Re: Increased ¹⁸F-2-Fluorodeoxysorbitol (¹⁸F-FDS) Activity in a Pituitary Spindle Cell Carcinoma

To the Editor

We read the report by Cheng et al describing ¹⁸F-2-fluorodeoxyglucose (¹⁸F-FDG) and ¹⁸F-2-fluorodeoxysorbitol (¹⁸F-FDS) PET to differentiate malignancy from inflammation in a 33-year old woman with a pituitary spindle cell carcinoma.¹ Li et al were the first to synthesize ¹⁸F-FDS, and reported visualization of both glioblastoma xenografts (U87MG) as well as inflammatory foci in mice, at early time-points after tracer injection.² It appears that the strong implication by Cheng et al that ¹⁸F-FDS has the potential to differentiate malignancy from inflammation is admirable in enthusiasm but may be factually inconsistent with the current state of the art.

Cheng et al reported PET SUVmax of 4.73 (¹⁸F-FDG) and 1.49 (¹⁸F-FDS) respectively, at the site of the pituitary spindle cell carcinoma.¹ However, Li et al reported that ¹⁸F-FDS exhibits very poor in vitro uptake by mammalian cells – 0.1% for ¹⁸F-FDS versus up to 40% for ¹⁸F-FDG in U87MG cells.² There are no transporters for ¹⁸F-FDS entry into mammalian cells and the substitution of the hydroxyl group by fluorine at the 2C-position completely abrogates the recognition by mammalian enzymes.³ Moreover, Li et al visualized tumors and inflammation in mice by ¹⁸F-FDS PET in the context of high background at earlier time points after tracer injection (5–60 min). Therefore, given the lack of specific-uptake mechanisms, they suggested that increased blood flow and leaky vasculature were responsible for visualization of xenografts (U87MG) as well as inflammatory foci.² These findings are also consistent with the data by Weinstein and Ordonez et al who also demonstrated the lack of ¹⁸F-FDS uptake in healthy mammalian or cancer cells.⁴ Interestingly, Weinstein and Ordonez et al reported substantial (~1000-fold higher than mammalian cells) and specific uptake of ¹⁸F-FDS in Gram-negative bacteria (Enterobacteriaceae). At 120 min after ¹⁸F-FDS injection in mice, infectious foci due to these bacteria were clearly visualized by PET, but no appreciable signal was noted at the sites of (sterile) inflammation or U87MG brain xenografts.⁴ Furthermore, dynamic ¹⁸F-FDS PET studies in mice with U87MG brain tumors demonstrated some initial signal, which however dissipated 60–120 min after tracer-injection.

In conclusion, ¹⁸F-FDS may be useful to detect infectious foci due to Enterobacteriaceae.⁴ While the sensitivity and specificity of this technique in the clinical setting remains to be determined, it is likely that ¹⁸F-FDS uptake at the sites of sterile inflammation and malignancy (as demonstrated by Cheng et al), is consistent with a non-specific blood pool effect; that is, capillary leak at the site of inflammation or malignancy.