I'd be willing to have a crack at this if I thought it could help (my background is in nuclear methods for materials characterisation). But I've a feeling that the difference in uptake between cancerous and normal tissue might not be large enough to make this especially useful. Fluorine-18 labelled glucose is used in PET imaging, but not for treatment as far as I'm aware. On the other hand, if a compound can be found that is more selective for the cancer in question, making it radioactive may offer a further improvement.

Isotope labeling for imaging requires only a low isotope purity. One radioactive glucose in a vast sea of normal glucose is just fine, because as I'm sure you know the energies of the emitted particles make a very distinct marker.

What about an enzymatic synthesis? Or even partial biosynthesis.

Just grow some sprouts in it?

(Not an endorsement of the idea)