Article
Hot-Spot areas in FET-PET delineate malignant tumour parts within suspected WHO grade II glioma
Search Medline for
Authors
Published: | May 20, 2009 |
---|
Outline
Text
Objective: Metabolic imaging such as O-(2-18F-fluoroethyl)-L thyrosine (FET) positron emission tomography (PET) often revealed heterogeneous findings in suspected WHO grade II glioma showing one or more HOT-SPOT area(s). To which extent these HOT-SPOTS correlate with malignant tumour parts, has never been elucidated; conversely, it still has to be shown, whether non-HOT-SPOT areas indicate more benign tumour parts. For prospective clarification, a correlative topographical analysis of FET-PET imaging and neuro-pathological findings was conducted.
Methods: Adult patients with an MRI-based suspicion of a supratentorial WHO Grade II glioma were included after prior informed consent. For stepwise non-invasive tumour grading of each tumour, maps of FET uptake kinetics were generated, as described elsewhere (Eur J Nucl Med Mol Imaging, 2007). Tumour parts with suspected malignant behaviour (HOT-SPOT areas) were outlined three-dimensionally. After co-registration of MRI and PET data, stereotactic serial biopsy was performed along a trajectory, which includes the HOT-SPOT volume (if present). Tissue samples were taken in 1 mm steps from within and outside the HOT-SPOT areas. Each of these was scored with regard to its position within the stereotactically defined tumour space, and evaluated independently by two neuro-pathologists unaware of the PET-findings. Grading was done according to WHO criteria and in addition, the Ki-67 labelling index was calculated.
Results: Thirty-seven patients were included. Nineteen samples per tumour (median) were evaluated. 28 tumours showed a benign kinetic (no HOT-SPOT) and stepwise histological evaluation indicated Grade II characteristics in 26 of these tumours. Two tumours exhibited Grade III characteristics. In 9 tumours HOT-SPOT areas were detected: Stepwise histological evaluation revealed Grade II characteristics outside and Grade III characteristics within the HOT SPOT in 6 tumours; in 3 tumours the entire tumour volume was overshadowed by the HOT-SPOT volume and Grade III characteristics were seen throughout these tumours. HOT-SPOT areas were associated with a significant higher Ki-67 labelling (15% vs. 4%, p<0.01). The labelling index outside the HOT-SPOT area was similar to that of tumours lacking HOT-SPOT areas at all (p=0.3).
Conclusions: Suspected low-grade gliomas are heterogeneously composed and HOT-SPOT areas delineate malignant tumour parts. These findings have implications for biopsy and treatment planning.