Sir,

Over the past several years, positron emission tomography (PET) has become an important diagnostic modality for evaluation of solid malignant lesions.1 In contrast with computerized tomography and magnetic resonance imaging, which detect structural abnormalities, PET scan relies on altered metabolism within a malignant cell to detect tissue abnormalities. Malignant cells have a higher rate of glucose uptake and therefore show preferential uptake of radiolabeled Fluorine-18 fluorodeoxyglucose (FDG).2 FDG undergoes spontaneous decay emitting positrons with a half life of 110 min. The anatomic localization of FDG entrapment is improved with sequential imaging of the corresponding areas using noncontrast computerized tomography (CT) and a fused image is generated for diagnostic evaluation.

We performed a pilot study on a consecutive prospective series of 10 patients who were clinically diagnosed with primary uveal melanoma. Each patient underwent whole body FDG-PET/CT scan as part of work up for metastasis prior to treatment on Siemens Biograph16 system (Hoffman Estates, IL, USA). The scan was performed after the patient had been fasting for at least four hours. FDG dose was approximately 10–12 mCi (370–444 MBq) given intravenously. The maximum standardized uptake value (SUV) was measured in cases where the primary tumour could be visualized. CT was carried out at the same setting. Correlation was attempted between tumour size and identification of the tumour by PET/CT scan.

Comment

Overall, the diagnostic accuracy of uveal melanoma is more than 99% based on indirect ophthalmoscopy, ultrasonographic, and angiographic studies.3 Therefore, the role of ancillary studies such as immunoscintigraphy,4 CT,5 and magnetic resonance imaging is used to establish the diagnosis6 is limited to the atypical cases.

In our series, uveal melanoma could be detected with FDG-PET/CT scan in six of 10 (60%) patients (Figure 1). Each tumour appeared to demonstrate relatively uniform uptake and the intensity of FDG uptake (as measured by maximum SUV) varied between tumours from 3.5 to 8.6. Tumours that were at least 13 mm in diameter, 4.5 mm in height, 125 mm2 in scleral contact area, or 565 mm3 volume were visualized (Table 1).

Figure 1
figure 1

Ophthalmic (a), ultrasonographic (b), and histopathologic (c) appearance of a choroidal melanoma. The tumour (18 mm in base and 13 mm in height) could be visualized by PET scan (d) (maximum SUV of 6.1) and a CT scan (e). Fused image of PET/CT scan localizes hypermetabolic activity to the intraocular tumour (f).

Full size image
Table 1 Correlation of tumour size and visualization on PET/CT scana
Full size table

PET evaluation of smaller lesions is limited.7, 8 It is conceivable that additional factors affecting tumour glucose metabolism such as tumour necrosis, intratumoural haemorrhage, and inflammation could influence tumour visualization.8 As PET/CT scan is a test of metabolic activity of the tissues, it may have a potential in evaluating tumour response following plaque radiotherapy especially in cases with questionable clinical response.9