Abstract
509
Objectives To determine the value of adding PET/CT to clinical assessment after curative-intent RT in patients with HNSCC.
Methods Patients with HNSCC receiving a post-RT PET/CT from 2005-2010 were retrospectively identified. RT was followed by clinical exam and at least one PET/CT 60-180 days after RT-end. The clinical exam immediately prior to the date of the PET/CT was scored as: i) disease free or ii) suspicious/definite for HNSCC. PET/CT scans were scored as: i) disease free, ii) equivocal, or iii) suspicious/definite for HNSCC. Final analysis was conducted in two ways: an equivocal PET/CT was considered either (1) positive or (2) negative. Sensitivity, specificity, PPV, NPV, accuracy, false negative (FN), and false positive (FP) rates of PET/CT were determined for each analysis. Referring physicians were prospectively sent surveys 7-10 days after a patient's PET/CT to assess the percent change in patient management from PET/CT results.
Results 352 patients met eligibility. PET/CT was performed a median of 14 weeks after RT. Median follow-up after RT was 30 months. Sensitivity, specificity, PPV and NPV were 85%, 80%, 56%, 95%, for Analysis 1, and 75%, 92%, 74%, and 93% for Analysis 2, respectively. Overall accuracy of PET/CT was 81% in Analysis 1 and 88% in Analysis 2. The FN and FP rates were 3.4% and 15.3% in Analysis 1, and 5.7% and 6% in Analysis 2, respectively. 76% of ordering physicians felt PET/CT led to a change in their intended patient management.
Conclusions A negative PET/CT ~14 weeks after RT reliably predicts disease absence. Patients with negative clinical exam and negative PET/CT may be offered less frequent follow-up than usual. The PPV of PET/CT is modest, requiring confirmation prior to radical surgery. Patients with an equivocal PET/CT should be followed with close clinical exam and imaging to reduce the possible harm from FNs. Physicians found PET/CT valuable in decision-making. Thus, PET/CT should be part of the post-treatment clinical algorithm for patients with HNSCC.