844-845 / 1152
Translation via modeling??
•
Models represent correlations
•
Correlations do not proof a mechanism
–
Cardiac dose metric may correlate to another
(lung) dose metric that was not considered
•
Lack of correlation does not proof / reject a
mechanism
–
Power?
–
Population/treatment characteristics?
–
Was the relevant dose metric considered and
did it vary sufficiently?
–
Medication?