T-cell-recruiting bispecific molecule therapy has yielded promising results in patients with hematologic malignancies, however, resistance and subsequent relapse remains a major challenge. T-cell exhaustion induced by persistent antigen stimulation or tonic receptor signaling has been reported to compromise outcomes of T-cell based immunotherapies. The impact of continuous exposure to bispecifics on T-cell function, however, remains poorly understood. In relapsed/refractory B-cell precursor acute lymphoblastic leukemia patients, 28-day continuous infusion with the CD19xCD3 bispecific molecule blinatumomab led to declining T-cell function. In an in vitro model system, mimicking 28-day continuous infusion with the half-life-extended CD19xCD3 bispecific AMG 562, we identified hallmark features of exhaustion arising over time. Continuous AMG 562 exposure induced progressive loss of T-cell function (day 7 vs day 28 mean specific lysis: 88.4% vs 8.6%; n = 6; p = .0003). Treatment-free intervals (TFIs), achieved by AMG 562 withdrawal, were identified as powerful strategy for counteracting exhaustion. TFIs induced strong functional reinvigoration of T cells (continuous vs TFI specific lysis on day 14: 34.9% vs 93.4%; n = 6; p < .0001) and transcriptional reprogramming. Furthermore, use of a TFI led to improved T-cell expansion and tumor control in vivo. Our data demonstrate the relevance of T-cell exhaustion in bispecific antibody therapy and highlight that T cells can be functionally and transcriptionally rejuvenated with TFIs. In view of the growing number of bispecific molecules being evaluated in clinical trials, our findings emphasize the need to consider and evaluate TFIs in application schedules to improve clinical outcomes.

The bispecific T-cell engager (BiTE®) blinatumomab is approved for treatment of relapsed/refractory B-cell precursor acute lymphoblastic leukemia and applied as continuous infusion over 28 days. The overall response rate to blinatumomab reported in clinical trials was 43 % and correlated to T-cell expansion (Zugmaier et al. 2015). In chronic viral infections, continuous antigen stimulation induces T-cell exhaustion, defined by phenotypic changes and functional impairment (Wherry 2011). Thus, we hypothesized that continuous BiTE® construct stimulation leads to T-cell exhaustion and that a treatment-free interval (TFI) reverses progressive T-cell dysfunction. To simulate continuous application of a BiTE® construct in vitro, T-cell long-term co-cultures were set up. Healthy donor T cells were stimulated in the presence of CD19+ OCI-Ly1 cells for 28 days with AMG 562, a half-life extended CD19 and CD3 specific BiTE® construct. T cells were harvested from the co-culture every 3-4 days between day 7 and 28 and assessed for markers of T-cell exhaustion: (1) AMG 562-mediated cytotoxicity of T cells was evaluated as specific lysis of CD19+ Ba/F3 target cells after 3 days, (2) T-cell expansion during the cytotoxicity assay was calculated as fold change (FC) of CD2+ counts, (3) Cytokine secretion of AMG 562-stimulated T cells was evaluated in co-culture supernatants by cytometric bead array (CBA) or after PMA/Ionomycine stimulation via intracellular cytokine staining (ICCS), (4) T-cell metabolic fitness was determined by Mito- and Glycolytic Stress Test using a Seahorse Analyzer, and (5) expression of the exhaustion-related transcription factor TOX was assessed by multiparameter flow cytometry. In order to assess the effect of a TFI on T-cell function, we cultured T cells and CD19+ OCI-Ly1 cells in the absence of AMG 562 from day 7-14 and 21-28 and compared their activity to T cells stimulated continuously with AMG 562. On day 7 of continuous (CONT) AMG 562 stimulation, we observed high cytotoxic and proliferative potential (% specific lysis=93±0.2, FC=2.9±0.2) as well as high IFN-g and TNF-a secretion analyzed by ICCS (% CD8+IFN-g+TNF-a+=23±6.7). However, cytotoxicity and proliferation decreased gradually until day 28 (% specific lysis=28±8.9; FC=0.6±0.1). CBA analysis confirmed decreasing secretion of IFN-g (day 3: 61113±12482, day 24: 3085±1351 pg/ml) and TNF-a (day 3: 1160±567, day 24: 43±7.6 pg/ml) as well as decreased IL-2 and granzyme B levels in culture supernatants. We furthermore observed highest mitochondrial fitness and basal glycolysis in T cells on day 7 of stimulation (basal OCR=2.2±0.6, maximal OCR=3.7±1.0, SRC=1.5±1.1 pmol/min/1000 cells, basal ECAR=2.0±0.4 mpH/min/1000 cells) which decreased until day 28 (basal OCR=0.4±0.2, maximal OCR=1.5±0.5, SRC=1.0±0.2 pmol/min/1000 cells, basal ECAR=0.5±0.2 mpH/min/1000 cells). In concordance, TOX increased during continuous stimulation (MFI ratio CD8+ day 7=6±0.8 to 12±0.8 on day 28). Strikingly, implementation of a TFI of 7 days led to superior cytotoxicity in T cells compared to continuously stimulated T cells (% specific lysis on day 14 CONT=34±4.2, TFI=99±2.2) and granzyme B production (CD8+; MFI ratio on day 14 CONT=124±11, TFI=303±34). Furthermore, increased proliferation during the cytotoxicity assay was observed in previously rested T cells (FC CONT=0.2±0.0, TFI=1.6±0.6). Although T cell function also decreased over time in TFI T cells, they maintained a strikingly higher cytotoxic potential (CONT=6±4.4, TFI=52±9.9) as well as higher granzyme B production (CONT=25±2, TFI=170±11) on day 28 compared to continuously stimulated T cells. In addition, TFI T cells showed increased IFN-g and TNF-a secretion after PMA/Ionomycine stimulation on day 28 (% CD8+IFN-g+TNF-a+ CONT=21±3.8, TFI=38±11.6). Our in vitro results demonstrate that continuous AMG 562 exposure negatively impacts T-cell function. Comprehensive analysis of T-cell activity in an array of functional assays suggests that continuous BiTE® construct exposure leads to T-cell exhaustion which can be mitigated through TFI. Currently, T cells from patients receiving blinatumomab are being analyzed to confirm the clinical relevance of our findings. Furthermore, RNA-Seq of continuously vs. intermittently AMG 562-exposed T cells will help us to understand underlying transcriptional mechanisms of BiTE® construct induced T-cell exhaustion. Disclosures Zieger: AMGEN Research Munich: Research Funding. Buecklein:Pfizer: Consultancy; Novartis: Research Funding; Celgene: Research Funding; Amgen: Consultancy; Gilead: Consultancy, Research Funding. Brauchle:AMGEN Inc.: Research Funding. Marcinek:AMGEN Research Munich: Research Funding. Kischel:AMGEN: Current Employment, Current equity holder in publicly-traded company, Patents & Royalties. Subklewe:Gilead Sciences: Consultancy, Honoraria, Research Funding; Pfizer: Consultancy, Honoraria; Morphosys: Research Funding; Seattle Genetics: Research Funding; AMGEN: Consultancy, Honoraria, Research Funding; Janssen: Consultancy; Roche AG: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Celgene: Consultancy, Honoraria.

BiTE ® (Bispecific T-cell Engager) constructs represent a novel immunotherapeutic strategy that recruits T cells against cancer cells independent of their TCR specificity. Currently, two CD33xCD3 BiTE ® antibody constructs (AMG 330 & AMG 673) are being investigated in phase I dose escalation trials in patients with relapsed/refractory Acute Myeloid Leukemia (AML) with early evidence of acceptable safety and anti-leukemic activity (Ravandi et al., ASH 2020; Subklewe et al., EHA 2020). So far, details of BiTE ® mediated T-cell engagement and information on parameters contributing to their efficacy need more investigation. Therefore, we aimed to characterize the interplay between target and effector cells to deepen our mechanistic understanding of BiTE ® construct mediated T-cell engagement. Previously, we have created a novel in vitro model system with murine Ba/F3 cells expressing human (hu) CD33 ± huCD80 ± huCD86 ± huPD-L1 to study T-cell proliferation and cytotoxicity induced by AMG 330. Using that system, we showed that expression of T-cell co-signaling receptors on target cells modulate AMG 330 induced T-cell activity (Marcinek et al., ASH 2018, EHA 2019). Here, we hypothesize that expression of costimulatory molecules impacts BiTE ® mediated immune synapse formation and consecutive downstream signaling in BiTE ® construct activated T cells. To study whether AMG 330 can induce synapse formation and TCR triggering we used a previously described reconstituted T-cell system, which consists of non-immune (HEK) cells introduced with genes encoding the TCR and other proteins (e.g. CD45) required for the regulation of TCR phosphorylation (James et al., Nature 2012). HEK-T cells were incubated with huCD33 transduced RajiB cells in presence of fluorescently labeled AMG 330 or a control BiTE® (cBiTE) construct to allow cell conjugation. A spinning disc confocal microscope system was used to image cells. To pinpoint the role of T-cell co-signaling receptors in immune synapse formation we incubated differentBa/F3 cell constructs or primary AML (pAML) cells with healthy donor T cells in the presence of AMG 330 and analyzed intensity of LFA-1 expression within the synapse using an Imaging Flow Cytometer. Furthermore, we determined phosphorylation of ZAP70, AKT and ERK in conjugated T cells after various time points by phosphoflow cytometry. We observed that AMG 330, in contrast to cBiTE®, induced TCR triggering reflected by exclusion of CD45 from the RajiB-T-cell-interface. Simultaneously clustering of CD33 occurred in AMG 330 induced cell-cell-interfaces (Fig. 1A/B). The percentage of conjugates formed with huCD33 + Ba/F3 cells was significantly higher in constructs expressing huCD86, compared to those expressing no costimulatory antigens or additional huPD-L1 (Mean % in huCD33 + Ba/F3: 2.8 vs. huCD33 + CD86 +.Ba/F3: 4.2 [p=0.0031] vs. huCD33 + huCD86 + PD-L1 + Ba/F3: 2.8 [p=0.0018]). This was accompanied by LFA-1 accumulation within the T-cell-Ba/F3 cell synapse (Mean of MFI in huCD33 + CD86 +.Ba/F3: 10,933 > huCD33 + huCD86 + PD-L1 + Ba/F3: 7,749 > huCD33 + Ba/F3: 7,028). For downstream signaling in T cells after engagement with Ba/F3 cell constructs in the presence of AMG 330, we observed that kinase phosphorylation was highest after 10 minutes in CD86 co-expressing Ba/F3 cells (Mean % of phosphorylation in T-cell conjugates with huCD33 + vs huCD33 + huCD86 + vs huCD33 + CD86 +.PD-L1 + Ba/F3: pERK 40.9 vs 54.3 [p=0.0064] vs 51.2 %; pAKT: 69.1 vs 81.5 [p=0.0642] vs 75.1 %; pZAP70: 6.9 vs 12.2 [p<0.0001] vs 7.7 % [p<0.0001]) (Fig. 1C). Finally, we evaluated if these finding could also be observed in pAML samples. For that, we determined LFA-1 expression intensity within AMG 330-induced pAML-T-cell synapses. We used CD33 + pAML samples with either high CD86 and no PD-L1 expression or vice versa. Comparing synapse formation between these samples, LFA-1 intensity was 4.6-fold higher in the CD86 + PD-L1 - sample compared to the CD86 - PD-L1 + pAML. Taken together, our data unravel molecular mechanisms of BiTE® construct induced immune synapse formation, highlighting the role of costimulatory molecules in this process. They support the notion that T cell co-signaling receptors like CD86 and PD-L1 modulate T-cell response in an early event manner. Prospective analyses in clinical trials are needed to validate the relevance of checkpoint molecule expression on target cells as a potential predictive biomarker for response. Figure 1 Figure 1. Disclosures Brauchle: Adivo: Current Employment. Lacher: Roche: Research Funding. Kischel: Amgen GmbH Munich: Current Employment. von Bergwelt: Roche: Honoraria, Research Funding, Speakers Bureau; Miltenyi: Honoraria, Research Funding, Speakers Bureau; Mologen: Honoraria, Research Funding, Speakers Bureau; Kite/Gilead: Honoraria, Research Funding, Speakers Bureau; Novartis: Honoraria, Research Funding, Speakers Bureau; Astellas: Honoraria, Research Funding, Speakers Bureau; MSD Sharpe & Dohme: Honoraria, Research Funding, Speakers Bureau; BMS: Honoraria, Research Funding, Speakers Bureau. Theurich: Amgen: Consultancy, Honoraria; BMS/Celgene: Consultancy, Honoraria; GSK: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; Sanofi: Consultancy, Honoraria; Takeda: Consultancy, Honoraria. Buecklein: Novartis: Consultancy, Other: congress and travel support, Research Funding, Speakers Bureau; Pfizer: Consultancy, Honoraria, Speakers Bureau; Miltenyi: Research Funding; Kite/Gilead: Consultancy, Honoraria, Other: Congress and travel support, Research Funding; BMS/Celgene: Consultancy, Research Funding; Amgen: Consultancy, Honoraria. Subklewe: Janssen: Consultancy; Seattle Genetics: Consultancy, Research Funding; Roche: Research Funding; Novartis: Consultancy, Research Funding, Speakers Bureau; Pfizer: Consultancy, Speakers Bureau; Klinikum der Universität München: Current Employment; Takeda: Speakers Bureau; MorphoSys: Research Funding; Miltenyi: Research Funding; Gilead: Consultancy, Research Funding, Speakers Bureau; Amgen: Consultancy, Research Funding, Speakers Bureau; BMS/Celgene: Consultancy, Research Funding, Speakers Bureau.

Bispecific antibodies represent a promising treatment option for acute myeloid leukemia (AML). We have recently described a novel T-cell bispecific antibody (TCB) targeting the intracellular tumor antigen Wilms tumor 1 (WT1) in the context of HLA-A*02 (Augsberger et al. Blood 2021). Based on these findings a multicenter first-in-human clinical trial was initiated in relapse/refractory AML (NCT04580121). Possible immune escape mechanisms against T-cell based immunotherapy are provided by the tumor microenvironment (TME) of the bone marrow by co-inhibition of T cells or stromal cells shielding leukemic cells from immune effector cells. To overcome the immunosuppressive effect of the TME and to enhance T-cell responses, we evaluated the combination of the WT1-TCB with an antibody fusion protein that targets a stromal antigen (Fibroblast-activation protein; FAP) and provides a positive costimulatory signal (4-1BBL) to T cells. FAP is upregulated on cancer-associated fibroblasts after remodulation of the bone marrow niche by leukemic cells, and the FAP specificity of the molecule therefore provides T-cell co-stimulation tightly restricted to the tumor niche. Efficacy of the combination (WT1-TCB + FAP-4-1BBL antibody fusion protein) was evaluated in co-culture assays over 4 days with primary HLA-A*02 + AML cells, healthy donor (HD) T cells and three NIH-3T3 fibroblast cell lines. NIH-3T3 cell lines were genetically modified to express low and high levels of FAP, respectively. Wild-type NIH-3T3 cells were included as control. Additionally, a control (Ctrl)-TCB and a Ctrl-4-1BBL antibody fusion protein recognizing a non-tumor target derived from the human germline repertoire were included. Enhancement of T-cell mediated cytotoxicity by the FAP-4-1BBL antibody fusion protein was evaluated by (1) specific lysis of primary AML cells, (2) upregulation of the T-cell activation markers CD25 and 4-1BB, (3) T-cell expansion calculated as fold change compared to day 0, and (4) Granzyme B-expression which was evaluated by intracellular staining. After 4 days of co-culture, with an E:T ratio of 1:2, we observed a mean specific lysis of 55.1±8.2% (±SEM; n=4) of primary AML cells mediated by HD T cells and WT1-TCB. Notably, this was reduced to 19.4±5.9% (±SEM; n=4) in the presence of NIH-3T3 cells. However, AML cell lysis was restored by the addition of the FAP-4-1BBL antibody fusion protein in the presence of high FAP expressing NIH-3T3 cells (mean specific lysis: 62.8±7.3%; ±SEM; n=4). Concomitantly, the FAP-4-1BBL antibody fusion protein led to increased expression of the activation molecules CD25 (MFI ratio: 22.1±5.3 vs. 10.4±1.3; ±SEM; n=4) and 4-1BB (MFI ratio: 10.4±6.0 vs. 2.1±0.3; ±SEM; n=4) on CD3 + T cells. Furthermore, lysis was accompanied by increased frequencies of granzyme B expressing T cells (45.0±2.5% vs. 16.1±5.3%; n=3). Importantly, the FAP-4-1BBL antibody fusion protein led to improved T-cell proliferation, especially of CD8 + T cells (fold change on day 4 vs day 0: 5.7±2.2 vs. 1.0±0.3; ±SEM; n=4). Overall similar observations were made in the presence of low FAP expressing NIH-3T3 cells. Taken together, we have established an in vitro model system mimicking the immunoprotective bone marrow TME using NIH-3T3 cells resulting in impaired AML cell lysis. Providing additional T-cell co-stimulation by a tumor-stroma targeted 4-1BB agonist, however, restored WT1-TCB-mediated cytotoxicity of primary AML cells in the presence of FAP expressing cell lines. Importantly, the combination overcame the immunosuppressive effect of the NIH-3T3 cells on T cells as further demonstrated by improved T-cell activation and expansion. The tumor-stroma targeted 4-1BB agonist therefore represents a promising combinatorial approach to enhance T-cell activity at the local tumor site and warrants further investigations in an in vivo model system. Disclosures Pulko: Roche: Current Employment, Current equity holder in publicly-traded company, Patents & Royalties. Claus: Roche: Current Employment, Current equity holder in publicly-traded company, Patents & Royalties. Buecklein: Pfizer: Consultancy, Honoraria, Speakers Bureau; Kite/Gilead: Consultancy, Honoraria, Other: Congress and travel support, Research Funding; Novartis: Consultancy, Other: congress and travel support, Research Funding, Speakers Bureau; Miltenyi: Research Funding; BMS/Celgene: Consultancy, Research Funding; Amgen: Consultancy, Honoraria. Umana: Roche: Current Employment, Current equity holder in publicly-traded company, Patents & Royalties. Klein: Roche: Current Employment, Current equity holder in publicly-traded company, Patents & Royalties. Subklewe: Novartis: Consultancy, Research Funding, Speakers Bureau; Klinikum der Universität München: Current Employment; Roche: Research Funding; Seattle Genetics: Consultancy, Research Funding; Pfizer: Consultancy, Speakers Bureau; Janssen: Consultancy; Takeda: Speakers Bureau; MorphoSys: Research Funding; Miltenyi: Research Funding; Gilead: Consultancy, Research Funding, Speakers Bureau; Amgen: Consultancy, Research Funding, Speakers Bureau; BMS/Celgene: Consultancy, Research Funding, Speakers Bureau.

Novel immunotherapeutic strategies like BiTE ® (bispecific T cell engager) constructs aim to eradicate neoplastic cells by TCR-independent T-cell activation, and therefore rely on the function of autologous T cells. Currently, their efficacy is also evaluated in heavily pre-treated patients with relapsed/refractory acute myeloid leukemia (AML). Previous data demonstrated dysfunction in CD8 + T cells of AML patients (Knaus et al 2018). Thus, we aimed to characterize the progressive modulation of T-cell activity over the course of AML progression to improve the optimal application of T-cell based immunotherapeutic approaches. Bone marrow mononuclear cells (BMMCs) from AML patients at time of initial diagnosis (ID), complete remission (CR), relapse (RL), as well as of age-matched healthy donors (HD) were analyzed for T-cell subset distribution and expression of exhaustion markers by flow cytometry. Additionally, T-cell function was assessed after stimulation with 1) CD3/CD28 beads; 2) AMG 330, a CD33/CD3 specific BiTE ® construct, after incubation with OCI-AML3 target cells; or 3) AMG 330 in an autologous ex vivo long-term culture system after incubation with primary AML cells (pAML). After 6 days, T cell proliferation, expression of effector molecules and cytokines, and AMG 330-mediated T-cell cytotoxicity were assessed by flow cytometry. Lastly, we performed longitudinal bulk RNA-sequencing on 5000 sorted T cells from 7 matched ID-RL primary AML samples. Immunophenotypic analysis of BM T-cell subsets revealed a shift from T NAIVE toward central/effector memory subsets during AML progression. We observed lower percentages of T NAIVE in RL (n=3) compared to CR (n=3) CD8 + T cells(11.8 vs. 45.2%, p=0.07; RL vs. CR). Conversely, RL patients showed increased percentages of CD8 + memory T cells (T CM: 23.4 vs. 6.7%; T EM: 29.4 vs. 20.2%; T EMRA: 35.3 vs. 27.8%; RL vs. CR). Further characterization of exhaustion markers exhibited a significantly higher percentage of both CD4 + and CD8 + T cells expressing 2B4 (CD244) in ID (n=19) and RL (n=13) compared to HD (n=10, both p < 0.001). A higher percentage of PD-1 + CD8 + and TIM-3 + CD4 + T cells was detected in both ID and RL relative to HD (all p < 0.05). However, a significantly increased percentage of CD8 + T cells expressing TIM-3 and CD160 was detected in ID relative to HD (p < 0.05). Intriguingly, RL CD4 + T cells demonstrated a significantly higher level of LAG3 compared to ID (p < 0.01). In line with phenotypic exhaustion features, ID (n=4) and RL (n=5) CD8 + T cells showed reduced proliferation compared to HD (n=4) CD8 + T cells after CD3/CD28 bead stimulation (both p < 0.01). Correspondingly, we observed a marked reduction in the expression of Granzyme B (GZMB) by CD8 + T cells (both p < 0.05). Interestingly, when compared to ID, RL CD4 + T cells showed decreased TNF-α secretion (p < 0.05). In contrast to these findings, AMG 330-mediated T cell cytotoxicity against OCI-AML3 target cells was superior with RL T cells compared to ID T cells (p < 0.001). The percentage of GZMB + CD8 + T cells strikingly enhanced in RL relative to ID (p < 0.01). In an autologous setting with pAML samples, T cells from RL patients (n=6) showed higher AMG 330-mediated cytotoxicity compared to ID (n=9) T cells (67.7 vs. 35.2; RL vs. ID). In our longitudinal RNA-sequencing, differentially expressed genes analysis detected 61 up- and 30 downregulated genes (log2 FC > 1 or < -1; p < 0.01) in RL T cells compared to their matched ID counterparts. Among the significantly upregulated genes in RL, we identified genes associated with memory T cell function (TP53INP2, DUSP4) and exhaustion (NR4A1, TOX2). Moreover, Gene set enrichment analysis showed significant enrichment of gene signatures associated to memory and exhausted T cells (normalized enrichment score (NES)=1.2 and 1.3; p-value= 0.026 and 0.008, respectively), depletion of oxidative phosphorylation (NES=-2.05; p adj < 0.0001) and protein secretion (NES=-1.49; p adj < 0.05) gene signatures in RL vs. ID T cells. Taken together, our data show that patient T cells acquire an activated/exhausted phenotype upon AML progression. However, this is not reflected in the T-cell effector functions upon AMG 330 stimulation, in contrast to bead stimulation. These observations may highlight the significant role of the AML target cells in shaping a T-cell response. To this end, we will further analyze the longitudinal communication between T cells and their corresponding AML blasts. Disclosures Brauchle: Adivo: Current Employment. Kischel: Amgen GmbH Munich: Current Employment. Buecklein: BMS/Celgene: Consultancy, Research Funding; Amgen: Consultancy, Honoraria; Kite/Gilead: Consultancy, Honoraria, Other: Congress and travel support, Research Funding; Miltenyi: Research Funding; Novartis: Consultancy, Other: congress and travel support, Research Funding, Speakers Bureau; Pfizer: Consultancy, Honoraria, Speakers Bureau. Subklewe: Novartis: Consultancy, Research Funding, Speakers Bureau; MorphoSys: Research Funding; Roche: Research Funding; Miltenyi: Research Funding; Seattle Genetics: Consultancy, Research Funding; Gilead: Consultancy, Research Funding, Speakers Bureau; BMS/Celgene: Consultancy, Research Funding, Speakers Bureau; Amgen: Consultancy, Research Funding, Speakers Bureau; Janssen: Consultancy; Pfizer: Consultancy, Speakers Bureau; Takeda: Speakers Bureau; Klinikum der Universität München: Current Employment.