Nivolumab timing as a major survival predictor in patients with stage IV non–small cell lung cancer.

9058 Background: Functional activity and trafficking of T(CD8) and other immune cells are regulated over the 24 hours by the circadian timing system. Nivolumab (NIV) binds to PD-1 receptors and targets T(CD8). These pharmacologic effects could be regulated by circadian clocks, hence suggesting possible daily changes in its efficacy, as substantiated for immune checkpoint inhibitors (ICIs) in two prior reports. Methods: Here we searched whether increasing the proportion of morning NIV infusions could critically improve efficacy in consecutive metastatic Non-Small Cell Lung Cancer (NSCLC) patients (pts). Pts received NIV (240 mg iv q 2 weeks) at a daily time that was ‘randomly’ allocated for each course by the day-hospital coordinators. The median time of actual NIV infusions was computed for each pt. The study population was split into three timing-groups: a ‘morning’ one, where the pts received at least 2/3 of NIV infusions before 12:54, i.e. the median time of all NIV infusions; an intermediate group, where the pts received at least 1/3 of NIV infusions before and 1/3 after 12:54; and an ‘evening’ group, where pts received at least 2/3 of NIV infusions after 12:54. CTCAE-toxicity rates, iRECIST-tumor responses, progression-free survival (PFS), and overall survival (OS) were computed according to NIV timing group. Results: 95 previously-treated stage IV NSCLC pts (M/F,79/16; PS 0-1, 96%; 41-83 years old) were retrospectively allocated to a ‘morning’ group (36 pts), an ‘intermediate’ group (24 pts), and an ‘evening’ group (35 pts). Pts received NIV as 2 nd line (76%). Tumor PD-L1 status was positive for 39 of 72 pts (54%). Main metastatic sites were bone (52% of the pts), pleura (41%), liver (25%), brain (24%), and adrenal gland (20%). Pt characteristics were similar in the 3 groups, except for liver metastases (41.7%, 8.3% and 25.7% for ‘morning’, ‘intermediate’ and ‘evening’ groups respectively, p = 0.010). Grade 2-4 fatigue was least in the ‘morning’ group (28%) vs 62% (‘intermediate’) and 40% (‘evening’) (p = 0.027). Median PFS (months) was 11.1 for the ‘morning’ group, 5.9 for the ‘intermediate’ group, and 3.1 for the ‘evening’ group (p = 0.002). Median OS (months) [95% C.L.] was 34.2 [ - ] for the ‘morning’ group, 15.3 [8.0 – 22.7] for the ‘intermediate’ group, and 12.4 [4.0 – 20.7] for the ‘evening’ group (p = 0.023). Respective 2-years survival rates were was 52.6%, 26.2% for the and 15.0% (p = 0.002). Multivariable analysis confirmed that the administration of >2/3 NIV in the morning predicted for longer PFS (Hazard ratio, 0.26 [0.14-0.51], p < 0.001) and OS (0.22 [0.10-0.51], p < 0.001). Conclusions: NIV was largely more effective in the ‘morning’ as compared to the ‘intermediate’ or ‘evening’ groups, with no apparent bias. Randomized and translational circadian timing studies are needed to unravel the mechanisms at work in the chronopharmacology of ICI’s, so as to minimize the risk of resistance and to maximize therapeutic benefits.