Introduction

Checkpoint agonism represents a promising class of therapies for the treatment of autoimmune and inflammatory diseases, including ulcerative colitis (UC), where unmet needs persist. Engagement of checkpoint receptor PD-1 by its natural ligand PD-L1 is critical for regulation of T cells. Rosnilimab is an investigational PD-1 agonist IgG1 antibody designed to optimize inhibitory signaling through PD-1, does not block PD-1 binding to PD-L1, and depletes PD-1^high pathogenic T cells while agonizing remaining PD-1+ T cells. The percentages of PD-1+ T cells in the peripheral blood of moderate and severe UC patients are higher than in mild UC patients (1-4). Reduction of elevated PD-1^high Tph cells in both UC colon and peripheral blood correlates with remission (2). Therapeutic reduction of PD-1^high Tph cells and inflammatory cytokines from PD-1+ T cells, by depletion and agonism signaling together, has the potential to reduce overactive immune responses and induce remission.

Aims & Methods

The potency of the agonism and depletion mechanisms of rosnilimab was evaluated in vitro using peripheral blood mononuclear cells (PBMCs) from UC patients. Rosnilimab is a human IgG1 antibody with Fc effector function. A version of rosnilimab that lacks Fc effector function was generated by mutating the IgG1 Fc domain at L234A and L235A (LALA). PBMCs from UC patients were activated with anti-CD3 and anti-CD28, cultured for several days with rosnilimab IgG1, rosnilimab IgG1-LALA, or isotype control and analyzed for the number of PD-1+ T cells and levels of inflammatory cytokines. In addition, the therapeutic potential of depletion and agonism of pathogenic PD-1+ T cells was evaluated in a mouse model of UC using a surrogate of rosnilimab. The surrogate consists of the fab domains of rosnilimab formatted with a mouse IgG2a Fc domain which enables immune cell depletion in mice. On Day 0, hPD1 CD4+CD45Rb+ T cells donor cells were adoptively transferred into Rag2-/- recipients to elicit UC-like disease. Starting on Day 21, 10 mpk of isotype control mIgG2a, rosnilimab-mIgG2a, or control anti-mIL-12 p40 were dosed intraperitoneally twice weekly for 4 weeks. This model system allowed assessment of the therapeutic potential of depletion and agonism mechanisms of rosnilimab on hPD1+ T cells, with a depletion-enabled mouse IgG2a isotype, in the context of T cell-driven gut inflammation and pathology.

Results

In PBMCs from UC patients, at therapeutically relevant concentrations, rosnilimab IgG1 reduced PD-1^high T cells by 90%, while rosnilimab IgG1-LALA reduced PD-1^high T cells by 14%. Isotype control did not reduce T cells. Furthermore, rosnilimab IgG1 reduced secretion of inflammatory cytokines interferon gamma (IFN-γ) and CXCL13 compared to rosnilimab IgG1-LALA or isotype control. In the hPD-1 mouse model of UC, rosnilimab mIgG2a reduced body weight loss, colon pathology, IFN-γ and CXCL13 as assessed by spatial transcriptomics of the colon, and infiltration of hPD-1 CD4+ T cells into the colon as assessed by immunofluorescence, compared to isotype control. In addition, rosnilimab mIgG2a demonstrated efficacy comparable to anti-IL-12 p40 antibody treatment.

Conclusion

Rosnilimab, a PD-1 checkpoint agonist antibody, functions to reduce pathogenic PD-1+ T cells by both depletion and agonism. These in vivo and in vitro results are consistent with the mechanisms of depletion and agonism and support the strong rationale for evaluating the therapeutic potential of rosnilimab in an ongoing phase 2 study in UC (NCT06127043).

References

1. Shi W, Zhang Y, Hao C, Guo X, Yang Q, Du J, Hou Y, Cao G, Li J, Wang H, Fang W. PeerJ. 2023; May 31;11:e15481.
2. Long Y, Xia C, Sun Y, Ma Y, Xu L, Song Y, Liu C. Immunol Lett. 2021; 233:2-10.
3. Roosenboom B, Horjus Talabur Horje CS, Smids C, Leeuwis JW, van Koolwijk E, Groenen MJM, Wahab PJ, van Lochem EG. Scand J Gastroenterol. 2021; 56(6):671-679.
4. Nguyen J, Finkelman BS, Escobar D, Xue Y, Wolniak K, Pezhouh M. Hum Pathol. 2022; 126:19-27.

Disclosure

All authors are employees and shareholders of AnaptysBio, Inc.

Introduction

Decreased vagus nerve activity and increased inflammatory immune cells are observed in ulcerative colitis (UC) (1, 2). Due to its anti-inflammatory and protective properties, vagus nerve stimulation (VNS) has garnered attention for UC therapy (3). Non-invasive transcutaneous auricular vagus nerve stimulation (NitaVNS), devoid of the side effects associated with invasive techniques, demonstrates protective anti-inflammatory effects in our mouse UC-like model of colitis (4), although the specific cellular mechanisms remain unclear.

Aims & Methods

To evaluate the effects of NitaVNS on the distribution of granulocytes, macrophages, natural killer (NK) cells, and CD4⁺ and CD8⁺ T cells in the colon tissue, spleen, and mesenteric lymph nodes of a UC-like colitis mouse model, NitaVNS was administered to C57BL/6 male mice daily for 10 min, beginning one day before colitis induction with 5% dextran sulfate sodium for 5 days in water (the control group received regular water). This continued until the end of the study, with no stimulation in sham mice. The daily disease activity index (DAI) was evaluated along with distal macroscopic scores at the time of sacrifice. The frequencies of granulocytes (Ly6G), macrophages (F4/80), NK cells (NK1.1), CD4⁺ T cells (CD3, CD4), and CD8⁺ T cells (CD3, CD8) were analyzed in the cecum, proximal, mid and distal colon, spleen, and mesenteric lymph nodes (MLNs) using flow cytometry.

Results

NitaVNS in colitic mice significantly improved the DAI and macroscopic scores related to rectal bleeding, colon bleeding, fecal consistency loss, and shortened colon length compared to sham-stimulated colitic mice. In contrast, no changes were observed in the non-colitic mice. While NitaVNS did not alter the frequency of Ly6G⁺ cells in the cecum, mid/distal colon, and MLNs, it significantly decreased the positivity of these cells in the proximal colon and spleen of mice with colitis without affecting non-colitic mice. NitaVNS significantly decreased the frequency of F4/80⁺ cells in the mid-colon but did not influence these cells in the cecum, proximal and distal colon, spleen, or MLNs. In colitic conditions, NitaVNS significantly increased NK cell frequency in the spleen without affecting non-colitic mice. Moreover, NitaVNS did not alter the frequency of NK cells in the colon or MLNs in colitic or non-colitic mice. Additionally, NitaVNS did not induce changes in CD4⁺ and CD8⁺ T cell distribution in the colon tissue, spleen, or mesenteric lymph nodes of colitic and non-colitic mice.

Conclusion

These results suggest that NitaVNS could potentially reduce inflammation in colitic mice, locally and systemically, by decreasing the frequency of granulocytes in the colon and spleen and colonic macrophages. Additionally, NitaVNS increased the number of splenic NK cells, potentially contributing to its systemic effects. These findings indicate promising therapeutic potential for NitaVNS.

References

1. Ghia JE, Blennerhassett P, Kumar-Ondiveeran H, Verdu EF, Collins SM. The vagus nerve: a tonic inhibitory influence associated with inflammatory bowel disease in a murine model. Gastroenterology. 2006;131(4):1122-30.
2. Wiese JJ, Manna S, Kühl AA, Fascì A, Elezkurtaj S, Sonnenberg E, Bubeck M, Atreya R, Becker C, Weixler B, Siegmund B, Patankar JV, Prüß MS, Schumann M. Myenteric Plexus Immune Cell Infiltrations and Neurotransmitter Expression in Crohn's Disease and Ulcerative Colitis. J Crohns Colitis. 2024 Jan 27;18(1):121-133.
3. Verma N, Mudge JD, Kasole M, Chen RC, Blanz SL, Trevathan JK, Lovett EG, Williams JC, Ludwig KA. Auricular Vagus Neuromodulation-A Systematic Review on Quality of Evidence and Clinical Effects. Front Neurosci. 2021 Apr 30;15:664740.
4. Hesampour F, Tshikudi D, Veysel Özden A, Bernstein CN, Ghia JE. A54 The impact of non-invasive transcutaneous auricular nerve stimulation (NitaVNS) on the development of experimental colitis in mice. J Can Assoc Gastroenterol. 2023 Mar 7;6(Suppl 1):30.

Disclosure

F. Hesampour: None Declared, D. Tshikudi: None Declared, A. Veysel Özden: The employee of the company providing the stimulator, C. N. Bernstein: None Declared, J-E. Ghia: None Declared.