Malignant gliomas are highly aggressive brain tumors. Surgical removal and chemoradiation of the tumor are the standard of care. Recently, the U.S. Food and Drug Administration (FDA) approved a compound called 5-aminolevulinic acid (5-ALA) as an imaging agent to aid in differentiating tumor from normal tissue during surgery. 5-ALA is a precursor in the heme biosynthesis pathway, which is inefficient in glioma cells because their strongly rewired metabolism does not rely on heme. When patients with malignant glioma ingest 5-ALA prior to surgery, the glioma cells fluoresce pink under a blue light due to their preferential uptake and conversion of 5-ALA to the final precursor in heme biosynthesis, the fluorescent molecule protoporphyrin IX (PpIX). We sought to investigate whether extracellular vesicles (EVs) released from PpIX-enriched glioma cells would fluoresce and be detectable in the blood of these patients.
We employed Amnis® Imaging Flow, which combines flow cytometry and microscopy to detect PpIX-positive EVs. We first determined the optimal 5-ALA dose to maximize fluorescence and minimize cell death. We used a combination of beads of different size (100-500nm) and liposomes with different emission spectra to ensure that the signal emitted in Channel 11 (~640nm) of the Amnis® output was indeed from PpIX, and that all other channels reported no signal. Controls also included lysis with Triton-X of liposomes and EVs.
Importantly, we showed that glioma cells released a significantly higher number of PpIX-positive EVs (247-fold increase) than normal endothelial cells (6-fold increase) after 5-ALA ingestion. We also used xenograft mouse models to show that the presence of PpIX-positive EVs in circulating plasma after 5-ALA ingestion correlated strongly with the presence of a primary brain tumor, while the signal from the plasma of normal control mice remained below background both before and after 5-ALA ingestion.
Finally, we tested the optimized assay in the plasma of patients with gliomas undergoing 5-ALA fluorescence guided surgery at the Massachusetts General Hospital. Samples were collected prior to 5-ALA intake as well as at the time of surgery, prior to tumor removal. Pre- and post-5-ALA plasma samples were kept in the dark to avoid bleaching of the PpIX signal, as were the patients for 24 hours post 5-ALA. We collected samples from 4 patients whose tumors were avidly fluorescent during surgery and 2 patients whose tumors showed minimal fluorescence. Interestingly, we detected PpIX-positive EVs only in the plasma samples from patients whose tumors were avidly fluorescent. Finally, when we compared the fold increase (pre/post-5-ALA) in PpIX-positive signal to the size of the tumor, we found a clear correlation, suggesting that the detected events are likely coming directly from the tumor. This is the first time intracranially derived EVs have been quantified in circulating plasma, and this development opens the door for many exciting studies that can shed light on brain-derived EV dynamics and half-life. For example, we detected between 3,000 and 8,000 PpIX-positive events per mL of plasma. Assuming each 1 mL of plasma contains roughly 1010 EV/mL, we can deduct that only 0.00008% of EVs in blood are of glioma tumor origin. Furthermore, this assay allows us to study EV dynamics in tumor patients undergoing therapy as well as determine the effects of medications such as dexamethasone on the release of EVs into the bloodstream.
Clinically, there is a major need for minimally invasive diagnosis of brain cancer, and characterizing circulating tumor-specific fluorescent EVs provides a window into the primary tumor’s presence and status. Detecting and characterizing fluorescent EVs after administering 5-ALA allows for diagnosis and potentially monitoring of malignant gliomas over time.
Jones PS, et al. Characterization of plasma-derived protoporphyrin-IX-positive extracellular vesicles following 5-ALA use in patients with malignant glioma. (2019) eBioMedicine 48:23-35. doi: 10.1016/j.ebiom.2019.09.025. PMID: 31628025.