Positron emission tomography (PET) response criteria in solid tumors were recently proposed as a standardized method for the metabolic and quantitative assessment of response to chemotherapy. However, use of these criteria is limited in many institutions because of the need for exclusive software. This study was designed to clarify whether tumor to normal esophageal (T/N) ratio on (18)F-fluorodeoxyglucose PET/computed tomography could predict response to neoadjuvant chemotherapy and stratify prognosis in patients with esophageal squamous cell carcinoma (ESCC). Clinicopathological data were collected for 73 patients with ESCC who received neoadjuvant chemotherapy with docetaxel, cisplatin, and 5-fluorouracil followed by curative resection. The right liver lobe and normal esophagus were utilized as reference tissues for diagnosing complete metabolic response (CMR). Statistical methods included Kaplan-Meier analysis and univariate and multivariate Cox proportional hazards regression analyses. CMR was achieved in 24 patients on the basis of maximum standardized uptake value (SUVmax) and in 11 on the basis of SUVmax evaluation with T/N ratio. Although prognosis was poorer in patients who achieved CMR than partial metabolic response based on SUVmax, the responses were significantly correlated with disease-free survival (DFS) based on SUVmax evaluation with T/N ratio (P=0.0011). Receiver operating characteristic curve analysis showed that SUVmax evaluation with T/N ratio was the best predictor of pGrade3. Multivariate analysis showed that SUVmax evaluation with T/N ratio was an independent predictor of DFS in patients with pGrade1 pathologic response. SUVmax evaluation with T/N ratio is useful for evaluating the effects of neoadjuvant chemotherapy in patients with ESCC.
Read full abstract