I sympathize with your situation and frustration. And yes, temozolomide and bevacizumab and whatever else they're giving your friend won't buy much time. Everyone knows that.
Part of the problem unfortunately is that glioblastoma, like many highly-aggressive cancers, is not a single disease but many. The cells mutate frequently, and each group of mutant cells (call them a "clone") is in business for itself to survive anything you throw at it. So something like your Vika virus idea, for instance, might have a 1 in a million chance of killing 99% of the cells, but the remaining 1% will be completely resistant and go on about their business. (Meanwhile, the virus is orders of magnitude more likely to cause more harm than good.)
Since you've known so many people who've had GBM, there may be value in investing in research that might help someone in the future. There are two broad moonshot approaches. First, the immune system is the central axis of cancer. Every cancer that grows and spreads is an example of the immune system failing to do its job (usually because the tumor has shut it down). If we can better understand how this happens, we can make pan-cancer drugs. They sort of already exist: one group is called immune checkpoint inhibitors. They basically unmask the tumor, allowing the immune system to identify it and take care of business. But they're only part of the solution, we obviously need more.
Another approach is further out there, and involves development of nanotechnology. Bacteria-sized machines small enough to get into cells could ideally be tailored to do pretty much whatever we want. This is a bit more of a long shot, but this is the sort of thing that would help solve the endless game of whack-a-mole that is cancer medicine today.