Evaluation of FDG PET in patients with cervical cancer

Abstract

Although many human cancers can be imaged by 2-[18F]-fluoro-2-deoxy-D-glucose (FDG) and PET, there is little clinical experience with FDG PET in cervical cancer. The purpose of this study was to evaluate the feasibility of FDG PET scans on patients with cervical cancer.

Methods: FDG PET scans were performed on 21 patients with histologically proven uterine cervical cancer (17 newly diagnosed, 4 recurrence). After two levels of transmission scanning, approximately 370 MBq FDG were injected, and dynamic scans over 60 min were obtained at the level of suspected tumors, followed by static scans. Postvoid scans were also obtained in 11 patients to minimize FDG activity in the urinary bladder. FDG uptake was interpreted visually and classified into 4 grades (0 = normal, 1 = probably normal, 2 = probably abnormal and 3 = definitely abnormal). For a semiquantitative index of FDG uptake in tumors, the standardized uptake value (SUV) corrected by predicted lean body mass (SUL) was calculated and compared. The detectability of lymph node metastases by PET was compared with that by CT.

Results: Of the 21 newly diagnosed or recurrent cancers, 16 (76%) were detected by FDG PET without use of postvoid imaging (i.e., interpreted as grade 2 or 3). The SULs of tumors ranged from 2.74-13.03, with a mean of 8.15 +/- 3.00 (SUV range 3.68-14.94, mean 10.31 +/- 3.19). There was no significant relationship between the SUL of cervical cancer and the clinical stage. Postvoid FDG PET images substantially reduced the tracer activity in the urinary bladder and improved the visualization of cervical cancers, with three additional cases detected using the postvoid images. In the 11 patients with postvoid imaging, all 11 cancers (100%) were detected. FDG PET detected lymph node metastases in 6 (86%) of 7 patients with known metastases, whereas CT was positive in 4 patients (57%), equivocal in 2 patients (29%) and negative in 1 patient (14%). All PET and CT scans were true-negative in the patients with no lymph node metastases (interpreted as grade 0 or 1 by PET, and as negative by CT).

Conclusion: These preliminary data demonstrate the feasibility of FDG PET imaging in patients with cervical cancer. FDG PET appears to be promising for detecting untreated or recurrent cervical cancers and lymph node metastases, although the excreted FDG in the urine remains problematic in some cases.