Imaging astrocytes


18F-THK5351 PET Can Identify Astrogliosis in Multiple Sclerosis Plaques. Ishibashi K, Miura Y, Hirata K, Toyohara J, Ishii K. Clin Nucl Med. 2019 Jul 24. doi: 10.1097/RLU.0000000000002751. 

A 26-year-old woman with relapsing-remitting multiple sclerosis (MS) underwent F-THK5351 PET during a remission period. PET imaging showed that small regions with elevated uptake of F-THK5351 were scattered in the brain and that the foci of F-THK5351 accumulations corresponded anatomically to MS plaques identified by MRI. Because F-THK5351 binds to monoamine oxidase B highly expressed in astrocytes, F-THK5351 accumulates in lesions undergoing astrogliosis. Hence, elevated uptake of F-THK5351 in the present case can represent ongoing astrogliosis in inactive MS lesions (plaques).

I read this post with a bit of excitement as it says that it is possible to image astrocytes, using a positron emission topography agent. This is a low grade radioactive drug that can bind to astrocytes. The PET scanner can detect this. They say this drug binds to the monoamine oxidase B receptor. Then I start to be more concerned. A quick look on “Brain Seq” ((CNS tissue RNA expression website) revealed that whilst this could be astrocyte selective, it is not astrocyte specific and could label other cells such as neurons. A quick look at “protein atlas” (protein expression website) for MAOB expression in the brain and it is clearly expressed by more than astrocytes.

This sounds like it could be like TSPO, hailed as being a microglial imager but clearly binding to more than just hot microglia,

Furthermore, they say this “images astrogliosis”, but in the absense of them killing the individual imaged to do some histology, how can they know this? This is a common problem of using imaging of an unknown pathology to define what the image is. Let’s see how this fairs.

About the author



Leave a Reply to MouseDoctor Cancel reply

    • You are looking for an astrocyte marker, with the resolution of PET you would be able to see accumulation of astrocytes, As for a marker I dont know which is best but Aquarporin 4 is expressed by astrocytes the where the FPKM for AQP4 is over 300-400.

      However I do remember staining mice for interleukin 6 about thirty years ago and it was amazing at picking up gliotic astrocytres and didn’t really come up on normal astrocytes. However, it also picks up macrophages and so isnt specific.

  • Many of the receptors targeted by current radiodiagnostic ligands are not 100% specific to the target of interest. However, this can be accounted for by the standardised uptake values and anatomical MRI data. Just because neurons also express MOAB doesn’t mean this tracer holds no diagnostic or prognostic value. The problem is that there are very few companies that are developing radiodiagnostics through to market. Most languish as research tools compounded by radiopharmacists and are never fully explored. Currently, the development costs far outweigh their market value.

    • Is it because they are radioactive no-one wants them. I remember having a radioactive source in the room next door. We had asked for it to be moved may times, when builders came we said don’t go in that room got a Geiger counter which started to hiss/buzz the builders were off the job in minutes the source was moved the next day.

      • In a pre-clinical setting sure but in a clinical setting most large hospitals have radiopharmacies that deal with beta emitters daily. There are other problems that hinder development of radiodiagnostics, theranostics etc. For example, just producing the radiometal can be difficult. Most reactors only spit out these metals on select days and few cyclotron produced metals are manufactured to GMP standard. Then you have the problem of half-life and getting enough product with sufficient specific activity to hospital in time (gallium generators excluded). This is all very costly. All for a diagnostic product that will never provide the return on investment the pharmaceutical pirates expect.

        Concerning the use of PET ligands in MS, there is evidence suggesting radiation exposure increases demyelination (but at much higher radiation doses than with PET). Nevertheless, safety could also be a concern…?

        • Most of the CNS PET ligands use radioisotopes with half lives in the order of minutes eg 18F 99mins, 11C 20 mins so an on-site cyclotron is needed, some but not many hospitals do have them. Technetium and gallium aside, not readily available. But specific ligands for targets also needed. Radiation effects unlikely to be a problem from amounts used and short half life.

  • “but in the absense of them killing the individual imaged to do some histology, how can they know this?”…..

    This is the biggest hurdle to overcome……

By MouseDoctor



Recent Posts

Recent Comments