Some great practice-changing studies did not involve corruption.. I'm thinking PORTEC2, Turrisi SCLC, Canadian breast.
This word "corruption" is odd in this context.
Systemic agents typically need clear phase 3 data with large, expensive studies to be approved by the FDA. That's both for approval and for each indication for use. This is in addition to research before this stage that is often performed in or with academia. Drug companies pay for all of this, typically with rates to academic centers higher than the rates paid by government funded studies.
New linear accelerators need no such trials. A report is filed with the FDA, the device is granted clearance for marketing, and rad oncs start buying it. The indication for use is very broad, something like can be used to radiate anything the doc says. The vendors don't need to pay for studies, and they don't. They might give a little money here and there to spread some influence, but it's peanuts compared to pharma money. In academics, we then publish what we can think to do with these devices, but it's typically retrospective data since nobody will pay for the prospective studies.
As for the government, they pay for studies and provide a lot of cooperative group funding, but NIH funding has not grown in a long time and is highly competitive. Getting large practice changing studies off the ground is a long process, mired in beuracracy and politics, and highly competitive. A lot of people try to go this route, since it essentially makes your academic career to be involved with these studies, but the pace of discovery using the NIH funded route is much slower.
So it's not corruption to get funding from pharma. It's a byproduct of the regulatory environment that we have. Pharma funding is how a lot of the research simply gets paid for.