Quantitative PET imaging of PD-L1 expression in xenograft and syngeneic tumour models using a site-specifically labelled PD-L1 antibody

Camilla Christensen,Andreas Kjaer,Carsten H Nielsen,Maria Z Alfsen,Lotte K Kristensen

doi:10.1007/s00259-019-04646-4

Camilla Christensen, Andreas Kjaer + Show 3 more

Open Access

https://doi.org/10.1007/s00259-019-04646-4

Copy DOI

Abstract

PurposeDespite remarkable clinical responses and prolonged survival across several cancers, not all patients benefit from PD-1/PD-L1 immune checkpoint blockade. Accordingly, assessment of tumour PD-L1 expression by immunohistochemistry (IHC) is increasingly applied to guide patient selection, therapeutic monitoring, and improve overall response rates. However, tissue-based methods are invasive and prone to sampling error. We therefore developed a PET radiotracer to specifically detect PD-L1 expression in a non-invasive manner, which could be of diagnostic and predictive value.MethodsAnti-PD-L1 (clone 6E11, Genentech) was site-specifically conjugated with DIBO-DFO and radiolabelled with 89Zr (89Zr-DFO-6E11). 89Zr-DFO-6E11 was optimized in vivo by longitudinal PET imaging and dose escalation with excess unlabelled 6E11 in HCC827 tumour-bearing mice. Specificity of 89Zr-DFO-6E11 was evaluated in NSCLC xenografts and syngeneic tumour models with different levels of PD-L1 expression. In vivo imaging data was supported by ex vivo biodistribution, flow cytometry, and IHC. To evaluate the predictive value of 89Zr-DFO-6E11 PET imaging, CT26 tumour-bearing mice were subjected to external radiation therapy (XRT) in combination with PD-L1 blockade.Results89Zr-DFO-6E11 was successfully labelled with a high radiochemical purity. The HCC827 tumours and lymphoid tissue were identified by 89Zr-DFO-6E11 PET imaging, and co-injection with 6E11 increased the relative tumour uptake and decreased the splenic uptake. 89Zr-DFO-6E11 detected the differences in PD-L1 expression among tumour models as evaluated by ex vivo methods. 89Zr-DFO-6E11 quantified the increase in PD-L1 expression in tumours and spleens of irradiated mice. XRT and anti-PD-L1 therapy effectively inhibited tumour growth in CT26 tumour-bearing mice (p < 0.01), and the maximum 89Zr-DFO-6E11 tumour-to-muscle ratio correlated with response to therapy (p = 0.0252).ConclusionPET imaging with 89Zr-DFO-6E11 is an attractive approach for specific, non-invasive, whole-body visualization of PD-L1 expression. PD-L1 expression can be modulated by radiotherapy regimens and 89Zr-DFO-6E11 PET is able to monitor these changes and predict the response to therapy in an immunocompetent tumour model.

Highlights

Immune checkpoint therapy has recently emerged as an effective way of evading the immunosuppressive tumour microenvironment allowing the immune system to eradicate tumours
We evaluated the ability of 89Zr-DFO6E11 to visualize and quantify the therapy-induced changes in programmed death ligand 1 (PD-L1) expression following radiotherapy and the predictive value of 89Zr-DFO-6E11 PD-L1 Positron emission tomography (PET) prior to immune checkpoint blockade of PD-L1
We found the maximum tumour-to-muscle ratios of 89Zr-DFO-6E11 CT26 tumour-bearing mice to correlate with response to antiPD-L1 therapy alone or in combination with radiotherapy

Summary

Introduction

Immune checkpoint therapy has recently emerged as an effective way of evading the immunosuppressive tumour microenvironment allowing the immune system to eradicate tumours. Clinical responses and long-term survival across several tumour types have anchored the clinical utility of immune checkpoint therapy. In many cancers, the response rates are not impressive with a large proportion of non-responding patients. Precise methods to reliably identify patients most likely to benefit from immune checkpoint inhibitors are needed as they have the potential to improve the overall response rates. Existing companion diagnostics to select patients eligible for anti-PD-L1 therapy include ex vivo assessment of PD-L1 expression in tumours by immunohistochemistry (IHC) [3, 4]. Adding to this, assessing responses to immunotherapy by standard RECIST criteria may be challenging compared with conventional chemotherapy as tumour cells are not killed directly and the fact that pseudo progression often is observed in patients receiving immune checkpoint inhibitors [7,8,9]

Methods

Results

Discussion

Conclusion