PET pharmacokinetics of [¹⁸F]F-AraG as a marker of T cell activation during immunotherapy with a STING agonist with and without anti-PD-1.

BACKGROUND

There is an urgent need for clinical biomarkers to monitor the mechanistic efficacy of immunotherapy in oncology. The [18F]F-AraG tracer, developed to image activated T cells, may provide the means to establish such a biomarker. This study extends previously reported pharmacokinetic findings in untreated NSCLC across additional tumor types undergoing immunotherapy, and examines [18F]F-AraG kinetics in various organs. Four patients with advanced solid tumors were included in this phase 1 trial investigating a stimulator of interferon genes agonist with anti-programmed cell death protein 1. Patients underwent a 90 min [18F]F-AraG PET/CT scan at baseline, one week and six weeks on-treatment. The tracer kinetics in tumors and organs were modeled using three compartment models. Model selection was based on the Akaike information criterion (AIC), visual assessment of the fit quality, and the reliability of estimated kinetic parameters. The relationship between the primary kinetic measures (influx rate (Ki), volume of distribution (VT)) and the simplified uptake metrics (standardized uptake values (SUV), tumor-to-blood ratio (TBR) and tumor-to-plasma ratio (TPR)) were evaluated using scatter plots.

RESULTS

The kinetics of all tumors and most organs were adequately described with the two-tissue irreversible model (2T3K), which was identified as the preferred model based on the AIC in 40/55 tumor regions. For most organs, the 2T3K model was selected. We observed a consistent linear trend across all metrics for the relationship between primary kinetic parameters and simplified uptake metrics. However, a patient-specific bias can be observed for SUV and TBR, which was corrected for when using the TPR.

CONCLUSION

This study confirms that [18F]F-AraG PET uptake in tumors is best described by a 2T3K model, with no treatment-induced effects observed. Quantification of [18F]F-AraG uptake in static scans is most reliable when normalized using the TPR.

TRIAL REGISTRATION: ClinicalTrials.gov at 14 February 2019, NCT03843359, https//clinicaltrials.gov/study/NCT03843359.