Actually, the ~25% numbers people have mentioned seem rather low, based on my experience. I think the project I was PI on a several years ago had indirect costs of around 39%. Not cancer or healthcare related, but enough to give me some perspective.
Usually an organization or university has an indirect cost rate that they charge for all clients. These are included in the bid, so if the indirect costs are excessive, the bid will not be competitive.
So what does indirect costs pay for? I don't really know how it's calculated. It probably varies from organization to organization. Some things covered by indirect costs at one place may be directly itemized at another. The most obvious things would be things like facility maintenance, utilities, Janitorial service, IT, etc. And, of course, you have your administrative services: personnel, purchasing, and other support staff.
For a lab, it may include overhead to operate and maintain equipment. It's not cheap keeping lab instruments running. Service contracts can run $10-15k/year per instrument. (And those can pay for themselves very quickly.) Some places may include that type of thing in their indirect costs. There are rules as to what can be charged to a grant. And then rules as to what equipment purchased with one grant can be used on the next grant. One project cannot subsidize another.
Anyway, capping the indirect cost rate is not likely to actually save any money at all, in my opinion. Bids and proposals will just be written differently. For example, instead of lumping IT into overhead, I suspect that the IT charges and fees will instead be itemized as a direct cost. The way this would work is that the IT department would charge the grant a fee for IT service. Similar things would likely be done with facilities overhead.
Is 50% indirect costs excessive? I really don't know. Like I said, I don't know how it's calculated, just what I have and have not seen as line items on grants. It would probably depend on the nature of the research. When I think about medical research, I tend to think in terms of a lab. And the major lab instruments cost $100,000-$300,000. I know of one lab that spent, as I recall, around $300,000 for the instrument to analyze for PFAS. Lab instruments and the lab buildings themselves create indirect expenses that aren't present in, say an engineering or architecture firm designing a building or a bridge.