Prognostic value of the timing of immune checkpoint inhibitors infusion in patients with advanced hepatocellular carcinoma.

Abstract

457 Background: Experimental and clinical data in melanoma and lung cancer suggest that the response to immune checkpoint inhibitors (ICI) may be influenced by circadian rhythms. However, similar findings are lacking in hepatocellular carcinoma (HCC). The aim of this study was to assess the impact of infusion timing of atezolizumab, an ICI, associated or not with bevacizumab, as first line systemic treatment in advanced HCC. Methods: We retrospectively analyzed all consecutive patients, who received, as first line systemic treatment, at least two cycles of atezolizumab +/- bevacizumab for advanced HCC in Paul Brousse Hospital, Villejuif, France, between January 2020 and October 2022. We split them in two groups: 1) patients receiving the first two cycles after 1 pm (afternoon group); 2) patients receiving at least one of the two first cycles before 1 pm (morning or mixed group). Cox regression models were used to assess the prognostic value of ICI infusion timing for the overall survival (OS). Results: A total of 131 patients were analyzed. 119 patients (90.8%) were male, with a median age of 70 years [35; 88]; 40 patients (31%) had performance status (PS) ≥ 1. 103 patients (79%) had cirrhosis; 61 patients (46.6%) had portal hypertension. 49 patients (37.4%) had metastatic HCC and 52 patients (40.6%) had portal vein thrombosis at diagnosis. The mean initial level of alpha-fetoprotein was 5,212.8 ng/mL (SD: 18,681.8). 36 patients (27.5%) received only atezolizumab. 41 patients were included in the first group and were compared to 90 patients included in the second group. There were no significant clinical differences between the two groups. The median OS was significantly lower in the afternoon group compared to the morning or mixed group (11.5 months vs 18.7 months, p = 0.015). In univariate analysis, PS ≥ 1 (p = 0.025), initial level of alpha-fetoprotein (p = 0.030), monotherapy by atezolizumab (p = 0.004) and the first two cycles of ICI infusion after 1 pm (p = 0.018) were significantly correlated with OS. The first two cycles of ICI infusion after 1 pm remains an independent factor significantly associated with OS in the multivariate Cox model, with HR 2.29, 95% CI: 1.27-4.14, p = 0.005 (Table). Conclusions: Our results suggest that the afternoon infusion of the first two cycles of ICI is associated with lower outcomes in patients with advanced HCC, independently of other prognostic factors. There may be confounding factors due to the variability of administration of further perfusions. Larger data are needed to confirm such findings. [Table: see text]