Metabolic activity by 18F-FDG-PET/CT to predict for early response after nivolumab in previously treated NSCLC.

Abstract

e20615 Background: Nivolumab, an immune checkpoint inhibitor and approved as an anti PD-1 drug, is administered to patients with previously treated non-small cell lung cancer (NSCLC). However, there are no established biomarkers to predict the efficacy of nivolumab. Here, we report a prospective study to investigate whether 2-[18F]-fluoro-2-deoxy-D-glucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) can predict therapeutic response after early-phase nivolumab treatment. Methods: Twenty-four patients who received nivolumab were enrolled from January to September 2016. 18F-FDG-PET/CT was performed before and 1-month after nivolumab therapy. SUVmax, SUVmean, metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were calculated in comparison with CT scan. We also analyzed the protein expression of PD-L1 and identified tumor infiltratrating lymphocytes (TILs) such as CD4-, CD8-, and Foxp3-positive cells within tumor specimens by immunohistochemistry. Correlation of response and nivolumab plasma concentration 1-month post therapy was analyzed. Results: Among the 24 patients, an early metabolic response to nivolumab was observed in 7 patients (29%) on SUVmax, 6 patients (25%) on MTV, and 8 patients (33%) on TLG, whereas 7 (29%) patients achieved a partial response (PR) by RECIST. Out of the 7 patients with confirmed PR, early metabolic response by SUVmax, SUVmean, MTV, and TLG were observed in 6 (85.7%), 5 (71.4%), 5 (71.4%) and 7 (100%) patients, respectively. However, only one patient (14.2%) yielded a PR by CT scan 1-month after nivolumab. All 9 (100%) patients with progressive metabolic disease by TLG had confirmed progressive disease (PD). Progression-free survival of more than 5 months was observed in all patients with early metabolic response by TLG. The morphological change by CT scan was not clearly identified as an early response. No significant correlation was found between metabolic response and nivolumab plasma concentration and expression of PD-L1 and TILs. Conclusions: Metabolic response calculation by TLG could be a better early-phase predictor of efficacy and survival after nivolumab in comparison with immune biomarkers such as PD-L1. Clinical trial information: UMIN000020814.