Seems unlikely to me, I have done a fair bit of radiation testing
and never seen a discrete capacitor let go. However these are
generally not in high ripple current applications and I derate at
least 2X the expected applied voltage. But there is no way that
the beta flux from these samples is going to heat the capacitor
to the point of failure - they would have self heated noticeably
as a large portion of alpha or beta never make it out of the
source mass before stopping.
I think it's more likely that the ionization from the radioactive
material let you start a spark-gap arc easier, so as to then
put more ripple current through the converter's filter cap
(which in-the-moment is the only source of arc step-current)
and cook it by repeated I^2*R (boil the electrolyte and you
are done).
Might suggest you look for alternatives to, or supplement the
boost converter's output capacitor that will tolerate more
ripple current (we're not talking the ripple from the boost
supply, but the ripple from the spark's negative differential
resistance behavior when it's kicked off).