Indication Of PET/CT For Endoscopic Resection Of Superficial Esophageal Squamous Cell Carcinomas

Background and Aims ESD has been accepted as the standard treatment forsurperficial esophageal carcinomas. Preoperative evaluation of invasion depth andlymph node metastasis are of great importance.18F-fluoro-deoxyglucose(FDG)PET-CT has been widely applied in diagnosis, staging and restaging, response tothreapy, recurrence and predicting prognosis of advanced esophageal carcinomas.However, few reports are available on PET/CT for superficial esophageal squamouscell carcinoma(SESCC), the clinical significance of which has been remainedcontroversial. Therefore, aims of the present study are to explore the characteristics ofFDG intake and its influence factors, as well as clincial significance of PET/CT ondecision-making of superficial esophageal squamous cell carcinoma, namelyendoscopic resection or surgical esophagectomy.Methods A total of2461consecutive patients with confirmed or highly-suspiciousesophageal carcinoma, who underwent endoscopic resection or esophagectomy fromJanuary2007to March2014in PLA General hospital, were analyzed, retrospectively.According to inclusion and exclusioncriteria,61SESCC patients (64lesions) withpreoperative18F-FDG PET/CTwere enrolled, endoscopic resection for52lesions andesophagectomy for12lesions.Data of them was recorded and analyzed.Results Of the64lesions,24lesions were visible and40were non-visible. Sensitivityof PET/CT for initial diagnosis of SESCC was37.5%(24/64). Univariate analysisshowed visibility and semi-quantitive FDG intake (SUVmax,SUVmax ratio A,SUVmax ratio B) were related to age, longitudinal diameter, circumferential degree,non-flat or nodal type and invasion depth (P<0.05). But Multivariate logisticregression analysis showed longitudinal diameter≥40mm and invsion depth≥sm2were the only two influence factors(OR value=23.2,10.4;P<0.05). Meantime,visiblityand SUVmax were closely related to≥sm2invasion(OR value=7.17,8.0;P<0.05).When visible lesions indicated esophagectomy andnon-visible for endoscopic resection, sensitivity(Sen), specificity(Spe) and correctclassfication rate(CCR) was82.3%(14/17),78.7%(37/47) and79.7%(51/64). With aSUVmax cutoff of2.30for clinical decision-making of SESCCs, Sen, Spe and CCRwas70.6%(12/17),76.6%(36/47) and75%(48/64), respectively. With Roedl PET/CTscoring system, SESCCs with invasion depth≥sm2score more in FDG uptakeintensity, focality and eddentricity than sm1(P<0.05). ROC curve indicated thatE-F score(eccentricity-focality score)≥2contributed to a highest diagnosticefficiency, with Sen, Spe and CCR of70.6%、87.2%and82.8%. Compared with EUS,which had a Sen, Spe and CCR of60%、77.8%and77.4%, PET/CT appeared to havea diagnostic efficiency.Conclusions FDG visibility and uptake are closely related to longitudinal diameter≥40mm and invsion depth≥sm2. Despite limited diagnostic value of primarySESCCs,18F-FDG PET/CT plays a important role in indicating esophagectomy orendoscopic resection for SESCCs. Compared with EUS,18F-FDG PET/CT appears tohave a higher accuracy in clinical decision-making of SESCCs. Besides, RoedlPET/CT socring system could be applied in differentiating≥sm2SESCCs from sm1. However, to reach a higher accuracy, some modifications are still required.