For the assessment of the efficacy of clinical gene therapy trials, different imaging modalities have been developed that enable a noninvasive assessment of location, magnitude, and duration of transduced gene expression in vivo. These imaging methods rely on a combination of an appropriate marker gene and a radiolabeled or paramagnetic marker substrate that can be detected by PET or MRI. Here, we assess whether the nucleoside analog 2'-fluoro-2'-deoxy-1beta-D-arabinofuranosyl-5-iodouracil (FIAU), a specific marker substrate for herpes simplex virus type 1 thymidine kinase (HSV-1-tk) gene expression, penetrates the blood-brain barrier (BBB) as an essential prerequisite for a noninvasive assessment of HSV-1-tk gene expression in gliomas.
Methods: No-carrier-added [(124)I]FIAU was synthesized by reacting the precursor 2'-fluoro-2'-deoxy-1beta-D-arabinofuranosyluracil (FAU) with carrier-free [(124)I]NaI. The course of biodistribution of [(124)I]FIAU was investigated in anesthetized cats (n = 3; organs) and in one patient with a recurrent glioblastoma (plasma and brain) by PET imaging over several hours (cats, 1-22 h) to several days (patient, 1-68 h). FIAU PET was performed in conjunction with multitracer PET imaging (cerebral blood flow and cerebral metabolic rate of O(2) in cats only; cerebral metabolic rate of glucose and [(11)C]methionine in all subjects). A region-of-interest analysis was performed on the basis of coregistered high-resolution MR images. The average radioactivity concentration was determined, decay corrected, and recalculated as percentage injected dose per gram of tissue (%ID/g) or as standardized uptake values (SUVs).
Results: The average chemical yield of [(124)I]FIAU synthesis was 54.6% +/- 6.8%. The chemical and radiochemical purities of [(124)I]FIAU were found to be >98% and >95%, respectively. In cats, the kinetic analysis of [(124)I]FIAU-derived radioactivity showed an early peak (1-2 min after injection) in heart and kidneys (0.20 %ID/g; SUV, 4.0) followed by a second peak (10-20 min after injection) in liver and spleen (0.16 %ID/g; SUV, 3.2) with subsequent clearance from tissues and a late peak in the bladder (10-15 h after injection). In the unlesioned cat brain, no substantial [(124)I]FIAU uptake occurred throughout the measurement (<0.02 %ID/g; SUV, <0.4). In the patient, [(124)I]FIAU uptake in normal brain was also very low (<0.0002 %ID/g; SUV, <0.16). In contrast, the recurrent glioblastoma revealed relatively high levels of [(124)I]FIAU-derived radioactivity (5-10 min after injection; 0.001 %ID/g; SUV, 0.8), which cleared slowly over the 68-h imaging period.
Conclusion: The PET marker substrate FIAU does not penetrate the intact BBB significantly and, hence, is not the marker substrate of choice for the noninvasive localization of HSV-1-tk gene expression in the central nervous system under conditions in which the BBB is likely to be intact. However, substantial levels of [(124)I]FIAU-derived radioactivity may occur within areas of BBB disruption (e.g., glioblastoma), which is an essential prerequisite for imaging clinically relevant levels of HSV-1-tk gene expression in brain tumors after gene therapy by FIAU PET. For this purpose, washout of nonspecific radioactivity should be allowed for several days.