Semiquantitative Volumetry by Sequential PET/CT May Improve Prediction of Complete Response to Neoadjuvant Chemoradiation in Patients With Distal Rectal Cancer

BACKGROUND: Previous studies using PET/CT imaging have failed to accurately identify complete responders to neoadjuvant chemoradiation among patients with rectal cancer. The use of metabolic parameters alone or imprecise delineation of baseline and residual tumor volumes may have contributed for these disappointing findings. OBJECTIVE: The purpose of this study was to determine the accuracy of complete response identification in rectal cancer after neoadjuvant chemoradiation by sequential PET/CT imaging with a decrease in tumor metabolism and volume using optimal tumor volume delineation. DESIGN: This was a retrospective comparison of prospectively collected data from a clinical trial (National Clinical Trial 00254683). SETTINGS: The study was conducted at a single research center. PATIENTS: Ninety patients with cT2-4N0-2M0 distal rectal cancer underwent sequential PET/CT at baseline and 12 weeks after neoadjuvant chemoradiation. Quantitative metabolic analysis (median and maximal standard uptake values), volumetric estimates (metabolic tumor volume), and composite estimates incorporating volume and quantitative data (total lesion glycolysis) were compared for the assessment of response to neoadjuvant chemoradiation using receiver operating characteristic curves. Individual standard uptake value thresholds were used according to response to neoadjuvant chemoradiation to match metabolic activity and optimize volume delineation. MAIN OUTCOME MEASURES: The accuracy of complete response identification by multiple volumetric and metabolic parameters using sequential PET/CT imaging was measured. RESULTS: Variation in total lesion glycolysis between baseline and 12-week PET/CT scans was associated with the best area under the curve (area under the curve = 0.81 (95% CI, 0.69-0.92)) when compared with standard uptake value or metabolic tumor volume for the identification of a complete responder. Patients with a >= 92% decrease in total lesion glycolysis between baseline and 12-week PET/CT scan had a 90% chance to harbor complete response. LIMITATIONS: This study was limited by its lack of interobserver agreement analysis. CONCLUSIONS: PET/CT scan using volume and metabolic estimates with individual standard uptake value thresholds for volume determination may provide a useful tool to predict response to neoadjuvant chemoradiation in distal rectal cancer.