Five year experience in the detection of recurrence of ovarian cancer with [F18]-FDG using an hybrid (CDET) gamma camera

Objective: The aim of this study was to evaluate the role of [F18]-FDG-CDET in the detection of recurrence of ovarian cancer. Methods and patients: After a fast of 6 hours, the patient (pt) was injected I. V. with 150-250 MBq of [F18]-FDG and imaging (whole-body scan and at least a tomoscintigram) was started 60 min. later, using a PICKER-MARCONI dual-head or triple-head CDET gamma camera. Between July 1997 and July 2001, 81 pts were studied for suspected recurrence of ovarian carcinoma. To date, the results of 62 pts are evaluable with reference to histology after surgery or concordance with conventional imaging and long-term follow-up. From these, 27 pts were referred for occult recurrence (OR) defined by an increase in serum CA-125 levels with negative CI and the remaining 35 pts for equivocal aspect at conventional imaging (ECI). Results: [F18]-FDG-CDET was true positive in 47 cases, 27 confirmed by histology after surgery (13 OC, 14 ECI) and the remaining 20 confirmed by evolution and concordance with CI. [F18]-FDG-CDET was true negative in 13 cases (6 OC, 7 ECI) as confirmed by spontaneous normalisation of CA-125 levels and no events during a 20-month follow-up for 11 pts and histology after surgery for 2 pts. [F18]-FDG CDET was false negative (FN) in 2 pts with ECI, 1 pt with a lymph node metastasis of less than 10 mm in size and 1 pt with continuing increase of CA-125 levels and still negative conventional imaging during a 8 month follow-up. No false positive results was reported in our study. In summary, the overall sensitivity, specificity and accuracy on a per patient basis were respectively 47/49 (96%), 13/13 (100%) and 60/62 (97%). The positive and negative predictive values were respectively 47/47 (100%) and 13/15 ( 87%). Conclusion: The present series of 62 patients gives valuable experience in CDET as compared to the cumulated data by Gambhir et al (J Nucl Med May 2001) reporting on a total of 357 patients using dedicated PET. Our CDET results are at least similar to those reported with dedicated PET cameras, which average values were Se=88%, Sp=90% PPV=85% and NPV=92% in this setting