Despite remarksble improvements in the treatment of pediatric acute limphoblastic leukemia (ALL) over last decades, relapse still caries a poor prognosis with considerable morbidity and mortality. New strategies and approaches are extremely needed. According to the results of previous protocol ALL-REZ-2014 despite the implementation of bortezomib, as well as idarubicin and navelbine, which have not been used in first-line therapy, the new protocol had demonstrated no significant improvement in reaching a steady remission, especially for B-ALL relapses. The therapy results for the high risk group patients still far from optimal due to refractoriness to chemotherapy, death from infectious complications, as well as acute chemotherapy toxicity. This article demonstrates first results of pilot protocol All-REZ-2016 for high risk ALL relapses, which was approved by the Independent Ethics Committee of the Dmitry Rogachev National Medical Research Center of Pediatric Hematology, Oncology, and Immunology. Between 01.08.2016 and 01.12.2018 42 patients from our center were enrolled in ALL-REZ-2016 study. Patients aged 1-18 years with a first relapse of ALL without t(9;22) translocation were eligible , we included only high risk groups (S3, S4, S5) patients in this study. CD3+ apheresis was made before treatment for all patients with B-cell phenotype. To determine the efficiency and toxicity of fludarabine- and clofarabine-including chemotherapy regimens after the preliminary phase with prednisone 60mg/m 2 , for patients with B-ALL randomization for blocks with fludarabine or clofarabine was performed. The next stage of the therapy was a course with blinatumomab (28 days) and weekly autologous lymphocyte infusions (x 4). ALL patients with T-ALL got 11 days of protocol N, and then studying the efficiency of clofarabine 52 mg/m 2 for TIII/TIV was planned, since nelarabine has been shown to be inefficient for this category. For the patients with TI/II ALL, nelarabine 650 mg/m2 therapy was given after protocol N. If remission was attained, a stem cell transplant was performed for all patients without delay. 42 patients were enrolled, the median age was 9.9 years (range 2.7-16.2), there was male predominance (63.8%), the majority had B-cell phenotype (61.7%). The estimated 2-year event-free survival (EFS) across all immunophenotypes was 47.7% (SE+/-8%) and a cumulative incidence of second relapse (CIR) was 27.1% (SE+/-7.6%). Patients with B-ALL had EFS 55% +/-10% and CIR 30.0% +/-9.9%, T-ALL had EFS 35% +/-13% and CIR 21.2+/-12%. We did not observe serious toxicity when using blinatumomab. The use of clofarabine and fludarabine was accompanied by deep immunosuppression and severe infectious episodes. Results for patients with B-cell relapses are much better then in previous protocol ALL-REZ-2014, using of autologous CD3+ lymphocytes infusion once a week during the continuous blinatumomab therapy is effective. For T-ALL relapses results remain unsatisfactory due to refractoriness, but number of patients is still small for final conclusions.