Abstract
BACKGROUND: Immune-checkpoint inhibitors (ICIs) have been increasingly used for non-small cell lung cancer (NSCLC) treatment in recent years. Although insufficient, the rate of programmed death-ligand 1 expression has been adopted as a predictor of ICI efficacy. We evaluated tumor growth rate as a clinically easy-to-use predictor of the therapeutic effect of ICIs. METHODS: This study is a single-institution retrospective study in Japan. NSCLC patients treated with nivolumab, pembrolizumab, or atezolizumab at Saitama Medical Center from January 1, 2016 to December 31, 2018 were enrolled, and followed until December 31, 2020. We defined and calculated the initial rapidity of tumor progression (IRP) as: the increase in the sum of the diameters of intrathoracic tumors and lymph nodes on two series of chest computed tomography (CT) scans (one obtained at an initial checkup and the other obtained immediately before the first treatment) divided by the number of days between these CT scans. Two coefficients were calculated: the maximal information coefficient (MIC) between IRP and time to treatment failure (TTF) using the Python package with minepy library, and the Spearman's rank correlation coefficient. RESULTS: A total of 55 patients (median age, 70 years; 47 men) were enrolled. The median TTF with ICIs was 126 days, and four patients continued to receive ICI treatment at the end of the follow-up. The MIC between IRP and TTF was 0.302 with weak correlation, and the Spearman's rank correlation coefficient was -0.347 (P=0.00938). CONCLUSIONS: The initial tumor growth rate had a negative linear correlation with the therapeutic effect of ICIs.