Introduction: Adult T-cell leukemia/lymphoma (ATL) is a lymphoid neoplasm of CD4+ T lymphocytes caused by the human T-cell leukemia virus type I (HTLV-1), which is classified into 4 clinical subtypes (ie, smoldering, chronic, acute, and lymphoma). Natural killer receptors (NKRs) were previously identified on T-cell lymphoma. Methods: NKRs expression was assessed by flow-cytometry and immunohistochemistry on peripheral blood samples (n = 55) and biopsies (n = 21) obtained from 50 ATL patients. Array-based analysis of genomic DNA methylation patterns of KIR3DL2 promoter was assessed. To explore the role of HTLV-1 on KIR3DL2 expression, KIR3DL2 and TAX mRNA expressions were assessed on primary ATL cells and on activated CD4+ T cells that were infected with HTLV-1 in-vitro. Ex-vivo autologous antibody dependent cell cytotoxicity (ADCC) was performed on sorted primary ATL cells with IPH4102, a monoclonal anti-KIR3DL2 antibody that has shown robust clinical activity in Phase I in patients with relapsed Sézary Syndrome (NCT02593045). Results: KIR3DL2 was the only detected NKR, mainly expressed on acute- compared to lymphoma- and chronic-types ATL (28/30 vs. 2/8 and 2/12 respectively; p = 0.001). KIR3DL2 expression correlated with the demethylation status in its promoter, and treatment by 5-Aza increased its expression from a positive basal state. TAX mRNA and KIR3DL2 expressions were correlated on primary ATL cells. HTLV-1 infection triggered KIR3DL2 protein expression by CD4+ cells. However, Tax alone is not sufficient to induce KIR3DL2 mRNA expression. Treatment with IPH4102 can selectively kill human KIR3DL2+ primary ATL cells ex-vivothrough ADCC with autologous NK cells. Conclusion: KIR3DL2 expression is mainly associated with acute-type ATL, the most frequent subtype with the poorest prognosis. Induction of KIR3DL2 gene transcription may be triggered by HTLV-1 infection followed by transcription maintenance due to DNA hypomethylation of the gene promoter. The benefit of targeting KIR3DL2 by IPH4102 should be further investigated in ATL patients. Keywords: human T-lymphotropic virus (HTLV); monoclonal antibodies (MoAb); peripheral T-cell lymphomas (PTCL). Disclosures: Cheminant, M: Research Funding: yes. Bruneau, J: Research Funding: yes. Sicard, H: Employment Leadership Position: yes; Stock Ownership: yes. Bonnafous, C: Employment Leadership Position: yes; Stock Ownership: yes. Hermine, O: Research Funding: yes.
Background: T-cell acute lymphoblastic leukemia (T-ALL/T-LL) is a class of highly aggressive cancers in children and young adult, marked by an aggressive behavior and poor clinical response, notably for refractory and relapsing (R/R) cases. As in numerous cancers, genetic lesions leading to aberrant PI3K signaling (PI3KSALT) are frequent in T-ALL and convey adverse outcomes from limited therapy response to early relapse and dismal survival rates. Aims: We aimed to explore the metabolic plasticity of PI3K-driven leukemia and unravel targets amenable to novel strategies to leverage the treatment and the outcome of this aggressive disease. Methods: Patients were enrolled in the GRAALL2003-2005 and FRALLE2000 trials. Patient primary samples were analyzed by pan-exome sequencing, arrayCGH, MLPA and RNA-sequencing. Patient-derived xenografts (PDX) were used for in vivo experimentation. PI3K signaling activity, glucose consumption and cell survival were evaluated by flow cytometry. Metabolomics analyses were carried out at our Metabolomics Core. Results: PI3KSALT patients had a poor response to corticoids, a shorter overall survival (5y-OS: 58% vs 74%, p = 0.007) and event-free survival (5y-EFS: 49% vs 65%, p = 0.008), and an increased incidence of relapse (5y-CIR: 39% vs 27%, p = 0.01). PI3KSALT define a subclass of aggressive T-ALL with a poor prognosis, urging innovative therapies for these patients. Primary samples and PDX of PI3KSALT leukemia have an hyperglycolytic profile. PI3KSALT T-ALL cell lines mimic this addiction to glucose. Surprisingly, while cell lines cannot survive glucose limitation, PI3KSALT PDX tolerate this starvation, underpinning their ability to rewire their metabolism to adapt to a nutrient-deprived microenvironment. We showed that PI3KSALT PDX have unique metabolic plasticity upon starvation. These blasts use glutaminolysis to cope with glucose limitation and sustain the TCA cycle, while wild-type PI3KS fail to do so (Fig 1a). Pharmaceutical inhibition of the PI3KS-mTOR pathway and glutamine metabolism presented a marked cytotoxicity ex vivo (Fig 1b). We proposed a therapeutic strategy for PI3KSALT T-ALL based on the targeting of mTOR and glutamine. Erwinase, an L-asparaginase with glutaminase activity, efficiently synergizes with Torisel and demonstrates tumor clearance and prolonged survival in vivo (Fig 1c-d). Critically, we report the case of five patients suffering from R/R PI3KSALT leukemia, with complex clinical history, treated with the association Erwinase/Torisel (ET). All five patients achieved a rapid response. Patients suffering from R/R T-ALL achieved negative minimal residual disease. Patients with R/R T-LL presented a significant decrease in the mediastinal mass (Fig 1e). After complete remission, three patients received consolidation therapy consisting of allogeneic SCT or donor lymphocyte infusion. Two patients remain alive, while two patients rapidly progressed after the discontinuation of ET treatment. Summary/Conclusion: R/R T-ALL/LL have dismal prognosis and outcomes, in part due to chemoresistance acquisition and limited therapeutic options. We propose a promising treatment combining an asparagine and glutamine degrader (Erwinase) with a PI3KS inhibitor (Torisel) that should be considered as a therapeutic option in a bridge-to-transplant approach for R/R T-ALL/LL with PI3KS deregulation. We show that metabolic plasticity conveys a unique and targetable vulnerability in PI3KS-driven leukemia that show promising results in pre-clinical and clinical settings.Keywords: Acute lymphoblastic leukemia, T cell acute lymphoblastic leukemia, Targeted therapy, PI3-K/AKT
Figure S1. Appendix S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
ATL carries a very bad prognosis because of intrinsic chemoresistance and severe immuno-suppression. In acute ATL, chemotherapy combinations improved response rate but failed to achieve a significant impact on survival. Patients with chronic and smoldering ATL have a better prognosis but long-term survival is poor when patients are managed with a watchful-waiting policy or with chemotherapy. The antiviral combination of zidovudine (AZT) and interferon-alpha (IFN) improves survival in leukemic subtypes of ATL (smoldering and chronic ATL and acute ATL with wild type p53) and should be considered as standard first line therapy. However, it is mandatory to 1) use it in leukemic forms as first line therapy and not after one or more cycles of chemotherapy; 2) start with high doses of both agents since reduced doses are often not effective. ATL lymphoma may benefit from initial chemotherapy combined to or followed by AZT/IFN and this approach should be tested in prospective clinical trials. Prophylaxis of opportunistic infections and intrathecal chemotherapy are mandatory. Yet, most patients relapse and alternative therapies are mandatory. IFN and arsenic trioxide induce Tax proteolysis, synergize to induce apoptosis in ATL cells and cure Tax-driven ATL in mice through specific targeting of leukemia initiating cell activity. Therefore, to prevent relapse, clinical trials assessing consolidative targeted therapies such as arsenic/IFN combination, histone deacetylase inhibitors or novel monoclonal antibodies particularly brentuximab in CD30 positive cases or the promising anti CCR4 antibodies, are mandatory after achieving CR with AZT/IFN. Finally, allogeneic BMT should be considered in suitable patients.
