Scientific Programme

he quality and quantity of tumor-infiltrating lymphocytes, particularly CD8⁺ T cells, are important parameters for control of tumor growth and response to immunotherapy. Here, we show in murine and human cancers that these parameters exhibit circadian oscillations, driven by both the endogenous circadian clock of leukocytes and rhythmic leukocyte infiltration, which depends on the circadian clock of endothelial cells in the tumor microenvironment. To harness these rhythms therapeutically, we demonstrate that efficacy of chimeric antigen receptor T cell therapy and immune checkpoint blockade can be improved by adjusting the time of treatment during the day. Furthermore, time-of-day dependent T cell signatures in murine tumor models predict overall survival in patients with melanoma and correlate with response to anti-PD1 therapy. Our data demonstrate the functional significance of circadian dynamics in the tumor microenvironment and suggest the importance of leveraging these features for improving future clinical trial design and patient care.

Introduction: TCR repertoire against tumors often lack high affinity TCRs, and the available T cells with low to moderate affinity frequently fall short in effectively eliminating cancer cells. In addition, immune response is hindered by regulatory T cells (Tregs).

Methods: We developed low and high affinity vaccines using the full agonist peptide p33 and its weak agonist peptide A4Y (Y-to-A substitution at position 4). In an additional low affinity model, we targeted B16F10 melanoma with our previously described multi-target germline personalized vaccine.

Results: Our results showed limited in vivo lytic cross-reactivity against the high-affinity p33 and the low-affinity A4Y peptides. Immunization with the low-affinity vaccine failed to hinder the progression of B16F10p33 tumors. Interestingly, when A4Y vaccination was combined with Treg depletion, the low-affinity T cell induced potent local and systemic immune-response and strong or complete blockade of tumor-growth. Furthermore, we found accumulation of protective tissue-resident memory T cells. The distance between infiltrating individual T cells in the tumor were significantly reduced, indicative of increased cell clustering, and T cell migration from blood vessels was increased following Treg depletion. Importantly, we could also show enhanced anti-tumor response when combining a low-affinity germline tumor-associated antigens vaccine with Treg depletion.
 

Conclusion: The data presented here not only revealed a novel function of Tregs in regulating the activity of low-affinity CD8+ T cells, but also promotes the development of innovative strategies of immunotherapy.

The intestinal epithelium plays a central role in human health and disease, and several chronic inflammatory disorders associate with a weakened epithelial barrier. The organoid model has been instrumental in studying primary epithelial cell behavior in homeostasis and disease. Recent advances in human organoid transplantation into mouse and human lay the base for studies of human epithelial cell behavior within tissue context, and for novel therapeutic approaches for diseases such as inflammatory bowel disease. It remained unclear how organoid transplantation into the colon affects epithelial phenotypes and protein expression, which is key to assess the suitability of this model. To address this, we employed Deep Visual Proteomics including AI-guided cell classification on images, microdissection, and high-sensitivity proteomics, on human colonic epithelial stem and differentiated cells in vivo, upon transplantation in situ, and organoids cultured in vitro. We find that transplanted organoids closely resemble human intestinal epithelial cells in vivo compared to organoids grown in vitro, indicating that organoid culture induces a transient shift in epithelial phenotypes, which is reversible upon reintroduction into the mucosa. Phenotypic differences between epithelial cells in vitro and in situ/in vivo were driven by hallmarks of high proliferation and lower functional differentiation in organoids. With this, we demonstrate that transplanted epithelial cells in situ represent a physiological, relevant model for studying functional aspects of mature colonocytes.

Introduction: Atopic dermatitis (AD) is a chronic inflammatory skin disease with a heterogeneous clinical phenotype, affecting up to 20% of children and 3% of adults worldwide. AD is characterized by barrier dysfunction, persistent Th2 inflammation, and skin dysbiosis due to S aureus overgrowth. Despite being colonize by S aureus, skin lesions in AD patients exhibit a conspicuous absence of neutrophils, potentially attributed to the report IL-4-impairment of neutrophil functions (migration, netosis, phagocytosis). However, there is not data on the impact of co-stimulation by opposing signals on neutrophils function.

Aim: To assess functional aspects of human neutrophils adaptation after in-vitro co-stimulation with IL-4 or IL-13 and IFN-g.

Methods: We measured STAT1 and STAT6 phosphorylation by FACS, netosis by Sytox green incorporation and ROS production by ampiflu-red detection.

Results: Combining IFN-g with IL-4 or IL-13 led to simultaneous STAT-6 and STAT-1 phosphorylation. Notably, IFN-g addition to IL-4 attenuated STAT-6 phosphorylation, while its addition to IL-13 significantly increased STAT1 phosphorylation, without affecting STAT6 phosphorylation. Netosis was reduced upon IL4-Ra signalling but was restore by the addition of IFN-g. However, IL4-Ra stimulation did not affect ROS production and the addition of IFN-g even increased it.

Conclusions: These data so far suggest a high degree of neutrophils plasticity and an immune regulatory effect of opposing signals in-vitro, which would be further evaluated in AD patients including RNA-seq and ATAC-seq profiling.

Background: Macrophage activation syndrome (MAS) is a life-threatening inflammatory condition. Elevated circulating levels of IL-18 are assumed to be pathogenic. IL-18 is initially produced as a pro-peptide, cleaved by caspase-1 (Casp1) upon inflammasome activation and naturally inhibited by IL-18 binding protein (IL-18BP) upon formation of high affinity complexes. We previously showed that severe MAS manifestations were dependent on IL-18 in IL-18BP knockout (KO), but not in WT, mice injected with CpG, a TLR9 agonist.

Aim: We aim to examine the role of inflammasome activation in CpG-induced MAS.

Methods: Il18bp^-/-, Casp1^-/-, Gsmd^-/-, Il18bp^-/-Casp1^-/-, Il18bp^-/-Gsmd^-/-, and WT littermates were injected i.p with CpG at days 0, 2, and 4 and followed until day 7 with daily measurement of body weight. Liver and spleen weight were assessed at day 7. Cytokines levels in organs and plasma were measured at mRNA and protein levels by RT-qPCR and ELISA.

Results: While the phenotype of Casp1^-/- and Gsmd^-/- did not differ from their WT littermates, we observed that the severity of MAS was markedly attenuated in Il18bp^-/-Casp1^-/- and Il18bp^-/-Gsmd^-/- compared to Il18bp^-/- littermates, including decreased body weight loss and splenomegaly. Free IL-18 levels at day 7 were decreased in dKO mice compared to Il18bp^-/- but were still detectable, suggesting the presence of IL-18 processing.

Conclusion: The severity of CpG-induced MAS in Il18bp^-/- mice is dependent on the maturation of IL-18 by Casp1 and released by Gasdermin-D. However, limited IL-18 cleavage occurs in the absence of Casp1.

Lupus panniculitis (LP) is a subtype of cutaneous lupus erythematosus that affects subcutaneous adipose tissue. The pathomechanisms underlying LP are largely unknown. In this study, we aimed to obtain deeper insights into LP through the characterization of cellular and molecular immune patterns. Using imaging mass cytometry (IMC), we analyzed the cellular infiltrates in deep skin biopsies from LP patients (n=8) and healthy controls (n=6). Concurrently, we performed nCounter nanostring technology to quantify the mRNA expression of immune-related markers. The IMC analysis revealed that T cells (CD3⁺) predominated in LP More precisely, they exhibited a predominantly skin-homing (CLA⁺), cytotoxic phenotype (CD8⁺, granzyme B⁺), indicating site-specific immune activation within skin tissue. B cells (CD20⁺) were also present. They constituted a significant portion of immune cells, alongside with the presence of dendritic cells and macrophages. Differential gene expression analysis revealed an upregulation in pathways associated with adaptive and innate immune responses. This included T cell receptor signaling (upregulation of CD247, LCK), lymphocyte activation (ZAP70, CD3D), cytokine signaling (STAT1, MYD88), and MHC antigen presentation (HLA-DRA, TAP1/2). Additionally, there was an upregulation of genes associated with the Complement system (C1QA/B, C4A/B). Our results suggest that a cytotoxic T- and B-cell-predominated immune response and complement activation is involved in LP. This points towards a pathogenesis that is different from other types of panniculitis.

In the phase 3 trial primary analysis, mRNA‑1345 was efficacious against RSV-associated lower respiratory tract disease (RSV‑LRTD) in adults ≥60 years, with no safety concerns. We report additional analyses when >90% of participants completed ≥6-month follow-up. This ongoing phase 3, multi-country, double-blind, placebo-controlled study (NCT05127434) randomized (1:1) adults ≥60 years to receive 1 dose of mRNA‑1345 (50 μg) or placebo. Primary objectives included safety and tolerability and vaccine efficacy (VE) against a first episode of RSV‑LRTD with ≥2 or ≥3 symptoms between 14 days and 12 months postinjection; key secondary efficacy objectives included prevention of RSV-associated acute respiratory disease (RSV-ARD). Efficacy against RSV-LRTD with shortness of breath (SOB) was assessed as a surrogate measure of more severe disease. The analysis included 36,157 participants (mRNA‑1345, n=18,112; placebo, n=18,045). mRNA‑1345 was well-tolerated; no safety concerns were identified. At a median follow-up of 8.6 months, mRNA‑1345 VE was 63.3%, 63.0%, and 53.9% against RSV-LRTD with ≥2 and ≥3 symptoms and RSV-ARD, respectively. Lower bounds of the 95% CI exceeded prespecified success criteria of 20% for all endpoints. VE was evident across RSV-A and RSV-B subtypes and was generally consistent across demographic and risk subgroups. VE was 74.6% (95% CI, 50.7-86.9) against RSV-LRTD with ≥2 symptoms, including SOB. mRNA-1345 was well-tolerated with no safety concerns, and demonstrated efficacy for the prevention of RSV disease through a median of 8.6 months among adults ≥60 years.

Aim: A successful strategy to cure HIV requires eradicating its reservoir, mainly represented by PD-1+ T follicular helper cells. Herein, we hypothesize that CAR-T cells redirected against PD-1 may represent an effective approach to target the HIV reservoir.

Methods: Second-generation CARs were cloned using the scFv of a blocking (bPD1-CAR) and a nonblocking (nbPD1-CAR) anti-PD-1 monoclonal antibody fused to an IgG4 hinge, CD28 transmembrane and 4-1BB signaling domains with a cherry reporter. Anti-CD19 CAR-T cells were used as controls. The killing activity was measured using FACS and luciferase-based assays. HIV resistance to R5 and X4 HIV strains was tested by selectively editing the CD4 receptor and/or CCR5 and CXCR4 co-receptors using CRISPR-Cas9 technologies.

Results: The PD-1 binding affinity and K_D values of the two anti-PD-1 mAbs were similar. Both anti-PD-1, but not CD19-control, CAR-T cells caused delayed depletion of PD-1+ CD4+ cherry negative T cells in ex-vivo cultures. Interestingly, the bPD1-CAR but not the nbPD1-CAR mediated tonic signaling, which was efficiently prevented by editing the endogenous PD-1. Both anti-PD-1 CAR-T cells equivalently killed PD-1^high transgenic cells, yet the nbPD1-CAR was significantly less efficient in killing wild-type PD-1^low target cells. CD4 editing was sufficient to confer HIV-resistance against R5 and X4 HIV strains without impairing the killing activity.

Conclusion: We successfully engineered two anti-PD-1 CARs with different functional activities. In-vivo experiments in humanized HIV-infected mice are undergoing.

Aim: MANDARA was a Phase 3, randomised, double-blind, 52-week study (NCT04157348) of benralizumab (n=70) vs mepolizumab (n=70) in patients with relapsing/refractory EGPA receiving standard of care. Non-inferiority was demonstrated for remission (BVAS=0 and oral glucocorticoid [OGC] ≤4mg/day) at both Weeks 36 and 48.

Methods: Post-hoc analyses of MANDARA assessed those achieving a more stringent definition of complete remission: BVAS=0 and OGC dose=0mg/day at both Weeks 36 and 48 and being relapse-free. Remission=sustained if criteria were met by Week 48 and maintained to end of the 52-week double-blind period. Investigators were encouraged to taper OGCs for patients who reached BVAS=0 as per standard practice and clinical judgement. HR and 95% CIs were estimated using a Cox regression model with Efron method to control for ties.

Results: Adjusted rates of complete remission at both Weeks 36 and 48: benralizumab, 23.5%; mepolizumab, 11.1% (difference: 12.47 [95% CI: 0.46, 24.48]; p=0.0418). Sustained remission was achieved by 65.7% and 64.3% of patients (HR: 1.19 [95% CI: 0.78, 1.81]; p=0.7793) in the benralizumab and mepolizumab groups; sustained complete remission was achieved by 35.7% and 22.9% patients (HR: 1.82 [95% CI: 0.97, 3.50]; p=0.0966), respectively.

Conclusions: Patients with EGPA achieve higher rates of complete remission with benralizumab vs mepolizumab when using a more stringent definition of the endpoint. These data highlight the possibility of achieving sustained treatment goals for patients with EGPA, including full tapering of OGCs and avoiding relapses.

Aim: To test ILIT Allergen intralymphatic immunotherapy (ILIT) with a chemically modified grass allergen administered with a biodegradable depot adjuvant MicroCrystalline Tyrosine.

Methods: Sixty ARC patients were randomized to receive ILIT with placebo or grass allergen (Polvac® Rye+Grass). Three ultrasound-guided injections were given at four-week intervals into inguinal lymph nodes. The dose was 1/15th of a maintenance dose for subcutaneous immunotherapy (SCIT). The primary outcome was a daily combined symptom medication score (cSMS) during the two following grass pollen seasons. Secondary outcomes comprised quality of life questionnaires, serology, lung-function tests, SPT and adverse events (AEs).

Results: Allergen ILIT was safe, with mild AEs in 48 out of 90 injections (53.8%) and three (3.2%) moderate AEs within the 40 minutes post-treatment observation. AEs were almost all erythema, wheal, or swelling at the injection site. All reactions retracted within 1-3 days. No SAEs. A significant reduction (>20%) of cSMS in the high grass pollen season was determined in the treatment group compared to placebo. The treatment effect increased with the measured seasonal pollen load. Allergen ILIT was associated with an increase in IgG4 towards timothy grass allergens and improved lung function (FEV1).

Conclusion: ILIT with a tyrosine-based SCIT product was well tolerated. The cSMS scoring two years after ILIT showed significant symptom amelioration in the allergen-treated. ILIT allows an ultrashort, effective pre-seasonal treatment, offering unique patient convenience and motivation in the treatment regimen.

Aim: OGCs have been the cornerstone of EGPA treatment but are associated with adverse events and frequent relapses during tapering. Benralizumab may represent a treatment option for patients that facilitates OGC-sparing while reducing disease burden.

Methods: MANDARA was a Phase 3, randomised, double-blind, 52-week study (NCT04157348) evaluating efficacy and safety of benralizumab vs mepolizumab in adults with relapsing/refractory EGPA. Investigators were encouraged to taper OGCs for patients with BVAS=0 as per standard practice and clinical judgement.

Results: Patients were randomised to benralizumab (n=70) or mepolizumab (n=70). A subset of patients achieved complete withdrawal of OGC during Weeks 48 through 52 (benralizumab, 41.4%; mepolizumab, 25.8% [difference: 15.69, 95% CI: 0.67, 30.71; p=0.0406]). More benralizumab-treated patients than mepolizumab-treated patients had a sustained 100% reduction in OGC use (100% reduction by Week 40, maintained through to Week 52: 24.3% vs 10.0%, respectively; HR: 2.97 [95% CI: 1.26, 7.77]). Similar proportions in both treatment groups achieved sustained ≥50% reduction in OGC use (≥50% reduction by Week 40, maintained through to Week 52: 77.1% vs 70.0% in benralizumab vs mepolizumab groups, respectively; HR: 1.17 [95% CI: 0.79, 1.74]; p=0.3852).

Conclusions: In the MANDARA study of patients with EGPA, treatment with either benralizumab or mepolizumab was associated with the ability to reduce OGC use. However, benralizumab-treated patients were more likely to fully eliminate use of OGC.

SLE is an inflammatory disorder characterized by the emergence of autoreactive cells. SLE patients exhibit reduced numbers of Natural Killer (NK) cells, decreased cytotoxicity, and impaired cytokine production. The molecular mechanisms underlying NK cell dysfunction in SLE remain elusive.

We compared NK cells from the peripheral blood of SLE patients and healthy controls. We found that SLE NK cells have an increased mitochondrial mass but decreased mitochondrial function, as evidenced by elevated superoxide levels and significant cristae disorganization. These alterations correlate with impaired NK cell functions.

The alterations observed are linked to an accumulation of mitochondrial DNA (mtDNA) in the cytosol of SLE NK cells, associated with a reduction in key mitochondrial clearance proteins (V-ATPase) and a higher lysosomal pH, in SLE NK cells. Inhibition of V-ATPase in healthy NK cells led to an increased mitochondrial mass by inhibiting lysosomal acidification, mirroring the phenotype of SLE NK cells. This suggests a potential link between impaired lysosomal acidification and cytosolic mtDNA accumulation. Treatment of SLE NK cells in vitro with Hydroxychloroquine (HCQ) and Urolithin A (UA) reduced mitochondrial mass and normalized lysosomal acidification, indicating a pathway for therapeutic intervention.

Our findings reveal a critical link between mitochondrial dysfunction and immunometabolic abnormalities in SLE NK cells, highlighting the potential of targeting mitochondrial and lysosomal pathways with HCQ and UA and representing a promising approach for treating SLE.