PET/CT as a predictor of survivals in patients with ovarian cancer following first line therapy

Abstract

1490

Objectives: The purpose of this study is to evaluate correlation between PET/CT parameters and survivals of patients with ovarian cancer following first line therapy.

Methods: From 2002 to 2007 Emory Healthcare database revealed 30 patients with ovarian cancer of stage IIIC and IV performed PET/CT for the restaging. PET/CT within 2 weeks following completion of chemotherapy were excluded. Demographics of patients, CA125, updated follow-up status including treatments were abstracted from medical records. We defined PET/CT parameters as 1. PET/CT positivity, maximum SUV (SUVmax), 3. volume of tumor with SUVmax, 4. total volume of tumor, 5. local or diffuse distribution of tumor foci, and 6. type of metastasis (local or peritoneal seeding without vs. with lymphatic and/or hematogenous metastasis). Kaplan-Meier survival curves were compared using the log-rank test.

Results: Sensitivity, specificity, and accuracy of PET/CT for diagnosis of recurrent ovarian cancer were 96.29%, 88.23%, and 93.18%. Overall survival (OS) was 29.87 - 47.91 months with 95% confidence interval (mean: 38.89; SD: 4.62). PET/CT parameters revealed no statistical significant correlation with OS. However, SUVmax showed a tendency of correlation with OS when divided at a cut-off value of SUVmax (SUVmax: 15) (p = 0.325). Also, local or diffuse distribution of tumor foci and type of metatasis showed a tendency of correlation with OS (p = 0.210 and p = 0.339).

Conclusions: In our study, PET/CT as a restaging was not a statistically significant prognostic factor. However, of PET/CT parameters SUVmax, distribution of tumor foci, and type of metastasis showed the tendency of correlation with OS. Further study with a large number of patients should be needed.