't. (The players, in general, had an interesting and memorable experience, from what I understand.)
Reflecting on it now, I think that not living up to my own expectations is not a problem in and of itself; there are probably a whole, different set of techniques available for running that kind of tightly-scripted, detailed autobiographical game and making it a cool experience. If I were to design and/or run a close-to-home game again, I would consider where on the spectrum it should lie, and adjust techniques accordingly.