Background Children with high-risk neuroblastoma have poor survival rates and urgently need more effective treatments with less side effects. Novel and improved immunotherapies may fill this need. However, despite their success in various adult cancers, CAR-T cells and immune checkpoint blockade show limited clinical efficacy in neuroblastoma. We aimed to provide a comprehensive overview of neuroblastoma’s immune environment and relevant immunoregulatory interactions, to identify strategies for improving immunotherapy efficacy.

Materials and Methods 25 tumor samples from 20 patients (17 with high-risk disease, 6 with MYCN amplification), were collected pre-treatment (n=10) or during resection surgery after induction chemotherapy (n=15). Samples were enzymatically digested, single-cell FACS sorted and sequenced by Cel-Seq2 protocol. In vitro killing assays were performed with luciferase-transduced patient-derived neuroblastoma organoids and adult healthy donor PBMCs. Checkpoint inhibition was tested in vivo in three syngeneic neuroblastomamodels (Neuro2a, N1E-115, N18) and one chemotherapy-resistant syngeneic model (Th-ALKF1174L/MYCN 129/SvJ).

Results Neuroblastomas were infiltrated by various immune cells, including dendritic cells, monocytes and four populations of macrophages. The latter showed an M2-like differentiation, associated with immunosuppressive and pro-tumorigenic features. Lymphoid cells consisted of NK, B, and different populations of T cells including highly suppressive Tregs. Of the two identified CD4⁺ non-Treg clusters, one cluster likely contained tumor-reactive cells and was significantly enriched for genes associated with T cell dysfunction in tumors, such as TIGIT and CTLA4. CD8⁺ T had significantly increased LAG3 and PDCD1 (PD-1) expression, also associated with T cell dysfunction. Overall, T cells showed increased signs of dysfunction/exhaustion particularly post-chemotherapy, with enhanced expression of immune checkpoint receptors. NK cells had impaired cytotoxic function (GZMB, PRF1, GNLY), particularly in pre-treatment tumors, which correlated with TGF-β1 signaling and a disbalance between inhibitory receptor genes TIGIT and CD96 and activating receptor CD226. To identify functionally relevant targets for reinvigorating T/NK cell function, we constructed an unsupervised interaction network. This analysis predicted an abundance of immunoregulatory interactions in the tumor microenvironment affecting T/NK cell function, which included, amongst others, CLEC2D-KLRB1, PD1-PDL1 and NECTIN2-TIGIT. Since also in T cells the TIGIT/CD226 balance proved disturbed, we tested combined TIGIT/PD-L1 blockade in vitro, whichsignificantly increased killing of patient-derived neuroblastoma organoids. Moreover, TIGIT/PD-L1 blockade in vivo in three syngeneic models induced complete remissions in a subset of animals and significantly improved survival. Lastly, addition of TIGIT blockade to the standard backbone treatment for relapse/refractory neuroblastoma patients significantly improved survival in a chemotherapy-resistant model mimicking relapse/refractory tumors.

Conclusions We provided a comprehensive atlas of neuroblastoma’s immune environment and identified TIGIT as a promising target for (combination) immunotherapy in neuroblastoma.

Disclosure Information J. Wienke: None. L.L. Visser: None. W.M. Kholosy: None. K.M. Keller: None. M. Barisa: None. S. Munnings-Tomes: None. E. Carlton: None. E. Poon: None. A. Rodriguez: A. Employment (full or part-time); Significant; Hoffman-La Roche. R. Bernardi: A. Employment (full or part-time); Significant; Hoffman-La Roche. F. van den Ham: None. S.R. van Hooff: None. K.P.S. Langenberg: None. F.C.P. Holstege: None. L. Chesler: B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Modest; Hoffman-La Roche. J. Anderson: B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Modest; Hoffman-La Roche. E. Ownership Interest (stock, stock options, patent or other intellectual property); Significant; Autolus Ltd. H.N. Caron: A. Employment (full or part-time); Significant; Hoffman-La Roche. T. Margaritis: None. M.M. van Noesel: None. J.J. Molenaar: B. Research Grant (principal investigator, collaborator or consultant and pending grants as well as grants already received); Significant; Hoffman-La Roche.