The present invention relates to an asymmetric bispeci fic antibody, directed towards cortical-derived T-ALL cancer cells expressing CDla, capable of recruiting, activating and redirecting normal T cells against cancer cells .

T-ALL is a hematologic neoplasm characteri zed by an aberrant proli feration of T cell progenitors starting from an intratimical di f ferentiation step that leads to the progressive infiltration of the bone marrow and lymphoid organs and the spread of immature leukemic T cells in the peripheral blood .

T-ALL is a hematologic, rare, orphan and very aggressive neoplasm characterized by frequently unfavorable evolution and poor prognosis in patients relapsed/ref ractory to conventional treatments . While the progress of immunotherapy has greatly improved the treatment of acute B-cell lymphoblastic leukemia (B-ALL ) , the lack of antigens with restricted expression on tumor T cells , therefore suitable for a selective targeting of the same saving the normal T- compartment , has signi ficantly hindered the development of new immunotherapy strategies towards this important clinical condition . It is therefore absolutely important to identi fy and use therapeutic targets that provide the rationale for the development of new therapeutic strategies .

The obj ect of the present invention is to provide a new bispeci fic antibody suitable to activate and redirect cytotoxic T-cells (bispeci fic T cell engager, BTCE ) , by means of CD3s binding CDla which is expressed by about 40% of T- ALL cases .

According to the present invention, a new asymmetric bispeci fic antibody, BTCE , is made against T-ALL, as defined in claim 1 .

For a better understanding of the present invention, a preferred embodiment is now described, by way of non-limiting example only, with reference to the figures , in which :

- Figures 1A- 1D show the experimental results relating to UMG2 antibody binding reactivity (mAb ) , in particular Figure 1A shows the reactivity of UMG2 on HEK293T cell lines that do not express CDla trans fected with a control vector ( left ) and on HEK293T/CDla trans fected with a plasmid encoding CDla ( right ) ; Figure IB shows the reactivity of UMG2 on patient-derived T-ALL cells ; Figure 1C shows the reactivity of UMG2 on cortical- derived T-ALL cell lines; Figure ID shows the binding of UMG2 on peripheral blood cells from healthy donors. Figures 2A-2F show experimental data relating to the in vitro cytotoxic activity of the UMG2/CDla-CD3 s BTCE construct; in particular, Figure 2A shows on the left: structure of UMG2-CD3s BTCE, on the right: targeting of CD3+ T cells towards CDla+ T-ALL cells mediated by the UMG2/CDla-CD3s BTCE construct evaluated by immunofluorescence microscopy analysis; Figure 2B shows data relating to the evaluation of the dissociation constant (Kd) of UMG2/CDla-CD3s BTCE evaluated by titration experiments on cortical T-ALL cell lines and CDla-expressing; Figure 2C shows the relative percentage (%) of induced cytotoxicity HEK293T cells transfected with an empty plasmid or with a plasmid encoding CDla; Figure 2D shows the relative percentage (%) of cytotoxicity on cortical-derived T-ALL, CDla+ and CDla- cell lines; Figure 2E shows data of the relative percentage (%) of cytotoxicity against primary CDla+ T-ALL cells (derived from patients) ; Figure 2F shows the increase in the concentration of CD107a on co-cultured T cells in an E:T ratio of 10 a 1 (10:1) both in T-ALL cell lines expressing UMG2 and primary cells derived from T-ALL patients; - Figure 3 shows experimental data relating to the in vitro functional activation of T cells in a concentration-dependent manner of UMG2/CDla-CD3s BTCE, assessed by activation of early and late markers (respectively CD69 and CD25) on both CD4 and CD8 positive T cells, by release of enzymes and cytotoxic molecules (granulisin, perforin and granzyme) , and production of cytokines (TNF-a, IFN-y and IL-2) assessed by co-culture experiments with HPB-ALL (A) , TALL-1 (B) and Jurkat (C) in the presence of increasing concentration (0.01-0.1-1 pg/ml) of UMG2/CDla-CD3s BTCE .

- Figure 4 shows the in vitro activity of T-cell-mediated UMG2/CDla-CD3s BTCE, in particular Figure 4A shows the level of NFAT1 protein on peripheral mononuclear blood cells (PBMCs) co-cultured with T-ALL cells in the presence of UMG2/CDla-CD3s BTCE or vehicle as a control; Figure 4B shows the proliferation of T cells co-cultured with HPB-ALL, TALL-1 and Jurkat cells respectively, in the presence of UMG2 /CDla-CD3 s BTCE; Figure 4C shows the extent of cytotoxicity towards neoplastic cells by PBMCs, to which CD4 and CD8 expressing lymphocytes, co- cultured with T-ALL cells and increasing concentrations of UMG2/CDla-CD3s BTCE have been subtracted. Figure 4D shows the relative percentage (%) of induced cytotoxicity on neoplastic cells by PBMCs, subtracted from CD56-expressing lymphocytes, or enriched with positive CD56, and PBMCs in the presence of Fc fragment blocker, co-cultured with Jurkat CDla-expressing T-ALL cells and increasing concentration (0.01-0.1- 1 pg/ml) of the BTCE construct;

- Figure 5 shows experimental data relating to the in vivo activity of UMG2/CDla-CD3s BTCE; in particular Figure 5A shows the evaluation of in vivo fluorescence in NSG immunosuppressed mice (NOD/SCID gamma, non-obese diabetic/severe combined immunodeficiency) , inoculated subcutaneously with HPB-ALL cells and treated with two weekly intraperitoneal doses of the BTCE construct at the concentration of 0.1 mg/kg or 0.5 mg/kg, or with the vehicle (PBS) , as control; Figure 5B shows the survival curves (Kaplan-Meier) of the treated mice compared to the control group (log-rank test, p<0.05) ;

- Figure 6 shows the combined treatment data for UMG2/CDla-CD3s BTCE and immune-checkpoint inhibitors (ICTs) . Figure 6A shows the expression of T-cell antitumor function exhaustion markers (PD-1, TIM3, and TIGIT) on CD4 and CD8 positive T cells co-cultured with CDla-expressing T-ALL cells in the presence of the BTCE construct in the presence of nivolumab and/or avelumab, compared to treatment with the BTCE construct alone. Figure 5B shows the relative percentage (%) of cytotoxicity on CDla-expressing T-ALL cells co-cultured with PBMC in the presence of the BTCE construct alone, nivolumab alone, avelumab alone, BTCE construct plus nivolumab or avelumab.

According to the invention, a new asymmetric bispecific antibody (UMG2/CDla-CD3s ) BTCE capable of recruiting T cells against cortical-derived T-ALL cells has been developed. In particular, the construct generated is an asymmetric 2+1, with a bivalent bond towards CDla and monovalent towards CD3s, made using the "knobs into holes" technology and starting from the sequence of scFv of the UMG2 antibody. The latter derives from the homonymous hybridoma obtained by long-term culture and numerous subclonings of the parental hybridoma UN5, already characterized and clustered as anti- CDla. The UMG2 hybridoma deposited on 17 September 2021, access number assigned by the International Depository Authority PD21003, produces the UMG2 antibody which has, compared to the parental antibody, high affinity for the epitope and a slight difference in the pattern of reactivity towards the target. The UMG2 antibody is therefore novel and directed towards a unique epitope of CDla, a glycoprotein highly expressed by T-ALL cells.

The Applicant has carried out experimental tests, some of which are shown in the figures, in which the in vitro activity of the UMG2 /CDla-CD3 s BTCE construct was evaluated on cancer cells co-cultured with peripheral blood mononuclear cells derived from healthy donors ( PBMCs ) in whole or depleted of CD4 /CD8 lymphocytes or enriched with CD56 lymphocytes with an E : T ratio equal to 10 : 1 . In vivo anti-tumor activity of UMG2 /CDla-CD3 s BTCE was assessed in a mouse NSG model reconstituted with human peripheral blood mononuclear cells (Hu-PBMC ) and inoculated with T-ALL cells . In vivo tumor growth was assessed by fluorescence imaging .

The Applicant investigated the targeting of CDla as a potential speci fic target for T-ALL . CDla is a transmembrane glycoprotein expressed on ~40% of cortical-derived T-ALL cases and on a population of cortical thymocytes but not on peripheral blood cells . On non-hematopoietic cells , only a subset of dendritic cells residing in the skin ( Langerhans cells , LC ) express the CDla antigen .

In order to develop a therapy against T-ALL, the Applicant has developed, according to the present invention, a new CD3 E-monovalent BTCE construct that simultaneously binds a single CDla epitope in order to recruit and trigger a potent T-cell-mediated anti-tumor response .

The in vitro and in vivo activity of this BTCE is reported below, supporting the translational value of CDla targeting strategies as a new therapy for patients with this aggressive disease . By long-term culture and several subcloning procedures , the UMG2 clone was selected from a murine hybridoma previously generated by the group of proponent researchers , which was subsequently characteri zed as an anti-CDla mAb . The UMG2 antibody recogni zes its epitope on CDla, while the af finity against the combinatorial site is slightly improved compared to the original parental clone . This clone was sequenced for generation of a bi-speci fic construct with a CD3 s monovalent arm .

To confirm that UMG2 recognizes a CDla epitope , the HEK293T cell line , negative for CDla expression, was trans fected with a plasmid encoding CDla or with a negative empty vector (EV) . After selection of the trans fected clone , speci fic binding of UMG2 or anti-CDla antibodies ( SK9 , BL6 HI 149 ) to CDla was assessed by flow cytometry . A strong binding of UMG2 was found on HEK293T cells expressing CDla while no reactivity was observed on HEK293T cells trans fected with the negative control ( Figure 1A) . Importantly, a similar reactivity model was observed for the other anti-CDla mAbs tested, further confirming the reactivity of UMG2 to CDla ( Complementary Figure 1A-B ) . In addition, to investigate whether UMG2 binds to an original CDla epitope , a competitive binding assay was performed between f luorochrome-conj ugated UMG2 and fluorochrome SK9 , BL6 , and anti-CDla HI 149- conj ugated mAbs . Interestingly, none of the anti-CDla mAbs compete with the UMG2 binding (supplementary Figure 2A) , thus indicating that UMG2 recognizes a previously undescribed CDla epitope.

Subsequently, the reactivity of UMG2 was evaluated on a panel of samples of T-ALL primary cells and cell lines. As shown in Figure IB, primary cortical T-ALL cells are recognized by UMG2. Among the 8 different cortical T-ALL cell lines tested, 3/8 are strongly positive for binding to labeled UMG2, 1/8 with lower intensity and 4 cell lines do not express the antigen (Figure 1C) . Interestingly, Ke-37, DND-41, and CCRF-CEM non-reactive cell lines with UMG2 are instead highly reactive to anti-CDla antibodies SK9, BL6, and HI149, indicating once again that the epitope recognized by UMG2 is unique and characterized by a narrow pattern of expression (Supplementary Figure 2B) compared to other anti- CDla mAbs .

Finally, the reactivity of UMG2 on peripheral blood cells from healthy donors was evaluated. No reactivity was found on the different subtypes of blood cells (T cells, B lymphocytes, NK, monocytes and neutrophils) as expected (Figure ID) , confirming a restricted expression pattern of the recognized epitope.

The structural characteristics of the asymmetric UMG2/CDla-CD3s BTCE, monovalent for the CD3s arm (2+1) , were developed to reduce the non-specific activation of T cells in the absence of a concomitant CDla binding, while the CDla reactivity domain was designed with a bivalent arm to enhance construct avidity for CDla-expressing cells (Figure 2A) . To this end, then (2+1) UMG2/CDla-CD3 s (clone L2K-07) BTCE was generated .

To define the apparent dissociation constant (Kd) of the new BTCE construct, titration experiments were performed. The average apparent Kd was estimated at 0.014 pg/mL, while binding saturation was achieved at concentrations of about 1 pg/mL (Figure 2B) .

The in vitro activity of the BTCE construct was first evaluated using HEK293T/EV and HEK293T/CDla cells, cultured together with PBMC as effector cells with ef f ector/target ratio (E:T) 10 : 1. Concentration-dependent T-cell cytotoxicity was observed to HEK293T/CDla cells but not to HEK293T/EV (Figure 2C) . To investigate whether BTCE induces a direct cytotoxic effect in the absence of T cells, the viability of HEK293T/CDla and HEK293T/EV cells was assessed after 72 hours of treatment with the BTCE construct alone. No difference in cell viability was observed between HEK293T/CDla and HEK293T/EV cells (Supplementary Figure 3A) .

To validate the translational relevance of these results, CDla-expressing T-ALL cells (Jurkat, HPB-ALL, TALL- 1, and PF-382) or non-expressing cells (Ke-37) , were cocultured with PBMC with an E : T ratio of 10:1 in the presence of increasing BTCE concentration. Treatment resulted in a strong cytotoxic effect against CDla-expressing T-ALL cells, while no significant cytotoxicity of T cells against CDla- negative Ke-37 cells was observed (Figure 2D) . On primary cells from T-ALL-carrying patients, a cytotoxic effect equal to 60% of CDla-expressing cancer cells was observed (Figure 2E) . Again, no direct cytotoxicity was observed in the absence of effector cells (Complementary Figure 3B) and treatment with the BTCE construct does not trigger a direct cytotoxic effect. In addition, CD107a expression assessed in a concentration-dependent manner of the BTCE construct is increased in co-cultured T cells, with an E:T ratio of 10:1, with both CDla-expressing T-ALL cell lines and with primary T-ALL cells, indicating cytotoxic degranulation induced by the BTCE construct (Figure 2F) .

To demonstrate activation of BTCE-guided T cells against CDla-expressing T-ALL leukemic cells, high expression (HPB-ALL and TALL-1) and medium CDla (Jurkat) T- ALL cell lines were co-cultured with PBMC with an E:T ratio of 10:1 in the presence of increasing concentration of the BTCE construct.

Importantly, UMG2/CDla-CD3s BTCE determines concentration-dependent T cell activation, as assessed by up-regulation of early and late activation markers CD69 and CD25 on both CD4/CD8 T cells, respectively, as well as by the release of granzyme, perforin, and granulysin, and the production of pro-inflammatory cytokines, such as TNF-a, IFN-y, and IL-2 (Figure 3 A-3 D) .

The CD3s downstream signaling pathway was explored to investigate T cell activation. In particular, NFAT1 protein levels in PBMCs co-cultured at a 10: 1 E:T ratio with CDla- expressing T-ALL cell lines were evaluated, in the presence of the BTCE construct or vehicle as a control. Treatment with the BTCE construct induced higher levels of NFAT1 protein than control (Figure 4A) , inducing T cell proliferation (Figure 4B) .

Finally, to demonstrate that the cytotoxic effect induced by UMG2/CDla-CD3s BTCE is indeed dependent on T cells, CDla-expressing T-ALL cells were exposed to an increasing concentration of UMG2/CDla-CD3s BTCE and cocultured with PBMCs in their entirety, or PBMCs deprived of the fraction of CD4 or CD8-expressing T cells. Importantly, minimal cytotoxic activity was observed in both private CD4 or CD8 positive T cells samples compared to PBMCs in total (Figure 4C) . In addition, to exclude the presence of the Fc domain in the BTCE construct from inducing recruitment of natural immunity effector cells through the Fc-FcyR interaction by inducing cytokine release syndrome, the mechanisms of action of the BTCE construct by employing an Fc fragment blocker were further investigated. Jurkat cells expressing CDla were co-cultured with an E : T ratio of 10 : 1 with either total PBMCs , or devoid of CD56-expressing T cells , or total PBMCs exposed to Fc blockers , or, finally, only CD56+ T cel ls . Considering that the minimal activity was found in the presence of only CD56+ T cells , both CD56+ T cell deprived PBMCs and PBMCs in total with blocked Fc, were able to achieve a cytotoxic activity comparable to the total PBMCs not depleted by the CD56+ component ( Figure 4D) .

Together these results indicate that the BTCE construct exerts T-cell-mediated cytotoxicity against CDla-expressing T-ALL cells that does not involve the activation of other cytotoxic cells .

Subsequently, the Applicant investigated the in vivo anti-tumor activity of the BTCE construct against CDla- expressing T-ALL cells . Fluorescent HPB-ALL cells were inoculated subcutaneously in immunocompromised NSG mice . Seven days after subcutaneous inoculation of HPB-ALL cells , human PBMCs derived from a healthy donor were inoculated into the blood for the purpose of reconstituting human cytotoxic ef fects in experimental animals . Then, three days after PBMC engraftment , mice were randomi zed to receive intraperitoneal BTCE construct at two administrations per week . Tumor growth was assessed with an in vivo imaging system that detects tumor cell fluorescence . Treated animals were divided into two groups to receive the 0.1 or 0.5 mg/kg dose of the BTCE construct, and both experimental groups showed a significant reduction in tumor growth at both doses (Figure 5A) . This effect resulted in increased survival of animals in the treated groups. Precisely, a median survival of 41 days was evaluated in the vehicle-only group while 62 and 63 days was the median survival in the animal groups receiving 0.1 mg/kg and 0.5 mg/kg of the BTCE construct, respectively (Figure 5B) .

It is known that chronic antigenic stimulation induces overexpression of PDL-l/PDL-2 immune-checkpoints on cancer cells and expression of T-function depletion markers, such as PD-1, TIM-3, LAG-3 and TIGIT on T cells, thereby compromising the anti-tumor response in vivo. The Applicant investigated that immune-checkpoint inhibitors used in combination with the BTCE construct can counteract T cell dysfunction by representing a promising field of investigation .

To assess whether chronic antigen stimulation induces T cell dysfunction on T cells, PBMC cells and labeled cells were co-cultured in the presence of the BTCE construct or vehicle, as a control. After 72 hours, T cells were again stimulated with 10: 1 E: T-labeled T-ALL cells in the presence of the BTCE construct, or anti-PD-1, or anti PDL1 alone or in combination or the vehicle (control) . A signi ficant reduction in T cell depletion markers was found on T cells co-cultured with CDla expressing cells treated with the combination of anti-PD-l /anti-PDL- 1 and the BTCE construct , compared to BTCE alone ( Figure 6A) .

To date , the treatment o f patients with ref ractory/relapsing T-ALLs is still an obj ective to be achieved . Although immunotherapeutic approaches have revolutioni zed the prospects of patients with B-ALL, no immunotherapy-based strategy is currently approved and available for the treatment of T-ALL .

BTCE constructs represent an emerging approach to cancer immunotherapy, which is based on the promotion of immune synapses between ef fector T cells and tumor target cells , by antibody binding domains for both tumor-associated and T-cell-speci f ic antigens . Thus , BTCE technology provides a novel functionality that is not present in any combination of parental antibodies .

There are several formats of BTCE constructs , ranging from very small proteins , consisting of two single-chain variable fragments ( fragment-based) , to larger immunoglobulin G ( IgG ) -like molecules .

According to the invention a construct formed from the structure of the UMG2 antibody, IgG2A with two CDla binding arms , with a single scFv, derived from an anti-CD3 mAb OKT3 has been developed . This structure ( asymmetric, 2 + 1 ) was chosen based on advantageous pharmacodynamic and pharmacokinetic considerations . First , the presence of bivalent binding domains for CDla can increase selective recognition and elimination of cortical T-ALL cells highly expressing the CDla antigen, saving healthy cells expressing it at low levels . Second, the monovalent arm for CDS is preferred to the bivalent binding to avoid non-speci fic activation of T cells by CDSe to minimi ze entrapment of the BTCE construct in normal tissues rich in T lymphocytes . Third, the presence of the neonatal Fc receptor ( FcRn) , which protects the IgG- like BTCE from degradation, confers a long plasma hal f-li fe ( days ) compared to the shorter plasma hal fli fe of the fragment-based BTCE (hours ) , which therefore must be continuously infused . In particular, in vitro , BTCE produced concentration-dependent T-cell activation, the release of inflammatory cytokines , and the induction of T- cell proli feration and activation, leading to lysis of T-ALL cells with an E : T-dependent ratio . Consistent with in vitro data, BTCE led to signi ficant antitumor activity and survival benefit in animals treated in the in vivo trial . Importantly, for the translational value of the BTCE construct , the Applicant found this activity in a concentration range comparable to the doses used for the first already known and clinically approved BTCE , blinatumomab . For what has been described sofar, the invention provides a potential new immunotherapeutic strategy for the treatment of T-ALL. Furthermore, BTCE-based therapy is an effective standard strategy that does not require the complex and expensive ex-vivo manipulation of effector cells as in the case of CAR-T for the re-direction of cellular immunity on the tumor. Finally, BTCEs are characterised by improved dose management, which reduces the risks associated with cytokine release syndrome (CRS) and other toxicities commonly associated with CAR-T therapy.

The experimental data obtained according to the invention support the feasibility and efficacy of a new BTCE targeted to T-ALL cells, with a very low potential risk of immunosuppression, supporting the concept of precision immunotherapy in these patients.

The UMG2/CDla-CD3s BTCE have been analyzed and sequenced. Hereby reported some results of the analytical results :

NAME bispecific mAb-CD3 (UMG2)

QUANTITY (mg) 42

TARGET (mg) 20

BONUS (mg) 22

CONC. (mg/ml) 4, 63

VOLUME (ml) 9

BUFFER PBS 100 mmol/1 L-arginine ENDOTOXIN (EU/mg) < 1 EU / mg

MONOMERICITY ( % ) 35 , 9%

OD 8 , 2

EXT . COEFF . 1 , 77

Sequences of UMG2 antibody are filed within this PCT application and are :

> PRO 39573_evi-5 UMG . 2 . VH3-hlv2 . HC . S354C . T366W- anti . CD3 . scFv

EVQLQQSGPELVEPGTSVRISCKTSGYTFTEYTIHWVKQSHGKSLEWIGHINPS NGGNSYNQRFKGTATLTADKSSSTAYMELRRLTSDDSAVYYCARWGWVYALDYWGQGTS VTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFP AVLQSSGLYSLSSWTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCP APELLGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKT KPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQV YTLPPCRDELTKNQVSLWCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLY SKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKGGGGSGGGGSGGGGSDIKL QQSGAELARPGASVKMSCKTSGYTFTRYTMHWVKQRPGQGLEWIGYINPSRGYTNYNQK FKDKATLTTDKSSSTAYMQLSSLTSEDSAVYYCARYYDDHYCLDYWGQGTTLTVSSVEG GSGGSGGSGGSGGVDDIQLTQSPAIMSASPGEKVTMTCRASSSVSYMNWYQQKSGTSPK RWIYDTSKVASGVPYRFSGSGSGTSYSLTISSMEAEDAATYYCQQWSSNPLTFGAGTKL ELK

> PRO 39574_evi-5 UMG . 2 . VH3- hlv2 . HC . Y349C . T366S . L368A. Y407V

EVQLQQSGPELVEPGTSVRISCKTSGYTFTEYTIHWVKQSHGKSLEWIGHINPS NGGNSYNQRFKGTATLTADKSSSTAYMELRRLTSDDSAVYYCARWGWVYALDYWGQGTS VTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFP

AVLQSSGLYSLSSWTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPC P APELLGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKT KPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQV CTLPPSRDELTKNQVSLSCAVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLV SKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

> PRO 39575_evi-5 UMG . 2 . VL-hk ( LEIK) . LG

DIQMTQTTSSLSASLGDRVTISCSASQDISNYLDWYQQKPDGTVRLLIYYTSSL HSGVPSRFSGSGSGTAFSLTI INLEPEDVATYYCHQYSNLPYTFGGGTKLEIKRTVAAP SVEI FPPSDEQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDST YSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC

In conclusion, it has been shown that the invention of the UMG2 /CDla-CD3 s BTCE construct could represent a safe and ef fective anti-T-ALL strategy to be investigated in a first- in-human clinical study as a maintenance treatment , like blinatumomab, for the elimination of minimal residual disease (MRD) or in ref ractory/relapsed patients expressing CDla to improve the poor rate of disease control obtained with nelarabine . Furthermore , taking into account the good prognosis of patients with T-ALL expressing CDla, the present invention can provide a framework for the incorporation of the new BTCE construct into a first-line chemo- free treatment as a bridge for allogeneic stem cell transplantation, thus opening up new opportunities for the treatment of T-ALL, which is still incurable today . In addition, the asymmetric bispeci fic antibody (UMG2 /CDla-CD3 s ) having the characteristics described is applicable to the immunological treatment o f other pathological conditions such as Langerhans histiocytosis characteri zed by high expression of CDla .

Finally, it is clear that the new asymmetric bispeci fic antibody (UMG2 /CDla-CD3 s ) for the immunological treatment of pediatric and adult cortical-derived CDla-expressing T-cell acute lymphoblastic leukemias ( T-ALL ) described and illustrated herein can be modi fied and varied without departing from the protective scope of the present invention, as defined in the attached claims .