Phase III Trial of PET/CT vs. CTSurveilance for Head and Neck Cancer

The null hypothesis is that patients screened by PET/CT will not have detection of disease recurrence any earlier than those screened by CT alone. The alternative hypothesis is that PET/CT surveillance will lead to detection of disease recurrence 3 months earlier than CT surveillance. Furthermore, to reject the null hypothesis, earlier detection must be associated with a cause-specific survival improvement of 10%. Primary endpoints will include time from the completion of definitive therapy to diagnosis of recurrent disease, and absolute survival within 3 years after completion of initial therapy. Duration of survival between diagnosis of recurrence and subsequent death will not be a primary endpoint because the investigators expect that PET/CT will offer an opportunity for earlier recognition of recurrence and be subject to lead-time bias. Duration of survival will be measured from completion of primary treatment until death. Note: the presence of residual disease at surgical consolidation does not constitute a recurrence event.