Intravenous Oncolytic Vaccinia Virus Therapy Results in a Differential Immune Response between Cancer Patients

Abstract

Pexa-Vec is an engineered Wyeth-strain vaccinia oncolytic virus (OV), which has been tested extensively in clinical trials, demonstrating enhanced cytotoxic T cell infiltration into tumours following treatment. Favourable immune consequences to Pexa-Vec include the induction of an interferon (IFN) response, followed by inflammatory cytokine/chemokine secretion. This promotes tumour immune infiltration, innate and adaptive immune cell activation and T cell priming, culminating in targeted tumour cell killing, i.e., an immunologically 'cold' tumour microenvironment is transformed into a 'hot' tumour. However, as with all immunotherapies, not all patients respond in a uniformly favourable manner. Our study herein, shows a differential immune response by patients to intravenous Pexa-Vec therapy, whereby some patients responded to the virus in a typical and expected manner, demonstrating a significant IFN induction and subsequent peripheral immune activation. However, other patients experienced a markedly subdued immune response and appeared to exhibit an exhausted phenotype at baseline, characterised by higher baseline immune checkpoint expression and regulatory T cell (Treg) levels. This differential baseline immunological profile accurately predicted the subsequent response to Pexa-Vec and may, therefore, enable the development of predictive biomarkers for Pexa-Vec and OV therapies more widely. If confirmed in larger clinical trials, these immunological biomarkers may enable a personalised approach, whereby patients with an exhausted baseline immune profile are treated with immune checkpoint blockade, with the aim of reversing immune exhaustion, prior to or alongside OV therapy.

Keywords: biomarkers; differential immune response; immune checkpoint blockade; immune exhaustion; immunotherapy; interferon response; oncolytic virus therapy; vaccinia virus.

Figure 1

Peripheral immune response to Pexa-Vec infusion. (A) Trial schema showing timing of virus infusion and peripheral blood sample collection. Differential IFN (B) and inflammatory cytokine (C) response to Pexa-Vec in Responder (black; n = 4) and Exhausted (white; n = 5) patients. Data is shown as fold-change from baseline (D1 pre); *** q-value < 0.001, **** q-value < 0.0001.

Figure 2

Immune cell redistribution in response to Pexa-Vec infusion. (A) Differential chemokine response in Responder (black; n = 4) and Exhausted (white; n = 4/5) patient plasma to Pexa-Vec infusion. Data is shown as fold-change difference from baseline (D1 pre) (**** q-value < 0.0001). (B) Total lymphocyte count and (C) individual immune cell populations (CD4⁺ T cells, CD8⁺ T cells, NK cells, NKT cells) both represented by fold-change difference between D2 and baseline (D1 pre) in Responder (black; n = 4) and Exhausted (white; n = 5 (A) and n = 4 (B)) patients (* p < 0.05). (D) Representative images of CD8 T cells in Responder and Exhausted tumour (CD8-positive cells are visualised by Fast Red staining). Bars represent 100 µm.

Figure 3

Differential activation of immune cell populations following Pexa-Vec infusion. Differential expression of (A) CD69 in immune cell populations, (B) NK CD107 expression (representing NK cell degranulation), and (C) PD-L1 expression in immune cell populations in Responder (black; n = 4) and Exhausted (white; n = 4/5) patients at baseline (D1 pre) and following Pexa-Vec infusion. Data is shown as fold-change difference from baseline (D1 pre) for % positive expression (* p < 0.05, ** p < 0.01).

Figure 4

Baseline indicators of predicted response to Pexa-Vec therapy. Differential baseline levels of (A) inflammatory cytokines and (B) chemokines in patient plasma and (C) PD-L1 expression on immune cell populations in Responder (black; n = 4) and Exhausted (white; n = 5) patient samples. Data is shown as pg/mL (A,B) or % positive expression for D1 pre samples (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). (D) Relative frequency of Tregs in PBMCs at baseline (D1 pre) in Responder (black; n = 4) and Exhausted (white; n = 5) patients. Data is shown as % Tregs of whole PBMCs; * p < 0.05.