The Battle within: AML´s p53 Strategies to Evade T-Cell Attack

Abstract

Bispecific T-cell recruiting antibody constructs (BsAb) have shown clinical efficacy with Blinatumomab, a CD3xCD19 BiTE ® construct used for the treatment of relapsed or refractory B-cell precursor ALL. Despite the success of BsAbs in B-cell malignancies, the translation to a myeloid setting has been unsuccessful, independently of the chosen target antigen. For example, targeting CD33 has failed to provide long-term benefits. Several reports have demonstrated that the diverse genetic makeup of AML confers resistance to T-cell based immunotherapy. Mutations of TP53 occur in 5-10% of patients with de novo AML and up to 50% in older patients with therapy related AML, thus highlighting its role as a potential resistance mechanism. We hypothesize, that genetic aberrations of TP53 in AML contribute to cell intrinsic and cell extrinsic resistance against T-cell based immunotherapy approaches. To study the impact of TP53 aberrations on T-cell based immunotherapy, MV4-11 p53 WT or MV4-11 p53 KD were cocultured with healthy donor T cells at an effector to target ratio of 1:6 with or without BsAbs. BsAb-mediated cytotoxicity was analyzed by specific lysis of the CD33 + target cells and T-cell proliferation was assessed by FarRed ® staining or CD2 + T-cell fold change calculated to day 0. Cytokine secretion was monitored in the coculture supernatants by cytometric beads arrays. To differentiate between secretome vs surfacome mediated differences, transwell assays were performed. To identify the T-cell immunosuppressive factor, secretome analysis were performed using the Olink® proteomics platform. To further dissect the effect of the p53 KD MV-411 on T cells, we performed RNA-Seq analysis of T cells after coculture with MV4-11 p53 KD vs p53 WT. As expected, BsAb-mediated cytotoxicity was reduced against p53 KD vs p53 WT MV4-11 (% specific lysis: p53 KD=47.86±7.37 vs p53 WT=76.42±3.82; p=0,0001). However, BsAb-mediated T-cell proliferation was also significantly decreased in cocultures with p53 KD vs p53 WT (% proliferated: 30.26±2.92 vs 44.25±4.08; p=0.002). In line with these findings, we observed a significant decrease in secretion of the proinflammatory cytokines IFNγ (p53 KD=2985,01±896,69 vs p53 WT=4009.15±764.97 pg/ml; p=0.068), TNF (29,38±8.16 vs 110.21±37.61 pg/ml; p=0.044) and IL-2 (1391.64±375.93 vs 2253.88±604.76 pg/ml; p=0.018). RNA-Seq analysis of T cells after coculture with MV4-11 p53 KD vs p53 WT revealed a decrease in the expression of genes associated with mitosis (such as E2F targets and mitotic spindle related genes). A lower mitotic rate provides a potential explanation for the reduction in T-cell proliferation after coculture with p53 KD. To dissect if the observed effects were due to the AML surfacome or the secretome, we performed transwell assays. Indeed, the negative impact of p53 KD AML cells on T-cell function was also observed without direct cell-cell contact. To gain further insights in the secretome of the AML cells, coculture supernatant was analysed using the Olink ® platform and demonstrated significantly higher secretion of IL-18 and LAP TGF-β1 (TGF-β1 latency associated protein) in the coculture with MV4-11 p53 KD vs MV4-11 p53 WT (normalized protein expression for IL-18: p53 KD=1.86±0.22 vs p53 WT=0.96±0.052; p=0.0153; for LAP TGF-β1: 6.52±0.18 vs 5.84±0.14; p=0.0394). The relevance of TGF-β on T-cell fitness was confirmed by the analysis of BsAb-mediated cytotoxicity with or without the addition of TGF-β. Addition of 3 ng/µl TGF-β to the coculture with p53 WT decreased the specific lysis significantly and was now comparable to the lysis of p53 KD. Confirmatory studies in primary AML samples with or without p53 aberrations further validated our findings with a clear reduction of BsAb-mediated cytotoxicity (% specific lysis: p53 KD=35.82±5.51 vs p53 WT=61.32±8.27; p=0.015), T-cell proliferation (CD2 + fold change: 3.17±0,61 vs 7.1±2.8; p=0.111) and secretion of proinflammatory cytokines. Taken together, we demonstrated that AML cells carrying a p53 aberration are less susceptible to BsAb-mediated cytotoxicity. Surprisingly, immune evasion was mediated by the AML secretome leading to a decrease in T-cell proliferation and cytokine secretion. Secretome analyses indicated TGF-β as a prominent cytokine leading to an impairment of T-cell function. Further studies are needed to understand the influence of p53 on the TGF-β pathway in AML cells.