(18)F-BMS986192 PET Imaging of PD-L1 in Metastatic Melanoma Patients with Brain Metastases Treated with Immune Checkpoint Inhibitors: A Pilot Study

Immune checkpoint inhibitors (ICIs) targeting programmed cell death protein-1 (PD-1)/programmed cell death ligand-1 (PD-L1) frequently induces tumor response in metastatic melanoma patients. However, tumor response often takes months and may be heterogeneous. Consequently, additional local treatment for nonresponsive metastases may be needed, especially in the case of brain metastases. Noninvasive imaging may allow the characterization of (brain) metastases to predict response. This pilot study uses (18)F-BMS986192 PET for PD-L1 expression to explore the variability in metastatic tracer uptake and its relation to tumor response, with a special focus on brain metastases. Methods: Metastatic melanoma patients underwent whole-body (18)F-BMS986192 PET/CT scanning before and 6 wk after starting ICI therapy. (18)F-BMS986192 uptake was measured in healthy tissues, organs, and tumor lesions. Tumor response was evaluated at 12 wk using CT of the thorax/abdomen and MRI of the brain. RECIST, version 1.1, was used to define therapy response per patient. Response per lesion was measured by the percentage change in lesion diameter. Toxicity was assessed according to Common Terminology Criteria for Adverse Events, version 4.0. Results: Baseline (18)F-BMS986192 PET/CT was performed in 8 patients, with follow-up scans in 4 patients. The highest tracer uptake was observed in the spleen, bone marrow, kidneys, and liver. Tracer uptake in tumor lesions was heterogeneous. In total, 42 tumor lesions were identified at baseline, with most lesions in the lungs (n = 21) and brain (n = 14). Tracer uptake was similar between tumor locations. (18)F-BMS986192 uptake in lesions at baseline, corrected for blood-pool activity, was negatively correlated with the change lesion diameter at response evaluation (r = −0.49, P = 0.005), both in intra- and extracerebral lesions. Receiver-operating-characteristic analysis demonstrated that (18)F-BMS986192 uptake can discriminate between responding and nonresponding lesions with an area under the curve of 0.82. At the follow-up scan, an increased (18)F-BMS986192 uptake compared with baseline scan was correlated with an increased lesion diameter at response evaluation. In the follow-up (18)F-BMS986192 PET scan of 2 patients, ICI-related toxicity (thyroiditis and colitis) was detected. Conclusion: In this pilot study, (18)F-BMS986192 PET showed heterogeneous uptake in intra- and extracerebral metastatic lesions in melanoma patients. Baseline (18)F-BMS986192 uptake was able to predict an ICI treatment–induced reduction in lesion volume, whereas the follow-up PET scan allowed the detection of treatment-induced toxicity.