Combining [11C]-AnxA5 PET Imaging with Serum Biomarkers for Improved Detection in Live Mice of Modest Cell Death in Human Solid Tumor Xenografts

Abstract

Background In vivo imaging using Annexin A5-based radioligands is a powerful technique for visualizing massive cell death, but has been less successful in monitoring the modest cell death typically seen in solid tumors after chemotherapy. Here we combined dynamic positron emission tomography (PET) imaging using Annexin A5 with a serum-based apoptosis marker, for improved sensitivity and specificity in assessment of chemotherapy-induced cell death in a solid tumor model.Methodology/Principal FindingsModest cell death was induced by doxorubicin in a mouse xenograft model with human FaDu head and neck cancer cells. PET imaging was based on 11C-labeled Sel-tagged Annexin A5 ([11C]-AnxA5-ST) and a size-matched control. 2-deoxy-2-[18F]fluoro-D-glucose ([18F]-FDG) was utilized as a tracer of tissue metabolism. Serum biomarkers for cell death were ccK18 and K18 (M30 Apoptosense® and M65). Apoptosis in tissue sections was verified ex vivo for validation. Both PET imaging using [11C]-AnxA5-ST and serum ccK18/K18 levels revealed treatment-induced cell death, with ccK18 displaying the highest detection sensitivity. [18F]-FDG uptake was not affected by this treatment in this tumor model. [11C]-AnxA5-ST gave robust imaging readouts at one hour and its short half-life made it possible to perform paired scans in the same animal in one imaging session.Conclusions/SignificanceThe combined use of dynamic PET with [11C]-AnxA5-ST, showing specific increases in tumor binding potential upon therapy, with ccK18/K18 serum measurements, as highly sensitive markers for cell death, enabled effective assessment of modest therapy-induced cell death in this mouse xenograft model of solid human tumors.

Comparison of methods for evaluating radiolabelled Annexin A5 uptake in pre-clinical PET oncological studies

“…However, in this case it was a priority to be able to correlate the estimations of cell death induction revealed in these tumours with PET with ex vivo tissue and serum analyses, which have been previously reported [6]. In the present study the methods for quantifying the PET data for these inter-individual comparisons are analysed in more depth.…”

“…AnxA5 does bind reversibly to the phosphatidylserine externalized after induction of cell death [27]. Uptake of the [methyl-11 C]-H 6 -AnxA5-ST-CH 3 in head and neck carcinoma xenografts has been observed to reach a maximum at around 30 min, plateau and tend to decrease slightly at the end of a 60 min scan [6]. These characteristics indicated that the compartmental methods for reversible [28,29] rather than irreversible tracers [30] could be more appropriate, which is further evaluated here (see results 3.2).…”

“…SUV was computed for every reconstructed time frame. When the last 25 minutes were used to include more data, the mean uptake was affected slightly, ≈ 2%, by the tendency of [methyl-11 C]-H 6 -AnxA5-ST-CH 3 uptake to decrease after 35 min [6]. Therefore only SUVs calculated at semi-equilibrium, 30 to 35 minutes after the PET scanning was initiated, were used in the comparisons.…”

“…In this study we examine the use of a method for quantifying the behaviour in tumour-bearing mice of a medium-sized (36 kDa) protein AnnexinA5 (AnxA5) labelled with positron-emitting carbon-11 through a C-terminal selenocysteine tag (ST) [6]. The biological basis for the molecular imaging strategy is that cells externalize phosphatidylserine (PS) when they die, especially through apoptosis and necrosis, and that the endogenous human protein AnxA5 binds selectively and with high affinity (1 nM) to PS [7].…”

“…In order to assess non-specific contributions related to the vascularity of the tumours, we used a control protein of similar size to the AnxA5 but with no apparent binding behaviour: mutated thioredoxin fused with green fluorescent protein (mTrx-GFP) which was also similarly labelled with carbon-11 [6]. The macro parameter results for these two proteins are compared to those obtained by conventional uptake assessments in non-treated and chemotherapy-treated tumourbearing mice.…”