Siplizumab combination therapy with belatacept or abatacept broadly inhibits human T cell alloreactivity in vitro

Combined antigen-specific T cell


Introduction
Upon antigen recognition, T cell responses are regulated by signaling through complementary receptors expressed on T cells, a process known as costimulation.The development of selective costimulation-blocking biologics could improve longterm outcomes in transplant recipients that currently rely on broadly immunosuppressive small molecule maintenance therapy.Although potent in the prevention of graft rejection, calcineurin inhibitors (CNIs), such as cyclosporine and tacrolimus, are associated with a negative impact on patient morbidity and mortality when administered chronically. 1 Costimulation blockers abatacept and belatacept are fusion proteins of the extracellular domain of cytotoxic T-lymphocyte-associated antigen 4 and a human IgG1Fc-fragment (CTLA4-Ig). 2 Both bind B7 (CD80/CD86) on antigen-presenting cells and thus prevent costimulation through the B7/CD28 pathway.In a large phase III de novo renal transplant trial, belatacept improved patient/graft survival, kidney function, and decreased the incidence of donor-specific antibodies when compared with a cyclosporine-based regimen. 3However, the benefits of belatacept use are counterbalanced by the increased risk and severity of acute rejection compared with CNIs. 3,4Thus, despite the approval of belatacept in 2011 for use in kidney transplant recipients, most patients today receive traditional tacrolimus-based maintenance immunosuppression regimens. 5lthough costimulatory blockade of a single pathway strongly reduces T cell activation, heterogeneous expression of costimulatory receptors on different subpopulations leaves some T cells unaffected.This may contribute to the relatively high rates of acute cellular rejection in clinical studies investigating belatacept-based immunosuppression.However, preclinical in vitro and in vivo evidence suggests that a combination of costimulation blockers, ie, dual costimulatory blockade, may be an effective modality to achieve more complete inhibition of T cell activation. 6,7Namely, blocking independent costimulatory pathways that tend to display inverse expression patterns may adequately target the alloreactive T cell subpopulations.One example is the combined blockade of CD28/B7 costimulation and CD2/CD58 costimulation.CD2 plays an important role in the formation and organization of the immunological synapse between T cells and antigen-presenting cells. 8Siplizumab is an investigational humanized antiCD2 IgG1 monoclonal antibody that is currently undergoing multiple phase II clinical trials.0][11][12][13][14][15][16][17] Herein, we investigated the effect of combining siplizumab with abatacept or belatacept on human T alloreactivity in mixed lymphocyte reactions (MLRs).Our results show that low doses of siplizumab suppress the proliferation of abatacept/belatacept-resistant T cells and enhance the selective depletion of memory T cells.Therefore, we propose a novel approach to combine blockade of CD28/B7 and CD2/CD58 costimulation as a potential strategy to replace standard CNI-based maintenance immunosuppression and improve long-term outcomes following transplantation.

Allogeneic mixed lymphocyte reaction
Peripheral blood mononuclear cells (PBMC) were isolated via Ficoll Paque Plus (Cytiva) density gradient centrifugation from buffy coats.Buffy coats were obtained from anonymous healthy adult donors from the Karolinska University Hospital blood bank, and PBMC isolation was performed within 24 hours of blood collection.In line with Swedish legislation section code 4x 3p SFS 2003:460 (Lag om etikpr€ ovning av forskning som avser m€ anniskor), an ethics approval was not needed since the blood bank provides buffy coats for research purposes only from anonymous donors that have given prior written informed consent, and biological material cannot be traced back to a specific individual.Human PBMC from 15 individual donors in total were isolated in 5 batches of 3 donors and mixed in pairs (donor 1 þ 2, donor 1 þ 3, and donor 2 þ 3) resulting in 15 different MLR combinations.Specifically, PBMC from 2 donors were mixed in PBS at a concentration of 1.5 to 2.0 Â 10 7 cells per mL and stained with violet proliferation dye 450 (VPD450; BD Biosciences) following the manufacturer's instructions.This study used a two-way allogeneic MLR, ie, PBMC from neither donor were inactivated via irradiation/ chemical treatment.Thus, PBMC from both donors functioned as responders and stimulators.As this study investigated the effect of antibodies/fusion proteins on T cell responses in general and not the responsiveness of a specific donor, this setup was deemed satisfactory as a measure of alloreactivity.
VPD450-stained PBMC were washed and resuspended in 10% heat-inactivated FBS (Gibco) in AIM V medium (Gibco, Thermo Fisher Scientific Inc).Resuspended PBMC were dispensed into round-bottom 96-well cell culture plates, and pure medium or medium supplemented with siplizumab and/or abatacept or belatacept was added to a final concentration of 2 Â 10 6 cells per mL in a final volume of 200 μL.MLRs were incubated at 37 o C, 5% CO 2 for 7 and 10 days, respectively.On days 5 or 6, 100 μL fresh culture medium (no additional antibody; final volume ¼ 300 μL) was added to each well.On day 8, 100 μL medium was aspirated from each well without disturbing the cell pellet and replaced with 100 μL fresh culture medium.

Flow cytometry
After 7 days of MLR, samples were blocked with an Fcreceptor binding inhibitor (Invitrogen; Thermo Fisher Scientific Inc) and then stained with antibodies against cell surface antigens.Full list of antibodies can be found in the respective Supplementary Tables.T cells were gated as CD3 þ CD56 -.T cell subpopulations were defined as follows: naïve T cells: CCR7 þ CD45RA þ ; central memory T cells: CCR7 þ CD45RA -; effector memory T cells: CCR7 -CD45RA -; terminally differentiated effector memory T cells: CCR7 -CD45RA þ .After 10 days of MLRs, intracellular FoxP3 staining was performed using the eBioscience FoxP3/transcription factor staining buffer set (Invitrogen; Thermo Fisher Scientific Inc) according to the manufacturer's instructions.T cells were gated as CD3 þ .Regulatory T cells (Tregs) were identified as Cell proliferation was assessed using VPD450 (VPD450high: nonproliferated; VPD450low: proliferated).Samples were stained in the dark at 4 o C and washed twice in saline solution, followed by analysis using a BD Celesta flow cytometer (BD Biosciences).

Graphs and statistical analysis
Visualization of results and statistical analysis of underlying data were carried out using GraphPad Prism 8 software (GraphPad Software).Data displayed in graphs can be found in the respective Supplementary Tables.Data were analyzed using one-way analysis of variance followed by Dunnett's multiple comparison test, with no antibody controls serving as the comparison data set.
Notably, monotherapy with each agent induced incomplete inhibition of CD4 T cell proliferation.In contrast, significant and near-complete inhibition of CD4 T cell proliferation was mediated by the combination of different doses of siplizumab with abatacept (100 μg/mL, P .0183)or belatacept (100 μg/mL, P .0061)(Fig. 1).Interestingly, combination therapy with abatacept or belatacept decreased the half maximum inhibitory concentration (IC50) for the suppressive effect of siplizumab on day 7 of MLRs from 0.0068 μg/mL (monotherapy) to 0.0015 μg/mL and 0.0011 μg/mL, respectively.Following 10 days of MLRs, the IC50 of siplizumab monotherapy was 0.0235 μg/mL and 0.0035 μg/mL or 0.0025 μg/mL for combination with 100 μg/mL Abatacept or 100 μg/mL Belatacept, respectively.As displayed in Figure 1 (right panels), a similar pattern was observed for inhibition of CD8 T cell proliferation.After 7 days of MLR, inhibition of CD8 T cell proliferation by siplizumab monotherapy had an IC50 of 0.0105 μg/mL whereas combination therapy with abatacept or belatacept lowered the IC50 of siplizumab to 0.0012 μg/mL in both combinations.Following 10 days of MLRs, siplizumab monotherapy had an IC50 of 0.024 μg/mL, which was lowered to 0.0017 μg/mL when combined with abatacept or belatacept.

T cell depletion
To assess the effects of combining siplizumab (a T cell depleting agent 9 ) and CTLA4-Ig (a nondepleting agent; Nulojix Public Assessment Report, European Medicines Agency, published July 7, 2011), T cell counts were determined using flow cytometry after 10 days of MLRs (n ¼ 9; Supplementary Tables 9  and 10).As shown in Figure 3 (left panel), high doses (10-100 μg/mL) of either abatacept (P .0278)or belatacept (P .0187)reduced CD4 T cell counts compared with untreated controls.

Naïve T cell enrichment
We have previously reported that siplizumab altered the relative abundance of specific T cell subsets by reducing the percentage of memory T cells in an MLR. 9 Here we assessed the effects of combining siplizumab and CTLA4-Ig on the percentage of naïve (CCR7 þ CD45RA þ ) T cells using flow cytometry after 7 days of MLRs (n ¼ 9; Supplementary Tables 11 and 12).As shown in Figure 4 (left panel), abatacept (0.1-100 μg/mL, P .0069),belatacept (0.1-100 μg/mL, P .0050)and siplizumab (0.012-0.11μg/mL or 1 μg/mL, P .0437)induced a significant enrichment in naïve CD4 T cells relative to untreated controls.
The addition of siplizumab (1 μg/mL) to either abatacept (P .0391)or belatacept (P .0408)replicated the effect of high-dose siplizumab monotherapy and resulted in a 2-fold increase in the percentage of naïve CD4 T cells compared with the untreated control after 7 days of MLR.

Treg enrichment
Additionally, we investigated how combination therapy with siplizumab and abatacept or belatacept affects Treg enrichment in MLRs treated with siplizumab.Following 10 days of MLRs, the percentage of Tregs among proliferated CD4 T cells was   13  and 14).
As previously reported and shown here in Figure 5 (left panel), siplizumab (0.037-1 μg/mL, P .0338) induced significant enrichment of Tregs among proliferated CD4 T cells. 9,10Whereas low doses of abatacept (10 -7 to 1 μg/mL) or belatacept (10 -7 to 10 -2 μg/mL) combined with siplizumab (1 μg/mL) still resulted in significant Treg enrichment, the effect was lost (comparable to the untreated control) once the concentration of abatacept or belatacept increased to the point where maximum inhibition of T cell proliferation was reached (10-100 μg/mL abatacept or 1-100 μg/mL belatacept).In addition, given the strong reduction in  proliferation and total T cell counts presented in Figures 1-3, we investigated Treg cell counts among proliferated CD4 T cells using flow cytometry after 10 days of MLRs (Fig. 5; right panel).Neither CTLA4-Ig nor CD2 monotherapy resulted in significantly different Treg cell counts when compared with the untreated control.Interestingly, siplizumab (1 μg/mL) combinations with high-dose CTLA4-Ig therapy trended toward decreased Treg cell counts, but the reduction was not statistically significant after 10 days of MLRs among the subjects tested (n ¼ 9).

Discussion
This study demonstrates that dual costimulation blockade with siplizumab and CTLA4-Ig (abatacept or belatacept) results in potent synergistic inhibition of both alloreactive CD4 and CD8 T cells.An increase in the potency of MLR inhibition of up to 14-fold was observed with the combination compared with siplizumab alone.Dual CD2/CD28 costimulation blockade additionally induced potent selective depletion of memory T cells.Specifically, the combination of siplizumab with abatacept reduced total CD8 T cell counts 20-fold, although it enriched the naïve CD8 T subset 3.5fold, when compared with the untreated control in the MLRs setting.
The adoption of belatacept as first-line immunosuppression in kidney transplant recipients has been hindered by the increased risk of acute rejection, mostly within the first year posttransplant, when compared with CNI-based therapy. 18Investigations into the mechanisms of belatacept-resistant rejection suggested the presence of alloreactive T cell clones that can escape CTLA4-Ig-mediated immune control.Specifically, belatacept-treated renal transplant recipients possessing a higher pretransplant frequency of CD28 þ CD4 þ T cells (that could rapidly downregulate CD28 upon ex vivo stimulation) were more likely to experience acute rejection. 19,20Additional immunophenotyping of kidney transplant recipients highlighted a multifunctional population of CD57 þ CD28 -CD2 þ CD4 þ T cells, associated with belatacept-resistant rejection and capable of infiltrating the graft. 21The heterogeneous and inverse expression of CD2 and CD28 on human T cells generated the hypothesis that dual pathway targeting should provide coverage against all alloreactive T cells when they are dependent on at least 1 of these 2 costimulatory pathways.Encouragingly, belatacept-resistant alloreactive T cells (CD2 hi CD28 -CD8 þ ) have been shown to be susceptible to the treatment with the antiCD2 fusion protein alefacept in vitro. 6The approach was extended into a preclinical nonhuman primate kidney transplantation model with promising tolerance induction results. 7Alefacept is a fusion protein of the extracellular domain of CD58 (LFA-3) and a human IgG1 Fc fragment, that depletes CD2-expressing cells through an FcγRIII-dependent mechanism. 22Notably, by constituting the natural ligand of CD2, alefacept can function as an agonist and loses its inhibitory function when Fc-silenced. 22,23Although it was previously marketed for the treatment of psoriasis, alefacept has been voluntarily withdrawn from the market, and there are currently no marketed CD2-targeting treatments available.The investigational antiCD2 antibody siplizumab, currently in multiple phase 2 clinical trials, depletes effector memory, although sparing naïve conventional T cells, and promotes the expansion of alloreactive Tregs in vitro. 9,105][16][17] Given its unique binding and mechanisms of action, siplizumab may offer a more effective rejection prevention option than alefacept when combined with CTLA4-Ig.
Alternative de novo kidney transplant regimens have been designed to lower acute rejection rates by combining belatacept with a T cell-depleting agent.Two belatacept-based regimens, one involving alemtuzumab (a lymphocyte-depleting antiCD52 monoclonal antibody) and a second using rabbit antithymocyte globulin induction, demonstrated the potential benefits of CNIfree regimens. 18However, acute rejection rates remained high, suggesting the need for broader control of alloreactive T cells.We have recently demonstrated in vitro that siplizumab may provide a superior tolerogenic effect compared with alemtuzumab and rabbit antithymocyte globulin, providing the rationale for the evaluation of a siplizumab þ belatacept kidney transplant induction regimen. 1 When concentrations of CTLA4-Ig exerting the maximum inhibitory effect together with siplizumab were used, the Treg enrichment seen with siplizumab monotherapy was diminished. 9,10Our results are congruent with previous in vitro studies that describe a dose-dependent inhibition of Treg generation in MLRs by belatacept. 24Tregs are dependent on CD28 costimulation, with CD86 acting as the dominant ligand that supports the utilization of IL-2 secreted by activated conventional T cells. 25The strong antiproliferative effect of the combination of CTLA4-Ig and siplizumab in vitro could have reduced interleukin (IL)-2 concentrations below the threshold needed for Treg generation/survival.In a murine transplant model, high doses of CTLA4-Ig therapy conferred graft protection even after Treg depletion, suggesting that a strong net suppressive effect can compensate for the role of Treg cells. 26oreover, the use of IL-2/antiIL2-complexes improved the efficacy of CTLA4-Ig by counterbalancing its unfavorable effect on Tregs. 26The effect of belatacept on Treg frequency and suppressive function remains inconclusive in the clinic, with limited numbers of patients immunophenotyped for this cell type. 27,28Thus, one potential mechanistic drawback of the approach tested herein is that the addition of high-dose CTLA4-Ig appears to counteract the Treg enrichment properties normally exerted by siplizumab.Further investigations of other costimulatory pathways or alternative methods of targeting CD28 may identify a siplizumab pairing strategy where similarly broad suppression of T cell activation can be achieved without suppressing Treg survival or function.One possible approach includes antibody-mediated direct blockade of CD28, although sparing the CTLA-4 and PD-L1 coinhibitory signals, that is under investigation in the clinic. 29lthough MLRs using primary human PBMCs provide a good mechanistic understanding of the therapeutics tested, they also constitute this study's largest inherent limitation.Further in vivo preclinical testing is impeded by siplizumab's phylogenetic restricted binding to human and chimpanzee CD2. 30 Siplizumab, however, follows an expected biodistribution for IgG1 monoclonal antibodies and has been confirmed to exert its effector functions in both circulation and secondary lymphoid organs, 13 thereby increasing the clinical translatability of these in vitro findings.
Overall, our results demonstrate that even low doses of siplizumab combined with CTLA4-Ig are sufficient to enhance the selective depletion of memory T cells and completely attenuate any allospecific proliferation.The combined understanding of siplizumab's unique tolerogenic mechanisms of action and the in vitro findings presented herein provide a clear rationale for the clinical translation of an immunosuppression regimen combining siplizumab and CTLA4-Ig to human kidney transplantation.