Interesting question. I would assume that it isn't possible to selectively destroy mitochondria in a cell. These organelles perform many essential functions to keep cells alive. On the other hand, strong mitochondrial perturbation might lead to the activation of cell death pathways which are co-controlled by mitochondria. The only option that might work for rather limited periods is massive induction of mitophagy. However, to my knowledge it is not possible to activate only one mode of autophagy without co-activating others.
I saw something similar in a recent study where they removed mitochondria with defects from a progenitor cell, the final effects were attenuated symptoms of aging
In the paper mentioned by Gisela you can find a paragraph "Parkin‐mediated mitochondria clearance" in the methods section. Maybe the approach used by the authors can be useful to you, too.
It was quite interesting thanks, sadly it showed me that the exact mechanisms of the different types of Autophagy are not that well understood and that the promotion of autophagy on a large scale could cause damage to the local immune machinery and eventually promote cancer.
Although many therapies use the mitochondria as a target I think it would be interesting to see a mixed treatment that uses a chemical that depletes mitochondria after the use of the main antineoplastic agent, since a damaged and undigested mitochondria can be so beneficial for the tumor development