Endorse different assertion in staging

[korikuzma]

I wonder, for the long run, if it would make more sense to endorse a different Assertion instead of trying to patch the fusion variant of this assertion into a gene variant. E.g. assertion 180 (only on prod) is a substitution therapy but that would require us to update the data on staging. We could also endorse an assertion with combination therapy and then we'd only need to patch the therapy interaction value.

Originally posted by @susannasiebert in #188 (comment)