A Curse in Disguise?
How many people die from viruses, a lot of times they either stay with you and make you miserable but dont kill you (aids, cold, stds), or kill you very quickly and burn out (ebola). I honestly can't think of anything that has a high death rate but doesnt burn itself out before it can spread to a large amount of people, other than the flu viruses. And even then the flu is usually only fatal to the old, young, or already weak. A large portion of people die from just failings of the body, heart disease, brain diseases, cancer, things like that, that are not directly caused by viruses. So all in all this is a godsend.
To expand on that, a virus is actually considered unsuccessful if it kills you. While a bacterium will feed on living or dead material without really caring what harm it does, a virus needs living tissue in order to reproduce, so they usually do whatever they can to avoid killing the animals (including humans) that they infect. The flu manages to spread itself very effectively in sneezes and things, but a normal healthy adult human won't die from it - they will be miserable for a while, spread the virus to the people around them, and then recover when their body manages to kill the virus.
So this isn't so much a life-saving drug as a drug that will make people happier.
The one exception I can think of would be HIV, which does kill people if it turns into AIDS. As a whole, that is actually a very unsuccessful virus. It is spread by sexual contact, which means that most people will only pass it on to one or two other people (unless someone like Sarkozy gets it), and it can often kill the people it infects. But that is, as I said, an exception.
I'd say HIV is very successful. The virus's selfish interest lies in keeping the host alive long enough to spread the virus to others. It does that, and even better, the host may remain unaware of the virus for some time while spreading it; that stealth is important because once the host becomes aware of hosting the virus, the host may behave in ways that don't spread the virus as much. When the host's behavior changes so, it makes that host less valuable to the virus — so it then doesn't matter so much (to the virus) whether the host dies.
Yes, but it can only be spread to one or two other people, and very slowly. If someone were born with HIV, it won't be passed on to anybody else for at least 16 years (assuming they do it all legally), and perhaps a lot more. Once the host does start having unprotected sex, it is likely that they will be in a long-term committed relationship (and trying for a baby), so the only person they can pass it on to would be their partner, and they hopefully aren't sleeping around. In a person's lifetime, they probably won't spread it to more than ten people, even assuming they were born with it. Add to that the chance that the host could quite easily find out about it from a doctor or die from it, and the virus doesn't really stand very good odds.
Re: Cancer: is cancer not a virus? it can be killed by heat, but if there is a way to tell the body to kill it then it would seem a good way to kill cancer.