Conventional MRI often fails to distinguish between progressive tumour and radiation injury, because both appear as mass lesions with unspecific Gd-DTPA enhancement. Furthermore, the sensitivity of FDG PET for the evaluation of malignant lesions in the brain is limited owing to high cortical uptake. The aim of this study was to assess the potential of alternative SPET tracers in the same group of patients. 35.2+/-20.1 months after stereotactic radiotherapy (59.3+/-4.2 Gy) of low-grade astrocytomas (median WHO II), 16 patients, presenting 25 Gd-DTPA-enhancing lesions on MRI, were examined by SPET. Lesions were classified as progressive tumour (PT, n=17) or non-PT (nPT, n=8) based on prospective follow-up (clinical examination, MRI, proton-MR spectroscopy) for 25.6+/-6.7 months after SPET. SPET scans were performed 15 and 60 min after injection of 694+/-67 MBq hexakis(2-methoxyisobutylisonitrile)(99m)Tc(I) (MIBI). 3-[(123)I]iodo-alpha-methyl- L-tyrosine (IMT) SPET was acquired 15 min after injection of 291+/-58 MBq IMT. Lesion-to-normal tissue ratios (l/n) for IMT (l/n(IMT)) and MIBI (l/n(MIBI)) were calculated using a reference region mirrored to the contralateral hemisphere. Using IMT, significantly higher ratios ( P<0.001) were found in PT (1.7+/-0.4) than in nPT (1.1+/-0.1). For MIBI, there was no statistically significant difference ( P=0.206) between PT (3.7+/-2.8) and nPT (1.8+/-1.8). Sensitivities for MIBI and IMT were 53% and 94%, and specificities 75% and 100%, respectively. Positive predictive values for MIBI and IMT respectively reached 80% and 100%, and negative predictive values were 46% and 90%. In conclusion, in contrast to MIBI, IMT showed almost no overlap between the PT and the nPT group. The sensitivity, specificity and predictive values of IMT SPET were obviously higher than those of MIBI SPET. IMT is considered to be a useful tracer for differentiating PT from nPT in the follow-up of irradiated low-grade astrocytomas.