Metabolic Rate of glucose (MRglu) but not standardized uptake value (SUV) can distinguish high risk from low risk uveal melanoma: experience in 20 patients

Aim: Uveal melanoma, the most common primary malignant intraocular tumour, is classified as epithelioid, mixed, and spindle cell type. Based on histopathology, tumours can be divided in high risk (HR, epithelioid and mixed cell type) with worse prognosis, and low risk (LR, spindle cell type) with better prognosis. Up to now, no data exist on the role of 18F-FDG-PET-CT for evaluating the metabolic rate of glucose (MRglu) in these three distinct histological type. Aims of our study were: 1) to assess the feasibility to obtain MRglu (ml/min/100gr) from dynamic 3D 18F-FDG-PET-CT and 2) to evaluate whether epithelioid, mixed, and spindle cell type have different rate of MRglu and/or different SUV values. Materials and Methods: Twenty patients (mean age 62.4± 14 years, 16 M) with large uveal melanoma candidate for surgery underwent ophthalmologic examination, ocular ultrasonography(US), metastatic work-up, and pre-operative dynamic 18F-FDG-PET-CT (185 MBq, PhilipsGXL). 18F-FDG-PET-CT data were analysed semi-quantitatively with SUV, and quantitatively with MRglu. ROIs were drawn over the lesion and over carotid artery for input function. ANOVA test was used for statistical analysis; correlations between SUVmax, SUVmean, MRglu, and tumour volume were also performed. Results: All melanoma (n=20) were visualised at 18F-FDG-PET-CT. The mean value of tumour volume, SUVmax, SUVmean, MRglu were: 1.15± 1.1cc3, 3.34± 1.53, 4.08± 1.94, and 23.73± 14.61ml/min/100gr, respectively. Histopathology proved: 6 epithelioid, 7 mixed, and 7 spindle cell type. Epithelioid and mixed cell type showed significant higher value only in MRglu parameter when compared with spindle cell type: p=0.008 and 0.03 respectively. No significant difference was found between epithelioid and mixed cell type in volume, SUVmax, SUVmean, and MRglu values. According to the risk level, MRglu was significant lower (p<0.013) in LR (n=13) when compared with HR (n=7): 29.4±14.2 mg/min/100gr vs 13.2±8.6mg/min/100gr, while no significant difference was found in SUV values. No correlation was found between SUV and MRglu, SUV and volume, and MRglu and volume. Conclusions: 18F-FDG-PET-CT is a sensitive technique to detect large uveal melanoma and dynamic acquisition allows to easily obtain MRglu. Our results show that epithelioid and mixed cell type uveal melanoma both including epithelioid cells, considered at high risk, have higher glucose consumption than spindle type, without epithelioid cells and considered at low risk. This behaviour is well expressed by different values of MRglu, which gradually increase with increasing malignancy. This finding is not revealed by SUV, which is an “estimate” of glucose consumption, but only by MRglu because it reflects the “real” glucose consumption.