A Novel Combination of Bevacizumab and Pembrolizumab Stimulates Tumour Immunity Through Vascular Normalisation in Humanised Mouse Model

Abstract

Background: Immunotherapy has dramatically changed the treatment landscape of cancer. Immunotherapies targeting the programmed death 1/programmed death ligand 1 pathway have been shown to lead to durable responses in patients with a variety of cancers. However, combination approaches are required to extend this benefit beyond a subset of patients. Studies have suggested that anti-angiogenic therapy can elicit or enhance tumor immunity response.Therefore, we hypothesised that combining immunotherapy with anti-angiogenic treatment may have a synergistic effect and may enhance the efficacy of both treatments.Methods: We evaluated the combination of bevacizumab (anti-vascular endothelial growth factor monoclonal antibody) and pembrolizumab (anti-programmed death 1 monoclonal antibody) in two mouse models of human non-small cell lung cancer cell lines (H1299 and A549). We monitored tumour growth and examined changes in the tumour vasculature, along with the frequency and phenotype of tumour-infiltrating lymphocytes.Results: The combination of bevacizumab and pembrolizumab synergistically inhibited tumour growth in vivo without overt toxicity in both cell lines. Combination therapy reprogrammed the immune microenvironment by increasing CD8+ cytotoxic T cell infiltration, thereby turning tumours with different phenotypes into inflamed (‘hot’) tumours. This is potentially mediated by vascular normalisation and endothelial cell activation, as demonstrated by a reduction in the number of microvessels and an increase in adhesion molecules.Conclusions: Taken together, our preclinical studies demonstrate that the combination of bevacizumab and pembrolizumab had a synergistic anti-tumour effect and provides a theoretical basis for translating basic research into clinical applications.

Guiding model‐driven combination dose selection using multi‐objective synergy optimization

“…For the less responsive p53‐wild type A549 NSCLC cells in ref. 21, this corresponds to a model‐predicted dosing range of 3.28 to 87.2 mg/kg for pembrolizumab and 0.19 to 4.76 mg/kg for bevacizumab. All synergy metrics are computed 5 days after the reported data collection window, which corresponds to day .…”

“…Here, we keep the schedule fixed at that used in ref. 21 and seek to find combination doses that solve the multi‐objective SoE and SoP optimization problem. The CI heatmaps using the four synergy metrics are found in Figure S5.…”

“…We also consider an example involving two existing anticancer drugs, a PD‐1 checkpoint inhibitor pembrolizumab, and an anti‐angiogenic agent bevacizumab. Qiao et al 21 reported that this combination inhibited tumor growth compared to monotherapy without overt toxicity in two mouse models of human non‐small cell lung cancer (NSCLC), the p53 wild‐type A549 cells and the p53‐deficient H1299 cells. The animals received either: (1) 10 mg/kg of pembrolizumab starting on day 3, then given every three days (Q3D) until five doses have been administered, (2) 1 mg/kg of bevacizumab starting on day 0, then given every three days until six doses have been administered, or (3) a combination of these two drugs using the same protocol.…”

“…(g) and (j) show fits to control data and treatment TGI using 1 mg/kg of bevacizumab as specified in ref. 21. (h) and (k) show fits to control data and combination TGI using 10 mg/kg of pembrolizumab and 1 mg/kg of bevacizumab as specified in ref.…”

“…(h) and (k) show fits to control data and combination TGI using 10 mg/kg of pembrolizumab and 1 mg/kg of bevacizumab as specified in ref. 21. PK, pharmacokinetic; TGI, tumor growth inhibition.…”