Impact of steroid premedication on atezolizumab (atezo)-induced immune cell activation: A comparative analysis of IMpassion130 and IMpassion131 peripheral blood mononuclear cells (PBMCs)

1083 Background: The immune checkpoint inhibitor (ICI) atezo showed disparate outcomes as first-line therapy for metastatic TNBC when combined with nab-paclitaxel (nPac) in IMpassion130 [Schmid 2018] vs solvent-based paclitaxel (Pac) in IMpassion131 [Miles 2021]. A key difference between the trials was use of steroid premedication for Pac in IMpassion131 but not for nPac in IMpassion130. In patients (pts) receiving ICIs, prior steroid exposure has been linked to worse outcome [Drakaki 2020]. Further, IMpassion130 and IMpassion131 subgroup analyses suggested reduced atezo treatment effect in taxane-pretreated pts. This post hoc biomarker study explored the impact of: 1) steroids on systemic immune cell activation with atezo+taxane; 2) prior taxane exposure on atezo-induced immune cell activation. Methods: PBMCs collected at baseline and at day 1, cycle 2 (wk 4) were selected from matched pts (RECIST responders, PD-L1+, no liver metastases) from IMpassion130 and IMpassion131. Single-cell RNAseq was performed and the transcriptomic profile of immune cells was analyzed using GSEA pathway analyses, cell proportion and TCR clonality. Results: CITEseq from 695,851 single cells was generated from 39 IMpassion130 PBMC pairs (29 atezo+nPac; 10 placebo [Pla]+nPac) and 35 IMpassion131 pairs (26 atezo+Pac; 9 Pla+Pac). At wk 4, atezo+nPac resulted in increased IFNα and IFNγ responses across multiple cell types (CD4+ and CD8+ T cells, B cells, NK cells and monocytes) and proliferation in NK cells, but reduced TNF signaling across multiple cell types. In contrast, 4 wks of Pla+nPac increased TNF signaling but decreased IFNα and IFNγ responses. In the presence of steroids, 4 wks of atezo+Pac also increased IFNα and IFNγ responses mainly in NK cells and monocytes, but not T cells, and reduced proliferative pathways across B and T cells and TNF signaling. Pla+Pac increased TNF signaling only in NK cells, but reduced proliferative signatures across cell types. The only on-treatment change differing significantly between atezo+nPac vs atezo+Pac was the increase in proliferation pathways in NK, T and B cells with atezo+nPac. There were no significant changes in proportions of cell subsets or TCR clonality. PBMCs from taxane-pretreated pts had higher RNA-based metabolic profile (OxPhos, DNA repair and IFNα). Taxane-naive but not taxane-pretreated immune cells had increased IFNγ and IFNα response after atezo+taxane. Conclusions: Our results suggest that atezo+taxane promotes IFNα and IFNγ responses and that steroid co-administration reduces proliferation pathways across immune cells. Prior taxane exposure was associated with an increased metabolic status, possibly rendering immune cells less sensitive to atezo-induced activation. The immune context of taxane-naive TNBC may result in more potent immune activation with atezo. Clinical trial information: NCT02425891 and NCT03125902.