Pathological validation of FLT PET-CT in delineating the biological target length of gross tumor in esophageal carcinoma / 中华放射肿瘤学杂志

ABSTRACT

Objective To establish a optimal method and threshold of 3-deoxy-3-fluorothymidine (FLT) PET-CT in delineating the biological target length of gross tumor in esophageal carcinoma, and to compare FLT PET-CT with other imaging modalities including esophagoseopy, esophagography, CT and flu-orodeoxyglucose (FDG) PET-CT. Methods Twenty-four patients with esophageal squamous cell carcinoma treated with radical surgery were enrolled. Before surgery, all the patients underwent FLT PET-CT, esepha-goscopy and esophagography. Twenty-two patients also received FDG PET-CT scan. Gross tumor volumes (GTV) were delineated using seven different threshold of FLT PET-CT: visual interpretation, standardized uptake value (SUV) 1.3, SUV 1.4, SUV 1.5, 20% of maximum standard uptake value (SUV_(max)), 25% SUV_(max), and 30% SUV_(max). Three different thresholds of FDG PET-CT were used, including visual interpre-tation, SUV 2.5, and 40% SUV_(max). The length of tumors on FLT PET-CT scan were measured and recorded as L_(FLTvis), L_(FLT1.3), L_(FLT1.4), L_(FLT1.5), L_(FLT20%), L_(FLT25%), and L_(FLT30%), respectively. The length of tumors on FDG PET-CT scan were recorded as L_(FDGvis), L_(FDG2.5), and L_(FDG40%), respectively. The length of tumors on CT, esophagography and esophagoscopy were recorded as L_(CT), L_(X-ray) and L_(Scopy). All of these results were com-pared with the length of gross tumor in the reseeted specimen measured by pathological examination (L_(Path)), Results The L_(Path) was (4.90±2.14) cm. The Length of tumors delineated by different methods, being from short to long, were L_(FDG40%), L_(Scopy), L_(X-ray),L_(FLT1.5),L_(CT),L_(FLT30%),L_(FLTvis),L_(FLT1.4),L_(FLT25%), L_(FDG2.5),L_(FDGvis),L_(FLT1.3),L_(FLT20%). The mean values were (3.85±1.52), (4.46±2.23), (4.63± 2.37), (4.64±2.38),(4.69± 1.85),(4.75±2.19) ,(4.85±2.33),(4.87±2.35),(5.05±2.20), (5.08± 2.19) ,(5.10±2.22), (5.21±2.40) and (5.53±2.17) cm,respectively. The correlation coefficients were 0.91,0.93,0.88, 0.95, 0.90, 0.81,0.96, 0.96, 0.80, 0.99, 0.99, 0.95 and 0. 79 , respective-ly. All the P values were 0. 000. L_(FLT1.4) of FLT PET-CT and L_(FDG2.5) of FDG PET-CT were found more ap-proximate to L_(Path). There was no significant difference between L_(FLT1.4) and L_(FDG2.5) (1= 1.23, P = 0.232), and the correlation coefficient was 0.96 (P = 0. 000). Conclusions Thresholds of SUV 1.4 on FLT PET-CT and SUV 2.5 on FDG PET-CT could optimally estimate the tumor length measured by pathological examina-tion, and could be objective and simple methods for semiquantitative analysis.