Intravesical BCG in Patients with Non-Muscle Invasive Bladder Cancer Induces Trained Immunity and Decreases Respiratory Infections - Beyond the Abstract

Bacillus calmette–guérin (BCG) is the recommended intravesical immunotherapy for patients with high-risk non-muscle invasive bladder cancer (NMIBC) but a substantial number of patients do not show a good clinical response to BCG. There is a need for an increased understanding of the mechanisms and characteristics that affect the efficacy of BCG, including the role of ‘trained immunity’. Vaccination with the BCG vaccine induces ‘trained immunity’.

Trained immunity is a de facto immune memory for innate immune cells and is characterized by long-term functional and epigenetic reprogramming. Trained immunity allows innate immune cells to respond with an increased cytokine response upon encountering a secondary stimulus or challenge.¹ Vaccination with BCG protects against a wide range of non-tuberculous infectious diseases² and this heterologous protection induced by the BCG vaccine is mediated via trained immunity.³ Currently it is unknown if, and for how long, trained immunity is induced after intravesical BCG treatment in NMIBC patients and if BCG-induced trained immunity affects the clinical response.

In our work recently published in the Journal for ImmunoTherapy of Cancer⁴ we extensively investigated the induction of trained immunity by BCG instillations in NMIBC. We i) evaluated the presence of a long-term induction of systemic trained immunity in blood monocytes of BCG-treated NMIBC patients, ii) evaluated whether BCG instillations result in reduced incidence of respiratory infections, as has been shown after BCG vaccination and would be expected in case of training, and iii) explored the relation between BCG-induced trained immunity and oncological outcome in NMIBC.

We set up a prospective observational study in 17 high-risk NMIBC patients and drew blood at 9 time points during the first year of BCG therapy (Figure 1). The peripheral blood mononuclear cells (PBMCs) were isolated and stimulated ex vivo with Lipopolysaccharide (LPS) and Pam3Cys (P3C). The fold-change in cytokine production compared to pre-BCG1 was used to quantify the trained immunity response. At pre-BCG7 and pre-BCG10 we observed increased production of TNF and IL-1β compared to pre-BCG1. Importantly, at these time points the patients had not received BCG for 6-12 weeks, hence indicating long term induction of trained immunity.

Furthermore, using RNA-seq and ChIP-seq data from a subset of Tribute patients and in vitro trained immunity models we investigated the phenotype of BCG-induced trained immunity during intravesical BCG treatment. Both in vivo and in vitro we observed increased gene expression of Guanylate Binding Proteins (GBP1/2/4/5) and AIM2, which indicated a primed AIM2-GBP inflammasome and an increased capacity to produce IL-1β. Using a motif enrichment analysis, we observed increased activity of interferon regulatory factors (IRF) transcription factors after induction BCG treatment. This corresponded with increased interferon gamma (IFN-γ) protein levels in plasma at BCG6 and pre-BCG7. These results indicated that changes in immune gene expression induced by BCG instillations are regulated via interferon signalling pathways.

Figure 1: The study schedule of the Tribute study. The arrows indicate the time points at which a blood sample was obtained. The blue arrow indicates the baseline time point (pre-BCG1). The purple arrows indicate the important time points for assessment of long-term trained immunity induction.

In 657 NMIBC patients from the prospective UroLife and BlaZIB cohorts we collected data on incidence of respiratory infections and analysed the association with BCG treatment. We found that BCG-exposed NMIBC patients had a decreased frequency of respiratory infections compared to non-BCG-treated NMIBC patients (odds ratio 0.63 (95% CI 0.40 to 1.01)), supporting the hypothesis of induction of trained immunity by BCG instillations. Note that this risk estimate is in line with the reduced risk for respiratory infections (hazard ratio 0.56 (95% CI 0.39-0.82)) that was recently described in a meta-analysis of BCG vaccination trials².

Obstacles in inclusion of patients for the TRIBUTE study (including BCG shortage and the COVID pandemic) prohibited the study of the relation between trained immunity and treatment outcome (i.e., recurrence and progression) in this cohort. We could however explore the relation between BCG-induced trained immunity and oncological NMIBC outcome using exome chip data of 215 BCG-treated NMIBC patients from the Nijmegen Bladder Cancer Study. We found indications that variants in genes that are known to affect BCG-induced trained immunity associate with recurrence-free survival and progression-free survival after BCG treatment. These genetic findings suggest a connection between trained immunity and clinical efficacy of BCG in NMIBC.

So, in all, we presented comprehensive evidence from in-vivo and ex-vivo studies that BCG instillations in patients with NMIBC induce trained immunity. We showed that this trained immunity phenotype results in an augmented long-term systemic innate immune response, which may boost antitumor and antipathogen immune responses, and the potential mediating role of the interferon signalling pathways and AIM2-GBP inflammasome. If the trained immunity response is indeed related to BCG efficacy, as suggested by our genetic findings, this offers avenues for improvement of the efficacy of intravesical BCG treatment and the treatment of NMIBC patients in general.

That this can indeed be potentially relevant has recently been illustrated. Sing et al. generated a re-engineered BCG overexpressing cyclic di-AMP that offered stimulator of interferon genes (STING)-dependent enhanced trained immunity in human and murine primary cells. This re-engineered BCG showed improved efficacy against bladder cancer in murine models compared to wildtype BCG⁵. The promising results of the QUILT 3.032 study were published recently and showed that N-803+BCG had good clinical efficacy in 160 BCG-unresponsive NMIBC patients (NCT03022825) ^6,7. N-803 is an IL-15 super agonist that promotes the proliferation and activation of NK cells and CD8+ T cells. The researchers hypothesize that increased efficacy is due to the synergistic effect of N-803 and BCG which could lead to increased trained immunity responses compared to BCG alone.

1. Department of Internal Medicine, Radboudumc, Nijmegen, The Netherlands
2. Department for Health Evidence, Radboudumc, Nijmegen, The Netherlands 

1. Netea, M. G. et al. Trained immunity: A program of innate immune memory in health and disease. Science (New York, N.Y.) 352, aaf1098, doi:10.1126/science.aaf1098 (2016).
2. Trunk, G., Davidović, M. & Bohlius, J. Non-Specific Effects of Bacillus Calmette-Guérin: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Vaccines (Basel) 11, doi:10.3390/vaccines11010121 (2023).
3. Arts, R. J. W. et al. BCG Vaccination Protects against Experimental Viral Infection in Humans through the Induction of Cytokines Associated with Trained Immunity. 23, 89-100 e105, doi:10.1016/j.chom.2017.12.010 (2018).
4. Van Puffelen, J. H. et al. Intravesical BCG in patients with non-muscle invasive bladder cancer induces trained immunity and decreases respiratory infections. Journal for ImmunoTherapy of Cancer 11, e005518, doi:10.1136/jitc-2022-005518 (2023).
5. Singh, A. K. et al. Re-engineered BCG overexpressing cyclic di-AMP augments trained immunity and exhibits improved efficacy against bladder cancer. Nature Communications 13, 878, doi:10.1038/s41467-022-28509-z (2022).
6. Chamie, K. et al. IL-15 Superagonist NAI in BCG-Unresponsive Non–Muscle-Invasive Bladder Cancer. NEJM Evidence 2, EVIDoa2200167, doi:10.1056/EVIDoa2200167 (2022).
7. Chamie, K. et al. Final clinical results of pivotal trial of IL-15RαFc superagonist N-803 with BCG in BCG-unresponsive CIS and papillary nonmuscle-invasive bladder cancer (NMIBC). Journal of Clinical Oncology 40, 4508-4508, doi:10.1200/JCO.2022.40.16_suppl.4508 (2022).