One challenge to using oncolytic viruses as a treatment for cancer is that the viruses may cause off-target effects. I'm curious to know how useful it would be to create an oncolytic virus that has the receptor for another virus encoded into its genome (kind of like a viroreceptor)? It would probably be best if the encoded receptor is a receptor not present in humans, so maybe a receptor present on the surface of another animal's cells. If the first virus replicated to a higher degree in cancer cells than normal cells, and/or if the first virus was largely cleared from the normal cells before the second virus was administered, then it seems the cancer cells would differentially express the receptor for the second virus, and that could limit the off-target effects of the second (perhaps more potent) virus. (Of course, there are other ways of limiting off-target effects - injecting the viruses intratumorally, using viruses that the host would mount a memory response against, etc. - feel free to include these in your answer if you'd like.)
I'm sure there may be challenges in practice (one that comes to mind is whether a given first virus would have the genomic space to carry a receptor for the second, in addition to other desired genes) but I'm curious to know what people about this idea in the spirit of a thought experiment, or if you could point me to any relevant papers etc.