Use of 18F-fluorodeoxyglucose positron emission tomography for diagnosis of uterine sarcomas.

2010 
We evaluated the use of 18 F-fluorodeoxyglucose positron emission tomography (FDG-PET) for the diagnosis of uterine sarcomas. FDG-PET combined with serum lactate dehydrogenase (LDH) levels were compared with FDG-PET alone for the diagnosis of leiomyosarcomas (LMS), which are the most difficult uterine sarcomas to diagnose. FDG-PET imaging of endometrial cancer (EC) was used as a reference. Immunoreactivity for glucose transporter-1 (GLUT-I) correlated with FDG uptake was evaluated in sarcomas and leiomyomas (LM), including cases not examined by FDG-PET. FDG was injected after at least 5 h fasting and standardized uptake values (SUVs) were analyzed quantitatively 50-70 min after injection. Immunohistochemical expression of GLUT-1 was studied in paraffin sections of tumors using anti-GLUT-1 antibodies and GLUT-1 expression scores were derived based on staining intensities. FDG-PET was performed preoperatively in a total of 53 patients including 10 with sarcomas, 19 with EC and 24 with LM. Immunohistochemical examination was performed in 17 sarcomas, 6 EC and 9 LM [6 usual LM, 1 uterine smooth muscle tumor of uncertain malignant potential (UMP), and 2 bizarre LM (BLM)]. SUVs for uterine sarcomas and EC were significantly higher (p=0.0001) than those for LMs. There were no significant differences in SUVs among ECs, carcino-sarcomas (CS) and LMS. Significant differences in SUVs existed between LM and LMS (p=0.003). However, the diagnostic accuracy for LMS was only 73%. The diagnostic accuracy of FDG-PET combined with serum LDH was 100%. GLUT-1 expression scores were significantly higher in sarcomas and EC than in LM (p<0.0001). Intermediate GLUT-1 scores were found in two of the three cases of UMP and BLM. In conclusion, FDG-PET is useful for diagnosing uterine sarcomas, while FDG-PET combined with serum LDH is useful for diagnosing LMS. Immunohistochemical examination of GLUT-1 confirmed the high FDG uptake in LMS patients.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    38
    Citations
    NaN
    KQI
    []