One of the interesting developments is precision medicine: we can now use genomic sequencing and AI to "search" for the treatment that is most likely to be effective against a particular individual's cancer rather than following the rough "standard of care" playbook (for example, depending on the type of cancer, the standard of care might be 6 months of chemo, and if that doesn't work try something else, repeat until cancer goes away or patient dies). The precision medicine approach at this point seems to be among the things that we try if the standard of care doesn't work, but hopefully one day it will be step one (chemo sucks and it is an expensive use of precious time). In addition to searching for treatments, we can also search for clinical trials for which a patient is a good fit--this helps both the research side (it's very hard to find qualified patients) and possibly patients who don't have better options (and of course the research itself benefits future cancer patients).