Abstract

To evaluate the accuracy offluorine-18-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT), to correctly determine initial tumor stage in treatment-naive gastric cancer patients and to analyze the factors influencing the risk of false negative results. The baseline 18F-FDG PET/CT scans of 111 previously untreated gastric cancer patients were retrospectively assessed. Sensitivity, specificity, positive (PPV) and negative prediction value (NPV) were evaluated. An array of clinical, pathological and metabolic variables was analyzed to identify factors contributing to the risk of a false positive (FP) and false negative (FN) PET/CT result in detecting primary and metastatic tumor sites. The sensitivity, specificity, PPV and NPV of PET/CT to visualize distant metastases were 76.4%; 86.7%; 83% and 81.2%, respectively. In 13 (11.7%) patients the PET/CT exam was able to identify metastatic sites not recognized in radiographic staging, significantly altering the initially planned management. Of 64 PET/CT studies negative for distant metastases, 12 (18.75%) were clinically confirmed to be false negative. The risk of acquiring a FN result for primary tumor was 10.8% (12/111) and the overall risk of any FN readout for either primary and metastatic sites was 18.9% (21/111). The factors that contributed to increased probability of a FN result for primary tumor detection were early primary tumor stage T1-T2 (+16.2%; χ2=5.0, P=0.025), female sex (+10.1%; χ2=5.71, P=0.017) and neutrophil count below 4.2k/μL (9.7%; χ2=6.1, P=0.014). Patients with non-intestinal Lauren histologic type (+18.7%; χ2=8.9, P=0.003) or signet-ring/mucinous carcinoma (+9.6%; χ2=7.7, P=0.005) had increased probability of PET/CT being unable to identify their distant metastases. Women and patients with low neutrophil count featured borderline insignificantly increased percentage of non-intestinal tumor histology (P=0.07 and P=0.057, respectively). Fluorine-18-FDG PET/CT is a valuable diagnostic method in gastric cancer patients which significantly contributes to determining the TNM stage and thus helps choose correct patient management. Histology and primary tumor stage as well as patient cohorts in which these factors may vary should be considered when evaluating results to decrease a chance of a false negative readout.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call